ASCO Educational Book 2020

American Society of Clinical Oncology
UNITE AND CONQUER:

ACCELERATING PROGRESS TOGETHER
A PEER-REVIEWED, MEDLINE-INDEXED PUBLICATION
Volume 40
VOLUME 40
Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

Copyright © 2020 American Society of Clinical Oncology. All rights reserved.
American Society
of Clinical Oncology
Educational Book
Unite and Conquer:
Accelerating Progress Together
Volume 40
2020
© 2020 by American Society of Clinical Oncology Alexandria, VA

Contents
Letter From the Editor in Chief .................................................................................................................................... 1
INVITED ARTICLES
Escalating and De-escalating Therapy for Early-Stage HER2-Positive Breast Cancer
Danielle File, Giuseppe Curigliano, and Lisa A. Carey................................................................................................ 3
Palliative Care Skills and New Resources for Oncology Practices: Meeting the Palliative Care Needs of Patients With
Cancer and Their Families
Anthony Back, Tara Friedman, and Janet Abrahm.................................................................................................. 14
Emerging Subtypes and New Treatments for Castration-Resistant Prostate Cancer
Benedito A. Carneiro, Tamara L. Lotan, Andre de Souza, and Rahul Aggarwal ................................................eArticle
Molecular Profiling in Drug Development: Paving a Way Forward
Suzanne F. Jones and Andrew J. McKenzie.....................................................................................................eArticle
POINTS OF VIEW
Prophylactic Cranial Irradiation for Small-Cell Lung Cancer: Time for a Reassessment
Martin J. Edelman ................................................................................................................................................... 24
BREAST CANCER
Genomic Alteration in Metastatic Breast Cancer and Its Treatment
Allen Li, Stephen M. Schleicher, Fabrice Andre, and Zahi I. Mitri ........................................................................... 30
The Evolution of Clinical Trials in Metastatic Breast Cancer: Design Features and Endpoints That Matter
Andrew D. Seidman, Julia Maues, Tiah Tomlin, Vishal Bhatnagar, and Julia A. Beaver .......................................... 44
What’s the Price? Toxicities of Targeted Therapies in Breast Cancer Care
Carey K. Anders, Nicole R. LeBoeuf, Lara Bashoura, Saadia A. Faiz, Afreen I. Shariff, and Alexandra Thomas ...... 55
Avoiding the Swell: Advances in Lymphedema Prevention, Detection, and Management
Sarah A. McLaughlin, Nicole L. Stout, and Mark V. Schaverien........................................................................ eArticle
CANCER PREVENTION, RISK REDUCTION, AND GENETICS

Cancer Prevention in Low-Resource Countries: An Overview of the Opportunity
Sailaja Kamaraju, Jeffrey Drope, Rengaswamy Sankaranarayanan, and Surendra Shastri ....................................... 72
Integrating Genetic and Genomic Testing Into Oncology Practice
Susan M. Domchek, Elaine Mardis, Jennifer W. Carlisle, and Taofeek K. Owonikoko.......................................eArticle
CARE DELIVERY AND REGULATORY POLICY

From Theory to Practice: Implementation of Strategies to Reduce Acute Care Visits in Patients With Cancer
Bobby Daly, Laura C. Michaelis, John D. Sprandio, Jonathan T. Kapke, Ravi Kishore Narra, Elizabeth Malosh,
Alice Zervoudakis, Jessie Holland, and Melissa Zablocki......................................................................................... 84
ii 2020 ASCO EDUCATIONAL BOOK | asco.org/edbook

Expanding Access to Chimeric Antigen Receptor T-Cell Therapies: Challenges and Opportunities
Ankit Kansagra, Stephanie Farnia, and Navneet Majhail ..................................................................................eArticle
CENTRAL NERVOUS SYSTEM TUMORS

Gray Areas in the Gray Matter: IDH1/2 Mutations in Glioma
Martin J. van den Bent, Ingo K. Mellinghoff, and Ranjit S. Bindra........................................................................... 96
The Practical Application of Emerging Technologies Influencing the Diagnosis and Care of Patients With Primary
Brain Tumors
Leland S. Hu, Daniel J. Brat, Orin Bloch, Shakti Ramkissoon, and Glenn J. Lesser .........................................eArticle
DEVELOPMENTAL THERAPEUTICS—IMMUNOTHERAPY
Best Foot Forward: Neoadjuvant Systemic Therapy as Standard of Care in Triple-Negative and HER2-Positive
Breast Cancer
Priyanka Sharma, Roisin M. Connolly, Evanthia T. Roussos Torres, and Alastair Thompson ............................eArticle
Beyond Tumor PD-L1: Emerging Genomic Biomarkers for Checkpoint Inhibitor Immunotherapy
Galina G. Lagos, Benjamin Izar, and Naiyer A. Rizvi ........................................................................................eArticle
Biomarkers in Precision Cancer Immunotherapy: Promise and Challenges
William B. McKean, Justin C. Moser, David Rimm, and Siwen Hu-Lieskovan................................................... eArticle
Leveraging Patient-Derived Models for Immunotherapy Research
Katerina Politi ................................................................................................................................................... eArticle
DEVELOPMENTAL THERAPEUTICS—MOLECULARLY TARGETED AGENTS AND TUMOR BIOLOGY

Antibody-Drug Conjugates: Patient and Treatment Selection
Shalini Makawita and Funda Meric-Bernstam ....................................................................................................... 105
Selective CDK4/6 Inhibitors: Biologic Outcomes, Determinants of Sensitivity, Mechanisms of Resistance, Combinatorial
Approaches, and Pharmacodynamic Biomarkers
Erik S. Knudsen, Geoffrey I. Shapiro, and Khandan Keyomarsi ............................................................................. 115
The Evolution of Antibody-Drug Conjugates: A Positive Inflexion Point
Anthony W. Tolcher............................................................................................................................................... 127
The Resurgence of Antibody Drug Conjugates in Cancer Therapeutics: Novel Targets and Payloads
Valentina Boni, Manish R. Sharma, and Amita Patnaik ....................................................................................eArticle
Strategic Combinations to Prevent and Overcome Resistance to Targeted Therapies in Oncology
Ozge Gumusay, Pietro Paolo Vitiello, Chiara Wabl, Ryan B. Corcoran, Alberto Bardelli, and Hope S. Rugo......eArticle
GASTROINTESTINAL CANCER—COLORECTAL AND ANAL

Controversies in Rectal Cancer Treatment and Management
Weijing Sun, Raed Al-Rajabi, Rodrigo O. Perez, Saquib Abbasi, Ryan Ash, and Angelita Habr-Gama................... 136
Redefining Colorectal Cancer by Tumor Biology
Mohamed E. Salem, Alberto Puccini, and Jeanne Tie ........................................................................................... 147
Colorectal Cancer in the Young: Epidemiology, Prevention, Management
Rebecca L. Siegel, Christopher Dennis Jakubowski, Stacey A. Fedewa, Anjee Davis, and Nilofer S. Azad .......eArticle
GASTROINTESTINAL CANCER—GASTROESOPHAGEAL, PANCREATIC, AND HEPATOBILIARY

Overcoming Resistance to Targeted Therapies in Gastrointestinal Cancers: Progress to Date and Progress to Come
Christopher Chen, Maria Di Bartolomeo, Salvatore Corallo, John H. Strickler, and Lipika Goyal ............................ 161
asco.org/edbook | 2020 ASCO EDUCATIONAL BOOK iii

The Treatment of Hepatocellular Carcinoma With Portal Vein Tumor Thrombosis
Motaz Qadan, Nishita Kothary, Bruno Sangro, and Manisha Palta ........................................................................ 174
Moving Beyond Chemotherapy for Pancreaticobiliary Tumors: Targeted and Immunotherapy Strategies
Rebecca Allen, Naama Halpern, Sandra Algaze, Talia Golan, Anthony B. El-Khoueiry,
and Rachna T. Shroff ....................................................................................................................................... eArticle
GENITOURINARY CANCER—KIDNEY AND BLADDER

Current Approaches to the Treatment of Advanced or Metastatic Renal Cell Carcinoma
Matthew Tenold, Praful Ravi, Monika Kumar, Alex Bowman, Hans Hammers, Toni K. Choueiri, and
Primo N. Lara Jr .................................................................................................................................................... 187
Renal Cell and Urothelial Carcinoma: Biomarkers for New Treatments
Andrew L. Schmidt, Arlene Siefker-Radtke, David McConkey, and Bradley McGregor .....................................eArticle
GENITOURINARY CANCER—PROSTATE, TESTICULAR, AND PENILE

Sorting Through the Maze of Treatment Options for Metastatic Castration-Sensitive Prostate Cancer
Brian Schulte, Alicia K. Morgans, Neal D. Shore, and Carmel Pezaro ................................................................... 198
Immune Checkpoint Blockade for Prostate Cancer: Niche Role or Next Breakthrough?
Daniel Vargas P. de Almeida, Lawrence Fong, Matthew B. Rettig, and Karen A. Autio.....................................eArticle
Recent Advances in the Management of High-Risk Localized Prostate Cancer: Local Therapy, Systemic Therapy, and
Biomarkers to Guide Treatment Decisions
Rana R. McKay, Felix Y. Feng, Alice Y. Wang, Christopher J. D. Wallis, and Kelvin A. Moses ..........................eArticle
GERIATRIC ONCOLOGY
Toward Modernization of Geriatric Oncology by Digital Health Technologies
Armin Shahrokni, Kah Poh Loh, and William A. Wood........................................................................................... 209
The Globalization of Geriatric Oncology: From Data to Practice
Ravindran Kanesvaran, Supriya Mohile, Enrique Soto-Perez-de-Celis, and Harpreet Singh ..............................eArticle
GLOBAL HEALTH
Resource-Stratified Guideline-Based Cancer Care Should Be a Priority: Historical Context and Examples of Success
Natasha Hunter, Naomi Dempsey, Fayez Tbaishat, Mohammad Jahanzeb, Sana Al-Sukhun, and
Julie R. Gralow ...................................................................................................................................................... 217
The Use of Health-Related Technology to Reduce the Gap Between Developed and Undeveloped Regions Around
the Globe
Wilfred Ngwa, Ian Olver, and Kathleen M. Schmeler ............................................................................................. 227
GYNECOLOGIC CANCER
A Review of Immune Checkpoint Blockade Therapy in Endometrial Cancer
Angela K. Green, Jacqueline Feinberg, and Vicky Makker .................................................................................... 238
Therapeutic Targets and Opportunities in Endometrial Cancer: Update on Endocrine Therapy and Nonimmunotherapy
Targeted Options
Helen J. MacKay, Victor Rodriguez Freixinos, and Gini F. Fleming........................................................................ 245
PARP Inhibitors for Ovarian Cancer: Current Indications, Future Combinations, and Novel Assets in Development to
Target DNA Damage Repair
Panagiotis A. Konstantinopoulos, Stephanie Lheureux, and Kathleen N. Moore ...............................................eArticle
iv 2020 ASCO EDUCATIONAL BOOK | asco.org/edbook

Immune Therapy Opportunities in Ovarian Cancer
Lana E. Kandalaft, Kunle Odunsi, and George Coukos .....................................................................................eArticle
HEAD AND NECK CANCER

Novel Strategies to Effectively De-escalate Curative-Intent Therapy for Patients With HPV-Associated Oropharyngeal
Cancer: Current and Future Directions
Katharine A. R. Price, Anthony C. Nichols, Colette J. Shen, Almoaidbellah Rammal, Pencilla Lang, David A. Palma,
Ari J. Rosenberg, Bhisham S. Chera, and Nishant Agrawal................................................................................... 257
Recent Advances in the Development of Biomarkers and Chemoradiotherapeutic Approaches for Nasopharyngeal
Carcinoma
Brigette B.Y. Ma, Yu-Pei Chen, Edwin P. Hui, Xu Liu, Allen K.C. Chan, Anthony T.C. Chan, and Jun Ma ............. 270
HEALTH SERVICES RESEARCH AND QUALITY IMPROVEMENT

How Do We Align Health Services Research and Quality Improvement?
Devika Das, Lalan Wilfong, Katherine Enright, and Gabrielle Rocque .................................................................... 282
Successful Strategies to Address Disparities: Insurer and Employer Perspectives
Manali I. Patel, Richard Snyder, and Otis Brawley................................................................................................. 291
Direct-to-Consumer Advertising for Cancer Centers and Institutes: Ethical Dilemmas and Practical implications
Fay J. Hlubocky, Daniel F. McFarland, Patricia A. Spears, Laura Smith, Bonnie Patten, Jeffery Peppercorn, and
Randall Holcombe............................................................................................................................................ eArticle
The Changing Health Insurance Coverage Landscape in the United States
K. Robin Yabroff, Samuel Valdez, Mireille Jacobson, Xuesong Han, and A. Mark Fendrick..............................eArticle
HEMATOLOGIC MALIGNANCIES
ADCs, BiTEs, CARs, and Small Molecules: A New Era of Targeted Therapy in Non-Hodgkin Lymphoma
Jeremy S. Abramson, Nilanjan Ghosh, and Sonali M. Smith ................................................................................. 302
Clinical Controversies in the Management of Smoldering Multiple Myeloma
Oliver C. Lomas and Irene M. Ghobrial.................................................................................................................. 314
Current Perspectives on Therapy for Chronic Lymphocytic Leukemia
Farrukh T. Awan, Othman Al-Sawaf, Kirsten Fischer, and Jennifer A. Woyach...................................................... 320
Recent Advances in Managing Acute Lymphoblastic Leukemia
Daniel J. DeAngelo, Elias Jabbour, and Anjali Advani............................................................................................ 330
The Changing Landscape of Treatment In Acute Myeloid Leukemia
Kristin Koenig, Alice Mims, Mark J. Levis, and Mary M. Horowitz.......................................................................... 343
The Role of Early Intervention in High-Risk Smoldering Myeloma
Nisha S. Joseph, Madhav V. Dhodapkar, and Sagar Lonial ................................................................................... 355
Treatment of Smoldering Multiple Myeloma: Expectant Observation Should Still Be the Standard
Rafael Fonseca and Miguel Gonzalez-Velez .......................................................................................................... 364
Newly Diagnosed Myeloma in 2020
Philippe Moreau, Cyrille Touzeau, Ravi Vij, Scott R. Goldsmith, and Ashley E. Rosko ......................................eArticle
LUNG CANCER
Checkpoint Blockade in Lung Cancer With Driver Mutation: Choose the Road Wisely
Antonio Calles, Jonathan W. Riess, and Julie R. Brahmer ..................................................................................... 372
asco.org/edbook | 2020 ASCO EDUCATIONAL BOOK v

Making Checkpoint Inhibitors Part of Treatment of Patients With Locally Advanced Lung Cancers: The Time Is Now
Mark G. Kris, Corinne Faivre-Finn, Tiana Kordbacheh, Jamie Chaft, Jia Luo, Anne Tsao, and
Stephen Swisher............................................................................................................................................... eArticle
Promising Immuno-Oncology Options for the Future: Cellular Therapies and Personalized Cancer Vaccines
Benjamin J. Solomon, Paul A. Beavis, and Philip K. Darcy ..............................................................................eArticle
MELANOMA/SKIN CANCERS
And Now for Something Completely Different: Immunotherapy Beyond Checkpoints in Melanoma
Isabella Claudia Glitza, Stephanie L. Goff, Merrick Ross, and Kim Margolin .......................................................... 386
Immunologic Characteristics of Nonmelanoma Skin Cancers: Implications for Immunotherapy
Evan T. Hall, Elena Fernandez-Lopez, Ann W. Silk, Reinhard Dummer, and Shailender Bhatia ............................ 398
PEDIATRIC ONCOLOGY
Crossing Oceans: Preclinical Collaboration to Improve Pediatric Drug Development
Gregory Reaman, Louis Stancato, Gilles Vassal, and John M. Maris...................................................................... 409
Cutting to the Front of the Line: Immunotherapy for Childhood Acute Lymphoblastic Leukemia
Jennifer L. McNeer, Rachel E. Rau, Sumit Gupta, Shannon L. Maude, and Maureen M. O’Brien....................eArticle
Before It’s Too Late: Multistakeholder Perspectives on Compassionate Access to Investigational Drugs for Pediatric
Patients With Cancer
Elena Gerasimov, Martha Donoghue, Josh Bilenker, Tanya Watt, Nancy Goodman, and
Theodore W. Laetsch........................................................................................................................................ eArticle
PROFESSIONAL DEVELOPMENT AND EDUCATION ADVANCES

On the Shoulders of Giants: The Evolution of Renal Cell Carcinoma Treatment—Cytokines, Targeted Therapy,
and Immunotherapy
Janice P. Dutcher, Ronan Flippot, Jaleh Fallah, and Bernard Escudier................................................................. 418
The Next Step in Your Career: Getting the Right Opportunity and the Right Contract
Claire F. Verschraegen, Yousif Abubakr, and Mark Lee......................................................................................... 436
SARCOMA
How Technology Is Improving the Multidisciplinary Care of Sarcoma
Inga-Marie Schaefer, Kelvin Hong, and Anusha Kalbasi ........................................................................................ 445
Secondary Sarcomas: Biology, Presentation, and Clinical Care
Vanessa Eulo, Harry Lesmana, Leona A. Doyle, Kim E. Nichols, and Angela C. Hirbe........................................... 463
SYMPTOMS AND SURVIVORSHIP

Bringing Life to Death: The Need for Honest, Compassionate, and Effective End-of-Life Conversations
Amy R. MacKenzie and Michelle Lasota................................................................................................................ 476
Dermatologic Adverse Events of Systemic Anticancer Therapies: Cytotoxic Chemotherapy, Targeted Therapy,
and Immunotherapy
Alana Deutsch, Nicole R. Leboeuf, Mario E. Lacouture, and Beth N. McLellan..................................................... 485
Optimizing Cardiovascular Health in Patients With Cancer: A Practical Review of Risk Assessment, Monitoring, and
Prevention of Cancer Treatment–Related Cardiovascular Toxicity
Susan F. Dent, Robin Kikuchi, Lavanya Kondapalli, Roohi Ismail-Khan, Christine Brezden-Masley, Ana Barac, and
Michael Fradley..................................................................................................................................................... 501
vi 2020 ASCO EDUCATIONAL BOOK | asco.org/edbook

Taboo Topics in Adolescent and Young Adult Oncology: Strategies for Managing Challenging but Important
Conversations Central to Adolescent and Young Adult Cancer Survivorship
Giselle K. Perez, John M. Salsman, Kaitlyn Fladeboe, Anne C. Kirchhoff, Elyse R. Park, and
Abby R. Rosenberg .......................................................................................................................................... eArticle
Medical Decision-Making in Oncology for Patients Lacking Capacity
Jonathan M. Marron, Kaitlin Kyi, Paul S. Appelbaum, and Allison Magnuson .................................................. eArticle
asco.org/edbook | 2020 ASCO EDUCATIONAL BOOK vii

Editor Roster
EDITOR IN CHIEF
Don S. Dizon, MD, FACP, FASCO
ASSOCIATE EDITORS
Nathan Pennell, MD, PhD, FASCO
Hope S. Rugo, MD, FASCO
2020 ASCO MEETINGS AND MEMBER PUBLICATIONS

EDITORIAL BOARD
Suzanne Cole, MD, FACP
Neelima Denduluri, MD
Anne Katz, PhD, RN, FAAN
Miriam A. Knoll, MD
Stephen Leong, MD
Jason Luke, MD, FACP
Rana R. McKay, MD
Travis Osterman, DO, MS
Mohamed E. Salem, MD
Antoinette R. Tan, MD, MHSc, FACP
Breelyn A. Wilky, MD
MANAGING EDITOR EDITORIAL OFFICE

Lindsay F. Pickell, MFA American Society of Clinical Oncology, Inc.
2318 Mill Road, Suite 800
EDITORIAL ASSISTANT
Alexandria, VA 22314
Meagan Foy
Tel: (571) 483-1300
PRODUCTION MANAGER Fax: (571) 366-9550
Donna Dottellis Email: [email protected]
viii 2020 ASCO EDUCATIONAL BOOK | asco.org/edbook

General Information
The 2020 ASCO Educational Book (Print ISSN: 1548- DISCLAIMER
8748 40; Electronic ISSN: 1548-8756) is published ASCO assumes no responsibility for errors or omis-
by the American Society of Clinical Oncology, Inc. sions in this publication. The reader is advised to
(“ASCO”). check the appropriate medical literature and the
product information currently provided by the man-
DESCRIPTION ufacturer of each drug to be administered to verify,
The ASCO Educational Book is an NLM-indexed, among other matters, the dosage, the method and
peer-reviewed collection of articles written by ASCO duration of administration, or contraindications. It is
Annual Meeting faculty and invited experts in on- the responsibility of the treating physician or other
cology. Published annually, each volume highlights health care professional, relying on the independent
the most compelling research and developments experience and knowledge of the patient, to de-
across the multidisciplinary fields of oncology with the termine drug dosages and the best treatment for the
goal of improving patient value and care. patient. The ideas and opinions expressed in this
publication do not necessarily reflect those of ASCO.
PERMISSION REQUESTS
The mention of any company, product, service, or
Requests for permission to reprint all or part of any
therapy mentioned does not constitute an endorse-
article published in this title should be directed to
ment of any kind by ASCO. ASCO assumes no re-
Permissions, American Society of Clinical Oncology,
sponsibility for any injury or damage to persons or
Inc., 2318 Mill Road, Suite 800, Alexandria, VA
property arising out of or related to any use of the
22314. Tel: (571) 483-1300; fax: (571) 366-9550; or
material contained in this publication.
email: [email protected].
COPYRIGHT 2020 INDIVIDUAL PRICES

Copyright © 2020 American Society of Clinical On- Print: $99
cology, Inc. All rights reserved. No part of this pub- Digital: Free Public Access
lication may be reproduced or transmitted in any form
or by any means, electronic or mechanical, including
IMPORTANT ACCESS AND PRICING NOTES
photocopy, recording, or any information storage and
The ASCO Educational Book print editions can be
retrieval system, without permission in writing from
purchased at the ASCO Store (shop.asco.org). The
ASCO.
ASCO Educational Book provides free online public
Copies of articles in this publication may be made for access to all digital editions at asco.org/edbook.
personal use. This consent is given on the condition,
Prices are subject to change without notice. Current
however, that the copier pay the stated per-copy fee
prices are in effect for back volumes. Single issues,
through the Copyright Clearance Center, Inc. (222
both current and back, exist in limited quantities and
Rosewood Drive, Danvers, MA 01923) for any copying
are offered for sale subject to availability.
beyond that permitted by Sections 107 or 108 of the
U.S. Copyright Law. This consent does not extend to For further information, email [email protected] or
other kinds of copying, such as copying for general call (888) 273-3508.
distribution, for advertising or promotional purposes, for
creating new collective works, or for resale.
2020 ASCO EDUCATIONAL BOOK | asco.org/edbook ix

2020 Annual Meeting Disclosure
As the continuing education provider for the Meeting, ASCO is committed to balance, objectivity, and
scientific rigor in the management of financial interactions with for-profit health care companies that could
create real or perceived conflicts of interest. Participants in the Meeting have disclosed their financial
relationships in accordance with ASCO’s Policy for Relationships with Companies; review the policy at asco.
org/rwc.
ASCO offers a comprehensive disclosure management system, using one disclosure for all ASCO activities.
Members and participants in activities use coi.asco.org to disclose all interactions with companies. Their
disclosure is kept on file and can be confirmed or updated with each new activity.
Please email [email protected] with specific questions or concerns.
x ASCO EDUCATIONAL BOOK | asco.org/edbook

2019–2020 Annual Meeting Education Committee
The Annual Meeting Education Committee initiates, develops, plans, and assesses the needs for the ASCO Annual
Meeting education program.
Tatiana Prowell, MD – Chair DEVELOPMENTAL THERAPEUTICS — MOLECULARLY

Taofeek Owonikoko, MD, PhD, MS – Chair-Elect TARGETED AGENTS AND TUMOR BIOLOGY
Colin D. Weekes, MD, PhD – Immediate Past Chair David S. Hong, MD – Track Leader
Ryan Bruce Corcoran, MD, PhD
BREAST CANCER Victoria Villaflor, MD
Stacy Moulder, MD, MS – Track Leader Kwok-Kin Wong, MD, PhD
Fabrice Andre, MD, PhD William C. Zamboni, PharmD, PhD
Neelima Denduluri, MD
Kelly Hunt, MD GASTROINTESTINAL CANCER —
Sara Hurvitz, MD COLORECTAL AND ANAL
Andrew D. Seidman, MD Weijing Sun, MD, FACP – Track Leader
Erica Stringer-Reasor, MD Nilofer Saba Azad, MD
Alexandra Thomas, MD, FACP Sharlene Gill, MD, MPH, MBA
Mohamed E. Salem, MD
CANCER PREVENTION, RISK REDUCTION,
AND GENETICS GASTROINTESTINAL CANCER —
Otis W. Brawley, MD, MACP, FASCO – Track Leader GASTROESOPHAGEAL, PANCREATIC,
Banu Arun, MD AND HEPATOBILIARY
Joseph Curry, MD James Posey, MD – Track Leader
Susan M. Domchek, MD Ian Chau, MD
Sailaja Kamaraju, MD Peter Enzinger, MD
Mark A. Lewis, MD Lipika Goyal, MD
Victoria Seewaldt, MD Shubham Pant, MD, MBBS
Gina M. Villani, MD, MPH Rachna T. Shroff, MD
CARE DELIVERY AND REGULATORY POLICY GENITOURINARY CANCER — KIDNEY AND BLADDER
Suzanne Cole, MD, FACP – Track Leader Primo Lara, MD – Track Leader
Edward Riker Arrowsmith, MD, MPH James Brugarolas, MD, PhD
Robert Michael Daly, MD, MBA Bradley McGregor, MD
Kurt Demel, MD, MBA Chana Weinstock, MD
Adam P. Dicker, MD, PhD, FASCO Evan Y. Yu, MD
Nicole Gormley, MD
GENITOURINARY CANCER — PROSTATE,
CENTRAL NERVOUS SYSTEM TUMORS TESTICULAR, AND PENILE
Priya Kumthekar, MD – Track Leader Matt Galsky, MD – Track Leader
Tracy Todd Batchelor, MD Mark T. Fleming, MD
Simon S. Lo, MD, FACR, FASTRO Joshua Meeks, MD, PhD
Ingo Mellinghoff, MD Alicia K. Morgans, MD, MPH
DEVELOPMENTAL THERAPEUTICS — GERIATRIC ONCOLOGY

IMMUNOTHERAPY Armin Shahrokni, MD, MPH – Track Leader
Scott Gettinger, MD – Track Leader Kah Poh (Melissa) Loh, MD
Todd Michael Bauer, MD Harpreet Singh, MD
Benjamin Besse, MD, PhD Tanya Wildes, MD
Charles G. Drake, MD, PhD
Lei Zheng, MD, PhD
2020 ASCO EDUCATIONAL BOOK | asco.org/edbook xi

GLOBAL HEALTH PEDIATRIC ONCOLOGY
Razvan Popescu, MD, MRCP – Track Leader Theodore W. Laetsch, MD – Track Leader
Sana A. Al Sukhun, MD, MSc Martha B. Donoghue, MD
Eduardo L. Cazap, MD, PhD, FASCO Stephen Hunger, MD
Eduard Vrdoljak, MD, PhD Margaret Macy, MD
GYNECOLOGIC CANCER PROFESSIONAL DEVELOPMENT AND EDUCATION

Amit Oza, MD – Track Leader ADVANCES
Kathleen N. Moore, MD Brian C. Boulmay, MD – Track Leader
Amanda J. Walker, MD Rachel J. Buchsbaum, MD
Jeremy P. Cetnar, MD
HEAD AND NECK CANCER Thomas H. Openshaw, MD, MS
Nishant Agrawal, MD – Track Leader Sorbarikor Piawah, MD, MPH
Muralidhar Beeram, MD, MBBS Vyshak Alva Venur, MD
Hisham Mehana, PhD
Lori Wirth, MD SARCOMA
William D. Tap, MD – Track Leader
HEALTH SERVICES RESEARCH AND QUALITY Mark Agulnik, MD
IMPROVEMENT Gina Z. D’Amato, MD
Michael Halpern, MD, PhD, MPH – Track Leader Katherine A. Thornton, MD
Christopher R. Friese, PhD, RN, AOCN
Fabian M. Johnston, MD SYMPTOMS AND SURVIVORSHIP
Dax Kurbegov, MD Amy R. MacKenzie, MD, FACP – Track Leader
Aaron Philip Mitchell, MD Debra L. Barton, PhD, AOCN, FAAN, RN
Katherine Virgo, PhD, FASCO Norah Lynn Henry, MD, PhD
Maryam B. Lustberg, MD
HEMATOLOGIC MALIGNANCIES Kevin C. Oeffinger, MD
Ashley Elizabeth Rosko, MD – Track Leader Jennifer S. Temel, MD
Jeremy S. Abramson, MD
Aref Al-Kali, MD LIAISONS
Nilanjan Ghosh, PhD, MBBS Smita Bhatia, MD, MPH, FASCO – Cancer Survivorship
Catriona HM Jamieson, MD, PhD Committee
Anuj Kumar Mahindra, MD Tithia Biswas, MD – Cancer Research Committee
Joseph Mikhael, MD Mary L. (Nora) Disis, MD, FASCO – Cancer Prevention
Committee
LUNG CANCER Matthew Katz, MD – Clinical Practice Guidelines
Taofeek Owonikoko, MD, PhD, MS – Track Leader Committee
Gideon M. Blumenthal, MD Jonathan Marron, MD, MPH – Ethics Committee
Shruti Jolly, MD Ray Page, DO, PhD, FACOI, FASCO – Clinical Practice
Rosalyn A. Juergens, MD, PhD Committee
Mark G. Kris, MD, FASCO Gabrielle Rocque, MD, MSPH – Quality of Care Council
Christopher S. Lathan, MD, MS, MPH Leonard Saltz, MD – Value Task Force
Carlos Sampaio-Filho, MD – International Affairs
MELANOMA/SKIN CANCERS Committee
Ryan J. Sullivan, MD – Track Leader K. Robin Yabroff, PhD – Health Equity Committee
Rodabe Navroze Amaria, MD
Paolo Antonio Ascierto, MD
Kenneth F. Grossmann, MD, PhD
Eduardo M. Sotomayor, MD
xii 2020 ASCO EDUCATIONAL BOOK |asco.org/edbook

ASCO Educational Book Expert Panel
The Expert Panel is a group of well-recognized international physicians and researchers in oncology and related fields
who serve as peer reviewers of the ASCO Educational Book.
Raafat Abdel Malek, MD, PhD, Vijay Chaudhary, MD, MPH Jennifer Gao, MD
FRCR Zhijian Chen, MD, PhD Jacqueline Garcia, MD
Reham Abdel-Wahab, MD, PhD Haiying Cheng, MD, PhD Ignacio Garrido-Laguna, MD, PhD
Ghassan Abou-Alfa, MBA, MD Rangaswamy Chintapatla, MD Margaret Gatti-Mays, MD, MPH
Maysa Abu-Khalaf, MD Ami Chitalia, MD Sumit Gaur, MD
Melissa Accordino, MD Rakesh Chopra, MD Sophia George, PhD
David Adelstein, MD, FACP Melvin Chua, PhD, MBBS, FRCR Fatemeh Ghazanfari Amlashi, MD
Neeraj Agarwal, MD James Cleary, MD Sharlene Gill, MD, MPH, MBA
Boniface Ago, FWACS Ira Cohen, PharmD Karthik Giridhar, MD
Thierry Alcindor, MD Suzanne Cole, MD, FACP Jason Gold, MD
Aref Al-Kali, MD Leandro Colli, MD, PhD Melanie Goldfarb, MD, FACS,
Gustavo Almeida, MD, PhD Jose Mauricio Correia Mota, MD, PhD FACE, MSc
Amarendra Amar, MD Jennie Crews, MD Jonathan Goldman, MD
Rui Amaral Mendes, PhD, DMD Suzanne Dahlberg, PhD Priscila Goncalves, MD
James Anderson, PhD Tong Dai, MD, PhD Wilson Gonsalves, MD
Daniel Fontes Argolo, MD Senthil Damodaran, MD, PhD Carlos Gonzales, MD
Avan Armaghani, MD Tu Dan, MD David Graham, MD, FASCO
Mehmet Artac, MD Arvind Dasari, MD, MBBS Michael Grossbard, MD
Joseph Baar, MD, PhD Elizabeth Davis, MD Stephanie Guarino, MD
Elke Backman, PharmD Nancy Davis, MD Carmen Guillen-Ponce, PhD, MD
Maria Baggstrom, MD Roselle De Guzman, MD Shuchi Gulati, MD FACP
Pedro Barata, MD Ramon De Mello, MD, PhD, FACP Shilpa Gupta, MD
Megan Baumgart, MD Jocelyn De Yao, MD Shilpi Gupta, MD
Kebede Begna, MD Wilfred Delacruz, MD, PhD Philip Haddad, MD, FACP, FCCP
Charles Bennett, MD, PhD, FASCO Sukhbinder Dhesy-Thind, MD Rami Haddad, MD, FACP
Adam Berger, MD Elie Dib, MD, FACP Missak Haigentz, MD
Cristiane Bergerot, PhD, MS Elizabeth Dickson, MD Hatem Halabi, MD
Eric Bernicker, MD Urshila Durani, MD Lee Hartner, MD
Ananta Bhatt, MD Olive Eckstein, MD David Hermel, MD
Gouri Bhattacharyya, MBBS, MD, Martin Edelman, MD Lisa Holle, PharmD
MNAMS, PhD, FRCP Nagi El Saghir, MD, FACP Scott Huntington, MD, MPH
Amit Bhoil, MD Yannick Eller, MD Jimmy Hwang, MD
Kathleen Bickel, MD, MPhil, MS Hamid Emamekhoo, MD Nahida Islam, MD
Marijo Bilusic, MD, PhD Lawson Eng, MD Neil Iyengar, MD
Burthia Booker, PhD Vinicius Ernani, MD Gopa Iyer, MD
Marc Braunstein, MD, PhD Francisco Esteva, MD, PhD Shivi Jain, MD
Kathryn Bruno, PhD Jaleh Fallah, MD Anshu Jain, MD
Joseph Bubalo, BCOP, PharmD Bryan Faller, MD Murali Janakiram, MD, MS
Venkata Bulusu, MD, FRCR Zhongling Feng, PhD, MD Mauro Janoski, MD, FACP
Kathryn Burns, PhD Mary Jo Fidler, MD Sanford Jeames, PhD
Timothy Call, MD Gunnar Folprecht, MD Joanne Jeter, MD
Eduardo Castanon Alvarez, MD Mario Fontes E Sousa, MD Maxine Jochelson, MD
Paul Celano, MD Marie Belle Dela Cruz Francia, MD Robin Jones, MD, MBBS, MRCP
Pavani Chalasani, MD Danielle Friedman, MD, MS Himanshu Joshi, MBBS, MPH, PhD
George Chang, MD, MS Shirish Gadgeel, MD Sheheryar Kabraji, MD
John Charlson, MD Leena Gandhi, MD, PhD Chandana Kakani, MD
Gurkamal Chatta, MD Apar Kishor Ganti, MD Kevin Kalinsky, MD, MS
2020 ASCO EDUCATIONAL BOOK | asco.org/edbook xiii

Muhammad Kamal, MD Jesus Anampa Mesias, MD, MS Ravi Ramjeesingh, MD, PhD, FRCPC
Jasmine Kamboj, MD Nabel Mir, MD Brian Ramnaraign, MD
Hyunseok Kang, MD Kriti Mittal, MD Devika Rao, MD
Asha Karippot, MD Kabir Mody, MD Arpit Rao, MD
Margaret Kasner, MD, MSCE Catherine Moltzan, MD, FRCPC Ruta Rao, MD
Varinder Kaur, MD Halle Moore, MD Bernardo Rapoport, MD
Uqba Khan, MD Daniel Morel, MD Kakil Rasul, FRCP
Hina Khan, MD Scott Morgan, MD Abram Recht, MD, FASCO
Kathleen Kiernan Harnden, MD Vicki Morrison, MD Matthew Reilley, MD
Alba Kihn-Alarcón, MD, MSc Lindsay Morton, PhD Tracy Rose, MD
Chul Kim, MD Yasser Mostafa Kamel, MD, MBA Anna Roshal, MD
Gretchen Kimmick, MD, MS Muhamad Alhaj Moustafa, MD Melanie Royce, MD, PhD
Randall Kimple, MD, PhD George Mridula, MD Moira Rushton, MD, FRCPC
Samuel Klempner, MD Sarbajit Mukherjee, MD, MS Joshua Sabari, MD
Kelsey Klute, MD Kalyan Mukherjee, MBBS, MD Adrian Sacher, MD, MMedSc
David Korones, MD Daniel Mulrooney, MD, MS Solmaz Sahebjam, MD
Smitha Krishnamurthi, MD Alex Mutombo Baleka, MD Sadia Saleem, MD
Jairam Krishnamurthy, MD Swapna Narayana, MD Mohamed Salem, MD
Abhishek Kumar, MBBS Sunita D. Nasta, MD Rachel Sanborn, MD
Arun Kumar, MD Elizabeth Nichols, MD Sergio Santillana, MD
Shaji Kumar, MD Sorush Niknamian, MPH, PhD Edgardo Santos, MD, FACP
Dax Kurbegov, MD Saroj Niraula, MD, MS Matthew P. Schlumbrecht, MD,
Bonnie Labdi, PharmD, BCOP, RPh Adaeze Nwosu Iheme, MD MPH, FACOG
Mario Lacouture, MD Paul Oberstein, MD Mina Sedrak, MD, MS
Primo Lara, MD Tonia Onyeka, MBBS Nagashree Seetharamu, MD
Sayeh Lavasani, MD Moshe Ornstein, MD, MA Amikar Sehdev, MD, MPH
Sayeh Lavasani, MD, Ana Laura Ortega Granados, MD Thomas Semrad, MD, FACP
Rachel Layman, MD Elmar Orujov, MD, MBA Rahul Seth, DO
Ticiana Leal, MD Travis Osterman, DO, MS Bhuvana Setty, MD
Thomas LeBlanc, MD, MA, Uwa Otabor, MD Nishi Shah, MBBS, MPH
MHS, FAAHPM Jose Pablo Leone, MD Manisha Shah, MD
Lynne Lederman, PhD Aparna Parikh, MD, MS Safi Shahda, MD
Byung Ha Lee, PhD Rahul Parikh, MD Ahmad Shams, MD
Richard Lee, MD Wungki Park, MD Mukesh Shanthilal, MD
Daniel Lenihan, MD, FACC Sandip Patel, MD Gregg Shepard, MD
Chi Lin, MD, PhD James Pauff, MD, PhD Rachna Shroff, MD
Jerry Liu, MD, PhD Katrina Pedersen, MD, MS Sandra Silberman, MD, PhD
Kejun Liu, MD Angela Pennisi, MD, PhD Charles Simone, MD
Shelly Lo, MD Herman Perroud, MD, PhD Eric Singer, MD, MA, FACS
Richard LoCicero, MD Namrata Peswani, MD Navneet Singh, MD, FACP, FCCP
Herbert Loong, MBBS Aby Philip, MBBS Donielle Sliwa, MD, MPH
Yohann Loriot, MD, PhD Maria Catherine Pietanza, MD Heloisa Soares, MD
Yung Lyou, MD, PhD Katja Pinker-Domenig, MD Maria Leticia Solari, MD,PhD
Jyoti Malhotra, MD, MPH Navin Pinto, MD Enrique Soto Perez De Celis, MD
Monica Malik, MD, DNB Priyanka Pophali, MBBS Adam Sowalsky, PhD
Joshua Mammen, MD Steven Powell, MD Alexander Spira, MD, PhD, FACP
Rami Manochakian, MD Amanda Przespolewski, DO Amirrtha Srikanthan, MD
Maddalena Marchesi, MD Luis Raez, MD, FACP, FCCP James Stevenson, MD
Peter Maslak, MD Devalkumar Rajyaguru, MD Tom Stinchcombe, MD
Bartomeu Massuti, MD Suresh Ramalingam, MD Steve Stricker, MS, PharmD
Aju Mathew, MD, MPhil, FACP Ramya Ramaswami, MBBS, Sriman Swarup, MBBS, MD
Heather McArthur, MD, MPH MRCP, MPH James Talcott, MD
Inderjit Mehmi, MD David Ramirez, MD Mala Talekar, MD
xiv 2020 ASCO EDUCATIONAL BOOK | asco.org/edbook

Gevorg Tamamyan, MD, PhD, MSc Athira Unnikrishnan, MD, MBBS Stephanie Weiss, MD
Antoinette Tan, MD, FACP Matias Valsecchi, MD, MS Young Whang, MD, PhD
Marios Konstantinos Tasoulis, MD, Catherine Van Poznak, MD, FASCO Nicole Williams, MD
PhD Noam VanderWalde, MD Melissa Wilson, MD, PhD
Sarah Temkin, MD Neil Vasan, MD, PhD Paul Wissel, MD, FACP
Min Yuen Teo, MD, MRCP Vamsidhar Velcheti, MD William Wong, MD
Haluk Tezcan, MD David Vesole, MD, PhD, FACP Antoinette Wozniak, MD
Ramya Thota, MBBS Irina Veytsman, MD Suayib Yalcin, MD
Swapna Thota, MD Praveen Vikas, MBBS Daniel Yang, MD
Aleksei Tikhonov, MSc Katherine Virgo, PhD, FASCO Dawei Yang, MD
Narhari Timilshina, PhD Evan Vosburgh, MD Kelly Khai Li Yap, MD
Gabriel Tinoco, MD Christine Walko, PharmD, BCOP, Sriram Yennu, MD, MS, FAAHPM
Sanja Trajkova, PhD FCCP Sri Yeruva, MD
Daniel Trifiletti, MD Ying Wang, MD Yousef Zakharia, MD
Janice Tsang, MD Yubao Wang, MD, PhD Sara Zaknoen, MD
Derek Tsang, MD Lixia Wang, PhD Marjorie Zauderer, MD, MS, FACP
Alice Tzeng, PhD James Ward, MD Alexandra Zimmer, MD
Nataliya Uboha, MD, PhD Benjamin Weinberg, MD Danielle Zimmerman, MD
2020 ASCO EDUCATIONAL BOOK | asco.org/edbook xv

Letter From the Editor in Chief
To my colleagues in the oncology care community:
At the time of this writing, we are facing an unprecedented public health crisis as cases of and deaths from
COVID-19 continue to rise in many cities around the world. Rather than us gathering at Chicago’s
McCormick Place to share our knowledge and celebrate discoveries in cancer research, we are meeting
virtually to practice physical distancing while McCormick Place is being converted into a field hospital with
over 2,500 beds for patients with COVID-19. With the change to a science program–only virtual Annual
Meeting this May, the ASCO Educational Book represents the important contributions of the Education
Program faculty until the oral presentations can be given.
The Ed Book’s purpose is to inform clinicians and researchers about the most compelling developments
in oncology, from applying breakthrough therapies in practice to addressing the needs of the growing number of
cancer survivors. Like the field of oncology, the Ed Book has considerably transformed over the last 4 decades.
What started in 1985 as a meeting resource for attendees has grown into a cherished and trusted medical text for
members of the cancer care team across the world. Almost half of Ed Book’s readership originates from outside
the United States, and its articles have been accessed over 400,000 times in the last year. The popularity of
Ed Book is attributable to its steadfast mission: to improve patient care and quality through clinician education.
In that spirit, my associate editors, Nathan Pennell, MD, PhD, FASCO, and Hope S. Rugo, MD, FASCO, and
I have the privilege of presenting the 40th volume of the NLM-indexed ASCO Educational Book.
Howard A. “Skip” Burris III, MD, FASCO, FACP, 2019–2020 ASCO President, chose “Unite and Conquer:
Accelerating Progress Together” as the theme for this year’s Meeting. This theme recognizes the rapid
evolution of cancer science and the urgent need to engage in information-sharing to improve the lives of our
patients—a need that is ever-more relevant in the era of COVID-19. It is my sincerest belief that Ed Book
plays a critical role in achieving this. The insight found within its pages comes from the brightest minds in
oncology who have been chosen to serve as Annual Meeting Education Program faculty. Most articles
represent a collaboration between session speakers and their invited co-authors, many of whom are junior
faculty and fellows. It takes tremendous effort and time to ‘“unite” and document the knowledge of many
into one scholarly review. I cannot thank the faculty enough for the generosity of their contributions to this
year’s edition. It is a gift to current and future oncology care team members.
I would also like to recognize the expert panel members who selflessly dedicated their time to perform thorough
and thoughtful peer reviews. Finally, a well-deserved message of appreciation to both Nate Pennell and Hope
Rugo for their dedication and camaraderie on this labor of love. I literally could not have done it without you.
The 2020 ASCO Educational Book articles, as well as articles from past volumes, are available online for free
public access and download at www.asco.org/edbook. We welcome your feedback and suggestions on how
we can improve the content, so please contact us at [email protected] with your comments.
From all of us at the ASCO Educational Book, please stay safe at this incredibly difficult time.
Sincerely,
Don S. Dizon, MD, FACP, FASCO

ASCO Educational Book Editor in Chief
2020 ASCO EDUCATIONAL BOOK | asco.org/edbook 1

INVITED ARTICLES
This year’s Invited Articles represent the 2020 ASCO Annual Meeting theme, “Unite and Conquer: Accel-
erating Progress Together.” These important contributions to the 40th volume of the ASCO Educational Book
focus on the promising areas of research that urgently need greater attention and have the greatest potential
to improve care and outcomes. Authors were nominated by the ASCO Educational Book Editors and 2020
Annual Meeting leadership.
ARTICLES
Escalating and De-escalating Therapy for Early-Stage HER2-Positive Breast Cancer

Danielle File, Giuseppe Curigliano, and Lisa A. Carey
Palliative Care Skills and New Resources for Oncology Practices: Meeting the Palliative Care Needs of
Patients With Cancer and Their Families
Anthony Back, Tara Friedman, and Janet Abrahm

BREAST CANCER
invited article overview Escalating and De-escalating Therapy for

Early-Stage HER2-Positive Breast Cancer
Danielle File, MD1; Giuseppe Curigliano, MD, PhD2; and Lisa A. Carey, MD, ScM1
Untreated, HER2+ disease is the most aggressive breast cancer phenotype; however, the development of
multiple highly effective HER2-targeting drugs has transformed treatment and survival. These drugs include
the anti-HER2 monoclonal antibodies trastuzumab and pertuzumab; small molecule inhibitors lapatinib,
neratinib, and tucatinib; and antibody-drug conjugates trastuzumab emtansine (T-DM1) and now trastuzumab
deroxtecan. More complex regimens using these drugs continue to improve outcomes, but the incremental
benefits of these advances are often modest. Improved outcomes came from the addition of HER2-targeted
therapies to conventional chemotherapy, beginning with trastuzumab, then pertuzumab added to trastuzumab,
or with neratinib given for the year after trastuzumab. Neoadjuvant, or preoperative, administration of che-
motherapy plus HER2-targeting allows surgical deescalation and tailoring treatment by pathologic complete
response (pCR) to therapy. Patients with pCR after conventional therapy have excellent outcomes; what we now
know is that the poorer outcomes associated with residual disease can be ameliorated with adjuvant T-DM1.
However, as we have developed more complex, effective, and expensive therapy to maximize outcomes, it is also
true that we are overtreating many patients. In stage I HER2+ breast cancer, there are excellent outcomes with
paclitaxel plus trastuzumab or T-DM1 alone. Higher clinical stage HER2+ disease is still treated aggressively,
although intrinsic subtype or activated immune tumor microenvironment may identify those with augmented
treatment response or better outcome. It is likely that future strategies to escalate and de-escalate treatment
with less chemotherapy, fewer anti-HER2 drugs, or shorter duration will depend upon integrated clinical and
genomic modeling.
INTRODUCTION binds to the HER2 extracellular domain, causing HER2

Breast cancer is a heterogeneous disease with unique downregulation and decreased cell growth and di-
tumor biology, treatment options, and clinical course in vision. Initial results in metastatic HER2+ breast cancer
patients with each immunohistochemical subtype. HER2+ were so promising that patient advocates picketed
breast cancer is a particularly aggressive subtype that Genentech for increased access to the drug via a lottery
affects 20% to 25% of patients with breast cancer. system. In 1998, trastuzumab became the first U.S.
Before the development of HER2-targeted therapies, Food and Drug Administration–approved targeted therapy
it was associated with the poorest outcomes, with fre- for the treatment of a solid tumor on the basis of the
quent, early relapse and high rates of visceral and central seminal clinical trial demonstrating improved time to
nervous system (CNS) metastasis.1,2 disease progression, higher objective response rate,
longer duration of response, improved median overall
The discovery of the HER2 oncogene and development
survival, and a 20% reduction in the risk of death when
Author affiliations of HER2-targeted therapy is a remarkable story. In
combined with standard chemotherapy in patients with
and support 1989, nearly 3 decades before The Cancer Genome
information (if HER2+ metastatic breast cancer.4
Atlas and other comprehensive genome sequencing
applicable) appear
projects, Slamon and colleagues 3 recognized that The definition of HER2 positivity has become more
at the end of this
article. HER2 gene amplification was strongly correlated with rigorous over time, allowing greater accuracy and treatment
Accepted on HER2 protein overexpression and that gene alteration effectiveness. Current ASCO/College of American
February 26, 2020 was associated with distinct clinical outcome. They Pathologists guidelines define HER2+ tumors as those
and published at hypothesized that HER2 was a gene involved in the that are 3+ on immunohistochemistry, corresponding
ascopubs.org on to a circumferential membrane staining that is com-
pathogenesis of some breast and ovarian cancers,
April 2, 2020:
DOI https://doi.org/
which represented an opportunity for a novel treatment plete, intense (readily appreciated using a low-power
10.1200/EDBK_ tailored to tumor biology. This led to the development of objective), and in more than 10% of tumor cells, or
100023 trastuzumab, a humanized monoclonal antibody that in situ hybridization that is positive on the basis of

File, Curigliano, and Carey
survival from 20.3 to 25.1 months, and nearly doubled the

response rate when administered first line.4 The benefits of
PRACTICAL APPLICATIONS
the addition of trastuzumab to chemotherapy were not
• Neoadjuvant systemic therapy is preferred for without some cost, however, most notably cardiac dys-
clinical stage II to III since it can decrease
function, which was observed in more than 25% of women
surgical extent, reducing need for axillary lymph
who received trastuzumab concurrently with an anthracy-
node dissection and allowing for adjuvant tai-
lored therapy. cline. Pertuzumab, which also binds to the extracellular
domain of the HER2 receptor in a mechanism comple-
• Outside of T1N0, optimal HER2-directed (neo)
mentary to that of trastuzumab, was subsequently approved
adjuvant regimens include at least two cytotoxic
on the basis of the CLEOPATRA trial, in which the first-line
agents plus trastuzumab (AC-TH or TCH);
adding pertuzumab improves outcomes in combination of docetaxel, trastuzumab, and pertuzumab
node-positive regardless of hormone receptor resulted in median progression-free survival (mPFS) of 18.
status (AC-THP, TCHP). 5 months and a stunning 16-month improvement in median
overall survival from 40.8 to 56.5 months compared with
• Tailored adjuvant treatment after neoadjuvant
therapy substitutes 14 cycles of T-DM1 in those docetaxel, trastuzumab, and placebo.6 Despite dual HER2
with residual disease rather than completion of receptor targeting, no significant increase in cardiotoxicity
1 year of trastuzumab-based therapy. was observed.
• Neratinib improved outcomes when given for
Small-Molecule Inhibitors
a second year after 1 year of chemotherapy plus
trastuzumab; the role of this drug after a Lapatinib is a reversible EGFR (HER1) and HER2 tyrosine
pertuzumab-containing regimen or after kinase inhibitor, initially approved for use after disease pro-
tailored therapy with T-DM1 is unclear. gression on trastuzumab, an anthracycline, and a taxane. In
• In systemically untreated T1N0 disease, the this setting, lapatinib combined with capecitabine increased
APT trial regimen with paclitaxel plus trastu- time to progression from 4.4 to 8.4 months compared with
zumab for 12 weeks followed by trastuzumab capecitabine alone.7 Preclinical studies also demonstrated
for a total of 1 year provides excellent long-term a synergistic interaction when lapatinib and trastuzumab are
outcomes. administered concurrently, prompting a subsequent trial
design to include lapatinib with or without trastuzumab in
patients who had previously experienced progression on at
a single-probe average HER2 copy number of at least 6. least one trastuzumab-containing regimen. In this trial, ad-
0 signals/cell or a dual-probe HER2/CEP17 (chromosome dition of trastuzumab to lapatinib resulted in a hazard ratio of
enumeration probe 17) ratio of 2.0 or greater with an av- 0.73 for PFS compared with lapatinib alone, with a nearly
erage HER2 copy number of 4.0 signals/cell or greater. If the doubled clinical benefit rate, demonstrating the value of
dual-probe HER2/CEP17 ratio is less than 2.0, but the av- ongoing trastuzumab even after progression on a trastuzumab-
erage HER2 copy number is 6.0 signals/cell or greater, this is containing regimen.8 In tumors that are positive for both
also considered to be HER2 overexpressing.5 hormone receptors (HRs; estrogen receptor) and HER2,
first-line lapatinib added to letrozole improved mPFS from
THE BEGINNING OF ESCALATION: HER2-DIRECTED 3.0 to 8.2 months, with a hazard ratio for risk of disease
THERAPEUTICS IN METASTATIC DISEASE progression of 0.71 compared with letrozole alone and a clinical
benefit rate of 48% versus 29%.9 Combined with the TAnDEM
Novel therapies and treatment escalation strategies were
trial, which added trastuzumab to anastrozole and also dem-
first introduced in the metastatic setting, then moved into
onstrated poor mPFS with the aromatase inhibitor alone (2.4
the neoadjuvant and adjuvant setting. HER2-targeted ther-
months) and a significant improvement with the addition of
apies have dramatically improved outcomes in both the
HER2-targeting therapy (to 4.8 months; p = .0016), these
metastatic and early breast cancer settings and include two
findings established the importance of cotargeting estrogen
monoclonal antibodies (trastuzumab and pertuzumab), three
receptor and HER2.10
small-molecule inhibitors (lapatinib, neratinib and tucatinib),
and two antibody-drug conjugates (T-DM1 and trastuzumab Neratinib, an irreversible pan-HER2 inhibitor, was superior
deruxtecan). to the reversible HER1/HER2 inhibitor lapatinib when added
to capecitabine in heavily pretreated patients in the NALA
Monoclonal Antibodies trial, with a near doubling of 1-year PFS.11 Neratinib was also
The addition of trastuzumab to standard chemotherapy in studied in the first-line metastatic setting in the NEfERT-T
HER2+ metastatic breast cancer increased median time to trial, where it failed to show superiority over trastuzumab
progression from 4.6 to 7.4 months, improved median overall when each were combined with paclitaxel. Both the NALA
4 2020 ASCO EDUCATIONAL BOOK | asco.org/edbook

Tailoring Therapy for Early HER2-Positive Breast Cancer
trial and the NEfERT-T trial demonstrated the CNS activity of ESCALATION IN THE (NEO)ADJUVANT SETTING
neratinib, with the neratinib arm of the NALA trial dem- Addition of HER2-targeted therapies to treatment regimens
onstrating delayed time to intervention for patients with CNS for HER2+ early breast cancer soon followed advances in
involvement and the neratinib-treated patients in NEfERT-T the metastatic setting. In 2005, trastuzumab was approved
showing a lower incidence of CNS recurrences and delay in in the adjuvant setting in combination with chemotherapy
time to CNS metastases.12 for patients with operable HER2+ breast cancer on the basis
Tucatinib, a third-generation HER2 small-molecule inhibitor, of results from a combined analysis of two trials.20 The Na-
also seems to be active in the CNS, as shown in the tional Surgical Adjuvant Breast and Bowel Project (NSABP)
HER2CLIMB trial, which randomly assigned patients who B-31 trial assessed the efficacy of doxorubicin and cyclo-
were previously treated with trastuzumab, pertuzumab, and phosphamide followed by paclitaxel with and without 1 year
T-DM1 to receive trastuzumab and capecitabine with either of trastuzumab, starting with the first dose of paclitaxel. The
tucatinib or placebo. This trial demonstrated significantly North Central Cancer Treatment Group trial N9831 included
improved mPFS from 5.6 to 7.8 months and mOS from 17.4 the same two study arms, as well as an arm in which tras-
to 21.9 months, with the proportion of patientes surviving tuzumab was initiated after the completion of chemotherapy.
2 years increasing from 26.6% to 44.9%. Approximately When the comparable study arms were analyzed jointly, an
45% of the trial population had CNS involvement; 1-year absolute difference in disease-free survival (DFS) of 12% at
PFS was 24.9% among patients with brain metastases who 3 years and a 37% reduction in the risk of death was ob-
were treated in the tucatinib-containing arm compared with served with the addition of trastuzumab to paclitaxel after the
0% of patients in the placebo arm.13 completion of doxorubicin and cyclophosphamide.20,21 A
subsequent treatment escalation trial, HERA, focused on the
Antibody-Drug Conjugates duration of trastuzumab, finding that the addition of 1 year of
Antibody-drug conjugates capable of targeting HER2, then adjuvant trastuzumab after locoregional therapy and neo-
delivering a cytotoxic payload provide a third HER2-targeted adjuvant or adjuvant chemotherapy resulted in an 8.4%
treatment strategy. T-DM1 is a monoclonal antibody that absolute benefit in DFS with no additional statistically sig-
binds to the HER2 extracellular domain and then delivers nificant benefit from completion of a second year of trastu-
the microtubule inhibitor DM1, a maytansine derivate, only zumab.22 Regarding the optimal chemotherapy regimen to
to HER2-overexpressing cells. In addition to the benefit of combine with trastuzumab, the BCIRG 006 trial tested both
the selective delivery of cytotoxic therapy resulting in fewer an anthracycline/taxane-based regimen (AC-TH) and a tax-
off-target adverse effects, T-DM1 was demonstrated to offer ane/platinum-based regimen (TCH), finding that both
a median survival advantage of more than 5 months in trastuzumab-containing treatment arms were associated
patients who were previously treated with trastuzumab and with superior 5-year DFS and overall survival compared with
a taxane, both in the EMILIA trial, where it was compared chemotherapy alone. The trial was not powered to compare
with lapatinib and capecitabine,14 and in the TH3RESA trial the two trastuzumab arms, but numerically AC-TH had
in which it was compared with the physician’s choice of slightly fewer breast cancer events and slightly more cardiac
therapy. 15 Trastuzumab deruxtecan (DS-8201a) is the and leukemia events.23
newest antibody-drug conjugate composed of an anti-HER2
On the basis of the benefits in the metastatic setting, multiple
antibody, a cleavable tetrapeptide-based linker, and a cy-
neoadjuvant trials tested dual HER2-targeting with either
totoxic topoisomerase I inhibitor payload. It has a particu-
lapatinib (NeoALTTO,24 CALGB 40601,25 NSABP B-4126) or
larly high drug-to-antibody ratio (8; more than double that
pertuzumab (NeoSphere,27,28 TRYPHAENA29) added to che-
of T-DM1). In heavily treated patients with prior exposure
motherapy plus trastuzumab. These trials were developed with
to both trastuzumab and T-DM1, the DESTINY-Breast01
pCR as the endpoint on the basis of the role of pCR as an
trial demonstrated an impressive objective response rate
intermediate biomarker for improved DFS. These trials dem-
of 61% with a median duration of response of 14.8 months
onstrated higher pCR rates with the addition of the second
and mPFS of 16.4 months. Enthusiasm was only slightly
agent and resulted in accelerated U.S. Food and Drug Ad-
tempered by the toxicity profile, which notably includes
ministration approval of pertuzumab in this setting in 2013.
interstitial lung disease with several treatment-related
deaths in the trial.16-18 Trastuzumab duocarmazine is Given successful combinations in the neoadjuvant and met-
another antibody-drug conjugate that delivers a DNA- astatic settings, pertuzumab was added to trastuzumab and
alkylating payload to HER2 overexpressing tumor cells. adjuvant chemotherapy in the APHINITY trial.30 In this trial, the
It demonstrated activity in heavily pretreated patients 3-year rate of invasive DFS (iDFS) was 92% in patients with
with HER2-expressing metastatic cancer, and possibly node-positive disease who received pertuzumab with adjuvant
HER2-low tumors, and is being investigated in a phase chemotherapy and trastuzumab compared with 90.2% in
III clinical trial. 19 those who were treated with placebo (hazard ratio, 0.81).

This benefit was statistically significant (p = .02) and was in salvage T-DM1 as the new standard of care for patients
maintained in the 6-year analysis with a 24% decrease in with HER2+ residual disease after neoadjuvant therapy.
iDFS, which translated to an absolute risk reduction in node- Given their activity in pretreated metastatic breast cancer,
positive patients of 4.5%. This improved iDFS was observed trials using both tucatinib and trastuzumab deruxtecan in
in both HR+ and HR HER2+ breast cancers. A significant the residual disease setting are in development. In partic-
overall survival benefit has not yet been observed in this trial, ular, the high rate of brain metastases in HER2+ breast
although there were fewer deaths in the pertuzumab arm at cancer, with little impact on preventing metastases at this
the 6-year analysis.31 Patients with node-negative disease site with T-DM1 in the KATHERINE trial, means that pre-
did not benefit from the addition of pertuzumab. vention of CNS metastases remains an unmet need. Given
Unlike pertuzumab, the addition of lapatinib to adjuvant the clear evidence of tucatinib efficacy in the CNS com-
trastuzumab did not result in improved outcomes in the partment in metastatic disease in the HER2CLIMB trial, this
ALTTO trial.32 The DFS hazard ratio of 0.84 did not meet will be an endpoint of COMPASS-RD, a National Cancer
prespecified statistical endpoints, and the drug was asso- Institute National Clinical Trials Network trial in development
ciated with increased diarrhea, rash, and hepatotoxicity. that will test tucatinib added to T-DM1 in the high-risk re-
Although results from adjuvant lapatinib were disappointing, sidual disease setting.
the addition of the second-generation HER2-targeted ty- TAILORING THERAPY INTENSITY
rosine kinase inhibitor neratinib for 1 year after the com-
pletion of chemotherapy and adjuvant trastuzumab led to With current treatment paradigms involving polychemotherapy,
a modest increase in 2-year iDFS from 91.6% to 93.9% in plus up to three anti–HER2-targeting regimens and 3-year
the ExteNET trial.33 This benefit persisted in the 5-year iDFS now close to 90%,35 it is clear that we are overtreating
analysis, with a 5-year iDFS rate of 90.2% compared with many, if not most, patients with early-stage HER2+ breast
87.7% in the placebo arm.34 In a subgroup analysis, the cancer. Selecting the optimal subgroups for treatment
hazard ratio for iDFS was 0.60 and was statistically signif- escalation remains a challenge. Clinicopathologic features
icant for patients with HR+ tumors compared with 0.95 in remain important for prognostication across breast cancer
patients with HR disease. The overall survival data from subtypes and continue to help risk stratify patients to guide
this study are not yet mature; current guidelines recom- treatment recommendations. Addition of adjuvant pertu-
mend consideration of adjuvant neratinib after the com- zumab offered benefit only in patients with involved axillary
pletion of trastuzumab in patients with node-positive HER2+ lymph node status in the APHINITY trial. HR status is also
breast cancer with high risk of recurrence, with European relevant, and the biology underlying HR+/HER2+ and
approval including only patients with HR+ disease. Clinical HR /HER2+ breast cancer is distinct, although the most
benefit in patients who have also received pertuzumab and/ recent updated analysis of the APHINITY trial found similar
or T-DM1 is not known. benefits in HR+ and HR subsets (the earlier analysis sug-
gested only HR benefited). The ExteNET trial suggested that
POSTNEOADJUVANT SALVAGE THERAPY the benefit of a second year of HER2-targeting therapy with
Standard treatment of early HER2+ breast cancer has neratinib after trastuzumab was limited to the HR+ subgroup.
shifted from the adjuvant setting to the neoadjuvant setting. Given that this was a somewhat surprising finding that was not
Although there are no data suggesting a relapse or survival hypothesized a priori, it is not part of the U.S. Food and Drug
benefit with this approach, systemic treatment before sur- Administration approval, although treating physicians may
gical intervention improves the possibility of breast-conserving limit use of neratinib to very high–risk patients with HR+
therapy, reduces the need for axillary dissection, and allows disease.
for the assessment of pathologic response at the time of In addition to using clinicopathologic features to determine
definitive surgery. This allows for more comprehensive risk treatment intensity, biomarkers are now being routinely
stratification and the delivery of salvage therapy in patients collected and analyzed as part of clinical trials in search of
with residual or refractory disease. genomic predictors of therapy benefit. In ExteNET, for ex-
Given that patients with residual disease after neoadjuvant ample, the significance of PIK3CA mutations was of interest,
therapy have substantially worse prognosis than patients given that neratinib inhibits the PI3K/Akt signaling pathway.
who achieve pCR, the KATHERINE trial assessed whether Tissue specimens from 42% of the participating patients in
escalation to salvage therapy with adjuvant T-DM1 rather this trial were analyzed for two hotspot mutations in exon
than the standard completion of 1 year of trastuzumab could 9 and one hotspot mutation in exon 20 and underwent
improve outcomes.35 In this study, 3-year iDFS was sig- fluorescence in situ hybridization for PIK3CA amplification.
nificantly higher with T-DM1, with a hazard ratio for invasive The benefit of neratinib over placebo was greater in PIK3CA-
disease or death of 0.50 compared with trastuzumab. This altered tumors than PIK3CA wild-type tumors; however, the
improved outcome was observed in all subsets and resulted interaction test was not significant. These intriguing data are

insufficient to support the use of PIK3CA alterations as and multidrug regimens should be considered with treat-
a predictive biomarker for response to neratinib on the basis ment escalation strategies, and caution should be exercised
of this study.36 in using therapies with high cost and low clinical benefit.
The development and approval of biosimilar therapies is
Without more sensitive and specific biomarkers to tailor
expected to improve drug costs for patients; however, the
therapy on the basis of the risk of recurrence and the
large-scale impact of this has not yet been quantified.
likelihood of treatment response, overtreatment remains
a concern in the setting of treatment escalation in early DE-ESCALATION STRATEGIES
breast cancer. Whereas the addition of pertuzumab did not
As in the treatment of HR+ early breast cancer, improved
significantly increase cardiac events over adjuvant trastu-
outcomes in HER2+ early breast cancer has raised interest
zumab alone, any added adverse effects of a therapy that
in possible de-escalation strategies. De-escalation studies
does not provide benefit has an unfavorable risk-benefit
are often difficult to perform, as noninferiority trials are large
ratio. Very few treatments are without any side effects, and
and expensive, and frequently, therapeutic guidelines and/
grade 3 diarrhea, for example, occurs more frequently with
or clinical practice include patients who were excluded or
the addition of pertuzumab. It is especially important to
under-represented in randomized clinical trials, such as
consider lasting side effects of therapies in a patient pop-
patients with small, node-negative HER2+ breast cancer. It
ulation being treated with curative intent in whom any
is also true that both doctor and patient bias reduce en-
permanent side effects would persist potentially for decades
thusiasm for de-escalating strategies when the current
with ongoing detriment to quality of life. Based on cohort
treatment is well tolerated, as is often the case with anti-
studies in the United States, financial toxicity is also be-
HER2 drugs, such as trastuzumab, pertuzumab, or T-DM1.
coming increasingly recognized as a prominent concern
that should be considered when recommending new The APT trial examined patients with small, node-negative,
therapies and multidrug treatment regimens. Using claims HER2+ breast cancers. In an uncontrolled design, this trial
data from women with private health insurance diagnosed assessed whether patients with early-stage tumors could be
from 2008 to 2012, Giordano and colleagues37 showed that treated with paclitaxel and trastuzumab alone, targeting an
the addition of trastuzumab to chemotherapy adds nearly iDFS of no worse than 92%. The investigators reported a 3-
$80,000 of insurance payments, nearly $700 to median year iDFS of 98.7% with only two of 406 patients experi-
out-of-pocket payments, and more than $2,000 to top encing distant metastatic disease.45 At 7 years, DFS was
quartile out-of-pocket payments. This leaves some patients 93% with four distant recurrences and an overall survival of
responsible for more than $7,000 in out-of-pocket expenses 95%.46 APT changed clinical practice, providing a lower-
to receive their therapy. Effects of employment disruption toxicity effective regimen for low–clinical risk tumors (most
and additional costs associated with travel, lodging, and patients in the trial had 2 cm node-negative cancers).
childcare contribute to a risk of bankruptcy for many pa- ATEMPT examined 1 year of T-DM1 in patients with stage I
tients, making it unsurprising that financial toxicity has been HER2+ breast cancer.47 In this study, in the 383 patients
associated with reduced quality of life.38 Financial toxicity is who were treated with T-DM1 based on a 3:1 randomization
also associated with early mortality,39 making identification design, 3-year DFS was 97.7% with only two distant re-
of patients at risk for financial toxicity even more important, currences; however, 17% of patients discontinued therapy
although difficult. Younger patients with non-Medicare in- as a result of toxicity compared with 6% of patients who were
surance and those with lower socioeconomic status40 and treated in a parallel arm with the APT paclitaxel and tras-
minority race are at increased risk; however, shorter time tuzumab regimen. Adjuvant T-DM1 had less neuropathy,
from diagnosis and higher stage also place patients at risk,41 less neutropenia, and fewer infusion reactions, but more
and insured status does not eliminate the risk of financial thrombocytopenia, alanine transaminase increases, and
distress.42 Patients with Medicare also remain at risk of fi- bilirubin increases. The T-DM1 group seldom experienced
nancial toxicity, with patient costs for stage I breast cancer asymptomatic decline in left ventricular ejection fraction
exceeding $8,500, not including many of the newer HER2- (1.3% vs. 6.1%) and virtually no congestive heart failure.
targeting therapeutics, which will increase the patient fi- T-DM1 thus may be an alternative to paclitaxel and trastu-
nancial burden significantly.43 This distress often persists zumab for select patients with stage I HER2+ breast cancer. It
after treatment completion and affects one in every three is worth noting, however, that both APT and ATEMPT in-
survivors of cancer.44 In a recent survey, 79% of patients cluded largely or exclusively T1N0 tumors, limiting the gen-
with breast cancer reported that they would prefer to know eralizability of both of these regimens to higher-risk subsets.
about the out-of-pocket costs for their treatment before they In larger or node-positive HER2+ tumors, the clearest dem-
are treated, and 40% agreed that their doctor should onstration of successful treatment de-escalation in HER2+
consider the out-of-pocket costs when making a treatment disease derives from the shift to a neoadjuvant approach,
recommendation.41 The financial impact of novel therapies which has profound effects on surgical management. In

an analysis of the surgical results of CALGB 40601, a phase III HR+/HER2+ tumors behave differently from HR /HER2+
trial of neoadjuvant paclitaxel with trastuzumab, lapatinib, or tumors, including having significantly lower pCR rates re-
both, 43% of patients felt by the surgeon to be candidates only gardless of regimen, less association of pCR to outcome,
for mastectomy converted to breast-conservation candi- and typically lower absolute benefit differences with aug-
dates after neoadjuvant treatment; breast conservation mented anti-HER2 therapy, with some exceptions. HR+/
was successful 80% of the time.48 Neoadjuvant therapy HER2+ breast cancer with HER2 fluorescence in situ hy-
also reduces the likelihood of nodal involvement by more bridization ratio of less than 5 or with higher ESR1 levels
than 10% in clinically node-negative disease and converts derive less benefit from adjuvant trastuzumab.57 Although
approximately 50% of node-positive to node-negative this may suggest opportunities for alternate treatment
disease, thereby allowing omission of axillary dissection.49 strategies in patients with HR+/HER2+ disease, at this time,
Sentinel lymphadenectomy in this setting is more chal- these have not been tested.
lenging and is only accurate enough if more than two An intriguing prospect is whether some tumors are so de-
nodes are examined using dual tracer and possibly pendent upon HER2 (HER2 oncogene addicted) that cure
clipping.49 can be effected without chemotherapy. In general, small
Another effort at de-escalation involved the duration of studies using all-biologic regimens demonstrate lower pCR
HER2-targeted adjuvant therapy. In PERSEPHONE, more rates than regimens incorporating chemotherapy. In Neo-
than 4,000 women received either 6 months or 1 year of Sphere, one arm of the trial included only trastuzumab plus
adjuvant trastuzumab, with 4-year DFS of 89.4% versus 89.8%, pertuzumab (HP) without chemotherapy administered to
meeting noninferiority endpoints.50 In addition, 5% less 107 patients; that arm had a pCR rate of 17%.27 In TBCRC
patients reported severe adverse events, and fewer patients 006, pCR was observed in 27% of 64 patients (21% in HR+/
required early treatment discontinuation as a result of HER2+ and 36% in HR /HER2+) after neoadjuvant lapa-
cardiotoxicity. The PHARE trial tested the same hypothesis tinib plus trastuzumab.58 In the follow-up randomized TBCRC
in more than 3,300 patients with a different result: the 023 trial, extended dual anti-HER2 blockade (plus endocrine
hazard ratio for DFS at 7.5 years was 1.28, worse than the therapy in HR+) administered for 24 weeks to 61 patients
noninferiority cutoff.51 Several other trials, including Short- versus 12 weeks to 33 patients increased the pCR rate
HER52 and SOLD,53 tested 9 weeks against 1 year, also with twofold, and threefold in the estrogen receptor-positive
conflicting results. A meta-analysis that included more than subgroup, although the study did not meet its primary
11,000 patients from randomized trials found a DFS hazard endpoint of increasing the pCR rate to 45%.59 The PAMELA
ratio of 1.28 (95% CI, 1.09 to 1.36), supporting longer du- trial administered lapatinib plus trastuzumab alone to 150
ration, although cardiac events were also more than twofold women with HER2+ disease, with pCR in 31% (18% in HR+,
more frequent and the hazard ratio was only 1.15 in the HR+/ 33% in HR ). Taken together, these studies suggest that
HER2+ subset.54 At this time, it is not clear which patients a small but real proportion of HER2+ breast cancers are so
could be treated with shorter durations, although for those sensitive to anti-HER2 therapy that chemotherapy-free
with cardiac disease or poor treatment tolerance, shorter treatment may be adequate. However, we do not yet know if
durations seem to provide most of the benefit of 1 year of pCR obtained in these circumstances carries the same
trastuzumab. excellent prognosis as when obtained with more aggressive
measures.
BIOMARKERS, PATHOLOGIC COMPLETE RESPONSE, AND Some recent insight into this question comes from KRIS-
DISEASE-FREE SURVIVAL TINE, a randomized trial of T-DM1 plus pertuzumab (T-DM1 +
There have been substantial efforts to identify genomic P) versus docetaxel, carboplatin, trastuzumab, and pertuzu-
predictors of pCR to guide treatment de-escalation in the mab (TCHP). Of note, in the adjuvant setting, both arms
neoadjuvant setting. The CHER-LOB study, which dem- continued the same HER2-directed regimen (T-DM1 + P; HP)
onstrated increased pCR rates in patients receiving lapatinib to 1 year. T-DM1 + P underperformed relative to the free
with trastuzumab and chemotherapy, also found that pCR cytotoxic-containing TCHP regimen with pCR rates of 44%
rate was improved in PIK3CA wild-type tumors, whereas the versus 56% and higher progression rates during the neo-
presence of a PIK3CA mutation could be used to identify adjuvant phase.60 However, 3-year iDFS exceeded 95% in
patients who were less likely to benefit from dual anti-HER2 both arms among those patients with pCR.61 These data
inhibition.55 A pooled analysis of nearly 1,000 patients that support the upcoming National Cancer Institute trial EA 1181,
included five prospective trials of lapatinib and trastuzumab COMPASS-pCR,62 which will formally test a de-escalated
confirmed PIK3CA mutations to be associated with a re- approach with neoadjuvant docetaxel, trastuzumab, and
duced rate of pCR, most notably in the HR+ cohort; how- pertuzumab followed by completion of 1 year of trastuzumab
ever, no conclusions can be drawn regarding relapse or and pertuzumab without additional chemotherapy in those
survival.56 achieving pCR.

Additional insight into the biology of HER2+ breast cancer studies consistently reveal a direct association of acti-
comes from efforts at RNA in addition to DNA profiling. vated immune cells in the peritumoral environment, and
HER2+ breast cancer actually represents multiple biologic higher pCR as well as improved DFS in HER2+ breast
intrinsic subtypes, with the majority made up of the HER2- cancer. A meta-analysis performed investigated the re-
enriched, Luminal A, and Luminal B subtypes. These lationship between baseline tumor-infiltrating lympho-
subtypes have clinical implications. In CALGB 40601,25 cytes and pCR in patients with HER2+ breast cancer who
NOAH,63 NSABP B-41,64 and NeoALTTO,65 pCR rates re- were treated with neoadjuvant chemotherapy plus tras-
gardless of regimen were higher in the HER2-enriched tuzumab and lapatinib alone or in combination. In
subtype than either Luminal subtype, often more than a combined analysis of more than 1,200 patients in five
double the pCR rate. However, in CALGB 40601, the HER2- studies meeting the inclusion criteria, tumor-infiltrating
enriched subtype was also significantly associated with lymphocytes were significantly associated with pCR with
poorer DFS. This apparent discordance was because of the an odds ratio of 2.46 irrespective of the neoadjuvant
markedly worse outcome among HER2-enriched tumors treatment regimen.68 These investigations suggest that,
with residual disease.66 The PAMELA trial used intrinsic just as anatomic risk can be leveraged to de-escalate
subtype of HER2+ breast cancers to predict which patients therapy as was done in the APT trial, biologic risk by
will benefit from an all-biologic dual anti-HER2 therapy intrinsic subtype and activated immune cells, among
approach. In this study, patients with HER2+ stage I to IIIA other biomarkers, may provide mechanisms to de-escalate
breast cancer were administered lapatinib plus trastuzumab anatomically higher-risk tumors.
with endocrine therapy if their tumor was also HR+. Sixty- SUMMARY AND FUTURE DIRECTIONS
seven percent of women in the trial had tumors with the
Anti-HER2 therapy paired with chemotherapy is an essential
HER2-enriched subtype, and 41% had pCR compared with
component of (neo)adjuvant treatment of HER2+ breast
10% of patients with luminal A, luminal B, basal-like, or
cancer. Neoadjuvant systemic therapy is the preferred
normal-like intrinsic subtypes.67 This represents one mo-
approach to stage II or III HER2+ tumors (Table 1) to im-
lecular approach for de-escalation by proposing a mecha-
prove surgical options, monitor effectiveness of systemic
nism for the selection of patients upfront who are most likely
treatment, obtain prognostic information, and tailor therapy
to benefit from a particular therapeutic strategy—in this
on the basis of the extent of residual disease. Combination
case, the possibility of achieving pCR without chemother-
chemotherapy with an anthracycline, an alkylator, and
apy. However, whether this strategy will result in maintained
a taxane, together with trastuzumab (and pertuzumab if
excellent outcomes has not yet been determined.
node positive or otherwise high risk) or nonanthracycline
In addition to intrinsic subtype, both tumor-infiltrating lym- regimens using docetaxel and carboplatin as the chemo-
phocytes and RNA-based immune activation signature therapy component are both reasonable approaches for
TABLE 1. Suggested Escalation and De-escalation Strategies in HER2+ Early-Stage Breast Cancer
Pathologic
Clinical Stage Initial Treatment Recommendation Stage De-escalation Strategy Escalation Strategy
Stage I cT1N0 Surgery PT1aN0 No systemic therapy —
pT1b-cN0 12 weeks Tp + H to 1 year —
Polychemotherapy + shorter —
duration (6 months) H
Stage II cT2-3N0 cT0- Neoadjuvant Rx polychemotherapy + H pCR Complete 1-year H (+ P if given) —
2N1 (+ P if node+)
Residual — T-DM1 14 cycles
disease
Stage III cT3N1 Neoadjuvant Rx pCR Complete 1-year H + P —
cT4N(any) cT(any) polychemotherapy + H + P
Residual — T-DM1 14 cycles
N2-3
disease
Stage II-III surgery first Neoadjuvant Rx recommended; if surgery pT2-3N0 — Polychemotherapy + 1-year H
performed first, then adjuvant Rx
N+ ER+ — Polychemotherapy + 1-year H + P
or followed by N for 1 year
N+ ER — Polychemotherapy + H + P
Abbreviations: C, carboplatin; ER, estrogen receptor; H, trastuzumab; N, neratinib; P, pertuzumab; polychemotherapy, anthracycline/taxane or taxane based
with at least two cytotoxic drugs; pCR, pathologic complete response; Rx, treatment; Tc, docetaxel; T-DM1, trastuzumab emtansine; Tp, paclitaxel.

stage II or III HER2+ tumors in either the adjuvant or regimens, with or without pertuzumab, is recommended.
neoadjuvant settings. For anatomic stage 1 (T1N0), HER2+ Patients who achieve a pCR with anti–HER2-based therapy
tumors paclitaxel plus trastuzumab is a good de-escalated should receive adjuvant trastuzumab or trastuzumab plus
option. pertuzumab as administered in their neoadjuvant regimen
Several trials have addressed the option using less than based on current recommendations; however, this repre-
12 months of adjuvant trastuzumab-based therapy in early- sents another potential area of de-escalation.
stage, HER2+ breast cancer. These studies have shown Future escalation strategies in early HER2+ breast cancer
a narrow reduction in recurrence risk with 12 months of are likely to test a variety of augmented anti-HER2 drugs,
therapy compared with shorter (3- or 6-month) durations. such as tucatinib and trastuzumab deruxtecan, and novel
Thus, the standard of care is 1 year of trastuzumab-based drugs in this subtype, such as CDK4/6 inhibitors in tumors.
treatment as the preferred duration while acknowledging Key considerations will focus on identifying areas of remaining
that the benefit of 12 months over 6 months is likely to be
poor prognosis in HER2+ disease and unmet needs, such as
modest.
prevention of CNS metastases.
Extended anti-HER2 therapy with neratinib in the adjuvant
De-escalation efforts will determine the roles of tumor fea-
setting after 1 year of trastuzumab may further reduce the
tures and microenvironmental influences, as well as leverage
likelihood of tumor recurrence. Use of neratinib in cases
circulating biomarkers, to predict pCR and excellent DFS
of node-positive, HR+/HER2+ breast cancers, especially
with minimalist approaches. These will include integrative
those with four or more affected lymph nodes, treated with
trastuzumab-based therapy, is an option. Routine use of modeling algorithms to incorporate clinical and biologic tumor
neratinib in patients who were previously treated with features to tailor treatment.
pertuzumab-based therapy or after T-DM1 in the residual Three decades of research and advances in clinical practice
disease setting has not been studied. have transformed HER2+ breast cancer from the worst
In patients with residual invasive HER2+ breast cancer after prognosis to among the best, with relatively low relapse rates
neoadjuvant systemic therapy, adjuvant T-DM1 therapy and deaths from this disease. However, anti-HER2 regi-
substantially reduced the risk of recurrence with an absolute mens are also the most expensive, lengthy, and complex,
benefit of 8% to 12% in risk reduction. On the basis of this warranting the current focus on efforts to rationally direct
finding, use of T-DM1 for women with residual invasive intensified treatment of those who need it and de-intensified
cancer after neoadjuvant therapy with trastuzumab-based treatments for those who are likely to do well with less.
AFFILIATIONS AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

1
University of North Carolina at Chapel Hill, Chapel Hill, NC AND DATA AVAILABILITY STATEMENT
2
European Institute of Oncology, IRCCS, University of Milano, Milano, Disclosures provided by the authors and data availability statement (if
Italy applicable) are available with this article at DOI https://doi.org/10.1200/
EDBK_100023.
CORRESPONDING AUTHOR
Lisa A. Carey, MD, University of North Carolina at Chapel Hill, 170
Manning Dr., Campus Box 7305, Chapel Hill, NC 27599-7305; email:
[email protected].
REFERENCES
1. Cossetti RJ, Tyldesley SK, Speers CH, et al. Comparison of breast cancer recurrence and outcome patterns between patients treated from 1986 to 1992 and from
2004 to 2008. J Clin Oncol. 2015;33:65-73.
2. Lin NU, Winer EP. Brain metastases: the HER2 paradigm. Clin Cancer Res. 2007;13:1648-1655.
3. Slamon DJ, Godolphin W, Jones LA, et al. Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science. 1989;244:707-712.
4. Slamon DJ, Leyland-Jones B, Shak S, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses
HER2. N Engl J Med. 2001;344:783-792.
5. Wolff AC, Hammond MEH, Allison KH, et al. Human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of
American Pathologists clinical practice guideline focused update. J Clin Oncol. 2018;36:2105-2122.
6. Swain SM, Baselga J, Kim SB, et al; CLEOPATRA Study Group. Pertuzumab, trastuzumab, and docetaxel in HER2-positive metastatic breast cancer. N Engl
J Med. 2015;372:724-734.

7. Geyer CE, Forster J, Lindquist D, et al. Lapatinib plus capecitabine for HER2-positive advanced breast cancer. N Engl J Med. 2006;355:2733-2743.
8. Blackwell KL, Burstein HJ, Storniolo AM, et al. Randomized study of lapatinib alone or in combination with trastuzumab in women with ErbB2-positive,
trastuzumab-refractory metastatic breast cancer. J Clin Oncol. 2010;28:1124-1130.
9. Johnston S, Pippen J Jr., Pivot X, et al. Lapatinib combined with letrozole versus letrozole and placebo as first-line therapy for postmenopausal hormone receptor-
positive metastatic breast cancer. J Clin Oncol. 2009;27:5538-5546.
10. Kaufman B, Mackey JR, Clemens MR, et al. Trastuzumab plus anastrozole versus anastrozole alone for the treatment of postmenopausal women with human
epidermal growth factor receptor 2-positive, hormone receptor-positive metastatic breast cancer: results from the randomized phase III TAnDEM study. J Clin
Oncol. 2009;27:5529-5537.
11. Saura C, Oliveira M, Feng Y, et al. Neratinib + capecitabine versus lapatinib + capecitabine in patients with HER2+ metastatic breast cancer previously treated
with 2 HER2-directed regimens: findings from the multinational, randomized, phase III NALA trial. J Clin Oncol. 2019;37:15s (suppl; abstr 1002).
12. Awada A, Colomer R, Inoue K, et al. Neratinib plus paclitaxel vs trastuzumab plus paclitaxel in previously untreated metastatic ERBB2-positive breast cancer: the
NEfERT-T randomized clinical trial. JAMA Oncol. 2016;2:1557-1564.
13. Murthy RK, Loi S, Okines A, et al. Tucatinib, trastuzumab, and capecitabine for HER2-positive metastatic breast cancer. N Engl J Med. 2020;382:597-609.
14. Verma S, Miles D, Gianni L, et al; EMILIA Study Group. Trastuzumab emtansine for HER2-positive advanced breast cancer. N Engl J Med. 2012;367:1783-1791.
15. Krop IE, Kim SB, Martin AG, et al. Trastuzumab emtansine versus treatment of physician’s choice in patients with previously treated HER2-positive metastatic
breast cancer (TH3RESA): final overall survival results from a randomised open-label phase 3 trial. Lancet Oncol. 2017;18:743-754.
16. Modi S, Saura C, Yamashita T, et al; DESTINY-Breast01 Investigators. Trastuzumab deruxtecan in previously treated HER2-positive breast cancer. N Engl J Med.
2020;382:610-621.
17. Tamura K, Tsurutani J, Takahashi S, et al. Trastuzumab deruxtecan (DS-8201a) in patients with advanced HER2-positive breast cancer previously treated with
trastuzumab emtansine: a dose-expansion, phase 1 study. Lancet Oncol. 2019;20:816-826.
18. Krop I, Saura C, Yamashita T, et al. Trastuzumab deruxtecan (T-DXd; DS-8201a) in subjects with HER2-positive metastatic breast cancer previously treated with
T-DM1: a phase 2, multicenter, open-label study (DESTINY-Breast01). Presented at: San Antonio Breast Cancer Symposium, San Antonio, TX; 2019. Abstract
GS1-03.
19. Banergi U, van Herpen CML, Saura C, et al. Trastuzumab duocarmazine in locally advanced and metastatic solid tumours and HER2-expressing breast cancer:
a phase 1 dose-escalation and dose-expansion study. Lancet Oncol. 2019;20:1124-1135.
20. Romond EH, Perez EA, Bryant J, et al. Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med. 2005;
353:1673-1684.
21. Perez EA, Romond EH, Suman VJ, et al. Trastuzumab plus adjuvant chemotherapy for human epidermal growth factor receptor 2-positive breast cancer: planned
joint analysis of overall survival from NSABP B-31 and NCCTG N9831. J Clin Oncol. 2014;32:3744-3752.
22. Piccart-Gebhart MJ, Procter M, Leyland-Jones B, et al; Herceptin Adjuvant (HERA) Trial Study Team. Trastuzumab after adjuvant chemotherapy in HER2-
positive breast cancer. N Engl J Med. 2005;353:1659-1672.
23. Slamon D, Eiermann W, Robert N, et al; Breast Cancer International Research Group. Adjuvant trastuzumab in HER2-positive breast cancer. N Engl J Med. 2011;
365:1273-1283.
24. de Azambuja E, Holmes AP, Piccart-Gebhart M, et al. Lapatinib with trastuzumab for HER2-positive early breast cancer (NeoALTTO): survival outcomes of
a randomised, open-label, multicentre, phase 3 trial and their association with pathological complete response. Lancet Oncol. 2014;15:1137-1146.
25. Carey LA, Berry DA, Cirrincione CT, et al. Molecular heterogeneity and response to neoadjuvant human epidermal growth factor receptor 2 targeting in CALGB
40601, a randomized phase III trial of paclitaxel plus trastuzumab with or without lapatinib. J Clin Oncol. 2016;34:542-549.
26. Robidoux A, Tang G, Rastogi P, et al. Lapatinib as a component of neoadjuvant therapy for HER2-positive operable breast cancer (NSABP protocol B-41): an
open-label, randomised phase 3 trial. Lancet Oncol. 2013;14:1183-1192.
27. Gianni L, Pienkowski T, Im YH, et al. Efficacy and safety of neoadjuvant pertuzumab and trastuzumab in women with locally advanced, inflammatory, or early
HER2-positive breast cancer (NeoSphere): a randomised multicentre, open-label, phase 2 trial. Lancet Oncol. 2012;13:25-32.
28. Gianni L, Pienkowski T, Im YH, et al. 5-year analysis of neoadjuvant pertuzumab and trastuzumab in patients with locally advanced, inflammatory, or early-stage
HER2-positive breast cancer (NeoSphere): a multicentre, open-label, phase 2 randomised trial. Lancet Oncol. 2016;17:791-800.
29. Schneeweiss A, Chia S, Hickish T, et al. Pertuzumab plus trastuzumab in combination with standard neoadjuvant anthracycline-containing and anthracycline-
free chemotherapy regimens in patients with HER2-positive early breast cancer: a randomized phase II cardiac safety study (TRYPHAENA). Ann Oncol. 2013;
24:2278-2284.
30. von Minckwitz G, Procter M, de Azambuja E, et al; APHINITY Steering Committee and Investigators. Adjuvant pertuzumab and trastuzumab in early HER2-
31. Piccart M, Procter M, Fumagalli D, et al. Interim overall survival analysis of APHINITY (BIG 4-11): a randomized multicenter, double-blind, placebo-controlled trial
comparing chemotherapy plus trastuzumab plus pertuzumab versus chemotherapy plus trastuzumab plus placebo as adjuvant therapy in patients with operable
HER2-positive early breast cancer. Presented at: San Antonio Breast Cancer Symposium, San Antonio, TX; 2019. Abstract GS1-04.
32. Piccart-Gebhart M, Holmes E, Baselga J, et al. Adjuvant lapatinib and trastuzumab for early human epidermal growth factor receptor 2-positive breast cancer:
results from the randomized phase III adjuvant lapatinib and/or trastuzumab treatment optimization trial. J Clin Oncol. 2016;34:1034-1042.
33. Chan A, Delaloge S, Holmes FA, et al; ExteNET Study Group. Neratinib after trastuzumab-based adjuvant therapy in patients with HER2-positive breast cancer
(ExteNET): a multicentre, randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol. 2016;17:367-377.

34. Martin M, Holmes FA, Ejlertsen B, et al; ExteNET Study Group. Neratinib after trastuzumab-based adjuvant therapy in HER2-positive breast cancer (ExteNET): 5-
year analysis of a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol. 2017;18:1688-1700.
35. von Minckwitz G, Huang CS, Mano MS, et al; KATHERINE Investigators. Trastuzumab emtansine for residual invasive HER2-positive breast cancer. N Engl J Med.
2019;380:617-628.
36. Chia SKL, Martin M, Holmes FA, et al. PIK3CA alterations and benefit with neratinib: analysis from the randomized, double-blind, placebo-controlled, phase III
ExteNET trial. Breast Cancer Res. 2019;21:39.
37. Giordano SH, Niu J, Chavez-MacGregor M, et al. Estimating regimen-specific costs of chemotherapy for breast cancer: observational cohort study. Cancer. 2016;
122:3447-3455.
38. Kale HP, Carroll NV. Self-reported financial burden of cancer care and its effect on physical and mental health-related quality of life among US cancer survivors.
Cancer. 2016;122:283-289.
39. Ramsey SD, Bansal A, Fedorenko CR, et al. Financial insolvency as a risk factor for early mortality among patients with cancer. J Clin Oncol. 2016;34:980-986.
40. Chino F, Peppercorn JM, Rushing C, et al. Out-of-pocket costs, financial distress, and underinsurance in cancer care. JAMA Oncol. 2017;3:1582-1584.
41. Greenup R. The financial costs and burden of breast cancer care. Session presented at: San Antonio Breast Cancer Symposium, San Antonio, TX, December 10,
2019.
42. Zafar SY, Peppercorn JM, Schrag D, et al. The financial toxicity of cancer treatment: a pilot study assessing out-of-pocket expenses and the insured cancer
patient’s experience. Oncologist. 2013;18:381-390.
43. Singleterry J. The Costs of Cancer: Addressing Patient Costs. American Cancer Society Cancer Action Network. https://www.fightcancer. org/sites/default/files/
Costs%20of%20Cancer%20-%20Final%20Web.pdf. Accessed March 22, 2020.
44. Yabroff KR, Dowling EC, Guy GP Jr., et al. Financial hardship associated with cancer in the United States: findings from a population-based sample of adult
cancer survivors. J Clin Oncol. 2016;34:259-267.
45. Tolaney SM, Barry WT, Dang CT, et al. Adjuvant paclitaxel and trastuzumab for node-negative, HER2-positive breast cancer. N Engl J Med. 2015;372:134-141.
46. Tolaney SM, Guo H, Pernas S, et al. Seven-year follow-up analysis of adjuvant paclitaxel and trastuzumab trial for node-negative, human epidermal growth factor
receptor 2-positive breast cancer. J Clin Oncol. 2019;37:1868-1875.
47. Tolaney SM, Trippa L, Barry WT, et al. TBCRC 033: a randomized phase II study of adjuvant trastuzumab emtansine (T-DM1) vs paclitaxel (T) in combination with
trastuzumab (H) for stage I HER2-positive breast cancer (BC) (ATEMPT). Presented at: San Antonio Breast Cancer Symposium, San Antonio, TX; 2019. Abstract
GS1-05.
48. Golshan M, Cirrincione CT, Sikov WM, et al. Impact of neoadjuvant therapy on eligibility for and frequency of breast conservation in stage II-III HER2-positive
breast cancer: surgical results of CALGB 40601 (Alliance). Breast Cancer Res Treat. 2016;160:297-304.
49. Pilewskie M, Morrow M. Axillary nodal management following neoadjuvant chemotherapy: a review. JAMA Oncol. 2017;3:549-555.
50. Earl HM, Hiller L, Vallier AL, et al; PERSEPHONE Steering Committee and Trial Investigators. 6 versus 12 months of adjuvant trastuzumab for HER2-positive early
breast cancer (PERSEPHONE): 4-year disease-free survival results of a randomised phase 3 non-inferiority trial. Lancet. 2019;393:2599-2612.
51. Pivot X, Romieu G, Debled M, et al; PHARE trial investigators. 6 months versus 12 months of adjuvant trastuzumab in early breast cancer (PHARE): final analysis
of a multicentre, open-label, phase 3 randomised trial. Lancet. 2019;393:2591-2598.
52. Conte P, Frassoldati A, Bisagni G, et al; Reader study level-I and level-II Groups. Nine weeks versus 1 year adjuvant trastuzumab in combination with
chemotherapy: final results of the phase III randomized Short-HER study. Ann Oncol. 2018;29:2328-2333.
53. Joensuu H, Fraser J, Wildiers H, et al. Effect of adjuvant trastuzumab for a duration of 9 weeks vs 1 year with concomitant chemotherapy for early human
epidermal growth factor receptor 2-positive breast cancer: the SOLD randomized clinical trial. JAMA Oncol. 2018;4:1199-1206.
54. Chen L, Zhou W, Hu X, et al. Short-duration versus 1-year adjuvant trastuzumab in early HER2 positive breast cancer: a meta-analysis of randomized controlled
trials. Cancer Treat Rev. 2019;75:12-19.
55. Guarneri V, Dieci MV, Frassoldati A, et al. Prospective biomarker analysis of the randomized CHER-LOB study evaluating the dual anti-HER2 treatment with
trastuzumab and lapatinib plus chemotherapy as neoadjuvant therapy for HER2-positive breast cancer. Oncologist. 2015;20:1001-1010.
56. Loibl S, Majewski I, Guarneri V, et al. PIK3CA mutations are associated with reduced pathological complete response rates in primary HER2-positive breast
cancer: pooled analysis of 967 patients from five prospective trials investigating lapatinib and trastuzumab. Ann Oncol. 2016;27:1519-1525.
57. Loi S, Dafni U, Karlis D, et al. Effects of estrogen receptor and human epidermal growth factor receptor-2 levels on the efficacy of trastuzumab: a secondary
analysis of the HERA trial. JAMA Oncol. 2016;2:1040-1047.
58. Rimawi MF, Mayer IA, Forero A, et al. Multicenter phase II study of neoadjuvant lapatinib and trastuzumab with hormonal therapy and without chemotherapy in
patients with human epidermal growth factor receptor 2-overexpressing breast cancer: TBCRC 006. J Clin Oncol. 2013;31:1726-1731.
59. Rimawi MF, Niravath PA, Wang T. TBCRC023: a randomized multicenter phase II neoadjuvant trial of lapatinib plus trastuzumab, with endcorine therapy and
without chemotherapy, for 12 vs. 24 weeks in patients with HER2 overexpressing breast cancer. Presented at: San Antonio Breast Cancer Symposium, San
Antonio, TX; 2014. Abstract S6-02.
60. Hurvitz SA, Martin M, Symmans WF, et al. Neoadjuvant trastuzumab, pertuzumab, and chemotherapy versus trastuzumab emtansine plus pertuzumab in
patients with HER2-positive breast cancer (KRISTINE): a randomised, open-label, multicentre, phase 3 trial. Lancet Oncol. 2018;19:115-126.
61. Hurvitz SA, Martin M, Jung KH, et al. Neoadjuvant trastuzumab emtansine and pertuzumab in human epidermal growth factor receptor 2-positive breast cancer:
three-year outcomes from the phase III KRISTINE study. J Clin Oncol. 2019;37:2206-2216.

62. NCT04266249. CompassHER2-pCR: Decreasing Chemotherapy for Breast Cancer Patients After Pre-surgery Chemo and Targeted Therapy. https://clinicaltrials.
gov/ct2/show/NCT04266249. Accessed March 22, 2020.
63. Prat A, Bianchini G, Thomas M, et al. Research-based PAM50 subtype predictor identifies higher responses and improved survival outcomes in HER2-positive
breast cancer in the NOAH study. Clin Cancer Res. 2014;20:511-521.
64. Swain SM, Tang G, Lucas PC, et al. Pathologic complete response and outcomes by intrinsic subtypes in NSABP B-41, a randomized neoadjuvant trial of
chemotherapy with trastuzumab, lapatinib, or the combination. Breast Cancer Res Treat. 2019;178:389-399.
65. Fumagalli D, Venet D, Ignatiadis M, et al. RNA sequencing to predict response to neoadjuvant anti-HER2 therapy: a secondary analysis of the NeoALTTO
randomized clinical trial. JAMA Oncol. 2017;3:227-234.
66. Fernandez-Martinez A, Tanioka M, Fan C, et al. Predictive and prognostic value of B-cell gene-expression signatures and B-cell receptor (BCR) repertoire in
HER2+ breast cancer: a correlative analysis of the CALGB 40601 clinical trial (Alliance). Presented at: European Society for Medical Oncology Annual
Conference, Barcelona, Spain; 2019. Abstract 174O.
67. Llombart-Cussac A, Cortés J, Paré L, et al. HER2-enriched subtype as a predictor of pathological complete response following trastuzumab and lapatinib without
chemotherapy in early-stage HER2-positive breast cancer (PAMELA): an open-label, single-group, multicentre, phase 2 trial. Lancet Oncol. 2017;18:545-554.
68. Solinas C, Ceppi M, Lambertini M, et al. Tumor-infiltrating lymphocytes in patients with HER2-positive breast cancer treated with neoadjuvant chemotherapy plus
trastuzumab, lapatinib or their combination: a meta-analysis of randomized controlled trials. Cancer Treat Rev. 2017;57:8-15.

PALLIATIVE CARE
invited article Palliative Care Skills and New Resources for

Oncology Practices: Meeting the Palliative Care
Needs of Patients With Cancer and Their Families
Anthony Back, MD1; Tara Friedman, MD2; and Janet Abrahm, MD3
overview
In its 2017 guideline, ASCO challenged members to integrate palliative care into their standard oncology
practices for all patients, throughout their cancer trajectory. However, partnering with palliative care experts
alone will not be enough to achieve that goal; there are too few experts now, and there will not be enough in the
future to meet the needs of patients with cancer and their families. Other strategies are required. Oncologists
can develop new communication skills that were not included in their fellowship curricula, skills that integrate
into their visits the subjects that palliative care clinicians discuss routinely with patients referred to them. In
this review, Dr. Back offers three questions matched to communication skills that can help oncologists explore
key areas: (1) What is happening? (2) How do you (and I) feel? and (3) What is important? and discusses the
“REMAP” strategy for making urgent medical decisions. Dr. Friedman reviews the impact of community-based
palliative care resources and telehealth on care quality, patient centeredness, and reducing costs. Community-
based palliative care services and telehealth are available to patients and families at home, during active
treatment. Dr. Abrahm reviews how patient-reported outcomes (PROs) completed at home can enhance
patients’ symptom control, quality of life, and toleration of treatment and decrease unplanned emergency
visits by alerting clinicians to patients’ severe symptoms, making appropriate referrals, or suggesting patients
contact their oncology team. She also provides an update on using PROs and natural language processing with
clinical decision support to create sophisticated palliative care assessments and treatment options in the
electronic health record during patients’ office visits.
INTRODUCTION patient-centered communication in cancer care: (1)

In its 2017 guideline, ASCO challenged members to fostering the clinician-patient relationship; (2) exchanging
integrate palliative care into their standard oncology information; (3) responding to emotions; (4) managing
practices for all their patients, throughout their cancer uncertainty; (5) making decisions; and (6) enabling
trajectory. Partnering with palliative care experts alone, patient self-management.1 But another, perhaps more
however, will not be enough to achieve that goal. In this clinician-friendly, way to approach communication is to
article, we offer other strategies. First, Dr. Back details pose a set of questions that point to what clinicians must
new palliative care communication tools that were not accomplish in every conversation. There are three easy
included in oncology fellowship curricula. These tools questions that can enable you to keep your commu-
can improve patient-centered communication and nication on track: (1) What is happening? (2) How do
enable oncologists to explore goals, values, and im- you (and I) feel? and (3) What is important?
portant aspects of their patient’s quality of life. Next, Answering the “What is happening?” question prompts
Author affiliations
Dr. Friedman reviews the impact of community-based us to prepare for a visit. The clinician’s task is to look at
and support
information (if
palliative care resources and telehealth on care quality, all of the data at hand (often a lot), curate the most
applicable) appear patient-centeredness, and reducing costs. Lastly,
important bits to discuss, and structure how to present
at the end of this Dr. Abrahm envisions integrating PROs with clinical
article.
it. Others consider this “information provision”—giving
decision support to create sophisticated assessments
Accepted on
serious news is one version of it2—but using a patient’s
and treatment options in the EHR during patients’ office
February 26, 2020 view by asking this question is helpful because it prompts
visits.
and published at us to filter and sort the information that we think is im-
ascopubs.org on COMMUNICATION SKILLS THAT INTEGRATE portant to get across into manageable and understand-
March 26, 2020: PALLIATIVE CARE
able units. The “What is happening?” question also
10.1200/EDBK_ In a 2015 monograph for the National Cancer Institute, prompts us to explain the information in a way that
100022 Epstein and Street identified a six-part framework for is understandable to that particular patient, taking into

Palliative Care Skills and New Palliative Care Resources for Oncologists
still involved.3 Responding to emotion is not just a nice thing

to do, it enables your patient to process their own emotions in
the moment and to absorb information, interact more with the
• Ask the patient: (1) What is happening? (2) How clinician, and make better decisions. The part of this question
do you (and I) feel? and (3) What is important?
that asks clinicians to ask themselves how they are feeling
and then explain the information in a way un-
is simply a reminder that we all have our own reactions to
derstandable to that particular patient—taking
into account their education, health literacy, patients and that accepting our own reactions can free us up
interest in medical detail, body language, tone to be more fully present with our patients.4
of voice, and facial expressions, which enable The “What is important?” question shapes how we talk
you to empathize, react, and respond about the care that we recommend and what we prescribe,
appropriately.
and it invites patient buy-in. Asking “What is important?”
• When time is shorter than they hoped, use requires that clinicians uncover values, to understand what
REMAP. REMAP stands for (1) Reframe: “We’re really matters to a patient. This is often called “eliciting
in a different place now”; (2) respond to values,” but the “What is important?” question prompts us
Emotion; (3) Map out their hopes and worries;
not to make assumptions or jump to conclusions; the values
(4) Align with the patient and family’s hopes;
that a patient consider the most important are often held
and (5) taking those into account, make a Plan
for next steps, including a resuscitation rec- closely and only spoken about with someone who feels
ommendation (full code vs. Do Not Resuscitate safe. The “What is important?” question also prompts us, in
or Intubate (DNR/I). another direction, to draw our patient’s attention to the
medical decisions that they must make or be prepared to
• Identify and collaborate with community palli-
ative care and telehealth providers in your area. make. As the clinician considers the medical realities the
Consult GetPalliativeCare.org (https:// patient is facing, along with what was learned about how the
getpalliativecare.org/howtoget/find-a-palliative- patient is feeling and what is important to that patient,
care-team) for a comprehensive list of providers clinicians can easily craft an informed plan of care, share it
in your area. with that patient, and ask for feedback.
• Explore PRO options for your practice site, in- Each of these three questions is matched to a set of com-
cluding tablets, mobile applications, and munication skills that enable a clinician in cancer care to
desktops communicating directly with your seamlessly integrate considerations of quality of life into
EHR.
everyday conversations (Table 1). A useful framework when
• Challenge your organization to identify available a medical decision is pending or even urgent is REMAP
technology through which your practice can (Table 2).5
incorporate a system of automatic alerts to
clinicians when the PRO indicates a serious For times when a referral to a palliative care specialist is
unaddressed symptom. needed, it is important to preemptively counter assumptions
that palliative care is about dying. Make the reason for
referral transparent and frame it in words the patient has
used, such as, “I think we could do a better job with your
account their education, their health literacy, and their in- pain if I bring in one of my colleagues who specializes in
terest in medical detail. this,” or “Palliative care is meant to improve quality of life
at any stage of cancer. They have helped a lot of my
The “How do you (your patient) and I (you, the clinician)
patients.”
feel?” question reminds us to listen and empathize. It is
easier to listen generously if you have already decided how Communication that enables patients to feel that their quality
to deal with the “What is happening?” question; you will of life can be as important as any other aspect of care is at
have more bandwidth and attention because you will not be the heart of high-quality cancer care. It takes practice and
trying to filter and sort information as you listen. But listening feedback,6,7 and there are more options than ever for cli-
itself is not just about the words; it is also about the body nicians to learn more (Table 3). A companion piece to this
language, the tone of voice, and the facial expressions. All of introduction was just published in the Journal of Clinical
those data enable you to empathize, to feel the patient’s Oncology.8
emotions (at least a part of their emotions) so that you can
react and respond appropriately. Remember that emotion COMMUNITY-BASED PALLIATIVE CARE AND TELEHEALTH
happens faster than cognition in any of us, and, once started, Since 1982, palliative care in the United States was largely
emotion cannot be turned off; even when we suppress our defined by the Medicare Hospice Benefit. The end of the
conscious experience of emotion, our minds and bodies are 20th century saw the development of nonhospice palliative

Back, Friedman, Abrahm
TABLE 1. Three Key Questions and Matching Communication Skills

Key Question Communication Skill What You Say
What’s happening? Put together an agenda for the visit “There are three things I want to cover today…is there anything you want to
add to the agenda?”
Establish what the patient already knows “Tell me what you’ve heard from Dr. Google.”
Use language your patient will understand when “Your scan shows that the cancer is getting worse. Could we talk about
giving serious news what that means?”
Ask explicitly about understanding “Have I been clear? Is there anything I should explain better?”
Ask if prognostic information is desired “Would you like to talk about what to expect in the future?”
How do you (and I) Noticing emotions in patients “Could you say what reactions are going through your mind?”
feel?
Acknowledging emotion openly “It sounds like you are frustrated [or disappointed, or any other emotion].”
Give patients time to collect their thoughts and speak “Take your time.”
Noticing one’s own emotions [Internally] “How do I feel right now? Can I bring whatever presence I can
muster to this moment?”
What’s important? Elicit values that drive medical decisions “What’s important to your quality of life now?”
Ask for stories that illustrate values “Tell me about what a ‘good day’ is like for you.”
Anticipate upcoming medical decisions “I’d like to make sure we are prepared for a decision that is likely to come
up.”
Emphasize that you and your team will be there for the “We will be here for you at every step.”
long haul
care, largely in the hospital setting. Most recently, progressive treatment can be identified and addressed. As a guest
interest in and demand for palliative care have fueled growth in the patient’s home over a period of months or years,
in local, regional, and multistate organizations that provide providers identify previously unrecognized symptoms, barriers
community-based palliative care (CBPC) beyond the con- to treatment compliance, financial stressors, safety concerns,
straints of a hospital or hospice program. Consistent with and caregiver burden. Palliative care providers engage the
ASCO recommendations, these programs serve those with patient, family, and/or health care proxy to explore the patient’s
advanced disease and/or those approaching the last year goals, values, and beliefs. Interdisciplinary teams collaborate
of life, when the physical, emotional, psychosocial, spiritual, with the patient’s oncology team, balancing quality of life with
and financial stress of advancing disease creates the wid- evidence-based disease-directed treatments that align with
est gaps in care. It is during this time that health care costs the patient’s evolving preferences.
rise exponentially, because patients are often caught in an Community-Based Palliative Care Models
escalating cycle of hospitalizations, emergency depart-
ment (ED) visits, appointments, testing, and complex The Medicare Hospice Benefit started almost 40 years ago,
medication regimens. Thus, more than 50% of all health care but, over the last 20 years, nonhospice palliative care de-
spending goes to caring for the sickest 5% of Americans, livery in the home setting has grown substantially. In-
largely in the last year of life.9 terdisciplinary teams (e.g., physicians, nurse practitioners,
nurses, and social workers), supported by centralized care
Community-Based Palliative Care coordinators and clinical triage, visit patients in their homes.
Community-based programs broaden the reach of palliative CBPC practices include comanagement with the primary
care, meeting patients and families in their homes, where care physician, oncologist, and other specialists. Clinicians
gaps in care and the stresses of cancer diagnosis and share diagnostic information, therapeutic plans, impressions
TABLE 2. REMAP: Five Steps for Goals-of-Care Conversations

1. Reframe the situation “We’re in a different place now. What we’ve been doing is no longer working.”
2. Expect emotion and acknowledge it “I can see that this is not what you were hoping for.”
3. Map out important values “Tell me what is making your life good and meaningful now.”
4. Align with your patient and their family “I hear that you value your relationships, and I want to do everything possible to support you in that.”
5. Plan treatments to match values “Based on what you’ve told me, here are some recommendations…. What do you think?”

TABLE 3. Where to Learn Communication Skills

Center to Advance Palliative If your institution is a Center to Advance Palliative Care member, you have access to a series of short communication courses
Care that can be done online, even on mobile devices (https://www.capc.org).
VitalTalk Many free videos that demonstrate communication skills, an online continuing medical education course on Delivering
Serious News, and in-person Mastering Tough Conversations workshops (https://www.vitaltalk.org)
Respecting Choices An approach to advance care planning that powered city-wide involvement in LaCrosse, Wisconsin (https://www.
respectingchoices.org)
Serious Illness Care Program A structured conversation guide and system implementation tools especially aimed at early upstream conversations (https://
www.ariadnelabs.org/areas-of-work/serious-illness-care/)
about response to treatment, and prognosis and coordinate of telehealth-based palliative care interventions for patients
care when there are new symptoms, changes in functional with cancer in rural areas.18
or nutritional status, or substantial changes in the patient’s
goals of care.
Collaborating With Community-Based Providers
Although hospices and home health agencies can offer Coordination and collaboration with the patient’s oncology
nonhospice palliative care visits, restrictions in licensure and team are key. Oncologists identify patients they believe
funding create substantial barriers to large-scale sustainable would benefit, tell the patients about the referral, and make
interdisciplinary palliative care offerings. Early CBPC from a referral to their local program(s). Informing the patient of
Kaiser Permanente and Sutter Health provided evidence the referral enhances enrollment rates. Sharing pertinent
for CBPC’s benefits.10 More recently, other independent diagnostic and therapeutic history, communication prefer-
organizations have partnered directly with payers to pro- ences, and patient-specific concerns fosters collaboration.
vide an extra layer of support to their sickest members.11-13 Leveraging the unique expertise and perspective of each
CBPC has demonstrated improved performance in care team facilitates goal alignment and ensures that the pa-
quality, patient centeredness, and cost reduction. Patients tient’s individual needs are met most effectively.
enrolled in CBPC programs demonstrate significantly higher
rates of actionable advance directive completion,14 increased How To Find a CBPC Program Near You
likelihood of dying in their place of choice,14 improved
Table 4 compares several CBPC and telehealth palliative
access to hospice care, and longer hospice length of
care programs. Consult GetPalliativeCare.org (https://
stay.11,14 Patients, caregivers, and clinicians report high
getpalliativecare.org/howtoget/find-a-palliative-care-team/)
degrees of satisfaction with care.15 Patients have fewer
for a comprehensive list of providers in your area.
unnecessary ED visits and hospitalizations and overall
lower costs of care.9,16
OPTIMAL MANAGEMENT OF PATIENT SYMPTOMS:
Telehealth INTEGRATING PROs, NATURAL LANGUAGE PROCESSING,
Although some programs use telephonic and/or video- AND CLINICAL DECISION SERVICES TO PROVIDE
assisted visits to supplement in-home visits, others are RECOMMENDATIONS TO CLINICIANS VIA THE EHR
meeting the growing need for palliative care, particularly The 2017 ASCO guideline recommended integrating pal-
in rural areas, with largely or entirely remote interventions. liative care into standard oncology care for patients with
Using telephonic or video-assisted visits, palliative care advanced cancer.21 Operationalizing this recommendation
clinicians can assess health literacy, encourage self-efficacy, is not simple. The number of board-certified specialists in
coordinate community resources, foster communication the United States remains inadequate to fill the current,
across the health care team, and facilitate advance care let alone the projected, need.22 As of 2017, estimates in-
planning. Palliative care social workers assess for psy- dicated one palliative care provider for 1,200 people with
chosocial stressors, social determinants of health, care- serious illness, whereas there was one oncologist for every
giver stress, and health care use. Physical symptoms can 141 people with cancer.23 Increasing delivery of primary
be assessed remotely by palliative care nurses. Some pro- palliative care (including basic symptom management and
grams use patient-reported vital sign monitoring through advance care planning) would likely limit the need for
digital platforms. In one study, patients reported advantages specialty intervention.24 One strategy to increase primary
in telemedicine, including access to clinicians, rapid re- palliative care for patients with cancer is the electronic
sponse time, and improvements in the efficiency and quality health record (EHR) using clinical decision support (CDS)
of care.17 Evidence Project ENABLE demonstrated the value and PROs.

TABLE 4. Exemplar Community-Based and Telehealth Palliative Care Programs

Organization Description Key Outcomes
Kaiser Permanente In- Team provides comprehensive assessment and management Increased patient satisfaction, more likely to die in place of
Home Palliative Care of physical, psychological, social, and spiritual needs. choice, less likely to use the ED or be hospitalized. Substantial
Emphasized pain and symptom control, advance care cost savings.14,15,19
planning, and care coordination across settings. Includes 24-
hour urgent visits.
Home Connections Cheektowaga Hospice, in upstate NY, offers their Home Three times more likely to enroll in hospice, substantially longer
Connections program to adult patients with advanced illness hospice length of stay, higher actionable advance directive
who are prognostically upstream of hospice. An completion, better symptom management, higher likelihood
interdisciplinary team provides symptom management, of dying at home, and higher levels of patient, caregiver, and
patient education, supportive advance care planning, clinician satisfaction.10,11,16
community support services, respite, and on-call clinical
support.
Four Seasons Hospice Four Seasons provides CBPC over a diverse 14-county area in MIPS measures. Improved patient satisfaction and increased
& Palliative Care western North Carolina. Videoconferencing supplements in- hospice use.17,20
person visits. Uses remote vital sign monitoring. Provides an
extra layer of support, including symptom management,
relief of suffering, whole person care, and support for family
members. Facilitates advance care planning and advance
directive completion.
Aspire Health In-Home Aspire Health provides interdisciplinary palliative care across More advance care planning discussions and advance directive
Program 26 states and Washington, DC. Provides care of the seriously completion rate, more likely to use hospice and for longer
ill focused on symptom management, patient-family periods of time, reduction in unnecessary ED visits and
communication, advance care planning, care coordination hospitalizations (Kate Lach, Director of Social Work, Aspire
with patient’s other medical providers, and support services. Health, personal communication, January 2020).
Includes 24/7 telephonic clinical support.
Aspire Telehealth Designed for those with advanced cancer, the telephonic
program offers social work-lead palliative care support. The
team assesses and addresses health literacy, social
determinants of health, health care use, and resource needs.
Facilitates advance care planning. The intervention offers
nursing and physician support as needed for symptom
management and collaboration with oncology teams.
ProHEALTH Care Housed within an ACO, ProHEALTH’s interdisciplinary team Substantial cost savings, fewer hospitalizations, increased use of
Support offers in-home and telepalliative care to care for those with hospice and for longer periods of time13
serious illness. Team focuses on home complex care,
including management of chronic diseases, attention to
practical aspects of daily living, and caregiver support.
Landmark Health Landmark Health offers palliative care for existing patients in Improved POLST completion rates, improved use of hospice,
their longitudinal complex care offering. Intervention less likely to use the ED or be hospitalized (Dr. Scott Mancuso,
includes some in-person visits, along with telephonic internal Chief Clinical Officer, Landmark Health, personal
consultation. For a small group of patients with advancing communication, January 2020).
conditions, palliative care is the primary intervention, offering
a focus on aggressive symptom control and expert
communication.
Abbreviations: ACO, Accountable Care Organization; ED, emergency department; MIPS, Merit-based Incentive Payment System; PC, palliative care; POLST:
Physician Orders for Life Sustaining Treatment.
Automating Processing of PROs Using Clincal Decision new infection symptoms, but they are very reluctant to call
Support in the EHR about pain.25 Patients who regularly report PROs that are
Pain relief was one of the first PROs to be collected regularly used by clinicians’ EHRs to alert members of their oncology
when the Joint Commission for the Accreditation of Hospitals team to their uncontrolled symptoms have enhanced patient-
clinician communication, clinician awareness of symptoms,
mandated that patients be asked “What is your pain level? Is
symptom management, and quality of life and patient sat-
this level of relief satisfactory to you?” When patients are at isfaction, and they tolerate treatment better, have fewer un-
home, their suffering may go undetected except by their loved planned admissions or ED visits for uncontrolled symptoms,26
ones. Patients call oncology providers about fevers or other and even show improved overall survival.18,27,28

Patient use cell phones, tablets, or computers linked to their can supplement PROs. They are especially important for
EHR to report their levels of distress directly to their clini- those patients who find it difficult to quantify symptoms
cians.26 Computer-driven algorithms monitor symptom se- electronically or in person.31
verity and send alerts to a triage nurse via the EHR when
pain is above a prespecified level (e.g., 7, or severe) or the Clinical Decision Support
trend of the symptom is worsening, and the nurse contacts PROs are an important first step in identifying patients in
the patient. The nurse assesses and makes suggestions distress and helping them to resolve that distress promptly,
regarding an ED or clinic visit, makes another referral, or but the next level of care involves using CDS based on
changes the symptom management regimen.26,28,29 Algo- national consensus guidelines. Clinical guidelines for symp-
rithms in the PROMIS-T system make appointments for the tom assessment and management, like those from the
patients when the need for social work, registered dietician, National Comprehensive Cancer Network, ASCO, and Mul-
or health educator services is detected.29 The algorithms tinational Association for Supportive Care in Cancer, are often
used in the eRapid system enable the EHR to provide not used in clinical practice.32 On average, it takes 5 years
severity-tailored advice or a recommendation to contact for a guideline to be adopted in practice.33 CDS facilitates
their oncology team.26 The mobile application used in ePAL more rapid dissemination of, and adherence to, these
led to decreased pain severity and notably fewer pain-related guidelines.34
hospital admissions and admissions through the ED.30
CDS provides clinicians, patients, and other health care
In the future, NLP techniques that search the text of the stakeholders with pertinent knowledge and/or person-specific
electronic record for patients with uncontrolled symptoms information, intelligently filtered or presented at appropriate
FIGURE 1. Symptom Assessment and Management Intervention (SAMI) Recommendations: Report for a Patient With Lung Cancer. SAMI for patients with lung
cancer (SAMI-L), a web-based system that is programmed with algorithms derived from national consensus guidelines,37 was sent deidentified patient-
specific data (transmitted remotely via the internet to a remote clinical decision support [CDS] system). The data included patient-reported outcomes
(PROs) completed by the patient with lung cancer that day, relevant laboratory data drawn directly from the electronic health record (EHR), and the results
of Research Assistant interview of the patient confirming the doses of medications being taken. Within 2 minutes, SAMI sends back a report containing
symptom assessments and suggested treatment recommendations for that patient to the clinician who is going to see the patient that day in the outpatient
clinic. The figure shows the top halves of the front and back of a sample report for a patient who had just completed radiation therapy for lung cancer, who
was suffering from uncontrolled dyspnea and moderate pain. The red boxes indicate, on the left, the front page of the report detailing the suggestions for
treating dyspnea, which the patient reported as severe; and on the right, on the back of the report, the longitudinal record of the patient’s treatments and
reports of symptoms, including dyspnea, over time, documenting the marked increase in dyspnea reported at this visit. Severe symptoms are indicated in
red; moderate, in yellow; and mild, in green.

times, to enhance health and health care.35 CDS takes the complement the adopted suggestions, and billing codes
information provided by the PROs or from NLP and, using that can be used for reimbursement of symptom-focused
algorithms developed from accepted national consensus services.
guidelines for symptom assessment and treatment along
with patient-specific data, “works behind the scenes” in the As an example, consider the care of Emilio Gonzales (not
EHR to capture, analyze, and make suggestions for as- a real patient). Emilio has lung cancer metastatic to the bone
sessment and treatment of patients’ pain and other and comes to clinic for his next cycle of therapy. While he
symptoms.36-38 waits to see Dr. Smith, he completes symptom assessment
questionnaires on his smartphone or tablet, which take
For CDS systems to improve practice, they must be part of
about 10 minutes. The contents of these questionnaires are
the workflow, provide specific recommendations, be pres-
uploaded automatically into the EHR, which can seam-
ent at the time and location of decision-making, and be
lessly access the CDS-generated algorithms for symptom
computer based.39-41 We incorporated these principles in
assessment and management recommendations. When
our study of CDS combined with PROs and provided tho-
Dr. Smith accesses Emilio’s chart, on her computer screen
racic oncologists with paper reports containing the se-
is an electronic, color-coded report (like Fig. 1), indicating
verity of each symptom along with recommendations for
the severity of each of his symptoms (e.g., pain, dyspnea,
treating pain, anxiety, depression, dyspnea, and fatigue
anxiety, depression, fatigue), what medications Emilio has
(Fig. 1).36,38
been taking, and relevant lab results. She sees that Emilio
In the next iteration of CDS, clinical guidelines will again be has been on a 50-µg/hour fentanyl patch for his pain from
used to create algorithms that generate patient-specific bone metastases and takes eight rescue doses of 5 mg of
evidence-based suggestions for symptom assessment and oxycodone per day, but his pain is still severe. She also sees
management using PROs and other data available in the that his dyspnea is mild but his anxiety is moderate. Dr.
EHR. In addition, these CDS systems will include validation Smith clicks that she accepts the recommendations for
from the clinician as to the doses of medications the patient increasing the fentanyl patch to 62.5 µg/hour and the
is taking, and the guidance report will be generated in the oxycodone to 10 mg and adding ibuprofen and a zoledronic
EHR for the clinician’s review during a remote or actual acid infusion. She also “agrees” to a social work referral for
patient visit. This report will provide patient-specific as- Emilio’s anxiety, which is exacerbated by his pain. As she
sessment and management recommendations along with clicks on each of these choices, a symptom management
“smart phrases” for automatic documentation of sugges- progress note is being generated automatically for her to
tions adopted, tailored patient educational materials to copy and paste into her documentation. She can also click
FIGURE 2. Integrating Patient-

Reported Outcomes (PROs) and
Natural Language Processing (NLP)
Data With Clinical Decision Support
(CDS)–generated Algorithms to
Improve Treatment of Symptoms
of Outpatients With Cancer
PROs completed by patients at
home or in the outpatient clinic
and NLP data derived from doc-
umentation of patient encounters
are uploaded to the electronic
health record (EHR). The EHR
can access CDS-generated algo-
rithms to do a number of things:
(1) alert a clinician if a patient
has an uncontrolled symptom;
(2) instruct the patient to con-
tact the clinician right away;
(3) make referrals directly to
social worker (SW), nutritionist,
registered dietician (RD), or other clinicians; or (4) provide comprehensive evidence-based symptom assessment and treatment recom-
mendations to the clinician to use during the patient’s visit that day (e.g., symptom assessment and management intervention in patients with
lung cancer, or SAMI-L).

to print out the relevant patient education materials, and CONCLUSION

she is provided with the appropriate Intenational Classi- By employing the communication techniques described
fication of Diseases, 10th revision, codes for billing. When above, partnering with community palliative care clinicians,
Emilio goes home, he will be invited to complete PROs
and collaborating in telehealth visits for patients, oncology
routinely through the EHR patient portal, or his symptoms will
clinicians can begin to integrate primary palliative care
be detected by NLP of text notes he sends through the portal.
principles and management into patient care. In the future,
The EHR will then send alerts based on symptom se-
verity or worsening trends or make referrals. Because his improvements in integration of PROs and NLPs into the
PROs will be uploaded automatically into the EHR, when EHR, along with novel programming that provides symptom
Dr. Smith talks with him, she can can continue to access assessment and management recommendations accessible
the CDS-generated algorithms for comprehensive, evidence- in the EHR during patient visits, may further integrate state-of-
based, symptom assessment and management recom- the-art palliative care and symptom management with the
mendations (Fig. 2). state-of-the-art oncology care that patients receive.

1
University of Washington School of Medicine, Seattle, WA AND DATA AVAILABILITY STATEMENT
2
Aspire Health, Newtown, PA Disclosures provided by the authors and data availability statement (if
3
Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA applicable) are available with this article at DOI https://doi.org/10.1200/
EDBK_100022.
Janet Abrahm, MD, Dana-Farber Cancer Institute, JF 2010, 450
Brookline Ave., Boston, MA 02215; email: [email protected].
REFERENCES
1. Epstein RM, Street RL. Patient-Centered Communication in Cancer Care: Promoting Healing and Reducing Suffering. Bethesda, MD: National Cancer Institute;
2007.
2. Baile WF, Buckman R, Lenzi R, et al. SPIKES---A six-step protocol for delivering bad news: application to the patient with cancer. Oncologist. 2000;5:302-311.
3. Back AL, Arnold RA, Tulsky JA. Mastering Communication With Seriously Ill Patients: Balancing Honesty With Empathy and Hope. New York, NY: Cambridge
University Press; 2009.
4. Meier DE, Back AL, Morrison RS. The inner life of physicians and care of the seriously ill. JAMA. 2001;286:3007-3014.
5. Childers JW, Back AL, Tulsky JA, et al. REMAP: a framework for goals of care conversations. J Oncol Pract. 2017;13:e844-e850.
6. Back AL, Arnold RM, Baile WF, et al. Efficacy of communication skills training for giving bad news and discussing transitions to palliative care. Arch Intern Med.
2007;167:453-460.
7. Fallowfield L, Jenkins V, Farewell V, et al. Efficacy of a Cancer Research UK communication skills training model for oncologists: a randomised controlled trial.
Lancet. 2002;359:650-656.
8. Back AL. Patient-clinician communication issues in palliative care for patients with advanced cancer. J Clin Oncol. Epub 2020 Feb 5.
9. Institute of Medicine. Dying in America: Improving Quality and Honoring Individual Preferences Near the End of Life. Washington, DC: The National Academies
Press; 2014.
10. Labson M, Sacco M, Weissman D, et al. Innovative models of home-based palliative care. Clev Clin J Med. 2013;80 (suppl 1):eS30-35.
11. Kerr CW, Tangeman JC, Rudra CB, et al. Clinical impact of a home-based palliative care program: a hospice-private payer partnership. J Pain Symptom Manage.
2014;48:883-92.e1.
12. Cohn J, Corrigan J, Lynne J, et al. Community-based models of care delivery for people with serious illness. https://nam.edu/community-based-models-of-care-
delivery-for-people-with-serious-illness/. Accessed January 25, 2020.
13. Lustbader D, Mudra M, Romano C, et al. The impact of a home-based palliative care program in an accountable care organization. J Palliat Med. 2017;20:23-28.
14. Brumley RD, Enguidanos S, Cherin DA. Effectiveness of a home-based palliative care program for end-of-life. J Palliat Med. 2003;6:715-724.
15. Brumley R, Enguidanos S, Jamison P, et al. Increased satisfaction with care and lower costs: results of a randomized trial of in-home palliative care. J Am Geriatr
Soc. 2007;55:993-1000.
16. Kerr CW, Donohue KA, Tangeman JC, et al. Cost savings and enhanced hospice enrollment with a home-based palliative care program implemented as
a hospice-private payer partnership. J Palliat Med. 2014;17:1328-1335.
17. Bonsignore L, Bloom N, Steinhauser K, et al. Evaluating the feasibility and acceptability of a telehealth program in a rural palliative care population: TapCloud for
Palliative Care. J Pain Symptom Manage. 2018;56:7-14.

18. Bakitas M, Lyons KD, Hegel MT, et al. Effects of a palliative care intervention on clinical outcomes in patients with advanced cancer: the Project ENABLE II
randomized controlled trial. JAMA. 2009;302:741-749.
19. Meyer H. Changing the conversation in California about care near the end of life. Health Aff (Millwood). 2011;30:390-393.
20. Bull J, Kamal AH, Harker M, et al. Tracking patients in community-based palliative care through the Centers for Medicare & Medicaid Services Healthcare
Innovation Project. J Palliat Med. 2017;20:1231-1236.
21. Ferrell BR, Temel JS, Temin S, et al. Integration of palliative care into standard oncology care: American Society of Clinical Oncology clinical practice guideline
update. J Clin Oncol. 2017;35:96-112.
22. Lupu D, Quigley L, Mehfoud N, et al. The growing demand for hospice and palliative medicine physicians: will the supply keep up? J Pain Symptom Manage.
2018;55:1216-1223.
23. Worster B, Swartz K. Telemedicine and palliative care: an increasing role in supportive oncology. Curr Oncol Rep. 2017;19:37.
24. Quill TE, Abernethy AP. Generalist plus specialist palliative care: creating a more sustainable model. N Engl J Med. 2013;368:1173-1175.
25. Cooley ME, Nayak MM, Abrahm JL, et al. Patient and caregiver perspectives on decision support for symptom and quality of life management during cancer
treatment: implications for eHealth. Psychooncology. 2017;26:1105-1112.
26. Basch E, Barbera L, Kerrigan CL, et al. Implementation of patient-reported outcomes in routine medical care. Am Soc Clin Oncol Educ Book. 2018;38:122-134.
27. Basch E, Deal AM, Dueck AC, et al. Overall survival results of a trial assessing patient-reported outcomes for symptom monitoring during routine cancer treatment.
JAMA. 2017;318:197-198.
28. Denis F, Basch E, Septans A-L, et al. Two-year survival comparing web-based symptom monitoring vs routine surveillance following treatment for lung cancer.
JAMA. 2019;321:306-307.
29. Wagner LI, Schink J, Bass M, et al. Bringing PROMIS to practice: brief and precise symptom screening in ambulatory cancer care. Cancer. 2015;121:927-934.
30. Kamdar M, Centi AJ, Agboola S, et al. A randomized controlled trial of a novel artificial intelligence-based smartphone application to optimize the management of
cancer-related pain. J Clin Oncol. 2018;36 (suppl; abstr 76).
31. Koleck TA, Dreisbach C, Bourne PE, et al. Natural language processing of symptoms documented in free-text narratives of electronic health records: a systematic
review. J Am Med Inform Assoc. 2019;26:364-379.
32. Borneman T, Piper BF, Sun VC, et al. Implementing the Fatigue Guidelines at one NCCN member institution: process and outcomes. J Natl Compr Canc Netw.
2007;5:1092-1101.
33. Balas EA, Boren SA. Managing Clinical Knowledge for Health Care Improvement. In Bemmel J, McCray AT (eds). Yearbook of Medical Informatics 2000: Patient-
Centered Systems. Stuttgart, Germany: Schattauer Verlagsgesellschaft mbH; 2000;65-70.
34. Latoszek-Berendsen A, Tange H, van den Herik HJ, et al. From clinical practice guidelines to computer-interpretable guidelines. A literature overview. Methods
Inf Med. 2010;49:550-570.
35. Osheroff JA, Pifer EA, Teich JM (eds), et al. Improving Outcomes With Clinical Decision Support: An Implementer’s Guide. Chicago: Healthcare Information and
Management Systems Society Press; 2005.
36. Cooley ME, Lobach DF, Johns E, et al. Creating computable algorithms for symptom management in an outpatient thoracic oncology setting. J Pain Symptom
Manage. 2013;46:911-924.e1.
37. Cooley ME, Blonquist TM, Catalano PJ, et al. Feasibility of using algorithm-based clinical decision support for symptom assessment and management in lung
cancer. J Pain Symptom Manage. 2015;49:13-26.
38. Lobach DF, Johns EB, Halpenny B, et al. Increasing complexity in rule-based clinical decision support: the Symptom Assessment and Management Intervention.
JMIR Med Inform. 2016;4:e36.
39. Bright TJ, Wong A, Dhurjati R, et al. Effect of clinical decision-support systems: a systematic review. Ann Intern Med. 2012;157:29-43.
40. Kawamoto K, Houlihan CA, Balas EA, et al. Improving clinical practice using clinical decision support systems: a systematic review of trials to identify features
critical to success. BMJ. 2005;330:765.
41. Lobach DF. The road to effective clinical decision support: are we there yet? BMJ. 2013;346:f1616.

invited article Emerging Subtypes and New Treatments for

Castration-Resistant Prostate Cancer
Benedito A. Carneiro, MD, MS1,2; Tamara L. Lotan, MD3; Andre de Souza, MD1,2; and Rahul Aggarwal, MD4
overview
Genomic characterization of metastatic castration-resistant prostate cancer (mCRPC) has been remodeling
the treatment landscape of this disease in the past decade. The emergence of molecularly defined subsets of
mCRPC is altering the treatment paradigm from therapeutics with nonspecific activity across the spectrum,
including androgen receptor (AR)-directed treatments, docetaxel, and cabazitaxel, to targeted approaches
directed at molecular subsets of disease. The meaningful benefit of PARP inhibitors in mCRPC carrying
mutations in DNA repair genes demonstrated in a phase III trial epitomizes this transition in the treatment
paradigm of mCRPC and brings new challenges related to how to sequence and integrate the targeted
therapies on top of the treatments with broad activity in all mCRPC. To enable and sustain the advance of
precision oncology in the management of mCRPC, genomic characterization is required, including somatic
and germline testing, for all patients with the ultimate goal of longitudinal molecular profiling guiding
treatment decisions and sequential treatments of this lethal disease. This article reviews the emerging
molecular subtypes of mCRPC that are driving the evolution of mCRPC treatment.
INTRODUCTION response to checkpoint inhibitors provides another

Advancement in the understanding of pathogenesis example of treatments guided by genomic subsets of
and refined molecular characterization of mCRPC has mCRPC.8,9 These advances challenge our research
transformed the treatment landscape of this disease in field to design the optimal integration of these novel
the past decade. The Cancer Genome Atlas described therapies into the existing framework anchored on
androgen signaling blockade and chemotherapy and
seven molecular subtypes of prostate cancer that in-
to continue to develop reliable biomarkers to guide
cluded four groups characterized by gene fusions in-
selection and sequence of treatments. This article
volving the E26 transformation-specific (ETS) family of
reviews the emerging molecular subtypes of mCRPC
transcription factors (ERG, ETV1/4, FLI1) and three
that are driving the evolution of mCRPC treatment.
subsets defined by gene mutations (FOXA1, SPOP,
and IDH1).1 Somatic and germline analyses of prostate CLINICALLY ACTIONABLE SUBTYPES OF mCRPC
cancer reveal clinically meaningful molecular alter- IN 2020
ations that can inform prognosis and expand treatment BRCA2-Mutated mCRPC
options.2-5 In a recent comprehensive analysis inte- Two seminal publications in 2005 from the Alan
grating genomic and transcriptomic information, his- Ashworth and Thomas Helleday groups demonstrated
tology, and clinical outcomes from 429 patients with that mutations in BRCA1/2 sensitized cells to inhibition
mCRPC, RB1 loss had the strongest association with of PARP and provided the scientific rationale for the
inferior clinical outcomes such as poor overall survival clinical development of PARP inhibitors (PARPi) in
Author affiliations and faster development of resistance to AR inhibitors.6 oncology.10,11 Tumor cells deficient in BRCA2 (BRCA2 / ),
and support The therapeutic vulnerability of mCRPC with mutations which is involved in double-stranded DNA repair, have
information (if in DNA repair genes to poly(ADP-ribose) polymerase
applicable) appear
a compromised homologous recombination (HR)
at the end of this
(PARP) inhibitors provided the rationale for the clinical DNA repair mechanism and rely on other DNA repair
article. development and incorporation of this class of drugs in pathways such as nonhomologous end-joining, which
Accepted on April the treatment landscape of mCRPC. The meaningful is mediated by PARP.12,13 These alternative repair
22, 2020 and benefit of PARP inhibitors in mCRPC-carrying mutations mechanisms are prone to errors and can result in
published at in DNA repair genes demonstrated in a recent phase III genetic instability that contributes to carcinogene-
ascopubs.org on May
22, 2020: DOI https://
trial epitomizes the evolution in the treatment paradigm sis.14 PARP inhibition in the setting of BRCA2 de-
doi.org/10.1200/ of mCRPC.7 The identification of microsatellite instability ficiency leads to frequent single-strand DNA breaks
EDBK_100025 (MSI) and CDK12 mutations as potential predictors of that cannot be repaired and result in collapsed replication
2020 ASCO EDUCATIONAL BOOK | asco.org/edbook e319

Carneiro et al
tumors displaying mutations or deletions in genes regulating

HR (i.e., BRCA1/2, ATM, FANCA, PALB2, CHEK2) had tu-
mor responses.18 All seven tumors with BRCA2 loss and four
• Treatment of mCRPC has evolved from thera- out of five tumors with ATM deleterious mutations had re-
pies that have broad activity across the disease
sponses to olaparib, as well as those with aberrations in
spectrum (i.e., AR-targeted therapies, chemo-
BRCA1, FANCA PALB2, and CHEK2. This study and sub-
therapy) to treatments directed to genomic-
defined subgroups of mCRPC. sequent larger cohorts demonstrated that 20%–25% of
mCRPC cases harbor somatic or germline alterations in DNA
• The clinical benefit of PARP inhibitors for
repair genes involved with HR and are potentially sensitive to
treatment of mCRPC with DNA repair gene
mutations will help drive the incorporation of PARPi,2,19 which is a significant increase from early estimates
genomic characterization of tumors in the that only 2%–5% of patients with mCRPC carrying germline
clinical practice and promote precision oncol- BRCA1/2 alterations would benefit from PARP inhibition.
ogy in the management of this disease. These results propelled the field to further investigate the
• To enable and sustain the advance of precision therapeutic potential of PARPi in mCRPC with phase III
oncology in the treatment of mCRPC, genomic trial results recently presented. The antitumor activity of
characterization including somatic plus germ- PARPi in mCRPC was confirmed in studies including
line testing for all patients and refinement of the several compounds in this class (i.e., olaparib, ruca-
predictive value of BRCA1/2 and non-BRCA1/2 parib, niraparib, veliparib, and talozaparib).7,20-24
DNA repair genes represent critical
Results of the first phase III 9study comparing olaparib to
components.
standard-of-care antiandrogen treatments among patients
• PSMA-targeted radioligand therapies hold the with mCRPC harboring mutations in HR genes confirmed
potential to represent another pillar in the
the clinical benefit of this strategy.7 Patients with mCRPC
armamentarium of mCRPC treatment both
carrying mutations in 15 predetermined HR genes (BRCA1,
as monotherapy and as a component of
combinatorial strategies. BRCA2, ATM, BRIP1, BARD1, CDK12, CHEK1, CHEK2,
FANCL, PALB2, PPP2R2A, RAD51B, RAD51C, RAD51D,
• Ongoing and future studies will inform the op-
RAD54L) were randomly assigned to receive olaparib or
timal strategy to incorporate PARP inhibitors
into early lines of treatment, and novel drug antiandrogens (enzalutamide or abiraterone plus predni-
combinations may offer treatment alternatives sone). Patients were included in two cohorts: cohort A,
to overcome resistance to PARP inhibitors. mCRPC with mutations in BRCA1, BRCA2, or ATM; or
cohort B, mCRPC with alterations in one of the remaining 12
genes. Combined results from both cohorts showed a sig-
forks, accumulation of DNA defects, lethal doubled-stranded nificant improvement in median progression-free survival
breaks, and ultimately cell-cycle arrest and apoptosis.12,13,15,16 (mPFS) among patients receiving olaparib (5.8 vs. 3.5
The exquisite sensitivity to PARPi as monotherapy promoted months; p , .0001). Olaparib reduced the risk of pro-
the concept of synthetic lethality and was rapidly translated to gression by approximately 66% among the patients in
the clinic. A phase I trial investigating the PARPi olaparib cohort A (mPFS, 7.3 vs. 3.5 months; hazard ratio, 0.34;
showed a favorable toxicity profile and antitumor activity among p , .0001) and led to a remarkable response rate (33% vs.
patients with ovarian, breast, and prostate cancers carrying 2.3%) in this heavily pretreated population (17%–20% of
germline BRCA1/2 mutations.17 One of the patients in this study patients had received enzalutamide and abiraterone,
with mCRPC harboring a BRCA2 mutation had resolution of 18%–24% had received docetaxel and cabazitaxel). Interim
bone metastases and a durable prostate-specific antigen (PSA) overall survival analyses also favored olaparib (17.5 vs. 14.2
response lasting more than 2 years. This signal of activity months; p = .0063). Treatment was well tolerated with an
prompted a broad effort for the clinical development of PARPi adverse event profile consistent with prior studies. Anemia
primarily for treating breast and ovarian cancers with germline was the most frequent grade 3 or higher adverse event
BRCA1/2 mutations. The pivotal observation that somatic (21%), followed by fatigue (2.7%) and vomiting (2%).
and/or germline alterations in other proteins involved with HR Notably, exploratory analyses of mPFS based on individual
were also associated with sensitivity to PARPi reinvigorated gene alterations suggested that olaparib had the most
the interest in these agents in the treatment of mCRPC significant impact among tumors with BRCA2 mutations
and created the possibility to benefit a broader patient
(10.8 vs. 3.48 months; 128 patients) compared with those
population using molecular alterations in DNA repair
harboring CDK12 (5 vs. 2.2 months; 89 patients), ATM (5.3
mechanisms as potential predictive biomarkers.
vs. 4.7 months; 86 patients), or BRCA1 (2 vs. 1.8 months;
A phase II study investigating olaparib for the treatment of 13 patients) alterations. Antitumor activity was observed in
mCRPC (TOPARP-A trial) showed that 88% of patients with tumors carrying other molecular alterations (i.e., CHEK2,
e320 2020 ASCO EDUCATIONAL BOOK | asco.org/edbook

Emerging Subtypes of Castration-Resistant Prostate Cancer and Novel Treatments
RAD51B, RAD54L), but the small size of these subgroups can lead to DNA mutations. DNA mismatch repair (MMR)
limits interpretation of the impact on mPFS. Nevertheless, is a conserved DNA repair mechanism critical to genome
these results highlighted the exquisite sensitivity of BRCA2- integrity. Deleterious mutations in MMR genes (i.e., MSH2,
mutated tumors to olaparib, which has been confirmed in MSH6, PMS2, MLH1) increase the rate of spontaneous
studies investigating other PARPi and fostered the interest in insertions/deletions and results in the MSI-high (MSI-H)
BRCA2-mutated prostate cancers as a relevant pathologic phenotype that contributes to the pathogenesis of sporadic
and clinical subgroup.20-23 and hereditary cancers.38-41 Mutations in MMR genes can
Prostate tumors with BRCA2 mutations have been associated occur in germline DNA (genomic underpinning of Lynch
with aggressive pathologic features, a distinct genomic pro- syndrome).42 MSI-related mutations frequently affect
file, and poor clinical outcomes. Approximately 1%–2% of genes regulating DNA repair (MSH3, MSH6, MRE11A),
patients with prostate cancer have germline BRCA2 muta- epigenetic mechanisms (HDAC2, ARID1A), apoptosis
tions; these are important to distinguish from BRCA2 somatic (BAX), and other cellular processes (TGFBR2, IGFR2,
alterations, which are more prevalent in advanced prostate XPO5). They can result in truncated proteins and neo-
tumors.25-27 Patients with germline mutations in BRCA2 antigens presented by major histocompatibility complex I
develop prostate cancer at a younger age, have a higher risk in the cell surface.43 The majority of MSI-H tumors (83%)
of lymph and distant metastases, and have decreased sur- across multiple cancer types have a higher tumor muta-
vival compared with those without BRCA2 mutations.28-32 tional burden associated with expression of neoantigens
These tumors are also frequently associated with intra- that promote immune cell infiltration and upregulation of
ductal carcinoma, which in itself portends poor prognosis.28,33 programmed death-ligand 1 (PD-L1), fostering antitumor
Detailed genomic analyses using whole-genome sequencing immune response.44-47 These observations supported the
of primary prostate tumors harboring deleterious BRCA2 hypothesis that MMR deficiency (dMMR)/MSI-H pheno-
mutations showed that these tumors have a unique molec- type could predict responses to checkpoint inhibitors
ular profile resembling CRPC more closely than sporadic across multiple histologies.48,49 In fact, 53% of patients
castration-sensitive tumors.34 BRCA2-mutated tumors had with dMMR tumors from 12 distinct histologies (the only
higher genomic instability with increased frequencies of patient with prostate cancer in this cohort had a complete
genomic rearrangements and single-nucleotide variants response) had radiographic tumor responses to pem-
compared with wild-type tumors.35 MYC amplification, brolizumab (anti–PD-1), including a 21% rate of complete
a molecular feature of aggressive prostate cancer, was more responses.49 These findings, combined with results of other
prevalent among BRCA2-mutated tumors in comparison trials, led to the approval of pembrolizumab for all dMMR/MSI-
with sporadic tumors controlled by Gleason score (72% vs. H tumors, the first tissue-agnostic U.S. Food and Drug Ad-
20%). Furthermore, BRCA2-mutated tumors and sur- ministration approval of a cancer treatment.50
rounding intraductal carcinoma shared common precursor A comprehensive next-generation sequencing (NGS)-based
clones. These results suggest that treatment-naı̈ve BRCA2- analysis of 1,033 patients with prostate cancer classified
mutated tumors have genomic features similar to those 3% (32 of 1,033) of these tumors as dMMR/MSI-H.9 These
acquired by mCRPC (genomic instability, alterations in tumors were defined by having deleterious germline or
MYC, mTOR, and WNT pathways) that contribute to their somatic alterations in MMR genes or MMR protein loss by
poor prognosis and aggressive biology.34 Although the immunohistochemistry (IHC). Twenty-two percent of the
clinical benefit of PARPi for the treatment of mCRPC car- dMMR/MSI-H cases (7 of 32) displayed germline mutations
rying BRCA2 mutations is well demonstrated, the role of in MMR genes that included MSH2 (5 patients), MSH6 (1
more aggressive treatment of metastatic BRCA2-mutated patient), and PMS2 (1 patient). Notably, two of six patients
tumors at earlier stages of disease using PARPi and/or with serial tumor specimens demonstrated acquisition of
combinations with platinum agents remains to be dem- somatic mutations in MMR genes leading to the dMMR/
onstrated. Ongoing studies are investigating the role of MSI-H phenotype during disease evolution. Eleven patients
PARPi in the treatment of localized BRCA2-mutated with dMMR/MSI-H mCRPC tumors received treatment with
prostate cancer, and more intensive surveillance strate- anti–PD-1/–PD-L1 or other checkpoint inhibitor combina-
gies and optimization of local therapy of BRCA2-mutated tions. PSA decline greater than 50% was observed in 54%
tumors have been proposed based on their worse clinical of patients (6 of 11), four patients had radiographic tumor
outcomes, particularly for those patients with concurrent responses, and one patient achieved stable disease lasting
intraductal carcinoma (NCT03432897).28,30,36,37 approximately 6 months leading to a durable clinical benefit
rate of 45.5% (5 of 11). These results are consistent with the
MSI-High Tumors rate of benefit observed in the study with pembrolizumab in
DNA replication and recombination processes cause the other dMMR/MSI-H tumors49 and underscore the role of
insertion of mismatched DNA bases that if left unrepaired MSI-H as an emerging biomarker in mCRPC.

Carneiro et al
Assessment of dMMR/MSI-H status is currently recom- clinical benefit.55-58 However, it reinforces the need for
mended by the National Comprehensive Cancer Network continued efforts to develop biomarkers for immune-based
for consideration of pembrolizumab treatment of patients therapies for mCRPC and novel combinatorial strategies.59-63
with mCRPC, but it is important to mention some limitations EMERGING THERAPEUTIC TARGETS IN SUBTYPES OF mCRPC
with this biomarker. Standard methodology has not been
established among the platforms available for assessment Treatment-Emergent Small-Cell Neuroendocrine
of dMMR/MSI-H status in prostate cancer (i.e., IHC analysis Prostate Cancer
of MMR proteins, polymerase chain reaction amplification De novo small-cell neuroendocrine prostate cancer is a rare
of specific microsatellite repeats, and NGS-based ap- clinical entity, observed in less than 1% of all newly di-
proaches). Somatic DNA sequencing of tumors can miss agnosed cases of prostate cancer.64 In contrast, treatment-
MMR mutations because exon-only sequencing does not emergent small-cell neuroendocrine prostate cancer (t-SCNC),
detect intronic rearrangements such as those involving the which arises in the castration-resistant setting after the
MSH2 and MSH6 genes, and most genomic assays do application of AR-targeted therapy, is more common.65-68
not report biallelic mutations in MMR, which could be The estimated prevalence of t-SCNC depends in part on the
inferred from IHC results.51,52 The standard polymerase diagnostic criteria but likely ranges between 6% and 15% of
chain reaction–based MSI detection method (five-marker all patients with mCRPC.65 The detection of t-SCNC in the
Bethesda panel), which was developed and validated for mCRPC setting is associated with poor prognosis, with an
colorectal cancer, has a low sensitivity (72%) compared estimated median survival of less than 20 months from the
with NGS-based MSI detection methods (93%–97%) in onset of mCRPC.65
prostate cancer.53 The presence of high tumor mutational Morphologically, the appearance of t-SCNC is similar to that
burden does not always correlate with MSI-H status and of de novo small-cell prostate cancer and other small-cell
should not replace it. Although the vast majority of MSI-H cancer types: densely packed sheets of cells without distinct
tumors (83%) have high tumor mutational burden, ap- glandular formation or prominent nucleoli.69 Immunohis-
proximately only 16% of tumors with high tumor mutational tochemical staining, when performed, frequently shows
burden are MSI-H according to an analysis that included 91 positive staining for neuroendocrine markers, including
cases of prostate cancer.47 These examples highlight the chromogranin, synaptophysin, and CD56. Ki-67, a marker
complementary role of these MSI detection platforms. PD- of tumor cell proliferation, is typically greater than 50%,
L1 expression in tumor or immune cells has not proven to be consistent with other high-grade neuroendocrine cancers.
a clinically useful biomarker for this disease, as suggested Genomically, the entity is characterized by frequent loss of
by a recent study evaluating pembrolizumab.54 function mutations and/or copy number loss of the tumor
The largest clinical trial reported to date investigated suppressor RB1, alone or co-occurring with loss of other
pembrolizumab monotherapy in mCRPC enrolled 258 pa- tumor suppressors including either PTEN or TP53.65,70,71
tients in three cohorts (cohorts 1 and 2, RECIST measurable Indeed, preclinical models have established a potential
disease PD-L1 positive or PD-L1 negative; cohort 3, pre- causal role for the dual loss of RB1 and TP53 in promoting
dominant bone disease irrespective of PD-L1 expression).54 trans-differentiation to t-SCNC.72 Clinically, dual loss of tu-
Pembrolizumab showed modest antitumor activity in this mor suppressors RB1, TP53, and PTEN is associated with
population, with an overall response rate of 5% in cohorts 1 aggressive phenotype and clinical features that are con-
and 2, but it included two patients with complete radio- sistent with t-SCNC. Transcriptionally, t-SCNC is charac-
graphic response and four patients with stable disease terized by significant epigenetic dysregulation with frequent
lasting beyond 6 months. Disease control rate and median upregulation of chromatin modifiers including EZH2.66,70,73
radiologic PFS per Prostate Cancer Working Group 3– Diagnostically, the detection of t-SCNC currently requires
modified RECIST were 20% and 2.1 months for all three histologic confirmation via tumor biopsy acquisition, which
cohorts, respectively. Whole-exome sequencing analysis of may not be feasible for many patients with bone-
tumors from six patients who had tumor responses did not predominant prostate cancer. Laboratory characteristics
show evidence of MSI according to the MSI Dection by NGS and site of metastasis do not reliably capture the presence of
assay. However, evidence of dMMR by IHC was observed t-SCNC. In a prior multi-institutional prospective study de-
among two patients who had durable response of more than scribing the clinical and genomic features of t-SCNC, there
2 years. Clinical outcomes were comparable between PD- was no difference in serum PSA between tumors harboring
L1–positive and PD-L1–negative tumors as defined by the t-SCNC vs. those that did not (median PSA, 64.8 vs. 46.2
combined positive score score. These results support ng/mL; p = .938).65 Similarly, the presence of visceral organ
previous observations that a small subset of patients with metastases did not serve as a reliable surrogate for presence
mCRPC (5%–15%) can achieve tumor responses to of t-SCNC. In contrast, genomic loss of RB1 is an event that
checkpoint inhibitors, and some of them will obtain durable seems to occur late in the disease pathway and may serve as

a potential surrogate for the identification of t-SCNC.74,75 histologic subtype with epigenetic modifiers, including
Recent developments in molecular imaging may eventually enhancer of zeste homolog 2 (EZH2) inhibitors. In a prior
prove to be a reliable surrogate for the identification of phase II study of enzalutamide plus the EZH2 inhibitor CPI-
t-SCNC. In particular, loss of prostate-specific membrane 1205 in unselected patients with mCRPC and prior treat-
antigen (PSMA) expression coupled with high fluorodeox- ment with abiraterone and/or enzalutamide, durable PSA
yglucose uptake (PSMA low/fluorodeoxyglucose high) on and objective tumor responses were observed.78 Ran-
paired PET may enrich for de-differentiated tumors with domized trials are underway to evaluate the epigenetic
potential neuroendocrine transformation. Liquid biopsies with modifiers in combination with AR-targeting agents to pre-
analysis of methylation profiles may also serve as a reliable vent and/or treat t-SCNC (NCT03460977, NCT04179864).
indicator of t-SCNC according to recent work from Beltran Among other emerging therapeutic targets is the atypical
and colleagues, if validated in future prospective studies.70 Notch ligand DLL3, which has been shown to be overex-
The current standard of care for the treatment of t-SCNC is pressed in 50%–60% of all t-SCNC tumors, at a similar
platinum-based doublet chemotherapy, most commonly prevalence as other high-grade neuroendocrine cancers.79
either platinum/taxane or platinum/etoposide. The first A prior phase Ib basket study evaluating the antibody-drug
randomized trial supporting the addition of platinum was conjugate rovalpituzumab tesirine in high-grade neuroen-
recently published by Aparicio and colleagues.76 In this docrine tumors, including t-SCNC, demonstrated pre-
randomized phase I/II study of cabazitaxel with or without liminary evidence of efficacy.80 Novel methods to target
carboplatin among patients with mCRPC, in patients with DLL3, including use of bispecific antibodies, are planned.
dual loss of RB1 with either TP53 or PTEN, detected either Finally, loss of AR activity and so-called de-differentiation
via circulating tumor DNA and/or tumor biopsy, a post hoc under intense and prolonged AR-targeted therapy does not
analysis demonstrated a statistically significant improve- result in a neuroendocrine phenotype in all cases. There are
ment in overall survival with combination chemotherapy likely multiple AR-independent pathways contributing to
versus cabazitaxel monotherapy (median overall survival, therapeutic resistance. Chief among these may be a re-
20.2 vs. 8.5 months; p = .0002). Conversely, in patients cently described double-negative phenotype, characterized
whose tumors did not harbor dual tumor suppressor loss, by Nelson and colleagues at the University of Washington by
overall survival was not improved with the addition of loss of both AR expression and negative neuroendocrine
carboplatin to cabazitaxel (median overall survival, 21.7 vs. markers (absence of positive immunohistochemical stain-
21.5 months; p = .702). Although the results were ex- ing for chromogranin and synaptophysin).81 Understanding
ploratory and require further prospective validation, it the pathogenesis, treatment targets, and disease outcomes
suggests the potential therapeutic benefit of the addition of of the double-negative phenotype is a subject of ongoing
carboplatin in patients with mCRPC with aggressive ge- research.82
notypic features. Targeting PI3K-AKT-mTOR Axis
Despite the potential therapeutic benefit of platinum-based Among the most common lost tumor suppressors in prostate
doublet chemotherapy, the duration of response and mPFS cancer is the phosphatase and tensin homolog (PTEN),
are modest. This is consistent with the short duration of which opposes the oncogenic PI3K-AKT-mTOR signaling
response observed in other high-grade neuroendocrine axis. PTEN loss occurs most commonly by large-scale ge-
cancers. Multiple emerging treatment strategies are being nomic deletions in prostate cancer. It is present in more than
evaluated. Among these include immune checkpoint in- 40% of mCRPC, whereas point mutations inactivating PTEN
hibitors such as the combination of ipilimumab and nivo- or leading to AKT-mTOR inactivation are much rarer.2 PTEN
lumab. In a recently published phase II basket study in loss is associated with higher-grade and more aggressive
patients with pretreated high-grade neuroendocrine can- disease in surgically treated patients, a finding that suggests
cers of various primary sites, including a subset with neu- that therapies targeting PI3K-AKT-mTOR signaling could be
roendocrine prostate cancer (two patients), an encouraging beneficial for lethal prostate cancer. Unfortunately, early
response rate of 44% was observed.77 Further prospective trials of single-agent mTOR and pan-PI3K inhibitors did not
studies are underway to evaluate the impact of immune show promising results, a finding potentially attributable to
checkpoint inhibitors, both in the first-line setting combined loss of feedback inhibition of upstream receptor tyrosine
with platinum-based chemotherapy and in the relapsed/ kinase and AR signaling.83-85
refractory disease setting. Additional combination immune
Notably, patients in the early trials described previously were
checkpoint inhibitor studies with novel targeted agents,
not selected by PTEN status, in part because validated
including PARP inhibitors, are also underway.
assays to assess PTEN status have only recently been
Given the significant epigenetic dysregulation observed developed. Because PTEN loss frequently occurs as
in t-SCNC, there is significant interest in targeting this a subclonal event in primary prostate tumors and bi-allelic

Carneiro et al
PTEN deletion status can be difficult to estimate by seas H2AX, BRCA1, CHEK2, and p53 that ultimately mediate
quencing, validated IHC assays have emerged as the most its effects in the DNA damage response (DDR), cell cycle
promising assays to detect PTEN loss in prostate arrest, and apoptosis98 (Fig. 1). ATM mutations or other
cancer.86,87 The correlation of these assays with genomic molecular dysfunctions directs the DDR to rely on ATR- and
studies is very high.87-89 Using IHC assays to assess PTEN DNA-PK–driven pathways creating therapeutic vulnerabil-
status, more recent studies of AKT inhibitors, such as the ities to their respective inhibitors.99 As a pivotal regulator of
ATP-competitive ipatasertib, have shown some promise in DDR and supported by preclinical data showing hyper-
biomarker-selected populations. In a recent phase II trial of sensitivity of ATM-deficient cells to PARPi,100 ATM muta-
ipatasertib with abiraterone, nearly one-half of the patients tions were postulated as relevant predictors of response to
had PTEN protein loss by IHC assay. When patients who PARPi in mCRPC. However, results from the TRITON2
were PTEN deficient were analyzed as a coprimary substudy showed that only 10% (2 of 19) of patients with ATM
group, the addition of ipatasertib to abiraterone improved alterations had radiographic responses to rucaparib and 4%
median radiologic PFS from 4.6 to 11.5 months (hazard (2 of 49) had PSA responses, results consistent with the
ratio, 0.39; 95% CI, 0.22–0.70). In contrast, PTEN-intact TOPARP-B study (8.3% RECIST response and 5% PSA
patients had no significant benefits from the addition of response).20,96 The phase III PROFOUND study also did not
ipatasertib (hazard ratio, 0.84; 95% CI, 0.51–1.37).89 An show a significant difference in mPFS among 86 patients
ongoing phase III trial is comparing the combination of with ATM alterations receiving olaparib or antiandrogens.
ipatasertib and abiraterone versus placebo plus abiraterone Potential explanations for these findings include the fact that
in the first-line treatment of mCRPC (NCT03072238). ATM has a myriad of functions beyond its role in DNA repair,
Capivasertib (AZD5363), another AKT inhibitor, is also the relatively limited characterization of ATM aberrations in
being investigated in combination with enzalutamide or these studies (distinct impact in DNA repair between ATM
abiraterone in mCRPC (NCT04087174).90 AKT inhibitors mutations producing kinase dead proteins vs. alterations
hold the potential to improve the outcomes of prostate resulting in lack of expression of ATM), and the potential
cancers carrying AKT1 E17K and other activating point compensation by other upstream regulators of DDR helping
mutations involving PIK3CA, PIK3R1, and AKT genes.91,92 preserve the HR (i.e., ATR, DNA-PK).101 Importantly, the
Prostate tumors with SPOP mutations can also display in- recent emergence of potent ATR inhibitors has renewed
creased AKT-mTORC1 signaling, and novel therapeutic interest in interrogating ATM status in prostate cancer and
strategies are under investigation for this subset of other malignancies,102-104 because initial phase I trials have
tumors. 93,94 shown favorable responses to ATR inhibitors that may be
specific to ATM-deficient tumors.105
Other DNA Damage Response Pathway Mutations: ATM,
CHEK2, BRCA1, CDK12 Alterations in cyclin-dependent kinase 12 (CDK12), docu-
mented in 3%–7% of patients with mCRPC, were associated
The initial hypothesis that mCRPC with alterations in DNA
with sensitivity to PARPi in preclinical studies based on its
repair genes beyond BRCA1/2 would be sensitive to PARPi
involvement with DDR, but its role as a predictor biomarker
was built on the premise that genomic alterations in these
in mCRPC remains elusive.8,106 In the PROFOUND study,
other genes (i.e., ATM, BRIP1, BARD1, CDK12, CHEK1,
the 61 patients carrying alterations in CDK12 achieved
CHEK2, FANCL, PALB2, PPP2R2A, RAD51B, RAD51C,
longer mPFS when treated with olaparib compared with the
RAD51D, RAD54L) would compromise HR and increase
28 patients receiving enzalutamide or abiraterone (5 vs. 2.2
the reliance on PARP-mediated repair mechanisms.15,95
months) without statistical significance documented.7 Other
However, emerging results from clinical trials suggest that
studies also did not demonstrate significant tumor re-
deleterious mutations in ATM, CDK12, and CHEK2 are
sponses in patients with CDK12 alterations receiving PARPi.
associated with lower rates of tumor response and clinical
No PSA or radiographic responses were observed among 20
benefit compared with alterations in PALB2, FANCA,
patients with CDK12 alterations receiving olaparib in the
BRIP1, and RAD51B, suggesting that further research is
TOPARP-B study (5 patients had a significant decline of
needed to better characterize the relative sensitivity of these
circulating tumor cells); only 1 of 15 patients treated with
non-BRCA gene alterations to PARPi.96
rucaparib had a PSA response in the TRITON2 study.20,96
tAaxia-telangiectasia mutated (ATM), together with ataxia Multicenter retrospective studies also showed a lack of
telangiectasia and Rad53-related (ATR) and DNA- activity of PARPi in this population. They revealed clinical
dependent protein kinase (DNA-PK), plays a central role and pathologic features associated with poor prognosis
in the HR pathway after recognition of double-stranded (i.e., higher Gleason score at presentation, shorter time to
breaks by sensor proteins (Mre11, Rad50, Nbs1).97 After development of CRPC and metastases, poor tumor re-
binding to double-stranded breaks, ATM undergoes auto- sponses to hormonal agents, taxanes, and PARPi).107,108
phosphorylation and activates downstream substrates such There is a growing appreciation for CDK12 as a predictor of

FIGURE 1. Overview of DNA Re-

pair Proteins and Therapeutic
Targets
(1) ATM binds to double-
stranded breaks, undergoes
auto-phosphorylation, which can
lead to cell cycle arrest and ap-
optosis mediated by CHEK2
and p53. (2) Single strand DNA
breaks or replication stress at the
replication fork activates ATR
which in turn phosphorylates
CHEK1, which holds the pro-
gression of the cell cycle from G2
to M. (3) PARP is essential to the
assembly of the protein complex
that repairs single strand breaks,
as a rescue pathway when the
BRCA2 system does not work.
PARP inhibition leads to accu-
mulation of single-strand DNA
breaks that cannot be repaired
and ultimately results in col-
lapsed replication forks, lethal
doubled stranded breaks, and
cell death. Dual CHEK1/2 in-
hibitor: LY2606368; ATR inhibitor:
ceralasertib; PARP inhibitors: ola-
parib, rucaparib, niraparib, veliparib,
and talazoparib. Abbreviations:
ATM, ataxia-telangiectasia mutated;
ATR, ataxia telangiectasia and Rad53-related.
response to checkpoint inhibitors instead.109 CDK12-mutated deficiency driven by BRCA1/2 alterations is an area of active
prostate tumors have increased T-cell infiltration and burden investigation, and combinatorial strategies are being ex-
of neoantigens resulted from recurrent gene fusions, which plored with the ultimate goal to decrease HR activity and
seems to support responses to checkpoint inhibitors observed promote vulnerability to PARPi.95
in this group of patients.8,108 Clinical trials are investigating the
Combination strategies to induce BRCAness/synthetic le-
efficacy of checkpoint inhibitors in this subset of mCRPC thality with PARPi or to overcome resistance Recognizing
(NCT03570619, NCT04104893, NCT04336943). that mutations in DDR genes will not be able to predict the
Despite their essential roles as effectors of DDR, mutations response to PARPi in all patients and the importance of the
in CHEK2 and BRCA1 were not associated with statistically HR-deficiency phenotype observed independently of DDR
significant PFS improvement with olaparib in the PRO- gene alterations, several combinations of PARPi with other
FOUND study. Only two of 12 patients with CHEK2 muta- targeted agents and immunotherapy are undergoing clinical
tions had tumor responses to rucaparib in the TRITON2; investigation (Table 1).
both patients had co-occurring mutations in DDR genes Activation of the PI3K/mTOR pathway has been well
(i.e., ATM, BRCA2), suggesting a limited prediction power of documented as a potential mechanism of resistance to
CHEK2-isolated alterations. Encouraging tumor responses PARPi, and its blockade impaired HR and sensitized cells to
were observed among patients with PALB2 (four of seven in PARP inhibition in several preclinical models.111-115 The
the TOPARP-B study, two of two in the TRITON2), BRIP1, combination of olaparib and the PI3K inhibitor buparlisib
RAD51B, RAD54L, and FANCA, but future analyses with (BKM120) evaluated among 70 patients with high-grade
a larger number of patients is necessary to draw definitive serous ovarian and breast cancers showed a promising
conclusions regarding the role of non-BRCA2 mutations as signal of efficacy in patients with BRCA wild-type and
biomarkers of response to PARPi in mCRPC.7,20,96,110 Fur- those carrying BRCA2 germline mutations.116 In models of
thermore, predictors of response to PARPi beyond HR prostate cancer, olaparib plus buparlisib showed significant

Carneiro et al
TABLE 1. Ongoing Clinical Trials Evaluating Novel Combinations for Treatment of mCRPC
Combinations Drugs Arms Clinical Trial Identifier
PARP Inhibitors Combinations
PARPi and anti–PD-1 Olaparib and pembrolizumab Olaparib plus pembrolizumab vs. enzalutamide or KEYLYNK-010 NCT03834519
abiraterone/prednisone phase III
Rucaparib and nivolumab Single arm Phase I/II NCT03572478
Niraparib, cetrelimab, and Niraparib plus cetrelimab, niraparib plus abiraterone QUEST phase I/II NCT03431350
abiraterone
Talazoparib and avelumab Single arm Javelin PARP NCT03330405
phase I/II
PARP and AR signaling Niraparib and abiraterone Niraparib plus abiraterone vs. abiraterone MAGNITUDE NCT03748641
inhibitors phase III
Tazaloparib and enzalutamide Tazaloparib plus enzalutamide vs. enzalutamide TALAPRO-2 NCT03395197
phase III
PARP and VEGF inhibitors Olaparib and cediranib Olaparib with or without cediranib Phase II NCT02893917
PARP and AKT inhibitors Rucaparib and ipatasertib Single arm Phase I/II NCT03840200
PARP and PI3K inhibitors Rucaparib and copanlisib Single arm Phase I/II NCT04253262
PARP and ATR inhibitors Olaparib and ceralasertib Single arm TRAP phase II NCT03787680
(AZD6738)
PARP and chemotherapy Talazoparib and temozolomide Single arm Phase I/II NCT04019327
Rucaparib, docetaxel and Single arm PLATI-PARP NCT03442556
carboplatin phase II
PARP and radioisotope Olaparib and radium-223 Olaparib plus radium-223 and radium-223 Phase I/II NCT03317392
Niraparib and radium-223 Single arm NiraRad phase Ib NCT03076203
177
Olaparib and Lu-PSMA Single arm Phase I NCT03874884
Combinations Targeting PI3K-AKT-mTOR Axis
PI3K β inhibitor and AZD-8186 and Docetaxel Single arm Phase I NCT03218826
docetaxel
AKT and AR signaling Ipatasertib and abiraterone Ipatasertib plus abiraterone vs. abiraterone IPATential150 NCT03072238
inhibitors phase III
AKT and anti–PD-L1 Ipatasertib and atezolizumab Single arm IceCAP phase I/II NCT03673787
Combinations With Checkpoint Inhibitors
177
Anti–PD-1 and Pembrolizumab and Lu- Single arm Phase Ib NCT03805594
radioisotope PSMA
Pembrolizumab and radium- Radium 223 and radium-223 plus pembrolizumab Phase II NCT03093428
223
Anti–PD-1 and Nivolumab and testosterone Single arm COMBAT-CRPC NCT03554317
testosterone phase II
Anti–PD-1 and AR Pembrolizumab and Pembrolizumab plus enzalutamide vs. enzalutamide KEYNOTE-991 NCT04191096
inhibitor enzalutamide phase III
Anti–PD-1 plus PARPi or Pembrolizumab, olaparib, Pembrolizumab plus olaparib or pembrolizumab plus KEYNOTE-365 NCT02861573
taxane or signaling docetaxel, enzalutamide, or docetaxel or pembrolizumab plus enzalutamide or phase Ib/II
inhibitors abiraterone pembrolizumab plus abiraterone
Nivolumab, rucaparib, Nivolumab plus rucaparib or nivolumab plus CheckMate 9KD NCT03338790
docetaxel, or enzalutamide docetaxel or nivolumab plus enzalutamide phase II
Taxane and PD-1 Docetaxel and pembrolizumab Pembrolizumab plus docetaxel vs. placebo plus KEYNOTE-921 NCT03834506
docetaxel phase III
(Continued on following page)

TABLE 1. Ongoing Clinical Trials Evaluating Novel Combinations for Treatment of mCRPC (Continued)
Combinations Drugs Arms Clinical Trial Identifier
Taxane and anti–PD-1 and Nivolumab, ipilimumab, Nivolumab plus ipilimumab or ipilimumab or CheckMate 650 NCT02985957
anti–CTLA-4 cabazitaxel cabazitaxel
Nivolumab, docetaxel Nivolumab plus docetaxel or docetaxel CheckMate 7DX NCT04100018
phase III
Abbreviations: mCRPC, metastatic castration-resistant prostate cancer; PARPI, PARP inhibitor; AR, androgen receptor; ATR, ataxia telangiectasia and
Rad53-related.
synergy and induced apoptosis and growth suppression treatment both as monotherapy and as a component of
in vitro, as well as marked tumor regression and improve- combinatorial strategies. PSMA is a transmembrane gly-
ment of PFS in a xenograft model of aggressive prostate coprotein expressed in prostate epithelium, as well as brain
cancer (Pten / p53 / ).117 These results provided the astrocytes, duodenal enterocytes, proximal renal tubule,
preclinical rationale for an ongoing phase I/II clinical trial and salivary gland epithelium.121 PSMA is expressed in up
combining rucaparib and copanlisib (PI3K inhibitor) for to 95% of prostate tumors, with expression levels in-
mCRPC in our institutions (NCT04253262). Another study creasing after castrate resistance, which makes it an
is evaluating the combination of copanlisib, olaparib, and attractive target for imaging modalities and therapy. PSMA-
durvalumab among patients with advanced solid tumors targeted radionuclide therapies consist of radiolabeled
carrying mutations in DDR genes or alterations in PTEN or small molecules or monoclonal antibodies. Small mole-
PIK3CA (NCT03842228). cules are based on glutamate-urea-lysine moieties tar-
Intrinsic or acquired resistance to PARPi also represents an geting the glutamate carboxypeptidase II pocket of PSMA.
area of active investigation and particularly important Conversely, antibodies target a domain distant from the
considering its growing clinical use for the treatment of active site of the enzyme. Small molecules (e.g., 617)
mCRPC. Based on increased reliance on the ATR pathway distribute diffusely through the body and can cause
in BRCA1/2-deficient cells resistant to PARPi,118 clinical nonhematologic adverse effects such as xerostomia, fa-
studies are exploring the combination of PARPi with ATR tigue, and nausea more often than antibodies. Antibodies
inhibitors in advanced solid tumors (NCT02264678, (e.g., J591) are not excreted by the kidney, do not ac-
NCT02723864, NCT03462342). Combinations of PARPi cumulate in lacrimal or salivary glands (as shown by 89Zr-
with ATR and other cell cycle checkpoint inhibitors such as J591 PET imaging), and have a longer half-life than small
WEE1 kinase and checkpoint kinase 1 (CHK1) have po- molecules, leading to myelosuppression because of the
tential synergy by compromising cell cycle arrest necessary bystander effect of its radionuclide.122
to repair DNA damage caused by PARPi.13,95 Ongoing 177
Lu-PSMA J591 monoclonal showed encouraging anti-
combination studies of olaparib and WEE1 and CHK1 tumor activity in early trials using a single-dose sched-
inhibitors might reveal promising results to mCRPC ule with myelosuppression representing the dose-limiting
treatment naı̈ve or resistant to PARPi (NCT02511795, toxicity.123,124 Phase II studies involving a total of 75
NCT03057145). patients showed PSA responses in up to 60% of patients,
Combinations of PARPi with chemotherapy agents have been with up to 26% achieving a PSA decline higher than 30%.
limited by efficacy and increased myelotoxicities. However, In the first trial, measurable disease was present in 26%
novel sequential strategies with cytotoxic agents aligned with of patients, with one (8.3%) of these achieving partial
biomarker-defined selection of patients might lead to better radiographic response and 66% showing stable dis-
outcomes as suggested by recent studies evaluating main- ease.123 Approximately 50% of patients developed severe
tenance olaparib after platinum-based chemotherapy among thrombocytopenia and 26% had severe neutropenia. A
patients with metastatic pancreatic cancer carrying germline dose-fractionated schedule allowed a higher dose to be
BRCA mutations, and another study investigating the com- delivered, leading to higher PSA responses (29% of the
bination of cisplatin/gemcitabine/veliparib in patients with patients achieved more than 50% PSA decline), but
germline BRCA/PALB2-mutated pancreatic cancers.119,120 grade 4 thrombocytopenia was documented in 58% of
patients. 125
PSMA-Based Treatments Small molecules targeting PSMA have shown significant
PSMA-targeted radionuclide therapies hold the potential to antitumor activity and less frequent myelosuppression. A
represent another pillar in the armamentarium of CRPC multicenter retrospective analysis including 145 patients

Carneiro et al
with mCRPC treated with 177Lu PSMA-617 showed that CONCLUSIONS

45% (45 of 99) of patients had a PSA decline of at least Remarkable clinical advances have been built on the ge-
50%, with 12% (18 of 145) of patients developing grade nomic characterization of mCRPC. Subsets of mCRPC
3–4 hematologic toxicities.126 A phase II study involving 43 defined by molecular features such as DNA repair gene
patients with mCRPC receiving up to four cycles of 177Lu mutations, for instance, have propelled a novel class of
PSMA-617 documented a PSA decline of at least 50% in therapies in this disease. Several ongoing clinical trials are
57% of patients (17 of 30), objective responses in nodal or exploring strategies to bring these agents to earlier lines of
visceral disease in 82% (14 of 17) of patients, and im- therapy, refine predictive biomarkers, and provide mecha-
provement of pain with a favorable toxicity profile (grade 3 nisms to overcome resistance. The emerging role of CDK12
or 4 thrombocytopenia in 13% of patients).127 These en- defining a subset of mCRPC with potential implications to
couraging results are supporting confirmatory studies and antitumor immune response holds the potential to advance
novel combinations. An ongoing study is investigating the the clinical benefits of checkpoint inhibitor or other immune-
combination of escalating doses of olaparib with one to based therapies in mCRPC. Genomic profiling of tumors
three cycles of 177Lu PSMA-617 (NCT03874884). Another aligned with progress in circulating tumor DNA technologies
trial is studying the association with pembrolizumab and incorporation of clinical outcomes remains a critical pillar
(NCT03805594), and a phase III trial will investigate, for to sustain the development of reliable biomarkers and pre-
the first time, the superiority of a radionuclide to cab- dictors of response capable of informing treatments and
azitaxel (NCT03392428; Table 1). optimal sequence in mCRPC while fostering discoveries.
AFFILIATIONS University, 164 Summit St., Providence, RI 02906; email: benedito_

1
Warren Alpert Medical School, Brown University, Providence, RI [email protected].
2
Lifespan Cancer Institute, Providence, RI
3
Department of Pathology, Johns Hopkins University, Baltimore, MD
4 AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
University of California, San Francisco, CA
AND DATA AVAILABILITY STATEMENT
Disclosures provided by the authors and data availability statement (if
applicable) are available with this article at DOI https://doi.org/10.1200/
CORRESPONDING AUTHOR EDBK_100025.
Benedito A. Carneiro, MD, MS, Lifespan Cancer Institute, Division of
Hematology/Oncology, The Warren Alpert Medical School of Brown
REFERENCES
1. Cancer Genome Atlas Research Network. The molecular taxonomy of primary prostate cancer. Cell. 2015;163:1011-1025.
2. Robinson D, Van Allen EM, Wu YM, et al. Integrative clinical genomics of advanced prostate cancer. Cell. 2015;161:1215-1228.
3. Pritchard CC, Mateo J, Walsh MF, et al. Inherited DNA-repair gene mutations in men with metastatic prostate cancer. N Engl J Med. 2016;375:443-453.
4. Barbieri CE, Baca SC, Lawrence MS, et al. Exome sequencing identifies recurrent SPOP, FOXA1 and MED12 mutations in prostate cancer. Nat Genet. 2012;
44:685-689.
5. Antonarakis ES, Lu C, Wang H, et al. AR-V7 and resistance to enzalutamide and abiraterone in prostate cancer. N Engl J Med. 2014;371:1028-1038.
6. Abida W, Cyrta J, Heller G, et al. Genomic correlates of clinical outcome in advanced prostate cancer. Proc Natl Acad Sci USA. 2019;116:11428-11436.
7. Hussain M, Mateo J, Fizazi K, et al. Profound: Phase 3 study of olaparib versus enzalutamide or abiraterone for metastatic castration-resistant prostate cancer
(mCRPC) with homologous recombination repair (HRR) gene alterations. Ann Oncol. 2019;30:v851-v934.
8. Wu YM, Cieslik M, Lonigro RJ, et al. Inactivation of CDK12 delineates a distinct immunogenic class of advanced prostate cancer. Cell. 2018;173:1770-1782.
9. Abida W, Cheng ML, Armenia J, et al. Analysis of the prevalence of microsatellite instability in prostate cancer and response to immune checkpoint blockade.
JAMA Oncol. 2019;5:471-478.
10. Farmer H, McCabe N, Lord CJ, et al. Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy. Nature. 2005;434:917-921.
11. Bryant HE, Schultz N, Thomas HD, et al. Specific killing of BRCA2-deficient tumours with inhibitors of poly(ADP-ribose) polymerase. Nature. 2005;
434:913-917.
12. Jelinic P, Levine DA. New insights into PARP inhibitors’ effect on cell cycle and homology-directed DNA damage repair. Mol Cancer Ther. 2014;13:1645-1654.
13. Haynes B, Murai J, Lee JM. Restored replication fork stabilization, a mechanism of PARP inhibitor resistance, can be overcome by cell cycle checkpoint
inhibition. Cancer Treat Rev. 2018;71:1-7.
14. Lord CJ, Ashworth A. BRCAness revisited. Nat Rev Cancer. 2016;16:110-120.

15. Lord CJ, Ashworth A. PARP inhibitors: synthetic lethality in the clinic. Science. 2017;355:1152-1158.
16. Ashworth A, Lord CJ. Synthetic lethal therapies for cancer: what’s next after PARP inhibitors? Nat Rev Clin Oncol. 2018;15:564-576.
17. Fong PC, Boss DS, Yap TA, et al. Inhibition of poly(ADP-ribose) polymerase in tumors from BRCA mutation carriers. N Engl J Med. 2009;361:123-134.
18. Mateo J, Carreira S, Sandhu S, et al. DNA-repair defects and olaparib in metastatic prostate cancer. N Engl J Med. 2015;373:1697-1708.
19. Arce S, Athie A, Pritchard CC, et al. Germline and somatic defects in DNA repair pathways in prostate cancer. Adv Exp Med Biol. 2019;1210:279-300.
20. Mateo J, Porta N, Bianchini D, et al. Olaparib in patients with metastatic castration-resistant prostate cancer with DNA repair gene aberrations (TOPARP-B):
a multicentre, open-label, randomised, phase 2 trial. Lancet Oncol. 2020;21:162-174.
21. Abida W, Bryce AH, Vogelzang NJ, et al. Preliminary results from TRITON2: A phase II study of rucaparib in patients (pts) with metastatic castration-resistant
prostate cancer (mCRPC) associated with homologous recombination repair (HRR) gene alterations. Ann Oncol. 2018;29(suppl 8). Abstract 793PD.
22. Smith M, Sandhu S, Kelly W, et al. Pre-specified interim analysis of Galahad: a phase II study of niraparib in patients (pts) with metastatic castration-resistant
prostate cancer (mCRPC) and biallelic DNA-repair gene defects (DRD). Ann Oncol. 2019;30:v851-v934.
23. de Bono JS, Mehra N, Higano CS, et al. TALAPRO-1: a phase II study of talazoparib (TALA) in men with DNA damage repair mutations (DDRmut) and metastatic
castration-resistant prostate cancer (mCRPC)—first interim analysis (IA). J Clin Oncol. 2020;37:6s(suppl; abstr 119).
24. Hussain M, Daignault-Newton S, Twardowski PW, et al. Targeting androgen receptor and DNA repair in metastatic castration-resistant prostate cancer: results
from NCI 9012. J Clin Oncol. 2018;36:991-999.
25. Edwards SM, Kote-Jarai Z, Meitz J, et al; British Association of Urological Surgeons Section of Oncology. Two percent of men with early-onset prostate cancer
harbor germline mutations in the BRCA2 gene. Am J Hum Genet. 2003;72:1-12.
26. Agalliu I, Karlins E, Kwon EM, et al. Rare germline mutations in the BRCA2 gene are associated with early-onset prostate cancer. Br J Cancer. 2007;97:826-831.
27. Kote-Jarai Z, Leongamornlert D, Saunders E, et al; UKGPCS Collaborators. BRCA2 is a moderate penetrance gene contributing to young-onset prostate cancer:
implications for genetic testing in prostate cancer patients. Br J Cancer. 2011;105:1230-1234.
28. Risbridger GP, Taylor RA, Clouston D, et al. Patient-derived xenografts reveal that intraductal carcinoma of the prostate is a prominent pathology in BRCA2
mutation carriers with prostate cancer and correlates with poor prognosis. Eur Urol. 2015;67:496-503.
29. Porter LH, Lawrence MG, Ilic D, et al. Systematic review links the prevalence of intraductal carcinoma of the prostate to prostate cancer risk categories. Eur Urol.
2017;72:492-495.
30. Castro E, Goh C, Olmos D, et al. Germline BRCA mutations are associated with higher risk of nodal involvement, distant metastasis, and poor survival outcomes
in prostate cancer. J Clin Oncol. 2013;31:1748-1757.
31. Akbari MR, Wallis CJ, Toi A, et al. The impact of a BRCA2 mutation on mortality from screen-detected prostate cancer. Br J Cancer. 2014;111:1238-1240.
32. Liede A, Karlan BY, Narod SA. Cancer risks for male carriers of germline mutations in BRCA1 or BRCA2: a review of the literature. J Clin Oncol. 2004;
22:735-742.
33. Isaacsson Velho P, Silberstein JL, Markowski MC, et al. Intraductal/ductal histology and lymphovascular invasion are associated with germline DNA-repair gene
mutations in prostate cancer. Prostate. 2018;78:401-407.
34. Taylor RA, Fraser M, Livingstone J, et al. Germline BRCA2 mutations drive prostate cancers with distinct evolutionary trajectories. Nat Commun. 2017;8:13671.
35. Quigley DA, Dang HX, Zhao SG, et al. Genomic hallmarks and structural variation in metastatic prostate cancer. Cell. 2018;174:758-769.
36. Cheng HH, Pritchard CC, Montgomery B, et al. Prostate cancer screening in a new era of genetics. Clin Genitourin Cancer. 2017;15:625-628.
37. Castro E, Goh C, Leongamornlert D, et al. Effect of BRCA mutations on metastatic relapse and cause-specific survival after radical treatment for localised prostate
cancer. Eur Urol. 2015;68:186-193.
38. Lynch HT, de la Chapelle A. Genetic susceptibility to non-polyposis colorectal cancer. J Med Genet. 1999;36:801-818.
39. Thibodeau SN, Bren G, Schaid D. Microsatellite instability in cancer of the proximal colon. Science. 1993;260:816-819.
40. Peltomäki P, Aaltonen LA, Sistonen P, et al. Genetic mapping of a locus predisposing to human colorectal cancer. Science. 1993;260:810-812.
41. Drake JW, Charlesworth B, Charlesworth D, et al. Rates of spontaneous mutation. Genetics. 1998;148:1667-1686.
42. Hause RJ, Pritchard CC, Shendure J, et al. Classification and characterization of microsatellite instability across 18 cancer types. Nat Med. 2016;22:1342-1350.
43. Baretti M, Le DT. DNA mismatch repair in cancer. Pharmacol Ther. 2018;189:45-62.
44. Schumacher TN, Schreiber RD. Neoantigens in cancer immunotherapy. Science. 2015;348:69-74.
45. Vogelstein B, Papadopoulos N, Velculescu VE, et al. Cancer genome landscapes. Science. 2013;339:1546-1558.
46. Segal NH, Parsons DW, Peggs KS, et al. Epitope landscape in breast and colorectal cancer. Cancer Res. 2008;68:889-892.
47. Chalmers ZR, Connelly CF, Fabrizio D, et al. Analysis of 100,000 human cancer genomes reveals the landscape of tumor mutational burden. Genome Med.
2017;9:34.
48. Le DT, Uram JN, Wang H, et al. PD-1 blockade in tumors with mismatch-repair deficiency. N Engl J Med. 2015;372:2509-2520.
49. Le DT, Durham JN, Smith KN, et al. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science. 2017;357:409-413.
50. Boyiadzis MM, Kirkwood JM, Marshall JL, et al. Significance and implications of FDA approval of pembrolizumab for biomarker-defined disease. J Immunother
Cancer. 2018;6:35.

Carneiro et al
51. Antonarakis ES. A new molecular taxonomy to predict immune checkpoint inhibitor sensitivity in prostate cancer. Oncologist. 2019;24:430-432.
52. Pritchard CC, Morrissey C, Kumar A, et al. Complex MSH2 and MSH6 mutations in hypermutated microsatellite unstable advanced prostate cancer. Nat
Commun. 2014;5:4988.
53. Hempelmann JA, Lockwood CM, Konnick EQ, et al. Microsatellite instability in prostate cancer by PCR or next-generation sequencing. J Immunother Cancer.
2018;6:29.
54. Antonarakis ES, Piulats JM, Gross-Goupil M, et al. Pembrolizumab for treatment-refractory metastatic castration-resistant prostate cancer: multicohort, open-
label phase II KEYNOTE-199 study. J Clin Oncol. 2020;38:395-405.
55. Graff JN, Alumkal JJ, Drake CG, et al. Early evidence of anti-PD-1 activity in enzalutamide-resistant prostate cancer. Oncotarget. 2016;7:52810-52817.
56. Hansen AR, Massard C, Ott PA, et al. Pembrolizumab for advanced prostate adenocarcinoma: findings of the KEYNOTE-028 study. Ann Oncol. 2018;
29:1807-1813.
57. Boudadi K, Suzman DL, Anagnostou V, et al. Ipilimumab plus nivolumab and DNA-repair defects in AR-V7-expressing metastatic prostate cancer. Oncotarget.
2018;9:28561-28571.
58. Beer TM, Kwon ED, Drake CG, et al. Randomized, double-blind, phase III trial of ipilimumab versus placebo in asymptomatic or minimally symptomatic patients
with metastatic chemotherapy-naive castration-resistant prostate cancer. J Clin Oncol. 2017;35:40-47.
59. Lu X, Horner JW, Paul E, et al. Effective combinatorial immunotherapy for castration-resistant prostate cancer. Nature. 2017;543:728-732.
60. Padmanee S, Pachynski RK, Narayan V, et al. Initial results from a phase II study of nivolumab (NIVO) plus ipilimumab (IPI) for the treatment of metastatic
castration-resistant prostate cancer (mCRPC; CheckMate 650). J Clin Oncol. 2019;37:7s(suppl; abstr 142).
61. Fong PCC, Retz M, Drakaki A, et al. Keynote-365 cohort C: pembrolizumab (pembro) plus enzalutamide (enza) in abiraterone (abi)-pretreated patients (pts) with
metastatic castrate resistant prostate cancer (mCRPC). J Clin Oncol. 2019;37:7s(suppl; abstr 171).
62. Massard C, Retz M, Hammerer P, et al. Keynote-365 cohort b: pembrolizumab (pembro) plus docetaxel and prednisone in abiraterone (abi) or enzalutamide
(enza)-pretreated patients (pts) with metastatic castrate resistant prostate cancer (mCRPC). J Clin Oncol. 2019;37:7s(suppl; abstr 170).
63. Yu EY, Massard C, Retz M, et al. Keynote-365 cohort a: pembrolizumab (pembro) plus olaparib in docetaxel-pretreated patients (pts) with metastatic castrate-
resistant prostate cancer (mCRPC). J Clin Oncol. 2019;37:7s(suppl; abstr 145).
64. Zaffuto E, Pompe R, Zanaty M, et al. Contemporary incidence and cancer control outcomes of primary neuroendocrine prostate cancer: a SEER database
analysis. Clin Genitourin Cancer. 2017;15:e793-e800.
65. Aggarwal R, Huang J, Alumkal JJ, et al. Clinical and genomic characterization of treatment-emergent small-cell neuroendocrine prostate cancer: a multi-
institutional prospective study. J Clin Oncol. 2018;36:2492-2503.
66. Beltran H, Prandi D, Mosquera JM, et al. Divergent clonal evolution of castration-resistant neuroendocrine prostate cancer. Nat Med. 2016;22:298-305.
67. Beltran H, Tomlins S, Aparicio A, et al. Aggressive variants of castration-resistant prostate cancer. Clin Cancer Res. 2014;20:2846-2850.
68. Aparicio AM, Shen L, Tapia EL, et al. Combined tumor suppressor defects characterize clinically defined aggressive variant prostate cancers. Clin Cancer Res.
2016;22:1520-1530.
69. Hu J, Han B, Huang J. Morphologic spectrum of neuroendocrine tumors of the prostate: an updated review. Arch Pathol Lab Med. 2020;144:320-325.
70. Beltran H, Romanel A, Conteduca V, et al. Circulating tumor DNA profile recognizes transformation to castration-resistant neuroendocrine prostate cancer.
J Clin Invest. 2020;130:1653-1668.
71. Aggarwal RR, Quigley DA, Huang J, et al. Whole-genome and transcriptional analysis of treatment-emergent small-cell neuroendocrine prostate cancer
demonstrates intraclass heterogeneity. Mol Cancer Res. 2019;17:1235-1240.
72. Mu P, Zhang Z, Benelli M, et al. SOX2 promotes lineage plasticity and antiandrogen resistance in TP53- and RB1-deficient prostate cancer. Science. 2017;
355:84-88.
73. Zhang Y, Zheng D, Zhou T, et al. Androgen deprivation promotes neuroendocrine differentiation and angiogenesis through CREB-EZH2-TSP1 pathway in
prostate cancers. Nat Commun. 2018;9:4080.
74. Nava Rodrigues D, Casiraghi N, Romanel A, et al. RB1 heterogeneity in advanced metastatic castration-resistant prostate cancer. Clin Cancer Res. 2019;
25:687-697.
75. Chen WS, Aggarwal R, Zhang L, et al; West Coast Prostate Cancer Dream Team. Genomic drivers of poor prognosis and enzalutamide resistance in metastatic
castration-resistant prostate cancer. Eur Urol. 2019;76:562-571.
76. Corn PG, Heath EI, Zurita A, et al. Cabazitaxel plus carboplatin for the treatment of men with metastatic castration-resistant prostate cancers: a randomised,
open-label, phase 1-2 trial. Lancet Oncol. 2019;20:1432-1443.
77. Patel SP, Othus M, Chae YK, et al. A Phase II Basket Trial of Dual Anti-CTLA-4 and Anti-PD-1 Blockade in Rare Tumors (DART SWOG 1609) in Patients with
Nonpancreatic Neuroendocrine Tumors. Clin. Cancer Res. 2020;26:2290-2296.
78. Taplin ME, Hussain A, Shah S, et al. ProSTAR: a phase Ib/II study of CPI-1205, a small molecule inhibitor of EZH2, combined with enzalutamide (E) or
abiraterone/prednisone (A/P) in patients with metastatic castration-resistant prostate cancer (mCRPC). J Clin Oncol. 2019;37:7s (suppl; abstr TPS335) .
79. Puca L, Gavyert K, Sailer V, et al. Delta-like protein 3 expression and therapeutic targeting in neuroendocrine prostate cancer. Sci Transl Med. 2019;
11:eaav0891.

80. Aggarwal R, Mansfield A, Bletran H, et al. Preliminary safety and efficacy of rovalpituzumab tesirine in patients with delta-like protein 3-expressing advanced
solid tumors. Ann Oncol. 2017;28:v142-v157.
81. Bluemn EG, Coleman IM, Lucas JM, et al. Androgen receptor pathway-independent prostate cancer is sustained through FGF signaling. Cancer Cell. 2017;
32:474-489.
82. Sahin I, Mega AE, Carneiro BA. Androgen receptor-independent prostate cancer: an emerging clinical entity. Cancer Biol Ther. 2018;19:347-348.
83. Kruczek K, Ratterman M, Tolzien K, et al. A phase II study evaluating the toxicity and efficacy of single-agent temsirolimus in chemotherapy-naı̈ve castration-
resistant prostate cancer. Br J Cancer. 2013;109:1711-1716.
84. Armstrong AJ, Halabi S, Healy P, et al. Phase II trial of the PI3 kinase inhibitor buparlisib (BKM-120) with or without enzalutamide in men with metastatic
castration resistant prostate cancer. Eur J Cancer. 2017;81:228-236.
85. Carver BS, Chapinski C, Wongvipat J, et al. Reciprocal feedback regulation of PI3K and androgen receptor signaling in PTEN-deficient prostate cancer. Cancer
Cell. 2011;19:575-586.
86. Khoury JD, Wang WL, Prieto VG, et al. Validation of immunohistochemical assays for integral biomarkers in the NCI-MATCH EAY131 clinical trial. Clin Cancer
Res. 2018;24:521-531.
87. Lotan TL, Wei W, Ludkovski O, et al. Analytic validation of a clinical-grade PTEN immunohistochemistry assay in prostate cancer by comparison with PTEN FISH.
Mod Pathol. 2016;29:904-914.
88. Lotan TL, Heumann A, Rico SD, et al. PTEN loss detection in prostate cancer: comparison of PTEN immunohistochemistry and PTEN FISH in a large
retrospective prostatectomy cohort. Oncotarget. 2017;8:65566-65576.
89. de Bono JS, De Giorgi U, Rodrigues DN, et al. Randomized phase II study evaluating Akt blockade with ipatasertib, in combination with abiraterone, in patients
with metastatic prostate cancer with and without PTEN loss. Clin Cancer Res. 2019;25:928-936.
90. Kolinsky MP, Rescigno P, Bianchini D, et al. A phase I dose-escalation study of enzalutamide in combination with the AKT inhibitor AZD5363 (capivasertib) in
patients with metastatic castration-resistant prostate cancer. Ann Oncol. 2020;31:619-625.
91. Boormans JL, Korsten H, Ziel-van der Made AC, et al. E17K substitution in AKT1 in prostate cancer. Br J Cancer. 2010;102:1491-1494.
92. Hyman DM, Smyth LM, Donoghue MTA, et al. AKT inhibition in solid tumors with AKT1 mutations. J Clin Oncol. 2017;35:2251-2259.
93. Yan Y, Ma J, Wang D, et al. The novel BET-CBP/p300 dual inhibitor NEO2734 is active in SPOP mutant and wild-type prostate cancer. EMBO Mol Med. 2019;
11:e10659.
94. Yan Y, An J, Yang Y, et al. Dual inhibition of AKT-mTOR and AR signaling by targeting HDAC3 in PTEN- or SPOP-mutated prostate cancer. EMBO Mol Med.
2018;10:e8478.
95. Pilié PG, Gay CM, Byers LA, et al. PARP inhibitors: extending benefit beyond BRCA-mutant cancers. Clin Cancer Res. 2019;25:3759-3771.
96. Abida W, Campbell D, Patnaik A, et al. Non-BRCA DNA Damage Repair Gene Alterations and Response to the PARP Inhibitor Rucaparib in Metastatic
Castration-Resistant Prostate Cancer: Analysis From the Phase II TRITON2 Study. Clin Cancer Res. Epub 2020 Feb 21.
97. Maréchal A, Zou L. DNA damage sensing by the ATM and ATR kinases. Cold Spring Harb Perspect Biol. 2013;5:5.
98. Lavin MF. Ataxia-telangiectasia: from a rare disorder to a paradigm for cell signalling and cancer. Nat Rev Mol Cell Biol. 2008;9:759-769.
99. Fok JHL, Ramos-Montoya A, Vazquez-Chantada M, et al. AZD7648 is a potent and selective DNA-PK inhibitor that enhances radiation, chemotherapy and
olaparib activity. Nat Commun. 2019;10:5065.
100. Ménisser-de Murcia J, Mark M, Wendling O, et al. Early embryonic lethality in PARP-1 Atm double-mutant mice suggests a functional synergy in cell proliferation
during development. Mol Cell Biol. 2001;21:1828-1832.
101. Yamamoto K, Wang Y, Jiang W, et al. Kinase-dead ATM protein causes genomic instability and early embryonic lethality in mice. J Cell Biol. 2012;198:305-313.
102. Karnitz LM, Zou L. Molecular pathways: targeting ATR in cancer therapy. Clin Cancer Res. 2015;21:4780-4785.
103. Charrier JD, Durrant SJ, Golec JM, et al. Discovery of potent and selective inhibitors of ataxia telangiectasia mutated and Rad3 related (ATR) protein kinase as
potential anticancer agents. J Med Chem. 2011;54:2320-2330.
104. Reaper PM, Griffiths MR, Long JM, et al. Selective killing of ATM- or p53-deficient cancer cells through inhibition of ATR. Nat Chem Biol. 2011;7:428-430.
105. Bono JSD, Tan DSP, Caldwell R, et al. First-in-human trial of the oral ataxia telangiectasia and Rad3-related (ATR) inhibitor BAY 1895344 in patients (pts) with
advanced solid tumors. J Clin Oncol. 2019;37:15s(suppl; abstr 3007).
106. Bajrami I, Frankum JR, Konde A, et al. Genome-wide profiling of genetic synthetic lethality identifies CDK12 as a novel determinant of PARP1/2 inhibitor
sensitivity. Cancer Res. 2014;74:287-297.
107. Reimers MA, Yip SM, Zhang L, et al. Clinical outcomes in cyclin-dependent kinase 12 mutant advanced prostate cancer. Eur Urol. 2020;77:333-341.
108. Antonarakis ES, Velho PI, Agarwal N, et al. CDK12-altered prostate cancer: clinical features and therapeutic outcomes to standard systemic therapies, PARP
inhibitors, and PD1 inhibitors. Ann Oncol. 2019;30:v325-v55.
109. Antonarakis ES. Cyclin-dependent kinase 12, immunity, and prostate cancer. N Engl J Med. 2018;379:1087-1089.
110. Sokolova AO, Yu EY, Cheng HH. Honing in on PARPi Response in Prostate Cancer: from HR Pathway to Gene-by-Gene Granularity. Clin Cancer Res. Epub 2020
Mar 31.
111. Cardnell RJ, Feng Y, Diao L, et al. Proteomic markers of DNA repair and PI3K pathway activation predict response to the PARP inhibitor BMN 673 in small cell
lung cancer. Clin Cancer Res. 2013;19:6322-6328.

Carneiro et al
112. Ibrahim YH, Garcı́a-Garcı́a C, Serra V, et al. PI3K inhibition impairs BRCA1/2 expression and sensitizes BRCA-proficient triple-negative breast cancer to PARP
inhibition. Cancer Discov. 2012;2:1036-1047.
113. Juvekar A, Burga LN, Hu H, et al. Combining a PI3K inhibitor with a PARP inhibitor provides an effective therapy for BRCA1-related breast cancer. Cancer
Discov. 2012;2:1048-1063.
114. Philip CA, Laskov I, Beauchamp MC, et al. Inhibition of PI3K-AKT-mTOR pathway sensitizes endometrial cancer cell lines to PARP inhibitors. BMC Cancer.
2017;17:638.
115. van de Ven AL, Tangutoori S, Baldwin P, et al. Nanoformulation of Olaparib Amplifies PARP Inhibition and Sensitizes PTEN/TP53-Deficient Prostate Cancer to
Radiation. Mol Cancer Ther. 2017;16:1279-1289.
116. Matulonis UA, Wulf GM, Barry WT, et al. Phase I dose escalation study of the PI3kinase pathway inhibitor BKM120 and the oral poly (ADP ribose) polymerase
(PARP) inhibitor olaparib for the treatment of high-grade serous ovarian and breast cancer. Ann Oncol. 2017;28:512-518.
117. González-Billalabeitia E, Seitzer N, Song SJ, et al. Vulnerabilities of PTEN-TP53-deficient prostate cancers to compound PARP-PI3K inhibition. Cancer Discov.
2014;4:896-904.
118. Yazinski SA, Comaills V, Buisson R, et al. ATR inhibition disrupts rewired homologous recombination and fork protection pathways in PARP inhibitor-resistant
BRCA-deficient cancer cells. Genes Dev. 2017;31:318-332.
119. Golan T, Hammel P, Reni M, et al. Maintenance olaparib for germline BRCA-mutated metastatic pancreatic cancer. N Engl J Med. 2019;381:317-327.
120. O’Reilly EM, Lee JW, Zalupski M, et al. Randomized, multicenter, phase II trial of gemcitabine and cisplatin with or without veliparib in patients with pancreas
adenocarcinoma and a germline BRCA/PALB2 mutation. J Clin Oncol. 2020;38:1378-1388.
121. Silver DA, Pellicer I, Fair WR, et al. Prostate-specific membrane antigen expression in normal and malignant human tissues. Clin Cancer Res. 1997;3:81-85.
122. Pandit-Taskar N, O’Donoghue JA, Durack JC, et al. A phase I/II study for analytic validation of 89Zr-J591 ImmunoPET as a molecular imaging agent for
metastatic prostate cancer. Clin Cancer Res. 2015;21:5277-5285.
123. Tagawa ST, Milowsky MI, Morris M, et al. Phase II study of Lutetium-177-labeled anti-prostate-specific membrane antigen monoclonal antibody J591 for
metastatic castration-resistant prostate cancer. Clin Cancer Res. 2013;19:5182-5191.
124. Bander NH, Milowsky MI, Nanus DM, et al. Phase I trial of 177lutetium-labeled J591, a monoclonal antibody to prostate-specific membrane antigen, in patients
with androgen-independent prostate cancer. J Clin Oncol. 2005;23:4591-4601.
125. Tagawa ST, Vallabhajosula S, Christos PJ, et al. Phase 1/2 study of fractionated dose lutetium-177-labeled anti-prostate-specific membrane antigen monoclonal
antibody J591 (177 Lu-J591) for metastatic castration-resistant prostate cancer. Cancer. 2019;125:2561-2569.
177
126. Rahbar K, Ahmadzadehfar H, Kratochwil C, et al. German multicenter study investigating Lu-PSMA-617 radioligand therapy in advanced prostate cancer
patients. J Nucl Med. 2017;58:85-90.
127. Hofman MS, Violet J, Hicks RJ, et al. [177Lu]-PSMA-617 radionuclide treatment in patients with metastatic castration-resistant prostate cancer (LuPSMA trial):
a single-centre, single-arm, phase 2 study. Lancet Oncol. 2018;19:825-833.

POINTS OF VIEW
The section contains articles describing emerging, highly debated, or controversial topics in cancer research,
treatment, and care to benefit patients and the field of oncology.
ARTICLES
Prophylactic Cranial Irradiation for Small-Cell Lung Cancer: Time for a Reassessment
Martin J. Edelman

LUNG CANCER
points of view Prophylactic Cranial Irradiation for Small-Cell

Lung Cancer: Time for a Reassessment
Martin J. Edelman, MD1
overview
Prophylactic cranial irradiation (PCI) has been an accepted part of the management of both limited and small-
cell lung cancer; however, the data that support its use in limited-stage disease is based on an analysis of trials
done before currently accepted approaches to staging (i.e., brain MRI and/or PET scanning) were available.
For extensive disease, data are available from two randomized studies that are in direct conflict. This article
explores the basic rationale for PCI and the evidence indicating that it is time to readdress the question of its
routine use.
INTRODUCTION a new drug: Evidence should be provided in controlled

Prophylactic cranial irradiation has been considered clinical trials.
the standard of care for the treatment of limited-stage PCI IN LSCLC
small-cell lung cancer (LSCLC) since the 1990s and for Although it is frequently stated that PCI improves the
extensive-stage disease (ESCLC) since 2007. However, curative potential of LSCLC, no randomized trial has
the evidence that support its use is quite limited and ever demonstrated this to be the case. However,
primarily based on studies performed before current a meta-analysis by Auperin and colleagues2 (The
staging techniques or using staging that many would Prophylactic Cranial Irradiation Collaborative Group),
consider inadequate. In addition, PCI is associated on the basis of individual patient data from seven
with increased expense, inconvenience, and potential randomized trials, found that overall survival (OS) was
toxicity. This paper summarizes the key arguments and improved in patients who received PCI compared with
evidence regarding PCI. those who did not with a relative risk of 0.84 (p = .01). A
total of 987 patients were evaluated from seven ran-
RATIONALE FOR PCI domized trials that enrolled between 1977 and 1994.
Small-cell lung cancer (SCLC) was noted to be highly The relative risk for CNS metastasis was more striking
responsive to chemotherapy and radiotherapy in the at 0.46 (p , .001). There was no effect on the cu-
1960s. This was followed by the development of mulative incidence of other metastases or locoregional
potentially curative regimens combining chemo- relapse; therefore, the conclusion was that PCI re-
therapy and radiation in the 1980s, with up to 25% of duced the incidence of CNS metastases and that the
patients experiencing durable benefit. Accompa- improvement in OS was solely a consequence of su-
nying these developments was the growing re- perior control in the CNS. Absolute reduction in mor-
alization that many patients (30%–50%) would tality was 5.4%. It is important to note that the criteria
experience relapse in the central nervous system for entry, treatment, and assessment of patients en-
(CNS).1,2 Drawing on experience in pediatric acute rolled in these studies were highly variable: some
Author affiliations leukemia, investigators developed PCI with the ra- patients with ESCLC were enrolled, chemotherapy and
and support tionale that sterilization of the CNS, a sanctuary site radiotherapy regimens differed with some patients
information (if for SCLC, could improve survival in treated LSCLC. In receiving induction chemotherapy or radiation or the
applicable) appear addition, there is the position that the reduction of combination, and the methods of establishing re-
at the end of this
article.
CNS metastases is inherently beneficial as this will sponse (chest x-ray, CT, bronchoscopy) differed.
Accepted on
prevent potentially devastating complications. For The basic rationale for PCI in LSCLC is that the CNS will
February 25, 2020 PCI to be recommended, one or both of the following represent either the sole site of relapse or will seed
and published at must be demonstrated: overall survival is improved, other areas as the initial site of relapse. Available data
ascopubs.org on May or quality of life is improved.
18, 2020: DOI https://
do not strongly favor either of these assumptions, at
doi.org/10.1200/ The burden of proof is on those who advocate for this least for the majority of patients. The Auperin analysis
EDBK_281041 approach. The standard should be similar to that for explicitly states that there was “no effect on other

PCI in SCLC
publication of the meta-analysis, PET scans and CT/PET

scans have become part of the standard methods of eval-
uation for patients with SCLC. Unquestionably, a substantial
• The recommendation for PCI in patients with fraction of patients who were deemed to have limited disease
limited SCLC is based upon data obtained be-
in the 1980s would now be found to have ESCLC. A retro-
fore modern staging methods such as MRI or
spective study of 280 patients with LSCLC who achieved
PET.
a partial or complete response from Korea found that use of
• Although PCI signficantly decreases the risk of initial PET scanning for determining stage eliminated the
brain metastases in both limited and extensive
value of PCI. Risk of brain metastasis as the first site of
disease, its role in improving survival is
treatment failure was lower in patients who received PCI
less clear.
versus those who did not, with the cumulative incidence of
• At a minimum, the use of PCI should be intracranial metastasis being 25.4% versus 38.9%, re-
adapted to risk. The risk of CNS metastases is
spectively (p = .014). The reduced risk of brain metastasis
low in patients with resectable (i.e., node
in PCI-treated patients, however, was only apparent in the
negative ) LSCLC, and those patients should not
be advised to undergo PCI. subgroup of patients who had no initial PET imaging, with
a cumulative incidence of intracranial metastasis being
• The risk of neurocognitive sequelae of PCI is
13.3% versus 37.0%, respectively (p = .020), but not
high in patients who are older (. age 70) or who
in patients who had an initial PET scan (34.3% vs. 41.1%;
have underlying comorbidities that would in-
crease risk (e.g., early Alzheimer’s disease, etc.). p = .243).5
Another issue is that the nature of LSCLC has changed in
that many patients are now detected with much lower
metastases or on local or regional recurrences.” Arria- volumes/stage of disease as a consequence of the wide-
gada and colleagues3 analyzed the data from two ran- spread availability of CT or as part of low-dose CT screening
domized trials conducted in the 1980s and found that for lung cancer. Patients with stage I and II disease may now
freedom from all events—CNS or other relapse—in PCI be treated surgically and have a low risk of CNS disease. A
versus non-PCI groups was 11% versus 17% at 5 years. retrospective study from Shanghai that evaluated 349 pa-
This result is similar to the magnitude of benefit demon- tients found a low risk of CNS disease for those with localized
strated in the Auperin meta-analysis, but also indicates the disease and no benefit in terms of cerebral recurrence, nor
fundamental limitations of the approach. survival for those who were treated with PCI.6 At Memorial
Sloan Kettering Cancer Center, 283 patients were evalu-
The meta-analysis is of questionable utility in 2020. All
ated. There was a relatively low risk of failure in the CNS of
aspects of staging have changed. Most notably, no patient
patients with low-stage disease (I and/or II; 12%) versus
had CNS evaluation using MRI. MRI has clearly demon-
those with locally advanced disease (III; 26%).7
strated superiority to CT for the detection of CNS disease.4
Remarkably, a fraction of patients included in the meta-
PCI IN ESCLC
analysis were from the pre-CT era when the only method of
imaging the CNS would have been pneumoencephalography. Slotman and colleagues8 from the European Organization
In an analysis of 481 patients at one center in the Neth- for the Research and Treatment of Cancer (EORTC) per-
erlands, the prevalence of brain metastases in the CT formed a randomized trial of PCI with ESCLC (Table 1). Two
era was 10% versus 24% in the MRI era. All of those in the hundred eighty-six patients were randomly assigned. All
CT era were symptomatic, whereas approximately one patients were required to have a good response to che-
half of those detected by MRI were asymptomatic. The motherapy. A variety of PCI dose/schedules were used:
number of patients classified as having extensive disease 20 Gy in five or eight fractions, 24 Gy in 12 fractions, 25 Gy
also increased in the MRI era as patients were frequently in 10 fractions, or 30 Gy in 10 or 12 fractions. Biologically
upstaged on the basis of the detection of asymptomatic equivalent doses for these schedules range from 25 to 39
disease.4 Gy. The primary end point for the trial was the development
of symptomatic brain metastases. There was a substantial re-
It is a reasonable hypothesis that the advantage for PCI in
duction in the risk of symptomatic brain metastases (40.4%
the meta-analysis came from the treatment of disease that
to 14.6% at 1 year) and an improvement in OS (hazard
would now clearly be recognized and treated with whole-
ratio, 0.68). Median OS was 6.74 months in the PCI arm
brain radiotherapy, whereas patients without evidence of
versus 5.42 months in the control arm. One-year survival
disease could be spared PCI.
was 27.1% in the PCI arm versus 13.3% in the control arm.
It must also be recognized that this migration of staging was This study resulted in recommendations for PCI in patients
not only for the CNS, but also for systemic disease. Since the with extensive disease who experienced a response to

Martin J. Edelman
therapy. Given the paucity of progress in ESCLC, it has been patients in the Japanese trial received second-line che-
regarded as an important therapeutic advance and in- motherapy, which clearly can extend survival and reduce
corporated into several guidelines.9,10 symptoms.12 Therefore, it is not that either study is correct or
incorrect, but that they are each correct for the different
However, over time, substantial concerns with the general
populations studied.
applicability of this study have emerged. First, the defi-
nition of response to chemotherapy was not defined by MORBIDITY OF DISEASE VERSUS PCI
standard criteria (i.e., RECIST), nor the type of chemo-
It is important to note that virtually all studies of PCI in SCLC
therapy to be used. Second, and most importantly, patients have demonstrated reduction in CNS metastases com-
did not have mandatory CNS staging. The protocol man- pared with controls. Although it is intuitively attractive that
dated evaluation of the CNS only if patients were symp- superior control of CNS disease would lead to better quality
tomatic. It is well established that a substantial fraction of of life, this has not been prospectively demonstrated. In
patients with ESCLC who are neurologically intact will have fact, most of the existing data focus on attempting to
CNS disease. demonstrate that there is not a decrement of neuro-
Takahashi et al11 led a Japanese consortium that randomly cognitive function and consequent quality of life after PCI.
assigned 224 patients to PCI versus observation. In contrast A major problem is that PCI dose and fraction have varied
to the EORTC study, all patients were required to undergo considerably over the years, as have methods of assess-
MRI at baseline. Chemotherapy was specified as a platinum ment of neurocognition.
doublet. In addition, there was regular follow up with CNS In non–small-cell lung cancer, PCI studies in patients with
imaging by MRI. The PCI dose/scheduled was uniform: stage III disease treated with concurrent chemoradiotherapy,
25 Gy in 10 fractions. They found that PCI decreased the a population with similar demographics (e.g., age, smoking
incidence of CNS metastases, but that OS was not improved history) to patients with LSCLC, found that, although there
(hazard ratio, 1.24; Table 1). was a reduction of CNS disease with PCI, there was a clear
The difference in initial staging strategies is critical to un- cognitive difference that favored the control arm.13,14
derstanding and applying the data from these studies. Both of the randomized trials of PCI in ESCLC have dem-
These studies use PCI in two fundamentally different onstrated that PCI was associated with increased symp-
populations, those with baseline (i.e., before any therapy) toms. In the Japanese trial, PCI was associated with
imaging of the brain regardless of symptoms and those with a slightly greater degree of acute toxicity—mostly grade
symptom-directed imaging. Other differences between trials 2—with no difference in mini-mental status exams. The
may also have contributed to the differing results. Almost all EORTC trial evaluated global quality of life with two EORTC
TABLE 1. Comparison of EORTC and Japanese Trials of PCI

Variable PCI Observation
Takahashi et al n = 113 n = 111
OS 11.6 (HR, 1.27; p = .094) 13.7
1 year, % 48.4 53.6
2 years, % 15.0 18.8
Incidence of brain metastases, % 48.0 69.0
PFS, months 2.3 2.4
Second-line chemotherapy administered, % 88 89
EORTC (Slotman) n = 143 n = 143
OS (months) 6.7 (HR, 0.68; p = .003) 5.4
1 year, % 27.1 13.3
2 years, % Not reported Not reported
Incidence of brain metastases at 1 year (symptomatic), % 14.6 40.4
PFS, months 3.4 2.8
Second-line chemotherapy administered, % Not reported Not reported
Abbreviations: EORTC, European Organization for the Research and Treatment of Cancer; PCI, prophylactic cranial irradiation; HR, hazard ratio; OS, overall
survival; PFS, progression-free survival.

PCI in SCLC
FIGURE 1. MAVERICK (SWOG1827): MRI Brain Surveillance Alone Versus MRI Surveillance and Prophylactic Cranial Irradiation—A Randomized Phase III
Trial in Small-Cell Lung Cancer
Abbreviations: OS, overall survival; PFS, progression-free survival; QOL, quality of life; SRS, stereotactic radiosurgery; WBRT, whole-brain radiation therapy.
instruments. They found that PCI was associated a greater trial. The Southwest Oncology Group has initiated a trial that
severity of symptoms of fatigue, appetite loss, nausea, and will test PCI in the modern era (Fig. 1).
leg weakness for up to 3 months. There was no substantial An issue that is frequently ignored is the increasing efficacy
difference in global health status, role, and emotional and of systemic therapy for the treatment of CNS disease. This
cognitive functioning; however, in all those assessments, has been recognized for some time. An editorial by Larry
patients with PCI had numerically lower scores throughout Einhorn, MD, in 1995 noted that in long-term survivors,
most of the time periods assessed, some of which bordered those receiving less effective chemotherapy regimens
on significance. In short, the prevention of symptomatic may have a higher incidence of CNS disease and may
CNS disease did not translate into improved quality of life therefore potentially derive greater benefit with PCI. 17
or neurocognitive function.15 Standard cytotoxic agents have been demonstrated
It is possible that developments in radiation delivery may be for decades to have activity against established CNS
able to decrease the toxicity of PCI. Alteration of the dose disease in SCLC 18 ; therefore, as treatment of SCLC
and schedule of PCI can clearly decrease potential mor- becomes increasingly effective, the need for CNS pro-
bidity. In addition, emerging approaches, such as hippo- phylaxis may actually decline. Anti–programmed death-
campal sparing and the use of protective agents (e.g., 1/programmed death ligand-1 agents have recently
memantine), may also improve the therapeutic index.16 been demonstrated to be effective in ESCLC and are
However, in the absence of compelling improvement in undergoing evaluation in LSCLC. 19 Of note, in the PA-
survival, PCI needs to demonstrate superiority versus ob- CIFIC study for stage III non–small-cell lung cancer,
servation alone and treatment after development of disease anti–programmed death-1 therapy was associated with
in terms of symptoms and quality of life. a decreased risk of CNS recurrence.20
The SWOG trial will address the role of PCI in this era with
LOOKING FORWARD modern staging and therapy; however, as with all things in
The basic rationale for PCI is that there are some patients medicine, the field will change over time. An improving
who obtain excellent systemic control of disease and those understanding of the biology of CNS metastases may allow
who experience relapse in the CNS that leads to death. us to enrich the population of patients who are at risk for
Given the questions regarding this treatment and strong developing such disease and more effectively determine
opinions on both sides, it is clearly time for a randomized who should get PCI.

Martin J. Edelman
AFFILIATION AUTHOR’S DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

1
Fox Chase Cancer Center, Philadelphia, PA AND DATA AVAILABILITY STATEMENT
Disclosures provided by the author and data availability statement (if
EDBK_281041.
Martin J. Edelman, MD, Fox Chase Cancer Center, 333 Cottman Ave.,
Philadelphia, PA 19111; email: [email protected].
REFERENCES
1. Einhorn LH, Bond WH, Joe BT, et al. Long-term results in combined-modality treatment of small cell carcinoma of the lung. 1978;5(3):309–313 a. Semin Oncol.
1978;5:309-313.
2. Aupérin A, Arriagada R, Pignon JP, et al; Prophylactic Cranial Irradiation Overview Collaborative Group. Prophylactic cranial irradiation for patients with small-cell
lung cancer in complete remission. N Engl J Med. 1999;341:476-484.
3. Arriagada R, Le Chevalier T, Rivière A, et al. Patterns of failure after prophylactic cranial irradiation in small-cell lung cancer: analysis of 505 randomized patients.
Ann Oncol. 2002;13:748-754.
4. Seute T, Leffers P, ten Velde GP, et al. Detection of brain metastases from small cell lung cancer: consequences of changing imaging techniques (CT versus MRI).
Cancer. 2008;112:1827-1834.
5. Choi M, Lee Y, Moon SH, et al. Effect of accurate staging using positron emission tomography on the outcomes of prophylactic cranial irradiation in patients with
limited stage small-cell lung cancer. Clin Lung Cancer. 2017;18:77-84.
6. Xu J, Yang H, Fu X, et al. Prophylactic Cranial Irradiation for Patients with Surgically Resected Small Cell Lung Cancer. Journal of Thoracic Oncology. 2017;
12:347-363.
7. Wu AJ, Gillis A, Foster A, et al. Patterns of failure in limited-stage small cell lung cancer: implications of TNM stage for prophylactic cranial irradiation. Radiother
Oncol. 2017;125:130-135.
8. Slotman B, Faivre-Finn C, Kramer G, et al; EORTC Radiation Oncology Group and Lung Cancer Group. Prophylactic cranial irradiation in extensive small-cell lung
cancer. N Engl J Med. 2007;357:664-672.
9. Kalemkerian GP, Loo BW, Akerley W, et al. NCCN guidelines insights: small cell lung cancer, version 2.2018. J Natl Compr Canc Netw. 2018;16:1171-1182.
10. Jett JR, Schild SE, Kesler KA, et al. Treatment of Small Cell Lung Cancer: Diagnosis and Management of Lung Cancer, 3rd Ed: American College of Chest
Physicians Evidence-Based Clinical Practice Guidelines. In Chest, Vol. 143. 2013;e400S-e419S.
11. Takahashi T, Yamanaka T, Seto T, et al. Prophylactic cranial irradiation versus observation in patients with extensive-disease small-cell lung cancer: a mul-
ticentre, randomised, open-label, phase 3 trial. Lancet Oncol. 2017;18:663-671.
12. von Pawel J, Schiller JH, Shepherd FA, et al. Topotecan versus cyclophosphamide, doxorubicin, and vincristine for the treatment of recurrent small-cell lung
cancer. J Clin Oncol. 1999;17:658-667.
13. Sun A, Bae K, Gore EM, et al. Phase III trial of prophylactic cranial irradiation compared with observation in patients with locally advanced non-small-cell lung
cancer: neurocognitive and quality-of-life analysis. J Clin Oncol. 2011;29:279-286.
14. De Ruysscher D, Dingemans AC, Praag J, et al. Prophylactic cranial irradiation versus observation in radically treated stage III non-small-cell lung cancer:
a randomized phase III NVALT-11/DLCRG-02 study. J Clin Oncol. 2018;36:2366-2377.
15. Slotman BJ, Mauer ME, Bottomley A, et al. Prophylactic cranial irradiation in extensive disease small-cell lung cancer: short-term health-related quality of life and
patient reported symptoms: results of an international Phase III randomized controlled trial by the EORTC Radiation Oncology and Lung Cancer Groups. [erratum:
J Clin Oncol. 2009;27:1002] J Clin Oncol. 2009;27:78-84.
16. Robin TP, Rusthoven CG. Strategies to preserve cognition in patients with brain metastases: a review. Front Oncol. 2018;8:415.
17. Einhorn LH III. The case against prophylactic cranial irradiation in limited small cell lung cancer. Semin Radiat Oncol. 1995;5:57-60.
18. Lee JS, Murphy WK, Glisson BS, et al. Primary chemotherapy of brain metastasis in small-cell lung cancer. J Clin Oncol. 1989;7:916-922.
19. Horn L, Mansfield AS, Szcze˛ sna A, et al; IMpower133 Study Group. First-line atezolizumab plus chemotherapy in extensive-stage small-cell lung cancer. N Engl
J Med. 2018;379:2220-2229.
20. Antonia SJ, Villegas A, Daniel D, et al; PACIFIC Investigators. Durvalumab after chemoradiotherapy in stage III non-small-cell lung cancer. N Engl J Med. 2017;
377:1919-1929.

BREAST CANCER

BREAST CANCER
Genomic Alteration in Metastatic Breast Cancer

and Its Treatment
Allen Li, MD, MS1; Stephen M. Schleicher, MD, MBA2; Fabrice Andre, MD, PhD3; and Zahi I. Mitri, MD, MS1
overview
Metastatic breast cancer (mBC) remains responsible for the majority of breast cancer deaths. Whereas clinical
outcomes have improved with the development of novel therapies, resistance almost inevitably develops,
indicating the need for novel therapeutic approaches for the treatment of mBC. Recent investigations into
mBC genomic alterations have revealed novel and potential therapeutic targets. Most notably, therapies
against PIK3CA mutation and germline BRCA1/2 mutations have solidified the role of targeted therapy in
mBC, with treatments against these alterations now approved by the U.S. Food and Drug Administration (FDA)
on the basis of clinical benefit for patients with mBC. Familiarity with relevant genomic alterations in mBC,
technologies for mutation detection, methods of interpreting genomic alterations, and an understanding of
their clinical impact will aid practicing clinicians in the treatment of mBC as the field of breast oncology moves
toward the era of precision medicine.
INTRODUCTION Technologies, Seattle, WA). Gene expression profile

Breast cancer remains the most common cancer type assays allow for recurrence risk prediction beyond
among women in the United States, with an estimated clinical and pathologic presentations and refinement of
268,600 new cases in 2019 and 41,760 deaths.1 adjuvant therapy.
Despite advances in detection and treatment, breast Whereas early-stage breast cancer has clearly derived
cancer remains the second leading cause of death for benefit from gene-expression profiling, the role of tu-
women in the United States, with most deaths attrib- mor genomics in mBC is rapidly evolving. The recent
uted to the development of metastatic disease.2,3 With successes of phosphoinositide 3-kinase (PI3K) in-
increasing treatment options for mBC, familiarity with hibitor for the treatment of PIK3CA-mutated hormone
the genomic landscape of mBC and how to incorporate receptor (HR)–positive mBC and of PARP inhibitors in
tumor genomic findings into patient care is critical. deleterious germline BRCA1/2-mutated mBC have
solidified the role of genomic testing to guide therapy
Breast cancer is recognized as a group of heteroge-
for patients with mBC.11-13 There is also growing evi-
neous diseases with variable clinical courses and
dence to confirm tumor genomics’ significance in
treatment responses. In clinical practice, breast can-
treatment response and resistance with direct clinical
cers are classified according to estrogen receptor,
impact.14,15 It is important for practicing clinicians to be
progesterone receptor, and HER2 status in different
familiar with particular genomic mutations in mBC and
subtypes that guide prognostication and treatment
their clinical implications. We present an overview of
selection. It is now recognized that heterogeneity exists
genomic alterations in mBC, technologies for de-
beyond the expression of those receptors. Genomic
tection, methods of interpretation, and their impact on
alterations, gene expression profiles, histology, and
clinical practice.
Author affiliations immune markers, among others,4-9 bring a new un-
and support derstanding of breast cancer and offer both prognostic GENOMIC LANDSCAPE OF METASTATIC
information (if and predictive value.7,10,11 Of particular importance are BREAST CANCER
applicable) appear
the tumor genomic alterations and gene expression With advances in next-generation sequencing tech-
at the end of this
article. profiles that have changed the standard of care in nology, research efforts have focused on providing
Accepted on March early-stage disease. Commercially available genomic genomic characterization of mBC tumors. It is recog-
4, 2020 and assays that provide gene-expression profiles include nized that molecular changes can occur throughout
published at Oncotype DX (Genomic Health, Redwood City, CA), the disease course. For receptor status alone, alter-
ascopubs.org on EndoPredict (Myriad Genetics, Salt Lake City, UT), ations are found after neoadjuvant chemotherapy16-18
March 26, 2020:
Breast Cancer Index (Biotheranostics, San Diego, CA), and between matched primary and metastatic
10.1200/EDBK_ MammaPrint (Agendia, Amsterdam, the Netherlands), lesions,19-21 with discordance rates of 10% to 30%.20
280463 and Predictor Analysis of Microarray 50 (NanoString Discordance of gene expression between primary and

Genomics of Metastatic Breast Cancer
mutation is found in different frequencies among breast

cancer subtypes, with HR-positive breast cancer having the
highest frequency of 40% of tumors in both systemic and
• Understand sequencing tools and their limita- early-stage diseases.23,24,26 Other mutations along the PI3K/
tions for the detecting genomic alterations, in-
AKT pathway are also present in HR-positive mBC, in-
cluding plasma cell-free DNA.
cluding AKT1 and PTEN. Like PIK3CA, these mutations are
• Appreciate and interpret sequencing results found most commonly in HR-positive mBC. Of interest,
using systemic framework in clinical decision- unlike PIK3CA mutation, AKT1 and PTEN mutations are
making for the treatment of metastatic breast
found more commonly in the metastatic setting across all
cancer.
subtypes, suggesting that those mutations may be acquired
• Describe an overview of actionable mutations in through treatment and disease progression.
metastatic breast cancer and matched targeted
therapies. Other common genomic alterations in HR-positive breast
cancer, including CDH1, GATA3, KMT2C, MAP3K1,
• Highlight mutations of potential significance
MAP2K4, NF1, and ERBB2, have been implicated as potential
and existing evidence.
driver mutations, mechanisms of resistance, or prognostic
markers.28-31 Given that these mutations are enriched in mBC,
metastatic tumors has also been documented using the this may present opportunities for the development of novel
Predictor Analysis of Microarray 50 assay, with a 36.9% targeted therapy approaches for this patient population.
discordance rate.21 Similarly, genomic alterations among ERBB2 mutations, which encode for HER2 and are enriched
mBC exhibit different profiles than that of early-stage dis- in metastatic HR-positive breast cancer compared with early-
ease, including acquisition of driver genomic alterations not stage disease, is one example.15,23 Similarly, NF1 loss-of-
present in primary disease,22-25 revealing different potential function mutations, which induce mitogen-activated protein
treatment targets. Table 1 outlines the frequency of com- kinase pathway activity and resulting treatment resistance,32
monly altered oncogenic driver genes among mBC from are enriched in mBCs.23,24,26 TP53 mutations are also noted in
various studies and compares them with a prior study of The HR-positive breast cancer but in a lower frequency compared
Cancer Genome Atlas cohort in primary early-stage breast with other subtypes—approximately 20% in early-stage dis-
cancer. Table 1 reflects that genomic pathway alterations ease and 30% in metastatic disease.23,24,26
may be shared across breast cancer subtypes. Below, we
Genomic Alterations in Metastatic HER2-Positive
outline the frequency of genomic mutations in mBC by
Breast Cancer
clinical subtypes.
HER2 is encoded by ERBB2, and ERBB2 amplification is
Genomic Alterations in Metastatic HR-Positive a common feature of HER2-positive breast cancer. The sig-
Breast Cancer nificance of ERBB2 amplification has been well established
Overexpression of HR and activation of its downstream and predicts a benefit from HER2-directed therapies in mBC.
pathway are key features of HR-positive breast cancer. In addition to amplification, ERBB2 hotspot mutations are also
Antiestrogen therapy as monotherapy or in combination documented. Activating ERBB2 mutations can result in altered
remains the standard of care in the first-line setting for HR- cell growth and differentiation in addition to the activation of the
positive mBC. Metastatic HR-positive breast cancers are mitogen-activated protein kinase pathway.33 The clinical role of
enriched in the mutation of the estrogen receptor 1 gene, ERBB2 hotspot mutations remains to be established, espe-
ESR1, in approximately 20% of tumors compared with less cially as recent studies suggest they are enriched in mBC
than 5% in early-stage disease.15,23 ESR1 encodes estrogen compared with early-stage disease.34
receptor, and its mutation at the ligand-binding domain of Given that up to one-half of HER2-positive breast cancer
the estrogen receptor results in ligand-independent con- also expresses HRs at various levels, genomic alterations
stitutive estrogen receptor activity and activation of down- may be shared between these two patient populations.35
stream signaling, which is known to lead to resistance to Specifically, the incidence of ERBB2, NF1, ESR1,
endocrine-based therapies.14 PI3KCA, GATA3, and ATM mutations is similar across
PIK3CA mutation results in the activation of the PI3K/AKT both of these groups and seem to be enriched in mBC
pathway, leading to dysregulation of key cellular functions.27 compared with early-stage disease.23,24,26 TP53 muta-
PIK3CA mutations are found in similar frequencies in both tions are observed in HER2-positive breast cancer in
primary and metastatic lesions and are observed across all approximately 55% of both early-stage and metastatic
subtypes in various degrees.15,23 The similar mutation fre- disease.23,24,26
quency between disease stages is consistent with the role of Brain metastasis remains a clinical challenge in HER2-
PIK3CA as a driver in oncogenic transformation. PIK3CA positive mBC. A number of studies have noted that

Li et al
TABLE 1. Common Genomic Alterations in Breast Cancer

Metastatic Primary
24 23
Razavi et al Angus et al The Cancer Genome Atlas26
ER+/HER2 HER2+ TNBC ER+/HER2 HER2+ TNBC ER+/HER2 ER /HER2+ TNBC

Gene (n = 679) (n = 145) (n = 81) (n = 279) (n = 77) (n = 58) (n = 330) (n = 75) (n = 86)
PIK3CA 38.6 37.9 13.6 45.9 45.5 19.0 43.9 30.7 10.5
TP53 29.7 55.2 90.1 31.5 58.4 79.3 20.6 54.7 79.1
GATA3 18.7 11.0 0.0 13.6 10.4 5.2 13.6 10.7 0.0
CDH1 18.3 11.7 1.2 12.2 3.9 8.6 9.1 2.7 1.2
ESR1 17.4 6.9 0.0 19.0 13.0 0.0
KMT2C 10.9 5.5 2.5 10.8 11.7 10.3
MAP3K1 9.1 2.8 0.0 7.9 7.8 1.7 10.9 2.7 0.0
ARID1A 7.8 2.8 7.4 8.2 9.1 5.2
AKT1 7.1 0.0 2.5 7.2 3.9 1.7 3.3 1.3 1.2
PTEN 7.1 3.4 11.1 14.0 5.2 13.8 4.8 0.0 1.2
ERBB2a 6.3 6.9 0.0 5.4 6.5 5.2
TBX3 6.2 4.1 0.0 7.5 2.6 0.0 3.3 0.0 1.2
FOXA1 5.6 5.5 0.0 6.1 7.8 3.5
NCOR1 5.3 2.8 4.9 6.1 6.5 6.9 3.9 1.3 1.2
NF1 5.2 11.0 6.2 11.1 9.1 8.6 3.3 1.3 2.3
MAP2K4 4.3 2.1 0.0 7.9 10.4 1.7 5.8 1.3 1.2
BRCA2 4.0 3.4 2.5 6.1 10.4 3.5
ATM 3.7 4.8 3.7 6.1 6.5 3.5
CBFB 3.5 0.7 1.2 2.9 1.3 0.0 2.1 0.0 1.2
RUNX1 3.5 5.5 2.5 1.8 0.0 1.7 4.5 1.3 0.0
RB1 3.4 0.0 11.1 3.6 3.9 8.6 0.9 1.3 4.7
BRCA1 2.5 2.1 3.7 2.2 6.5 1.7
SF3B1 2.5 0.7 1.2 2.1 1.3 0.0
CTCF 2.4 0.0 0.0 3.3 1.3 1.2
PTPRD 1.9 4.1 4.9 2.4 2.7 2.3
CDKN1B 1.5 0.7 1.2 4.3 0.0 6.9 1.5 0.0 0.0
CHEK2 1.5 2.8 1.2
PALB2 1.3 0.7 2.5
GPS2 1.2 0.0 0.0 2.2 2.6 0.0 1.2 1.3 1.2
PIK3R1 1.2 2.8 7.4 2.7 5.3 1.2
NTRK1 0.9 0.7 1.2
NOTE. Data are presented as percentages.

Abbreviations: ER, estrogen receptor; HER2, human epidermal growth factor receptor 2; TNBC, triple-negative breast cancer.
a
Mutation only, excluding amplification.
HER2-positive mBC brain metastases harbor mutations Genomic Alterations in Metastatic Tripe-Negative
different from the primary tumor. These include BRAF, Breast Cancer
FGFR2, EGFR, and KIT, among others.36,37 These mutations Triple-negative breast cancer (TNBC), which is character-
may be potential novel therapeutic targets for brain me- ized by a lack of estrogen receptor, progesterone receptor,
tastasis in HER2-positive mBC. and HER2 expression, has a worse prognosis than other

breast cancer subtypes. Despite the recent approval of system. Conflicting results exist regarding the level of TMB
immune checkpoint blockade in metastatic TNBC, re- between primary and metastatic breast cancers. One study
sistance inevitably develops, and survival outcomes remain found mildly increased TMB in mBC compared with primary
poor.10,11 Genomic profiling of TNBC has revealed that tumors in HR-positive disease and no differences in other
several genetic alterations are present at higher prevalence subtypes. Another study suggested higher TMB in HR-
than in other breast cancer subtypes. TP53 tumor sup- positive and TNBC mBC compared with early-stage
pressor gene mutations are found in more than one-half of disease.22,24 TMB remains investigational in mBC.
metastatic TNBC tumors and reach up to more than 90% in
In individual patients, tumor genomic discrepancies may
certain reports.24 TP53 mutations are frequent in early-stage
exist between primary and metastatic tumors,23,47,48 the
TNBC as well, nearing 70% to 80%.23,26 Targeted therapy
primary tumor and brain metastases,36,37,49 before and after
that restores TP53 function remains an important unmet
treatment resistance, and even within a single tumor.50 The
need in oncology. Other than TP53, other mutations noted
genomic evolution may result from clonal selection from
include PIK3CA, RB1, and PTEN but at a much lower
therapeutic pressure, among other factors. Serial genomic
frequency.23,24,26
testing may be warranted in certain clinical settings, such as
The homologous recombination system has garnered at- at the time of disease progression on a certain therapy.
tention as a possible therapeutic target, especially in TNBC. Furthermore, whereas large-scale studies of genomic in-
Homologous recombination is critical in maintaining DNA cidence provide an overview, these studies differ in the type
integrity, especially from double-strand breaks. Germline of treatments received, sites and timing of biopsies, and
mutations of BRCA1/2, which are known to cause homol- sequencing methods. Current somatic mutation data in the
ogous recombination deficiency (HRD), are found in 10% to metastatic setting can inform particular targeted therapy,
20% of TNBC, and somatic mutations of BRCA1/2 are namely the use of PI3K inhibitors, as well as mechanisms of
found in 3% to 5% of TNBC.38 It is now recognized that resistance. Last, information on individual tumor genomic
BRCA1/2 wild-type tumors can still exhibit HRD as a result alterations provides the opportunity for patients to partici-
of other genetic or epigenetic mutations within the ho- pate in molecular- and genetic-driven clinical trials—for
mologous recombination machinery, termed BRCAness.39 instance, multiple clinical trials are ongoing for ESR1-
Multiple somatic mutations that can result in HRD have mutated breast cancer.51,52
been implicated, including BRCA1/2, RAD51, PALB2, ATR,
CHK1, WEE1, and PLK1, among others.39-41 These muta- CLINICAL DETECTION OF GENOMIC ALTERATIONS
tions are observed in all mBC subtypes but in relatively low Tissue Testing
frequencies. Unlike germline mutations, somatic mutation detection
Microsatellite Instability and Tumor Mutational Burden relies on sequencing of tumor DNA. Maturation of next-
generation sequencing (NGS) technology allows for the
In addition to specific genomic alterations, evaluation of timely and accurate sequencing of a large number of genes.
tumor microsatellite instability (MSI) that leads to defects in Different NGS platforms exist for a variety of applications,
DNA mismatch repair has become a standard of care in including whole-genome and whole-exome sequencing of
metastatic solid tumors. Patients with tumors that harbor tumors, targeted genome profiling, transcriptome/RNA se-
MSI are candidates for treatment with the immune quencing, and others. NGS for the detection of tumor ge-
checkpoint inhibitor pembrolizumab.42 Clinical testing for nomic mutations is now commonplace. The most common
MSI involves immunohistochemistry and polymerase chain clinical approach is targeted genome profiling, which se-
reaction testing for four proteins of the mismatch repair quences a preselected panel of oncogenes of interest,
pathway: MSH2, MSH6, MLH1, and PMS2. MSI has been particularly those with matched targeted therapies. Onco-
documented in breast cancer but at a lower frequency genic panel NGS testing is now available commercially and
compared with other tumors. One study reported that 0.9% in various academic settings. Examples of oncogene se-
of primary TNBC has MSI.43 Another study using a The quencing services include FoundationOne (Foundation
Cancer Genome Atlas cohort observed MSI in 1.53% of Medicine, Cambridge, MA), Oncomine (ThermoFisher
breast cancer of all subtypes.44 Scientific, Waltham, MA), CANCERPLEX (KEW, Cambridge,
Other than MSI, tumor mutational burden (TMB), which MA), Caris Molecular Intelligence (Caris Life Science, Irving,
quantifies the total number of somatic coding mutations in TX), Tempus (Tempus Labs, Chicago, IL), MSK-IMPACT
a tumor, is also a developing predictive biomarker for im- (Memorial Sloan Kettering Cancer Center, New York, NY),
mune checkpoint inhibitor therapy in oncology.45,46 TMB is and OmniSeq Advance (Roswell Park Cancer Institute,
presented as the number of mutations per megabase. In- Buffalo, NY). The MSK-IMPACT and FoundationOne assays
creased TMB is theorized to produce more tumor-specific currently carry FDA approval, and the Caris assay carries an
mutant epitopes that can be recognized by the immune FDA Breakthrough Device designation.

Li et al
Oncogenes included in the various panels may be disease Although advanced sequencing techniques have made the
site specific or generalized. The number of genes in a panel detection of key genomic alterations possible, before con-
may range from 20 to more than 400. MSI and TMB testing sidering testing, it is first critical to determine whether se-
are also often included in NGS oncogene panels. When quencing is clinically indicated and how results would affect
selecting a sequencing service, one must confirm that the treatment decisions. Although the cost of sequencing has
genes of interest are included in the panel. Furthermore, decreased, the financial ramifications must still be con-
caution is required when interpreting sequencing results. sidered. Furthermore, targeted therapy approvals are
Whereas the sequencing reports often provide insights, it is commonly accompanied by associated approved assays,
important to recognize that different mutations within which should receive priority in considering which assays to
a particular gene of interest can result in different functional use. Last, evidence associated with treatment decisions on
impact. For example, more than 10 different PIK3CA point the basis of genomic alteration discovered in sequencing
mutations have been documented; a number of them, like needs to be evaluated critically.
E542K, are known gain-of-function mutations that affect
tumor biology, whereas a number of other mutations are Interpretation of Genomic Mutations
either neutral or inconclusive.53 It is also important to rec-
With the widespread availability of commercial and research
ognize that targeted therapies that are beneficial in a par-
sequencing, interpreting genomic results may not be
ticular disease may not translate to other disease sites.
a simple task. Novel genomic alterations can be found
Besides tumor somatic mutations, incidental pathogenic
incidentally, and results can be complex, such as a different
germline variants may be identified through tumor se-
mutation of a known oncogene or a new gene altogether. As
quencing, which can carry ramifications for the patient and
mentioned above, it is important to distinguish neutral
family. One study reported the rate of incidental pathogenic
variants from pathologic mutations with clinical significance.
germline variant detection in tumor NGS to be 2.3%.54
Furthermore, sometimes multiple actionable mutations
Additional medical genetics consultation and potential
coexist with no available guidelines to help prioritize them.
germline testing should be offered when pathogenic
There is a need to systemically evaluate the importance of
germline variants are detected.
genomic mutations to overcome the challenges associated
Liquid Biopsies with interpreting genomic results.
Sequencing is traditionally performed on tumor tissue, One publicly available framework that helps with the in-
which is limited by sample availability and biopsy procedural terpretation of specific genomic alterations is OncoKB
risk. On the basis of these limitations, the utility of se- (http://oncokb.org),57 a comprehensive precision oncology
quencing cell-free tumor DNA (cfDNA) within the plasma knowledge base with evidence-based information on indi-
from lysed tumor cells is being explored. Targeted NGS vidual somatic mutations and structural alterations that
sequencing of plasma cfDNA has been proposed both as an integrates clinical, biologic, and therapeutic information
alternative to tumor tissue sequencing and as a complement along with guidelines and recommendations derived from
to increase detection rate. Sequencing cfDNA is well studied FDA labeling, National Comprehensive Cancer Network
in metastatic non–small cell lung cancer. A prospective guidelines, etc. The base is organized by gene, alteration,
study comparing tissue sequencing with cfDNA sequencing tumor type, and clinical implication. OncoKB uses a system
using a commercial cfDNA service (Guardant360; Guardant structured by different levels of evidence to consistently
Health) in patients with treatment-naı̈ve metastatic non– describe the clinical utility of individual mutations. Level 1
small cell lung cancer found a high concordance rate for the offers the strongest evidence, which includes specific gene
detection of actionable mutations.55 The sensitivity of cfDNA mutations that are recognized by the FDA with matched
sequencing can depend on the amount of tumor DNA FDA-approved therapy. In addition to actionable mutations,
shedding, which can be influenced by disease type, tumor OncoKB also classifies mutations that have been shown to
burden, and therapy. It has been suggested that cfDNA may confer resistance to specific targeted therapies into three
account for tumor heterogeneity by potentially capturing levels on the basis of the strength of evidence. Level re-
genomic alterations present at multiple sites compared with sistance 1 offers the strongest evidence and predicts re-
tissue sampling, which is limited from a single biopsy site.56 sistance to FDA-approved drugs, with examples like RAS
The screening kit for PIK3CA mutations using plasma mutation in metastatic colorectal cancers and EGFR T790M
cfDNA has received FDA approval; if the test is negative in mutation in metastatic non–small cell lung cancer. As of this
plasma, testing of tumor tissue should be pursued. Se- article, OncKB has annotated 668 genes with 5,125 al-
quencing of cfDNA can also be considered when se- terations across 46 tumor types, including 25 alterations of
quencing is indicated but tissue biopsy is risk prohibited or level 1, 14 of level 2, 28 of level 3, 20 of level 4, and 11 of
not feasible. level resistance 1/resistance 2.57

Similarly, the European Society for Medical Oncology genomic alterations in breast cancer ranked by ESCAT59
(ESMO) Translational Research and Precision Medicine and OncoKB as of January 2020.
Working Group has proposed a classification system of
molecular alterations based on the available evidence that CLINICAL IMPLICATIONS OF GENOMIC ALTERATIONS IN
allows for a framework of clinical interpretation and the METASTATIC BREAST CANCER
ranking of genomic alterations. The ESMO Scale for Clinical ERBB2 Amplification: Tier IA (ESCAT), Level 1 (OncoKB)
Actionability of Molecular Targets (ESCAT) classification is
Whereas NGS is able to detect ERBB2 amplification, it is
the framework designed to aid oncologists in prioritizing
more commonly determined in the clinical setting using
potential targets for clinical use when interpreting a gene
immunohistochemistry (IHC) or fluorescence in situ hy-
sequencing panel. The level of evidence in ESCAT is des-
bridization (when IHC is equivocal). ERBB2 amplification
ignated into tiers on the basis of existing regulatory ap-
predicts the benefit of HER2-directed therapy. Of note,
provals as well as clinical and preclinical data. Tier I is the
discordance between fluorescence in situ hybridization and
highest tier and indicates that a genetic target is suitable for
IHC exists. One recent study found that fluorescence in situ
routine clinical use with a specific therapy recommendation
hybridization may be negative for amplification in 15.9% of
based on the specific molecular alteration. The tiers are
the tumors with a HER2 IHC of 3+.60 Other mechanisms of
further divided by type of supporting evidences.58 Figure 1
HER2 overexpression exist, such as chromosome 17 an-
highlights the ranking systems by levels of evidence for
euploidy, which explains HER2 positivity without ERBB2
OncoKB and ESCAT in relation to standard of care.
amplification.61 Unlike ERBB2 mutations (discussed be-
Both OncoKB and ESCAT offer a consistent way by which to low), ERBB2 amplification indicates increased copy num-
classify mutations and provide associated evidence-based ber of the gene. Multiple clinical trials have confirmed the
clinical guidance. As new evidence becomes available, the role of different HER2-directed therapies in the treatment of
ranking of a given mutation may change with time. The web- ERBB2/HER2-amplified breast cancer across multiple set-
based nature of OncoKB allows for frequent updates and tings. Trastuzumab, pertuzumab, and docetaxel demon-
easy access. Numerous other evidence-based classifica- strated improved overall survival compared with trastuzumab
tions exist with different focuses. Table 2 lists the annotated plus docetaxel, making the regimen a standard first-line
FIGURE 1. OncoKB and ESMO

Scale for Clinical Actionability of
Molecular Targets Framework of
Evaluating Genomic Alterations
Ranked by Level of Evidence
Abbreviations: ESCAT, Euro-
pean Society for Medical On-
cology Scale for Clinical
Actionability of Molecular Tar-
gets; ESMO MCBS, European
Society for Medical Oncology
Magnitude of Clinical Benefit
Scale; FDA, U.S. Food and Drug
Administration.

Li et al
TABLE 2. Common mBC Genomic Alterations With Their Respective OncoKB and ESCAT Ranking
Genomic Alteration OncoKB Ranking57 ESCAT Ranking59 FDA-Approved Therapies in mBC
ERBB2 amplification 1 IA Variousa
Germline BRCA1/2 mutations 2 IA Olaparib, talazoparib
PIK3CA mutations 1 IA Alpelisib
Microsatellite instability 1 IC Pembrolizumab
NTRK translocations 1 IC Larotrectinib, entrecitinib
ESR1 mutations 3 IIA —
PTEN loss 4 IIA —
AKT1 mutations 3 IIB —
ERBB2 mutations 3 IIB —
MDM2 N/A IIIA —
Somatic BRCA1/2 mutations N/A IIIA —
ERBB3 mutations N/A IIIB —
ARID1A/B N/A IVA —
ATR/ATM/PALB2 4 IVA —
CDH1 N/A IVA —
IGF1R N/A IVA —
INPP4B loss N/A IVA —
MAP2K4/MAP3K1 N/A IVA —
MT4 N/A IVA —
MYC N/A IVA —
NF1 4 IVA —
PIK3R1 N/A IVA —
RUNX1/CBFB N/A IVA —
SF3B1 N/A IVA —
TP53 N/A IVA —
Abbreviations: ESCAT, European Society for Medical Oncology Scale for Clinical Actionability of Molecular Targets; FDA, U.S. Food and Drug Administration;
mBC, metastatic breast cancer; N/A, not applicable.
a
Trastuzumab, pertuzumab, lapatinib, neratinib, ado-trastuzumab emtansine, and fam-trastuzumab deruxtecan-nxki are approved in ERBB2-amplified mBC
in different settings.
therapy in ERBB2/HER2-amplified mBC.62 Additional estrogen receptor–dependent cellular signaling. The com-
HER2-directed therapies include the tyrosine kinase in- bination of PI3K inhibition plus fulvestrant is synergistic
hibitors lapatinib and neratinib63,64 and the antibody-drug in vivo, suggesting that PI3K inhibition in PIK3CA-mutated
conjugates trastuzumab emtansine and trastuzumab tumors can overcome resistance to antiestrogen therapies.70
deruxtecan-nxki, the latter being the most recent HER2 The phase III randomized SOLAR-I compared the combination
agent to receive FDA approval.65-67 An exhaustive overview of the PI3K inhibitor alpelisib and fulvestrant with fulvestrant
of HER2-directed therapy is beyond the scope of this article. alone in patients with HR-positive, HER2-negative mBC that
had progressed on prior endocrine therapy.71 This study re-
PIK3CA: Tier IA (ESCAT), Level 1 (OncoKB) ported that the combination of alpelisib and fulvestrant had
The PIK3CA gene codes for the catalytic subunit of PI3K.68 A superior progression-free survival (PFS) compared with ful-
gain-of-function mutation can cause the activation of vestrant alone, leading to its FDA approval in HR-positive
multiple downstream signaling cascades, including the mBC.12 These findings stress the importance of clinical test-
PI3K/AKT/mTOR pathway that promotes cell survival and ing for PI3KCA mutation for patients with HR-positive mBC
proliferation.69 Initial preclinical studies found that, in who experience progression on first-line endocrine therapy. In
PIK3CA-mutated breast cancers, PI3K inhibitors had only addition to alpelisib, other novel PI3K inhibitors are in various
modest single-agent activity but significantly induced stages of development.72,73

Of interest, PI3KCA mutations have not been found to be partners and may not all be oncogenic. Cross-referencing
predictive of response for all inhibitors of the PI3K/AKT/ with a database like OncoKB to confirm fusion activity is
mTOR pathway. A phase II study of the PI3K inhibitor important. Furthermore, hotspot mutations of NTRK1
taselisib in combination with fulvestrant in patients with HR- G595R and NTRK3 G623R, level resistance 2 (OncoKB),
positive, HER2-negative disease found that clinical activity carry likely resistance to larotrectinib.78 The frequency of
was independent of PIK3CA mutation status.74 Similar re- NTRK fusions in mBC is low; one study that examined
sults were reported in the FAKTION trial that evaluated the 12,214 consecutive patients with mBC found that 0.13% of
AKT inhibitor capivasertib in combination with fulvestrant in tumors harbored NTRK gene fusions.81 When considering
HR-positive mBC, in which improvement in PFS was irre- therapy options for mBC with NTRK fusions, optimal
spective of PI3K mutation status.75 In addition, a retro- treatment sequencing is undetermined.
spective analysis of PIK3CA mutation status in cfDNA in the
cohort of the BOLERO-2 trial, which examined the mTOR BRCA1/2: Germline Tier IA, Somatic Tier IIIA (ESCAT),
inhibitor everolimus in combination with exemestane versus Level 2 (OncoKB)
placebo plus exemestane for patients with HR-positive, Germline mutation of BRCA1/2 results in HRD. PARP en-
HER2-negative mBC, found that the PFS benefit from zymes are essential for DNA single-strand break repair.
everolimus was maintained irrespective of the PIK3CA Tumors with germline HRD rely more heavily on PARP
mutation status.76 enzymes for DNA repair8,82; therefore, inhibition of PARP
MSI: Tier IC (ESCAT), Level 1 (OncoKB) enzymes leads to persistence of DNA single-strand breaks
and eventual cell death through synthetic lethality. The
Pembrolizumab, a PD-1 antibody, has received tumor-
randomized, controlled, phase III OlympiAD trial compared
agnostic approval for advanced solid tumors with MSI, in-
olaparib, a PARP inhibitor, with single-agent chemotherapy
cluding mBC. Approval was based on data from 149 pa-
of physician’s choice in patients with HER2-negative mBC
tients with MSI-solid tumors who were enrolled across five
that harbored a germline BRCA1/2 mutation. The majority of
uncontrolled, multicohort, multicenter, single-arm clinical
patients in the study had received prior chemotherapy for
trials, with the majority of patients having colorectal cancer.
their metastatic disease. The trial demonstrated significant
Among patients included, two had mBC and both achieved
improvement in PFS in the olaparib group compared with
a partial response (durations of 7.6 months and 15.9
the chemotherapy group (7.0 months vs. 4.2 months;
months).42 As mentioned above, MSI incidence in mBC is
hazard ratio, 0.58; p , .001). The ORR in the olaparib group
low compared with other solid tumors. Pembrolizumab is
was 59.9% compared with 28.8% in the chemotherapy
a reasonable treatment option in patients with MSI mBC;
group.11 On the basis of this study, olaparib received FDA
however, with multiple treatments available in mBC, optimal
approval for the treatment of HER2-negative mBC with
sequencing of therapy must be further explored.
germline BRCA1/2 mutations in patients who were pre-
NTRK Fusion: Tier IC (ESCAT), Level 1 (OncoKB) viously treated with chemotherapy in the neoadjuvant,
NTRK1, NTRK2, and NTRK3 genes encodes the three adjuvant, or metastatic setting.83 Similarly, the randomized,
transmembrane tropomyosin receptor kinase (Trk) receptor controlled, phase III EMBRACA trial compared the PARP
proteins, Trk A, B, and C (TrkA, TrkB, and TrkC). Fusion of inhibitor talazoparib with single-agent chemotherapy of
NTRK induces constitutively active protein function, physician’s choice in patients with mBC harboring germline
resulting in an oncogenic driver.77 Larotrectinib, a TRK BRCA1/2 mutations. Approximately 35% of patients re-
inhibitor, has gained FDA approval for the treatment of solid ceived talazoparib in the first-line setting, with the majority
tumors that harbor a NTRK gene fusion. The approval was receiving one or two lines of prior systemic therapy for their
based on an early-phase clinical trial of 55 patients with metastatic disease. The median PFS was significantly longer
advanced solid tumors harboring a NTRK fusion, of which in the talazoparib arm than in the chemotherapy arm
one patient had mBC. Whereas the independent review of (8.6 months vs. 5.6 months; hazard ratio, 0.54; p , .0001).13
the overall response rate (ORR) was 75%, the one patient On the basis of this study, talazoparib received FDA ap-
with mBC did not achieve a response.78 Another TRK in- proval for the treatment of HER2-negative mBC with
hibitor, entrectinib, is also approved for treatment of NTRK- germline BRCA1/2 mutations, regardless of prior chemo-
positive solid tumors. Pooled analysis of ALKA, STARTRK-1, therapy use.84 Indirect comparison of the two PARP in-
and STARTRK-2 single-arm studies in patients with NTRK hibitors approved suggest similar benefits with different
gene fusion includes 11% (6 patients) with mBC and an adverse effect profiles.85 PARP inhibitors offer a targeted
ORR of 83%.79 Of note, among breast cancer subtypes, therapy option for patients with mBC with germline BRCA1/
NTRK fusions are found most commonly among secretory 2 mutations.
breast carcinoma and mammary analog secretory carci- The role of PARP inhibitors for the treatment of mBC with
noma.80 Furthermore, NTRK fusion has multiple fusion somatic BRCA1/2 mutations is under investigation. Studies

Li et al
of PARP inhibitors used in ovarian cancer suggest similar breast cancer.99 Patients with ESR1 mutations receiving
therapeutic benefit in patients with both germline and so- fulvestrant had improved PFS compared with those re-
matic BRCA mutations.86-88 Furthermore, one study found ceiving exemestane (hazard ratio, 0.52; 95% CI, 0.30 to
that niraparib, a PARP inhibitor, has demonstrated activity 0.92). Patients with wild-type ESR1 had similar PFS re-
not only in patients with germline BRCA1/2 mutations but gardless of treatment arm.100 In the PALOMA3 trial, which
also in those with HRD-positive tumors with wild-type BRCA compared the combination of palbociclib and fulvestrant
as well as those with somatic mutations.89 Multiple clini- with fulvestrant alone in HR-positive mBC, there was no
cal trials (ClinicalTrials.gov identifiers: NCT03990896, difference in PFS between ESR1 wild-type and ESR1-
NCT03286842, and NCT04053322) examining the efficacy mutated mBCs.100 Currently, there are no targeted thera-
of PARP inhibitors in mBC beyond germline BRCA1/2 pies for ESR1-mutated mBC, but ESR1 status may inform
mutation are underway. Enrollment in these trials should be tumor resistance to aromatase inhibitors. Optimal treatment
encouraged to understand the role of PARP inhibitors for the of ESR1-mutated mBC is under investigation using novel
treatment of mBC carrying HRD and somatic BRCA endocrine therapies, including trials investigating novel
mutations. nonsteroidal selective estrogen receptor modulators, such
as lasofoxifene versus fulvestrant, for the treatment of HR-
ERBB2 Mutation: Tier IIB (ESCAT), Level 3 (OncoKB)
positive mBC with acquired ESR1 mutation (ClinicalTrials.
Multiple activating ERBB2 mutations have been found in all gov identifier: NCT03781063).
mBC subtypes and are likely driver mutations in mBC.
Mutations may affect the extracellular domain, trans- PTEN Mutation: Tier IIA (ESCAT), Level 4 (OncoKB)
membrane domain, or tyrosine kinase domains, leading to Oncogenic PTEN mutations that result in loss of function
activation of HER2 signaling pathways, even without ERBB2 lead to downstream signaling of the AKT/mTOR pathway,
amplification.90-92 In vitro studies found that certain kinase altering cellular functions. Capivasertib (AZD5363), an AKT
domain mutations and extracellular mutations that lead to inhibitor, plus paclitaxel was evaluated against paclitaxel
enhanced HER2 signaling may have increased sensitivity to alone in a phase II randomized trial (PAKT trial) for the
HER2-directed therapy.92,93 Different ERBB2 mutations at treatment of patients with treatment-naı̈ve metastatic TNBC.
the tyrosine kinase inhibitor binding site, like L755S, has In that study, the combination of capivasertib plus paclitaxel
been linked to tyrosine kinase inhibitor resistance in vitro.94 had a more pronounced impact on PFS in PIK3CA/AKT1/
A phase II trial examining neratinib, an irreversible pan-HER PTEN-altered tumors (hazard ratio, 0.30; PFS, 9.3 months
tyrosine kinase inhibitor, for the treatment of ERBB2- vs. 3.7 months) compared with the intent-to-treat pop-
mutated, nonamplified mBC in heavily pretreated patients ulation (hazard ratio, 0.74; PFS, 5.9 months vs. 4.2
demonstrated a 36% clinical benefit rate.95 Similarly, an- months).101 Of note, this study also showed that mutations
other phase II trial examining the activity of neratinib in along this pathway are not mutually exclusive.101 Another
ERBB2-mutated mBC reported a clinical benefit rate of AKT inhibitor, ipatasertib, plus paclitaxel was evaluated in
31% (90% CI, 13% to 55%), including one complete re- a phase II randomized, placebo-controlled, double-blind
sponse, one partial response, and three stable diseases.96 A phase II trial (LOTUS trial) in locally advanced or meta-
phase II multihistology trial (SUMMIT) examined neratinib static TNBC in the first-line setting. Ipatasertib with pacli-
for the treatment of advanced solid tumors with somatic taxel demonstrated a more pronounced impact on PFS in
ERBB2/3 mutations. The trial included 25 patients with PIK3CA/AKT/PTEN-altered tumors (hazard ratio, 0.44; PFS,
mBC with ERBB2 mutations and reported an ORR of 32% 9.0 months vs. 4.9 months) compared with the intent-to-
(95% CI, 15% to 54%).97 Currently, neratinib has FDA treat population (hazard ratio, 0.60; PFS, 6.2 months vs. 4.9
approval only for the treatment of early-stage breast cancer months).102 The question remains as to which alteration
with ERBB2/HER2 amplification. Clinical trials investigating along this pathway is most oncogenic and whether the
the significance of ERBB2 mutation in mBC and its ap- different alterations benefit equally from targeted therapy.
propriate targeted therapy use should be encouraged.
AKT1 Mutations: Tier IIB (ESCAT), Level 3 (OncoKB)
ESR1 Mutation: Tier IIA (ESCAT), Level 3A (OncoKB) As mentioned above, capivasertib (AZD5363) plus pacli-
Different activating ESR1 mutations can be found in mBC, taxel has activity in PIK3CA/AKT1/PTEN-altered metastatic
especially after prior exposure to antiestrogen therapy, with TNBC.101 Capivasertib was also evaluated in advanced solid
a few selected mutations resulting in increased resistance to tumors harboring AKT1 E17K activating mutations. The trial
endocrine therapy.98 The impact of plasma-detected ESR1 included 20 patients with HR-positive mBC in addition to
mutation on treatment outcomes was examined in two patients with other cancer types, with a median of five lines
phase III clinical trials, SoFEA and PALOMA3. In the SoFEA of prior therapy. The median PFS was 5.5 months (95% CI,
trial, fulvestrant with or without anastrozole was compared 2.9 months to 6.9 months) in patients with HR-positive
with exemestane for the treatment of advanced, HR-positive mBC.103 Clinical trials investigating other AKT inhibitors and

their combinations are underway. Ipatasertib, an AKT in- may contribute to HRD, such as ATR/ATM/PALB2. Their role
hibitor, in combination with paclitaxel is being investigated in disease management remains investigational only.
for the treatment of patients with PIK3CA/AKT1/PTEN- CONCLUSION
altered advanced breast cancer (ClinicalTrials.gov identifier:
NCT03337724). Another study evaluating ipatasertib plus Approvals for alpelisib and PARP inhibitors for the treatment
palbociclib and fulvestrant versus placebo plus palbociclib of mBC with PIK3CA mutations and germline BRCA1/2
and fulvestrant in HR-positive, HER2-negative mBC is mutations, respectively, have highlighted the importance of
also underway (IPATunity150; ClinicalTrials.gov identifier: incorporating mBC genomic profiles into treatment con-
NCT04060862). The scientific premise of AKT inhibitors for sideration. Tumor-agnostic treatment indications for MSI
mBC harboring genetic lesions within the PI3K/AKT/mTOR and NTRK fusions provide additional treatment options for
pathway is sound. Enrollment of clinical trials investigating patients whose disease harbors those genetic alterations.
their efficacy should be considered for those patients. Whereas oncogene sequencing assays are readily available,
their application and result interpretation require caution.
Other ESCAT Tier III and IV Genetic Alterations Recognizing that current frontline treatments for mBC are
MDM2 amplifications and ERBB3 mutations are tier III based on well-founded evidence, tumor genomic testing
genomic mutations for mBC; tier III mutations either have should be performed in the appropriate setting to help guide
a matched drug-alteration with clinical benefit in another therapy. Furthermore, optimal therapy sequencing for the
tumor type or belong to the same families of tier I genes but treatment of mBC harboring actionable mutations remains
lack associated clinical evidence.58,59 MDM2 negatively reg- undefined. Last, recent studies of the genomic landscape of
ulates p53 and is found to be amplified in approximately one- mBC have provided additional insight into mBC biology and
half of liposarcomas. MDM2 inhibitors in the setting of lip- revealed other potential therapeutic targets that require
osarcoma are being investigated in early-phase clinical additional investigation. Many of those genomic targets are
trials.104 ERBB3 encodes for HER3; its clinical significance promising with supportive evidences. Enrollment of bio-
and response to inhibitors require more investigation. A host of marker- and genomic-driven clinical trials should be en-
ESCAT tier IV genomic alterations exist, including those that couraged when appropriate.
AFFILIATIONS CORRESPONDING AUTHOR

1
Department of Hematology Oncology, Oregon Health & Science Zahi I. Mitri, MD, MS, Oregon Health & Science University, 3181 SW Sam
University, Knight Cancer Institute, Portland, OR Jackson Park Rd., Mail Code: OC14HO, Portland, OR 97239; email:
2
Tennessee Oncology, OneOncology, Nashville, TN [email protected].
3
Gustave Roussy Cancer Center, Université Paris Saclay, Villejuif, France
AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
EDBK_280463.
REFERENCES
1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin. 2019;69:7-34.
2. Early Breast Cancer Trialists’ Collaborative Group (EBCTCG). Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year
survival: an overview of the randomised trials. Lancet. 2005;365:1687-1717.
3. Soni A, Ren Z, Hameed O, et al. Breast cancer subtypes predispose the site of distant metastases. Am J Clin Pathol. 2015;143:471-478.
4. Turner N, Tutt A, Ashworth A. Hallmarks of ‘BRCAness’ in sporadic cancers. Nat Rev Cancer. 2004;4:814-819.
5. Lehmann BD, Jovanović B, Chen X, et al. Refinement of triple-negative breast cancer molecular subtypes: implications for neoadjuvant chemotherapy selection.
PLoS One. 2016;11:e0157368.
6. Rampurwala M, Wisinski KB, O’Regan R. Role of the androgen receptor in triple-negative breast cancer. Clin Adv Hematol Oncol. 2016;14:186-193.
7. Chan JJ, Tan TJY, Dent RA. Are there any clinically relevant subgroups of triple-negative breast cancer in 2018? J Oncol Pract. 2018;14:281-289.
8. Lehmann BD, Bauer JA, Chen X, et al. Identification of human triple-negative breast cancer subtypes and preclinical models for selection of targeted therapies.
J Clin Invest. 2011;121:2750-2767.
9. Liu MC, Pitcher BN, Mardis ER, et al. PAM50 gene signatures and breast cancer prognosis with adjuvant anthracycline- and taxane-based chemotherapy:
correlative analysis of C9741 (Alliance). NPJ Breast Cancer. 2016;2:15023.

Li et al
10. Schmid P, Adams S, Rugo HS, et al; IMpassion130 Trial Investigators. Atezolizumab and nab-paclitaxel in advanced triple-negative breast cancer. N Engl
J Med. 2018;379:2108-2121.
11. Robson ME, Tung N, Conte P, et al. OlympiAD final overall survival and tolerability results: olaparib versus chemotherapy treatment of physician’s choice in
patients with a germline BRCA mutation and HER2-negative metastatic breast cancer. Ann Oncol. 2019;30:558-566.
12. André F, Ciruelos E, Rubovszky G, et al; SOLAR-1 Study Group. Alpelisib for PIK3CA-mutated, hormone receptor-positive advanced breast cancer. N Engl
J Med. 2019;380:1929-1940.
13. Litton JK, Rugo HS, Ettl J, et al. Talazoparib in patients with advanced breast cancer and a germline BRCA mutation. N Engl J Med. 2018;379:753-763.
14. Robinson DR, Wu Y-M, Vats P, et al. Activating ESR1 mutations in hormone-resistant metastatic breast cancer. Nat Genet. 2013;45:1446-1451.
15. Chandarlapaty S, Chen D, He W, et al. Prevalence of ESR1 mutations in cell-free DNA and outcomes in metastatic breast cancer: a secondary analysis of the
BOLERO-2 clinical trial. JAMA Oncol. 2016;2:1310-1315.
16. van de Ven S, Smit VT, Dekker TJ, et al. Discordances in ER, PR and HER2 receptors after neoadjuvant chemotherapy in breast cancer. Cancer Treat Rev. 2011;
37:422-430.
17. Parinyanitikul N, Lei X, Chavez-MacGregor M, et al. Receptor status change from primary to residual breast cancer after neoadjuvant chemotherapy and
analysis of survival outcomes. Clin Breast Cancer. 2015;15:153-160.
18. Niikura N, Tomotaki A, Miyata H, et al. Changes in tumor expression of HER2 and hormone receptors status after neoadjuvant chemotherapy in 21,755 patients
from the Japanese breast cancer registry. Ann Oncol. 2016;27:480-487.
19. Lindström LS, Karlsson E, Wilking UM, et al. Clinically used breast cancer markers such as estrogen receptor, progesterone receptor, and human epidermal
growth factor receptor 2 are unstable throughout tumor progression. J Clin Oncol. 2012;30:2601-2608.
20. Schrijver WAME, Suijkerbuijk KPM, van Gils CH, et al. Receptor conversion in distant breast cancer metastases: a systematic review and meta-analysis. J Natl
Cancer Inst. 2018;110:568-580.
21. Lluch A, González-Angulo AM, Casadevall D, et al. Dynamic clonal remodelling in breast cancer metastases is associated with subtype conversion. Eur J Cancer.
2019;120:54-64.
22. Bertucci F, Ng CKY, Patsouris A, et al. Genomic characterization of metastatic breast cancers [published correction appears in Nature. 2019;572:E7]. Nature.
2019;569:560-564.
23. Angus L, Smid M, Wilting SM, et al. The genomic landscape of metastatic breast cancer highlights changes in mutation and signature frequencies. Nat Genet.
2019;51:1450-1458.
24. Razavi P, Chang MT, Xu G, et al. The genomic landscape of endocrine-resistant advanced breast cancers. Cancer Cell. 2018;34:427-438.e6.
25. Yates LR, Knappskog S, Wedge D, et al. Genomic evolution of breast cancer metastasis and relapse. Cancer Cell. 2017;32:169-184.e7.
26. Cancer Genome Atlas Network. Comprehensive molecular portraits of human breast tumours. Nature. 2012;490:61-70.
27. Stemke-Hale K, Gonzalez-Angulo AM, Lluch A, et al. An integrative genomic and proteomic analysis of PIK3CA, PTEN, and AKT mutations in breast cancer.
Cancer Res. 2008;68:6084-6091.
28. Gustin JP, Miller J, Farag M, et al. GATA3 frameshift mutation promotes tumor growth in human luminal breast cancer cells and induces transcriptional changes
seen in primary GATA3 mutant breast cancers. Oncotarget. 2017;8:103415-103427.
29. Gala K, Li Q, Sinha A, et al. KMT2C mediates the estrogen dependence of breast cancer through regulation of ERα enhancer function. Oncogene. 2018;
37:4692-4710.
30. Xue Z, Vis DJ, Bruna A, et al. MAP3K1 and MAP2K4 mutations are associated with sensitivity to MEK inhibitors in multiple cancer models. Cell Res. 2018;
28:719-729.
31. Corso G, Veronesi P, Sacchini V, et al. Prognosis and outcome in CDH1-mutant lobular breast cancer. Eur J Cancer Prev. 2018;27:237-238.
32. Creighton CJ, Hilger AM, Murthy S, et al. Activation of mitogen-activated protein kinase in estrogen receptor alpha-positive breast cancer cells in vitro induces an
in vivo molecular phenotype of estrogen receptor alpha-negative human breast tumors. Cancer Res. 2006;66:3903-3911.
33. Kamaruzman NI, Aziz NA, Poh CL, et al. Oncogenic signaling in tumorigenesis and applications of siRNA nanotherapeutics in breast cancer. Cancers (Basel).
2019;11:E632.
34. Christgen M, Bartels S, Radner M, et al. ERBB2 mutation frequency in lobular breast cancer with pleomorphic histology or high-risk characteristics by molecular
expression profiling. Genes Chromosomes Cancer. 2019;58:175-185.
35. Lal P, Tan LK, Chen B. Correlation of HER-2 status with estrogen and progesterone receptors and histologic features in 3,655 invasive breast carcinomas. Am
J Clin Pathol. 2005;123:541-546.
36. De Mattos-Arruda L, Ng CKY, Piscuoglio S, et al. Genetic heterogeneity and actionable mutations in HER2-positive primary breast cancers and their brain
metastases. Oncotarget. 2018;9:20617-20630.
37. de Oliveira Taveira M, Nabavi S, Wang Y, et al. Genomic characteristics of trastuzumab-resistant Her2-positive metastatic breast cancer. J Cancer Res Clin
Oncol. 2017;143:1255-1262.
38. Hahnen E, Hauke J, Engel C, et al. Germline mutations in triple-negative breast cancer. Breast Care (Basel). 2017;12:15-19.
39. Hoppe MM, Sundar R, Tan DSP, et al. Biomarkers for homologous recombination deficiency in cancer. J Natl Cancer Inst. 2018;110:704-713.
40. Byrum AK, Vindigni A, Mosammaparast N. Defining and modulating ‘BRCAness’. Trends Cell Biol. 2019;29:740-751.

41. Saini P, Li Y, Dobbelstein M. Wee1 is required to sustain ATR/Chk1 signaling upon replicative stress. Oncotarget. 2015;6:13072-13087.
42. Marcus L, Lemery SJ, Keegan P, et al. FDA approval summary: pembrolizumab for the treatment of microsatellite instability-high solid tumors. Clin Cancer Res.
2019;25:3753-3758.
43. Kurata K, Kubo M, Mori H, et al. Microsatellite instability in triple negative breast cancers. Cancer Research. 2019;79:P1-06-11.
44. Bonneville R, Krook MA, Kautto EA, et al. Landscape of microsatellite instability across 39 cancer types. JCO Precis Oncol. 2017;2017:1-15.
45. Goodman AM, Kato S, Bazhenova L, et al. Tumor mutational burden as an independent predictor of response to immunotherapy in diverse cancers. Mol Cancer
Ther. 2017;16:2598-2608.
47. Ng CKY, Bidard FC, Piscuoglio S, et al. Genetic heterogeneity in therapy-naı̈ve synchronous primary breast cancers and their metastases. Clin Cancer Res.
2017;23:4402-4415.
48. Schrijver WAME, Selenica P, Lee JY, et al. Mutation profiling of key cancer genes in primary breast cancers and their distant metastases. Cancer Res. 2018;
78:3112-3121.
49. Tyran M, Carbuccia N, Garnier S, et al. A comparison of DNA mutation and copy number profiles of primary breast cancers and paired brain metastases for
identifying clinically relevant genetic alterations in brain metastases. Cancers (Basel). 2019;11:E665.
50. Kato T, Park JH, Kiyotani K, et al. Integrated analysis of somatic mutations and immune microenvironment of multiple regions in breast cancers. Oncotarget.
2017;8:62029-62038.
51. Laine M, Fanning SW, Greene M, et al. Lasofoxifene as a potential treatment for ER+ metastatic breast cancer. J Clin Oncol. 2019;37:1056.
52. Bidard FC, Sabatier R, Berger F, et al. PADA-1: a randomized, open label, multicentric phase III trial to evaluate the safety and efficacy of palbociclib in
combination with hormone therapy driven by circulating DNA ESR1 mutation monitoring in ER-positive, HER2-negative metastatic breast cancer patients. J Clin
Oncol. 2018;36 (suppl; abstr TPS1105).
53. Shimoi T, Hamada A, Yamagishi M, et al. PIK3CA mutation profiling in patients with breast cancer, using a highly sensitive detection system. Cancer Sci. 2018;
109:2558-2566.
54. Meric-Bernstam F, Brusco L, Daniels M, et al. Incidental germline variants in 1000 advanced cancers on a prospective somatic genomic profiling protocol. Ann
Oncol. 2016;27:795-800.
55. Leighl NB, Page RD, Raymond VM, et al. Clinical utility of comprehensive cell-free DNA analysis to identify genomic biomarkers in patients with newly diagnosed
metastatic non-small cell lung cancer. Clin Cancer Res. 2019;25:4691-4700.
56. Chung JH, Pavlick D, Hartmaier R, et al. Hybrid capture-based genomic profiling of circulating tumor DNA from patients with estrogen receptor-positive
metastatic breast cancer. Ann Oncol. 2017;28:2866-2873.
57. Chakravarty D, Gao J, Phillips SM, et al. OncoKB: a precision oncology knowledge base. JCO Precis Oncol. 2017;10.1200/PO.17.00011.
58. Mateo J, Chakravarty D, Dienstmann R, et al. A framework to rank genomic alterations as targets for cancer precision medicine: the ESMO Scale for Clinical
Actionability of molecular Targets (ESCAT). Ann Oncol. 2018;29:1895-1902.
59. Condorelli R, Mosele F, Verret B, et al. Genomic alterations in breast cancer: level of evidence for actionability according to ESMO Scale for Clinical Actionability
of molecular Targets (ESCAT). Ann Oncol. 2019;30:365-373.
60. Payandeh M, Sadeghi M, Sadeghi E, et al. Is there any concordance between of IHC with FISH in HER2-positive breast cancer patients? Int J Hematol Oncol
Stem Cell Res. 2017;11:43-48.
61. Nassar A, Khoor A, Radhakrishnan R, et al. Correlation of HER2 overexpression with gene amplification and its relation to chromosome 17 aneuploidy: a 5-year
experience with invasive ductal and lobular carcinomas. Int J Clin Exp Pathol. 2014;7:6254-6261.
62. Swain SM, Kim SB, Cortés J, et al. Pertuzumab, trastuzumab, and docetaxel for HER2-positive metastatic breast cancer (CLEOPATRA study): overall survival
results from a randomised, double-blind, placebo-controlled, phase 3 study. Lancet Oncol. 2013;14:461-471.
63. Geyer CE, Forster J, Lindquist D, et al. Lapatinib plus capecitabine for HER2-positive advanced breast cancer. N Engl J Med. 2006;355:2733-2743.
65. Modi S, Saura C, Yamashita T, et al; DESTINY-Breast01 Investigators. Trastuzumab deruxtecan in previously treated HER2-positive breast cancer. N Engl
J Med. 2020;382:610-621.
66. Verma S, Miles D, Gianni L, et al; EMILIA Study Group. Trastuzumab emtansine for HER2-positive advanced breast cancer. N Engl J Med. 2012;
367:1783-1791.
67. Krop IE, Kim SB, González-Martı́n A, et al; TH3RESA study collaborators. Trastuzumab emtansine versus treatment of physician’s choice for pretreated HER2-
positive advanced breast cancer (TH3RESA): a randomised, open-label, phase 3 trial. Lancet Oncol. 2014;15:689-699.
68. Gymnopoulos M, Elsliger MA, Vogt PK. Rare cancer-specific mutations in PIK3CA show gain of function. Proc Natl Acad Sci USA. 2007;104:5569-5574.
69. Bader AG, Kang S, Zhao L, et al. Oncogenic PI3K deregulates transcription and translation. Nat Rev Cancer. 2005;5:921-929.
70. Bosch A, Li Z, Bergamaschi A, et al. PI3K inhibition results in enhanced estrogen receptor function and dependence in hormone receptor-positive breast
cancer. Sci Transl Med. 2015;7:283ra51.

Li et al
71. Juric D, Loibl S, Andre F, et al. Alpelisib (ALP) with fulvestrant (FUL) in patients (pts) with PIK3CA-mutated hormone receptor-positive (HR+), human epidermal
growth factor receptor-2-negative (HER2-) advanced breast cancer (ABC): primary or secondary resistance to prior endocrine therapy (ET) in the SOLAR-1 trial.
J Clin Oncol. 2019;37 (suppl; abstr 1038).
72. Edgar K, Hanan E, Staben S, et al. Preclinical characterization of GDC-0077, a specific PI3K alpha inhibitor in early clinical development. Cancer Res. 2017;
77:156 (abstr 156).
73. Patnaik A, Appleman LJ, Tolcher AW, et al. First-in-human phase I study of copanlisib (BAY 80-6946), an intravenous pan-class I phosphatidylinositol 3-kinase
inhibitor, in patients with advanced solid tumors and non-Hodgkin’s lymphomas. Ann Oncol. 2016;27:1928-1940.
74. Dickler MN, Saura C, Richards DA, et al. Phase II study of taselisib (GDC-0032) in combination with fulvestrant in patients with HER2-negative, hormone
receptor-positive advanced breast cancer. Clin Cancer Res. 2018;24:4380-4387.
75. Jones RH, Casbard A, Carucci M, et al. Fulvestrant plus capivasertib versus placebo after relapse or progression on an aromatase inhibitor in metastatic,
oestrogen receptor-positive breast cancer (FAKTION): a multicentre, randomised, controlled, phase 2 trial. Lancet Oncol. 2020;21:345-357.
76. Moynahan ME, Chen D, He W, et al. Correlation between PIK3CA mutations in cell-free DNA and everolimus efficacy in HR+, HER2- advanced breast cancer:
results from BOLERO-2. Br J Cancer. 2017;116:726-730.
77. Amatu A, Sartore-Bianchi A, Siena S. NTRK gene fusions as novel targets of cancer therapy across multiple tumour types. ESMO Open. 2016;1:e000023.
78. Drilon A, Laetsch TW, Kummar S, et al. Efficacy of larotrectinib in TRK fusion-positive cancers in adults and children. N Engl J Med. 2018;378:731-739.
79. Demetri GD, Paz-Ares L, Farago AF, et al. Efficacy and safety of entrectinib in patients with NTRK fusion-positive (NTRK-fp) tumors: pooled analysis of
STARTRK-2, STARTRK-1 and ALKA-372-001. Presented at: ESMO 2018 Congress; October 2018; Munich, Germany.
80. Cocco E, Scaltriti M, Drilon A. NTRK fusion-positive cancers and TRK inhibitor therapy. Nat Rev Clin Oncol. 2018;15:731-747.
81. Ross J, Chung J, Elvin J, et al. NTRK fusions in breast cancer: clinical, pathologic and genomic findings. Cancer Res. 2018;78 (suppl; abstr P2-09-15).
82. Burstein MD, Tsimelzon A, Poage GM, et al. Comprehensive genomic analysis identifies novel subtypes and targets of triple-negative breast cancer. Clin Cancer
Res. 2015;21:1688-1698.
83. Caulfield SE, Davis CC, Byers KF. Olaparib: a novel therapy for metastatic breast cancer in patients with a BRCA1/2 mutation. J Adv Pract Oncol. 2019;
10:167-174.
84. Guney Eskiler G. Talazoparib to treat BRCA-positive breast cancer. Drugs Today (Barc). 2019;55:459-467.
85. McCrea C, Hettle R. Indirect treatment comparison of the efficacy and safety of olaparib 300 mg tablets BID and talazoparib 1 mg once daily in the treatment of
patients with germline BRCA-mutated (gBRCA) HER2-negative metastatic breast cancer. J Clin Oncol. 2019;37 (suppl; abstr e12570).
86. Oza AM, Tinker AV, Oaknin A, et al. Antitumor activity and safety of the PARP inhibitor rucaparib in patients with high-grade ovarian carcinoma and a germline or
somatic BRCA1 or BRCA2 mutation: integrated analysis of data from Study 10 and ARIEL2. Gynecol Oncol. 2017;147:267-275.
87. Swisher EM, Lin KK, Oza AM, et al. Rucaparib in relapsed, platinum-sensitive high-grade ovarian carcinoma (ARIEL2 Part 1): an international, multicentre,
open-label, phase 2 trial. Lancet Oncol. 2017;18:75-87.
88. Ledermann J, Harter P, Gourley C, et al. Olaparib maintenance therapy in patients with platinum-sensitive relapsed serous ovarian cancer: a preplanned
retrospective analysis of outcomes by BRCA status in a randomised phase 2 trial. Lancet Oncol. 2014;15:852-861.
89. Mirza MR, Monk BJ, Herrstedt J, et al; ENGOT-OV16/NOVA Investigators. Niraparib maintenance therapy in platinum-sensitive, recurrent ovarian cancer.
N Engl J Med. 2016;375:2154-2164.
90. Bose R, Kavuri SM, Searleman AC, et al. Activating HER2 mutations in HER2 gene amplification negative breast cancer. Cancer Discov. 2013;3:224-237.
91. Wang T, Xu Y, Sheng S, et al. HER2 somatic mutations are associated with poor survival in HER2-negative breast cancers. Cancer Sci. 2017;108:671-677.
92. Greulich H, Kaplan B, Mertins P, et al. Functional analysis of receptor tyrosine kinase mutations in lung cancer identifies oncogenic extracellular domain
mutations of ERBB2. Proc Natl Acad Sci USA. 2012;109:14476-14481.
93. Wang SE, Narasanna A, Perez-Torres M, et al. HER2 kinase domain mutation results in constitutive phosphorylation and activation of HER2 and EGFR and
resistance to EGFR tyrosine kinase inhibitors. Cancer Cell. 2006;10:25-38.
94. Kancha RK, von Bubnoff N, Bartosch N, et al. Differential sensitivity of ERBB2 kinase domain mutations towards lapatinib. PLoS One. 2011;6:e26760.
95. Ma CX, Bose R, Gao F, et al. Phase II trial of neratinib for HER2 mutated, non-amplified metastatic breast cancer (HER2mut MBC). J Clin Oncol. 2016;34 (suppl;
abstr 516).
96. Ma CX, Bose R, Gao F, et al. Neratinib efficacy and circulating tumor DNA detection of HER2 mutations in HER2 nonamplified metastatic breast cancer. Clin
Cancer Res. 2017;23:5687-5695.
97. Hyman DM, Piha-Paul SA, Won H, et al. HER kinase inhibition in patients with HER2- and HER3-mutant cancers [published correction appears in Nature.
2019;566:E11-E12]. Nature. 2018;554:189-194.
98. Toy W, Weir H, Razavi P, et al. Activating ESR1 mutations differentially affect the efficacy of ER antagonists. Cancer Discov. 2017;7:277-287.
99. Johnston SR, Kilburn LS, Ellis P, et al; SoFEA investigators. Fulvestrant plus anastrozole or placebo versus exemestane alone after progression on non-steroidal
aromatase inhibitors in postmenopausal patients with hormone-receptor-positive locally advanced or metastatic breast cancer (SoFEA): a composite, mul-
ticentre, phase 3 randomised trial. Lancet Oncol. 2013;14:989-998.
100. Fribbens C, O’Leary B, Kilburn L, et al. Plasma ESR1 mutations and the treatment of estrogen receptor-positive advanced breast cancer. J Clin Oncol. 2016;
34:2961-2968.

101. Schmid P, Abraham J, Chan S, et al. Capivasertib plus paclitaxel versus placebo plus paclitaxel as first-line therapy for metastatic triple-negative breast cancer:
the PAKT trial. J Clin Oncol. 2020;38:423-433.
102. Kim SB, Dent R, Im SA, et al; LOTUS investigators. Ipatasertib plus paclitaxel versus placebo plus paclitaxel as first-line therapy for metastatic triple-negative
breast cancer (LOTUS): a multicentre, randomised, double-blind, placebo-controlled, phase 2 trial. Lancet Oncol. 2017;18:1360-1372.
103. Hyman DM, Smyth LM, Donoghue MTA, et al. AKT inhibition in solid tumors with AKT1 mutations. J Clin Oncol. 2017;35:2251-2259.
104. Dembla V, Somaiah N, Barata P, et al. Prevalence of MDM2 amplification and coalterations in 523 advanced cancer patients in the MD Anderson phase 1 clinic.
Oncotarget. 2018;9:33232-33243.

BREAST CANCER
The Evolution of Clinical Trials in Metastatic

Breast Cancer: Design Features and Endpoints
That Matter
Andrew D. Seidman, MD1; Julia Maues, MA1; Tiah Tomlin, MS1; Vishal Bhatnagar, MD2; and Julia A. Beaver, MD3
overview
The evolution of thought in assessing benefit in clinical trials of systemic therapy for metastatic breast cancer
(MBC) is well documented, with most agents garnering regulatory approval based either on an advantage in
overall survival (OS), time to progression (TTP), or progression-free survival (PFS) over an existing standard of
care or objective response rate (ORR). Previous guidance for industry on clinical trial endpoints for the
approval of cancer drugs and biologics was provided by the U.S. Food and Drug Administration (FDA) in 2007
and recently updated in 2018. The more recent FDA guidance recognizes that advances in science are
facilitating the development of oncology products, which “may also result in the identification of additional
endpoints that may be used to support approval of oncology products.” This article critically addressed the
evolution of thought on the advancement of clinical trials in MBC, from various stakeholder perspectives.
Despite the term “stakeholder,” the objective of all co-authors and parties concerned is to promote and inform
the optimal design, conduct, and reporting of clinical trials for women with advanced breast cancer toward
improving and extending lives. This article provides an overview of the evolving perspectives on this issue from
the physician, regulatory agency, and patient and/or advocate points of view.
INTRODUCTION to assess benefit from the clinical trial intervention.9 A

The evolution of thought in assessing benefit in clinical working group convened by the National Cancer In-
trials of systemic therapy for MBC is well documented,1-6 stitute Breast Cancer Steering Committee conducted
with most agents garnering regulatory approval based a detailed curated systematic review of the literature
either on an advantage in OS, TTP, or PFS over an on clinical trial endpoints for metastatic breast cancer
existing standard of care or objective response rate to generate more refined, granular evidence-based
(ORR). Previous guidance for industry on clinical trial recommendations for endpoints for clinical trials in
endpoints for the approval of cancer drugs and biologics MBC.10 The primacy of OS or PFS was parsed by
was provided by the FDA in 20077 and recently updated biologic subtype (triple negative, HR /HER2+, HR+/
in 2018.8 The more recent FDA guidance recognizes that HER2+, and HR+/HER2 ), as well as by line of
advances in science are facilitating the development of therapy for MBC (first, second, third, and beyond;
oncology products, which “may also result in the iden- Fig. 1). This working group, composed of medical
tification of additional endpoints that may be used to oncologists, expert disease-specific FDA liaisons,
support approval of oncology products.” This article biostatisticians, and patient advocates, generated key
critically addressed the evolution of thought on the ad- position statements with a high level of agreement that
vancement of clinical trials in MBC from various stake- underscored the need for examination and validation
Author affiliations
holder perspectives. Despite the term “stakeholder,” the of novel metrics to assess benefit in clinical trials:
and support
information (if
objective of all of the co-authors and parties concerned is
to promote and inform the optimal design, conduct, and 1. Toxicity can outweigh small gains in PFS as well as OS.
applicable) appear
at the end of this reporting of clinical trials for women with advanced breast 2. The balance of PFS gain and toxicity burden, as well
article. cancer toward improving and extending lives. as patient-reported outcome (PRO) and/or quality-of-
Accepted on March life (QoL) effect should be considered individually, in
4, 2020 and parallel when PFS is the preferred endpoint.
published at Novel Metrics to Assess Benefit in Clinical Trials 3. Insufficient patient-based research exists to gauge
ascopubs.org on
MBC is heterogeneous both in its biology and ex- outpatients regarding incremental PFS gain versus
March 31, 2020:
pected outcomes. Consequently, the expected post- toxicity burden.
10.1200/EDBK_ progression survival for a patient exiting a clinical trial 4. Reporting the area under the time-toxicity curve
280451 may be paramount in the choice of a primary endpoint may provide a more meaningful assessment of

Evolution of Clinical Trials of MBC
tolerability then reporting percentage with worse grade

experienced.
5. Graphic displays of toxicity over time can communicate
• Clinical trials for advanced breast cancer have complex ideas with precision and clarity.
evolved with input from various stakeholders.
6. The PRO version of Common Terminology Criteria for
• The heterogeneity in biology and natural his- Adverse Events (CTCAE) yields direct insight into patient
tory of metastatic breast cancer argues for experience and provides a comprehensive perspective of
situation-specific design and endpoint the benefits and risks of treatment.
considerations.
In the context of these positions, there have been several
• Novel metrics that assess multiple efficacy and
contemporary efforts to optimally characterize benefit be-
toxicity endpoints to gauge clinical benefit have
been and continue to be evaluated. tween regimens in comparative clinical trials that consider
primary and secondary endpoints, as well as toxicity, PRO,
• The patient’s voice as assessed not only in
and QoL data, as available. One such instrument is the
patient reported outcomes but also in the de-
European Society of Clinical Oncology Magnitude of Clinical
sign of clinical trials, will ensure that all patients’
interests are represented, including historically Benefit Scale (ESMO-MCBS).
under-represented populations.
ESMO-MCBS
• Graphic representation of adverse events, such
as those that capture area under the time- Cherny et al11 have developed and validated a stan-
toxicity curve, as well as visual depiction of dardized, reproducible tool for scoring clinical benefit in
benefit and harm, can augment the un- the treatment of solid tumors in randomized comparative
derstanding of the human impact of a clinical trials (RCTs). Specific forms are used for studies of new
trial intervention. agents in the management of cancers without curative
intent, further refined by whether the primary outcome of
FIGURE 1. Biologic and Setting-

Specific Considerations for End-
points in RCTs for Metastatic
Breast Cancer
(A) Working group consensus on
preferred endpoints by biologic
subtype and line of therapy. (B)
Hypothetical scenarios for ex-
pected postprogression survival
and choice of preferred endpoint.10
Abbreviations: OS, overall sur-
vival; PFS, progression-free sur-
vival; RR, response rate; TNMBC,
triple-negative metastatic breast
cancer.

Seidman et al
interest is (1) OS, (2) PFS/TTP, or (3) QoL, toxicity, or ASCO Conceptual Framework to Assess the Value of
response rate (RR). Within this construct, using a 5- Cancer Treatment Options
point scoring system, benefit is prioritized, and direct ASCO’s Value in Care Task Force chose to define value in
endpoints such as OS and QoL take precedence over cancer care by emphasizing clinical benefit (efficacy),
less reliable surrogates such as PFS or RR. In studies in toxicity (safety), and cost (efficiency).16 Within their frame-
which OS has primacy, scoring considers hazard ratio work, clinical benefit and toxicity are combined to generate
and median survival gain; adjustments are made based a “net health benefit” score. A point-based scoring system
on QoL and toxicity considerations. In studies in which was created for improvements in OS or PFS (e.g., 1 point if
PFS/TTP has primacy, upgrading or downgrading of OS or PFS improved by 0% to 24%, 2 points if improved
scores can occur based on secondary outcomes, such 25% to 49%, 3 points if improved 50% to 75%, and 4 points
as toxicity data, improvement in OS, or data derived from if improved 76% to 100%). The net health benefit toxicity
QoL evaluation. Interpretation of the evidence for benefit calculation assessed the percentage difference in grade 3
derived from PFS and RR may be influenced by sec- to 5 toxicities for the new regimen compared with the
ondary outcome data. Examples of metrics that affect standard comparator. Bonus points could be assigned for
scoring for toxicity assessment include toxic death evidence of palliation of symptoms. Value (not the focus of
rate . 2%, hospitalization rate for toxicity . 10%, severe our present article) was derived from assessing net health
congestive heart failure rate . 4%, and grade 3 neu- benefit in relation to cost.
rotoxicity . 10%. Assessment of PFS is sensitive to both
In the context of assessing clinical benefit, one can regard
hazard ratios and absolute changes.
data inputs as either a “lumper” or a “splitter.” From
When PFS, not OS, is the primary endpoint, stakeholder a regulatory perspective, the totality of evidence is con-
perspective may be more important. There are little patient- sidered by examining several toxicity and efficacy metrics
derived data regarding the meaningfulness of PFS gains siloed from one another (splitting). In the deliberations of the
from the patient perspective.12 Hurvitz et al13 performed National Cancer Institute (NCI) Breast Cancer Steering
a multistage study that involved direct patient interviews and Committee Working Group on meaningful and appropriate
developed a questionnaire that assessed overall health and endpoints for clinical trials and metastatic breast cancer,10
well-being, cancer-related anxiety, the impact of treatment there was moderate support for the development and val-
on physical and emotional status, and the tradeoff between idation of a single metric that could capture key relevant
incremental gain in PFS and “cost” in terms of toxicity or efficacy and toxicity measures into one understandable and
negative impact on QoL. In scenarios in which PFS and OS meaningful result (lumping). The Working Group envisioned
are not closely correlated, it will be interesting to examine the that such a metric would encompass both relative and
relationship between PFS and formal PROs going forward, absolute gains in efficacy measures (e.g., PFS/OS), an “area
as these are currently integrated into several contemporary under the curve” assessment of toxicity over time,17 and
clinical trials.14 PROs, all together. Such an approach might allow for
How do newer metrics of clinical benefit stack up against comparisons of standard regimens used in the same setting
the standard of regulatory approval? Tibau et al15 col- that have not been and may never be studied in RCTs, such
lected data on efficacy, safety, and QoL using the ESMO- as the use of fulvestrant with a cyclin-dependent kinase 4/6
MCBS form from RCTs leading to FDA approval for solid inhibitor versus exemestane and everolimus after progres-
tumors. They determined that only 38.8% of RCTs per- sion on a nonsteroidal aromatase inhibitor for HR+/HER2
formed in the palliative setting met the ESMO-MCBS MBC. One current visual approach to display treatment
threshold for meaningful benefit (45.6% of drugs re- features collectively is the National Comprehensive Cancer
ceived regular approval and 18.5% of those received Center Network’s Evidence Blocks, which grade efficacy,
accelerated approval). For studies with coprimary end- safety, evidence quality, evidence consistency, and af-
points (e.g., PFS/OS), the authors chose to use OS as their fordability in a visual format.
benchmark for comparison with MCBS, which introduced The composite endpoint known as clinical benefit rate
potential bias. Notably, over time there has been an in- encompasses the proportion of patients who have a con-
crease in the number of trials meeting the MCBS firmed complete or partial response (CR or PR) and also
threshold for “substantial improvement” and “high level includes patients with stable disease (SD) for at least
of clinical benefit" (i.e., a MCBS score of 4 of 5 or 5 of 5). a period of time (e.g., for hormone-sensitive malignancies,
This effect seemed to be influenced by more frequent this is typically 6 months), or CR + PR + SD ( 6 months).
observation of improved QoL or reduced toxicity, as op- Some trials have used a shorted 4-month period of SD to the
posed to improvement in relative efficacy. same definition based on the expected natural history of the

disease subtype and/or scenario. This is one of the many PROs: Secondary, Primary, or Coprimary?
metrics that may be considered “in totality” to gauge the In a 2016 review of published literature regarding the use of
utility of a specific treatment. patient-reported outcomes in advanced breast cancer
clinical trials, Turner-Bowker et al14 identified 25 articles
Graphic and/or Visual Representation of Benefit and Harm that reported studies that used PROs to evaluate treatment
The notion that a patient might care more that she spent benefit and/or toxicity in MBC. Nineteen of these publica-
6 months with grade 2 diarrhea or neuropathy versus 6 days tions reported on phase III trials, 4 reported on phase II
emphasizes the importance of capturing time-dependent trials, with 10 publications focused solely on PRO efficacy
toxicity, which can be best captured graphically.17 Visual results. A total of 11 different PRO questionnaires were
displays of multivariate health care outcome data may identified across these publications; the most commonly
provide more meaning than when data are presented in text reported were EORTC QLQ-C30 (52%) and the FACT-B
or tabular format.18,19 Techniques that allow data visuali- (28%). The trials involved endocrine therapy for HR+ MBC,
zation can be powerful tools to synthesize information and HER2-targeted therapies for HER2+ MBC, chemothera-
provide insight into complex data.18 Waterfall plots offer peutic agents, and CDK 4/6 inhibitors. Most publications
a representation of a change in tumor volume that provides (84%) described or referred to the statistical test(s) used to
more granular information than that which may be gleaned analyze PRO data. Five studies (20%) reported on how
by RECIST criteria response proportion alone. Swimmer missing PRO data were handled. Of publications that re-
plots can illustrate individual patient time-dependent out- ported specific PRO results, at least some improvement
comes such as PFS or OS and can be further refined with from baseline was reported in eight publications (34.8%),
unique patient identifiers or subset identifiers. In a recent with no change in PRO scores reported in six publications
analysis of PALOMA-2 data, Rugo et al20 demonstrated that (26.1%), and deterioration in nine publications (39.1%; four
the combination of palbociclib and letrozole “provided publications did not specify change from baseline).
significant clinical benefit by prolonging median PFS … The Setting International Standards in Analyzing Patient
regardless of whether patients achieved an objective re- Reported Outcomes and Quality of Life Endpoint Data for
sponse” (Fig. 2). Other techniques include spider plots and Cancer Clinical Trials (SISAQOL) Consortium was estab-
radar charts (Fig. 3).19 A radial plot displays several metrics lished to address the need to develop standards, guidelines,
simultaneously in a circular format composed of radii ra- and recommendations for analysis of PRO data in cancer
diating from the center, with a length of a specific radius RCTs. In their systematic review conducted in 2006 and
corresponding to the measured value; these are sometimes updated in 2018, 66 articles that reflected RCTs for PRO
connected to form an enclosed shape. The radar plot, data in MBC were identified, involving 26,905 patients. The
a specific radial plotting technique, has found use for sample sizes ranged from 66 to 1,102 patients, with an
graphing multivariate data in health care. 21,22 Scenarios average of 407 patients per trial. Twelve (18%) of the studies
can be imagined in which multiple efficacy and/or tox- were considered to be practice-changing trials.27 Only eight
icity endpoints of interest could be displayed graphically (12%) of these trials reported a specific PRO research
for different regimens, allowing for an easily digestible hypothesis, with heterogeneity in statistical methods used,
comparison. and a mixture of longitudinal and cross-sectional tech-
One increasingly common representation of efficacy, a niques. Forty-eight trials (73%) did not report how missing
histogram known as the waterfall plot, may potentially ex- data were handled.
aggerate response rates.26 RECIST 1.1 requires a confir- Thus, it seems apparent that although PROs are universally
matory scan documenting a . 30% reduction in tumor valued to “provide care that is respectful of an responsive to
measurement to count as a response, whereas waterfall individual patient preferences, needs, and values ensuring
plots show the single best subsequent scan. Thus, not every that patient values guide clinical decisions,”28 careful co-
bar below the 30% line is a RECIST criteria response. Also, ordination of future efforts between medical investigators,
because the response rate is calculated based on the in- industry, regulatory agencies, and patient advocates will be
tention to treat the denominator, the waterfall plot only in- necessary to improve and enhance the integration of this
cludes patients able to be assessed for response. Patients important metric to inform the benefit of new systemic
who lack post-baseline scans are not evaluable for response agents for breast cancer. Electronic tablet and internet-
and excluded from the waterfall plot. Nevertheless, waterfall based versions of validated tools may increase accessibil-
plots do provide a valuable visual to production of changes ity and ease of data transfer and reliability of PRO data.29,30
in tumor burden that are not evident based on routine With further optimization and validation, it is conceivable
conventional response reporting (e.g., CR, PR, SD, pro- that PROs, which are largely considered secondary end-
gression of disease). points RCTs for MBC, could be promoted to coprimary

Seidman et al
endpoints. The favorable validity, reliability, and responsiveness

of the NCI’s patient-perspective CTCAE—PRO-CTCAE31—is
currently being assessed in ongoing RCTs toward this end.
Can Real-World Data Inform Future Clinical Trial Design
and/or Endpoints?
It is often stated that the benefits and harms of a new
treatment as assessed in the context of a carefully defined
clinical trial population cautiously assembled through the
filter of dozens of eligibility criteria may not reflect the
eventual benefits and harms in a broader, real-world
population that has not been so vetted. With the wide-
spread adoption of electronic health records and the
growing wealth of digital information about patients, there is
now an opportunity to both learn how clinical trial outcomes
reflect real-world outcomes, and likewise, how real-world
experience can inform future clinical trial design.32 For
electronic health records to derive data to yield reliable real-
world evidence, it must be of known and sufficiently high
quality.33 Real-world data can be generated using experi-
mental designs similar to those used in conventional clinical FIGURE 2. Waterfall Plots for the Best Percentage Change in Tumor Size
trials; these data support active pharmacological vigilance, at Any Time in Patients With Measurable Disease at Baseline Who Re-
provide insights into the natural history of disease, and ceived (A) Palbociclib Plus Letrozole or (B) Placebo Plus Letrozole
inform the development of future control arms of clinical Despite fairly similar waterfall plot representation of responses, patients
trials.34 The use of such evidence has the potential to allow receiving letrozole + palbociclib had a substantially longer median PFS
than patients receiving letrozole + placebo (50).20
the efficient answers to important clinical questions, saving
Abbreviations: ITT, intent-to-treat; PFS, progression-free survival;
time and money while yielding answers relevant to broader
RECIST, Response Evaluation Criteria in Solid Tumors.
populations of patients than would be possible, as well as a
Defined according to RECIST version 1.1 and excludes patients with
a specialized research environment.35 bone-only disease and other patients with nonmeasurable disease be-
Response and progression are assessed differently by yond bone-only. For each treatment aim, the total number is based on
RECIST criteria in the clinical trial setting compared with the ITT population, excluding patients with missing values for change
from baseline at all visits postbaseline and those with nonmeasurable
outside a clinical trial. As such, changes in radiographic
disease. Red lines indicate response criteria based on RECIST Version
disease that the trigger stops current therapy and switches
1.1. Upper Red Line: Progressive disease defined as at least a 20%
to the next also vary. A growing number of clinical settings in increase in the sum of diameters of target lesions. Lower Red Line: Partial
which “objective progression” does not necessarily indicate response defined as at least a 30% decrease in the sum of diameters of
treatment failure or must change therapy supports the need target lesions.
for reassessing criteria for progression.36 Oxnard et al36
classified the scenarios into four groups: tumor marker
progression, focal progression amenable to local therapy, phase III trial, the median PFS for letrozole and palbociclib
indolent or asymptomatic progression, and progression was shorter than that previously reported from phase II and
while on immunotherapy. In breast cancer, the develop- III trials. Differences in response assessment and different
ment of systemic agents that may be effective both within criteria for progression between the clinical trial and the real-
the central nervous system and for non–central nervous world settings may partially contribute to this discordance
system disease leads to the requirement to examine efficacy (vide supra). Furthermore, the authors noted that their real-
collectively and in both compartments separately. world population included patients with more previous
In a recent real-world cohort study of patients who received chemotherapy for early-stage breast cancer as well as lower
letrozole plus palbociclib for metastatic breast cancer, 411 performance status than in the PALOMA clinical trials.
women treated from February 2015 to December 2017
were retrospectively reviewed, examining PFS as the pri- FDA Perspective
mary endpoint.37 A modest 8% of patients permanently Approval of drugs by the FDA requires substantial evidence
discontinued palbociclib because of adverse events. Al- of safety and effectiveness.7,38 Two pathways exist for drug
though the estimated PFS in patients treated with fulvestrant approval: traditional and accelerated approval. Although
and palbociclib was comparable to the previously reported traditional approval does not require an overall survival

better than available therapies on the endpoint studied,

and states that confirmatory trials may be required and that
if benefit is not verified, the drug is subject to potential
withdrawal from marketing.
Although direct clinical outcomes, such as how a patient
feels, functions, or survives, are considered most mean-
ingful, clinically relevant surrogate or intermediate out-
comes such as tumor measures are also considered
important. FDA has acknowledged that with a favorable
benefit–risk profile, the delay of tumor progression or
substantial decrease in tumor size can also benefit patients
and demonstrate efficacy.41 The most commonly used
endpoints in oncology to primarily support drug approval by
the two approval pathways include OS, PFS, and ORR.42,43
Although overall survival is considered the gold standard
and allows for incorporation of a meaningful and relevant
clinical outcome with a low risk of biased assessment, there
are limitations to the use of OS as the primary endpoint. In
the context of MBC in which there are multiple available
therapies and patients are generally cycled through lines of
therapy with the potential (particularly in the HER2+ and
HR+ spaces) for a long post-treatment survival period, OS
may not be practical. In addition, allowing patients to cross-
over on progression or be aware of their treatment arm, may
confound the assessment of OS, and in contrast, may not be
feasible or ethical to disallow.44 Alternatively, PFS is a ra-
diographic measurement that incorporates a time-to-event
analysis. Although there is the potential for investigator
bias, an objectively measured PFS by blinded central
review can minimize this risk and allow for a more feasible
FIGURE 3. Visual Representations of Outcome Metrics
design in which poststudy treatments do not confound the
(A1) Example of radar plot assessing 5 dimensions associated with
a health care system.23 (A2) Example of a radar plot addressing common
primary endpoint. For instance, waiting for OS benefit for
symptoms of generalized anxiety disorder.24 (B) Example of spider plot approval of a drug with PFS benefits could delay approval
and associated waterfall plot in a trial of PD-1 blockade in advanced for years. PFS can be considered an endpoint in support
colorectal cancer.25 of accelerated or traditional approval depending on the
magnitude, study design, and measurement characteris-
tics. ORR is also a radiographic measurement that relies
improvement, endpoints for traditional approval allow for upon tumor shrinkage and is able to reflect the direct
demonstration of direct clinical benefit and do not require activity of a drug. Therefore, as opposed to PFS and OS,
a therapy to be better than other available therapies. In ORR is interpretable in a single-arm trial. The clinical
many diseases, drugs are therefore able to be approved meaningfulness of ORR depends on the clinical scenario
based on a comparison with a placebo-treatment arm. and the associated duration of response. Further re-
However, in oncology it may not be ethical to allow for finement of specific criteria for meaningful endpoints in
treatment with a placebo arm; therefore, many trials are the context of checkpoint inhibitor therapy for MBC (e.g.,
conducted using an add-on design or a replacement in triple negative MBC) is ongoing; the issue of endpoint
design that permits demonstration that the drug controls surrogacy is complex and disease- and/or scenario-
the tumor or improves survival compared with a standard dependent.45 ORR can correlate poorly with OS.46 As-
of care. Accelerated approval, created by legislation in sessment of biomarker changes (e.g., cell-free DNA) may
1992 spurred on by activism from the HIV crisis, allows for complement conventional efficacy assessment going
approval based on either a surrogate endpoint reasonably forward.
likely to predict clinical benefit or an intermediate end- In the last 30 years, the FDA has approved 25 drugs for 34
point in the direct causal pathway of a more meaningful indications for the treatment of patients with metastatic
endpoint.39,40 Accelerated approval requires a drug to be breast cancer.43,47,48 These approvals have used both the

Seidman et al
traditional (n = 27) and accelerated approval pathways (n =

7) and have used many of the most common endpoints. For
accelerated approval, three indications were supported by
ORR with supportive duration of response, and four used
PFS (or time to progression). One of these drugs, bev-
acizumab, initially approved via accelerated approval based
on PFS, was subsequently removed from the market be-
cause further clinical evidence did not verify benefit and the
benefit–risk profile was no longer viewed favorably. For
traditional approvals, 20 indications were supported by PFS
(or time to progression), 3 by OS, 3 by ORR, and 1 in-
corporated both clinical and pharmacokinetic endpoints.
Novel endpoints and better defining the patient experience
are ongoing areas of interest for the FDA. In the case of
MBC, endpoints that better capture the patient experience
and help put the efficacy endpoint into context of tolerability
can assist with the evaluation of efficacy. Many trials ex-
amine PRO measurements, including measurements of
a patient’s health-related QoL, physical functioning, or
disease-related symptoms. Unfortunately, because of lim-
itations of health-related QoL, including the potential for
confounders (health-related QoL components can be af-
fected by life circumstances outside health-related con-
cerns) and methodological challenges (defining clinically
meaningful thresholds), the use of health-related QoL as
a key endpoint in cancer trials has been uncommon but is
frequently used as an exploratory endpoint for descriptive
purposes. Focusing PRO analyses on domains that are FIGURE 3. (Continued).
proximal to the disease and treatment being studied, such
as cancer-related symptoms, treatment-related symp-
tomatic adverse effects, physical function, role function, reasons, OS is not achieved easily or within a short time
and overall side effects may provide a more precise un- period, such that alternative endpoints have been used. The
derstanding of the patient experience. 49 These PROs main alternative endpoint to OS is PFS, which is captured
often support and supplement radiographic and survival more quickly than OS and is not influenced by subsequent
endpoints in regulatory decision-making. From a meth- treatments.50 PFS is often criticized because an improve-
odological perspective, having clearly stated and well- ment in PFS does not always predict an improvement in OS.
defined PRO objectives in cancer clinical trials will lead For patients going through scan after scan, the benefit of not
to more interpretable results. Newer technologies, other experiencing progression should not be minimized and can
clinical outcome assessments, and real-world studies be more meaningful if it results in lower pain levels. With the
also have the potential to better delineate meaningful gains in PFS recently observed for the addition of CDK 4/6
and interpretable endpoints and continue to advance inhibitors to endocrine therapy, the endpoint of time to
the field. chemotherapy or chemotherapy-free survival (with cen-
soring for deaths) has emerged as one means to illustrate
Patient and/or Advocate Perspective a potential benefit of this new treatment approach. It will be
interesting to formally capture and understand how patients
The last decade has seen a plethora of drugs available for
regard this endpoint in the context of the continuum of care
people living with MBC. Unfortunately, the disease remains
for HR+ advanced breast cancer.
incurable. Currently available drugs have the goal of pro-
longing lives, with the additional goal of keeping patients’ Although it is understandable that the primary endpoint of
QoL as high as possible. To be approved by the FDA, a drug a clinical trial must be measurable and be able to determine
needs to be tested for safety and efficacy in clinical trials. efficacy with statistical certainty, it is also important to
Therefore, the design of such trials, and specifically, the understand that the QoL of a patient while on therapy can be
choice of endpoints, is critical. Many consider the gold more important than living longer. PRO measures are
standard endpoint to be OS, but for previously described a means for patient reporting on the status of their health

FIGURE 4. Alignment in Stakeholder Per-

spectives on Clinical Trial Design and
Endpoints
Graphic depiction of the authors’ view on
the alignment of viewpoints and priorities
for future clinical trials in advanced breast
cancer. We feel that we (B) is a closer
approximation to the current state than (A).
condition directly, without it being interpreted by a clinician phase III trials in advanced breast cancer,53 optimizing the
or others.51 It is key that the patient voice is included when chances of a positive trial. Finally, efficient and cost-
developing PRO measurements and that the information effective evaluation of new targeted therapies linked to
collected is used in a specified analysis, ideally as a sec- biomarkers using a Bayesian adaptive design in the phase II
ondary endpoint, not simply an exploratory one. When metastatic setting is emerging as an attractive clinical trial
patients can report what matters to them in a format that is design strategy.54
convenient for them, are told why the questions are being Another Reason to Consider PROs
asked, and are informed about the results and how they are
used, patients become partners in the research and are With clinical trials being the cornerstone in developing and
more likely to dedicate time to answer questions. validating new cancer therapies and treatment options for
metastatic breast cancer patients, there is still a looming
Clinical Trial Design gap. According to data presented in the FDA’s Drug Trial
The discussion of trial design in the FDA Perspective section Snapshot Reports, less than 5% of African American pa-
of this article raises important points about the ethics of trials tients were involved in clinical trials for 24 of 31 cancer
with a placebo arm. The general public (and patients) often drugs approved since 2015.55,56 It has also been noted by
does not know that in the United States, placebo arm trials the American Cancer Society that for most cancers, African
are unacceptable (unless surveillance is standard of care).44 American patients in the United States have the highest
Clinical trials are the pillar of cancer clinical research and death rate and the shortest survival rates than any other
the main mechanism for introducing new treatments in ethnic group. With this disparity in mind, perhaps we
medicine. Therefore, it is key that the public be educated should consider looking at endpoints from a different
about the importance of clinical trials, including that patient perspective. Clinical trialists must address how a possible
outcomes are, on average, better with clinical trials, even if lack of diversity in a clinical trial might affect or predict the
on a control arm. Once a patient is educated on the benefits outcomes that are intended to be applied for the benefit of
of clinical trial participation and offered participation, they all patients. Can we truly say that the new and innovative
say yes more than 50% of the time.52 Thus, serious and drugs that are coming to the market are successful if there
thoughtful, efforts to increase health care and clinical reis inadequate representation of a particular population with
search literacy are sorely needed. Other barriers to partic- worse expected outcomes? Perhaps the investigational
ipation include trials not being available for the cancer type agent was successful, but only for a subset of the general
and stage of the patient at the treating center (55.6%) and population. When identifying endpoints of a clinical trial, it
not being eligible for available trials (21.5%).52 Efforts to is important to consider that all ethnic groups are well
broaden eligibility criteria and decentralize clinical trials are represented.
currently in place. How Can We Ensure Diverse Representation?
Clinical trial designs increasingly are powered to enable It is critical to note the lack of diversity in clinical trials and
unbalanced randomization in the direction of the in- the implications this has for people of color. First, the latest
vestigational arm as opposed to the control arm (e.g., 2:1 cancer therapies have not been proven to be effective in
randomization). Although the benefit of the experimental minority populations. Second, not enough therapies have
arm cannot be known in advance of the trial, this design been advanced for cancers that disproportionally affect
feature holds appeal to many patients hoping to receive minorities. Another issue that affects clinical trial partici-
a promising agent. Lessons learned from the neoadjuvant pation is the financial toxicity experienced by cancer pa-
setting (e.g., in the I-SPY 2 trial) can inform the design of tients. When patients are offered trials, costs to the

Seidman et al
patient—medical expenses not covered by trial or in- CONCLUSION

surance, travel and lodging, and indirect costs of time spent We hold a few truths to be self-evident: (1) recent clinical
at clinic or infusion center—must be disclosed. Broadening trials in advanced breast cancer have led to prolonged
eligibility criteria to adequately represent special pop- survival; (2) efficacy often comes at a cost (toxicity), (3)
ulations (older adults, those with brain metastases and situation-specific considerations for the most meaningful
organ dysfunction, racial/ethnic minorities, those with HIV) and appropriate endpoint for a clinical trial in advanced
deserves careful attention. breast cancer are warranted, and (4) capturing the patient’s
Using PROs may be regarded as one useful way to help voice, and all patients’ voices, including those under-
address diversity in drug development to ensure that clinical represented in clinical trials, will ensure that the “winner” in
trials not only reflect the unique biology of the patient and a randomized clinical trial will provide benefit to the broadest
their tumors but also give voice to the needs of patients from and most diverse population.
diverse backgrounds. Not having adequate representation There is room for improvement, and the path forward is best
in clinical trials can lead to a subset of patients who receive paved by thoughtful collaboration between oncologist in-
therapies that are either not effective or not tolerable, vestigators, industry, the FDA, and the patient. We believe
leading to a poor quality of life. By collecting PROs over time that we are much more aligned in our view than some might
from the African American patient population, such data suspect (Fig. 4). Because it can take years to design, open,
could potentially be beneficial for the general population and fully enroll patients in a large clinical trial, standards of
when attempting to recruit patients. The more data col- care can change during that time and can make the eli-
lected, the easier it will be to address this disparity. gibility for a given trial obsolete by the time the trial is
Finally, the discussion of clinical trial endpoints is not one completed. Careful attention to the changing landscape of
to be won by one statistic. To quote Jane Perlmutter, “Every standard therapy for breast cancer is thus required, and
cancer patient is the same, we want to live long and we alterations to an ongoing clinical trial may warrant consid-
want to live well, and every cancer patient is unique.” eration for amendments to that trial to ultimately produce
Therefore, a trial should be assessed on multiple end- a meaningful and actionable result. Finally, one should not
points. A composite endpoint could be possible, in which fear, but rather embrace the rigorous evaluation of novel
each patient sets different parameters and/or values for metrics to assess benefit and harm in the context of future
each variable. clinical trials and use them if and/or when validated.

1
My Style Matters, Inc., Atlanta, GA Andrew D. Seidman, MD, Breast Medicine Service, Memorial Sloan
2
Oncology Center of Excellence, U.S. Food and Drug Administration, Kettering Cancer Center, 300 East 66th Street, New York, NY 10065;
Silver Spring, MD email: [email protected].
3
U.S. Food and Drug Administration, Silver Spring, MD
EDBK_280451.
REFERENCES
1. Cortazar P, Justice R, Johnson J, et al. US Food and Drug Administration approval overview in metastatic breast cancer. J Clin Oncol. 2012;30:1705-1711.
2. Petrelli F, Barni S. Surrogate endpoints in metastatic breast cancer treated with targeted therapies: an analysis of the first-line phase III trials. Med Oncol. 2014;
31:776.
3. Raphael J, Verma S. Overall survival (OS) endpoint: an incomplete evaluation of metastatic breast cancer (MBC) treatment outcome. Breast Cancer Res Treat.
2015;150:473-478.
4. Conde-Estévez D, Tusquets I, Servitja S, et al. An overview of randomized clinical trials in metastatic breast cancer: variables affecting regulatory drug approval.
Anticancer Drugs. 2014;25:992-997.
5. Song SY, Seo H, Kim G, et al. Trends in endpoint selection in clinical trials of advanced breast cancer. J Cancer Res Clin Oncol. 2016;142:2403-2413.
6. Hurvitz SA. Evolving options for the treatment of metastatic breast cancer: progression-free survival as an endpoint. Cancer Treat Rev. 2011;37:495-504.
7. Guidance for Industry. Clinical trial endpoints for the approval of cancer drugs and biologics. https://www.fda.gov/media/71195/download. Accessed March 20,
2020.

8. U.S. Department of Health and Human Services. Clinical trial endpoints for the approval of cancer drugs and biologics: guidance for industry. www.fda.gov/
media/71195/download. Accessed February 29, 2020.
9. Broglio KR, Berry DA. Detecting an overall survival benefit that is derived from progression-free survival. J Natl Cancer Inst. 2009;101:1642-1649.
10. Seidman AD, Bordeleau L, Fehrenbacher L, et al. National Cancer Institute Breast Cancer Steering Committee Working Group report on meaningful and
appropriate endpoints for clinical trials in metastatic breast cancer. J Clin Oncol. 2018;36:3259-3268.
11. Cherny NI, Sullivan R, Dafni U, et al. A standardised, generic, validated approach to stratify the magnitude of clinical benefit that can be anticipated from anti-
cancer therapies: the European Society for Medical Oncology Magnitude of Clinical Benefit Scale (ESMO-MCBS). Ann Oncol. 2015;26:1547-1573.
12. Kaklamani VG. Clinical implications of the progression-free survival endpoint for treatment of hormone receptor positive advanced breast cancer. Oncologist.
2016;21:922-930.
13. Hurvitz SA, Lalla D, Crosby RD, et al. Use of the metastatic breast cancer progression (MBC-P) questionnaire to assess the value of progression-free survival for
women with metastatic breast cancer. Breast Cancer Res Treat. 2013;142:603-609.
14. Turner-Bowker DM, Hao Y, Foley C, et al. The use of patient-reported outcomes in advanced breast cancer clinical trials: a review of the published literature. Curr
Med Res Opin. 2016;32:1709-1717.
15. Tibau A, Molto C, Ocana A, et al. Magnitude of clinical benefit of cancer drugs approved by the US Food and Drug Administration. J Natl Cancer Inst. 2018;
110:486-492.
16. Schnipper LE, Davidson NE, Wollins DS, et al; American Society of Clinical Oncology. American Society of Clinical Oncology statement: a conceptual framework to
assess the value of cancer treatment options. J Clin Oncol. 2015;33:2563-2577.
17. Thanarajasingam G, Atherton PJ, Novotny PJ, et al. Longitudinal adverse event assessment in oncology clinical trials: the Toxicity over Time (ToxT) analysis of
Alliance trials NCCTG N9741 and 979254. Lancet Oncol. 2016;17:663-670.
18. Saary MJ. Radar plots: a useful way for presenting multivariate health care data. J Clin Epidemiol. 2008;61:311-317.
19. Stafoggia M, Lallo A, Fusco D, et al. Spie charts, target plots, and radar plots for displaying comparative outcomes of health care. J Clin Epidemiol. 2011;
64:770-778.
20. Rugo HS, Finn RS, Gelmon K, et al. Progression-free survival outcome is independent of objective response in patients with estrogen receptor-positive, human
epidermal growth factor receptor 2-negative advanced breast cancer treated with palbociclib plus letrozole compared with letrozole: analysis from PALOMA-2.
Clin Breast Cancer. In press.
21. Kosaka A, Osaku M, Watabiki Y, et al. QOL of breast conserved patients analyzed by radar chart. Biotherapy. 1995;9:1023-1030.
22. Thaker NG, Ali TN, Porter ME, et al. Communicating value in health care using radar charts: case study of prostate cancer. J Oncol Pract. 2016;12:813-820.
23. McCarthy M, Wilkinson D, Gonzalez-Izquierdo A, et al. Measuring quality in cancer services. Report for the National Co-ordinating Centre for NHS Service Delivery
and Organisation R & D (NCCSDO). www.researchgate.net/publication/228673375. Accessed February 20, 2020.
24. Leonard K, Abramovitch A. Cognitive functions in young adults with generalized anxiety disorder. Eur Psychiatry. 2019;56:1-7.
25. Le DT, Uram JN, Wang H, et al. PD-1 blockade in tumors with mismatch repair deficiency. N Engl J Med. 2015;372:2509-2520.
26. Kim MS, Prasad V. Assessment of accuracy of waterfall plot representations of response rates in cancer treatment published in medical journals. JAMA Netw
Open. 2019;2:e193981.
27. Pe M, Dorme L, Coens C, et al; Setting International Standards in Analyzing Patient-Reported Outcomes and Quality of Life Endpoints Data Consortium
(SISAQOL). Statistical analysis of patient-reported outcome data in randomised controlled trials of locally advanced and metastatic breast cancer: a systematic
review. Lancet Oncol. 2018;19:e459-e469.
28. Institute of Medicine. Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: National Academy Press; 2001.
29. Matthies LM, Taran FA, Keilmann L, et al. Electronic patient-reported outcome tool for the FACT-B (Functional Assessment of Cancer Therapy-Breast)
questionnaire for measuring the health-related quality of life in patients with breast cancer: reliability study. J Med Internet Res. 2019;21:e10004.
30. Wallwiener M, Matthies L, Simoes E, et al. Reliability of an e-PRO tool of EORTC QLQ-C-30 for management of health-related quality of life in patients with breast
cancer: prospective randomized trial. J Med Internet Res. 2017;19:e322.
31. Dueck AC, Mendoza TR, Mitchell SA, et al; National Cancer Institute PRO-CTCAE Study Group. Validity and reliability of the US National Cancer Institute’s patient-
reported outcomes version of the common terminology criteria for adverse events (PROO-CTCAE). JAMA Oncol. 2015;1:1051-1059.
32. Berger ML, Curtis MD, Smith G, et al. Opportunities and challenges in leveraging electronic health record data in oncology. Future Oncol. 2016;12:1261-1274.
33. Curtis MD, Griffith SD, Tucker M, et al. Development and validation of a high-quality composite real-world mortality endpoint. Health Serv Res. 2018;
53:4460-4476.
34. Khozin S, Blumenthal GM, Pazdur R. Real-world data for clinical evidence generation in oncology. J Natl Cancer Inst. 2017;109:djx187.
35. Sherman RE, Anderson SA, Dal Pan GJ, et al. Real-world evidence - what is it and what can it tell us? N Engl J Med. 2016;375:2293-2297.
36. Oxnard GR, Morris MJ, Hodi FS, et al. When progressive disease does not mean treatment failure: reconsidering the criteria for progression. J Natl Cancer Inst.
2012;104:1534-1541.
37. Varella L, Eziokwu AS, Jia X, et al. Real-world clinical outcomes and toxicity in metastatic breast cancer patients treated with palbociclib and endocrine therapy.
Breast Cancer Res Treat. 2019;176:429-434.

Seidman et al
38. Code of Federal Regulations. §21-314.126. https://www.ecfr.gov/cgi-bin/text-idx?SID=8eea61dec0ef9b17f1fa0d0d3a7736c6&mc=true&node=se21.5.314_

1126&rgn=div8. Accessed January 7, 2020.
39. Code of Federal Regulations. §314.510. https://www.ecfr.gov/cgi-bin/text-idx?SID=7e3afa31ba9326a86c1c08ca7cb22f4c&mc=true&node=se21.5.314_
1510&rgn=div8. Accessed January 7, 2020.
40. US Food and Drug Administration. Guidance for Industry: Expedited Programs for Serious Conditions–Drugs and Biologics. 2014. www.fda.gov/downloads/
drugs/guidancecomplianceregulatoryinformation/guidances/ucm358301.pdf. Accessed January 7, 2020.
41. Blumenthal GM, Pazdur R. Response rate as an approval end point in oncology: back to the future. JAMA Oncol. 2016;2:780-781.
42. US Food and Drug Administration. Guidance for Industry: Clinical Trial Endpoints for the Approval of Cancer Drugs and Biologics. 2008. www.fda.gov/regulatory-
information/search-fda-guidance-documents/clinical-trial-endpoints-approval-cancer-drugs-and-biologics. Accessed January 9, 2020.
43. Beaver JA, Howie L, Pelosof L, et al. A 25-year experience of US Food and Drug Administration accelerated approval of malignant hematology and oncology drugs
and biologics: a review. JAMA Oncol. 2018;4:849-856.
44. U.S. Department of Health and Human Services. Placebos and blinding in randomized controlled cancer clinical trials for drug and biological products: guidance
to industry. www.fda.gov/media/130326/download. Accessed January 9, 2020.
45. Zhang J, Liang W, Liang H, et al. Endpoint surrogacy in oncological randomized controlled trials with immunotherapies: a systematic review of trial-level and arm-
level meta-analyses. Ann Transl Med. 2019;7:244.
46. Ritchie G, Gasper H, Man J, et al. Defining the most appropriate primary endpoint in phase 2 trials of immune checkpoint inhibitors for advanced solid cancers:
a systematic review and meta-analysis. JAMA Oncol. 2018;4:522-528.
47. Sun J, Wei Q, Zhou Y, et al. A systematic analysis of FDA-approved anticancer drugs. BMC Syst Biol. 2017;11 (suppl 5):87.
48. U.S. Food and Drug Administration. Hematology/oncology (cancer) approvals & safety notifications. www.fda.gov/drugs/resources-information-approved-drugs/
hematologyoncology-cancer-approvals-safety-notifications. Accessed January 10, 2020.
49. Kluetz PG, Slagle A, Papadopoulos EJ, et al. Focusing on patient-reported outcomes in cancer clinical trials: symptomatic adverse events, physical function, and
disease-related symptoms. Clin Cancer Res. 2016;22:1553-1558.
50. Saad ED, Katz A, Hoff PM, et al. Progression-free survival as surrogate and as true end point: insights from the breast and colorectal cancer literature. Ann Oncol.
2010;21:7-12.
51. Spears P. Presentation to the Metastatic Breast Cancer Alliance Patient Advocate Advisory Group. www.youtube.com/watch?v=TkojulLYShI&feature=youtu.be.
Accessed on January 23, 2019.
52. Unger JM, Vaidya R, Hershman DL, et al. Systematic review and meta-analysis of the magnitude of structural, clinical, and physician and patient barriers to
cancer clinical trial participation. J Natl Cancer Inst. 2019;111:245-255.
53. Rugo HS, Olopade OI, DeMichele A, et al; I-SPY 2 Investigators. Adaptive randomization of veliparib-carboplatin treatment in breast cancer. N Engl J Med. 2016;
375:23-34.
54. Wason JMS, Abraham JE, Baird RD, et al. A Bayesian adaptive design for biomarker trials with linked treatments. Br J Cancer. 2015;113:699-705.
55. Propublica. Black patients miss out on promising cancer drugs. www.propublica.org/article/black-patients-miss-out-on-promising-cancer-drugs. Accessed
February 20, 2020.
56. U.S. Food and Drug Administration. 2018 Drug trials snapshots: summary report. www.fda.gov/media/120253/download. Accessed February 20, 2020.

BREAST CANCER
What’s the Price? Toxicities of Targeted

Therapies in Breast Cancer Care
Carey K. Anders, MD1; Nicole R. LeBoeuf, MD, MPH2; Lara Bashoura, MD3; Saadia A. Faiz, MD3; Afreen I. Shariff, MD4; and
Alexandra Thomas, MD5
overview
Agents with mechanisms novel to breast cancer care have been approved to treat breast cancer. These agents
include drugs that target cyclin-dependent kinases, phosphoinositide 3-kinase PI3KCA gene mutations,
PARP, checkpoint regulation, and novel antibody-drug conjugates. However, these novel approaches bring
a risk of toxicities quite different from those of conventional cytotoxic chemotherapy. Here, we review these
agents and discuss related adverse events, with particular attention to endocrine, pulmonary, and derma-
tologic toxicities. Endocrine toxicities associated with novel cancer therapies for breast cancer are distinct and
often present with symptoms related to the specific hormonal deficiencies and rarely hormonal excess. Given
the complex and sometimes irreversible nature of these toxicities, once recognized, transdisciplinary
management with an endocrinologist experienced with managing drug-related toxicities is encouraged. Drug-
related pneumonitis is a serious concern with new targeted therapies. Presentation may not be easily dis-
tinguished, and a multidisciplinary team approach can optimize patient care. Heightened awareness is crucial
for early detection and treatment. Management should follow recommendations provided by the National
Cancer Institute Common Terminology Criteria for Adverse Events and agent-specific guidelines. Cutaneous
toxicities from anticancer therapies represent a common and often poorly characterized challenge for patients
with breast cancer. Although our understanding of dermatologic effects from novel therapies continues to
improve, the breadth of toxicities spans all dermatologic conditions. Targeted therapies offer effective and
often novel therapeutic strategies for patients with breast cancer but also bring new adverse event profiles. In
this era, it will be important both to closely follow monitoring recommendations and to remain vigilant for
emerging toxicities.
INTRODUCTION women with advanced hormone receptor (HR)–

In recent years, several therapies with disease targets positive, HER2-negative breast cancer. Multiple
new to breast oncology have been approved for use. studies have demonstrated improved progression-free
This era of newer targeted agents brings the promise of survival (PFS) with the addition of a CDK inhibitor to
enhanced therapeutic efficacy, increasing our ability antiestrogen therapy, and more recent results revealed
both to cure breast cancer and to prolong survival for an overall survival benefit with the addition of
patients with advanced disease. These newer treat- ribociclib1 and abemaciclib.2 Importantly, the dedi-
ments, however, can be associated with adverse cated study of premenopausal and perimenopausal
events, some of which are quite distinct from the well- women in the MONALEESA-7 trial found that when
described, generally predictable side effects associ- treated with ovarian function suppression with either
ated with chemotherapy (Tables 1 and 2). Here, we an aromatase inhibitor or tamoxifen and ribociclib,
review newly approved agents with mechanisms novel these women also had overall survival benefit from the
Author affiliations
to breast oncology and discuss the presentation and addition of CDK inhibition.1 A recent meta-analysis
and support
management of toxicities, with particular attention to found that, relative to antiestrogen therapy alone,
information (if
applicable) appear endocrine, pulmonary, and dermatologic toxicities. combinations with palbociclib, ribociclib, and abe-
at the end of this maciclib each had superior PFS with a consistent
article.
OVERVIEW OF NEWER TARGETED THERAPIES IN hazard ratio of approximately 0.40.3 In this work, none
Accepted on March
BREAST ONCOLOGY of the three agents were superior to the others with
2, 2020 and regard to PFS or the portion of patients achieving
published at Cyclin-Dependent Kinase Inhibitors a response to therapy, supporting the practice of
ascopubs.org on May
18, 2020: DOI https://
Cyclin-dependent kinase (CDK) inhibitors, used to- considering an agent’s specific toxicity profile when
doi.org/10.1200/ gether with antiestrogen therapy, have become the selecting a CDK inhibitor for an individual patient.
EDBK_279465 standard in the first- and second-line settings for Abemaciclib is also approved for use as a single agent

Anders et al
which dose reductions and monitoring are recommended,

respectively.10,11
• Recent years have seen new treatments ap- Phosphoinositide 3-Kinase Inhibitors
proved for the treatment of breast cancer, some Alpelisib, a recently approved phosphoinositide 3-kinase
of which have toxicity profiles quite distinct from (PI3K) inhibitor, targets mutations in the PIK3CA gene,
those of traditional antineoplastic therapies. which are observed in approximately 40% of HR-positive
• Endocrine glands are particularly vulnerable to breast cancers.12,13 The SOLAR-1 trial enrolled patients with
immune-related toxicities, resulting in variable advanced HR-positive breast cancer that had progressed on
clinical presentations that are specific to the prior endocrine therapy. The median PFS was 11.0 months
endocrine gland affected. Endocrine toxicities for patients with PIK3CA-mutated tumors who received
can range from subacute to life-threatening fulvestrant with alpelisib compared with 5.7 months for
presentations requiring multidisciplinary care;
those who received fulvestrant with placebo (hazard ratio for
high clinical and biochemical vigilance is rec-
progression or death, 0.65; 95% CI, 0.50–0.85).14 Patients
ommended for timely intervention.
were excluded if they had received prior fulvestrant, and
• Drug-induced pneumonitis is a clinical di- only a small portion of patients had received prior CDK
agnosis, and exclusion of other potential and
inhibitors. The sequencing of therapy targeting PIK3CA-
concomitant etiologies is paramount in
mutated tumors and the role of these agents in over-
management.
coming CDK inhibitor resistance is the subject of ongoing
• dAEs from ICIs cover the entire breadth of in- studies. Notable adverse events seen with alpelisib were
flammatory skin disorders and are best man-
hyperglycemia, rash, and diarrhea, with permanent agent
aged collaboratively with a focus on the specific
discontinuation attributable to adverse events seen for
dermatologic diagnosis presenting as toxicity.
25.0% of patients who received alpelisib compared with
• Following monitoring guidelines and remaining 4.2% of those treated with placebo.14
vigilant for emerging toxicities will be important
in this era of new drugs and drug classes. PARP Inhibitors
Two oral inhibitors of PARP, olaparib and talazoparib, are
available for treatment of breast cancer for patients with
for patients whose disease has progressed while receiving
metastatic breast cancer and deleterious or suspected
endocrine therapy and chemotherapy.
deleterious germline BRCA mutations. Tumors that lack
CDK inhibitors are frequently associated with cytopenias, functional BRCA are vulnerable to PARP inhibition because
particularly neutropenia; in studies of palbociclib and ribo- this “dual hit” loss of DNA repair tools leads to the inability to
ciclib, neutropenia occurred in 75% to 80% of patients, with repair DNA damage, resulting in synthetic lethality.15 In the
grade 3/4 neutropenia seen in more than 50% of patients.4,5 OlympiAD trial, patients who received olaparib, following
Regular, upfront monitoring is recommended and can later be chemotherapy given in the neoadjuvant, adjuvant, or
adjusted.6 Neutropenia associated with palbociclib appears metastatic setting, had a median PFS of 7.0 months
as a result of cell cycle inhibition with resumption of neutrophil compared with 4.2 months (hazard ratio for disease pro-
production with agent withdrawal, likely a class effect, in gression or death, 0.58; 95% CI, 0.43–0.80) for those who
contrast to chemotherapy, which induces neutrophil apo- received single-agent chemotherapy.16 The magnitude of
ptosis.7 In the clinic, unlike with chemotherapy, CDK the benefit was greater for patients with triple-negative
inhibitor–induced neutropenia is infrequently associated with breast cancer than for those with HR-positive breast can-
febrile neutropenia and responds rapidly to drug withdrawal. cer. In the EMBRACA trial, the median PFS was longer for
As such, granulocyte colony-stimulating factors are not rec- those who received talazoparib (8.6 months; hazard ratio for
ommended for routine use with CDK inhibitor–induced disease progression or death, 0.54; 95% CI, 0.41–0.71)
neutropenia. The frequency of grade 3/4 neutropenia also rather than single-agent chemotherapy (5.6 months).17
decreases with extended therapy, suggesting that cumulative Although there did appear to be activity among patients
toxicity is unlikely and that dose adjustments adequately treat who had already received platinum compounds, neither trial
neutropenia.8 Furthermore, CDK inhibitor dose reductions do compared PARP inhibition directly with platinum.
not appear to affect therapeutic efficacy, with the PALOMA-3
Overall, PARP inhibitors were better tolerated than che-
trial showing that dose reduction of palbociclib for grade 3/4
motherapy, and the mostly low-grade adverse events were
neutropenia did not adversely affect PFS.9
managed with dose reductions or supportive care. Rates of
In this drug class, other agent-specific adverse events of treatment discontinuation were low, and cumulative expo-
note include diarrhea with abemaciclib and QTC pro- sure with extended cycles did not increase anemia.18
longation and liver function abnormalities with ribociclib, for Patient-reported outcomes analyses support the improved

What’s the Price? Toxicities of Targeted Therapies in Breast Cancer
TABLE 1. Overview of Adverse Events: Chemotherapy and Biologic Targeted Therapies

Adverse Event Chemotherapy Novel Targeted Therapies
Profile Most patients affected Can be well tolerated, but also associated with uncommon, sometimes serious, adverse events
Limited organ systems Wider range of organ systems can be affected
Understanding Well described Generally known, but evolving
Timing Largely during treatment Variable, and can manifest early, during, and after treatment
Long-term effects well described Long-term effects less well described
Predictability Predictable Less predictable
quality of life for patients treated with PARP inhibitors rel- 25 months (hazard ratio, 0.68; 95% CI, 0.55–0.85) for
ative to chemotherapy.19,20 Both olaparib and talazoparib treatment with T-DM1 relative to treatment with capecita-
are rarely associated with myelodysplastic syndrome or bine and lapatinib.26 Overall survival benefit was also seen
acute myeloid leukemia. Talazoparib did have higher rates with longer follow-up, even with treatment crossover.27
of grade 3/4 hematologic toxicities than single-agent che- T-DM1 has also been approved for adjuvant use in pa-
motherapy, notably anemia, for which dose interruption and tients with residual disease following neoadjuvant taxane
reduction is recommended.21 and trastuzumab-based therapy based on the results of the
KATHERINE trial, which demonstrated improved invasive
Immune Checkpoint Inhibitors
disease-free survival at 3 years of 88.3% with T-DM1 and
Immune checkpoint inhibitors (ICIs) that target tumor im- 77.0% with trastuzumab (hazard ratio, 0.50; 95% CI,
mune evasion have come into widespread use. Earlier work 0.39–0.64).28 DS-8201 links a topoisomerase I inhibitor with
evaluating these agents in breast cancer had led to mixed an HER2-directed antibody and received accelerated ap-
results.22 The IMpassion130 study, however, demonstrated proval by the U.S. Food and Drug Administration (FDA) in
improved PFS of 7.5 versus 5.0 months (hazard ratio, 0.62; December 2019 based on the results of the DESTINY-
95% CI, 0.49–0.78) and overall survival of 25.0 versus Breast01 study. This phase II study, which enrolled 184
15.5 months (hazard ratio, 0.62; 95% CI, 0.45–0.86) in the patients, reported a 60.9% (95% CI, 53.4%–68.0%) re-
first-line setting for metastatic triple-negative breast cancers sponse rate for heavily pretreated patients, with a median
that overexpress PD-L1, favoring treatment with combined response duration of 14.8 months (95% CI, 13.8–16.9).29
atezolizumab and nab-paclitaxel over single-agent nab-
paclitaxel. This landmark study led to the approval of this Both antibody-drug conjugates are associated with rates of
combination.23 left ventricular dysfunction that do not appear to exceed that
of single-agent trastuzumab and similar monitoring is rec-
ICIs can be well tolerated, but they also have the potential for ommended. T-DM1 is associated with frequent low-grade
a wide range of adverse events largely based on their ability hepatic transaminase elevation, and rarely hepatotoxicity,
to unleash systemic autoimmune phenomena. In the IM- thrombocytopenia, and death have occurred. Notably,
passion130 trial, withdrawal of any agent as a result of patients receiving T-DM1 can experience hemorrhage in the
adverse events was 15.9% in the atezolizumab arm and absence of anticoagulation and thrombocytopenia, which
8.2% in the placebo arm. Approaches to the management has rarely been life-threatening. DS801 is more commonly
of toxicities associated with checkpoint blockade are associated with adverse effects seen with cytotoxic che-
addressed in guidelines from ASCO24 and the Society for motherapy, such as cytopenias, nausea, vomiting, diarrhea,
Immunotherapy of Cancer.25 Importantly, immune-mediated and alopecia, suggesting off-target exposure to the anti-
toxicity from checkpoint inhibitors can occur early, during, neoplastic payload. Patients treated with DS8201 in the
and even after treatment with these agents. DESTINY-Breast01 study were at high risk for interstitial
Antibody-Drug Conjugates lung disease, which occurred in 13.6% of patients and
resulted in four deaths (see the section on pulmonary
Two approved antibody-drug conjugates, ado-trastuzumab
toxicities of targeted therapies). Further investigation and
emtansine (T-DM1) and [Fam] trastuzumab deruxtecan
postmarketing experience will be essential to further define
(DS-8201), both harness a monoclonal antibody targeting
the benefits and toxicity profile of this compound.
HER2 to deliver the antineoplastic payload directly to the
tumor. T-DM1 is approved for use in patients who have
previously received a taxane and trastuzumab in the met- ENDOCRINE TOXICITIES OF TARGETED THERAPIES IN
astatic setting or whose disease recurred within 6 months BREAST CANCER CARE
of adjuvant therapy. In this patient population, the EMILIA Historically, and as the survival rate for childhood cancer
trial demonstrated improved overall survival of 31 versus has substantially improved, the management of treatment-

Anders et al
related endocrinopathies has largely been managed by our who present with ICI-related toxicities, including hypergly-
pediatric oncology colleagues who have become adept at cemia (www.nccn.org). For patients who present with hy-
treating the common, late effects of cancer treatments.30-32 perglycemia, one of the first decision points is the presence
In this setting, essentially any endocrine organ can be or absence of DKA. If DKA is diagnosed, recommendations
adversely affected, including the pituitary gland, repro- are made to hold immunotherapy, initiate insulin, and treat
ductive organs, thyroid, pancreas, and adrenal glands. In DKA as per institutional guidelines, with an endocrinology
contrast to management of endocrine toxicities among the consultation in the inpatient setting. Immunotherapy
pediatric age group, treating the rapidly growing number of rechallenge would not be considered until DKA has cor-
adult patients developing these toxicities is a newer but rected and glucose levels have stabilized. This decision
rapidly developing skill set for adult oncologists. Given the to continue therapy with the same agent must be in-
advent of effective targeted agents and immunotherapy dividualized to each patient’s global health, severity of
agents that are associated with notable improvements in toxicity, and cancer burden. For patients who develop hy-
patient outcomes, there is a more recent need for careful perglycemia in the absence of DKA, recommendations are
attention to recognizing and managing “off-target” effects of to continue immunotherapy and administer medical therapy
therapies. For the purposes of this review, the focus will be as per institutional guidelines. The treatment of T1DM is
centered on the incidence and diagnosis, management, similar to standard-of-care practice in which insulin is ini-
and prognosis of hyperglycemia, thyroid, and adrenal tiated at the time of diagnosis. It is recommended that
endocrinopathies associated with cancer therapies. oncologists pre-identify endocrinologists in their practice
who can help guide critical management in these life-
Hyperglycemia threatening scenarios. In contrast, if the patient develops
Immunotherapy and hyperglycemia Targeted therapy for type 2 diabetes mellitus as a consequence of these ther-
breast cancer can have a dual effect on insulin by reducing apies, the treatment can be initiated with consultation with
the secretion of insulin from the pancreatic islet cells, es- an endocrinologist who may recommend lifestyle changes,
sentially acquired type 1 diabetes mellitus (T1DM), or can metformin, or oral hypoglycemic agents with or without
induce insulin resistance by mechanisms similar to those in insulin depending on the degree of hyperglycemia. In either
type 2 diabetes mellitus.33 Globally, targeted breast cancer scenario, signs or symptoms of pancreatic exocrine in-
therapy that leads to reduced insulin secretion falls into the sufficiency should also be monitored and supplemented
immunotherapy category, whereas treatments that lead to with pancrealipase, as needed. Finally, multidisciplinary
insulin resistance are among the inhibitors of the PI3K and team management between oncology, endocrinology, and
mTOR pathway. Specific to immunotherapy, including both nutrition is paramount.
inhibitors of the PD/PD-L and CTLA-4 pathways, these
PI3K inhibition and hyperglycemia Both the phosphatidy-
agents can lead to rapid-onset hyperglycemia within days to
linositide kinase and mTOR are key oncologic signaling
weeks of the first dose of PD inhibition. Although this
pathways important in breast cancer, for which there are
phenomenon is relatively rare and was reported for only one now two FDA-approved agents, alpelisib and everolimus,
patient (0.2%) treated with atezolizumab in the IMpas-
respectively.35 Although targeting these pathways in con-
sion130 trial, it is critical to recognize these symptoms
junction with endocrine therapy is associated with im-
because they can be associated with life-threatening con-
provements in outcome for patients with HR-positive,
sequences such as diabetic ketoacidosis (DKA).34 Inhibitors
HER2-negative advanced breast cancer, targeting this
of PD/PD-L and CTLA-4 pathways block immune inhibitory
pathway also interferes with the normal process of insulin
mechanisms that predispose patients to autoimmunity; in
signaling, leading to insulin resistance. Essentially, following
the setting of treatment-related T1DM, this is attributable
insulin binding to the cell surface insulin α receptor, the
to autoimmune destruction of pancreatic β cells. Overall, tyrosine kinase intracellular component of the insulin β
autoimmune T1DM is more common among those receiving
receptor is activated and phosphorylates insulin receptor
inhibitors of PD/PD-L pathways than among other thera-
substrate 1. Phosphorylated insulin receptor substrate 1,
pies.33 For patients who present with T1DM following receipt
in turn, phosphorylates phosphatidylinositol 4,5-bisphos-
of PD-1 antibody treatment, C-peptide levels are typically
phate, phosphatidylinositol (3,4,5)-trisphosphate, and ulti-
low or undetectable and antibodies such as glutamic acid
mately AKT and protein kinase B. Phosphorylated AKT and
decarboxylase-65 are positive, similar to classic T1DM.34
phosphorylated protein kinase B are required for membrane
Antibodies, when positive, are helpful to establish a di-
translocation of glucose transporter 4, required of insulin-
agnosis, but it is not uncommon to see antibody-negative mediated glucose transport across the cell membrane.
T1DM with immunotherapy.34
Thus, disruption in this pathway through inhibition of PI3K
The National Comprehensive Cancer Network (NCCN) blocks this pathway upstream of glucose transporter 4,
provides systematic guidelines on the treatment of patients resulting in hyperglycemia. In the SOLAR-1 study, the rate of

TABLE 2. Novel Targeted Therapies for Treatment of Breast Cancer

Label Warnings, Precautions,** and
Class Agent Key Trial(s) FDA-Approved Therapeutic Space* Other Notable Effects
CDK inhibitor Abemaciclib MONARCH 2, Advanced HR-positive, HER2-negative breast Diarrhea
MONARCH 3 cancer, with antiestrogen therapy or as single Neutropenia
agent Interstitial lung disease/pneumonitis
Hepatotoxicity
Venous thromboembolism
Increase in creatinine without decrease
in glomerular filtration
Palbociclib PALOMA-2, Advanced HR-positive, HER2-negative breast Neutropenia
PALOMA-3 cancer, with antiestrogen therapy Interstitial lung disease/pneumonitis
Approved for men; precautions for
male patients with female partners
of reproductive potential
Ribociclib MONALEESA-2, Advanced HR-positive, HER2-negative breast Interstitial lung disease/pneumonitis
MONALEESA-7 cancer, with antiestrogen therapy QT interval prolongation
QT interval prolongation with tamoxifen
Hepatobiliary toxicity
Neutropenia
PI3K inhibitor Alpelisib SOLAR-1 Advanced PIK3CA-mutated, HR-positive, HER2- Severe hypersensitivity
negative breast cancer, with antiestrogen Severe cutaneous reactions
therapy Hyperglycemia
Pneumonitis
Diarrhea
PARP inhibitor Olaparib OlympiAD In advanced HER2-negative breast cancer, for Myelodysplastic syndrome/acute
patients with deleterious or suspected myeloid leukemia (, 1.5%)
deleterious germline BRCA mutation after Pneumonitis (, 1%)
chemotherapy in the (neo)adjuvant or metastatic
setting
Talazoparib EMBARCA In advanced HER2-negative breast cancer, for Myelodysplastic syndrome/acute
patients with deleterious or suspected myeloid leukemia (, 0.3%)
deleterious germline BRCA mutation Myelosuppression
Anemia, grade 3/4 (39%)
Thrombocytopenia, grade 3/4 (15%)
Immune Atezolizumab IMpassion130 Advanced triple-negative breast cancer in Immune-mediated pneumonitis
checkpoint combination with nab-paclitaxel for tumors that Immune-mediated hepatitis
inhibitor express PD-L1 with the companion diagnostic Immune-mediated colitis
test Immune-mediated endocrinopathies
Infections
Infusion-related reactions
Antibody-drug Ado-trastuzumab EMELIA, (i) In HER2-positive, advanced breast cancer, for Hepatotoxicity
conjugate emtansine KATHERINE patients who previously received trastuzumab Cardiac toxicity
and a taxane Pulmonary toxicity
(ii) In HER2-positive early breast cancer, for Infusion-related reactions
adjuvant treatment for patients with residual Hemorrhage
invasive disease after neoadjuvant treatment Thrombocytopenia
Neurotoxicity
[Fam-] trastuzumab DESTINY- Advanced HER2-positive breast cancer, after two Interstitial lung disease
deruxtecan (DS- Breast01 or more anti-HER2–based regimens in the Neutropenia
8201) metastatic setting Left ventricular dysfunction
Abbreviations: FDA, U.S. Food and Drug Administration; CDK, cyclin-dependent kinase; HR, hormone receptor; PI3K, phosphoinositide 3-kinase.
*Advanced indicates local advanced unresectable and/or metastatic.
**All also include embryo-fetal toxicity.

Anders et al
hyperglycemia of any grade was 63.7%, and the rate of Other Endocrinopathies
grade 3 hyperglycemia was 32.7% for patients receiving Although a detailed review is beyond the scope of this
alpelisib. Moreover, grade 4 hyperglycemia was seen in article, it is important to mention other endocrinopathies
3.9% of patients who received alpelisib versus 0.3% in the associated with immunotherapy in the treatment of breast
placebo arm of the SOLAR-1 study. cancer, specifically abnormalities with the thyroid, adrenal,
Given the frequency of hyperglycemia with alpelisib, the and pituitary glands. In the IMpassion130 study, among
medical oncology community is becoming more facile at the immune-related adverse events (irAEs) of special in-
expecting and managing this toxicity. Prior to treatment terest, any-grade hypothyroidism was reported for 78 of
initiation, it is important to counsel patients to follow the 452 patients (17.3%) treated with atezolizumab, com-
American Diabetes Association recommendations and to pared with 19 of 438 patients (4.3%) in the placebo arm.23
provide monitoring equipment to help temporize and There were no grade 3 or 4 events in either arm. Hyper-
identify hyperglycemia. 36 Although there is no formal thyroidism, any grade, was reported in 4.4% of the patients
guideline for management of PI3K inhibitor–associated who received atezolizumab and 1.4% of those who re-
hyperglycemia, recommendations from peer-reviewed received placebo; one patient (0.2%) who received atezo-
views and institutional experience provide the bases for our lizumab had grade 3 and 4 hyperthyroidism, but no
recommendations. Once therapy is initiated, even mild hyperthyroidism was reported among patients in the pla-
increases in fasting plasma glucose ( 126 mg/dL) should cebo arm. Immune-related adrenal insufficiency was re-
prompt initiation of 500 mg of metformin XR once daily with ported in four patients (1%, all grade) and one patient
titration to a maximum dose of 2,000 mg daily. Many on- (0.2%, grade 3 or 4) in the atezolizumab group but not in
cology providers are initiating 500 mg of metformin XR for patients in the placebo group. There were no reports of
patients at risk for hyperglycemia 7 days prior to the initiation immune-related hypophysitis. With the exception of hy-
pothyroidism, these events are quite rare but should be
of alpelisib; the XR formulation is favored to minimize the
considered for patients with advanced breast cancer il-
overlapping toxicities of diarrhea and nausea associated
lustrating signs or symptoms of each of these respective
with both agents. If fasting plasma glucose is maintained
endocrinopathies. The NCCN guidelines provide excellent
below 250 mg/dL, continuing alpelisib at the same dose is
guidance on thyroid replacement, thyroid suppres-
acceptable, but additional therapy should be started in
sion, and management of adrenal and pituitary hor-
consultation with an endocrinologist. If glucose levels re-
mone replacement (www.nccn.org). Finally, frequent
main persistently high (i.e., 250 mg/dL), then alpelisib
communication with our endocrinology colleagues is
therapy should be interrupted until the levels improve and
encouraged.
therapy should be reinitiated at lower doses. At any point
when desired blood glucose targets are not achieved, the PULMONARY TOXICITIES OF TARGETED THERAPIES IN
addition of an insulin-sensitizing agent such as a pioglita- BREAST CANCER CARE
zone and insulin-sparing agents like sodium glucose Targeted therapy has emerged as the preferred treatment of
transporter-2 inhibitors (e.g., canagliflozin, empagli- breast cancer, but data regarding pulmonary toxicities are
flozin), dipeptidyl peptidase 4-inhibitors (e.g., sitagliptin), sparse. In this section, we will discuss the clinical pre-
and glucagon-like peptide-1 receptor agonists (e.g., daily sentation and diagnostic work-up for pulmonary toxicities
liraglutide, weekly semaglutide) can be considered prior to and review the medical literature available to date.
initiation of insulin.
Pulmonary toxicity may present in various ways. Respiratory
If fasting blood glucose does not improve to 126 mg/dL or symptoms may present insidiously or acutely, and com-
lower within 21 days of aggressive treatment, alpelisib plaints may range from mild cough, wheezing, dyspnea,
should be discontinued. Upon discontinuation of alpelisib, or pleurisy to severe hypoxic respiratory insufficiency.37
a careful taper of hypoglycemic agents and insulin is re- Nonspecific symptoms, including fatigue, fever, and de-
quired to avoid hypoglycemia, as blood glucose may nor- creased appetite, may also occur. Along with a thorough
malize within 24 to 72 hours of alpelisib discontinuation, in history and physical examination, the initial work-up typi-
our experience. Close monitoring and patient education is cally involves imaging with a chest radiograph followed by
advised. Patients with severe hyperglycemia (i.e., plasma CT. Drug-induced pneumonitis often involves both lungs,
glucose 500 mg/dL) should be evaluated for DKA (al- and infiltrates can be described as ground glass or con-
though rare in this setting) and aggressively treated with solidative. These parenchymal processes may be diffuse or
hydration and insulin therapy as well as electrolyte repletion localized, and they can manifest with an upper or lower lobe
in the inpatient setting. If fasting blood glucose does not predominance.38 It is imperative to evaluate for other po-
improve within 24 hours, treatment with alpelisib should be tential or concomitant processes such as aspiration or in-
permanently discontinued.36 fectious pneumonia, pulmonary edema, radiation-induced

FIGURE 1. Radiographic Images

of Drug-Induced Pneumonitis
(A) Localized consolidation (black
arrow) with air bronchograms
consistent with organizing pneu-
monia or infectious pneumonia.
(B) Ground glass opacities on the
right lung base, and interstitial
infiltrate in the left lung base with
fibrotic changes (black arrow-
heads) consistent with non-
specific interstitial pneumonitis.
(C) Diffuse alveolar damage with
interstitial edema, subtle ground
glass (right upper lobe, black
arrowhead), and dense alveolar
infiltrate (left upper lobe, black
arrow). (D) Chronic radiation-
induced lung injury with sub-
pleural blebs and scarring (left
upper lobe, four small arrows)
and new ground glass infiltrates
(right upper lobe, arrow).
lung injury, and progression of underlying malignancy. disease, so they will be considered the same entity for the
Frequently, infection versus drug-induced pneumonitis purpose of this review. Interstitial lung disease usually
becomes the final competing diagnosis. Bronchoscopy refers to a broad category of diseases characterized by
with bronchoalveolar lavage may help confirm infection, various amounts of infiltration and pathologic changes
but declaring pulmonary toxicity is a clinical diagnosis. within the lung resulting in fibrosis or scarring.39 These
Histopathologic specimens (via transbronchial biopsy inflammatory changes can occur because of a variety of
or other modality) are not required, but they may be useful sources but may also be idiopathic. The patterns are
distinguished based on specific radiographic distribution
in the evaluation of focal or nonresolving opacities.
of infiltrates and pathologic correlation (Fig. 1). Interstitial
Pulmonary function testing provides an additive test to
lung disease patterns commonly encountered after inciting
assess lung injury objectively and discern response in its
exposures include nonspecific interstitial pneumonitis, dif-
management.37
fuse alveolar damage, organizing pneumonia, and hyper-
Pulmonary toxicities from various therapies have been sensitivity pneumonitis.39,40 The pathophysiology of drug-
described as drug-induced pneumonitis or interstitial lung induced pneumonitis is not well understood. It is thought to
TABLE 3. Grading for Pneumonitis* per the National Cancer Institute Common Terminology Criteria for Adverse Events
Grade
Grade Grade 1 Grade 2 Grade 3 Grade 4 5
Presentation Asymptomatic Symptomatic, limiting Severe symptoms, limiting self-care Life-threatening respiratory Death
instrumental activities of activities of daily living insufficiency
daily living
Intervention No intervention Medical intervention indicated Medical intervention indicated (hospital Urgent medical intervention
indicated admission, oxygen supplementation) indicated (ventilatory support)
Observation only Cessation or dose adjustment Discontinue therapy Discontinue therapy
Consider corticosteroid use Corticosteroid use Corticosteroid use or other
immunosuppressive treatment
*Pneumonitis is a disorder characterized by inflammation focally or diffusely affecting lung parenchyma. Grading is based on the National Cancer Institute
Common Terminology Criteria for Adverse Events (version 5.0).41

Anders et al
case reports also highlighted the importance of early de-

tection given this potentially fatal complication.45,46
In early 2019, the Japanese Pharmaceutical and Medical
Devices Agency reported that abemaciclib was associated
with 14 serious cases of pneumonitis and three of the cases
contributed to death.47 At the end of 2019, the FDA
reviewed all clinical trials for the three approved CDK4/6
inhibitors and found that pneumonitis was identified in 1%
to 3% of cases and less than 1% had a fatal outcome.48
Subsequently, the FDA package insert for adverse events was
updated to include rare but life-threatening pneumonitis.
PI3K Inhibitors
PI3K inhibitors used in the treatment of hematologic ma-
lignancies have been associated with serious and fatal
pneumonitis for up to 5% of patients receiving these
medications.49 Alpelisib was recently approved for treat-
ment of PIK3CA-mutated HR-positive advanced breast
cancer. In large clinical trials, reported pulmonary toxicity
was rare with all grades of pneumonitis (0.7% in grade 3, 0.
4% in grade 4).14 The FDA warned that pneumonitis oc-
curred for 1.8% of patients receiving alpelisib. The most
recent guidelines recommend monitoring patients for re-
FIGURE 2. Eczematous Patches on the Trunk, Presenting With Pruritus spiratory symptoms while they are receiving therapy,
and Background Xerosis in the Setting of PD-1 Inhibitor Therapy withholding medication, and administering treatment with
steroids in cases of severe pneumonitis.
be related to a direct or indirect cytotoxic effect, oxidative Immune Checkpoint Inhibitors

injury from free oxygen radicles, or immune-mediated
mechanisms. ICIs remove the inhibitory effects of the Immunotherapy has substantially improved survival while
immune system, which augments the immune response simultaneously exposing patients to the potential for toxic-
against the tumor but may also result in immune-mediated ities and irAEs.37 ICI-related pneumonitis is a rare but po-
side effects. Severity of pneumonitis is graded according to tentially life-threatening event. In large clinical trials, the
the National Cancer Institute Common Terminology Cri- incidence of ICI-related pneumonitis is 3% to 5% but
teria for Adverse Events (Table 3).41 Management includes can be up to 10% among those receiving combination
cessation or dose reduction of the associated agent, im- therapy.38,50,51
munosuppressive therapy, and supportive measures. Risk factors for developing pneumonitis include prior ra-
Rechallenge with the offending agent may be considered, diation therapy to the chest, history of lung disease, use
but discussion regarding substantial risk versus benefit of combination therapy, and tumor histology type. In-
must occur prior to the treatment.37 Further studies are terestingly, the risk of pneumonitis has not been dose de-
needed to better understand the pathophysiology, in- pendent, and there was no association with tobacco use.52
cidence, and therapy of pneumonitis related to the Reported radiographic patterns include organizing pneu-
classes of medications discussed below. monia, nonspecific interstitial pneumonitis, and rarely dif-
fuse alveolar damage for patients presenting with acute
CDK4/6 Inhibitors
pneumonitis. The severity of pneumonitis is variable, but
Initial large studies evaluating the effect of CDK4/6 inhibitors reported cases were mainly low grade, improved with
did not report pneumonitis as an adverse event. However, in cessation of treatment, and resolved with immunosup-
2018, Gong et al42 described grade 3 pneumonitis with pressive therapy.52 Despite the initial observations that irAEs
palbociclib requiring dose modification, and this resolved could be associated with improved survival, a recent report
within 3 months of drug withdrawal. Subsequent updates to evaluating the effect of development of immune-related
the MONALESSA-2 and MONALESSA-3 trials reported rare pneumonitis found that this was associated with an in-
deaths related to acute respiratory failure.43,44 Additional creased risk of death.53

FIGURE 3. Lichenoid Eruptions

(A) Violaceous, flat-topped scaly
papules consistent with a lichenoid
eruption secondary to PD-1 in-
hibitory therapy. (B) Checkpoint
inhibitor–induced bullous lichen
planus–like eruption. (C) Thick,
violaceous papules and plaques
of hypertrophic lichen planus–
like eruption. (D) Erosive variant
of lichen planus–like toxicity
In the IMpassion130 trial, 14 patients (3%) were diagnosed rates were much higher with DS8201, with all-grade
with immune-related pneumonitis and only one patient had pneumonitis occurring in 13.6%, grade 3 in 0.5%, and
grade 3 or 4 pneumonitis.23 In the second interim analysis, death in 2.2% of patients.29 The median time of onset was
these numbers slightly increased; 16 patients (4%) had any- 193 days (range, 42–535). During the study, all potential
grade pneumonitis and two patients (, 1%) had grade 3 or drug-related pneumonitis cases were reviewed by an in-
4 pneumonitis.54 These rates are consistent with previously dependent, multidisciplinary committee, and a detailed
published data from large lung cancer trials.55-57 algorithm established according to published guidelines
Practice guidelines for the diagnostic work-up and man- was followed for any patient with respiratory symptoms.60
agement of immune irAEs were recently published to At the time of analysis, 20 patients who had grade 2
standardize therapy.58 A multidisciplinary approach to op- pneumonitis or higher were treated with corticosteroids;
timize patient care has been endorsed.59 After resolution of seven of these patients required hospitalization. Because of
pneumonitis, current guidelines recommend rechallenge the high rate of fatal pneumonitis, close monitoring of re-
with therapy as an option for grade 2 pneumonitis but only in spiratory symptoms with high suspicion for drug-related
select cases of grade 3. Rechallenge is not recommended interstitial lung disease is recommended.
for grade 4 pneumonitis. Further research is needed to
DERMATOLOGIC TOXICITIES OF TARGETED THERAPIES IN
evaluate biomarkers that can either identify at-risk patients
BREAST CANCER CARE
or support the diagnostic work-up to confirm pneumonitis.
Dermatologic adverse events (dAEs) from targeted anti-
Antibody-Drug Conjugates cancer agents are common, affect up to 80% of patients,
Pneumonitis has also been described with antibody-drug lead to dose alterations, and affect treatment outcomes
conjugates. Reported rates with T-DM1 were 1.2% in the and quality of life.61-64 In breast cancer care, dAEs are less
EMILIA trial and 1.1% in the KATHERINE trial.26,27 These frequent with CDK and PARP inhibitors. Toxicities from

Anders et al

The widespread use of ICIs has allowed for improved
understanding of patterns and classification and man-
agement of toxicities from these agents. Diverse irAEs may
affect any organ system,66,67 with the skin being the earliest
and most common organ affected. dAEs from PD-L1 in-
hibitor monotherapy occur for about 20% of patients,
whereas anti–CTLA-4 monotherapy is associated with
higher rates of dAEs (44%–59%), anti–PD-1 therapy is
intermediate (34%–42%), and combination therapy is as-
sociated with the highest incidence (59%–72%).68,69 These
skin toxicities span T-cell– and antibody-mediated dis-
orders and may occur during or after cessation of
therapy; proposed mechanisms include flares of pre-
existing or predisposed organ-specific autoimmunity,
environmental antigenic triggers (infection, other drugs,
microbiome profiles, etc.), or cross-reactivity between tu-
mor- and organ-specific antigens. Although available
guidelines broadly address the approach to irAEs of the
skin, given the extent of possible toxicities, immune
suppression–sparing approaches are generally optimized
through multidisciplinary care.
Pruritus is among the most common toxicities from ICIs,
affecting almost one-half of patients. Itch may occur with or
without rash and may be focal or diffuse.68 Although it is
typically low grade, pruritus can impact quality of life.
The treatment ladder consists of dry skin care with emollients
and antihistamines, followed by trials of doxepin, gabapentin,
pregabalin, naloxone, mirtazapine, or aprepitant,70-72 with the
FIGURE 4. Psoriasis Plaques
approach selected based on comorbidities and prescriber
(A) Well-demarcated red plaques with micaceous scales on the legs,
discretion rather than factors related to itch itself.
consistent with plaque-type psoriasis from checkpoint blockade. (B)
Small papules with white scales consistent with the guttate variant of Primarily captured as “maculopapular rash” in clinical trials,
psoriasis. distinct patterns of inflammatory toxicity are now known to
be associated with ICIs. When appropriately classified,
antibody-drug conjugates mirror the traditional toxicities in
disease-specific interventions can limit systemic steroid use
the skin, hair, and nails associated with their cytotoxic
and allow for mechanism-based interventions. Eczematous
payload. PI3K inhibitors appear to be associated with
eruptions from ICIs have a mean time to occurrence of 6
isoform target-specific skin toxicities, with rash affecting
months, affect nearly 20% of patients, and present as
35.6% of patients treated with the p110α inhibitor alpe-
pruritic, thin, or edematous scaly erythematous macules
lisib.14 With the approval of atezolizumab in combination
and plaques73,74; these eruptions resemble atopic, num-
with nab-paclitaxel and given the ongoing efforts to im-
mular, or contact dermatitis (Fig. 2). Mild cases are man-
prove the response of breast cancer to checkpoint
aged with topical steroids and antihistamines. More
blockade, understanding the presentation of dAEs in this
extensive cases may be managed collaboratively using
setting is of paramount importance and thus covers the
phototherapy and agents such as dupilumab. Systemic
majority of this review. Specific and timely classification
steroids are rarely required.
of cutaneous toxicities, regardless of the offending agent,
allows for integrated and mechanism-targeted management Morbilliform or classic maculopapular eruptions are most
of dAEs, ultimately limiting interruption of the desired an- common for patients treated with CTLA-4 inhibitors
ticancer regimen. Referring oncologists and oncoderma- (29%–50%) 69,75 and are rare with PD-1/PD-L1 inhibitor
tologists frequently disagree as to whether anticancer monotherapy. Morbilliform eruptions tend to occur within
therapy needs be interrupted because of dAEs65 and col- the first 2 to 4 weeks of treatment, and they are usually mild
laboration across specialties is needed to produce optimal and self-limiting with management including topical ste-
outcomes. roids, emollients, and antipruritis agents. Systemic

Among the additional cutaneous dAEs that may occur,

vitiligo occurs most frequently for patients with melanoma, is
associated with prognosis and survival benefit, and requires
no specific treatment beyond sun protection. Bullous
pemphigoid has been reported88-93 up to 20 months after
initiating therapy. It may present as urticarial or eczematous
plaques, usually with discrete tense bullae on an in-
flammatory base (Fig. 5). Bullous pemphigoid almost always
flares and persists with steroid taper, and steroid-sparing
approaches such as rituximab, omalizumab, dapsone,
methotrexate, and plasma exchange94-97 have been used
successfully and should be added early. Rare cases of
cutaneous sclerosis,98,99 dermatomyositis,100,101 cutaneous
lupus,102,103 and vasculitides104-106 have been reported;
FIGURE 5. Checkpoint Inhibitor–Induced Bullous Pemphigoid With Tense beyond systemic steroids, these are approached in col-
Bullae and Inflammatory Hemorrhagic Erosions and Crusts at Sites of laboration with dermatology and rheumatology.
Ruptured Bullae
Importantly, given combination approval, taxanes, including
steroids are reserved for high-grade eruptions or systemic nab-paclitaxel, cause a distinct variant of hand-foot syn-
hypersensitivity. drome that occurs for up to 10% of patients.107 Although
classic hand-foot syndrome and irAEs primarily affect the
Lichenoid eruptions are the most common specific rash
palms and soles, taxanes cause erythematous patches and
type to occur for patients receiving PD-1/PD-L1 inhibitors
scaly plaques on the dorsal surfaces of the hands and
and present as pruritic, flat-topped, often violaceous,
feet.108 Onycholysis or lifting of the nails occurs, which can
discrete papules and plaques on the trunk and extremities
be painful and limit activities of daily living.109 Both skin and
(Fig. 3A).73,74,76 This subtype has been reported as early
nail toxicity can be prevented or decreased substantially by
as 3 days and more than 13 months after the initiation
cooling the hands and feet beginning 15 minutes before,
of ICIs. Rarely, bullous (Fig. 3B), hypertrophic (Fig. 3C),
throughout, and for 15 minutes after infusion.110 Skin
and erosive (Fig. 3D) variants may occur and oral lichenoid
changes can be treated with midpotency topical steroids.
changes may occur with or without cutaneous find-
ings.77-80 Treatment consists of topical corticosteroids for dAEs From Other Targeted Agents
limited disease; oral steroids may be used for severe
PI3Kα is ubiquitously expressed and cutaneous toxicities
variants or for symptom management while steroids and
from alpelisib are common, affecting 36% of patients, with
sparing agents such as phototherapy or acitretin are
10% experiencing high-grade rash.14 Described as mac-
added.68,81
ulopapular, this toxicity has not yet been further classified in
New onset or flares of preexisting psoriasis can occur with the literature. Although PI3Kγ/δ and pan-PI3K inhibitors
ICIs.82-84 All variants of psoriasis have been reported, in- have been reported to induce psoriasis and a pityriasis rubra
cluding classic well-demarcated red plaques with shiny pilaris-like eruption,111,112 perhaps owing to the effect of
white scales (Fig. 4A), guttate papules (Fig. 4B), sebop- inhibiting these isoforms on T regulatory cells, alpelisib (a
soriasis, and palmoplantar, nail, and inverse psoriasis.
Pi3Kα isoform inhibitor) appears to induce a different
Psoriatic arthritis may co-occur.85 Limited disease can be
morphology. Patients present with pink urticarial-appearing
managed with topical steroids and topical vitamin D ana-
macules and papules that may have scales. Unlike true
logs. Narrow-band ultraviolet B therapy can be used for
urticaria, they persist but generally fade over 1 to 2 weeks
more severe or diffuse disease.74,83 Systemic treatments
and respond to topical steroids and/or antihistamines. In our
such as acitretin, methotrexate, and apremilast have also
experience, in the absence of severe symptoms, patients
been used with success.74,83,86,87 Importantly, although
may be rechallenged without recurrence of rash. Pro-
systemic steroids are recommended in the NCCN guidelines
phylactic antihistamines may prevent or diminish the de-
for inflammatory rashes, they are not used in the majority of
velopment of this rash, suggesting a mechanism other than
cases of idiopathic psoriasis because tapering can pre-
a classic delayed type hypersensitivity, perhaps mediated
cipitate a disease flare. Thus, absent total body involvement
via histamine-releasing pathways.
or electrolyte abnormalities, systemic steroids should be
avoided. In severe cases or with concurrent arthritis, the Finally, T-DM1 is associated with cutaneous and mucosal tel-
decision for further biologic therapy should be made angiectases that are histologically consistent with vascular
collaboratively. ectasias. The lesions themselves are of no medical consequence;

Anders et al
however, they may be a manifestation of T-DM1–induced liver specific agents and to remain vigilant for postmarket effects,
injury, which may require elastography to capture.113,114 such as the pulmonary toxicity described years after the
approval of CDK inhibitors. Finally, as we seek to balance
CONCLUSION the promise of therapeutic efficacy with treatment-related
The agents described here offer exciting and effective adverse events, it will be beneficial to develop and maximize
therapeutic strategies to treat breast cancer. The coming the use of predicative biomarkers, biomarkers of response,
years will certainly bring more innovative compounds di- and response-based imaging so we can come closer
rected at these targets as well as at other novel targets. In to delivering precisely the amount of therapy needed for
this era of rapidly emerging new drugs and drug classes, it tumor eradication and ideally no more than the needed
will be critical to follow recommended monitoring for therapy.

1
Division of Medical Oncology, Duke Cancer Institute, Department of Alexandra Thomas, MD, Division of Hematology and Oncology,
Medicine, Duke University School of Medicine, Durham, NC Department of Internal Medicine, Wake Forest University School of
2
Department of Dermatology, Center for Cutaneous Oncology, Dana- Medicine, 1 Medical Center Blvd., Winston-Salem, NC 27157; Twitter:
Farber/Brigham and Women’s Cancer Center, Harvard Medical School, @A_THOMASMD; email: [email protected].
Boston, MA
3
Department of Pulmonary Medicine, University of Texas MD Anderson
Cancer Center, Houston, TX AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
4
Division of Endocrinology, Metabolism, and Nutrition, Department of AND DATA AVAILABILITY STATEMENT
Medicine, Duke University School of Medicine, Durham, NC Disclosures provided by the authors and data availability statement (if
5
Division of Hematology and Oncology, Department of Internal Medicine, applicable) are available with this article at DOI https://doi.org/10.1200/
Wake Forest University School of Medicine, Winston-Salem, NC EDBK_279465.
REFERENCES
1. Im SA, Lu YS, Bardia A, et al. Overall survival with ribociclib plus endocrine therapy in breast cancer. N Engl J Med. 2019;381:307-316.
2. Sledge GW Jr, Toi M, Neven P, et al. The effect of abemaciclib plus fulvestrant on overall survival in hormone receptor-positive, ERBB2-negative breast cancer
that progressed on endocrine therapy-MONARCH 2: a randomized clinical trial. JAMA Oncol. Epub 2019 Sep 29.
3. Giuliano M, Schettini F, Rognoni C, et al. Endocrine treatment versus chemotherapy in postmenopausal women with hormone receptor-positive, HER2-
negative, metastatic breast cancer: a systematic review and network meta-analysis. Lancet Oncol. 2019;20:1360-1369.
4. Hortobagyi GN, Stemmer SM, Burris HA, et al. Ribociclib as first-line therapy for HR-positive, advanced breast cancer. N Engl J Med. 2016;
375:1738-1748.
5. Finn RS, Martin M, Rugo HS, et al. Palbociclib and letrozole in advanced breast cancer. N Engl J Med. 2016;375:1925-1936.
6. Spring LM, Zangardi ML, Moy B, et al. Clinical management of potential toxicities and drug interactions related to cyclin-dependent kinase 4/6 inhibitors in
breast cancer: practical considerations and recommendations. Oncologist. 2017;22:1039-1048.
7. Hu W, Sung T, Jessen BA, et al. Mechanistic investigation of bone marrow suppression associated with palbociclib and its differentiation from cytotoxic
chemotherapies. Clin Cancer Res. 2016;22:2000-2008.
8. Finn RS, Crown JP, Ettl J, et al. Efficacy and safety of palbociclib in combination with letrozole as first-line treatment of ER-positive, HER2-
negative, advanced breast cancer: expanded analyses of subgroups from the randomized pivotal trial PALOMA-1/TRIO-18. Breast Cancer Res.
2016;18:67.
9. Verma S, Bartlett CH, Schnell P, et al. Palbociclib in combination with fulvestrant in women with hormone receptor-positive/HER2-negative advanced
metastatic breast cancer: detailed safety analysis from a multicenter, randomized, placebo-controlled, phase III study (PALOMA-3). Oncologist. 2016;
21:1165-1175.
10. Abemaciclib [package insert]. Indianapolis, IN: Eli Lilly and Company; 2019.
11. Ribociclib [package insert]. East Hanover, NJ: Novartis Pharmaceuticals Corporation; 2019.
12. Cancer Genome Atlas Network. Comprehensive molecular portraits of human breast tumours. Nature. 2012;490:61-70.
13. Goncalves MD, Hopkins BD, Cantley LC. Phosphatidylinositol 3-kinase, growth disorders, and cancer. N Engl J Med. 2018;379:2052-2062.
14. André F, Ciruelos E, Rubovszky G, et al; SOLAR-1 Study Group. Alpelisib for PIK3CA-mutated, hormone receptor-positive advanced breast cancer. N Engl
J Med. 2019;380:1929-1940.

16. Robson M, Im SA, Senkus E, et al. Olaparib for metastatic breast cancer in patients with a germline BRCA mutation. N Engl J Med. 2017;
377:523-533.
17. Litton JK, Rugo HS, Ettl J, et al. Talazoparib in patients with advanced breast cancer and a germline BRCA mutation. N Engl J Med. 2018;
379:753-763.
19. Robson M, Ruddy KJ, Im SA, et al. Patient-reported outcomes in patients with a germline BRCA mutation and HER2-negative metastatic breast cancer receiving
olaparib versus chemotherapy in the OlympiAD trial. Eur J Cancer. 2019;120:20-30.
20. Ettl J, Quek RGW, Lee KH, et al. Quality of life with talazoparib versus physician’s choice of chemotherapy in patients with advanced breast cancer and germline
BRCA1/2 mutation: patient-reported outcomes from the EMBRACA phase III trial. Ann Oncol. 2018;29:1939-1947.
21. Talazoparib [package insert]. New York, NY: Pfizer Laboratories; 2019.
22. Emens LA. Breast cancer immunotherapy: facts and hopes. Clin Cancer Res. 2018;24:511-520.
J Med. 2018;379:2108-2121.
24. Brahmer JR, Lacchetti C, Schneider BJ, et al; National Comprehensive Cancer Network. Management of immune-related adverse events in patients treated with
immune checkpoint inhibitor therapy: American Society of Clinical Oncology clinical practice guideline. J Clin Oncol. 2018;36:1714-1768.
25. Puzanov I, Diab A, Abdallah K, et al; Society for Immunotherapy of Cancer Toxicity Management Working Group. Managing toxicities associated with immune
checkpoint inhibitors: consensus recommendations from the Society for Immunotherapy of Cancer (SITC) Toxicity Management Working Group. J Immunother
Cancer. 2017;5:95.
367:1783-1791.
27. Diéras V, Miles D, Verma S, et al. Trastuzumab emtansine versus capecitabine plus lapatinib in patients with previously treated HER2-positive advanced breast
cancer (EMILIA): a descriptive analysis of final overall survival results from a randomised, open-label, phase 3 trial. Lancet Oncol. 2017;18:732-742.
28. von Minckwitz G, Huang CS, Mano MS, et al; KATHERINE Investigators. Trastuzumab emtansine for residual invasive HER2-positive breast cancer. N Engl
J Med. 2019;380:617-628.
J Med. 2020;382:610-621.
30. Hudson MM, Ness KK, Gurney JG, et al. Clinical ascertainment of health outcomes among adults treated for childhood cancer. JAMA. 2013;309:2371-2381.
31. Brignardello E, Felicetti F, Castiglione A, et al. Endocrine health conditions in adult survivors of childhood cancer: the need for specialized adult-focused follow-
up clinics. Eur J Endocrinol. 2013;168:465-472.
32. Mostoufi-Moab S, Seidel K, Leisenring WM, et al. Endocrine abnormalities in aging survivors of childhood cancer: a report from the Childhood Cancer Survivor
Study. J Clin Oncol. 2016;34:3240-3247.
33. Shariff AI, Syed S, Shelby RA, et al. Novel cancer therapies and their association with diabetes. J Mol Endocrinol. 2019;62:R187-R199.
34. Hughes J, Vudattu N, Sznol M, et al. Precipitation of autoimmune diabetes with anti-PD-1 immunotherapy. Diabetes Care. 2015;38:e55-e57.
35. Simioni C, Martelli AM, Zauli G, et al. Targeting the phosphatidylinositol 3-kinase/Akt/mechanistic target of rapamycin signaling pathway in B-lineage acute
lymphoblastic leukemia: an update. J Cell Physiol. 2018;233:6440-6454.
36. Nunnery SE, Mayer IA. Management of toxicity to isoform alpha-specific PI3K inhibitors. Ann Oncol. 2019;30 (suppl 10):x21-x26.
37. Jain A, Shannon VR, Sheshadri A. Immune-related adverse events: pneumonitis. Adv Exp Med Biol. 2018;995:131-149.
38. Nishino M, Hatabu H, Hodi FS, et al. Drug-related pneumonitis in the era of precision cancer therapy. JCO Precis Oncol. 2017;1:1-12.
39. Travis WD, Costabel U, Hansell DM, et al; ATS/ERS Committee on Idiopathic Interstitial Pneumonias. An official American Thoracic Society/European Re-
spiratory Society statement: update of the international multidisciplinary classification of the idiopathic interstitial pneumonias. Am J Respir Crit Care Med. 2013;
188:733-748.
40. Müller NL, White DA, Jiang H, et al. Diagnosis and management of drug-associated interstitial lung disease. Br J Cancer. 2004;91 (suppl 2):S24-S30.
41. National Cancer Institute. Common Terminology Criteria for Adverse Events (CTCAE), version 5.0. https://ctep.cancer.gov/protocolDevelopment/electronic_
applications/ctc.htm. Accessed March 15, 2020.
42. Gong J, Cho M, Yu KW, et al. A single institution experience with palbociclib toxicity requiring dose modifications. Breast Cancer Res Treat. 2018;168:381-387.
43. Hortobagyi GN, Stemmer SM, Burris HA, et al. Updated results from MONALEESA-2, a phase III trial of first-line ribociclib plus letrozole versus placebo plus
letrozole in hormone receptor-positive, HER2-negative advanced breast cancer. Ann Oncol. 2018;29:1541-1547.
44. Slamon DJ, Neven P, Chia S, et al. Phase III randomized study of ribociclib and fulvestrant in hormone receptor-positive, human epidermal growth factor
receptor 2-negative advanced breast cancer: MONALEESA-3. J Clin Oncol. 2018;36:2465-2472.
45. Ahsan I, Malik F, Jafri S. Palbociclib related pnemotoxicity: a rare side effect. Am J Respir Crit Care Med. 2017;195 (abstr A5546).
46. Jazieh KA, Budd GT, Dalpiaz N, et al. Can CDK4/6 inhibitors cause fatal lung injury? Expert Rev Anticancer Ther. 2019;19:917-919.

Anders et al
47. Japanese Pharmaceuticals and Medical Devices Agency. Investigation results on abemaciclib. www.pmda.go.jp/files/000229644.pdf. Accessed March 15,
2020.
48. U.S. Food and Drug Administration. FDA warns about rare but severe lung inflammation with Ibrance, Kisqali, and Verzenio for breast cancer. www.fda.gov/
drugs/drug-safety-and-availability/fda-warns-about-rare-severe-lung-inflammation-ibrance-kisqali-and-verzenio-breast-cancer. Accessed March 15, 2020.
49. Cheson BD, O’Brien S, Ewer MS, et al. Optimal management of adverse events from copanlisib in the treatment of patients with non-Hodgkin lymphomas. Clin
Lymphoma Myeloma Leuk. 2019;19:135-141.
50. Naidoo J, Wang X, Woo KM, et al. Pneumonitis in patients treated with anti-programmed death-1/programmed death ligand 1 therapy. J Clin Oncol. 2017;
35:709-717.
51. Khunger M, Rakshit S, Pasupuleti V, et al. Incidence of pneumonitis with use of programmed death 1 and programmed death-ligand 1 inhibitors in non-small
cell lung cancer: a systematic review and meta-analysis of trials. Chest. 2017;152:271-281.
52. Suresh K, Voong KR, Shankar B, et al. Pneumonitis in non-small cell lung cancer patients receiving immune checkpoint immunotherapy: incidence and risk
factors. J Thorac Oncol. 2018;13:1930-1939.
53. Suresh K, Psoter KJ, Voong KR, et al. Impact of checkpoint inhibitor pneumonitis on survival in NSCLC patients receiving immune checkpoint immunotherapy.
J Thorac Oncol. 2019;14:494-502.
54. Schmid P, Rugo HS, Adams S, et al; IMpassion130 Investigators. Atezolizumab plus nab-paclitaxel as first-line treatment for unresectable, locally advanced or
metastatic triple-negative breast cancer (IMpassion130): updated efficacy results from a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet
Oncol. 2020;21:44-59.
55. Socinski MA, Jotte RM, Cappuzzo F, et al; IMpower150 Study Group. Atezolizumab for first-line treatment of metastatic nonsquamous NSCLC. N Engl J Med.
2018;378:2288-2301.
J Med. 2018;379:2220-2229.
57. Fehrenbacher L, Spira A, Ballinger M, et al; POPLAR Study Group. Atezolizumab versus docetaxel for patients with previously treated non-small-cell lung cancer
(POPLAR): a multicentre, open-label, phase 2 randomised controlled trial. Lancet. 2016;387:1837-1846.
58. Brahmer JR, Lacchetti C, Thompson JA. Management of immune-related adverse events in patients treated with immune checkpoint inhibitor therapy:
American Society of Clinical Oncology clinical practice guideline summary. J Oncol Pract. 2018;14:247-249.
59. Naidoo J, Zhang J, Lipson EJ, et al. A multidisciplinary toxicity team for cancer immunotherapy-related adverse events. J Natl Compr Canc Netw. 2019;
17:712-720.
60. Kubo K, Azuma A, Kanazawa M, et al; Japanese Respiratory Society Committee for formulation of Consensus statement for the diagnosis and
treatment of drug-induced lung injuries. Consensus statement for the diagnosis and treatment of drug-induced lung injuries. Respir Investig. 2013;
51:260-277.
61. Rosen AC, Balagula Y, Raisch DW, et al. Life-threatening dermatologic adverse events in oncology. Anticancer Drugs. 2014;25:225-234.
62. Rosen AC, Case EC, Dusza SW, et al. Impact of dermatologic adverse events on quality of life in 283 cancer patients: a questionnaire study in a dermatology
referral clinic. Am J Clin Dermatol. 2013;14:327-333.
63. Barbu MA, Niţipir C, Voiosu T, et al. Impact of dermatologic adverse reactions on QOL in oncologic patients: results from a single-center prospective study. Rom
J Intern Med. 2018;56:96-101.
64. Hackbarth M, Haas N, Fotopoulou C, et al. Chemotherapy-induced dermatological toxicity: frequencies and impact on quality of life in women’s cancers. Results
of a prospective study. Support Care Cancer. 2008;16:267-273.
65. Barrios DM, Phillips GS, Freites-Martinez A, et al. Outpatient dermatology consultations for oncology patients with acute dermatologic adverse events impact
anticancer therapy interruption: a retrospective study. J Eur Acad Dermatol Venereol. Epub 2019 Dec 19.
66. Johnson DB, Chandra S, Sosman JA. Immune checkpoint inhibitor toxicity in 2018. JAMA. 2018;320:1702-1703.
67. Postow MA, Sidlow R, Hellmann MD. Immune-related adverse events associated with immune checkpoint blockade. N Engl J Med. 2018;378:158-168.
68. Sibaud V. Dermatologic reactions to immune checkpoint inhibitors: skin toxicities and immunotherapy. Am J Clin Dermatol. 2018;19:345-361.
69. Collins LK, Chapman MS, Carter JB, et al. Cutaneous adverse effects of the immune checkpoint inhibitors. Curr Probl Cancer. 2017;41:125-128.
70. Wu J, Lacouture ME. Pruritus associated with targeted anticancer therapies and their management. Dermatol Clin. 2018;36:315-324.
71. Kwatra SG, Ständer S, Kang H. PD-1 blockade-induced pruritus treated with a mu-opioid receptor antagonist. N Engl J Med. 2018;379:1578-1579.
72. Ito J, Fujimoto D, Nakamura A, et al. Aprepitant for refractory nivolumab-induced pruritus. Lung Cancer. 2017;109:58-61.
73. Hwang SJ, Carlos G, Wakade D, et al. Cutaneous adverse events (AEs) of anti-programmed cell death (PD)-1 therapy in patients with metastatic melanoma:
a single-institution cohort. J Am Acad Dermatol. 2016;74:455-461.
74. Coleman E, Ko C, Dai F, et al. Inflammatory eruptions associated with immune checkpoint inhibitor therapy: a single-institution retrospective analysis with
stratification of reactions by toxicity and implications for management. J Am Acad Dermatol. 2019;80:990-997.
75. Welborn M, Kubicki SL, Garg N, et al. Retrospective chart review of cutaneous adverse events associated with tremelimumab in 17 patients. Am J Clin Dermatol.
2018;19:899-905.

76. Tetzlaff MT, Nagarajan P, Chon S, et al. Lichenoid dermatologic toxicity from immune checkpoint blockade therapy: a detailed examination of the clini-
copathologic features. Am J Dermatopathol. 2017;39:121-129.
77. Shi VJ, Rodic N, Gettinger S, et al. Clinical and histologic features of lichenoid mucocutaneous eruptions due to anti-programmed cell death 1 and anti-
programmed cell death ligand 1 immunotherapy. JAMA Dermatol. 2016;152:1128-1136.
78. Guggina LM, Yanes DA, Choi JN. Inverse lichenoid drug eruption associated with nivolumab. JAAD Case Rep. 2017;3:7-9.
79. Sibaud V, Eid C, Belum VR, et al. Oral lichenoid reactions associated with anti-PD-1/PD-L1 therapies: clinicopathological findings. J Eur Acad Dermatol
Venereol. 2017;31:e464-e469.
80. Strickley JD, Vence LM, Burton SK, et al. Nivolumab-induced lichen planus pemphigoides. Cutis. 2019;103:224-226.
81. Fixsen E, Patel J, Selim MA, et al. Resolution of pembrolizumab-associated steroid-refractory lichenoid dermatitis with cyclosporine. Oncologist. 2019;
24:e103-e105.
82. Johnson A, Shulman L, Kachajian J, et al. Access to care in Vermont: factors linked with time to chemotherapy for women with breast cancer-a retrospective
cohort study. J Oncol Pract. 2016;12:e848-e857.
83. Bonigen J, Raynaud-Donzel C, Hureaux J, et al; Groupe de Recherche sur le Psoriasis and the Groupe Cancérologie Cutanée of the Société Française de
Dermatologie the GEM Resopso, Apsoderm and the Groupe Français de Pneumo-Cancérologie. Anti-PD1-induced psoriasis: a study of 21 patients. J Eur Acad
Dermatol Venereol. 2017;31:e254-e257.
84. Chia PL, John T. Severe psoriasis flare after anti-programmed death ligand 1 (PD-L1) therapy for metastatic non-small cell lung cancer (NSCLC). J Immunother.
2016;39:202-204.
85. Kostine M, Rouxel L, Barnetche T, et al; FHU ACRONIM. Rheumatic disorders associated with immune checkpoint inhibitors in patients with cancer-clinical
aspects and relationship with tumour response: a single-centre prospective cohort study. Ann Rheum Dis. 2018;77:393-398.
86. Lidar M, Giat E, Garelick D, et al. Rheumatic manifestations among cancer patients treated with immune checkpoint inhibitors. Autoimmun Rev. 2018;
17:284-289.
87. Fattore D, Annunziata MC, Panariello L, et al. Successful treatment of psoriasis induced by immune checkpoint inhibitors with apremilast. Eur J Cancer. 2019;
110:107-109.
88. Hanley T, Papa S, Saha M. Bullous pemphigoid associated with ipilimumab therapy for advanced metastatic melanoma. JRSM Open. 2018;
9:2054270418793029.
89. Kuwatsuka Y, Iwanaga A, Kuwatsuka S, et al. Bullous pemphigoid induced by ipilimumab in a patient with metastatic malignant melanoma after unsuccessful
treatment with nivolumab. J Dermatol. 2018;45:e21-e22.
90. Naidoo J, Schindler K, Querfeld C, et al. Autoimmune bullous skin disorders with immune checkpoint inhibitors targeting PD-1 and PD-L1. Cancer Immunol
Res. 2016;4:383-389.
91. Biolo G, Caroppo F, Salmaso R, et al. Linear bullous lichen planus associated with nivolumab. Clin Exp Dermatol. 2019;44:67-68.
92. Zumelzu C, Alexandre M, Le Roux C, et al. Mucous membrane pemphigoid, bullous pemphigoid, and anti-programmed death-1/programmed death-ligand 1:
a case report of an elderly woman with mucous membrane pemphigoid developing after pembrolizumab therapy for metastatic melanoma and review of the
literature. Front Med (Lausanne). 2018;5:268.
93. Haug V, Behle V, Benoit S, et al. Pembrolizumab-associated mucous membrane pemphigoid in a patient with Merkel cell carcinoma. Br J Dermatol. 2018;
179:993-994.
94. Damsky W, Kole L, Tomayko MM. Development of bullous pemphigoid during nivolumab therapy. JAAD Case Rep. 2016;2:442-444.
95. Lopez AT, Geskin L. A case of nivolumab-induced bullous pemphigoid: review of dermatologic toxicity associated with programmed cell death protein-1/
programmed death ligand-1 inhibitors and recommendations for diagnosis and management. Oncologist. 2018;23:1119-1126.
96. Ridpath AV, Rzepka PV, Shearer SM, et al. Novel use of combination therapeutic plasma exchange and rituximab in the treatment of nivolumab-induced bullous
pemphigoid. Int J Dermatol. 2018;57:1372-1374.
97. Siegel J, Totonchy M, Damsky W, et al. Bullous disorders associated with anti-PD-1 and anti-PD-L1 therapy: a retrospective analysis evaluating the clinical and
histopathologic features, frequency, and impact on cancer therapy. J Am Acad Dermatol. 2018;79:1081-1088.
98. Barbosa NS, Wetter DA, Wieland CN, et al. Scleroderma induced by pembrolizumab: a case series. Mayo Clin Proc. 2017;92:1158-1163.
99. Tjarks BJ, Kerkvliet AM, Jassim AD, et al. Scleroderma-like skin changes induced by checkpoint inhibitor therapy. J Cutan Pathol. 2018;45:615-618.
100. Sheik Ali S, Goddard AL, Luke JJ, et al. Drug-associated dermatomyositis following ipilimumab therapy: a novel immune-mediated adverse event associated with
cytotoxic T-lymphocyte antigen 4 blockade. JAMA Dermatol. 2015;151:195-199.
101. Kudo F, Watanabe Y, Iwai Y, et al. Advanced lung adenocarcinoma with nivolumab-associated dermatomyositis. Intern Med. 2018;57:2217-2221.
102. Shao K, McGettigan S, Elenitsas R, et al. Lupus-like cutaneous reaction following pembrolizumab: an immune-related adverse event associated with anti-PD-1
therapy. J Cutan Pathol. 2018;45:74-77.
103. Liu RC, Sebaratnam DF, Jackett L, et al. Subacute cutaneous lupus erythematosus induced by nivolumab. Australas J Dermatol. 2018;59:e152-e154.
104. Daxini A, Cronin K, Sreih AG. Vasculitis associated with immune checkpoint inhibitors-a systematic review. Clin Rheumatol. 2018;37:2579-2584.
105. Padda A, Schiopu E, Sovich J, et al. Ipilimumab induced digital vasculitis. J Immunother Cancer. 2018;6:12.
106. Comont T, Sibaud V, Mourey L, et al. Immune checkpoint inhibitor-related acral vasculitis. J Immunother Cancer. 2018;6:120.

Anders et al
107. Rzepecki AK, Franco L, McLellan BN. PATEO syndrome: periarticular thenar erythema with onycholysis. Acta Oncol. 2018;57:991-992.
108. Lacouture ME. Dermatologic Principles and Practice in Oncology: Conditions of the Skin, Hair, and Nails in Cancer Patients. Hoboken, NJ: John Wiley and Sons;
2014.
109. Lau CP, Hui P, Chan TC. Docetaxel-induced nail toxicity: a case of severe onycholysis and topic review. Chin Med J (Engl). 2011;124:2559-2560.
110. Scotté F, Tourani JM, Banu E, et al. Multicenter study of a frozen glove to prevent docetaxel-induced onycholysis and cutaneous toxicity of the hand. J Clin
Oncol. 2005;23:4424-4429.
111. Dewan AK, Sowerby L, Jadeja S, et al. Pityriasis rubra pilaris-like erythroderma secondary to phosphoinositide 3-kinase inhibition. Clin Exp Dermatol. 2018;
43:890-894.
112. Dewan AK, Gupta S, Bach DQ, et al. Psoriasiform eruptions secondary to phosphoinositide 3-kinase inhibition. JAAD Case Rep. 2019;5:401-405.
113. Sibaud V, Niec RE, Schindler K, et al. Ado-trastuzumab emtansine-associated telangiectasias in metastatic breast cancer: a case series. Breast Cancer Res
Treat. 2014;146:451-456.
114. Milam P, Berger M, Ramaswamy B, et al. Spider telangiectases and palmar erythema as harbingers of structural liver changes in three breast cancer patients on
ado-trastuzumab emtansine. J Clin Aesthet Dermatol. 2019;12:23-26.

BREAST CANCER
Avoiding the Swell: Advances in Lymphedema

Prevention, Detection, and Management
Sarah A. McLaughlin, MD, FACS1; Nicole L. Stout, DPT, CLT-LANA, FAPTA2; and
Mark V. Schaverien, MD, MB, ChB, MRCS, MSc, Med, FRCS(Plast)3
overview
Lymphedema is a feared complication of cancer treatments that negatively affects survivors’ quality of life.
The true incidence of lymphedema is difficult to determine given its long latency period. As the number of
survivors of cancer is increasing in the United States, lymphedema is poised to be a major health concern. The
most noteworthy risk factor for lymphedema is comprehensive lymph node dissection. The last 2 decades have
witnessed a dramatic shift in cancer treatment in an attempt to systematically de-escalate therapeutic in-
terventions, specifically seeking to shift treatment away from routine lymph node dissection in favor of sentinel
node biopsy or radiation strategies, thereby reducing the risk for lymphedema while maintaining survival
outcomes. A growing body of robust evidence supports prospective screening and thereby a prospective
surveillance model (PSM) for early diagnosis and intervention for the prevention and treatment of lymphe-
dema. Finally, investigators are actively evaluating the effectiveness of contemporary surgical procedures in
ameliorating the symptoms and disability of patients with lymphedema and reducing the risk of future
episodes of cellulitis, with outcomes of surgery significantly better than with conservative therapy alone. In this
article, we review the current data surrounding these initiatives.
INTRODUCTION populations without breast cancer. Regardless, given

Lymphedema occurs when protein-rich lymphatic fluid that the National Cancer Institute suggests over 19
accumulates in the interstitial tissue, causing swelling. million survivors of cancer will be living in the United
In the developed world, cancer treatments remain the States by 2024, lymphedema is poised to be a major
most common cause of secondary lymphedema de- public health concern.2
velopment. Clinicians caring for patients with cancer CANCER SURGERY AND RADIATION CONSIDERATIONS
should recognize the clinical manifestations of lym- For years, physicians believed lymph node removal
phedema, as its incidence is likely higher than re- was therapeutic. As such, patients willingly endured
ported. Further, lymphedema confers a profound considerable risk and anxiety for lymphedema,
negative impact on the functional status and overall choosing survival over long-term morbidity. However,
quality of life of survivors of cancer. Current advances since the dawn of the 21st century, landmark clinical
in lymphedema science center around reducing trials in surgical and radiation oncology have chal-
lymphedema risk by minimizing and personalizing lenged this long-held notion that more comprehensive
cancer treatment strategies, prevention and early di- treatment and nodal extirpation equates to improved
agnosis, and novel therapeutic interventions for those survival. Instead, these trialists have argued that most
affected with lymphedema. cancers are systemic diseases, benefit from multi-
Author affiliations The true incidence of lymphedema is unknown, likely modality therapy, allow for personalized approaches,
and support due to its prolonged latency requiring diligent long- and support lymph node removal in most cases as
information (if term follow-up and lack of consistent definitions used a diagnostic tool as opposed to a therapeutic in-
applicable) appear tervention in the setting of early-stage or biologically
by clinicians for diagnosis. Although lymphedema has
at the end of this
long been considered synonymous with breast cancer favorable disease.
article.
Accepted on March treatment, it has been reported after most cancers Further, advancements in cancer treatments and
29, 2020 and requiring lymph node removal.1 It is interesting to note documented low rates of isolated regional nodal re-
published at that incidence of lymphedema varies across cancer currences have reinforced lymph node evaluation as
ascopubs.org on type, even if the affected nodal basin is consistent. a likely diagnostic and not a therapeutic. This con-
April 21, 2020:
Also, lymphedema risk appears higher when the at- clusion is critical, as it is clear that the extent of nodal
10.1200/EDBK_ risk limb is a lower extremity. Finally, few published surgery and the synergistic effect of surgery and ra-
280471 manuscripts on cancer-related lymphedema focus on diation on the affected lymphatic basin remain the

McLaughlin, Stout, and Schaverien
validated these trials, noting the incidence of lymphedema

to be 6.3% with a range of 0% to 23%.3 Interestingly, the
American College of Surgeons Oncology Group (ACOSOG)
• In the developed world, secondary lymphe- Z0010 trial7 reported higher rates of subjective lymphedema
dema is most commonly caused by cancer
after both SLNB (objective 12%; subjective 24%) and ALND
treatments. Although lymphedema is frequently
linked to breast cancer, it can occur after lymph (objective 40%; subjective 41%), which underscore the
nodes have been removed or radiated in any controversies on how lymphedema should be defined and
lymph node basin. the value of both objective and subjective measures. Re-
gardless, it is clear that less extensive axillary surgery results
• Stepwise surgical trials have evaluated the role
of minimizing lymph node dissection in favor of in less lymphedema, and therefore, surgical research has
sentinel node biopsy or radiation therapy alone. focused on how to extend SLNB to more patients.
This de-escalation in care has resulted in Further review of the randomized SLNB clinical trials gave
a lower sustained incidence of lymphedema. way to critical study on the extent of nodal positivity among
• The PSM for lymphedema is a clinically effec- patients proceeding to ALND. Specifically, it was recognized
tive intervention to promote early detection and that in about 70% of patients undergoing ALND, the sentinel
management of lymphedema, as well as a fea- nodes were the only positive lymph nodes. This finding
sible practice for oncology providers. empowered the ACOSOG to investigate whether every pa-
• Surgical treatments can be broadly categorized tient with a positive SLNB benefits from completion ALND.
into physiologic procedures, including lym- The ACOSOG Z0011 trial prospectively and randomly se-
phovenous bypass and VLNT, and debulking lected women undergoing breast-conserving surgery with
procedures, including SAL with controlled
one or two positive sentinel lymph nodes to either com-
compression therapy and direct excision.
pletion ALND or no further axillary surgery.8 They reported
• Evidence supports the efficacy of surgical regional recurrence rates of less than 1% in both arms and
treatments for lymphedema at reducing the no differences between overall survival (SLNB: 86.3% vs.
excess limb volume, reducing the occurrence of
ALND: 83.6%; noninferiority p = .02) or disease-free sur-
cellulitis, decreasing the need for conservative
therapy, improving patient quality of life, and vival (SLNB: 80.2 vs. ALND: 78.2%; p = .32) at 10 years,
improving physical function. thus proving in this population of patients with early-stage
breast cancer that SLNB alone was not inferior to ALND.
Although fiercely debated for several years, Z0011 has since
primary drivers of lymphedema development. Acknowl- been widely adopted after its initial publication in 2011,
edging lymphedema is a serious concern after treatment of further extending the benefits of minimal axillary exploration
many different cancers, a focused review of the stepwise (SLNB) and lower lymphedema risk to more patients.
progression of clinical trials in breast cancer research
After the practice-changing success of Z0011, the next
provides a robust framework of level 1 evidence for the
major question was how to further reduce axillary node
systematic de-escalation of lymph node treatment. Although
treatment to patients presenting with clinically positive
these trials aimed to promote personalized intervention
nodes at diagnosis and the efficacy of SLNB after neo-
strategies and evaluate survival, most also addressed
adjuvant therapy. The ACOSOG 1071, SENTINA, and SN
lymphedema risk and incidence as secondary endpoints.
FNAC trials confirmed the feasibility of SLNB in this
BREAST CANCER SURGICAL TRIALS population.9-11 The subsequent trials (Alliance 11202 and
Axillary dissection remained the standard of care for breast NRG B-51) are now actively evaluating the role of selective
cancer lymph node assessment until the early 2000s. ALND with or without axillary radiation in patients treated
Shaitelman et al3 reviewed 5,354 patients, finding a pooled with neoadjuvant chemotherapy who remain pathologically
incidence of lymphedema of 28% (range 11%–57%) in node positive or downstage to being node negative. Lym-
those undergoing axillary lymph node dissection (ALND). phedema is a secondary endpoint.
The breast cancer community adopted sentinel lymph node Finally, although not currently being evaluated in a pro-
biopsy (SLNB) for axillary assessment around the year spective, randomized clinical trial, the benefit of targeted
2000. The randomized controlled trials designed to evaluate axillary dissection is being studied by breast surgeons.
the efficacy of SLNB (all mandated backup ALND) not only Specifically, patients presenting with a clinically positive
demonstrate accurate axillary staging but also less lym- axillary node undergo needle biopsy to confirm metastatic
phedema after SLNB when compared with ALND, 0% to 7% disease and to leave a marking clip. After neoadjuvant
versus 12% to 16%, respectively.4-6 A pooled analysis of chemotherapy, the node is localized for surgical excision,
6,711 patients with breast cancer undergoing SLNB and targeted axillary dissection is performed in addition to

SLNB. Targeted axillary dissection reduces the false- disease, or receipt of adjuvant systemic therapy (all p ,
negative rate of SLNB in these high-risk patients and .02).12
therefore helps better target which patients might benefit MELANOMA SURGERY TRIALS
from completion ALND due to chemotherapy-resistant re-
sidual disease.9 Its effect on lymphedema has not specifi- Melanoma research focusing on de-escalation nodal
cally been documented but it is expected to parallel that of strategies to minimize treatment morbidity has paralleled
SLNB. This evolution in breast cancer clinical trials serves as breast cancer research. Overall, the pooled incidences of
the gold standard for intervention de-escalation strategies lymphedema after melanoma surgery are reported as 4.1%
documenting sustained excellent survival rates while sys- after SLNB, 3% after ALND, and 18% after inguinofemoral
tematically reducing lymphedema risk lymph node dissection.3 Hyngstrom et al13 prospectively
followed 182 patients with melanoma for 12 months and
BREAST CANCER RADIATION TRIALS found, similar to breast cancer data, that lymph node dis-
section resulted in significantly more lymphedema than
Although surgeons have focused on de-escalation of sur-
SLNB (odds ratio 3.18; p , .01). Data from the Multicenter
gery, radiation oncologists have investigated the equipoise
Selective Lymphadenectomy Trial I (MSLT 1) prospectively,
of axillary radiation and axillary dissection in the setting of
randomly selected patients undergoing wide excision of
a positive sentinel node. The After Mapping of the Axilla:
a primary melanoma to receive SLNB or observation alone.
Radiotherapy Or Surgery? (AMAROS) trial10 randomly se-
Those with a positive SLNB completed immediate or early
lected patients with a positive SLNB to recevie standard
lymph node dissection.14 If patients developed a regional
ALND or axillary radiation therapy. The trial reported sig-
nodal recurrence, they completed delayed lymph node
nificantly less lymphedema after axillary radiation therapy
dissection. This allowed restriction of lymph node dissection
(objective 5%; subjective 11%) when compared with ALND
to only those patients who might benefit, again validating
(objective 13%; subjective 23%). A recent prospective
SLNB. At 5-year median follow-up, patients undergoing
study of 1,811 patients evaluated breast cancer–related
immediate/early lymph node dissection had less lymphe-
lymphedema (BCRL) risk according to extent of axillary
dema than those undergoing delayed lymph node dissec-
surgery and axillary regional lymph node radiation
tion (12.4% vs. 20%; p = .04). Further, inguinofemoral
(RLNR).11 At 5 years, the authors found the following
dissection was significantly more likely to result in lym-
cumulative lymphedema incidences: SLNB alone, 7.7%;
phedema than ALND (26% vs. 9%; p , .001). The DeCoG
SLNB plus RLNR, 10.8%; ALND alone, 29%; and ALND
and MSLT 2 trials build upon the results of MSLT 1.
plus RLNR, 38.7%. Interestingly, in the setting of similar
local control rates, multivariable analysis found no note- EMERGING STRATEGIES FOR SURGICAL PREVENTION
worthy difference in lymphedema rates between axillary OF LYMPHEDEMA
surgery groups regardless of use of RLNR, and the ALND Beyond limiting surgical dissection, surgeons are actively
groups consistently had higher lymphedema risks than the investigating prevention techniques that can be combined
SLNB groups. These data lend further support to the with routine axillary surgery to further limit lymphedema risk.
findings of the AMAROS trial, documenting that in patients Axillary reverse mapping (ARM) seeks to preserve upper-
with only one or two positive sentinel lymph nodes, lym- extremity lymphatics and nodes, identifying them separately
phedema could be significantly reduced if axillary regional from those draining the breast and thereby reducing lym-
nodal radiation replaced ALND. phedema. The surgeon isolates the lymph nodes draining
The combination of ALND and comprehensive nodal ra- the breast with technetium and those draining the arm with
diation can act synergistically to nearly double the risk of blue dye. A systematic review found lymphedema in 0% to
lymphedema. In a systematic review, Cormier et al reviewed 6% of patients undergoing ARM plus SLNB and 5.9% to
radiation and reported lymphedema risk after breast or 24% of patients undergoing ARM plus ALND.15 Concerns
chest wall radiation to be 14.5% but was 31.5% after breast/ surrounding ARM include reliable and consistent ARM
chest wall plus supraclavicular radiation and 41.4% when identification rates, crossover lymph nodes/lymphatics
a posterior axillary boost was added.1 Shah et al12 also (breast SLNB is also ARM node), and feasibility axillae with
evaluated the impact of radiation on survivors of breast heavy tumor burden. The Alliance A221702 trial is currently
cancer completing breast-conserving surgery and whole- evaluating SLNB or ALND with and without ARM to formally
breast radiation, citing less lymphedema risk than Cormier evaluate the feasibility and utility of ARM (NCT03927027).
et al1 (8.3% to 14.7%), but noted lymphedema risk varied A few surgeons are performing the lymphatic microsurgical
according to whether supraclavicular, posterior axillary, or preventive healing approach (LYMPHA). LYMPHA seeks
internal mammary nodes were treated. Risk factors for to identify arm lymphatics in the axillary field and then
lymphedema included having more nodes removed, perform lymphatic to venous anastomoses via microsurgical
extracapsular extension, advanced nodal status, grade 3 techniques when a competent venous valve is present.

Boccardo et al16 published the first series noting lymphe- Baseline Assessment
dema in 4% of patients receiving LYMPHA after ALND that Prior to the onset of breast cancer treatment, an assessment
increased to 10.5% if they included those patients with of the individual’s baseline is warranted. The baseline as-
transient postoperative lymphedema. More recently, Feld- sessment establishes a premorbid level of function and
man et al17 found lymphedema in 8% of patients at 24 should obtain clinical measures of the limb and any self-
months, which increased to 12.5% when those with tran- reported limb problems. Ideally, this preoperative visit is
sient lymphedema were included. Collectively, these studies conducted in conjunction with a presurgical physician office
suggest some benefit to immediate lymphovenous anas- visit or nurse case-manager visit to reduce the individual’s
tomosis during ALND; however, further investigation is burden of multiple tests and appointments. This engage-
needed, as the added operative time and need for spe- ment also provides an opportunity for education about what
cialized microsurgical training must be considered if lymphedema is and why prospective monitoring is war-
LYMPHA is to be widely adopted for all patients ranted. Proactive education and awareness about condi-
undergoing ALND. tions such as lymphedema are perceived by patients to be of
RISK AND SURVEILLANCE high importance.24,25
Although SLNB procedures have reduced the extent of The baseline assessment should include valid tests that
axillary clearance, there remains a 6% to 25% risk for have demonstrated efficacy in enabling early identification
developing BCRL.3 This risk is attributed to the extent of of limb changes. Superior tests for clinical screening include
axillary dissection and whether radiation therapy (whole optoelectronic perometry and bioelectrical impedance
breast with or without regional nodal irradiation) is applied. analysis.26 These screening measures are highly sensitive
Notably, the onset of BCRL is commonly slow and pro- and specific to identifying early lymphedema when used in
gressive. Most cases present more than 6 months after the PSM for BCRL21; there is also good concordance between
breast surgical procedure and antineoplastic therapies, with these measures.27,28 The baseline measure is requisite to
reported incidence escalating for at least the first 3 years standardize and optimize early identification of lymphe-
after surgery.18 Moreover, the risk of arm morbidity reducing dema, leading to more precise diagnoses.29
physical function beyond the completion of cancer treat- A holistic view of the individual’s level of physical function is
ments is also relevant and necessitates attention.19 warranted and should be assessed at baseline. Although the
Knowing that this latent time to onset exists, there is an PSM is ideal for early detection and management of lym-
unprecedented opportunity to characterize an individual’s phedema, other common breast cancer treatment–related
risk and monitor their limb prospectively to identify self- problems such as shoulder morbidity and fatigue occur
reported symptoms and clinically measurable changes prevalently and can be identified and managed proactively
associated with early-onset lymphedema.20 When early using PSM.30,31 Suggested patient-reported outcome
lymphedema is identified, it can be treated conservatively, measures for upper-quadrant function include the Dis-
possibly preventing the progression to a more advanced, abilities of the Arm, Shoulder, and Hand questionnaire and
chronic condition.21 To optimize early identification and the Functional Assessment of Cancer Therapies Breast
management, a standardized methodology for surveillance + 4.32
is necessary.22 Interval Screening
THE PROSPECTIVE SURVEILLANCE MODEL Through the trajectory of breast cancer treatment, different
The PSM is recognized as an optimal framework to guide and varied antineoplastic therapies are introduced, each
clinical implementation of a screening methodology for with side effects that may affect the risk profile for the
early identification and management of breast cancer development of lymphedema. Especially concerning is re-
treatment–related impairments.22,23 Using the PSM in duced physical activity and exercise during treatment and
clinical practice enables the early identification and treat- the tissue changes associated with radiotherapy.33,34 Con-
ment of lymphedema, sometimes in a subclinical stage, sidering the varied nature of treatment side effects and the
when intervention can prevent the progression to a more need for individualized patient pathways, a tailored ap-
chronic form of the condition.21 The PSM suggests a clinical proach is warranted and should be driven by the individual’s
pathway that starts with a baseline assessment of the in- risk for lymphedema. The timing of the repeated interval
dividual’s limb at the point of breast cancer diagnosis, prior assessment ranges in the literature from 3-month intervals
to initiating cancer treatments, and proceeds with interval through the first year of active medical treatment of in-
screening at punctuated intervals through the duration of dividuals at higher risk to 6-month intervals for those with
treatment in an effort to identify clinically meaningful less risk and lower potential symptom burden.23,35 The
change from baseline. Figure 1 highlights the PSM clinical interval follow-up schedule should be planned in accor-
pathway. dance with the individual’s treatment pathway and should

FIGURE 1. Breast Cancer–Related Lymphedema Prospective Surveillance Model

Abbreviations: +/ , with or without; CDT, complete decongestive therapy; MLD, manual lymphatic drainage; PRN, as necessary.
minimize the burden of additional or unnecessary ap- reactance in the tissue, derived from an electrical wave that
pointments outside of the cancer care plan. Therefore, is sent though the tissue, to determine if early fluid con-
incorporating these repeated screening measures into gestion is occurring. Prospective studies suggest that bio-
oncology practice is critical. This can be achieved by electrical impedance analysis, when standardized through
introducing a clinically integrated rehabilitation pro- a PSM, enables early identification of BCRL.38,39
fessional into the cancer care clinical pathway,36 in-
Upon early identification of BCRL, it is imperative that the
corporating screening through navigation pathways, 37
clinical pathway include evidence-based clinical in-
or instituting system-specific processes for repeated
terventions to reduce limb volume, prescribe individualized
measures.
self-management strategies to prevent the progression of
The critical element of the repeated interval screening is severity of lymphedema, and reduce the impact on func-
assuring that a sensitive threshold for diagnosis is stan- tion. Application of compression garments, fitted and pre-
dardized and that clinical triage pathways are developed to scribed on an individualized basis,21,40 is recommended
promote intervention by a qualified lymphedema specialist along with self-manual lymphatic drainage techniques,41
when warranted. Subclinical lymphedema was first identi- and exercise42 is a recommended intervention for early
fied by Stout et al21 and described as a 3% or higher in- lymphedema.
crease in the affected limb from the baseline measure with
consideration for the contralateral limb. However, the in- Outcomes and Feasibility of the PSM
dividuals diagnosed through these initial studies demon- The emergence of evidence regarding PSM largely favors
strated clinically meaningful change at 5% volume change. the model as a clinically effective intervention to promote
The standard for clinical diagnosis is recognized as 5% or early detection and management of lymphedema, as well as
higher volume change from baseline. Using circumferential a feasible practice for oncology providers. Additionally, the
measurement methodology with tape measures, optoelec- PSM construct is favorable to early identification and
tronic perometry, or water displacement requires attention management of breast cancer–related upper-extremity
to the contralateral limb to control for weight gain or loss morbidity, and therefore, a holistic approach to measuring
when assessing change over time.26 Bioelectrical imped- the individual’s function over time should be considered
ance analysis, however, uses a ratio of resistance and part of this standard of care.43,44

The clinical outcomes overwhelmingly suggest that early Diagnosis and Staging
identification of lymphedema is best achieved using the For patients with symptoms and signs consistent with
PSM and standardizing measurement methodology, in- a lymphedema diagnosis and with clinically remarkable
cluding timing, assessment measures, and clinical in- volume and/or bioimpedance spectroscopy differences
terventions. Recent guidelines recommend PSM as between the limbs (or compared with preoperative
a standard of care to improve the detection and treatment of measurements),56,57 the presence of dermal backflow on
lymphedema.45,46 The 2019 National Comprehensive contrast-enhanced imaging of the lymphatic system is
Cancer Network Guidelines for Survivorship update rec- diagnostic for lymphedema. The severity and distribution
ognizes the subclinical stage of lymphedema and provides of this backflow correlates closely with the pathologic
guidance aligned with the PSM for early detection and condition of the lymphatic vessels and the stage of the
management of lymphedema.47 The feasibility of the model disease.58 Imaging modalities specific to the lymphatic
has also been examined and found to be favorable to system include indocyanine green fluorescent lymphog-
implementation.44,48 Additionally, preliminary cost analysis raphy, magnetic resonance lymphography, and radioiso-
demonstrates that PSM may be a cost-mitigating strategy tope lymphoscintigraphy.
when early intervention strategies are effective to prevent
the progression to chronic, later-stage lymphedema.49,50 Patient Selection for Surgical Intervention
Implementation as a Standard of Care For patients presenting early following a lymphedema di-
agnosis, a trial of conservative therapy, including daily use of
Greater attention is being directed toward proactively sup- a compression garment, directed by a lymphedema ther-
porting individuals throughout the continuum of cancer apist for 3 to 6 months is typically instituted and may be
care. 51 The concepts of risk stratification, prospective curative in some patients with upper-extremity lymphe-
monitoring for symptom burden, and early detection and dema.59 In patients with persistent lymphedema following
management of morbidity are gaining broad recognition this, surgical intervention is indicated in those who are
across survivorship research and are being heralded as medically fit for it.60
critical components to optimize cancer outcomes.52 The
PSM is a framework that could support larger imple- Patients presenting with advanced-stage lymphedema with
mentation efforts around cancer survivorship care. Although considerable pitting edema, in particular those with re-
the early research on PSM was directed to breast cancer, current bouts of cellulitis, benefit from complete de-
scaling this model to other cancer disease types is aligned congestive therapy until the maintenance phase is achieved
with current initiatives and could provide a clinical pathway before proceeding with surgery.61 Patients with recurrent
to prospectively assess broad domains of function, including episodes of cellulitis may benefit from prophylactic
cognitive, sexual, and psychosocial, in addition to physical, antibiotics.
function.53 The extension of the PSM has been proposed Those with untreated or uncontrolled primary cancer or
and is currently being studied in this regard.54,55 locoregional recurrence are not generally candidates for
SURGICAL INTERVENTIONS FOR THE TREATMENT OF surgical intervention and are better served by nonsurgical
SYMPTOMATIC LYMPHEDEMA management. Patients with a very high body mass index can
develop lymphedema spontaneously and should be man-
Background aged by weight-loss interventions, which can result in
A growing body of evidence supports the effectiveness of substantial improvements in lymphedema, prior to being
surgical procedures in ameliorating the symptoms and considered for surgical intervention.62
disability of patients with lymphedema and reducing the risk Algorithms for Surgical Management
of future episodes of cellulitis with outcomes of surgery
significantly better than with conservative therapy alone. An algorithmic approach to surgical intervention for lym-
These surgical procedures can be broadly categorized as phedema with standardization improves outcomes.60 Evi-
physiologic or debulking. Physiologic surgeries, including dence supports that lymphovenous bypass is indicated in
the lymphovenous bypass or vascularized lymph node earlier-stage lymphedema in which there are still patent
transplant (VLNT) procedures, aim to restore lymphatic fluid lymphatic vessels, VLNT in advanced lymphedema, and
drainage in the affected extremity. Vascularized lymph node debulking procedures for advanced lymphedema with
flaps may be harvested from regional lymphatic basins considerable soft tissue excess. Combination of these pro-
(axillary, inguinal, or cervical) or from within the abdomen. cedures extends indications for physiologic surgery to those
Once established, the chronic lymphedema phenotype is with advanced-stage chronic lymphedema phenotypes.63
characterized by hypertrophy of fibroadipose tissue that can Lymphovenous bypass procedure Using a fluorescent lym-
only be removed directly by suction-assisted lipectomy phography imaging system, intradermal injection of indoc-
(SAL) or excisional procedures. yanine green into the webspaces of the affected extremity

allows the lymphatic vessels to be mapped, and discrete the limb volume, improving function and quality of life, and
obstructed lymphatic vessels distal to areas of dermal decreasing the risk of future infections.70,71 As this procedure
backflow, ranging in caliber from 0.3 to 0.8 mm, are provides only minimal physiologic improvement of the lym-
identified and targeted for supermicrosurgical anastomosis phatic system, patients must continue wearing compression
to adjacent small venules.58 This procedure requires spe- garments lifelong to prevent recurrence. For patients with
cialist surgical techniques, specialist instruments and 11- large-volume advanced fibrotic disease, SAL is ineffective at
0 nylon suture, and a high-powered surgical microscope. extracting this tissue, and excisional techniques are required.
These include staged direct excision (modified Homan
Vascularized lymph node transplant procedures VLNT procedure) and, in extreme cases, excision and skin grafting
procedures are indicated in advanced presentation (Charles procedure).
lymphedema to import new lymphatic function into an
affected extremity. Although the exact mechanisms of CLINICAL OUTCOMES OF SURGICAL PROCEDURES
action of vascularized lymph node flap transplantation are Clinical outcome studies, including randomized controlled
yet to be fully elucidated, spontaneous lymphatic recon- trial evidence, support the efficacy of surgical treatment of
nection mediated by lymphangiogenic growth factor se- lymphedema at reducing the occurrence of cellulitis, de-
cretion from the transplanted lymph nodes (bridging creasing the need for compression garments and other
mechanism),64 as well as with establishment of new conservative therapy, reducing the excess limb volume,
lymphaticovenous drainage within these driven by perfu- improving patient quality of life, and improving physical
sion gradients between the arterial inflow and venous function.63,70,72-74 Research supports that physiologic pro-
outflow (pumping mechanism), have been demonstrated cedures, including lymphovenous bypass and VLNT, are
in both the clinical and experimental settings.65 These effective at decreasing the symptoms of lymphedema, re-
procedures involve microvascular anastomosis of a func- ducing the risk of future infections, and decreasing the
tional lymph node flap into an extremity, either to an anatomic amount of time spent daily for lymphedema care; selected
(orthotopic) or nonanatomical (heterotopic) location, to re- patients may be able to discontinue use of their com-
store lymphatic flow. Orthotopic VLNT to the axilla has the pression garments. Several studies have confirmed the
additional advantage of allowing for radical scar release and efficacy and long-term stability of large-volume SAL
decompression of the subclavian vein, where there is venous debulking with controlled compression therapy for reducing
insufficiency, which may decrease the lymphatic load with limb volume to that similar to the unaffected side for both the
subsequent improvement of the lymphedema. upper and lower extremities and improving patient quality of
Vascularized lymph node flaps may be harvested from life.63,70 Additionally, the incidence of cellulitis is dramati-
within the superficial inguinal (groin), lateral thoracic, cally reduced postoperatively.71 Outcomes for these surgical
supraclavicular, or submental regional lymph node ba- interventions are significantly better than with maximal
sins.60 For patients wishing to avoid any risk of iatrogenic conservative therapy alone.69,72
donor extremity lymphedema or visible donor site scars,
ASSESSING OUTCOMES OF SURGICAL INTERVENTION
intra-abdominal lymph node flap options are increasingly
being performed, in particular the omental (gastroepiploic) Outcome metrics for lymphatic surgery include limb vol-
vascularized lymphatic flap, which may be harvested lap- ume, incidence of cellulitis, physiologic downstaging, and
aroscopically to reduce donor-site morbidity.66 In patients patient-reported outcomes. Change in limb volume is most
undergoing postmastectomy breast reconstruction, VLNT commonly measured by limb circumferential measure-
may be performed by transferring a deep inferior epigastric ments (including derived volumetric calculations) or by
artery perforator flap with a chimeric groin lymph node using a perometer (optoelectronic limb volumeter). Bio-
transplant.67 Vascularized lymph node transfer from the impedance spectroscopy can also be used to comparatively
inguinal or axillary regions must be performed under reverse measure the extracellular fluid, and a growing body of
lymphatic mapping guidance to reduce the risk of iatrogenic evidence supports its use in response to intervention.
donor extremity lymphedema.68 Physiologic downstaging can be evaluated either using
radioisotope lymphoscintigraphy or by indocyanine green
Debulking procedures Once established, the chronic lym- lymphography. The Lymphedema Quality of Life scale,
phedema phenotype is characterized by hypertrophy of Upper Limb Lymphedema 27 questionnaire, and Lym-
fibroadipose soft tissues, which can only be removed by phedema Life Impact Scale are validated tools specific for
direct excision. Traditional excisional surgeries that resulted patient-reported outcomes in patients with lymphedema.
in unacceptable scarring and morbidity have been replaced,
except in the most severe cases, by minimally invasive SAL CONCLUSION
with postoperative controlled compression therapy.69 SAL Secondary lymphedema in the developed world is most
with controlled compression therapy is effective at reducing commonly attributed to cancer treatment. Serial clinical

trials in many cancer subtypes are actively seeking to de- microsurgery for prevention and treatment of lymphe-
escalate nodal treatment strategies, recognizing that dema are actively being studied. These may provide
comprehensive nodal extirpation frequently does not add durable long-term therapeutic options for at-risk and
benefit to long-term survival. Widespread adoption of pro- affected patients with lymphedema. It is critical that all
spective lymphedema screening models and consistent clinicians treating patients with cancer understand
patient education are needed to raise awareness and early available resources for lymphedema prevention and
intervention in at-risk patients. Exciting advancements in treatment.

1
Department of Surgery, Mayo Clinic, Jacksonville, FL Sarah A. McLaughlin, MD, FACS, Department of Surgery, Mayo Clinic,
2
Department of Hematology Oncology, West Virginia University Cancer 4500 San Pablo Rd., Jacksonville, FL 32224; email: mclaughlin.sarah@
Institute, Morgantown, WV mayo.edu.
3
Department of Plastic Surgery, The University of Texas MD Anderson
Cancer Center, Houston, TX
EDBK_280471.
REFERENCES
1. Cormier JN, Askew RL, Mungovan KS, et al. Lymphedema beyond breast cancer: a systematic review and meta-analysis of cancer-related secondary lym-
phedema. Cancer. 2010;116:5138-5149.
2. DeSantis CE, Lin CC, Mariotto AB, et al. Cancer treatment and survivorship statistics, 2014. CA Cancer J Clin. 2014;64:252-271.
3. Shaitelman SF, Cromwell KD, Rasmussen JC, et al. Recent progress in the treatment and prevention of cancer-related lymphedema. CA Cancer J Clin. 2015;
65:55-81.
4. Ashikaga T, Krag DN, Land SR, et al; National Surgical Adjuvant Breast, Bowel Project. Morbidity results from the NSABP B-32 trial comparing sentinel lymph
node dissection versus axillary dissection. J Surg Oncol. 2010;102:111-118.
5. Purushotham AD, Upponi S, Klevesath MB, et al. Morbidity after sentinel lymph node biopsy in primary breast cancer: results from a randomized controlled trial.
J Clin Oncol. 2005;23:4312-4321.
6. Veronesi U, Paganelli G, Viale G, et al. A randomized comparison of sentinel-node biopsy with routine axillary dissection in breast cancer. N Engl J Med. 2003;
349:546-553.
7. Teshome M, Ballman KV, McCall LM, et al. Long-term incidence of lymphedema after sentinel lymph node dissection for early stage breast cancer: ACOSOG
Z0010. Presented at: Society of Surgical Oncology 67th Annual Cancer Symposium. Phoenix, AZ; 2014. Abstract 3.
8. Giuliano AE, Ballman KV, McCall L, et al. Effect of axillary dissection vs no axillary dissection on 10-year overall survival among women with invasive breast cancer
and sentinel node metastasis: the ACOSOG Z0011 (Alliance) randomized clinical trial. JAMA. 2017;318:918-926.
9. Caudle AS, Yang WT, Krishnamurthy S, et al. Improved axillary evaluation following neoadjuvant therapy for patients with node-positive breast cancer using
selective evaluation of clipped nodes: implementation of targeted axillary dissection. J Clin Oncol. 2016;34:1072-1078.
10. Donker M, van Tienhoven G, Straver ME, et al. Radiotherapy or surgery of the axilla after a positive sentinel node in breast cancer (EORTC 10981-22023
AMAROS): a randomised, multicentre, open-label, phase 3 non-inferiority trial. Lancet Oncol. 2014;15:1303-1310.
11. Naoum GE, Roberts SA, Shui AM, et al. Quantifying the impact of regional lymph node irradiation on lymphedema risk in breast cancer patients treated with SLNB
or ALND: long-term results from a prospective screening trial. Int J Radiat Oncol. 2019;105:S42.
12. Shah C, Wilkinson JB, Baschnagel A, et al. Factors associated with the development of breast cancer-related lymphedema after whole-breast irradiation. Int
J Radiat Oncol Biol Phys. 2012;83:1095-1100.
13. Hyngstrom JR, Chiang YJ, Cromwell KD, et al. Prospective assessment of lymphedema incidence and lymphedema-associated symptoms following lymph node
surgery for melanoma. Melanoma Res. 2013;23:290-297.
14. Faries MB, Thompson JF, Cochran A, et al; MSLT Cooperative Group. The impact on morbidity and length of stay of early versus delayed complete lym-
phadenectomy in melanoma: results of the Multicenter Selective Lymphadenectomy Trial (I). Ann Surg Oncol. 2010;17:3324-3329.
15. Ahmed M, Rubio IT, Kovacs T, et al. Systematic review of axillary reverse mapping in breast cancer. Br J Surg. 2016;103:170-178.
16. Boccardo FM, Casabona F, Friedman D, et al. Surgical prevention of arm lymphedema after breast cancer treatment. Ann Surg Oncol. 2011;18:2500-2505.
17. Feldman S, Bansil H, Ascherman J, et al. Single institution experience with lymphatic microsurgical preventive healing approach (LYMPHA) for the primary
prevention of lymphedema. Ann Surg Oncol. 2015;22:3296-3301.

18. Armer JM, Ballman KV, McCall L, et al. Lymphedema symptoms and limb measurement changes in breast cancer survivors treated with neoadjuvant che-
motherapy and axillary dissection: results of American College of Surgeons Oncology Group (ACOSOG) Z1071 (Alliance) substudy. Support Care Cancer. 2019;
27:495-503.
19. Shamley D, Robb K. An early warning surveillance programme for detecting upper limb deterioration after treatment for breast cancer: a novel technology
supported system. BMC Cancer. 2015;15:635.
20. Bulley C, Gaal S, Coutts F, et al. Comparison of breast cancer-related lymphedema (upper limb swelling) prevalence estimated using objective and subjective
criteria and relationship with quality of life. Biomed Res Int. 2013;2013:1-8.
21. Stout Gergich NL, Pfalzer LA, McGarvey C, et al. Preoperative assessment enables the early diagnosis and successful treatment of lymphedema. Cancer. 2008;
112:2809-2819.
22. Stout NL, Binkley JM, Schmitz KH, et al. A prospective surveillance model for rehabilitation for women with breast cancer. Cancer. 2012;118(Suppl):2191-2200.
23. Shah C, Arthur DW, Wazer D, et al. The impact of early detection and intervention of breast cancer-related lymphedema: a systematic review. Cancer Med. 2016;
5:1154-1162.
24. Levangie PK, Santasier AM, Stout NL, et al. A qualitative assessment of upper quarter dysfunction reported by physical therapists treated for breast cancer or
treating breast cancer sequelae. Support Care Cancer. 2011;19:1367-1378.
25. Binkley JM, Harris SR, Levangie PK, et al. Patient perspectives on breast cancer treatment side effects and the prospective surveillance model for physical
rehabilitation for women with breast cancer. Cancer. 2012;118(Suppl):2207-2216.
26. Levenhagen K, Davies C, Perdomo M, et al. Diagnosis of upper quadrant lymphedema secondary to cancer: clinical practice guideline from the oncology section
of the American Physical Therapy Association. Phys Ther. 2017;97:729-745.
27. Bundred NJ, Stockton C, Keeley V, et al; Investigators of BEA/PLACE studies. Comparison of multi-frequency bioimpedance with perometry for the early detection
and intervention of lymphoedema after axillary node clearance for breast cancer. Breast Cancer Res Treat. 2015;151:121-129.
28. Jain MS, Danoff JV, Paul SM. Correlation between bioelectrical spectroscopy and perometry in assessment of upper extremity swelling. Lymphology. 2010;
43:85-94.
29. Sun F, Skolny MN, Swaroop MN, et al. The need for preoperative baseline arm measurement to accurately quantify breast cancer-related lymphedema. Breast
Cancer Res Treat. 2016;157:229-240.
30. Springer BA, Levy E, McGarvey C, et al. Pre-operative assessment enables early diagnosis and recovery of shoulder function in patients with breast cancer. Breast
31. Gerber LH, Stout N, McGarvey C, et al. Factors predicting clinically significant fatigue in women following treatment for primary breast cancer. Support Care
Cancer. 2011;19:1581-1591.
32. Davies C, Ryans K, Levenhagen K, et al. Breast Cancer EDGE Task Force outcomes: quality of life and functional outcome measures for secondary lymphedema
in breast cancer survivors. Rehabil Oncol. 2014;32:7-12.
33. Peterson LL, Ligibel JA. Physical activity and breast cancer: an opportunity to improve outcomes. Curr Oncol Rep. 2018;20:50.
34. Warren LE, Miller CL, Horick N, et al. The impact of radiation therapy on the risk of lymphedema after treatment for breast cancer: a prospective cohort study. Int
35. Stout NL, Silver JK, Raj VS, et al. Toward a national initiative in cancer rehabilitation: recommendations from a subject matter expert group. Arch Phys Med
Rehabil. 2016;97:2006-2015.
36. Barnes CA, Stout NL, Varghese TK Jr., et al. Clinically integrated physical therapist practice in cancer care: a new comprehensive approach. Phys Ther. 2020;
100:543-553.
37. Stout NL, Sleight A, Pfeiffer D, et al. Promoting assessment and management of function through navigation: opportunities to bridge oncology and rehabilitation
systems of care. Support Care Cancer. 2019;27:4497-4505.
38. Whitworth PW, Cooper A. Reducing chronic breast cancer-related lymphedema utilizing a program of prospective surveillance with bioimpedance spectroscopy.
Breast J. 2018;24:62-65.
39. Kilgore LJ, Korentager SS, Hangge AN, et al. Reducing breast cancer-related lymphedema (BCRL) through prospective surveillance monitoring using bio-
impedance spectroscopy (BIS) and patient directed self-interventions. Ann Surg Oncol. 2018;25:2948-2952.
40. Javid SH, Anderson BO. Mounting evidence against complex decongestive therapy as a first-line treatment for early lymphedema. J Clin Oncol. 2013;
31:3737-3738.
41. Bahtiyarca ZT, Can A, Ekşioğlu E, et al. The addition of self-lymphatic drainage to compression therapy instead of manual lymphatic drainage in the first phase of
complex decongestive therapy for treatment of breast cancer-related lymphedema: a randomized-controlled, prospective study. Turk J Phys Med Rehabil. 2018;
65:309-317.
42. Schmitz KH, Ahmed RL, Troxel AB, et al. Weight lifting for women at risk for breast cancer-related lymphedema: a randomized trial. JAMA. 2010;304:2699-2705.
43. Rafn BS, Hung S, Hoens AM, et al. Prospective surveillance and targeted physiotherapy for arm morbidity after breast cancer surgery: a pilot randomized
controlled trial. Clin Rehabil. 2018;32:811-826.
44. Singh C, De Vera M, Campbell KL. The effect of prospective monitoring and early physiotherapy intervention on arm morbidity following surgery for breast cancer:
a pilot study. Physiother Can. 2013;65:183-191.

45. Armer JM, Hulett JM, Bernas M, et al. Best-practice guidelines in assessment, risk reduction, management, and surveillance for post-breast cancer lym-
phedema. Curr Breast Cancer Rep. 2013;5:134-144.
46. Shah C, Vicini FA, Arthur D. Bioimpedance spectroscopy for breast cancer related lymphedema assessment: clinical practice guidelines. Breast J. 2016;
22:645-650.
47. Sanft T, Denlinger CS, Armenian S, et al. NCCN guidelines insights: survivorship, version 2.2019. J Natl Compr Canc Netw. 2019;17:784-794.
48. Blaney JM, McCollum G, Lorimer J, et al. Prospective surveillance of breast cancer-related lymphoedema in the first-year post-surgery: feasibility and comparison
of screening measures. Support Care Cancer. 2015;23:1549-1559.
49. Chance-Hetzler J, Armer J, Van Loo M, et al. Prospective lymphedema surveillance in a clinic setting. J Pers Med. 2015;5:311-325.
50. Stout NL, Pfalzer LA, Springer B, et al. Breast cancer-related lymphedema: comparing direct costs of a prospective surveillance model and a traditional model of
care. Phys Ther. 2012;92:152-163.
51. Alfano CM, Mayer DK, Bhatia S, et al. Implementing personalized pathways for cancer follow-up care in the United States: proceedings from an American Cancer
Society-American Society of Clinical Oncology summit. CA Cancer J Clin. 2019;69:234-247.
52. Jacobs LA, Shulman LN. Follow-up care of cancer survivors: challenges and solutions. Lancet Oncol. 2017;18:e19-e29.
53. Alfano CM, Zucker DS, Pergolotti M, et al. A precision medicine approach to improve cancer rehabilitation’s impact and integration with cancer care and optimize
patient wellness. Curr Phys Med Rehabil Rep. 2017;5:64-73.
54. Alfano CM, Pergolotti M. Next-generation cancer rehabilitation: a giant step forward for patient care. Rehabil Nurs. 2018;43:186-194.
55. Cheville AL, Mustian K, Winters-Stone K, et al. Cancer rehabilitation: an overview of current need, delivery models, and levels of care. Phys Med Rehabil Clin N
Am. 2017;28:1-17.
56. Armer JM, Stewart BR. A comparison of four diagnostic criteria for lymphedema in a post-breast cancer population. Lymphat Res Biol. 2005;3:208-217.
57. Ridner SH, Dietrich MS, Cowher MS, et al. A randomized trial evaluating bioimpedance spectroscopy versus tape measurement for the prevention of lym-
phedema following treatment for breast cancer: interim analysis. Ann Surg Oncol. 2019;26:3250-3259.
58. Chang DW, Suami H, Skoracki R. A prospective analysis of 100 consecutive lymphovenous bypass cases for treatment of extremity lymphedema. Plast Reconstr
Surg. 2013;132:1305-1314.
59. Akita S, Nakamura R, Yamamoto N, et al. Early detection of lymphatic disorder and treatment for lymphedema following breast cancer. Plast Reconstr Surg.
2016;138:192e-202e.
60. Schaverien MV, Coroneos CJ. Surgical treatment of lymphedema. Plast Reconstr Surg. 2019;144:738-758.
61. Yamamoto R, Yamamoto T. Effectiveness of the treatment-phase of two-phase complex decongestive physiotherapy for the treatment of extremity lymphedema.
Int J Clin Oncol. 2007;12:463-468.
62. Nitti MD, Hespe GE, Kataru RP, et al. Obesity-induced lymphatic dysfunction is reversible with weight loss. J Physiol. 2016;594:7073-7087.
63. Carl HM, Walia G, Bello R, et al. Systematic review of the surgical treatment of extremity lymphedema. J Reconstr Microsurg. 2017;33:412-425.
64. Suami H, Scaglioni MF, Dixon KA, et al. Interaction between vascularized lymph node transfer and recipient lymphatics after lymph node dissection-a pilot study
in a canine model. J Surg Res. 2016;204:418-427.
65. Aschen SZ, Farias-Eisner G, Cuzzone DA, et al. Lymph node transplantation results in spontaneous lymphatic reconnection and restoration of lymphatic flow.
Plast Reconstr Surg. 2014;133:301-310.
66. Nguyen AT, Suami H, Hanasono MM, et al. Long-term outcomes of the minimally invasive free vascularized omental lymphatic flap for the treatment of
lymphedema. J Surg Oncol. 2017;115:84-89.
67. Saaristo AM, Niemi TS, Viitanen TP, et al. Microvascular breast reconstruction and lymph node transfer for postmastectomy lymphedema patients. Ann Surg.
2012;255:468-473.
68. Dayan JH, Dayan E, Smith ML. Reverse lymphatic mapping: a new technique for maximizing safety in vascularized lymph node transfer. Plast Reconstr Surg.
2015;135:277-285.
69. Brorson H, Svensson H. Liposuction combined with controlled compression therapy reduces arm lymphedema more effectively than controlled compression
therapy alone. Plast Reconstr Surg. 1998;102:1058-1067, discussion 1068.
70. Hoffner M, Ohlin K, Svensson B, et al. Liposuction gives complete reduction of arm lymphedema following breast cancer treatment-a 5-year prospective study in
105 patients without recurrence. Plast Reconstr Surg Glob Open. 2018;6:e1912.
71. Lee D, Piller N, Hoffner M, et al. Liposuction of postmastectomy arm lymphedema decreases the incidence of erysipelas. Lymphology. 2016;49:85-92.
72. Dionyssiou D, Demiri E, Tsimponis A, et al. A randomized control study of treating secondary stage II breast cancer-related lymphoedema with free lymph node
transfer. Breast Cancer Res Treat. 2016;156:73-79.
73. Ozturk CN, Ozturk C, Glasgow M, et al. Free vascularized lymph node transfer for treatment of lymphedema: A systematic evidence based review. J Plast Reconstr
Aesthet Surg. 2016;69:1234-1247.
74. Sharkey AR, King SW, Ramsden AJ, et al. Do surgical interventions for limb lymphoedema reduce cellulitis attack frequency? Microsurgery. 2017;37:348-353.

CANCER PREVENTION, RISK
REDUCTION, AND GENETICS

Cancer Prevention in Low-Resource Countries: An

Overview of the Opportunity
Sailaja Kamaraju, MD, MS1; Jeffrey Drope, PhD2; Rengaswamy Sankaranarayanan, MD3; and Surendra Shastri, MD4
overview
Rising trends in the incidence of cancer in low- and middle-income countries (LMICs) add to the existing
challenges with communicable and noncommunicable diseases. While breast and colorectal cancer in-
cidence rates are increasing in LMICs, the incidence of cervical cancer shows a mixed trend, with rising
incidence rates in China and sub-Saharan Africa and declining trends in the Indian subcontinent and South
America. The increasing frequencies of unhealthy lifestyles, notably less physical activity, obesity, tobacco
use, and alcohol consumption are causing a threat to health care in LMICs. Also, poorly developed health
systems tend to have inadequate resources to implement early detection and adequate basic treatment.
Inequalities in social determinants of health, lack of awareness of cancer and preventive care, lack of efficient
referral pathways and patient navigation, and nonexistent or inadequate health care funding can lead to
advanced disease presentation at diagnosis. This article provides an overview of opportunities to address
cancer control in LMICs, with a focus on tobacco control, vaccination for cervical cancer, novel tools to assist
with early detection, and screening for breast and other cancers.
INTRODUCTION emphasis on potential opportunities for cancer pre-

Recent reports on the global burden of cancer estimate vention in LMICs.
18.1 million new cases and 9.6 million cancer deaths
ADDRESSING THE BARRIERS TO SCREENING FOR THE
worldwide in 2018 (Fig. 1).1 As the burden of com-
MOST COMMON CANCERS
municable disease and cardiovascular disease con-
tinues, the incidence of cancer is on the rise in LMICs.2 Tobacco Use in LMICs
Regardless of the geographic area, social determinants Tobacco use continues to be the largest preventable
of health (SDH) seem to play a crucial role in cancer risk factor for the main noncommunicable diseases,
control and its outcomes across the globe.3 Even in including cancer.7,8 Cigarette smoking increases the
higher-income nations, factors related to SDHs play risk of at least 12 cancers, including acute myeloid
a vital role in obtaining cancer screening, including leukemia and cancers of the bladder, colon and rec-
sociodemographic factors, health literacy, cultural in- tum, esophagus, kidney, larynx, liver, lung, oral cavity
fluences, lack of beliefs in preventive health, socio- and pharynx, pancreas, stomach, and uterine cervix.7-9
economic status, and lack of access to health care.4 More than 24% of all cancer deaths globally are at-
Also, patients’ attitudes, beliefs (including fear of can- tributable to tobacco use.7,9 Currently, more than 1
cer), and an inadequate number of facilities equipped billion people in the world use tobacco products, and if
with cancer screening tools all prevent timely nothing is done to help them quit and to prevent young
screening.5 Unfortunately, limited allocation of re- people from starting, more than half of these people will
sources to health care and a lack of implementation succumb to premature deaths from tobacco-related
Author affiliations
and support of government policies ultimately affect prevention diseases. In 2019 alone, the estimated death toll from
information (if and the disease burden. Although the incidence of tobacco-related disease was more than 8 million.8 The
applicable) appear cancer is increasing in LMICs, the cost of cancer is economic burden is also staggering, approximately $2
at the end of this trillion (in 2016 purchasing power parity) from both
article.
limiting its affordability. 6 Therefore, a focus on
cancer prevention, early detection, and use of novel higher health care costs and lost productivity.10 More-
Accepted on March
and cost-effective methods of screening are in- over, the burden is increasing in LMICs, where the
2, 2020 and
published at valuable in lowering cancer incidence and the re- majority of tobacco users now live.
ascopubs.org on lated financial burden on a country’s economy. However, there is encouraging news for public health:
April 2, 2020:
Herein, we describe the global burden of cancers Tobacco control has a well-developed package of
10.1200/EDBK_ related to tobacco use and other common cancers of evidence-based interventions that are proven to work.
280625 the cervix, breast, and gastrointestinal tract, with an These interventions are enshrined internationally in the

Cancer Prevention in Low-Resource Settings
two, Algeria and Mauritania, using the resources to address

cancer specifically.11
• Cancer incidence in LMICs is rapidly in- A second effective provision is graphic warning labels on
creasing, demanding the dedication of more tobacco packaging. Research consistently shows that ed-
resources and the implementation of cost- ucating users, particularly young people, through packaging
effective prevention measures. promotes quitting and reduces initiation.13 It is inexpensive
• Cancer awareness and education through because governments place the financial burden on the
community health workers and navigators can tobacco companies while retaining full control over the
improve cancer prevention, screening, and messages and images on the package. In recent years,
early detection in low-resource settings. many countries have moved a step further to plain, stan-
• Government support for comprehensive dardized packaging of tobacco products, which removes
tobacco control programs—including higher all colors and logos. Because these characteristics have
taxes, graphic warnings, marketing bans, been proven to appeal especially to children and youth,
and accessible and free tobacco cessation their removal is having demonstrable positive public health
programs—will reduce multiple tobacco- effects.14
related cancers.
A third proven intervention is the banning of tobacco in-
• Prophylactic HPV vaccines have a significant dustry marketing.15 It is well documented that one key part
impact in reducing the risk of cervical cancer if
of the tobacco industry’s success in attracting and main-
deployed successfully to young girls (ages 9 to
taining customers is their relentless marketing. Removing
14).
their ability to communicate to the public undermines their
• Novel screening methods such as mobile efforts to find new customers and maintain existing ones.
mammographic units, detection of immuno-
The challenge is that the industry finds many creative ways
chemical fecal occult blood, and Papanicolaou
to communicate. For example, much tobacco marketing
test will assist in early detection of breast, co-
lorectal, and cervical cancers, respectively. is done now through point-of-sale displays. Another major
strategy is price discounting. Governments must be vigilant
in addressing the tobacco industry’s marketing innovations.
Framework Convention on Tobacco Control (FCTC), the
A fourth provision is smoke-free laws that prohibit smoking
world’s first public health treaty under the auspices of the
in public and workplaces. Exposure to second-hand smoke
World Health Organization (WHO).11 Coming into force in
causes substantial morbidity and mortality annually, in-
2005, the treaty now has 181 parties, and it undergirds the
cluding an estimated 1.2 million deaths.8 Approximately
global approach to tobacco control. In most LMICs, the main
one-third of females and one-fifth of males are exposed to
provisions of the FCTC are barely implemented, which
second-hand smoke. Smoke-free laws not only protect
leaves plenty of room for improvement.
nonsmokers, especially children, but they de-normalize
The most effective provision is the increase in excise taxes smoking, and it is well established that smokers are more
on tobacco products. The mechanism is simple: When likely to quit and youth are less likely to start when smoking
governments raise taxes, companies raise prices to maintain is rarely observed in public.16
profits, and higher prices lead to lower consumption.
An important fifth intervention is the use of mass media
Typically, a 10% increase in tobacco prices leads to a 4% to
campaigns to inform the public. Whereas incorporating
8% decrease in consumption.12 This decreased consump-
learning about the dangers of tobacco use in school cur-
tion comes in large part from young people not initiating
ricula is important and useful, targeted mass media cam-
tobacco use. In fact, tax research unequivocally demon-
paigns are proving to be a better investment value. Young
strates two important dynamics: Young people and those in
people are particularly responsive to social media sources,
lower socioeconomic groups are more affected by changes in
from which they typically get most of their information.17
prices. Thus, lower-income people typically reap a dispro-
portionate amount of the reward from these policies. Consumption of Smokeless Tobacco
The other major benefit of tobacco excise tax increases is Although smoking remains the most common and deadly
the increased revenue for government. Because tobacco form of tobacco use, smokeless tobacco use is a major
products are typically inelastic (the decline in consumption cancer risk factor for a number of cancers, most notably
is less than the increase in price), governments typically get those of the head and neck.18 Globally, more than 350
the double benefit of higher revenue and reduced con- million people use a wide variety of smokeless tobacco
sumption. Nearly three dozen countries report investing products.19 Though most smokeless users live in South and
these new tobacco tax revenues in health care, including Southeast Asia (. 80%), the prevalence of smokeless

Kamaraju et al
tobacco use is above 10% in nearly 30 countries around the interventions to eliminate cervical cancer as a public
globe, and use is particularly concentrated in lower socio- health problem in the 21st century.26
economic groups and increasingly among youth.20 Recent
research suggests that the accessibility and affordability of Global burden of cervical cancer The estimated age-
these products is exacerbating this growing public health standardized incidence rate of cervical cancer globally is
challenge. Most troubling is the lack of policy progress to about 13.1 per 100,000 women (range, 2 to 75 per 100,000
tackle it, wherein interventions such as taxation and youth women), based on cancer data from 185 countries.27 There
access lag behind other tobacco control efforts.21 were approximately 570,000 new cases of cervical cancer
and 311,000 cervical cancer deaths in 2018.27 Cervical
There are three main barriers to successful tobacco control. cancer was the fourth most common cancer in women,
First, the tobacco industry spends billions of dollars to ranking after breast cancer (2.1 million cases), colorectal
market its deadly products. The messages of “cool” and cancer (0.8 million), and lung cancer (0.7 million). It was the
“youthful” are ubiquitous and require perseverance by to- leading cause of cancer-related death in women in eastern,
bacco control proponents to counter. At the same time, the western, middle, and southern Africa. The highest incidence
tobacco industry actively lobbies policy makers against was reported from Eswatini (former Swaziland), with ap-
public health provisions and even aggressively litigates proximately 6.5% of women developing cervical cancer
regulations and policies that seek to reduce consumption. before age 75. More than one-third of the global burden of
Last, weak governance, particularly in many LMICs, con- cervical cancer was contributed by China (106,000 cases;
tinues to be a major obstacle. Too often, governments are 48,000 deaths) and India (97,000 cases; 60,000 deaths).
slow to react or do not make any effort to protect citizens Globally, the average age at diagnosis of cervical cancer was
from the obvious dangers posed by tobacco use. The FCTC age 53, ranging from age 44 (Vanuatu) to age 68 (Singapore).
has strong provisions in its Article 5.3 that help governments Cervical cancer ranked among the top three cancers af-
to insulate themselves from the pressure and influence of fecting women younger than age 45 in 146 (79%) of 185
the tobacco industry, most notably by actively keeping the countries assessed.
industry away from policy making.22 Certain countries in West Asia and Maghreb with low in-
cidence of cervical cancer (fewer than 5 in 100,000) have
Cervical Cancer Prevention and Control a low frequency of sexually transmitted infections and low
Cervical cancer is an important public health problem among prevalence of HPV infection, probably because of socio-
middle-aged women in LMICs, where screening programs do economic factors that reduce risky sexual behaviors.28 A
not exist or are ineffective, and a high risk of HIV infection recent report hypothesized that overtransmission of a pro-
exists. It is well established that persistent infection with one tective genetic variant, TP53 codon 72 proline allele, may
of the 13 high-risk types of HPV may lead to the development also be responsible for the low risk in the West Asian
population.29 On the other hand, the high incidence of
of high-grade cervical precancerous lesions such as cervical
cervical cancer in sub-Saharan Africa, Latin America, and
intraepithelial neoplasia grade 3 (CIN 3) and grade 2 (CIN 2)
South Asia seems to be caused by increased background
or adenocarcinoma in situ, which if left undetected or un-
risk attributable to high rates of HPV persistence and HIV
treated may lead to the development of invasive cervical
transmission.28,30 Because the burden of cervical cancer in
cancer.23 Among the high-risk types, HPV 16 and HPV 18
sub-Saharan Africa is substantial and is increasing, there is
cause 70% to 75% of cervical cancers and 50% to 60% of
a great need for improved prevention to reduce morbidity
cervical precancerous lesions.24,25 This knowledge has led to and mortality from this disease.31
the development of two major approaches to cervical cancer
prevention: HPV vaccination to prevent HPV infection and Information on cancer trends over time enables dynamic
testing for high-risk HPV to facilitate the detection of cervical planning for changing patterns of cancer at the national
cancer precursor lesions. level. In Asia, a disturbing pattern is the increasing trend in
cervical cancer incidence rates and the narrowing urban–
There is conclusive evidence that cervical precancerous rural differences in incidence after several years of falling
lesions can be prevented by HPV vaccination, and cervical rates in China32,33 and in sub-Saharan Africa.34,35 The an-
cancer can be prevented with early detection and treatment nual percentage increase in age-standardized incidence
of precancerous lesions. Given the effective and affordable rates is about 9.2% in China.33 In a pooled analysis of
interventions for preventing cervical cancer, and the fact several prevalence studies, the overall prevalence of high-
that implementation of both prevention and early detection risk HPV infection was reported to range from 15% to 20%
can lead to avoidance of cervical cancer, WHO has called in China.36,37 On the other hand, cervical cancer incidence
on member states to commit and implement a global rates have been declining with a concomitant increase in
initiative to scale up preventive, screening, and treatment breast cancer, particularly in urban populations in India,

FIGURE 1. Estimated Number of New Cancer Cases by World Area, 2018
possibly because of socioeconomic development, reduced cervical cancer. It recommends two doses 6 or 12 months
parity, and increasing age at marriage.38 apart for 9- to 14-year-old girls and three doses over a 6-
month period for those aged 15 and older and for immu-
Cervical cancer in immunocompromised women Precancerous
nocompromised patients irrespective of age.40 There is
cervical lesions and cervical cancer are more aggressive
and progressive in immunocompromised patients. Given recent evidence that even a single dose of HPV vaccine
the high prevalence of precancerous cervical lesions among might protect against cervical neoplasia.41 HPV vaccination
HIV-positive women, prevention measures are extremely has been followed by sustained protection against CIN2 or
important in HIV-positive women.39 CIN3 and adenocarcinoma in situ associated with HPV 16/
18 infection in young women who were not initially infected
Currently bivalent (targeting HPV 16 and 18), quadrivalent
with high-risk HPV or HPV 16/18, and the occurrence of
(targeting HPV 6, 11, 16, and 18), and nonavalent (targeting
severe adverse events or adverse pregnancy outcomes was
HPV 6, 11, 16, 18, 31, 33, 45, 52, and 58) prophylactic
not significantly higher in recipients of HPV vaccines.32
vaccines effective in HPV-naive patients have been licensed
in many countries, and new HPV vaccines are undergoing Primary prevention of cervical cancer through HPV vacci-
evaluation in India and China. The Indian vaccine is ex- nation is a cost-effective preventive measure. HPV vacci-
pected to be licensed in 2021. The WHO recommends that nation is being implemented in a number of sub-Saharan
HPV vaccination be included in national immunization African countries with high vaccine coverage, improved
programs as part of a comprehensive strategy to prevent community sensitization, and strong commitment from

Kamaraju et al
national governments (e.g., Rwanda, Botswana, Zambia), diagnosis, and effective treatment of early invasive cancer
despite various challenges.42 HPV vaccination has been and political commitment and allocation of adequate re-
associated with negative media coverage and negative sources are critical for elimination in LMICs.
advocacy in a number of countries.43,44 Denmark provides
an excellent example of how a country grappled with Global Burden of Breast Cancer Breast cancer continues to
negative media coverage of HPV vaccination and the be the most common cancer in women both in developed
steadying impact of action by national authorities.43,44 These and developing countries. According to the GLOBOCAN
important national experiences provide valuable leads for 2018 estimates, WHO reports 626,679 women lost their
successful vaccination implementation for countries be- lives to breast cancer in 2018.48 Although the incidence of
ginning to plan for vaccine introduction. breast cancer is higher in high-income countries, with 74.4
per 100,000 in Western Europe, compared with 34.4 per
Screening for cervical cancer Cervical cancer can be 100,000 in Asia, the overall incidence is increasing in
prevented by screening for cervical precancerous lesions developing countries.49 In 2018, among the 2.1 million
and treating them by screening with cytology or visual in- cases of female breast cancers diagnosed worldwide, ap-
spection with 3% to 5% acetic acid (VIA) or HPV testing. proximately 911,000 cases (rate of 34 per 100,000) were
Cytology screening traditionally has been the cornerstone of reported from Asia. Correspondingly, the overall outcomes
cervical screening in high-income countries, where it has of breast cancer vary with survival rates of 80% or more in
been followed by substantial decline in incidence and North America, compared with 40% or less in low-income
mortality. countries, demanding continued efforts across the contin-
The difficulty in implementing cytology or HPV screening in uum of cancer care in LMICs.49
LMICs has led to the implementation of VIA screening and
Epidemiology, prevention, and early detection of breast
treatment of VIA-positive women with cryotherapy or thermal
cancer The rising incidence of breast cancer among the
ablation.5 The real-time result of screening with VIA allows
younger women in LMICs is concerning. Whereas the median
a single-visit approach combining screening and treatment,
age of breast cancer for women in developed countries is 63,
thereby maximizing compliance to treatment. Given the
it is more common among younger women between age 45
feasibility of introducing VIA screening in public health ser-
and 50 and in India, Mexico, and Arab countries.50 Given that
vices, several sub-Saharan African countries have imple-
the majority of LMICs have no screening guidelines in place,
mented it. However, a major challenge with VIA screening
high-risk screening and chemoprevention options seem
has been the difficulties in quality assurance and in main-
unfeasible. In this context, other preventive and cost-effective
taining a consistent screening performance in the long run.
measures are optimal: mass cancer awareness and edu-
The fact that HPV testing is more sensitive, accurate, and cation camps, community health workers and navigator-based
reproducible in detecting high-grade lesions and more ef- care delivery models, clinical breast examination (CBE), mobile
fective in preventing future invasive cancers than screening technology, novel screening methods, addressing other risk
with liquid-based or conventional cytology, particularly in factors such as obesity, and lifestyle modifications.51-53 Al-
the post-HPV vaccination scenario, has led to increasing though breast cancer prevention trials incorporating high-risk
implementation of HPV screening in national programs.45 surveillance with imaging and risk reduction strategies with
Recent European guidelines strongly recommend primary prophylactic mastectomy have shown success in BRCA gene
HPV testing over cytology screening. The possibility of doing mutation carriers, the uptake of these services is not cost-
HPV testing by using self-collected cervical cell specimens effective in LMICs.54-56
facilitates reaching women who otherwise would not par-
Despite the lack of collective agreement on the benefits of
ticipate in screening. The high negative predictive value and
CBE, it continues to be an important tool and a standard of
the possibility of longer screening intervals for HPV-negative
practice along with screening mammography.57,58 Although
women are other advantages of HPV screening. It has been
the success of screening mammography with a 19% re-
shown that even a single round of VIA or HPV screening has
duction in mortality for women age 40 to 70 appears
been followed by significant reduction in cervical cancer
promising, overdiagnosis continues to be a challenge.59
mortality in low-income countries.46,47
Mammography is a potential option for screening, but the
Elimination of cervical cancer The goal set by WHO to majority of LMICs lack screening guidelines and equipped
reduce the incidence rate of cervical cancer to less than 4 in facilities for cancer screening. Eventually, many patients
100,000 women globally by vaccinating 90% of girls by age succumb to advanced stages of disease at diagnosis and
15, screening 70% of women aged 35 to 45, and treating at high mortality rates. Also, incomplete treatments and poor
least 90% of precancerous lesions detected by screening adherence continue to add to the burden at a societal level.
requires substantial investment by national governments. A To explore the use of triennial CBE in reducing the rate of
judicious combination of HPV vaccination, screening and advanced breast cancer and mortality, Sankaranarayanan

et al53 conducted a cluster randomized clinical trial in substantial morbidity, mortality, and financial burden could
Trivandrum, India. In this trial, 115,652 women age 30 to 69 be avoided by early detection.71-73 Head and neck cancers
in 275 clusters were randomly assigned to receive CBE or no impose catastrophic economic burden on families of pa-
screening. CBE was performed by trained community health tients in LMICs.74 One study of the economic effects of head
workers, who functioned as liaisons between the healthy and neck cancers in South Asia estimated that US$16.9
women and physicians. This study demonstrated higher billion was lost on account of these cancers every year in the
numbers of early-stage breast cancers in the CBE group region.75 Another study conducted by Patterson et al re-
compared with the control group (18.8 and 8.1 per 100,000 ported an anticipated global cumulative loss of US$535
women, respectively).53 billion during the years 2018 to 2030 due to head and neck
cancers.76 Surgery remains the treatment of choice for these
Mobile mammography: a potential tool to improve access cancers in LMICs since chemotherapy is very expensive and
and screening in LMICs Previous studies reported mobile radiation facilities are limited.77,78 Even surgical care is fairly
mammographic units as valuable screening tools in im- limited and usually not specialized to handle head and neck
proving access for those with transportation barriers.51,60-68 cancers.79
However, the utility of and adherence to screening mam-
mography on fixed versus mobile units depended on soci- Liver Cancer
odemographic factors and women’s perceived risk of breast LMICs together contributed 83% of the global annual in-
cancer.61,64 Whereas the users of fixed mammography units cidence of liver cancers and 82% of the global annual
were noted to be from stable socioeconomic backgrounds, mortality from liver cancers in 2018, with China contributing
tended to be insured, and were adherent to screening,60,61,64 more than 50% of these.80 Poor survival rates make liver
mobile units’ users tended to be from underserved com- cancer the seventh most common cause of cancer mor-
munities with transient lifestyles and living situations and tality worldwide (after lung, breast, prostate, colon, non-
were less adherent.65 Vang et al67,68 reported critical gaps and melanoma skin cancers, and stomach cancers).
the challenges of the mobile units, such as the quality, in- Chronic infection with hepatitis B and hepatitis C viruses,
complete imaging, higher recalls, limited use of newer alcohol, tobacco, aflatoxins, diabetes, and obesity are all
technology (e.g., film-screen, full-field digital, tomosynthesis), associated with the disease. Cirrhosis is usually a precursor to
and the lack of diagnostic imaging. Despite the expanding liver cancer.81-83 Hepatitis B vaccination, universal safety
use of mobile mammographic units in several parts of the precautions for preventing transmission of blood-borne in-
world, such as sub-Saharan Africa,62 the association with fections, blood product safety, supply of sterile needles and
sociodemographic factors and the recall rates continue to syringes to injectable drug users, and safe sex practices have
raise challenges for screening adherence.66 Rodrı́guez- been shown to reduce the transmission of hepatitis B and
Cuevas et al66 conducted a mobile mammography pro- C.84,85 Observational studies claim that coffee, statins, met-
gram in Mexico in 2005 and 2006 for women older than age formin, and aspirin have a protective effect against liver
40. Among the 96,828 women submitted to mammography, cancer.86-89 Coffee even finds a mention in the 2018 clinical
only 1% of mammograms in the Breast Imaging Reporting practice guidelines of the European Association for the Study
and Data System were 0, 4, or 5. Among women with ab- of the Liver.90 Several LMICs have incorporated hepatitis B
normal mammograms, 27.7% had breast cancer (ductal vaccines into their national infant immunization programs,
carcinoma in situ, 2.1%; stage I, 29.4%; stage II, 42.2%), and some of these countries have reported declining rates of
and older age (age 50 and older) was found to be the only risk liver cancer.91 Sonography is the standard surveillance test
factor for having an abnormal mammogram. In LMICs with recommended by the American Association for the Study of
existing infrastructure, multidisciplinary evaluation of mobile Liver Diseases, the European Association for the Study of the
mammographic units as a potential cost-effective opportunity Liver, and the Asia Pacific Association for the Study of the
will serve as a new avenue for screening. Liver.91-93 Serum alpha-fetoprotein is sometimes used as an
Head and Neck Cancers adjunct to sonography.94
Head and neck cancers (including cancers of the lip and Esophagus, Stomach, and Colorectal Cancers
oral cavity, larynx, oropharynx, hypopharynx, sinuses, and LMICs contributed 75% of the global annual incidence and
salivary glands) contribute 4.9% and 4.7%, respectively, to mortality from esophageal, stomach, and colorectal cancers
the global cancer incidence and mortality.69 An estimated combined in 2018.80 Contrast radiography and upper en-
76% of head and neck cancer cases and 84% of head and doscopy have been used for screening stomach cancers.
neck cancer deaths occur in LMICs,69 and high rates of Observational studies suggest that the screening has con-
tobacco consumption, including smokeless tobacco, are tributed to early detection and mortality reduction, but
a crucial concern for increasing rates of these cancers.70 there are no data from large randomized trials.95-103 The value
Although head and neck cancers are largely preventable, of screening asymptomatic people for gastric cancer is

Kamaraju et al
controversial even in areas with a high incidence of gastric As studies of single-dose HPV vaccination continue to ac-
cancer.104 There are only limited data suggesting that such cumulate data,41,119-121 two-dose nonavalent HPV vacci-
an approach may decrease the incidence of gastric cancer nation as the standard of care (at least 6 months apart and
in high-incidence areas.105-108 Therefore, routine radio- before the 15th birthday) for girls (and boys if economically
graphic, endoscopic, and Helicobacter pylori screenings of feasible) should be next on the list of LMIC health priori-
asymptomatic healthy people are not recommended. ties.122 Along with HPV vaccination, population-based
cervical screening with VIA/HPV testing at 5-year intervals
Tobacco, alcohol, malnutrition, and HPV are identified as risk
between the ages of 30 and 65 (two or three lifetime
factors for squamous cell carcinoma of the esophagus.109
screens)123,124 would help LMICs meet the WHO goal of
Long standing gastroesophageal reflux disease, with Barrett
global cervical cancer elimination. Despite the Global Alli-
esophagus as a usual precursor, is associated with in-
ance for Vaccines and Immunization subsidies of HPV
creased risk of esophageal adenocarcinoma.110-112 Benefits
vaccine, LMICs are still slow in the rollout of HPV vaccines,
of screening for Barrett esophagus have been debated,
because of the costs and logistics associated with HPV
without a firm consensus. The logistics of implementing such
vaccine delivery. International health agencies must work
a screening strategy and the subsequent management of
together to make HPV vaccination affordable in LMICs.
patients diagnosed with Barrett esophagus present enormous
challenges in LMICs. Overall, the support for radiographic and Because the age cohorts for breast cancer in women are much
endoscopic esophageal screening has been modest.113 younger in LMICs, mammography as a population-based
screening tool is likely to result in more harm than benefit.
Fecal occult blood test (FOBT) screening for colorectal
Based on the current evidence, breast self-examination, CBE,
cancers have shown mortality reduction.114 Sigmoidoscopy,
and sonography also cannot be recommended for population-
colonoscopy, and double-contrast barium enema are used
based breast cancer screening in LMICs.125 Breast cancer
to triage FOBT screening positives. Thailand implemented
control programs in LMICs should therefore be currently
a successful pilot project demonstrating the feasibility of
limited to awareness programs with increased facilities to di-
conducting an organized FOBT colorectal cancer screening
agnose and treat symptomatic referrals.
program in LMICs, setting an example for other LMICs.115
Screening and early detection of liver cancer continue to be
DISCUSSION challenging in LMICs. Although liver sonography has been
The incidence and mortality of cancer are increasing in used with some success in Asian countries, it is not feasible
LMICs; however, more than 70% of the commonly occurring in African countries. Therefore, all LMICs should adopt
cancers in LMICs have evidence-based prevention, screening, a universal hepatitis B vaccination policy for infants. Ad-
and early detection interventions. LMICs should therefore ditionally, blood safety, universal safety precautions, and
invest in low-cost cancer awareness, prevention, screening, safe sex practices should be promoted, given the additional
and early detection programs that are locally feasible. Large health benefits. Esophagus, stomach, and colorectal can-
investments in human resources would be needed for such cers are also increasing in LMICs, but these could be at-
programs. tributed to a combination of variables including, diet,
physical activity, increasing lifespan, and improved di-
Effective tobacco control measures should top the health
agnosis of the disease and disease registration processes.
agenda in LMICs. Several LMICs already have WHO
Radiographic and endoscopic screening for these cancers,
FCTC–based legislation in place, and they should ensure
recommended in some high-income countries, is not fea-
tougher enforcement.116,117 Smokeless tobacco use, which
sible in LMICs because of inadequate numbers of service
is very common in South Asia, is under-researched, and
providers and necessary infrastructure. Therefore, focusing
investments should be made in developing evidence-based
on tobacco and alcohol control, and FOBT screening where
interventions for smokeless tobacco control. Although in-
feasible, might be the best strategies for LMICs.
creasing taxes and pack warnings118 on tobacco products
are effective evidence-based strategies, additional chal- CONCLUSION
lenges exist because the majority of tobacco products sold Cancer incidence in LMICs is on the rise, requiring re-
in LMICs are nonpackaged (e.g., single cigarettes) and sources and implementation of cost-effective prevention
hence escape regulatory controls. Banning tobacco ad- measures. In addition, the increasing prevalence of obesity
vertising and marketing (including surrogate advertising and and lack of physical activity among people living in urban
marketing), enacting and enforcing tough smoke-free laws, areas in LMICs will eventually result in sharp increases in
and promoting effective mass media and social media noncommunicable diseases, including cancers. Investing in
campaigns targeting teenagers and youth would yield op- programs to aggressively address this problem is necessary
timal returns on investments. now to prevent this future catastrophe.


1
Medical College of Wisconsin, Milwaukee, WI Sailaja Kamaraju, MD, MS, Medical College of Wisconsin, 9200 W.
2
American Cancer Society, Atlanta, GA Wisconsin Ave., Milwaukee, WI 53226; email: [email protected].
3
World Health Organization International Agency for Research on Cancer,
Lyon, France
4
University of Texas, MD Anderson Cancer Center, Houston, TX
EDBK_280625.
REFERENCES
1. Bray F, Ferlay J, Soerjomataram I, et al. Global cancer statistics: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA
Cancer J Clin. 2018;68:394-424.
2. Remais JV, Zeng G, Li G, et al. Convergence of non-communicable and infectious diseases in low- and middle-income countries. Int J Epidemiol. 2013;
42:221-227.
3. Viswanath K, Emmons KM. Message effects and social determinants of health: its application to cancer disparities. J Commun. 2006;56 (suppl 1):S238-S264.
4. Marmot M. Health equity, cancer, and social determinants of health. Lancet Glob Health. 2018;6:s29.
5. Sankaranarayanan R. Screening for cancer in low- and middle-income countries. Ann Glob Health. 2014;80:412-417.
6. Fitzmaurice C, Abate D, Abbasi N, et al; Global Burden of Disease Cancer Collaboration. Global, regional, and national cancer incidence, mortality, years of life
lost, years lived with disability, and disability-adjusted life-years for 29 cancer groups, 1990 to 2017: a systematic analysis for the Global Burden of Disease
Study. JAMA Oncol. Epub 2019 Sep 27.
7. U.S. Department of Health and Human Services. The Health Consequences of Smoking: 50 Years of Progress. A Report of the Surgeon General. Atlanta, GA:
Department of Health and Human Services. Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion;
2014.
8. Stanaway JD, Afshin A, Gakidou E; GBD 2017 Risk Factor Collaborators. Global, regional, and national comparative risk assessment of 84 behavioural,
environmental and occupational, and metabolic risks or clusters of risks for 195 countries and territories, 1990–2017: a systematic analysis for the Global
Burden of Disease Study 2017. Lancet. 2018;392:1923-1994.
9. Wild CP, Weiderpass E, Stewart BW, (eds). World Cancer Report. Cancer Research for Cancer Prevention. Lyon, France: World Health Organization International
Agency for Research on Cancer; 2020.
10. Drope JSN, Cahn Z, Drope J, et al. The Tobacco Atlas. Atlanta: American Cancer Society and Vital Strategies. The Tobacco Atlas. Atlanta, GA: American Cancer
Society and Vital Strategies; 2017.
11. World Health Organization. WHO Report on the Global Tobacco Epidemic 2017. Geneva, Switzerland: World Health Organization: 2017.
12. Chaloupka FJ. The economics of tobacco taxation. Presented at: 2009 Tobacco Summit; April 16, 2009; Fairbanks, AK.
13. O’Hegarty M, Pederson LL, Nelson DE, et al. Reactions of young adult smokers to warning labels on cigarette packages. Am J Prev Med. 2006;30:467-473.
14. Wakefield M, Coomber K, Zacher M, et al. Australian adult smokers’ responses to plain packaging with larger graphic health warnings 1 year after imple-
mentation: results from a national cross-sectional tracking survey. Tob Control. 2015;24 (suppl 2):ii17-ii25.
15. Henriksen L. Comprehensive tobacco marketing restrictions: promotion, packaging, price and place. Tob Control. 2012;21:147-153.
16. Kelly BC, Vuolo M, Frizzell LC, et al. Denormalization, smoke-free air policy, and tobacco use among young adults. Soc Sci Med. 2018;211:70-77.
17. Freeman B, Potente S, Rock V, et al. Social media campaigns that make a difference: what can public health learn from the corporate sector and other social
change marketers? Public Health Res Pract. 2015;25:e2521517.
18. Niaz K, Maqbool F, Khan F, et al. Smokeless tobacco (paan and gutkha) consumption, prevalence, and contribution to oral cancer. Epidemiol Health. 2017;
39:e2017009.
19. Sinha DN, Gupta PC, Kumar A, et al. The Poorest of Poor Suffer the Greatest Burden From Smokeless Tobacco Use: A Study From 140 Countries. Nicotine Tob.
Res. 2018;20:1529-1532.
20. Sinha DN, Kumar A, Bhartiya D, et al. Smokeless Tobacco Use Among Adolescents in Global Perspective. Nicotine Tob. Res. 2017;19:1395-1396.
21. Mehrotra R, Yadav A, Sinha DN, et al. Smokeless tobacco control in 180 countries across the globe: call to action for full implementation of WHO FCTC
measures. Lancet Oncol. 2019;20:e208-e217.
22. Prevent20. Home page. https://wecanprevent20.org/. Accessd March 24, 2020.
23. World Health Organization International Agency for Cancer Research. Human papillomaviruses. IARC Monogr Eval Carcinog Risks Hum. 2005;90.
24. Bzhalava D, Guan P, Franceschi S, et al. A systematic review of the prevalence of mucosal and cutaneous human papillomavirus types. Virology. 2013;
445:224-231.

Kamaraju et al
25. de Sanjose S, Quint WG, Alemany L, et al; Retrospective International Survey and HPV Time Trends Study Group. Human papillomavirus genotype attribution in
invasive cervical cancer: a retrospective cross-sectional worldwide study. Lancet Oncol. 2010;11:1048-1056.
26. World Health Organization: WHO director-general calls for all countries to take action to help end the suffering caused by cervical cancer. 2018. www.who.int/
reproductivehealth/call-to-action-elimination-cervical-cancer/en/. Accessd March 24, 2020.
27. Arbyn M, Weiderpass E, Bruni L, et al. Estimates of incidence and mortality of cervical cancer in 2018: a worldwide analysis. Lancet Glob Health. 2020;
8:e191-e203.
28. World Health Organization. Global Health Observatory (GHO) data. 2019. www.who.int/gho/hiv/en/. Accessd March 24, 2020.
29. Alsbeih GA, Al-Harbi NM, Bin Judia SS, et al. Reduced rate of human papillomavirus infection and genetic overtransmission of TP53 72C polymorphic variant
lower cervical cancer incidence. Cancer. 2017;123:2459-2466.
30. Jemal A, Center MM, DeSantis C, et al. Global patterns of cancer incidence and mortality rates and trends. Cancer Epidemiol Biomarkers Prev. 2010;
19:1893-1907.
31. Vaccarella S, Laversanne M, Ferlay J, et al. Cervical cancer in Africa, Latin America and the Caribbean and Asia: regional inequalities and changing trends. Int
J Cancer. 2017;141:1997-2001.
32. Arbyn M, Xu L. Efficacy and safety of prophylactic HPV vaccines. A Cochrane review of randomized trials. Expert Rev Vaccines. 2018;17:1085-1091.
33. Li X, Zheng R, Li X, et al. Trends of incidence rate and age at diagnosis for cervical cancer in China, from 2000 to 2014. Chin J Cancer Res. 2017;29:477-486.
34. Chokunonga E, Borok MZ, Chirenje ZM, et al. Trends in the incidence of cancer in the black population of Harare, Zimbabwe 1991-2010. Int J Cancer. 2013;
133:721-729.
35. Mboumba Bouassa RS, Prazuck T, Lethu T, et al. Cervical cancer in sub-Saharan Africa: a preventable noncommunicable disease. Expert Rev Anti Infect Ther.
2017;15:613-627.
36. Li K, Li Q, Song L, et al. The distribution and prevalence of human papillomavirus in women in mainland China. Cancer. 2019;125:1030-1037.
37. Zhu B, Liu Y, Zuo T, et al. The prevalence, trends, and geographical distribution of human papillomavirus infection in China: the pooled analysis of 1.7 million
women. Cancer Med. 2019;8:5373-5385.
38. Badwe RA, Dikshit R, Laversanne M, et al. Cancer incidence trends in India. Jpn J Clin Oncol. 2014;44:401-407.
39. Weldegebreal F, Worku T. Precancerous cervical lesion among HIV-positive women in sub-Saharan Africa: a systematic review and meta-analysis. Cancer
Contr. 2019;26:1073274819845872.
40. World Health Organization. Human papillomavirus vaccine, WHO position paper, May 2017: recommendations. Vaccine. 2017;35:5753-5755.
41. Sankaranarayanan R, Joshi S, Muwonge R, et al; Indian HPV vaccine study group. Can a single dose of human papillomavirus (HPV) vaccine prevent cervical
cancer? Early findings from an Indian study. Vaccine. 2018;36:4783-4791.
42. Black E, Richmond R. Prevention of cervical cancer in sub-Saharan Africa: the advantages and challenges of HPV vaccination. Vaccines (Basel). 2018;6:E61.
43. Hansen PR, Schmidtblaicher M, Brewer NT. Resilience of HPV vaccine uptake in Denmark: decline and recovery. Vaccine. 2020;38:1842-1848.
44. Sankaranarayanan R, Basu P, Kaur P, et al. Current status of human papillomavirus vaccination in India’s cervical cancer prevention efforts. Lancet Oncol.
2019;20:e637-e644.
45. Maver PJ, Poljak M. Primary HPV-based cervical cancer screening in Europe: implementation status, challenges, and future plans. Clin Microbiol Infect. Epub
2019 Sep 17.
46. Sankaranarayanan R, Esmy PO, Rajkumar R, et al. Effect of visual screening on cervical cancer incidence and mortality in Tamil Nadu, India: a cluster-
randomised trial. Lancet. 2007;370:398-406.
47. Sankaranarayanan R, Nene BM, Shastri SS, et al. HPV screening for cervical cancer in rural India. N Engl J Med. 2009;360:1385-1394.
48. International Agency for Research on Cancer. World Health Organization. GLOBOCAN. http://gco.iarc.fr/. Accessed March 24, 2020.
49. Coleman MP, Quaresma M, Berrino F, et al; CONCORD Working Group. Cancer survival in five continents: a worldwide population-based study (CONCORD).
Lancet Oncol. 2008;9:730-756.
50. Tfayli A, Temraz S, Abou Mrad R, et al. Breast cancer in low- and middle-income countries: an emerging and challenging epidemic. J Oncol. 2010;
2010:490631.
51. Kamaraju S, DeNomie M, Visotcky A, et al. Increasing mammography uptake through academic–community partnerships targeting immigrant and refugee
communities in Milwaukee. WMJ. 2018;117:55-61.
52. Kamaraju S, Olson J, DeNomie M, et al. Community breast health education for immigrants and refugees: lessons learned in outreach efforts to reduce cancer
disparities. J Cancer Educ. 2019;34:1092-1096.
53. Sankaranarayanan R, Ramadas K, Thara S, et al. Clinical breast examination: preliminary results from a cluster randomized controlled trial in India. J Natl
Cancer Inst. 2011;103:1476-1480.
54. Metcalfe KA, Birenbaum-Carmeli D, Lubinski J, et al; Hereditary Breast Cancer Clinical Study Group. International variation in rates of uptake of preventive
options in BRCA1 and BRCA2 mutation carriers. Int J Cancer. 2008;122:2017-2022.
55. Paluch-Shimon S, Cardoso F, Sessa C, et al; ESMO Guidelines Committee. Prevention and screening in BRCA mutation carriers and other breast/ovarian
hereditary cancer syndromes: ESMO Clinical Practice Guidelines for cancer prevention and screening. Ann Oncol. 2016;27 (suppl 5):v103-v110.

56. Rosner BA, Colditz GA, Hankinson SE, et al. Validation of Rosner-Colditz breast cancer incidence model using an independent data set, the California Teachers
Study. Breast Cancer Res Treat. 2013;142:187-202.
57. Smith RA, Cokkinides V, Brawley OW. Cancer screening in the United States, 2012: a review of current American Cancer Society guidelines and current issues in
cancer screening. CA Cancer J Clin. 2012;62:129-142.
58. US Preventive Services Task Force. Screening for breast cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med. 2009;
151:716-726.
59. Pace LE, Keating NL. A systematic assessment of benefits and risks to guide breast cancer screening decisions. JAMA. 2014;311:1327-1335.
60. Katapodi MC, Lee KA, Facione NC, et al. Predictors of perceived breast cancer risk and the relation between perceived risk and breast cancer screening: a meta-
analytic review. Prev Med. 2004;38:388-402.
61. Elewonibi BR, Thierry AD, Miranda PY. Examining mammography use by breast cancer risk, race, nativity, and socioeconomic status. J Immigr Minor Health.
2018;20:59-65.
62. Apffelstaedt JP, Hattingh R, Baatjes K, et al. Results of a pilot programme of mammographic breast cancer screening in the Western Cape. S Afr Med J. 2014;
104:297-298.
63. Brooks SE, Hembree TM, Shelton BJ, et al. Mobile mammography in underserved populations: analysis of outcomes of 3,923 women. J Community Health.
2013;38:900-906.
64. Chen YR, Chang-Halpenny C, Kumarasamy NA, et al. Perspectives of mobile versus fixed mammography in Santa Clara County, California: a focus group study.
Cureus. 2016;8:e494.
65. Moyce SC, Schenker M. Migrant workers and their occupational health and safety. Annu Rev Public Health. 2018;39:351-365.
66. Rodrı́guez-Cuevas S, Guisa-Hohenstein F, Labastida-Almendaro S. First breast cancer mammography screening program in Mexico: initial results 2005–2006.
Breast J. 2009;15:623-631.
67. Scheel JR, Tillack AA, Mercer L, et al. Mobile versus fixed facility: Latinas’ attitudes and preferences for obtaining a mammogram. J Am Coll Radiol. 2018;
15:19-28.
68. Vang S, Margolies LR, Jandorf L. Mobile mammography participation among medically underserved women: a systematic review. Prev Chronic Dis. 2018;
15:E140.
69. International Agency for Research on Cancer. Cancer Today. Cancer Fact Sheets. https://gco.iarc.fr/today/fact-sheets-cancers. Accessed on March 18, 2020.
70. Asthana S, Labani S, Kailash U, et al. Association of Smokeless Tobacco Use and Oral Cancer: A Systematic Global Review and Meta-Analysis. Nicotine Tob.
Res. 2019;21:1162-1171.
71. Sankaranarayanan R, Ramadas K, Thara S, et al. Long term effect of visual screening on oral cancer incidence and mortality in a randomized trial in Kerala,
India. Oral Oncol. 2013;49:314-21.
72. Sankaranarayanan R, Ramadas K, Thomas G, et al. Effect of screening on oral cancer mortality in Kerala, India: a cluster-randomised controlled trial. The
Lancet. 2005;365:1927-1933.
73. Subramanian S, Sankaranarayanan R, Bapat B, et al. Cost-effectiveness of oral cancer screening: results from a cluster randomized controlled trial in India. Bull.
World Health Organ. 2009;87:200-206.
74. Jan S, Kimman M, Peters SAE, et al; ACTION Study Group. Financial catastrophe, treatment discontinuation and death associated with surgically operable
cancer in South-East Asia: Results from the ACTION Study. Surgery. 2015;157:971-982.
75. Alkire BC, Bergmark RW, Chambers K, et al. Head and neck cancer in South Asia: Macroeconomic consequences and the role of the head and neck surgeon.
Head Neck. 2016;38:1242-1247.
76. Patterson RH, Fischman VG, Wasserman I, et al. Global Burden of Head and Neck Cancer: Economic Consequences, Health, and the Role of Surgery.
Otolaryngol Head Neck Surg. 2020;162:296-303.
77. Wilson BE, Jacob S, Yap ML, et al. Estimates of global chemotherapy demands and corresponding physician workforce requirements for 2018 and 2040:
a population-based study. Lancet Oncol. 2019;20:769-780.
78. Atun R, Jaffray DA, Barton MB, et al. Expanding global access to radiotherapy. Lancet Oncol. 2015;16:1153-1186.
79. Meara JG, Leather AJM, Hagander L, et al. Global Surgery 2030: evidence and solutions for achieving health, welfare, and economic development. Lancet.
2015;386:569-624.
80. World Health Organization International Agency for Research on Cancer. Cancer Today. https://gco.iarc.fr/today/home. Accessed March 24, 2020.
81. Nordenstedt H, White DL, El-Serag HB. The changing pattern of epidemiology in hepatocellular carcinoma. Dig Liver Dis. 2010;42 (suppl 3):S206-S214.
82. Parikh S, Hyman D. Hepatocellular cancer: a guide for the internist. Am J Med. 2007;120:194-202.
83. Venook AP, Papandreou C, Furuse J, et al. The incidence and epidemiology of hepatocellular carcinoma: a global and regional perspective. Oncologist. 2010;15
(suppl 4):5-13.
84. Franco E, Bagnato B, Marino MG, et al. Hepatitis B: epidemiology and prevention in developing countries. World J Hepatol. 2012;4:74-80.
85. Franco E, Meleleo C, Serino L, et al. Hepatitis A: epidemiology and prevention in developing countries. World J Hepatol. 2012;4:68-73.
86. Bravi F, Tavani A, Bosetti C, et al. Coffee and the risk of hepatocellular carcinoma and chronic liver disease: a systematic review and meta-analysis of prospective
studies. Eur J Cancer Prev. 2017;26:368-377.

Kamaraju et al
87. Sahasrabuddhe VV, Gunja MZ, Graubard BI, et al. Nonsteroidal anti-inflammatory drug use, chronic liver disease, and hepatocellular carcinoma. J Natl Cancer
Inst. 2012;104:1808-1814.
88. Singh S, Singh PP, Singh AG, et al. Statins are associated with a reduced risk of hepatocellular cancer: a systematic review and meta-analysis. Gastroenterology.
2013;144:323-332.
89. Zhou YY, Zhu GQ, Liu T, et al. Systematic review with network meta-analysis: antidiabetic medication and risk of hepatocellular carcinoma. Sci Rep. 2016;
6:33743.
90. European Association for the Study of the Liver. EASL clinical practice guidelines: management of hepatocellular carcinoma. J Hepatol. 2018;69:182-236.
91. Childs L, Roesel S, Tohme RA. Status and progress of hepatitis B control through vaccination in the South-East Asia region, 1992–2015. Vaccine. 2018;36:6-14.
92. Marrero JA, Kulik LM, Sirlin CB, et al. Diagnosis, staging, and management of hepatocellular carcinoma: 2018 practice guidance by the American Association
for the Study of Liver Diseases. Hepatology. 2018;68:723-750.
93. Tan CH, Low SC, Thng CH. APASL and AASLD consensus guidelines on imaging diagnosis of hepatocellular carcinoma: a review. Int J Hepatol. 2011;
2011:519783.
94. Chang TS, Wu YC, Tung SY, et al. Alpha-fetoprotein measurement benefits hepatocellular carcinoma surveillance in patients with cirrhosis. Am J Gastroenterol.
2015;110:836-844.
95. Choi KS, Jun JK, Suh M, et al. Effect of endoscopy screening on stage at gastric cancer diagnosis: results of the National Cancer Screening Programme in Korea.
Br J Cancer. 2015;112:608-612.
96. Fukao A, Tsubono Y, Tsuji I, et al. The evaluation of screening for gastric cancer in Miyagi Prefecture, Japan: a population-based case-control study. Int
J Cancer. 1995;60:45-48.
97. Hisamichi S, Sugawara N, Fukao A. Effectiveness of gastric mass screening in Japan. Cancer Detect Prev. 1988;11:323-329.
98. Inaba S, Hirayama H, Nagata C, et al. Evaluation of a screening program on reduction of gastric cancer mortality in Japan: preliminary results from a cohort
study. Prev Med. 1999;29:102-106.
99. Kunisaki C, Ishino J, Nakajima S, et al. Outcomes of mass screening for gastric carcinoma. Ann Surg Oncol. 2006;13:221-228.
100. Mizoue T, Yoshimura T, Tokui N, et al; Japan Collaborative Cohort Study Group. Prospective study of screening for stomach cancer in Japan. Int J Cancer. 2003;
106:103-107.
101. Murakami R, Tsukuma H, Ubukata T, et al. Estimation of validity of mass screening program for gastric cancer in Osaka, Japan. Cancer. 1990;65:1255-1260.
102. Oshima A, Hirata N, Ubukata T, et al. Evaluation of a mass screening program for stomach cancer with a case-control study design. Int J Cancer. 1986;
38:829-833.
103. Tsuji I, Fukao A, Sugawara N, et al. Cost-effectiveness analysis of screening for gastric cancer in Japan. Tohoku J Exp Med. 1991;164:279-284.
104. Leung WK, Wu MS, Kakugawa Y, et al; Asia Pacific Working Group on Gastric Cancer. Screening for gastric cancer in Asia: current evidence and practice. Lancet
Oncol. 2008;9:279-287.
105. Ford AC, Forman D, Hunt RH, et al. Helicobacter pylori eradication therapy to prevent gastric cancer in healthy asymptomatic infected individuals: systematic
review and meta-analysis of randomised controlled trials. BMJ. 2014;348:g3174.
106. Saito K, Arai K, Mori M, et al. Effect of Helicobacter pylori eradication on malignant transformation of gastric adenoma. Gastrointest Endosc. 2000;52:27-32.
107. Uemura N, Mukai T, Okamoto S, et al. Effect of Helicobacter pylori eradication on subsequent development of cancer after endoscopic resection of early gastric
cancer. Cancer Epidemiol Biomarkers Prev. 1997;6:639-642.
108. Wong BC, Lam SK, Wong WM, et al; China Gastric Cancer Study Group. Helicobacter pylori eradication to prevent gastric cancer in a high-risk region of China:
a randomized controlled trial. JAMA. 2004;291:187-194.
109. Glade MJ. Food, nutrition, and the prevention of cancer: a global perspective. American Institute for Cancer Research/World Cancer Research Fund, American
Institute for Cancer Research, 1997. Nutrition. 1999;15:523-526.
110. Wijnhoven BP, Tilanus HW, Dinjens WN. Molecular biology of Barrett’s adenocarcinoma. Ann Surg. 2001;233:322-337.
111. Lagergren J, Bergström R, Lindgren A, et al. Symptomatic gastroesophageal reflux as a risk factor for esophageal adenocarcinoma. N Engl J Med. 1999;
340:825-831.
112. Skacel M, Petras RE, Gramlich TL, et al. The diagnosis of low-grade dysplasia in Barrett’s esophagus and its implications for disease progression. Am
J Gastroenterol. 2000;95:3383-3387.
113. Sami SS, Ragunath K, Iyer PG. Screening for Barrett’s esophagus and esophageal adenocarcinoma: rationale, recent progress, challenges, and future di-
rections. Clin Gastroenterol Hepatol. 2015;13:623-634.
114. Hewitson P, Glasziou P, Watson E, et al. Cochrane systematic review of colorectal cancer screening using the fecal occult blood test (hemoccult): an update. Am
J Gastroenterol. 2008;103:1541-1549.
115. Khuhaprema T, Sangrajrang S, Lalitwongsa S, et al. Organised colorectal cancer screening in Lampang Province, Thailand: preliminary results from a pilot
implementation programme. BMJ Open. 2014;4:e003671.
116. Myers ML. The FCTC’s evidence-based policies remain a key to ending the tobacco epidemic. Tob Control. 2013;22 (suppl 1):i45-i46.
117. Shastri A, Shastri SS. Cancer screening and prevention in low-resource settings. Nat Rev Cancer. 2014;14:822-829.

118. Francis DB, Mason N, Ross JC, et al. Impact of tobacco-pack pictorial warnings on youth and young adults: a systematic review of experimental studies. Tob
Induc Dis. 2019;17:41.
119. Brotherton JM, Budd A, Rompotis C, et al. Is one dose of human papillomavirus vaccine as effective as three? A national cohort analysis. Papillomavirus Res.
2019;8:100177.
120. Sankaranarayanan R, Prabhu PR, Pawlita M, et al; Indian HPV Vaccine Study Group. Immunogenicity and HPV infection after one, two, and three doses of
quadrivalent HPV vaccine in girls in India: a multicentre prospective cohort study. Lancet Oncol. 2016;17:67-77.
121. Whitworth HS, Gallagher KE, Howard N, et al. Efficacy and immunogenicity of a single dose of human papillomavirus vaccine compared to no vaccination or
standard three and two-dose vaccination regimens: a systematic review of evidence from clinical trials. Vaccine. 2020;38:1302-1314.
122. Bergman H, Buckley BS, Villanueva G, et al. Comparison of different human papillomavirus (HPV) vaccine types and dose schedules for prevention of HPV-
related disease in females and males. Cochrane Database Syst Rev. 2019;11:CD013479.
123. Jeronimo J, Castle PE, Temin S, et al. Secondary prevention of cervical cancer: ASCO resource-stratified clinical practice guideline. J Glob Oncol. 2016;
3:635-657.
124. Lee H, Hur S, Jang H, et al. Cost-utility of a two-dose human papillomavirus vaccination programme added to cervical cancer screening compared with cervical
cancer screening alone in Korea. Asian Pac J Cancer Prev. 2019;20:425-435.
125. Lauby-Secretan B, Scoccianti C, Loomis D, et al; International Agency for Research on Cancer Handbook Working Group. Breast-cancer screening--viewpoint of
the IARC Working Group. N Engl J Med. 2015;372:2353-2358.

Integrating Genetic and Genomic Testing Into

Oncology Practice
Susan M. Domchek1; Elaine Mardis2; Jennifer W. Carlisle3; and Taofeek K. Owonikoko3
overview
Genetic information, both germline and somatic, is an increasingly important consideration in therapeutic
decision-making in cancer. Germline mutations in genes associated with increased cancer risk can identify
those individuals without cancer who may benefit from enhanced screening and prevention strategies. In
individuals with cancer, germline and somatic mutations may help to guide local and systemic management
decisions. Here, we review considerations of these issues in selected cancer types.
INTRODUCTION The FDA approved olaparib in December 2014 for the

Genetic testing, whether by germline or somatic (tu- treatment of recurrent germline BRCA1/2 mutation–
mor) sequencing, frequently occurs in the setting of associated ovarian cancer in the setting of three or
cancer. Identification of germline mutations in cancer- more lines of therapy.7 Since that time, multiple PARP
susceptibility genes potentially influences treatment of inhibitors have been approved for ovarian cancer,
the current cancer with regard to local and systemic including in the recurrent and first-line maintenance
management decisions, may provide information about settings, with the most dramatic results seen in tumors as-
the development of future cancers (and, thus, allow for sociated with germline or somatic BRCA1/2 mutations.6,8
secondary prevention), and has important implications Germline genetic testing for BRCA1 and BRCA2 mutations
for family members in terms of risk, enhanced screening, is recommended by ASCO9,10 and the National Com-
and prevention of disease.1,2 In advanced cancer, the prehensive Cancer Network for all patients with ovarian
use of somatic testing to identify potential therapeutic cancer, followed by somatic testing for BRCA1/2 mu-
targets has increased rapidly (Table 1), with further tations if germline testing is negative.9
growth expected.3,4 Mutations identified on tumor se- Olaparib and talazoparib are FDA approved for the
quencing or by blood-based (“liquid”) biopsies may be treatment of HER2/neu germline BRCA1 and BRCA2
acquired somatically (and, therefore, be in the tumor mutation–associated metastatic breast cancer. Two
only) or be inherited through the germline. It is important randomized phase III trials—OlympiAD and EMBRA-
to put processes in place to identify potential germline CA—demonstrated the superiority of these PARP in-
findings and perform germline testing to ensure optimal hibitors compared with standard-of-care chemotherapy
value for the patient and their family.5 Examples of the with respect to progression-free survival and overall re-
integration of genomic information into the care of pa- sponse rate.11-13 Importantly, both OlympiAD and the
tients with cancer include germline testing for BRCA1/2 EMBRACA studies also demonstrated improved quality
in breast, ovarian, pancreatic, and prostate cancer; of life with the use of a PARP inhibitor compared with
evaluation of mismatch repair in endometrial cancer; chemotherapy. Thus, assessment of BRCA1/2 status in
and somatic sequencing in lung cancer. patients with metastatic breast cancer is recommended.6
Individuals with BRCA1/2 mutation–associated meta-
BRCA1 AND BRCA2 MUTATIONS AND PARP INHIBITORS static pancreatic cancer7,14,15 or prostate cancer7,16,17
Author affiliations also benefit from PARP inhibitors. In the phase III
Multiple PARP inhibitors have been approved by the
and support
information (if
U.S. Food and Drug Administration (FDA) as cancer POLO study, compared with placebo, olaparib mainte-
applicable) appear therapeutics. PARP inhibitors have several putative nance led to an improvement in progression-free
at the end of this mechanisms of action, including the induction of double- survival14 in germline BRCA1/2 mutation–associated
article. stranded breaks after stalling and collapse of the DNA metastatic pancreatic cancer, leading to FDA approval
Accepted on April 8, replication forks. Tumors in which there is a defect in for this indication. Rucaparib, olaparib, and niraparib
2020 and published
homologous DNA repair (and, thus, defect in repair of have shown promising results in BRCA1/2 mutation–
at ascopubs.org on
XXXX, XX: DOI https://
double-stranded breaks), such as those associated associated metastatic prostate cancer,16,17 although they
doi.org/10.1200/ with mutations in BRCA1 and BRCA2, appear to be are not yet approved by the FDA for this indication. The
EDBK_280607 particularly susceptible to PARP inhibitors. National Comprehensive Cancer Network recommends

Domchek et al
microsatellite loci that are commonly assayed in polymerase

chain reaction–based testing. The analyte for these tests is
DNA extracted from formalin-fixed paraffin-embedded en-
• Identification of germline mutations in cancer- dometrial cancer samples that are submitted by community
susceptibility genes may influence cancer
providers for testing and signed-out clinical results. We pro-
treatment.
duce NGS libraries using automation and include universal
• Somatic testing aimed at identifying therapeutic molecular indices on the individual library adapters, permit-
targets may reveal potentially germline findings, ting sample multiplex input to the hybrid capture step and
which are important to recognize.
a resulting economy of scale. High-depth NGS data coverage
is analyzed by a tumor-only variant-detection pipeline that
genetic testing for all individuals with pancreatic cancer or provides information on variant allele fraction for each iden-
metastatic prostate cancer. Whether PARP inhibitors will tified variant, to assign a putative somatic or germline status.
demonstrate benefit in germline BRCA1/2 mutation carriers Initially, we will confirm the suspect germline variants with
with noncanonical tumors is the subject of ongoing study. a CAP-validated assay from Color Genomics (Burlingame, CA).
Ultimately, with sufficient training data in hand from these
The use of multigene panel testing allows for cost-effective
comparisons, we will design and test a machine learning–
sequencing of genes beyond BRCA1 and BRCA2, many of
based predictor to supplant the need for secondary germline
which, like BRCA1 and BRCA2, are also involved in ho-
testing. For patients enrolled who have a confirmed Lynch
mologous DNA report (e.g., ATM, BRIP, BARD1, CHEK2,
syndrome variant detected and signed out, genetic counseling
PALB2, RAD51C, and RAD51D).18 Studies are ongoing, but
will be offered.
efficacy of PARP inhibitors in tumors associated with muta-
tions in these genes should not be assumed. Although early The read coverage data from each microsatellite locus are
studies suggest benefit in PALB2 mutation–associated tu- evaluated by modeling the variable repeat lengths obtained at
mors, data may be less compelling for tumors associated with each locus and analyzed by a machine learning–based pre-
CHEK2 and ATM mutations.17 Thus, studies are needed to dictor that was trained on data from known microsatellite-stable
determine which tumor types and mutations in which genes and -instable endometrial cancer samples. We also used these
may predict response to PARP inhibitors. training data to bootstrap a range of scores from this predictor
that indicate indeterminate microsatellite status, thereby pro-
NEXT-GENERATION SEQUENCING PANEL APPROACHES TO viding higher confidence in the prediction of microsatellite-stable
NEW THERAPEUTIC INDICATIONS IN and -instable samples. For patients with a clinical microsatellite-
ENDOMETRIAL CANCERS instable status, treatment with anti–PD-1 checkpoint blockade
Large-scale cancer discovery in endometrial cancer con- therapy (pembrolizumab) will be indicated.2
ducted through the The Cancer Genome Atlas19 provided We have used this assay to test more than 250 endometrial
numerous insights into the underlying genomics of this cancer cancer samples submitted by Ohio community oncologists,
type. When combining these results with existing knowledge with continued testing and the aforementioned developments
and data from recent clinical trials, new therapeutic and planned over the next few years. In addition to the clinical
patient monitoring indications have emerged. We sought to testing being performed for these patient samples, other loci
combine these new directions into a stand-alone College of are captured by the OPTEC panel. Seventy-two genes are
American Pathologists (CAP)-validated next-generation se- being evaluated for mutational status, and there are probes
quencing (NGS) panel test with accompanying computational spaced at approximately 30 kb to evaluate copy number
analyses and to offer the test to an Ohio-wide network of alterations along all chromosomes. These additional data will
gynecological oncology providers. This approach provides be studied in the context of correlative science aims, with
important germline testing for Lynch syndrome genes with a future goal of offering expanded testing to patients with
accompanying therapeutic choice indications and enables endometrial cancer that includes new genetically indicated
the identification of patients with microsatellite-instable dis- therapeutic developments. All novel variants detected in the
ease who may benefit from anti–PD-1 checkpoint blockade course of clinical and research testing by the OPTEC panel will
therapy.20 Finally, we also assay for germline and/or somatic be contributed to ClinVar.
POLE variants, which are indicative of women with the best
disease outcomes, despite their hypermutated status. GENOMIC TESTING IN THE STANDARD MANAGEMENT OF
The resulting Ohio Prevention and Treatment of Endometrial LUNG AND THYROID CANCER
Cancer (OPTEC) assay is based on IDT xGEN Lockdown Molecular testing has become a standard and critical part of
hybrid capture technology, providing more than 500-fold NGS the diagnostic work-up for patients with advanced non–
read data coverage across all exons of Lynch syndrome genes small cell lung cancer (NSCLC) to identify optimal therapy
(PMS2, MLH1, MSH6, MSH2) and POLE, as well as 21 and to avoid potential toxicities of ineffective therapies.

TABLE 1. Examples of FDA Approvals Based on Specific Germline or Somatic Alterations

Target Drug Tumor Type FDA Approval for Germline or Somatic
dMMR/MSI-H Pembrolizumab Tumor site IHC or MSI: somatic implications for those with germline mutations in MLH1, MSH2,
agnostic MSH6, PMS2
NTRK fusion Larotrectinib Tumor site Somatic
agnostic
Entrectinib
EGFR mutations Erlotinib Lung cancer Somatic
Gefitinib
Afatinib
Osimertinib
BRCA1 and BRCA2 Olaparib* Ovarian cancer Germline and somatic
mutations
Breast cancer Germline
Pancreas cancer Germline
Rucaparib* Ovarian cancer Germline or somatic
Talazoparib Breast cancer Germline
Abbreviations: FDA, U.S. Food and Drug Administration; dMMR, mismatch repair deficient; MSI, microsatellite instability; IHC, immunohistochemistry; MSI-I,
microsatellite instability-high.
*Olaparib, rucaparib, and niraparib also have approval in ovarian cancer beyond BRCA1/2 mutations.6
Because diagnostic samples in NSCLC are often obtained inherited RET genetic syndrome. This typically presents in
via minimally invasive techniques with fine needle aspiration the third or fourth decade of life and is associated with other
and/or core needle biopsies, the tissue available for testing is neoplastic conditions, including parathyroid adenomas and
limited. Starting with targeted sequencing for specific mu- pheochromocytomas.25 Because of the later onset of dis-
tations in the EGFR mutation, the standard has now evolved ease, thyroidectomy is recommended by age 5. Familial
into a broader panel of genomic tests. There are currently MTC is considered a variant of MEN2A and is similarly
more than 20 unique genomic alterations that are recog- associated with mutations in exons 10, 11, and 13, which
nized as potential therapeutic targets in lung cancer. At least confer later onset of disease.26 Patients with MEN2B may
five of these alterations, namely EGFR, ALK, ROS1, BRAF have associated pheochromocytoma, marfanoid habitus, or
and TRK, have been validated with approved agents as mucosal neuromas and, importantly, have earlier onset and
standard-of-care therapies. Investigational agents are also more aggressive MTC. The mutations common in MEN2B
in clinical development for other emerging targets of in- are located in exons 14 through 16, and because of their
terest. Comprehensive genomic testing in multiple large high risk, prophylactic thyroidectomy is recommended at
series identifies an oncogenic driver alteration in more than age 1 year. Laboratory-based screening for pheochromo-
50% of patients with NSCLC who are tested. Approximately cytomas and hyperparathyroidism in the case of MEN2A is
30% of these patients harbor genetic alterations for which recommended annually for all patients with MEN.
there are approved targeted therapies.21-23 Aberrations in Practical Considerations for Testing
other genes for which targeted therapies are in advanced
stages of development include RET rearrangements, MET Until fairly recently, genomic analysis of tumor tissue was
amplifications or exon 14 skip rearrangements, ERBB2 the primary route for molecular testing in solid tumors.
Blood-based testing, or so-called “liquid biopsy,” is rapidly
(HER2) mutations, and KRAS mutations.
growing in popularity and serves an important role in di-
Actionable genetic mutations in thyroid cancer involve agnostic and treatment decisions.27 Optimal use of these
a single gene, the RET proto-oncogene, in patients with technologies requires understanding the benefits and lim-
medullary thyroid carcinoma (MTC). Although the majority itations of these platforms. Tissue testing has the inherent
of patients with MTC have sporadic acquired mutations in benefit of providing histologic and phenotypic information,
RET, autosomal-dominant germline mutations account for as well as validated processing protocols. This is particularly
up to 25% of cases, with important implications for family relevant in lung cancer, when histologic transformation from
testing and genetic counseling. Sporadic MTC is commonly non–small cell histology to small cell histology significantly
seen in middle-aged adults, and appropriate genetic testing changes the therapeutic options.28 As the understanding of
can unveil germline mutations in up to 6%.24 Multiple molecular mechanisms of resistance has grown, so has the
endocrine neoplasia 2A (MEN2A) is the most common need for serial testing, which highlights many of the pitfalls

Domchek et al
associated with invasive biopsies. Drawbacks include the risk alterations, which have recently been used to identify four
of procedural complications, variable operator skill, patient distinct subtypes of small cell lung cancer. Prior to this work,
concerns, and sampling issues related to tumor heterogeneity. DNA-based profiling showed uniform loss of the tumor
Venous blood collection is safe and easy, enabling serial suppressor genes TP53 and RB1 but no potential thera-
testing. Moreover, circulating tumor DNA (ctDNA) analysis peutic oncogene targets.33 Dominant expression of the
can give a more complete profile of tumor alterations, albeit transcription factors ASCL1, NEUROD1, YAP1, and POU2F3
with the assumption that tumor DNA is shed equally. have been shown to be mutually exclusive in animal models
The use of ctDNA is evolving in lung cancer. In a prospective and human tumors.34 The two most common subtypes of
phylogenetic NGS analysis of patients with early-stage lung small cell lung cancer are tumors high in ASCL1 or NEU-
cancer, median variant allele frequency correlated with ROD1, which have typical neuroendocrine features. The
tumor volume.29 Patients were followed clinically with ex- subtypes defined by high YAP1 or POU2F3 are less common
amination and chest radiographs; ctDNA detection pre- and are associated with tumors that lack staining of typical
ceded clinical relapse by a median of 70 days. Ongoing and neuroendocrine markers.35 The clinical and therapeutic im-
planned clinical trials are assessing its clinical utility in plications of these distinct molecular subtypes of small cell
identifying minimal residual disease following surgery, to lung cancer are under active investigation.
tailor adjuvant therapy for those patients who need it, and to
avoid toxicities in patients who have been cured. There are also CONCLUSIONS
signs of clinical utility in the setting of advanced or metastatic
disease. Testing for the T790M resistance mutation using Germline and somatic sequencing in cancer is already of
ctDNA has been well established in patients with EGFR-mutant significant value in specific cancer types, with multiple FDA
lung cancer.30 Early ctDNA clearance has been shown to approvals of medications targeted to specific genomic alter-
correlate with outcomes in patients with targetable EGFR ations. Although most of these FDA approvals are disease
mutation or T790M mutation taking osimertinib and in patients restricted, several are “tumor agnostic” (pembrolizumab in
with nondriver-mutant NSCLC receiving immunotherapy.31,32 tumors with deficient mismatch repair and larotrectinib and,
subsequently, entrectinib in tumors with NTRK fusions). So-
RNA Sequencing matic sequencing needs to be examined with considerations of
Yet another layer of molecular profiling is interrogating whether findings could be germline, because this may have
intronic alterations, such as the exon 14 skip mutation in significant implications for the family, as well as the patient.
NSCLC and gene expression level differences or epigenetic New techniques will lead to even greater improvements.

1
Basser Center for BRCA, Abramson Cancer Center, University of Susan M. Domchek, Basser Center for BRCA, Abramson Cancer Center,
Pennsylvania, Philadelphia, PA University of Pennsylvania, 3400 Civic Center Blvd., Philadelphia, PA
2
Institute for Genomic Medicine, The Ohio State University, Columbus, 19104; email: [email protected].
OH
3
Department of Hematology and Medical Oncology, Winship Cancer
Institute of Emory University, Atlanta, GA
EDBK_280607.
REFERENCES
1. Parkes A, Arun BK, Litton JK. Systemic treatment strategies for patients with hereditary breast cancer syndromes. Oncologist. 2017;22:655-666.
2. Domchek SM. Risk-reducing mastectomy in BRCA1 and BRCA2 mutation carriers: a complex discussion. JAMA. 2019;321:27.
Oncol. 2016;27:795-800.
4. Schrader KA, Cheng DT, Joseph V, et al. Germline variants in targeted tumor sequencing using matched normal DNA. JAMA Oncol. 2016;2:104-111.
5. Clark DF, Maxwell KN, Powers J, et al. Identification and Confirmation of Potentially Actionable Germline Mutations in Tumor-Only Genomic Sequencing. JCO
Precis Oncol. 2019;3.
6. Banerjee SN, Lord CJ. First-line PARP inhibition in ovarian cancer - standard of care for all? Nat Rev Clin Oncol. 2020;17:136-137.

7. Kaufman B, Shapira-Frommer R, Schmutzler RK, et al. Olaparib monotherapy in patients with advanced cancer and a germline BRCA1/2 mutation. J Clin Oncol.
2015;33:244-250.
8. Moore K, Colombo N, Scambia G, et al. Maintenance olaparib in patients with newly diagnosed advanced ovarian cancer. N Engl J Med. 2018;379:2495-2505.
9. Konstantinopoulos PA, Norquist B, Lacchetti C, et al. Germline and somatic tumor testing in epithelial ovarian cancer: ASCO Guideline. J Clin Oncol. 2020;
38:1222-1245.
10. Daly MB, Pilarski R, Yurgelun MB, et al. NCCN Guidelines Insights: Genetic/Familial High-Risk Assessment: Breast, Ovarian, and Pancreatic, Version 1.2020.
J Natl Compr Canc Netw. 2020;18:380-391.
11. Robson M, Im SA, Senkus E, et al. Olaparib for metastatic breast cancer in patients with a germline BRCA mutation. N Engl J Med. 2017;377:523-533.
13. Litton JK, Rugo HS, Ettl J, et al. Talazoparib in patients with advanced breast cancer and a germline BRCA mutation. N Engl J Med. 2018;379:753-763.
15. Shroff RT, Hendifar A, McWilliams RR, et al. Rucaparib monotherapy in patients with pancreatic cancer and a known deleterious BRCA mutation. JCO Precis
Oncol. 2018;2018.
16. Mateo J, Carreira S, Sandhu S, et al. DNA-repair defects and olaparib in metastatic prostate cancer. N Engl J Med. 2015;373:1697-1708.
17. Abida W, Campbell D, Patnaik A, et al. Non-BRCA DNA damage repair gene alterations and response to the PARP inhibitor rucaparib in metastatic castration-
resistant prostate cancer: analysis from the phase 2 TRITON2 study. Clin Cancer Res. Epub 2020 Feb 21.
18. Robson M, Domchek S. Broad application of multigene panel testing for breast cancer susceptibility-Pandora’s Box is opening wider. JAMA Oncol. 2019;5:1687.
19. Kandoth C, Schultz N, Cherniack AD, et al; Cancer Genome Atlas Research Network. Integrated genomic characterization of endometrial carcinoma. Nature.
2013;497:67-73.
21. Pakkala S, Ramalingam SS. Personalized therapy for lung cancer: striking a moving target. JCI Insight. 2018;3.
22. Barlesi F, Mazieres J, Merlio JP, et al; Biomarkers France contributors. Routine molecular profiling of patients with advanced non-small-cell lung cancer: results
of a 1-year nationwide programme of the French Cooperative Thoracic Intergroup (IFCT). Lancet. 2016;387:1415-1426.
23. Kris MG, Johnson BE, Berry LD, et al. Using multiplexed assays of oncogenic drivers in lung cancers to select targeted drugs. JAMA. 2014;311:1998-2006.
24. Elisei R, Romei C, Cosci B, et al. RET genetic screening in patients with medullary thyroid cancer and their relatives: experience with 807 individuals at one center.
J Clin Endocrinol Metab. 2007;92:4725-4729.
25. Raue F, Frank-Raue K. Multiple endocrine neoplasia type 2: 2007 update. Horm Res. 2007;68(suppl 5):101-104.
26. Wells SA Jr., Asa SL, Dralle H, et al; American Thyroid Association Guidelines Task Force on Medullary Thyroid Carcinoma. Revised American Thyroid Association
guidelines for the management of medullary thyroid carcinoma. Thyroid. 2015;25:567-610.
27. Merker JD, Oxnard GR, Compton C, et al. Circulating tumor DNA analysis in patients with cancer: American Society of Clinical Oncology and College of American
Pathologists Joint Review. J Clin Oncol. 2018;36:1631-1641.
28. Jiang SY, Zhao J, Wang MZ, et al. Small-cell lung cancer transformation in patients with pulmonary adenocarcinoma: a case report and review of literature.
Medicine (Baltimore). 2016;95:e2752.
29. Abbosh C, Birkbak NJ, Wilson GA, et al; PEACE consortium. Phylogenetic ctDNA analysis depicts early-stage lung cancer evolution. Nature. 2017;545:446-451.
30. Jenkins S, Yang JC, Ramalingam SS, et al. Plasma ctDNA analysis for detection of the EGFR T790M mutation in patients with advanced non-small cell lung
cancer. J Thorac Oncol. 2017;12:1061-1070.
31. Raja R, Kuziora M, Brohawn PZ, et al. Early reduction in ctDNA predicts survival in patients with lung and bladder cancer treated with durvalumab. Clin Cancer
Res. 2018;24:6212-6222.
32. Song Y, Ma S, Shi M, et al. Predictive and prognostic values of circulating tumor DNA (ctDNA) clearance in osimertinib treated advanced non-small cell lung
cancer cohort. J Clin Oncol. 2019;37(15 suppl):3036.
33. George J, Lim JS, Jang SJ, et al. Comprehensive genomic profiles of small cell lung cancer. Nature. 2015;524:47-53.
34. Rudin CM, Poirier JT, Byers LA, et al. Molecular subtypes of small cell lung cancer: a synthesis of human and mouse model data. Nat Rev Cancer. 2019;
19:289-297.
35. Huang YH, Klingbeil O, He XY, et al. POU2F3 is a master regulator of a tuft cell-like variant of small cell lung cancer. Genes Dev. 2018;32:915-928.

CARE DELIVERY AND
REGULATORY POLICY

From Theory to Practice: Implementation of

Strategies to Reduce Acute Care Visits in Patients
With Cancer
Bobby Daly, MD, MBA1,5; Laura C. Michaelis, MD2; John D. Sprandio, MD3,4; Jonathan T. Kapke, DO4; Ravi Kishore Narra, MD2;
Elizabeth Malosh2; Alice Zervoudakis, MD1; Jessie Holland, RN6; and Melissa Zablocki, MPH5
overview
Patients with cancer frequently seek acute care as a result of complications of their disease and adverse
effects of treatment. This acute care comes at high cost to the health care system and often results in
suboptimal outcomes for patients and their caregivers. The Department of Health and Human Services has
identified this as a gap in our care of patients with cancer and has called for quality-improvement efforts to
reduce this acute care. We highlight the efforts of three centers—a community practice, an academic
practice, and a cancer center—to reduce acute care for their patients. We describe the foundational prin-
ciples, the practice innovation and implementation strategy, the initial results, and the lessons learned from
these interventions. Each of the described interventions sought to integrate evidence-based best practices for
reducing unplanned acute care. The first, a telephone triage system, led to 82% of calls being managed at
home and only 2% being directed to an emergency department (ED) or hospital. The second, a 24-hour
continuity clinic, led to a 26% reduction in ED utilization for patients with cancer. The third, a digital symptom
monitoring and management program for high-risk patients on active treatment, led to a 17% reduction in ED
presentations. There is a need for innovative care delivery models to improve the management of symptoms for
patients with cancer. Future research is needed to determine the elements of these models with the greatest
impact and how successful models can be scaled to other institutions.
INTRODUCTION of patients with cancer and poses a substantial fi-

Patients with cancer frequently seek acute care, de- nancial burden for the health care system.6 Acute care
fined as ED visits or hospitalizations, due to disease services account for approximately 48% of spending
complications and adverse effects of treatment. These and 67% of regional spending variation in Medicare
acute care visits are rising dramatically. ED visits for patients with advanced cancer, and a hospitalization
patients with cancer related to complications of sys- related to a treatment complication averages $22,000.7,8
temic or radiotherapy increased at a 5.5-fold higher Aside from the fiscal considerations, acute care visits
rate over 10 years compared with overall ED visits.1 A adversely affect patients and caregivers. DHHS stated
study of the Nationwide Emergency Department that these visits reduce patients’ quality of life by af-
Sample Healthcare Cost and Utilization Project found fecting their physical and emotional well-being, dis-
that from 2006 to 2012 adult cancer–related ED visits rupting their treatment schedules, decreasing their
were over three times more likely to result in an in- desire to engage in work and social activities, and
patient admission than noncancer-related visits.2 increasing the burden on their families.6 To incentivize
Similarly, a multicenter prospective cohort study that strategies to close this care gap, the Centers for
Author affiliations
and support included 18 EDs affiliated with the Comprehensive Medicare and Medicaid Services developed a new
information (if Oncologic Emergencies Research Network found that quality measure, OP-35. This measure evaluates
applicable) appear 57% percent of patients presenting with active can- providers based on ED visits and admissions for pa-
at the end of this
cer were admitted. 3 The most common presenting tients receiving chemotherapy for 10 conditions that
article.
symptoms were pain, shortness of breath, abdominal could potentially be preventable through improved
Accepted on March
4, 2020 and pain, and nausea. It has been asserted that many of outpatient care: anemia, dehydration, diarrhea, eme-
published at these ED visits are avoidable with timely outpatient sis, fever, nausea, neutropenia, pain, pneumonia, and
ascopubs.org on May care.4,5 sepsis (Fig. 1).6
18, 2020: DOI https://
doi.org/10.1200/ This acute care has been identified by the Department The goal in creating OP-35 was to focus resources and
EDBK_281139 of Health and Human Services (DHHS) as a gap in care attention on preventable acute care in patients with

Daly et al
visits. Historically, care processes have not fully leveraged

PRACTICAL APPLICATIONS technology to drive productivity and consistency of care
delivery.
• This article details the use of a telephone triage
system to keep patients out of the hospital. Because of the variable processes and technologies in the
• This article describes the design and imple- office setting, there exist substantial inconsistencies in the
mentation of a 24-hour cancer continuity clinic way that PROs and symptoms are collected and presented
at an academic medical center. to clinicians, as well as how the care team’s explanations,
predictions, and plans are documented into an Institute of
• This article reviews the design and imple-
mentation of a digital symptom monitoring and Medicine–compatible oncology care management plan.
management program for high-risk patients on The consumability and accuracy of care management plans
anticancer treatment. are highly provider dependent, causing difficulty with
addressing whether escalating symptoms occurring be-
tween visits are related to chemotherapy, comorbidities, the
cancer and encourage quality improvement efforts that disease itself, noncompliance with previous supportive
promote more effective communication and coordination recommendations, or combinations of the above. The in-
of care, as well as more effective prevention and treatment ability to quickly establish clinical visibility, devise a differ-
practices.6 Recently, Handley et al reviewed efforts to ential diagnosis, and confidently execute care leads to
reduce avoidable acute care10 and identified five best costly, time-consuming, and unnecessary ED visits and
practices that had success across the primary measures potentially dangerous hospital admissions.
of acute care for patients with cancer: ED visits, acute
hospitalizations, and 30-day rehospitalizations. These Practice Innovation/Implementation Strategy
strategies are (1) identify patients at high risk for un- The components of the intervention at Consultants in
planned acute care, (2) enhance access and care co- Medical Oncology and Hematology (CMOH) are described
ordination, (3) standardize clinical pathways for symptom below (Fig. 2).
management, (4) develop urgent cancer care tactics, and
(5) use early palliative care.10 Varying levels of evidence Processing PROs through collection, decision making, and
support these interventions but the authors assert that documentation CMOH developed a highly efficient physi-
integrating several of these strategies is likely to have the cian graphic user interface to support the care team in the
greatest impact. collection, grading, and longitudinal presentation of PROs
and symptoms. Providers are no longer responsible for
In this article, we describe the efforts of three centers—a recording and mining data, but rather synthesizing it to
community practice, an academic practice, and a cancer deliver factual explanations, predictions, and plans at the
center—to integrate elements of these best practices into point of care. The supportive processes were reengineered,
cancer care delivery at their institutions and review the such that the grading and documentation of PROs would
implementation strategies necessary to embed them into drive every aspect of the outpatient visit, bringing all
the workflow. For each care-delivery intervention, we de- clinical issues into focus for optimal management. CMOH
scribe the foundational principles, the practice innovation providers are now empowered to facilitate shared de-
and implementation strategy, the initial results, and the cisions in the context of the underlying disease, comorbid
lessons learned. Each center made substantial capital and conditions, goals of therapy, and the patient’s wishes. The
human resource investments to address this critical cancer result is a care management plan that is consumable by
care delivery issue, and the findings could have implications everyone.
for other organizations hoping to keep their patients at home
and work and out of the ED. Symptom-management guidelines Organizations like ASCO,
the National Comprehensive Cancer Network, the Oncology
CONSULTANTS IN MEDICAL ONCOLOGY AND HEMATOLOGY: Nursing Society, and the American Cancer Society have
TELEPHONE TRIAGE TO KEEP PATIENTS OUT OF made concerted efforts to provide evidence-based options
THE HOSPITAL for the management of common or predictable symptoms
Foundational Principles related to cancer care. These guidelines require a contex-
tual clinical decision to use a specific treatment option from
Effective symptom management is a daunting challenge in
multiple possible solutions. These served as the basis of
cancer care. Central to meeting that challenge is the reliable
discussion and eventual consensus in the development of
longitudinal collection of patient-reported outcomes (PROs)
symptom algorithms.
and the grading of active symptoms in the context of
comorbidities, medications, psychosocial factors, and dis- Practice-wide development of data-driven symptom
ease state during face-to-face encounters and between algorithms Symptom algorithms have three domains:

Strategies to Reduce Acute Care Visits in Patients With Cancer
FIGURE 1. Conditions in Patients

With Cancer Potentially Prevent-
able by Improved Outpatient Care
Ten conditions identified by the
Centers for Medicare and Med-
icaid Services that could be po-
tentially prevented through ap
propriately managed outpatient
care and communication.9
general assessment, nursing, and physician. General as- backup available. After connecting with the patient, a nurse
sessment provides high-level feedback regarding the reviews the care management plan from the most recent
severity and location of symptoms that would require an visit, which contains every detail of previous symp-
ED or unscheduled office visit (e.g., severe pain or tom management recommendations, recent medication
symptoms of dehydration). The nursing domain narrows changes, and the entire treatment history and plan in one
the focus to detailed clinical decision trees with specific digestible summary.
care recommendations agreed upon by the practice and
The nurse then follows the symptom algorithm decision
executed as standing orders. The standardized algorithm
trees down branches to possible outcomes: management
allows the triage nurse team to operate at the level of their
recommendations, same-day office visits, or ED evaluations.
license, dispensing medical advice or confidently de-
A process for nurse follow-up calls is triggered for at-home
termining the need to escalate to the physician domain by
management. The entire interaction is documented in the
ordering an emergent, unscheduled physician or ED
electronic medical record, with immediate physician noti-
evaluation.
fication of triage activity, importantly, without physician
The consistency of actions between the recommendations workflow interruption. The documented triage call is also
across the care team, triage algorithms, and written pa- presented to the physician at the next face-to-face
tient instructions minimizes contradictory directives. encounter.
Symptom algorithms are continuously re-evaluated, and
outcomes data govern subsequent changes to the Recording the disposition Before closing out the call, the
standing orders. nurse documents the disposition of the clinical call into
structured data. Along with the number of calls and the time
Patient and family education and engagement Patients and of incoming calls (office hours vs. evening or weekend),
their family members are fully engaged at the first visit and at CMOH analyzes the percentage of calls in each of three
every step, becoming partners in their care. The practice categories: (1) managed at home, (2) directed to the ED or
serves as the point of first contact for all matters, with direct admission, or (3) resulting in an unscheduled office,
symptoms reported early in the day via the nurse triage primary care, specialty, or imaging visit. Continuous review
phone service. If the patient feels the onset of a symptom at of outcomes defines the next quality-improvement project,
8:00 AM, call by 8:15 AM. Waiting until 4:00 PM will result in
driving efficacy.
a costly and lengthy ED intervention and, potentially, un-
necessary imaging and admissions. Initial Results
Executing symptom triage A consistent level of service at Reinforcement to “call early and often” has promoted the con-
a central location promotes efficiency, and, in turn, patient solidation (roughly 75%) of symptom calls between 7:30 AM
use. One nurse manages triage calls, with immediate and 5:00 PM on Monday through Friday, minimizing

Daly et al
FIGURE 2. Symptom Management Nurse Phone Triage
evening and weekend calls. In 2019, CMOH nurses triaged FROEDTERT HOSPITAL/MEDICAL COLLEGE OF WISCONSIN:
more than 4,350 clinical calls, with about 82% of calls 24-HOUR CONTINUITY CLINIC
being managed at home and only 2% directed to the ED Foundational Principles
or hospital.
High ED utilization among patients with cancer is partially
Based on social media reviews, patient engagement and attributed to a lack of oncology-specific alternatives to ED
access to quality symptom management drive patient and care. Expanded access to same-day oncology-specific
family satisfaction, further cementing the patient-practice outpatient care could reduce ED utilization, hospital ad-
relationship. missions, and cost of care. The concept of oncology-specific
same-day care has gained popularity at high-volume cancer
Lessons Learned centers nationally, although no formal enumeration or
Having an informed, trained, and supportive staff reliably registry for these centers is available. Given this back-
executing symptom management create a physician-led ground, and with the aims of reducing ED utilization, re-
care team, giving physicians time back without reducing ducing readmissions, and improving care satisfaction, a
their oversight of clinical care. The resultant data derived 24-hour Continuity Clinic (24hrCC) at Froedtert Hospital/
from a predictable and more uniform approach to man- Medical College of Wisconsin (FH/MCW) was developed.
agement of specific symptoms increases provider effi- This clinic opened in November of 2016 and provides
ciency. Prompt patient-coherent delivery of explanations, a physician or advanced practice provider (APP) staff en-
predictions, and symptom management plans faciliates vironment 24/7 for nonemergent care of patients treated for
reliance on the telephone triage service, as evidenced in cancer by a medical oncologist, hematologist, radiation
the reduction of independent patient presentations to oncologist, or surgical oncologist within the Froedtert Health
an ED. system. The clinic is considered to be outpatient and

requires a direct referral from one of our on-staff physicians, follow-up has a swollen leg and needs to be evaluated for
setting it apart from traditional urgent-care clinics. a venous thromboembolism) or via a phone triage (e.g.,
a patient calling the phone triage line on a Saturday morning
Practice Innovation/Implementation Strategy with a new rash on the legs). Before either patient is seen in
The Clinical Cancer Center at FH/MCW annually treats more the 24hrCC, the clinic attending, covering MD/APP, or fellow
than 4,500 patients who are newly diagnosed with cancer. would confirm there is capacity for the patient, specify who
In fiscal year 2018, there were over 297,000 outpatient visits will oversee care and disposition, and be available for calls
to the cancer center. When this project was conceived, there on this patient for the duration of their time in clinic. All
were 150 to 185 oncology-related visits monthly to our ED, patient-directed materials on this center clarify that referral
with an admission rate exceeding 55%. Common chief is required and patients should not appear without being
complaints included pain, shortness of breath, weakness, directed by their oncologist or hematologist.
dehydration, and uncontrolled nausea and vomiting.
Initial Results
The Medical College of Wisconsin expanded inpatient
Since implementation of the 24hrCC, ED utilization for pa-
cancer care services to a newly constructed inpatient area in
tients with cancer has fallen by 26%, defined as total number
late 2016. This included moving all inpatients with cancer to
of ED visits divided by the total number of completed out-
rooms located on two floors of the new building. Each floor
patient provider and outpatient treatment visits for all patients
was designed to contain 32 single-occupancy inpatient
with cancer. Various factors may have contributed to this
rooms. During strategic planning, leadership of the He-
observation, including increased provider awareness of re-
matology/Oncology Division met with hospital administrators
ducing unnecessary acute care use, more vigilant outpatient
and negotiated for two of those inpatient rooms to be set
management, and increased patient education regarding
aside for construction of our 24hrCC. The clinic was housed
preventative care and early intervention. However, in-
near the entrance to the new unit. By removing the wall
creased 24hCC utilization appears to correlate with de-
between the two rooms, the hospital created one large 670-
creased ED utilization. Of note, this may mean that the
foot space for 24-hour care of patients with cancer. The
patients directed to the ED present with higher acuity.
finished space incorporates four patient bays, two 50-foot
patient bathrooms, wall oxygen and suction, and a dedicated With respect to the inpatient admission rate, patients
nurse charting space. Locating the 24hrCC in proximity to the evaluated in the ED were twice as likely to be admitted to the
inpatient wards allows face-to-face communication between hospital compared with those seen in the 24hrCC. Certainly,
nurses when a patient is admitted from the clinic and means emergent ED encounters often appropriately require hos-
that the inpatient nocturnist staff can oversee the clinic op- pital admission; however, analysis of nonemergent patient
erations at night without losing geographic proximity to the visits also demonstrates that oncology-trained 24hrCC staff
inpatients whose care they manage. may have more experience and greater efficiency managing
the unique needs of patients with cancer than ED providers.
Staffing the clinic requires two tiers of providers: nursing
Additionally, oncology subspecialists staffing the 24hrCC
care and MD/APP staff oversight. Nursing care is the
are more likely than their ED counterparts to have prior
foundational component. An oncology nursing resource
experience with the patient. This continuity of care may
pool was created, and nurses were cross-trained for in-
better position a 24hrCC provider to understand the pa-
patient and outpatient care environments. All nurses who
tient’s clinical history and avoid an unnecessary hospital
elected to be part of this pool were on-boarded to the in-
admission. Finally, because 24hrCC providers are in-
patient wards and the outpatient oncology infusion center.
tegrated into the overall cancer care team, they may be
Of note, electronic health records differ between inpatient
better equipped to facilitate outpatient care to minimize
and outpatient care and required training on both electronic
unneeded hospitalization.
environments.
We also analyzed the overall admission rate for FH/MCW
Medical coverage for the clinic during weekly daytime hours
patients with cancer since the 24hrCC was instituted. In
is provided by a dedicated APP, who is overseen by the
a 15-month sample, postintervention, including patients
patient’s outpatient oncologist. After hours, weekends and
seen in the ED and in the 24hrCC, the inpatient admission
overnight, clinical care is managed by the inpatient core of
rate was 25% less than the admission rate during a baseline
hospitalists, nurse practitioners, physician assistants, and
preintervention period. The absolute decrease in admission
on-call hematologist/oncologists.
rate was 12%, with an admission rate of 34% during the
A key clarification for all patients and providers is that the postintervention period and 46% during the baseline pe-
24hrCC is not an urgent care model. All patients must be riod. These findings suggest that accessibility to the 24hrCC
referred. This may be from the outpatient clinic (e.g., was associated with an overall reduction in hospital ad-
a patient seen by her outpatient oncologist for breast cancer mission for our entire patient population.

Daly et al
Additionally, for select oncology patients, nonemergent digital management and engagement. InSight Care is
same-day care provided in the 24hrCC was associated with a technology-enabled program that identifies patients ini-
decreased imaging, electrocardiogram, and laboratory uti- tiating anticancer therapy who are at high risk for potentially
lization compared with care provided in the ED. One po- preventable acute care visits, monitors symptoms of en-
tential factor contributing to this observation is that the rolled patients daily, and intervenes as necessary. The
24hrCC is equipped to perform basic investigations, in- program seeks to enable seamless communication with
cluding imaging, and is staffed with a specialized care team patients about symptoms wherever they are and whenever
with experience in oncology that may be more likely to they are in need.
pursue a targeted diagnostic workup, which likely mini- During the build phase of the program, six specific patient
mized charges. The increased confidence FH/MCW on- experience transformation principles served as the foun-
cology providers have in sending patients who are sick, but dation for approaching the work. These principles guided
not acutely, to the 24hrCC rather than to the ED also helps decision making in the program and included the following:
avoid unnecessary ED visits.
1. From directive to empowering: promoting self-efficacy
The median hospital charge was $2,521 less for patients through patient education and coaching,
admitted from the 24hrCC compared with patients admitted 2. From isolated to supported: expanding the team of care
from the ED, and the median hospital charge was $1,162 and putting them within the patient’s reach,
less for patients discharged from the 24hrCC compared with 3. From complex to intuitive: reducing burden to patients
patients discharged from the ED. Cost differences were and making the patient experience intuitive, thus in-
possibly driven by less resource utilization in the 24hrCC creasing patient engagement,
than the ED. 4. From reactive to proactive: there before the need is
Lessons Learned known,
5. From information to insight: making data actionable, and
Despite numerous positive outcomes, hurdles to optimi-
6. From unknown to transparent: mitigating the fear of the
zation remain. Staffing was a consistent challenge. As the
unknown.
overall inpatient census climbed, the physicians and APPs
who helped to care for inpatients became busier, stretching With these as foundational principles, a digital symptom
their ability to oversee the care in the 24hrCC. In addition, monitoring and management program was created to keep
budgeting workforce is difficult because 24hrCC volumes are patients, caregivers, and the care team continuously con-
unpredictable. Currently, a full-time daytime physician as- nected. By keeping our patients in our line of sight, as well as
sistant helps manage patients and develops standard op- building on data-driven insights, we aimed to better deliver
erating procedures for communication, triage, and workforce. predictive, anticipatory, and proactive care.
Four years into this endeavor, the next goal is to determine Practice Innovation/Implementation Strategy
the feasibility of transitioning from a referred-care model to InSight Care leverages three interlocking elements of
an urgent-care model. Several key differences exist between innovation11 (Fig. 3).
these models of care. In the referred-care model, the vol-
ume and nature of complaints of patients is controlled. A novel risk-prediction model InSight Care utilizes a ma-
Patients are never in a waiting room because referrals only chine learning predictive analytics framework built from
occur when beds are available. Patients with symptoms 10,000 observations of patients starting anticancer treat-
consistent with emergent problems (i.e., head trauma in ment and refined to predict the risk of a potentially pre-
a patient with thrombocytopenia) can be triaged to the ED, ventable acute care visit based on 270 characteristics from
instead of initiating treatment at the 24hrCC and trans- the electronic medical record. A risk score is integrated into
ferring. Finally, moving to an urgent care model will likely the InSight Care digital platform via a web application,
require 24/7 oversight by physicians or APPs trained for this riskExplorer, which displays the top 10 characteristics
environment and the care goals. contributing to a patient’s risk. Patients in the top risk
quartile account for . 50% of preventable inpatient bed
MEMORIAL SLOAN KETTERING CANCER CENTER: INSIGHT days and would potentially benefit from more intensive
CARE—DIGITAL SYMPTOM MONITORING AND symptom monitoring and management.
MANAGEMENT FOR HIGH-RISK PATIENTS ON
ANTICANCER TREATMENT Digital monitoring Through a secure patient portal, enrolled
patients received a daily symptom assessment querying
Foundational Principles common symptoms associated with a potentially prevent-
There is an opportunity for a new model of cancer care able acute care visit. The dedicated InSight Care team,
delivery that provides proactive, coordinated, and partici- which consists of oncology practice nurses and APPs, was
patory care to patients on anticancer treatment through alerted if a patient reported a mild/moderate symptom

FIGURE 3. Project Vision: Ex-

tend Our Reach to Transform Our
Care
Abbreviations: EMR, electronic
medical record, NP, nurse prac-
titioner; PA, physician assistant.
(“yellow alert”) or a severe symptom (“red alert”), with red defined as enrollment of . 25% of patients starting anti-
alerts requiring an immediate response. Prior research by neoplastic therapy. Response rate was evaluated as the
Basch et al12,13 conducted at Memorial Sloan Kettering proportion of daily assessments completed, with a goal of
Cancer Center found that managing symptoms proactively, . 50% assessment completion in the 6 months after treat-
in patients receiving chemotherapy, through weekly ment start. Reported symptom types and the number of red
assessments led to a 30% enhancement in quality of life, and yellow alerts generated were also tracked. Acceptability
7% fewer ED visits, and a 5-month improvement in overall and perceived valued were assessed through qualitative
survival. However, focusing on high-risk patients and pro- interviews and observations by the Memorial Sloan Kettering
viding more intensive monitoring through daily symptom Cancer Center Strategy and Innovation Group’s human-
assessments facilitated by a constantly monitored mobile centered design team. Finally, acute care was tracked by
application could further enhance the impact. presentation rates of enrolled patients to the Memorial Sloan
Kettering Cancer Center Urgent Care Center compared with
Digital team-based care The InSight Care team acted as an model-identified high-risk patients who were not enrolled in
extension of the primary oncology team. However, unlike the the program.
primary oncology team, InSight Care clinicians engaged
with patients exclusively through a digital platform. Using an From October 15, 2018 to July 10, 2019, 342 patients at the
internal web-based application, the Symptom Tracker, the pilot site initiated intravenous anticancer therapy, and 100
team monitored symptom trends and their relationship (29%) were enrolled. Average age of enrolled patients was
to treatment and connected with patients 24/7 as needed 66 years (range, 31–87), and 45% were female. Thirty
by phone, through the patient portal, or by televisit. If percent of patients had a thoracic malignancy, followed by
a patient needed to be seen, the the right setting (e.g., a gastrointestinal malignancy (22%), a head and neck
outpatient clinic, a symptom care clinic, or the ED) was malignancy (22%), and other malignancies (26%). Re-
determined. sponse rate (completion of the daily assessments) in the first
30 days of the program was 67%; it was 60% at days 31
Initial Results through 90 and 56% at days 91 through 180. Ninety-three
The pilot program evaluated feasibility, acceptability, and percent of patients generated at least one red alert, with
perceived value, as well as the initial effect on acute care 74% of patients generating a red alert for pain, 53% gen-
usage. The program was designed to care for the quartile of erating it for activity, and 25% generating it for nausea. The
patients most likely to require a potentially preventable most frequent moderate/mild symptoms were constipation
acute care visit at treatment initiation. Feasibility was (85%), pain (73%), and activity (66%).

Daly et al
Patient interviews suggest that they valued: Lessons were also learned in the return of the patient-
1. Speedy responses: Patients appreciated rapid feedback reported data to the clinical teams. The InSight Care
from InSight Care after submitting assessments reporting team was on the front lines of receiving this information and
symptoms. “One time I included this issue in my survey, was tasked with synthesizing the data and intervening based
they reached out in 5 minutes. Nothing but good.” on patient reports. Determining the best way to return these
2. The feeling of having a safety net: Patients were reas- data to the providers and patients is still under development.
sured by having 24/7 access to clinicians. “The best part Providers could benefit from using these data to understand
is I don’t feel alone in this. They [InSight Care team] have how patients are responding to therapy and as a potential
been a lifesaver for me.” early warning sign of progression, whereas patients could
3. Convenience: Patients appreciated the ability to address use the data to better anticipate treatment toxicities. Opti-
issues without an in-person visit. “That was good. I didn’t mizing the presentation of these data to maximize value in
have to get dressed and go to another doctor’s appointment.” clinical care beyond symptom management interventions
remains a need.
In the InSight Care cohort, 22 of 100 patients had an Urgent
Care Center visit, for a presentation rate of 22% within Finally, human-machine interactions require further de-
6 months of enrollment. In comparison, 28 high-risk pa- velopment. The machine-learning model was used as
tients did not enroll in InSight Care during this same time a decision-support tool to assist oncologists in identifying
period, and 11 presented to the Urgent Care Center, for patients at high-risk for a potentially preventable ED visit to
a presentation rate of 39% within 6 months. This was not enroll in the program. However, how providers interact with
designed as an effectiveness study, but it provides a signal the model is still being analyzed with the goal of improving
that the intervention reduced acute care visits. machine-model collaborations in the future.
Lessons Learned DISCUSSION
The goal of the pilot was for continuous learning to make In creating the outpatient quality measure on unplanned ED
improvements prior to consideration of scaling it to other visits and admissions for patients on chemotherapy, the
areas of the institution. Two key lessons emerged. The first DHHS hoped to stimulate providers to address the origin of
was the need for the InSight Care team to build trust with the this care gap. They identified the root causes characterized
primary oncology teams. The InSight Care team was the first below. Each intervention addresses these root causes
point of contact for high-risk patients with symptoms, and through different means and uses a spectrum of the best
the primary oncology teams had to feel confident that these practices highlighted by Handley et al.10 Although none of
patients would be well taken care of and that the in- these interventions were tested against randomized alter-
terventions proposed by the InSight Care clinicians were natives, each has relied on important metrics to assess
aligned with their recommendations. This trust was built by feasibility and success. The metrics include presentation to
hiring nurses and nurse practitioners with experience in the ED, cost of care, and patient satisfaction. 6
oncology practice and/or supportive care. Many of these 1. The delayed onset of side effects that patients must
providers had several years of experience at our institution; manage at home: Chemotherapy and other antineo-
however, trust developed gradually and required that the plastics have side effects that are often predictable. Many
InSight Care team spend time in participating oncology organizations have created guidelines that assist pro-
clinics to foster relationships. Weekly huddles with the viders in anticipating and managing these toxicities.
participating providers and InSight Care clinicians facilitated However, patients are often at home, beyond the reach of
direct communication between the teams and identified providers, when side effects emerge. The described
clinic preferences. The second operational learning was that interventions shine a light on these toxicities, making
expertise by disease was essential for management of patient them accessible to those who can help. The interventions
symptoms. Initially, we had not organized the InSight Care facilitate patient reporting of these side effects; so they
providers by disease; rather, they cared for patients across are not struggling with them in isolation. The telephone
disease types. This led to inefficiencies in workflow (by having triage intervention focused on patient and family edu-
to communicate with multiple providers) and prevented cation and engagement, emphasizing frequent com-
development of a deep knowledge of particular malignancies munication to create a norm that moves symptoms from
and their therapies. Thus, providers were reorganized into a condition to be managed at home alone to a condition
disease-focused pods, in which they cared for patients with that should be addressed in collaboration with the on-
particular malignancies. This disease focus also enabled the cology team. FH/MCW recognized that many non-
InSight Care nurses and nurse practitioners to build stronger emergent conditions may be efficiently managed outside
relationships with the primary oncology teams and continue of an ED, at lower cost, and with a better patient ex-
to enhance their disease specific knowledge. perience. By creating the 24hrCC, patients’ conditions

could be addressed by a team that was familiar with their were escalated appropriately to the physician, and the
care, which is less likely to result in unnecessary testing nurse was empowered to schedule office visits as
or admissions. Thus, patients were not limited to staying necessary to address severe complaints, such as
home or going to an ED. Finally, InSight Care, through its escalating pain. In this way, the patient was directed to
mobile platform, built a system with a low barrier to the provider who can deliver the proper care at the
patient communication with a knowledgeable provider. right time. The 24hrCC was designed to be housed
Proactive symptom reporting enabled the detection of within the cancer care floor connecting patients to
early warning signs and prevented the escalation of oncology subspecialists who have familiarity with the
symptoms at home. patient. The 24hrCC providers were also integrated
2. Patients assuming little can be done about symptoms and into the overall cancer care team and, thus, were
not seeking assistance: Each intervention addressed this equipped to help manage care transitions. Finally, the
second root cause through developing programs that InSight Care team worked as an extension of the
were responsive to patient complaints. CMOH executed primary oncology team. The InSight Care team fol-
data-driven symptom algorithms that ensured consis- lowed patients longitudinally and, thus, gained an
tency of action across the care team and that patients extensive knowledge of the patient, their clinical
were given clear direction on symptom management, history, and their treatment. The digital tools, in-
and outcome data was under constant review to optimize cluding Symptom Tracker, were integrated into the
efficacy. These algorithms empowered nurses to act electronic health record, which allowed for further
independently and decisively, allowing for immediate grounding of the patient’s symptoms in their overall
care delivery and providing patients with assurance that clinical context.
symptom reporting led to an effective response. Through
its 24hrCC, FH/MCW created an environment that was
responsive to patient symptoms. The clinic was designed CONCLUSIONS
as a direct referral so that the primary oncology team In December of 2016, Congress passed the 21st Century
communicated directly with the provider in the 24hrCC to Cures Act authorizing $1.8 billion in funding for the Cancer
discuss the care plan and specify who oversees patient Moonshot initiative. A blue ribbon panel identified “im-
care in clinic. For InSight Care, daily symptom assess- proving management of symptoms across cancer treat-
ment were tied to symptom alerts that notified the cli- ments” as a research priority. Research that advances
nicians of any new or escalating symptoms. These cancer care delivery models and helps provide patients with
symptoms were then addressed in real time by a dedi- improved symptom care outside of the ED is needed. There
cated team. By having a nurse practitioner and a nurse remains a myriad of unaswered questions in the field of
on the team, supportive care medications, diagnostic improving quality of symptom care for patients with cancer.
tests, and education could be provided immediately in Strategies may need to differ based on geography (e.g., rural
response to these alerts, ensuring reliable care without vs. urban settings), and there may be interventions that are
interrupting the physician’s workflow. more appropriate for patients with solid tumors than for
3. Limited access to and communication with providers who those with hematologic malignancies. Workforce shortages,
can tailor care to the individual: Cancer care is in- building constraints, or health literacy may mean some
creasingly complex, and patients want their symptoms settings are more likely to benefit than others. The in-
addressed by providers who understand their disease terventions described reflect integration of those strategies
and treatment course. Each intervention created with the most evidence; however, further work is needed to
a system whereby complaints were escalated to understand the elements of these models that have the
a knowledgeable provider. In the telephone triage greatest impact and how these elements can be imple-
intervention, the calls requiring a medical intervention mented by, and disseminated to, other providers.
AFFILIATIONS
6
Department of Nursing, Memorial Sloan Kettering Cancer Center, New
1
Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, NY
York, NY
2
Division of Hematology and Oncology, Medical College of Wisconsin,
Milwaukee, WI
3
Consultants in Medical Oncology and Hematology, Broomall, PA
4
University of Wisconsin Cancer Center at ProHealth Care, Waukesha, WI
5
Department of Strategy and Innovation, Memorial Sloan Kettering Cancer
Center, New York, NY

Daly et al
CORRESPONDING AUTHOR AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Bobby Daly, MD, MBA, Memorial Sloan Kettering Cancer Center, One Dag AND DATA AVAILABILITY STATEMENT
Hammarskjold Plaza, 885 2nd Ave., New York, NY 10017; email: Disclosures provided by the authors and data availability statement (if
[email protected]. applicable) are available with this article at DOI https://doi.org/10.1200/
EDBK_281139.
REFERENCES
1. Jairam V, Lee V, Park HS, et al. Treatment-related complications of systemic therapy and radiotherapy. JAMA Oncol. 2019;5:1028-1035.
2. Rivera DR, Gallicchio L, Brown J, et al. Trends in adult cancer-related emergency department utilization: an analysis of data from the nationwide emergency
department sample. JAMA Oncol. 2017;3:e172450.
3. Caterino JM, Adler D, Durham DD, et al. Analysis of diagnoses, symptoms, medications, and admissions among patients with cancer presenting to emergency
departments. JAMA Netw Open. 2019;2:e190979.
4. Panattoni L, Fedorenko C, Greenwood-Hickman MA, et al. Characterizing potentially preventable cancer- and chronic disease-related emergency department
use in the year after treatment initiation: a regional study. J Oncol Pract. 2018;14:e176-e185.
5. Daly B, Nicholas K, Gorenshteyn D, et al. Misery loves company: presenting symptom clusters to urgent care by patients receiving antineoplastic therapy. J Oncol
Pract. 2018;14:e484-e495.
6. Department of Health and Human Services; Centers for Medicare & Medicaid Services: 42 CFR Parts 405, 412, 413, and 489. 2016. https://www.gpo.gov/fdsys/
pkg/FR-2016-08-22/pdf/2016-18476.pdf. Accessed May 15, 2017.
7. Brooks GA, Li L, Uno H, et al. Acute hospital care is the chief driver of regional spending variation in Medicare patients with advanced cancer. Health Aff
(Millwood). 2014;33:1793-1800.
8. Kolodziej M, Hoverman JR, Garey JS, et al. Benchmarks for value in cancer care: an analysis of a large commercial population. J Oncol Pract. 2011;7:301-306.
9. Centers for Medicare & Medicaid Services: Admissions and Emergency Department Visits for Patients Receiving Outpatient Chemotherapy Measure Technical
Report. Prepared by Mathematica Policy Research under subcontract to Yale New Haven Health Services Corporation/Center for Outcomes Research and
Evaluation, March 2016. https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/HospitalQualityInits/Measure-Methodology.html.
Accessed on March 23, 2020.
10. Handley NR, Schuchter LM, Bekelman JE. Best practices for reducing unplanned acute care for patients with cancer. J Oncol Pract. 2018;14:306-313.
11. Daly B, Baldwin-Medsker A, Perchick W. "Using Remote Monitoring to Reduce Hospital Visits for Cancer Patients." Harvard Business Review, November 2, 2019.
https://hbr.org/2019/11/usingremote-monitoring-to-reduce-hospital-visits-forcancer-patients.
12. Basch E, Deal AM, Kris MG, et al. Symptom monitoring with patient-reported outcomes during routine cancer treatment: a randomized controlled trial. J Clin
Oncol. 2016;34:557-565.
JAMA. 2017;318:197-198.

Expanding Access to Chimeric Antigen Receptor

T-Cell Therapies: Challenges and Opportunities
Ankit Kansagra, MD1; Stephanie Farnia, MPH2; and Navneet Majhail, MD, MS3
overview
Chimeric antigen receptor (CAR) T-cell therapy is a major advancement in the treatment of lymphoid ma-
lignancies, especially diffuse large B-cell lymphoma and acute lymphoblastic leukemia (ALL). Since the U.S.
Food and Drug Administration (FDA) approval of two CAR T-cell therapies, axicabtagene ciloleucel and
tisagenlecleucel, experience has highlighted various barriers to their broader access and use, including
challenges related to manufacturing a patient-specific product, high costs and inadequate reimbursement,
incomplete or nonsustained disease responses, and potential for causing life-threatening toxicities. Research
on disparities, application, and practice of hematopoietic cell transplantation (HCT) can inform opportunities
to address similar barriers to use of CAR T-cell therapies that are currently available as well as other cellular
therapies that are expected to become available in the near future. To ensure optimal patient outcomes, these
therapies should preferably be administered at centers that have experience and established quality processes
and practices. We review opportunities for centers, manufacturers, payers, and policy makers to address
barriers to care. We also provide a summary of available and alternative payments models for commercial CAR
T-cell and other cellular therapies.
INTRODUCTION pipeline and are expected to be available commercially

Axicabtagene ciloleucel (Axi-cel; Yescarta; Kite in the near future.
Pharma, Santa Monica, CA) and tisagenlecleucel OUTCOMES OF AXI-CEL AND TISA-CEL
(Tisa-cel; Kymriah; Novartis Pharmaceuticals, East
Hanover, NJ) are two CD19-directed CAR T-cell Both cellular therapy products provide sustained re-
mission in selected patients with relapsed/refractory
products that are currently approved by the FDA
disease who previously had very limited to no options
and the European Medicines Agency for the treatment
for disease control. Tisa-cel is approved for the treat-
of pre–B-cell ALL1 and specific subtypes of relapsed/
ment of patients up to age 25 who have ALL that is
refractory aggressive B-cell non-Hodgkin lymphoma.2
refractory or in second or later relapse. The phase II
Although these approvals have been transformative in
global ELIANA trial enrolled patients with relapsed/
the use of cellular immunotherapy in lymphoma and
refractory B-cell ALL, between ages 3 and 23 (75
leukemia, both drugs are complex cell therapy prod-
patients), who had not received prior anti-CD19 ther-
ucts and need specialized setup for production, ad-
apy. Patients received lymphodepleting chemotherapy
ministration, and management of toxicities. These before receiving a median dose of 3.1 106 cells/kg
products are also expensively priced relative to other CAR T cells.4 At the time of publication, the overall
cancer therapies at $375,000–$475,000 per dose, remission rate of patients who received an infusion of
with resultant restrictions on coverage and re- CAR T cells was 81%, with the median duration of
Author affiliations imbursement.3 These issues have led to concerns remission not reached. In a subsequent updated
and support regarding access and appropriate use. Furthermore, analysis, patients receiving Tisa-cel continued to show
information (if Axi-cel and Tisa-cel are approved by the FDA as drugs, sustained remission, with 66% overall survival; median
applicable) appear but, ultimately, they are cellular therapy products that
at the end of this
overall survival was not reached at the end of 24
article.
require experienced teams that can apply high-fidelity months.5 Since its approval, the Center for Interna-
Accepted on March
quality processes to ensure safe and reliable admin- tional Blood and Marrow Transplant Research
20, 2020 and istration, similar to what routinely occurs in HCT. (CIBMTR) has reported data on 159 patients treated
published at Clinical experience with the two commercially available with Tisa-cel, demonstrating similar efficacy and safety
ascopubs.org on agents provides an opportunity to understand access compared with the pivotal ELIANA trial. Overall, 88% of
April 28, 2020:
and economic issues around cellular therapies, which patients achieved complete remission. Considering the
10.1200/EDBK_ will facilitate the implementation of other CAR T-cell caveat that the ELIANA trial and the CIBMTR study
279151 and cellular therapies that are in the developmental used different toxicity grading criteria, severe (grade 3

Kansagra, Farnia, and Majhail
worse) CRS and neurologic events scored using the Lee

criteria occurred in 13% and 28% patients, respectively.
Overall survival at 18 months was 52%.9 Since its approval,
• CAR T-cell technology is currently in its infancy, real-world experience with Axi-cel has been described in
with only CD19-targeted therapy available in
a multicenter U.S. cohort of 295 patients reported to the
clinical practice.
CIBMTR. In this real-world experience, despite having
• Next-generation CAR T-cell and other cellular a larger proportion of patients with transformed and double-
therapies and manufacturing improvements will hit lymphoma, an older patient population compared with
help reduce production delays and improve
ZUMA-1 trial, and patients with worse performance status,
efficiency.
response rates and toxicities were comparable to that re-
• “Off-the-shelf” allogeneic CAR T-cell therapy ported in the ZUMA-1 trial.10
may provide a novel solution for improving
access to CAR T-cell therapy. Since the commercialization of these two CAR T-cell
products, several groups have reported real-world experi-
• Collaboration between national professional
societies, quality accreditation organizations, ence of efficacy and toxicity with Axi-cel and Tisa-cel,
outcome registries, and regulatory agencies is which largely mirrors the results of pivotal clinical trials
crucial in balancing access and optimizing (Table 1).11,12 Various guidelines within the United States2,13
outcomes. and Europe14 are published for appropriate patient selection
• Innovative payment models and newer payer for CAR T-cell therapy, management of toxicities, and short-
policies are urgently needed to limit financial and long-term monitoring of these patients. The guidelines
toxicity to patients and ensure adequate provide an important framework to individual CAR T-cell
reimbursement. treatment centers to develop protocols and standard op-
erating procedures for safe delivery of immune effector cell
therapy.
or worse) cytokine release syndrome (CRS) occurred in
47% and 14% patients, respectively.6 CLINICAL ADMINISTRATION OF AXI-CEL AND TISA-CEL
Tisa-cel is also FDA approved for the treatment of adult The process for treating patients with Axi-cel or Tisa-cel
patients with relapsed/refractory diffuse large B-cell lym- begins with patient referral and assessment, followed by
phoma. This approval was based on results from the phase a review of insurance coverage and appropriate authori-
II JULIET clinical trial, in which 111 patients (median age, zation; workup to assess disease status and eligibility for
56) were treated with Tisa-cel (median dose, 3.0 108 CAR treatment; apheresis and lymphocyte collection; CAR T-cell
T cells) after receiving lymphodepleting chemotherapy. manufacturing, lymphodepleting chemotherapy, and cell
Among the 93 patients treated in the United States and infusion; and, ultimately, care during recovery and man-
included in the efficacy analysis, the best overall response agement of short-term and long-term toxicities. Overall, it is
rate was 52%, including complete response in 40% of a complex process, and breakdown in the integrity of any
patients. The median progression-free survival had not been aspect of this process (e.g., apheresis, manufacturing, in-
reached for patients who had achieved complete response. fusion) can lead to serious patient harm and/or can com-
The incidences of CRS and neurotoxicity were 58% (23% promise patient outcomes. Furthermore, close coordination
grade 3–4) and 21% (12% grade 3–4), respectively. No of several teams within and outside the institution is required
therapy-related mortality was reported.7 The CIBMTR has to ensure safe and reliable delivery (e.g., treatment team,
subsequently reported real-world data on the use of Tisa-cel apheresis, cell processing laboratory, intensive care unit
in 70 patients with diffuse large B-cell lymphoma and team, and manufacturer). Given these issues, the care of
demonstrated safety and efficacy similar to what was ob- patients receiving Axi-cel and Tisa-cel in the United States is
served in the pivotal JULIET trial.8 presently limited to transplant centers with Foundation for
Axi-cel was FDA approved in October 2017 for the treatment the Accreditation of Cellular Therapy (FACT) accreditation
of patients with refractory high-grade B-cell non-Hodgkin for administering immune effector cell therapies. Of note,
lymphoma who have not experienced response or relapse centers without transplant programs can achieve immune
after two prior treatment regimens. In the phase I/II ZUMA-1 effector cell therapy accreditation through FACT. The re-
trial, 101 patients (77 with diffuse large B-cell lymphoma, quirement for FACT accreditation is driven primarily by
24 with other B-cell non-Hodgkin lymphoma) received centers, payers, and some regulators, because it reassures
autologous CAR T cells (median dose, 2 106 cells/kg) after several stakeholders, including patients, that the treat-
lymphodepletion with fludarabine/cyclophosphamide. The ment is being provided in centers with experienced teams
authors observed a response rate of 82%, and the median and appropriate quality processes in place to optimize
duration of response was 8.1 months. Severe (grade 3 or outcomes.

Access to Chimeric Antigen Receptor T-Cell Therapies
TABLE 1. Summary of Pivotal Trials and Postapproval Observational Studies of Axi-cel and Tisa-cel
Registration Trial Postmarketing Studies
9 7
ZUMA 1 JULIET Jacobson et al12 Nastoupil et al11 Pasquini et al10 Jaglowski et al8
Characteristics (Axi-cel) (Tisa-cel) (Axi-cel) (Axi-cel) (Axi-cel) (Tisa-cel)
Demographics
No. leukapheresed 119 165 104 295 NR NR
No. infused 108 111 91 274 533 63
Median (range) age, 59 (23–76) 56 (22–76) 64 (21–80) 60 (21–83) 61 (19–86) 65 (19–89)
years
ECOG PS 0-1 100% 100% 90% 81% 80% 82%
High risk IPI ( 3) 44% NR 46% 55% NR 46%
Bridging therapy 0% 92% 40% NR NR NR
Double/triple hit 11%* 27% 24% 23% 36% 30%
Prior autologous HCT 23% 49% 27% 33% 32% 22%
Outcomes
Median follow up, 27.1 14 5.6 3.9 6.2 5.8
months
Best ORR 83% 52% 71% 81% 74% 60%
Best CR 58% 40% 44% 57% NR 38%
Grade 3 CRS** 11% 22% 16% 7% 10% 4.3%
Grade 3 32% 12% 39% 33% 22% 4.3%
neurotoxicity†
Tocilizumab use 43% 14% 67% 63% NR 41%
Steroid use 26% NR 64% 64% NR 9%
Abbreviations: Axi-cel, axicabtagene ciloleucel; Tisa-cel, tisagenlecleucel; NR, not reported; ECOG PS, Eastern Cooperative Oncology Group performance
status; IPI, International Prognostic Index; HCT, hematopoietic cell transplantation; ORR, overall response rate; CR, complete response; CRS, cytokine
release syndrome.
*There were five patients with double/triple hit lymphoma out of 47 accessed.
**ZUMA-1 used Lee criteria and JULIET used Penn criteria for grading CRS.
†ZUMA-1 and JULIET used common terminology criteria for adverse events (CTCAE), version 4.03, for neurotoxicity.
OPPORTUNITIES FOR INCREASING ACCESS AND USE OF CAR opportunity of treatment earlier in the trajectory of relapse
T-CELL THERAPY when, possibly, the patient’s performance status is more
Access to and appropriate use of CAR T-cell therapy is optimal and facilitates coordination of logistics, such as
dependent on the complex interplay of several factors and insurance authorization and collection. In the case of
stakeholders, most prominent among whom are referring lymphoma, the ideal situation is to start planning for CAR
physicians, manufacturers, payers, and centers that ad- T-cell therapy at the time of first relapse, as patients start
minister these therapies. Some important factors that may salvage chemotherapy with a plan for autologous HCT, but
serve as a barrier to access are summarized below. We to quickly change course to Axi-cel or Tisa-cel if there is
focus primarily on areas in which delays in treatment occur inadequate response or disease progression after one to two
with contemporary CAR T-cell products, with the expectation cycles. Barriers to referral for CAR T-cell therapy have not
that improving technology (e.g., off-the-shelf cell therapies) been well described but likely mirror those known occur with
will alleviate some of issues. As noted previously, experience HCT, including perceptions and knowledge of referring
from HCT can be applied to enhance our understanding of physicians and health care disparities related to race/eth-
these factors and ways to address them.15-18 nicity, socioeconomic status, and distance, among other
factors.16,18 Education of referring clinicians to ensure timely
Patient Referral and Selection referral and to facilitate the process of promptly getting
To maximize a patient’s chance of receiving CAR T-cell a patient in to a treatment center will have an immediate
therapy, it is crucial for early referral of patients to a cen- impact in this area and requires a concerted effort between
ter that offers these therapies. Timely referral offers the treatment centers, professional societies, manufacturers,

and policy makers. This is going to be increasingly relevant control disease, whereas others may become ineligible and
as more CAR T-cell and other cellular therapy products die from rapidly progressive disease.
come to the market, both in the research and routine care
Simplifying the manufacturing process to achieve large-
spaces.
scale production of CAR T-cell therapy without compro-
Closely related to patient referral is selection of appropriate mising its personalized nature is a global challenge.
patients for CAR T-cell therapy. Patient selection criteria are Biopharmaceutical companies are more adapted to large-
continually evolving, and the challenge of selecting an volume production of drugs with a supply chain and
appropriate patient for commercial therapy in the United economies of scale. To shorten the time of production of
States ultimately falls on the individual center. Most CAR CAR T-cells, various approaches are being studied. A closed
T-cell therapy centers follow the HCT model, in which automated system (e.g., CliniMACS Prodigy; Miltenyi Biotec,
a multidisciplinary team reviews the disease-related mor- North Rhine-Westphalia, Germany) that does not require
bidity of the patient and alternative treatment options, the infrastructure and resources for a good manufacturing
keeping under consideration patient-related factors (e.g., practice facility is now available and has been shown to be
medical comorbidities, signs of active infections, perfor- feasible in early-phase clinical trials.19 However, the regu-
mance status, and disease trajectory). latory and quality assurance aspects for scaling this platform
to a multi-institutional level are not well understood. To
Payer Authorization further reduce production times and improve the scalability
An initial step for patients with private health insurance in of CAR T-cell production, companies/institutions are ex-
the United States is obtaining payer authorization to proceed ploring nonviral approaches (e.g., transposon/transposase-
with evaluation and, ultimately, infusion and care associated based system).20 Another attractive approach is off-the-shelf
with CAR T-cell therapy. Given the complexity and expense allogeneic CAR T-cell therapies generated using various
of these therapies, there is greater scrutiny from payers to genome-editing technologies, like transcription activator-
ensure that appropriate patient and center selection criteria like effector nuclease, CRISPR-Cas9, and zinc finger
are met. Treatment center and payer processes can add nuclease. 21,22 Natural killer cells engineered to express
a week or more to the treatment timeline for CAR T-cell a CAR have also been shown to be active in early-phase
therapies, which may be a challenge for patients with active trials. 23
and rapidly progressing disease. Most large payers and
experienced treatment centers now understand the urgency Toxicities of CAR T-Cell Therapy
for moving ahead with evaluation and treatment and have CRS and immune effect cell-associated neurotoxicity syn-
processes in place to expedite approval, although there is drome are two notable side effects of CAR T-cell therapy. To
variation in selection criteria and timelines among payers. mitigate risks of these toxicities and implement proper
Hence, there continues to be an opportunity to educate and monitoring strategies, the FDA has restricted Axi-cel and
collaborate with payers to streamline the approval process Tisa-cel to a Risk Evaluation and Mitigation Strategy pro-
for CAR T-cell therapies. This is especially relevant for gram. A second major issue has been differences in the
products for new indications and other cellular therapies grading of CAR T-cell therapy toxicity in clinical trials,
that are expected to be available in the coming years. Also, limiting our ability to understand the risk of toxicity across
and as noted above, timely referral to the treatment center products and enable cross-trial comparisons. To facilitate
facilitates and provides sufficient time to both centers and uniform reporting, the American Society for Transplantation
payers to accomplish this process. and Cellular Therapy has recently developed a grading scale
for CRS and immune effect cell-associated neurotoxicity
Wait Time During CAR T-Cell Manufacturing syndrome toxicity assessment.24 Multiple strategies are
There are various steps involved in the manufacturing of currently under investigation to mitigate toxicities with CAR
autologous CAR T-cell products, including leukapheresis, T-cell therapy. As the burden of disease has shown to be an
transportation to the manufacturing facility, and then important risk factor, a risk-adapted therapy with either
transportation of the product back to the treatment center debulking therapy to reduce disease burden or reduction of
after completion of quality checks. In ZUMA-1, the median the dose of CAR T-cells has been proposed. Second, cy-
time from leukapheresis to product delivery was 17 days, tokine release prophylaxis with tocilizumab has been
and in JULIET, the median time from enrollment to infusion studied, with a reduction in the incidence of severe CRS;
was 54 days.7,9 In the real-world experience with Axi-cel, the however, it had no benefit on immune effect cell-associated
median time from leukapheresis to initiation of lymphode- neurotoxicity syndrome.25 Last, multiple safety switches are
pleting chemotherapy was 21.5 days.11 This period can be being tested in CAR constructs to reduce toxicities through
psychologically daunting for the patients awaiting treatment, more precise control of cell proliferation and activity.26
and some patients may require bridging chemotherapy to Next-generation CAR T-cell products presently under

development are expected to be associated with a lower experienced and established personnel and infrastructure.
risk of CRS and immune effect cell-associated neurotox- FACT accreditation is one such mechanism that has been
icity syndrome. fundamental in advancing the safe and effective delivery of
immune effector (e.g., CAR T-cell) therapy and ensures that
Patient and Caregiver Out-of-Pocket Costs
the treatment center has met a minimal set of criteria for
Caregiver availability is an essential concern for patients their administration. A key point to note is that institutions
considering and receiving CAR T-cell therapy, especially other than transplant centers can apply for and achieve
because patients frequently need help with adherence to FACT accreditation if they can satisfactorily demonstrate
their care plan, monitoring of toxicity, and emotional support implementation of their standards. As noted previously, Axi-
during this intense process while dealing with active dis- cel and Tisa-cel are not yet available through all transplant
ease. Lack of a caregiver can be a barrier to HCT and has centers in the United States; hence, there is an opportunity
been shown to influence decisions to proceed with the to expand access while maintaining quality, because most
transplant, especially for recipients of allogeneic or out- transplant centers in the United States are FACT accredited.
patient HCT.27,28 In addition, per the Risk Evaluation and Given the complexity of CAR T-cell therapies and the quality
Mitigation Strategy program, patients are required to stay in processes required for their safe administration, it is ad-
close proximity to the treatment center for approximately visable to pursue administration of CAR T-cell therapies in
4 weeks after infusion, which can add to patient financial centers that have demonstrated some level of quality ex-
burden. Many potential solutions, including financial as- pertise, such as FACT accreditation.
sistance for caregivers, psychosocial and emotional sup-
port, and respite care, have been incorporated at treatment PAYMENT MECHANISMS FOR CAR T-CELL THERAPIES
centers, often tapping into resources used for HCT re- With costs ranging between $373,000 and $475,000 per
cipients. Patient assistance programs from drug manu- infusion, exclusive of patient care costs (e.g., evaluation,
facturers and payers that cover transportation and lodging apheresis, chemotherapy, and postinfusion care, including
can go a long way to reduce the burden on patients/families management of complications), there is keen interest in how
and improve access to therapy. to pay for CAR T-cell therapies. Contemporary payment
Geographic Access to CAR T-Cell Therapy Centers models are ill designed to address the reimbursement gap,
which remains unsustainable and a major challenge for
Distance to a treatment center may serve as a barrier to
broad patient access.
access for some patients. However, most of the population
resides within a reasonable driving distance from an HCT Discussion of alternative payment mechanisms or meth-
center. In a detailed evaluation of geographic access to HCT odologies has increased proportionately to the growth of the
services, Delamater et al29 found that 46.7% of the U.S. pipeline for high-cost cell and gene therapies. The models
population resides within 30 minutes of a transplant center. are a study in opposites, because they range from pack-
Access increases noticeably to 65.9% and 77.1% of the aging everything together in a one-time payment meant to
population when considering 60- and 90-minute travel encompass all care within a set period (bundled pay-
times to an HCT facility, respectively. These figures increase ments, global payments, case rates) to several variations of
to 85.3%, 90.6%, and 93.9% of the U.S. population for 120-, unbundling— spreading out payments over time, often with
150- and 180-minute access. As an example, fewer than 100 the payments contingent on clinical outcomes or member
centers are certified to give Axi-cel in the United States at eligibility (outcomes-based agreements [OBAs], value-
present. Considering that there are approximately 200 based payments, milestone-based contracts [MBCs]).
transplant centers in the country, there is an opportunity to Characteristics of the first commercial CAR T-cell products
expand availability of CAR T-cell therapies to other experi- have primarily required the specialized care teams and
enced and FACT-accredited transplant centers. inpatient structures previously associated with HCT, thus
also making CAR T-cell therapy payment a prime candidate
BALANCING QUALITY VERSUS ACCESS for the case rate or bundled payment mechanism associ-
A question that comes up in the context of widening access ated with HCT and with which these clinical and financial
to CAR T-cell therapies is the site of administration. To teams are already well accustomed. A survey of payment
optimize patient outcomes, a sophisticated infrastructure is structures has not been completed, but payer and provider
required to handle the scheduling logistics, from patient anecdotes indicate that global payment structures are, in
referral to infusion and beyond. Additionally, given that they fact, being used for a majority of CAR T-cell therapy epi-
are cell products, a robust set of quality processes must be sodes of care. These structures may start in the form of
followed to maintain the chain of custody and chain of single patient/case agreements, outlining the clinical ex-
identity for the CAR T-cell product. Hence, it is critical that pectations and associated payment for one patient or
cellular therapies be administered in centers that have a series of patients. With time and experience, both payers

and providers may be interested in transitioning from single providers. The general structure of the Novartis OBA in-
patient/case agreements to a broader contract outlining volves an agreement between provider and Novartis that
a global episode of care. the provider will participate in the OBA program for all el-
Future products with different clinical safety profiles may igible patients, regardless of their insurance payer. In this
challenge the current propensity toward case rates. Prod- structure, the provider will not receive an invoice for the
ucts with low rates of toxicity may lead to the provision of product if the treated patient does not achieve the antici-
CAR T-cell therapy in the outpatient setting and/or by pated complete remission status by the 35th day after in-
oncology teams without ties to a hematopoietic stem cell fusion. Novartis limits the use of the OBA to the pediatric
transplant operational structure. In these situations, toxicity- indication for Tisa-cel, which is priced at $475,000 com-
related follow-up care could be delivered by a different set of pared with $373,000 for large-cell lymphoma. Currently,
specialists than those initially administering the therapy or at there is no standard OBA in place by Novartis or Kite outside
a different institution entirely. Payers may elect to continue the pediatric indication; however, four states have been
with global payments by requiring the administering pro- successfully granted approval by the Centers for Medicare
vider to handle settlement among external teams or sites of and Medicaid Services for the use of an outcomes-based
care, or they may consider allowing all nonadministration rebate agreement with CAR T-cell therapy manufacturers
aspects of care to flow through the system per usual claims, for their beneficiaries.31 In the case of the Novartis OBA, the
depending on the contractual relationships it has with full price of the product is negated, as the provider is not
network providers. Given the price of the CAR T-cell invoiced and therefore does not bill the payer for the ac-
products therapies, there will likely continue to be quisition of the product. This type of full product cost waiver
restrictions around which providers can administer the is possible because of the use of the buy-and-bill drug
therapy and what administrative fees or markups can acquisition mechanism and the clinical evaluation point
be billed through to the payer. after the treatment of only 30 days. Hospital billing de-
partments may struggle with the delay this evaluation
Two other payment mechanisms, MBCs and OBAs, are also process creates for their standard billing practices, even
currently represented in CAR T-cell payment structures and when the timeframe is contained to 30–60 days after
may see growth in use. The MBC is a variation on the treatment. In other OBAs, a therapy is paid in full at the time
bundled payment/global payment model. MBCs define of treatment, and the clinical evaluation point may be
distinct phases of care; examples include cell collection, months to years in the future, creating challenges in tracking
product administration and hospitalization, and follow-up patients and limiting the potential rebate to between 17%
care, and they scope a series of payments according to each and 23% because of the operationalization of Medicaid Best
episode. The completion of each phase would trigger Price reporting requirements.
payment to the provider according to the contract. Although
seemingly very similar to the global payment contract, the A survey conducted by the Massachusetts Institute of
use of milestones acknowledges that a patient may progress Technology’s FoCUS team found that use of a specific fi-
to some but not all of the phases of care, whether because of nancing mechanism or payment mechanism in CAR T-cell
product manufacturing failure, progression, or death. MBCs therapy, other cellular immunotherapies, and gene thera-
allow for payment to the provider for those services already pies will continue to be driven by therapy-specific factors
provided while retaining the rest of the funds if the full and variations in payer type32 (i.e., commercial fully insured
treatment episode is not completed. The use of MBCs also lives vs. government payer) as well as by total therapy price
allows for the possibility that a patient may transition in- and provider acquisition models. Providers and payers will
surance coverage and payer responsibility during the need to be flexible in their use of models as therapies and
treatment episode. Alternate definitions of MBCs include their corresponding care models mature.31
payers using them to structure a rebate agreement with CONCLUSION
a manufacturer based on the achievement of an expected Approval of Axi-cel and Tisa-cel heralds the beginning of
clinical milestone (i.e., the lack of such a milestone would a new era in immunotherapy and paves the way for further
trigger a rebate). For purposes of this paper, MBCs will be development and commercialization of future cellular
used for the first definition (that of not paying until a set of therapies. As described in this article, there are several
services is complete) versus the use of an outcomes-based challenges to broader but appropriate use of these thera-
rebate mechanism or OBA.30 pies. With essential collaborations between professional
Shortly after Novartis received FDA approval for the use of societies, quality accreditation organizations, outcome
Tisa-cel in children and young adolescents with B-cell ALL, registries, and regulatory agencies, we can optimize their
it announced an optional OBA program for use by certified use and improve access.


1
Department of Medicine, University of Texas Southwestern Medical Navneet Majhail, MD, MS, 9500 Euclid Ave., CA60, Cleveland, OH
Center, Dallas, TX 44195; Twitter: @BldCancerDoc; email: [email protected].
2
Center for Clinical Value, Blue Cross Blue Shield Association, Chicago, IL
3
Blood and Marrow Transplant Program, Cleveland Clinic, Cleveland, OH
EDBK_279151.
REFERENCES
1. Kansagra AJ, Frey NV, Bar M, et al. Clinical utilization of chimeric antigen receptor T-cells (CAR-T) in B-cell acute lymphoblastic leukemia (ALL): an expert opinion
from the European Society for Blood and Marrow Transplantation (EBMT) and the American Society for Blood and Marrow Transplantation (ASBMT). Bone
Marrow Transplant. 2019;54:1868-1880.
2. Jain T, Bar M, Kansagra AJ, et al. Use of chimeric antigen receptor T cell therapy in clinical practice for relapsed/refractory aggressive B cell non-Hodgkin
lymphoma: an expert panel opinion from the American Society for Transplantation and Cellular Therapy. Biol Blood Marrow Transplant. 2019;25:2305-2321.
3. Hernandez I, Prasad V, Gellad WF. Total costs of chimeric antigen receptor T-cell immunotherapy. JAMA Oncol. 2018;4:994-996.
4. Maude SL, Laetsch TW, Buechner J, et al. Tisagenlecleucel in children and young adults with B-cell lymphoblastic leukemia. N Engl J Med. 2018;378:439-448.
5. Stephan G, Maude SL, Rives S, et al. Updated analysis of the efficacy and safety of tisagenlecleucel in pediatric and young adult patients with relapsed/refractory
(r/r) acute lymphoblastic leukemia. Paper presented at: 60th American Society of Hematology Annual Meeting and Exposition; December 2018; San Diego, CA.
6. Stephan G, Hu Z-H, Zhang Y, et al. Tisagenlecleucel chimeric antigen receptor (CAR) T-cell therapy for relapsed/refractory children and young adults with acute
lymphoblastic leukemia (ALL): real world experience from the Center for International Blood and Marrow Transplant Research (CIBMTR) and Cellular Therapy
(CT) Registry. Paper presented at: 61st American Society of Hematology Annual Meeting and Exposition; December 2019; Orlando, FL.
7. Schuster SJ, Bishop MR, Tam CS, et al; JULIET Investigators. Tisagenlecleucel in adult relapsed or refractory diffuse large B-cell lymphoma. N Engl J Med. 2019;
380:45-56.
8. Jaglowski S, Hu Z, Zhang Y, et al. Tisagenlecleucel chimeric antigen receptor (CAR) T-cell therapy for adults with diffuse large B-cell lymphoma: real world
experience from the Center for International Blood & Marrow Transplant Research Cellular Therapy registry. Blood. 2019;134 (suppl 1):766.
9. Neelapu SS, Locke FL, Bartlett NL, et al. Axicabtagene ciloleucel CAR T-cell therapy in refractory large B-cell lymphoma. N Engl J Med. 2017;377:2531-2544.
10. Pasquini M, Locke FL, Herrera AF, et al. Post-marketing use outcomes of an anti-CD19 chimeric antigen receptor T cell therapy, axicabtagene ciloleucel, for the
treatment of large B cell lymphoma in the United States. Blood. 2019;134 (suppl 1):764.
11. Nastoupil LJ, Jain MD, Spiegel JY, et al. Axicabtagene ciloleucel (Axi-cel) CD19 chimeric antigen receptor (CAR) T-cell therapy for relapsed/refractory large B-cell
lymphoma: real world experience. Blood. 2018;132 (suppl 1):627.
12. Jacobson CA, Hunter B, Armand P, et al. Axicabtagene ciloleucel in the real world: outcomes and predictors of response, resistance, and toxicity. Blood. 2018;
132(suppl 1):92.
13. Kansagra AJ, Frey NV, Bar M, et al. Clinical utilization of chimeric antigen receptor T cells in B bell acute lymphoblastic leukemia: An expert opinion from the
European Society for Blood and Marrow Transplantation and the American Society for Blood and Marrow Transplantation. Biol Blood Marrow Transplant. 2019;
25:e76-e85.
14. Yakoub-Agha I, Chabannon C, Bader P, et al. Management of adults and children undergoing chimeric antigen receptor T-cell therapy: best practice rec-
ommendations of the European Society for Blood and Marrow Transplantation (EBMT) and the Joint Accreditation Committee of ISCT and EBMT (JACIE).
Haematologica. 2020;105:297-316.
15. Paulson K, Brazauskas R, Khera N, et al. Inferior access to allogeneic transplant in disadvantaged populations: a center for international blood and marrow
transplant research analysis. Biol Blood Marrow Transplant. 2019;25:2086-2090.
16. Majhail NS, Omondi NA, Denzen E, et al. Access to hematopoietic cell transplantation in the United States. Biol Blood Marrow Transplant. 2010;16:1070-1075.
17. Hong S, Rybicki LA, Corrigan D, et al. Community risk score for evaluating health care disparities in hematopoietic cell transplantation. Biol Blood Marrow
Transplant. 2018;24:877-879.
18. Majhail NS, Nayyar S, Santibañez ME, et al. Racial disparities in hematopoietic cell transplantation in the United States. Bone Marrow Transplant. 2012;
47:1385-1390.
19. Zhu F, Shah N, Xu H, et al. Closed-system manufacturing of CD19 and dual-targeted CD20/19 chimeric antigen receptor T cells using the CliniMACS Prodigy
device at an academic medical center. Cytotherapy. 2018;20:394-406.
20. Hudecek M, Ivics Z. Non-viral therapeutic cell engineering with the Sleeping Beauty transposon system. Curr Opin Genet Dev. 2018;52:100-108.
21. Majzner RG, Mackall CL. Clinical lessons learned from the first leg of the CAR T cell journey. Nat Med. 2019;25:1341-1355.
22. Li J, Li W, Huang K, et al. Chimeric antigen receptor T cell (CAR-T) immunotherapy for solid tumors: lessons learned and strategies for moving forward. J Hematol
Oncol. 2018;11:22.

23. Liu E, Marin D, Banerjee P, et al. Use of CAR-transduced natural killer cells in CD19-positive lymphoid tumors. N Engl J Med. 2020;382:545-553.
24. Yáñez L, Sánchez-Escamilla M, Perales MA. CAR T cell toxicity: current management and future directions. HemaSphere. 2019;3:e186.
25. Locke FL, Neelapu SS, Bartlett NL, et al. Preliminary results of prophylactic tocilizumab after axicabtageneciloleucel (axi-cel; KTE-C19) treatment for patients with
refractory, aggressive non-Hodgkin lymphoma (NHL). Paper presented at: the 59th American Society of Hematology Annual Meeting and Exposition; December
2017; Atlanta, GA.
26. Boyiadzis MM, Dhodapkar MV, Brentjens RJ, et al. Chimeric antigen receptor (CAR) T therapies for the treatment of hematologic malignancies: clinical
perspective and significance. J Immunother Cancer. 2018;6:137.
27. Faucher C, Le Corroller Soriano AG, Esterni B, et al. Randomized study of early hospital discharge following autologous blood SCT: medical outcomes and hospital
costs. Bone Marrow Transplant. 2012;47:549-555.
28. Preussler JM, Mau LW, Majhail NS, et al. Caregiver availability and patient access to hematopoietic cell transplantation: social worker perspectives inform
practice. Support Care Cancer. 2019;27:4253-4264.
29. Delamater PL, Uberti JP. Geographic access to hematopoietic cell transplantation services in the United States. Bone Marrow Transplant. 2016;51:241-248.
30. Novartis. Kymriah: highlights of the outcome-based agreement and list of participating treatment centers. https://www.hcp.novartis.com/products/kymriah-oba/
outcome-based-agreement/. Accessed: February 17, 2020.
31. MIT-CBI. Tools for implementation of precision financing solutions within Medicaid plans. https://newdigs-dev.mit.edu/milestone-based-contracts-with-
medicaid-plans/. Accessed February 17, 2020.
32. MIT-CBI. Payer perspectives on financing and reimbursement of one-time high-cost durable treatments. https://newdigs.mit.edu/sites/default/files/MIT%
20FoCUS%20Payer%20Perspectives%202019F210v044.pdf. Accessed February 17, 2020.

CENTRAL NERVOUS SYSTEM
TUMORS

Gray Areas in the Gray Matter: IDH1/2 Mutations

in Glioma
Martin J. van den Bent, MD1; Ingo K. Mellinghoff, MD2,3; and Ranjit S. Bindra, MD, PhD4,5
Since the first discovery of isocitrate dehydrogenase (IDH ) mutations in cancer, considerable progress has
overview
been made in our understanding of their contribution to cancer development. For glioma, this has helped to
identify two diagnostic groups of tumors (oligodendroglioma and astrocytoma IDHmt) with distinct clinical
characteristics and that are now diagnosed by the presence of the IDH mutations. The metabolic changes
occurring as the consequence of the altered substrate affinity of the mutant IDH protein results in a cascade of
intracellular changes, also inducing a relative sensitivity to chemotherapy and radiotherapy compared with
IDHwt tumors. Pharmacologic blockade of the mutant enzyme with first-in-class inhibitors has been effi-
cacious for the treatment of IDH-mutant acute myeloid leukemia (AML) and is currently being evaluated in
phase III trials for IDH-mutant glioma (INDIGO) and cholangiocarcinoma (ClarIDHy). It seems likely that
acquired resistance to mutant IDH inhibitors will eventually emerge, and combination therapies to augment
the antitumor activity of mutant IDH inhibitors have already been initiated. Approaches to exploit, rather than
inhibit, the unique metabolism of IDH-mutant cancer cells have emerged from laboratory studies and are now
also being tested in the clinic. Results of these clinical trials are eagerly awaited and will likely provide new key
insights and direction of the treatment of IDH-mutant human cancer.
OVERVIEW OF GLIOMA-ASSOCIATED IDH1/2 gliomas had a better outcome compared with histo-
MUTATIONS (MARTIN J. VAN DEN BENT) logically similar tumors without IDH1/2 mutations.6,7
In 2008, a genome-wide sequencing study observed Further studies showed that a classification of glial
unknown mutations in the gene encoding isocitrate tumors based on their molecular profile resulted in
dehydrogenase (IDH1) in 18 (12%) of 149 glioblas- a much stronger prognostication compared with classic
toma samples.1 Remarkably, these mutations occurred histology.8-10 This resulted in the revised 2016 World
Health Organization (WHO) classification of brain
in young patients with glioblastomas that had pro-
tumors in which diffuse glioma are now classified
gressed from low-grade gliomas to what was known as
according to their IDH1/2 mutational and 1p/19q
secondary glioblastoma. Moreover, patients with glio-
status.11 In the aftermath of that fundamental change,
blastoma with IDH1 mutations had a much better
ongoing discussions (cIMPACT-NOW) have further
survival compared with patients with glioblastomas
moved the classification by renaming low-grade as-
without IDH1 mutations. Shortly afterward, a series of
trocytoma with mutations observed in classic glio-
studies showed that these mutations occurred in up to
blastoma (gain of 7 combined with loss of 10, and/or
80% of low-grade gliomas, and, in some IDH1wt glial EGFR amplification, and/or TERT promoter mutations
tumors, mutations in the IDH2 gene were found that only) by “low-grade astrocytoma with molecular fea-
carried a similar prognostic significance.2-4 Early clinical tures of glioblastoma.”12 These tumors indeed carry
Author affiliations studies suggested the occurrence of IDH mutations a prognosis that is as dismal as glioblastoma, and the
and support was an early event in gliomagenesis, occurring before question is whether they should not be labeled as
information (if the development of a 1p/19q codeletion.5 With more such.13 A further cIMPACT report on diffuse glioma
applicable) appear clinical data emerging, it became clear that virtually
at the end of this
entity proposes to revise the name of “glioblastoma,
article.
all 1p/19q codeleted tumors have an IDH1/2 muta- IDHmt” to “astrocytoma grade 4, IDHmt,” and adds
Accepted on
tion and that nearly all these tumors have MGMT additional molecular criteria to that grade 4 diag-
February 26, 2020 promoter methylation. Other mutations were found nosis (i.e., the presence of homozygous deletion of
and published at associated with IDH1/2 mutations, in particular, mu- CDKN2A).14-16 The rationale for this modification is the
ascopubs.org on tations in the TP53 and ATRX genes in tumors without similarity at the molecular level between these grade 4
March 19, 2020:
1p/19q codeletion and mutations in the TERT pro- tumors and IDH1/2-mutated grade 2 and 3 astrocy-
10.1200/EDBK_ moter region in IDH1/2-mutated tumors with 1p/19q toma (e.g., MGMT promoter methylation, sensitivity
280967 codeletion. Clinical studies showed that IDH1/2mt to radiotherapy) and the overall better outcome in this

IDH Mutations in Glioma
is to catalyze the oxidative decarboxylation of isocitrate into

α-ketoglutarate. During that enzymatic reaction, nicotin-
amide adenine dinucleotide phosphate (NADP+) functions
• Grade 4 tumors with IDH mutations will now be for IDH1 and IDH2 as the electron acceptor. The IDH1 gene
labeled astrocytoma, IDHmt grade 4.
is located in the cytoplasm and peroxisomes and IDH2 in
• IDH-mutated tumors share a metabolic back- the mitochondria. The mutant IDH protein has an altered
ground: the accumulation of 2HG and a re- substrate affinity of the enzyme and catalyzes the conver-
duction of α-ketoglutarate.
sion of α-ketoglutarate into 2-hydrogylutarate (2HG), during
• 2HG is an oncometabolite and is responsible for which process NADP+ is reduced to nicotinamide adenine
many of the early oncogenic events in astro- dinucleotide phosphate hydrogen. Because of this, the
cytoma and oligodendroglioma.
intracellular concentration of 2HG increases and the in-
• Drugs inhibiting the mutant IDH protein are tracellular concentration of α-ketoglutarate decreases.
effective in AML and cholangiocarcinoma and Subsequent studies found that 2HG acts as an oncome-
are currently being tested in glioma.
tabolite by altering many cellular functions resulting in
• Other metabolic alterations in IDH mutations genome-wide CpG island hypermethylation, increased
present other possible avenues for treatment, double-stand DNA breaks, decrease in NADP, and loss of
such as PARP inhibitors.
function because of depletion of α-ketoglutarate–dependent
enzymes.30-36 From these data, two opposite treatment
strategies have been developed: reduce the amount of
group. Other molecular indicators of poor prognosis have
intratumoral 2HG by directly inhibiting the function of
been proposed, but larger validation cohorts are needed
mutant IDH enzyme, or do not reduce the amount of 2HG
to clarify the role of CDK4 amplification, increased copy number
alterations, and absence of CpG island hypermethylation.14,17,18 and instead exploit the the cellular consequences of 2HG
Even in the molecular era, classic clinical features beyond accumulation, in particular, the impaired DNA repair
histology still remain of prognostic relevance. Age of the patient, mechanisms that may underly the increased sensitivity
tumor size, and the presence of contrast enhancement have of IDH-mutant gliomas to radiotherapy and alkylating
been identified as indicators of worse outcome.19-21 chemotherapy.
For treatment-related factors, residual tumor after surgery

impacts outcome, in particular, grade 2 and 3 IDHmt INHIBITING THE FUNCTION OF THE MUTANT IDH1/2 ENZYME
glioma, with much better survival in completely resected (INGO K. MELLINGHOFF)
patients.22,23 All studies on adding chemotherapy after ra- The discovery of somatic IDH mutations in glioblastoma and
diotherapy have shown improved outcome in patients with lower-grade gliomas jettisoned the mutant metabolic en-
oligodendroglioma or astrocytoma IDHmt. The first reports zyme to the top of the list of new drug targets for diffuse
were inconsistent to the question of whether it was the 1p/ glioma.1-3 Enthusiasm for direct inhibition of the mutant
19q status that was associated with improved outcome after enzyme was based on several observations. First, cancer-
adjuvant chemotherapy in grade 2 and 3 glioma, the IDH associated IDH mutations cluster in key arginine residues
mutations that are invariably present in 1p/19q codeleted within the enzyme’s active site (R132 of IDH1 and R140 or
tumors, or MGMT promoter methylation that is usually R172 of IDH2), raising the possibility of developing mutant-
present in IDHmt tumors.7,24,25 More recently, the benefit selective inhibitors with a wider therapeutic window. Sec-
to chemotherapy in IDHmt astrocytoma lacking 1p/19q ond, mutant IDH1 (mIDH1) appeared to play a prominent
codeletion was confirmed in the CATNON study, which was role in the pathogenesis of lower-grade glioma given the
a study on anaplastic astrocytoma without 1p/19q code- extraordinarily high prevalence of IDH mutation in this
letion.26 A similar benefit in grade 2 astrocytoma IDHmt was disease (70%–80%), the distinct DNA hypermethylation
observed in a post hoc study on a subset of patients from the pattern associated with IDH mutations,31,34 and the per-
pivotal PCV (procarbazine, lomustine, vincristine) in low- sistence of IDH-mutant tumor cell clones throughout the
grade glioma trial.27,28 Tumor grade (grade 2 vs. 3) has some disease course.37,38 Third, 2HG, the direct product of the
impact but less so than in the pre-WHO 2016 era.14,29 mutant enzyme, was sufficient to phenocopy the cancer-
The early observation on the strong association between promoting effects of the mutant IDH enzyme in a reversible
outcome and IDH mutations initiated further studies into the fashion,39,40 providing rationale for maximal 2HG inhibition
different clinical characteristics of these tumors. Functional as biologically relevant and readily quantifiable pharma-
studies resulted in the understanding that IDH-mutated cologic strategy. Last, inhibition of the mutant IDH enzyme
tumors represent a completely different entity on the showed antitumor activity in experimental models of glioma
molecular level. The function of all isocitratedehydrogenases and leukemia.41,42

van den Bent, Mellinghoff, and Binja
Nonetheless, the paucity of experimental models that faith- Vorasidenib (AG-881) is a first-in-class dual inhibitor of
fully recapitulate the genetic alterations, tumor microenvi- mutant IDH1 and 2, which was designed for enhanced brain
ronment, and growth patterns of IDH-mutant low-grade penetrance. Vorasidenib was well tolerated and showed
glioma has made it challenging to establish the role of preliminary evidence of antitumor activity in a now completed
mutant IDH in established tumors. In vitro studies using phase I trial in patients with nonenhancing glioma (Clin-
IDH1-mutant cancer cell lines, which often harbor geneti- icalTrials.gov identifier: NCT02481154).53 In the above-
cally more complex genomes, suggest that several factors mentioned perioperative trial (ClinicalTrials.gov identifier:
might relieve mutant IDH cancer cells from their de- NCT03343197), vorasidenib resulted in consistent and
pendency on the mutant enzyme for survival.43 dose-dependent 2HG suppression with greater than 90%
reduction at 50 mg daily compared with untreated controls.
The clinical development of inhibitors of mIDH proceeded
Preliminary efficacy data show objective tumor responses
most expeditiously in patients with AML, another cancer
(approximately 30%) and durable disease control with
type found to harbor mutations in IDH1 and, more com-
postoperative treatment. Clinical responses were also re-
monly, IDH2.44,45 Enasidenib, the first-in-class inhibitor of
ported, with the mutant IDH1 inhibitor DS-1001b responses
mIDH2, produced clinical responses in approximately 40%
in a phase I study in Japan.54
of patients with advanced mIDH2 AML.46,47 Ivosidenib, the
first-in-class inhibitor of the mutant IDH1 enzyme, similarly The collective data from the early clinical trials with mIDH
induced clinical and molecular remissions in patients with inhibitors in glioma suggest that lower-grade gliomas that
advanced mIDH1 AML.48 Both drugs were recently ap- do not enhance on MRI after contrast injection (so-called
proved in the United States for the treatment of mIDH AML. nonenhancing) may be susceptible to the antitumor activity
of mIDH inhibitors. Because contrast enhancement in low-
The role of mutant IDH in glioma and other solid tumors is grade gliomas is often associated with transformation to
currently under investigation. A phase I study with ivosi- a higher tumor grade and acquisition of additional genetic
denib in subjects with advanced solid tumors, including alterations, it seems plausible that the role of mIDH for
glioma, with an IDH1 mutation (ClinicalTrials.gov identifier: tumor maintenance is greatest at the earliest disease stage
NCT02073994) was initiated in March 2014 across 12 and before additional radiation and chemotherapy.55 This
study sites in the United States and one in France and finding is reminiscent of the enhanced responsiveness of
included 66 with glioma. Twelve of 66 (18.2%) patients had early-stage chronic myelogenous leukemia to targeted BCR-
glioblastoma; the remainder had lower-grade glioma. The ABL kinase inhibition.56 This conclusion led to the design of
median number of prior systemic therapies was two (range, a randomized, placebo-controlled phase III trial of AG-881
one to six), and 49 of 66 patients had received prior ra- in participants with residual or recurrent grade 2 glioma with
diotherapy. No dose-limiting toxicities were reported, and an IDH1 or IDH2 mutation (INDIGO; ClinicalTrials.gov
the maximum tolerated dose was not reached. A dose of identifier: NCT04164901). This trial is currently open to
500 mg once daily was selected for expansion based on the accrual and will enroll patients who have not yet received
pharmacokinetic/pharmacodynamic data from all solid tu- radiation or chemotherapy and are under radiographic
mor cohorts and preliminary clinical activity observed in the surveillance (watch and wait) following their initial surgery.
dose-escalation phase. As of the data cutoff, patients with
nonenhancing gliomas had a median treatment duration of EXPLOITING ONCOMETABOLITE-INDUCED DEFECTS IN
18.4 months (range, 1.4–47.2 months), and 15 patients IDH1/2-MUTANT GLIOMAS (RANJIT S. BINDRA)
(22.7%) remained on treatment. An exploratory analy- As mentioned previously, there are more than 70 α-
sis showed a reduction in tumor volume growth rates ketoglutarate (αKG)-dependent dioxygenases in the cell,
(i.e., compared with pretreatment growth rates) and tumor which perform a diverse range of functions including epi-
shrinkage in several patients.49 Interestingly, tumor shrink- genetic regulation, DNA repair, and metabolism. Mutant
age was primarily noted in tumors that did not enhance on IDH1/2-induced 2HG typically exceeds a concentration of 1
MRI after contrast injection, despite the fact that these to 2 mM in cells, and thus it is likely that nearly all of these
tumors presumably have an intact blood–brain barrier. A dioxygenases are inhibited, albeit to varying degrees, by
subsequent clinical trial (ClinicalTrials.gov identifier: such high levels of oncometabolite production.57 Inhibition
NCT03343197) showed greater than 90% inhibition of of even a single αKG-dependent dioxygenase, or subset of
the mutant IDH enzyme by ivosidenib in on-treatment these proteins, can induce profound changes in the cell. For
biopsies from patients with nonenhancing gliomas.50 example, the TET family of 5 methylcytosine hydroxylases
Both studies have completed enrollment. Similar to the are αKG-dependent dioxygenases and thus are inhibited by
experience in glioma, ivosidenib was well tolerated and 2HG.58 Loss of these proteins promotes a CpG island
showed antitumor activity in patients with IDH1-mutant methylator phenotype, which leads to silencing of numerous
advanced cholangiocarcinoma.51,52 loci in the genome.31,59 Based on these findings, an alternative

therapeutic strategy has been proposed to exploit vulner- glioma with IDH1/2 mutations (ClinicalTrials.gov identifier:
abilities associated with oncometabolite production rather NCT03581292).
than suppressing mutant IDH1/2 neomorphic enzymatic
activity. This type of approach overlaps with the concept of Alkylating Agents
synthetic lethality, in which targeting one of two related It has also been shown that mutant IDH1/2-induced 2HG
pathways does not affect viability, but targeting both impairs alkylation damage repair via inhibition of the ALKBH
pathways simultaneously leads to cell death.60 Synthetic family of αKG-dependent dioxygenases, ALKBH2 and
lethality gained momentum after the publication of two ALKBH3, leading to enhanced sensitivity to multiple
seminal papers in the early 2000s, which reported that alkylating agents, including temozolomide.71,72 It should
mutations in the homologous recombination (HR) genes, be noted that IDH1/2 mutations frequently co-occur with
BRCA1 and BRCA2, confer exquisite sensitivity to PARP MGMT promoter silencing, which is linked to the phe-
inhibitors.61,62 As multiple PARP inhibitors are now U.S. nomenon of 2HG-induced CpG island methylator pheno-
Food and Drug Administration approved for the treatment of type via TET inhibition discussed previously. MGMT loss
BRCA1/2-mutant cancers, the concept of synthetic lethal also confers alkylator sensitivity, but this pathway is distinct
tumor targeting is now well accepted as a potential effica- from ALKBH, and thus loss of both pathways in IDH1/2-
cious strategy in oncology. In the following, we discuss four mutant gliomas likely results in sensitivity to this class of
therapeutic classes of drugs that have been proposed for drugs, which is at least additive. Collectively, these findings
use to exploit oncometabolite-induced vulnerabilities as- reveal two pathways via which oncometabolites induce
sociated with IDH1/2-mutant gliomas. alkylator sensitivity: (1) direct inhibition of ALKBH2/3 by
PARP Inhibitors 2HG and (2) MGMT silencing via 2HG-induced TET in-
hibition and CpG island methylator phenotype. This also
In 2017, the Bindra and Glazer Laboratories were the first to
suggests that mutant IDH1/2 inhibitors should not be given
report that mutant IDH1/2 induces an HR defect that
concurrently with alkylators, given the potential for antag-
confers PARP inhibitor sensitivity.33 Mechanistically, it
onistic interactions, at least based on the effects of 2HG on
was found that 2HG-induced suppression of two histone
ALKBH2/3 function.
demethylases, KDM4A and KDM4B, induced an HR
defect similar to that associated with BRCA1/2 mutations.
BCL2 Family Inhibitors
Soon after this initial discovery, several other groups re-
ported a similar synthetic lethal interaction between PARP A synthetic lethal interaction between mutant IDH1/2 and
inhibition and mutant IDH1/2 across a wide range of tumor members of the BCL-2 family was first reported in AML
types.63-65 Of note, the phenomenon of oncometabolite- tumor models3 but has since been shown in IDH1/2-mutant
induced HR suppression was further extended to tumor- glioma.73,74 BCL2 inhibitors have been extensively tested in
associated mutations in other tricarboxylic acid cycle the clinic, and one drug is now U.S. Food and Drug Ad-
genes, such as fumarate and succinate.66 This novel ap- ministration approved for chronic lymphocytic leukemia
proach to target IDH1/2-mutant cancers with PARP inhibitors (venetoclax).75 As such, this strategy can be readily and
is now being tested in the clinic for glioma (ClinicalTrials. feasibly tested in the clinic for IDH1/2-mutant gliomas in the
gov identifiers: NCT03749187 and NCT03914742), as well future.
as many other tumor types, including chondrosarcoma and
cholangiocarcinoma (ClinicalTrials.gov identifiers: NCT02576444 NAMPT Inhibitors
and NCT03212274). These trials collectively are testing Tateishi and colleagues also reported that IDH1/2-mutant
the efficacy of two PARP inhibitors: olaparib and BGB290. gliomas are sensitive to a class of drugs that target nico-
Both of these drugs are notable for high potency, which tinamide adenine dinucleotide (NAD) metabolism, called
likely can be attributed to their action as PARP-trapping nicotinamide phosphoribosyltransferase (NAMPT) in-
PARP inhibitors.67 Relevant to glioma, BGB290 appears to hibitors. 43 Mechanistically, they demonstrated that IDH1/2
demonstrate relatively high levels of blood–brain barrier mutations silence a key gene in NAD metabolism, nicotinate
penetration in animal models.68 Olaparib appears to phosphoribosyltransferase (NAPRT), which leads to de-
penetrate at least T1-postcontrast-enhancing areas of creased levels of NAD+ and consequent NAMPT inhibitor
disease in glioma, based on the results of the OPARATIC sensitivity. In a subsequent study, the authors demonstrated
trial (ClinicalTrials.gov identifier: NCT01390571).69 This that temozolomide and NAMPT inhibitor combinations were
has prompted its evaluation in two additional glioblastoma particularly active in IDH1/2-mutant glioma models.76
trials, PARADIGM-1 and -2, which do not select for pa- Multiple potent NAMPT inhibitors have been developed
tients based on IDH1/2 mutation status.70 Another, albeit and were tested previously in numerous clinical trials, and
less potent, PARP inhibitor, veliparib, is also being tested there is currently one drug actively being tested in patients:
in a cohort of pediatric patients with newly diagnosed KPT-9274 (ClinicalTrials.gov identifier: NCT02702492).77

As such, there is great potential to test these drugs in IDH1/ with intratumoral 2HG accumulation in glioma are reversible
2-mutant glioma in future clinical trials. following inhibition of the mutant enzyme. Approaches to
CONCLUSIONS AND FUTURE DIRECTIONS exploit, rather than inhibit, the unique metabolism of IDH-
mutant cancer cells have emerged from laboratory studies
Since the first discovery of IDH mutations in cancer, con- and are now also being tested in the clinic. Results of these
siderable progress has been made in our understanding of clinical trials are eagerly awaited and will likely provide key
their contribution to cancer development. For glioma, this new insights and direction on the treatment of IDH-mutant
has helped to identify two diagnostic groups of tumors human cancer.
(oligodendroglioma and astrocytoma IDHmt) with distinct
clinical characteristics. Pharmacologic blockade of the mu- ACKNOWLEDGMENT
tant enzyme with first-in-class inhibitors has been efficacious I.K. Mellinghoff’s research is supported by the National In-
for the treatment of IDH-mutant AML and is currently being stitutes of Health (NIH; Grants 1 R35 NS105109 01 and
evaluated in phase III trials for IDH-mutant glioma (INDIGO) P30CA008748), the National Brain Tumor Society, Cycle of
and cholangiocarcinoma (ClarIDHy). It seems likely that Survival, and the Memorial Sloan Kettering Neuro-Oncology
acquired resistance to mutant IDH inhibitors will eventually Research in Translation Program. R. Bindra’s research is
emerge, and combination therapies to augment the antitu- supported by sources including the NIH (Grant R01
mor activity of mutant IDH inhibitors have already been CA215453-03), Oligo Nation, CureSearch, Rising Tide
initiated. This will require a much deeper understanding of Foundation, Gray Foundation, and Leukemia and Lymphoma
which of the many cellular and molecular effects associated Society.

1
Department of Neurology, Brain Tumor Center at Erasmus MC Cancer Martin J. van den Bent, MD, Department of Neuro-Oncology, Brain Tumor
Institute, Rotterdam, Netherlands Center at Erasmus MC Cancer Institute, Dr Molenwaterplein 401,
2
Human Oncology and Pathogenesis Program, Department of Neurology, Rotterdam 3015GD, the Netherlands; email: m.vandenbent@
Memorial Sloan Kettering Cancer Center, New York, NY erasmusmc.nl.
3
Department of Pharmacology, Weill Cornell Medical College, New York, NY
4
Departments of Therapeutic Radiology and Pathology, Yale School of
Medicine, New Haven, CT
5
Brain Tumor Center, Yale Cancer Center, New Haven, CT
EDBK_280967.
REFERENCES
1. Parsons DW, Jones S, Zhang X, et al. An integrated genomic analysis of human glioblastoma multiforme. Science. 2008;321:1807-1812.
2. Balss J, Meyer J, Mueller W, et al. Analysis of the IDH1 codon 132 mutation in brain tumors. Acta Neuropathol. 2008;116:597-602.
3. Yan H, Parsons DW, Jin G, et al. IDH1 and IDH2 mutations in gliomas. N Engl J Med. 2009;360:765-773.
4. Hartmann C, Meyer J, Balss J, et al. Type and frequency of IDH1 and IDH2 mutations are related to astrocytic and oligodendroglial differentiation and age: a study
of 1,010 diffuse gliomas. Acta Neuropathol. 2009;118:469-474.
5. Watanabe T, Nobusawa S, Kleihues P, et al. IDH1 mutations are early events in the development of astrocytomas and oligodendrogliomas. Am J Pathol. 2009;
174:1149-1153.
6. Sanson M, Marie Y, Paris S, et al. Isocitrate dehydrogenase 1 codon 132 mutation is an important prognostic biomarker in gliomas. J Clin Oncol. 2009;
27:4150-4154.
7. van den Bent MJ, Dubbink HJ, Marie Y, et al. IDH1 and IDH2 mutations are prognostic but not predictive for outcome in anaplastic oligodendroglial tumors:
a report of the European Organization for Research and Treatment of Cancer Brain Tumor Group. Clin Cancer Res. 2010;16:1597-1604.
8. Dubbink HJ, Atmodimedjo PN, Kros JM, et al. Molecular classification of anaplastic oligodendroglioma using next-generation sequencing: a report of the
prospective randomized EORTC Brain Tumor Group 26951 phase III trial. Neuro-oncol. 2016;18:388-400.
9. Wang XW, Ciccarino P, Rossetto M, et al. IDH mutations: genotype-phenotype correlation and prognostic impact. BioMed Res Int. 2014;2014:540236.
10. Brat DJ, Verhaak RG, Aldape KD, et al; Cancer Genome Atlas Research Network. Comprehensive, integrative genomic analysis of diffuse lower-grade gliomas.
N Engl J Med. 2015;372:2481-2498.
11. Louis DN, Ohgaki H, Wiestler OD, et al. WHO Classification of Tumours of the Central Nervous System, 4th edition. Lyon, France; World Health Organization;
2016.

12. Brat DJ, Aldape K, Colman H, et al. cIMPACT-NOW update 3: recommended diagnostic criteria for “Diffuse astrocytic glioma, IDH-wildtype, with molecular
features of glioblastoma, WHO grade IV. Acta Neuropathol. 2018;136:805-810.
13. Tesileanu CMS, Dirven L, Wijnenga MMJ, et al. Survival of diffuse astrocytic glioma, IDH1/2-wildtype, with molecular features of glioblastoma, WHO grade IV:
a confirmation of the cIMPACT-NOW criteria. Neuro-oncol. Epub 2019 Oct 22.
14. Shirahata M, Ono T, Stichel D, et al. Novel, improved grading system(s) for IDH-mutant astrocytic gliomas. Acta Neuropathol. 2018;136:153-166.
15. Appay R, Dehais C, Maurage CA, et al; POLA Network. CDKN2A homozygous deletion is a strong adverse prognosis factor in diffuse malignant IDH-mutant
gliomas. Neuro-oncol. 2019;21 (Supplement_3):1519-1528.
16. Reis GF, Pekmezci M, Hansen HM, et al. CDKN2A loss is associated with shortened overall survival in lower-grade (World Health Organization Grades II-III)
astrocytomas. J Neuropathol Exp Neurol. 2015;74:442-452.
17. Ceccarelli M, Barthel FP, Malta TM, et al; TCGA Research Network. Molecular profiling reveals biologically discrete subsets and pathways of progression in diffuse
glioma. Cell. 2016;164:550-563.
18. Aoki K, Nakamura H, Suzuki H, et al. Prognostic relevance of genetic alterations in diffuse lower-grade gliomas. Neuro-oncol. 2018;20:66-77.
19. Suchorska B, Schüller U, Biczok A, et al. Contrast enhancement is a prognostic factor in IDH1/2 mutant, but not in wild-type WHO grade II/III glioma as confirmed
by machine learning. Eur J Cancer. 2019;107:15-27.
20. Halani SH, Yousefi S, Velazquez Vega J, et al. Multi-faceted computational assessment of risk and progression in oligodendroglioma implicates NOTCH and PI3K
pathways. NPJ Precis Oncol. 2018;2:24.
21. Tom MC, Varra V, Leyrer CM, et al. Risk factors for progression among low-grade gliomas after gross total resection and initial observation in the molecular era. Int
22. Wijnenga MMJ, French PJ, Dubbink HJ, et al. The impact of surgery in molecularly defined low-grade glioma: an integrated clinical, radiological, and molecular
analysis. Neuro-oncol. 2018;20:103-112.
23. Aibaidula A, Chan AK, Shi Z, et al. Adult IDH wild-type lower-grade gliomas should be further stratified. Neuro-oncol. 2017;19:1327-1337.
24. Cairncross JG, Wang M, Jenkins RB, et al. Benefit from procarbazine, lomustine, and vincristine in oligodendroglial tumors is associated with mutation of IDH.
J Clin Oncol. 2014;32:783-790.
25. van den Bent MJ, Erdem-Eraslan L, Idbaih A, et al. MGMT-STP27 methylation status as predictive marker for response to PCV in anaplastic oligodendrogliomas
and oligoastrocytomas. A report from EORTC study 26951. Clin Cancer Res. 2013;19:5513-5522.
26. Van Den Bent MJ, Erridge S, Vogelbaum MA, et al. Second interim and first molecular analysis of the EORTC randomized phase III intergroup CATNON trial on
concurrent and adjuvant temozolomide in anaplastic glioma without 1p/19q codeletion. J Clin Oncol. 2019;37 (suppl; abstr 2000).
27. Buckner JC, Shaw EG, Pugh SL, et al. Radiation plus Procarbazine, CCNU, and Vincristine in Low-Grade Glioma. N Engl J Med. 2016;374:1344-1355.
28. Bell EH, Won M, Fleming JL, et al. Updated predictive analysis of the WHO-defined molecular subgroups of low-grade gliomas within the high-risk treatment arms
of NRG Oncology/RTOG 9802. J Clin Oncol. 2019;37 (suppl; abstr 2002).
29. Reuss DE, Mamatjan Y, Schrimpf D, et al. IDH mutant diffuse and anaplastic astrocytomas have similar age at presentation and little difference in survival:
a grading problem for WHO. Acta Neuropathol. 2015;129:867-873.
30. Lai A, Kharbanda S, Pope WB, et al. Evidence for sequenced molecular evolution of IDH1 mutant glioblastoma from a distinct cell of origin. J Clin Oncol. 2011;
29:4482-4490.
31. Turcan S, Rohle D, Goenka A, et al. IDH1 mutation is sufficient to establish the glioma hypermethylator phenotype. Nature. 2012;483:479-483.
32. Johannessen TC, Prestegarden L, Grudic A, et al. The DNA repair protein ALKBH2 mediates temozolomide resistance in human glioblastoma cells. Neuro-oncol.
2013;15:269-278.
33. Sulkowski PL, Corso CD, Robinson ND, et al. 2-Hydroxyglutarate produced by neomorphic IDH mutations suppresses homologous recombination and induces
PARP inhibitor sensitivity. Sci Transl Med. 2017;9.
34. Noushmehr H, Weisenberger DJ, Diefes K, et al; Cancer Genome Atlas Research Network. Identification of a CpG island methylator phenotype that defines
a distinct subgroup of glioma. Cancer Cell. 2010;17:510-522.
35. Molenaar RJ, Botman D, Smits MA, et al. Radioprotection of IDH1-mutated cancer cells by the IDH1-mutant inhibitor AGI-5198. Cancer Res. 2015;
75:4790-4802.
36. Clark O, Yen K, Mellinghoff IK. Molecular pathways: isocitrate dehydrogenase mutations in cancer. Clin Cancer Res. 2016;22:1837-1842.
37. Barthel FP, Johnson KC, Varn FS, et al; GLASS Consortium. Longitudinal molecular trajectories of diffuse glioma in adults. Nature. 2019;576:112-120.
38. Miller AM, Shah RH, Pentsova EI, et al. Tracking tumour evolution in glioma through liquid biopsies of cerebrospinal fluid. Nature. 2019;565:654-658.
39. Losman JA, Looper RE, Koivunen P, et al. (R)-2-hydroxyglutarate is sufficient to promote leukemogenesis and its effects are reversible. Science. 2013;
339:1621-1625.
40. Ye D, Ma S, Xiong Y, et al. R-2-hydroxyglutarate as the key effector of IDH mutations promoting oncogenesis. Cancer Cell. 2013;23:274-276.
41. Wang F, Travins J, DeLaBarre B, et al. Targeted inhibition of mutant IDH2 in leukemia cells induces cellular differentiation. Science. 2013;340:622-626.
42. Rohle D, Popovici-Muller J, Palaskas N, et al. An inhibitor of mutant IDH1 delays growth and promotes differentiation of glioma cells. Science. 2013;
340:626-630.
43. Tateishi K, Wakimoto H, Iafrate AJ, et al. Extreme vulnerability of IDH1 mutant cancers to NAD+ depletion. Cancer Cell. 2015;28:773-784.

44. Mardis ER, Ding L, Dooling DJ, et al. Recurring mutations found by sequencing an acute myeloid leukemia genome. N Engl J Med. 2009;361:1058-1066.
45. Kosmider O, Gelsi-Boyer V, Slama L, et al. Mutations of IDH1 and IDH2 genes in early and accelerated phases of myelodysplastic syndromes and MDS/
myeloproliferative neoplasms. Leukemia. 2010;24:1094-1096.
46. Yen K, Travins J, Wang F, et al. AG-221, a first-in-class therapy targeting acute myeloid leukemia harboring oncogenic IDH2 mutations. Cancer Discov. 2017;
7:478-493.
47. Amatangelo MD, Quek L, Shih A, et al. Enasidenib induces acute myeloid leukemia cell differentiation to promote clinical response. Blood. 2017;130:732-741.
48. DiNardo CD, Stein EM, de Botton S, et al. Durable remissions with ivosidenib in IDH1-mutated relapsed or refractory AML. N Engl J Med. 2018;378:2386-2398.
49. Mellinghoff IK, Ellingson BM, Touat M, et al. Ivosidenib in IDH1-mutated advanced glioma. J Clin Oncol. In press.
50. Mellinghoff IK, Cloughesy TF, Wen PY, et al. A phase 1, open-label, perioperative study of ivosidenib (AG-120) and vorasidenib (AG-881) in recurrent, IDH1-
mutant, low-grade glioma: updated results. Neuro-oncol. 2019;21 (Suppl 6):vi28-vi29.
51. Lowery MA, Burris HA III, Janku F, et al. Safety and activity of ivosidenib in patients with IDH1-mutant advanced cholangiocarcinoma: a phase 1 study. Lancet
Gastroenterol Hepatol. 2019;4:711-720.
52. Abou-Alfa GK, Macarulla MT, Javle M, et al. LBA10_PR-ClarIDHYP: a global, phase III, randomized, double-blind study of ivosidenib vs placebo in patients with
advanced cholangiocarcinoma with an isocitrate dehydrogenase 1 (IDH1) mutation. Ann Oncol. 2019;30 (suppl 5):v851-v934.
53. Mellinghoff IK, Penas-Prado M, Peters KB, et al. Phase 1 study of AG-881, an inhibitor of mutant IDH1/IDH2, in patients with advanced IDH-mutant solid tumors,
including glioma. J Clin Oncol. 2018;36 (suppl; abstr 2002).
54. Natsume A, Wakabayashi T, Miyakita Y, et al. Phase I study of a brain penetrant mutant IDH1 inhibitor DS-1001b in patients with recurrent or progressive IDH1
mutant gliomas. J Clin Oncol. 2019;37 (suppl; abstr 2004).
55. Jonsson P, Lin AL, Young RJ, et al. Genomic correlates of disease progression and treatment response in prospectively characterized gliomas. Clin Cancer Res.
2019;25:5537-5547.
56. Druker BJ, Talpaz M, Resta DJ, et al. Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med. 2001;
344:1031-1037.
57. Losman JA, Kaelin WG Jr. What a difference a hydroxyl makes: mutant IDH, (R)-2-hydroxyglutarate, and cancer. Genes Dev. 2013;27:836-852.
58. Xu W, Yang H, Liu Y, et al. Oncometabolite 2-hydroxyglutarate is a competitive inhibitor of α-ketoglutarate-dependent dioxygenases. Cancer Cell. 2011;19:17-30.
59. Figueroa ME, Abdel-Wahab O, Lu C, et al. Leukemic IDH1 and IDH2 mutations result in a hypermethylation phenotype, disrupt TET2 function, and impair
hematopoietic differentiation. Cancer Cell. 2010;18:553-567.
60. O’Neil NJ, Bailey ML, Hieter P. Synthetic lethality and cancer. Nat Rev Genet. 2017;18:613-623.
61. Bryant HE, Schultz N, Thomas HD, et al. Specific killing of BRCA2-deficient tumours with inhibitors of poly(ADP-ribose) polymerase [published correction
appears in Nature. 2007 447:346]. Nature. 2005;434:913-917.
63. Lu Y, Kwintkiewicz J, Liu Y, et al. Chemosensitivity of IDH1-mutated gliomas due to an impairment in PARP1-mediated DNA repair. Cancer Res. 2017;
77:1709-1718.
64. Molenaar RJ, Radivoyevitch T, Nagata Y, et al. IDH1/2 mutations sensitize acute myeloid leukemia to PARP inhibition and this is reversed by IDH1/2-mutant
inhibitors. Clin Cancer Res. 2018;24:1705-1715.
65. Philip B, Yu DX, Silvis MR, et al. Mutant IDH1 promotes glioma formation in vivo. Cell Rep. 2018;23:1553-1564.
66. Sulkowski PL, Sundaram RK, Oeck S, et al. Krebs-cycle-deficient hereditary cancer syndromes are defined by defects in homologous-recombination DNA repair.
Nat Genet. 2018;50:1086-1092.
67. Murai J, Zhang Y, Morris J, et al. Rationale for poly(ADP-ribose) polymerase (PARP) inhibitors in combination therapy with camptothecins or temozolomide based
on PARP trapping versus catalytic inhibition. J Pharmacol Exp Ther. 2014;349:408-416.
68. Gupta SK, Carlson BL, Schroeder MA, et al. Inhibition of PARP activity by BGB-290 potentiates efficacy of temozolomide in patient derived xenografts of
glioblastoma multiforme. Cancer Res. 2015;75 (15 suppl):3505.
69. Halford SER, Cruickshank G, Dunn L, et al. Results of the OPARATIC trial: a phase I dose escalation study of olaparib in combination with temozolomide (TMZ) in
patients with relapsed glioblastoma (GBM). J Clin Oncol. 2017;35 (suppl; abstr 2022).
70. Fulton B, Short SC, James A, et al. PARADIGM-2: two parallel phase I studies of olaparib and radiotherapy or olaparib and radiotherapy plus temozolomide in
patients with newly diagnosed glioblastoma, with treatment stratified by MGMT status. Clin Transl Radiat Oncol. 2017;8:12-16.
71. Wang P, Wu J, Ma S, et al. Oncometabolite D-2-hydroxyglutarate inhibits ALKBH DNA repair enzymes and sensitizes IDH mutant cells to alkylating agents. Cell
Rep. 2015;13:2353-2361.
72. Chen F, Bian K, Tang Q, et al. Oncometabolites d- and l-2-hydroxyglutarate inhibit the AlkB family DNA repair enzymes under physiological conditions. Chem Res
Toxicol. 2017;30:1102-1110.
73. Chan SM, Thomas D, Corces-Zimmerman MR, et al. Isocitrate dehydrogenase 1 and 2 mutations induce BCL-2 dependence in acute myeloid leukemia. Nat
Med. 2015;21:178-184.
74. Karpel-Massler G, Ishida CT, Bianchetti E, et al. Induction of synthetic lethality in IDH1-mutated gliomas through inhibition of Bcl-xL. Nat Commun. 2017;8:1067.

75. Lampson BL, Davids MS. The development and current use of BCL-2 inhibitors for the treatment of chronic lymphocytic leukemia. Curr Hematol Malig Rep.
2017;12:11-19.
76. Tateishi K, Higuchi F, Miller JJ, et al. The alkylating chemotherapeutic temozolomide induces metabolic stress in IDH1-mutant cancers and potentiates NAD+
depletion-mediated cytotoxicity. Cancer Res. 2017;77:4102-4115.
77. Yaku K, Okabe K, Hikosaka K, et al. NAD metabolism in cancer therapeutics. Front Oncol. 2018;8:622.

The Practical Application of Emerging

Technologies Influencing the Diagnosis and Care
of Patients With Primary Brain Tumors
Leland S. Hu, MD1; Daniel J. Brat, MD, PhD2; Orin Bloch, MD3; Shakti Ramkissoon, MD, PhD, MMSc4,,6; and Glenn J. Lesser, MD, FACP5
overview
Over the past decade, a variety of new and innovative technologies has led to important advances in the
diagnosis and management of patients with primary malignant brain tumors. New approaches to surgical
navigation and tumor localization, advanced imaging to define tumor biology and treatment response, and the
widespread adoption of a molecularly defined integrated diagnostic paradigm that complements traditional
histopathologic diagnosis continue to impact the day-to-day care of these patients. In the neuro-oncology
clinic, discussions with patients about the role of tumor treating fields (TTFields) and the incorporation of next-
generation sequencing (NGS) data into therapeutic decision-making are now a standard practice. This article
summarizes newer applications of technology influencing the pathologic, neuroimaging, neurosurgical, and
medical management of patients with malignant primary brain tumors.
AN INTEGRATED MOLECULAR APPROACH TO Integrated Diagnosis and the Development of the

CLASSIFICATION AND GRADING OF DIFFUSE GLIOMAS 2016 WHO Classification
Diffuse gliomas are primary central nervous system The molecular era for diffuse glioma classification be-
neoplasms that are characterized by the widespread gan with the identification of mutations in isocitrate
infiltration of brain and spinal cord structures by in- dehydrogenase (IDH) 1 in a small set of glioblastomas
dividual tumor cells displaying cytologic and histologic (GBMs) in 2008.3 Patients with IDH mutant GBMs were
features of glial differentiation. These tumors affect younger, developed tumors that evolved from a lower-
patients of all ages and arise throughout the neuro-axis grade glioma, and had much better prognosis than
but are most common in older adults and occur most patients with IDH wild-type GBMs. A year later, an
frequently in the cerebral hemispheres. From the early article on the IDH mutational status of all forms of
1900s and up until the 2016 World Health Organi- diffuse gliomas reported that more than 70% of his-
zation (WHO) brain tumor classification, diffuse glio- tologic grade 2 and 3 diffuse gliomas (oligoden-
mas were classified based upon the morphologic drogliomas, astrocytomas, and oligoastrocytomas)
and secondary GBMs (those that had progressed from
features of neoplastic cells.1 Those that had micro-
a lower-grade glioma) had IDH mutations.4 In contrast,
scopic features of astrocytic differentiation were as-
very few primary (de novo) GBMs harbored IDH mu-
trocytomas, those with features of oligodendrocytes
tations. Moreover, IDH mutations were associated with
were oligodendrogliomas, and those with both were
favorable clinical outcomes in patients with grade 2, 3,
oligoastrocytomas. Within a class, grading was based
or 4 astrocytomas. These findings led to our current
upon the presence of morphologic features that cor- understanding that IDH wild-type and IDH mutant
Author affiliations responded with patient outcome, including mitotic
and support
diffuse gliomas represent distinct diseases.
activity, microvascular hyperplasia, and necrosis.
information (if Another groundbreaking discovery was the identifi-
applicable) appear Significant concerns arose regarding the reproducibility cation of mutations in the histone H3F3A and
at the end of this and predictive value of morphologic classification and
article. HIST1H3B genes in high-grade diffuse gliomas that
grading, because the criteria were subjective and occurred in the pons and thalamus, predominantly in
Accepted on March
23, 2020 and depended on features such as nuclear roundness, children. The most common were mutations at the
published at regularity, and chromatin pattern, as well as cytoplas- H3K27 locus, found in 70% to 80% of pontine
ascopubs.org on mic clearing, eosinophilia, and size. Multiple studies gliomas.5,6 Of interest, the type of H3 mutation dictates
April 23, 2020:
documented the relatively poor interobserver agree- the tumor location, as well as its clinical features.
10.1200/EDBK_ ment for the diagnosis of diffuse gliomas based on Nearly all diffuse gliomas with an H3 mutation at the
280955 histologic assessment.2 K27 locus have a brainstem or thalamic location and

Hu et al
IDH mutant, 1p/19q codeleted—had the longest patient

survivals, distinct gene expression, and methylation profiles
and harbored mutations in CIC, FUBP1, Notch1, and the
• The incorporation of molecular information, TERT promoter. Those with IDH mutations but lacking
including IDH mutation status and the presence
1p/19q codeletions were characterized by TP53 mutations
or absence of 1p/19q codeletion, has led to the
and losses of ATRX and are now considered IDH mutant
concept of an integrated pathologic diagnosis
and a new WHO classification that provides astrocytomas, with associated survivals slightly shorter than
more accurate prognostic and predictive in- oligodendrogliomas. Those diffuse gliomas in adults that
formation for providers and their patients with lacked IDH mutation harbored genetic alterations similar to
primary malignant gliomas. IDH wild-type GBMs: EGFR amplification, PTEN and
• Advances in surgical navigation technology, CDKN2A deletions, and mutations in Rb1, NF1, and TP53.
coupled with techniques to better distinguish Clinical outcomes were similar to IDH wild-type GBMs, with
and selectively treat tumor-infiltrating normal median survival of less than 18 months.
brain tissue, has led to enhanced diagnostic, With overwhelming clinical and molecular evidence from
cytoreductive, and therapeutic results from multiple large-scale multiplatform investigations, the 2016
neurosurgical intervention in patients with
WHO classification incorporated genetic alterations into the
malignant brain tumors.
definitions of brain tumor entities and a part of an integrated
• Novel MRI and radiogenomics techniques can diagnosis.1 The diffuse gliomas included IDH wild-type and
help to distinguish true tumor growth from IDH mutant diffuse astrocytoma, WHO grade 2; IDH wild-
pseudoprogression and address intratumoral
type and IDH mutant anaplastic astrocytoma, WHO grade 3;
heterogeneity to localize genetically distinct,
and IDH wild-type and IDH mutant GBM, WHO grade 4. The
highly perfused, and metabolically active sites
for biopsy or resection. integrated diagnosis of oligodendrogliomas and anaplastic
oligodendrogliomas required the presence of IDH mutation
• Personalized therapeutic decision making, in-
and 1p/19q codeletion. The diagnosis of IDH mutant as-
cluding whether to use TTFields and/or po-
trocytomas (grades 2, 3, and 4) is supported by mutations in
tentially active targeted or immunotherapeutic
agents based on NGS of patients’ tumors, is now TP53 and loss or mutation of ATRX. It should be noted that
routine in the neuro-oncology clinic. the 2016 WHO classification did not include an integrated
diagnosis for oligoastrocytomas, because there was no
evidence to suggest a molecular definition of these “mixed”
typically arise in young children. H3 mutations that occur at tumors.
the G34 position lead to high-grade cerebral hemispheric
gliomas in adolescent and young adult patients. Moreover, Recent Advances in Addition to 2016 WHO Classification:
the mutations that co-occur with H3 mutations, such as cIMPACT-NOW
TP53, DAXX, ATRX, and ACVR1, also play a role in de- Although the incorporation of genetic alterations into in-
termining clinical features and precise tumor location along tegrated diagnoses represented a large step forward, it
the neural axis.7 Although pontine and thalamic high-grade became evident that the pace of change in molecular on-
diffuse gliomas are best known for their occurrence in cology research far exceeded the updates in the WHO
children, it is now evident that diffuse midline gliomas with classification and required a sanctioned mechanism that
histone H3K27 mutation also occur in adults, suggesting could more readily adopt these changes. cIMPACT-NOW
that the midline location of high-grade diffuse gliomas is (the Consortium to Inform Molecular and Practical Ap-
more tightly associated with H3K27 mutations than is pa- proaches to CNS Tumor Taxonomy) was formed by a group
tient age.8 The “diffuse midline glioma, H3K27 mutant, of experienced neuropathologists with clinical advisory
WHO grade 4” was added to the 2016 WHO classification, boards and steering committees representing the neuro-
representing the first brain tumor type that was graded oncology community; its primary goals were the evaluation
based on the presence of a specific mutation.1 and incorporation of neuro-oncology advances into clinical
practice.11 Working groups were formed to address practice
In 2015, two independent studies demonstrated that the
gaps and classification shortcomings.
lower-grade gliomas—those of histologic grade 2 and
3—could be classified more precisely based on molecular IDH mutant astrocytomas
criteria than by histology alone. In particular, the presence The grading criteria for diffuse astrocytic gliomas in the
or absence of IDH mutations and 1p/19q codeletions in this 2016 WHO classification were developed prior to the un-
group of diffuse gliomas divided them into three clinically derstanding of molecularly distinct entities, and the same
relevant classes.9,10 Those that had IDH mutations and grading criteria were used for IDH mutant and IDH wild-type
1p/19q codeletions—now considered oligodendroglioma, gliomas. Multiple retrospective studies have concluded that

Emerging Technologies and the Care of Patients with Brain Tumors
histologic grading may not stratify risk for patients with IDH For more than a century, the diagnosis of diffuse gliomas
mutant astrocytomas in the WHO grade 2 and 3 categories; was based primarily on histologic appearance, yet recent
others have demonstrated that traditional grading schemes molecular genetic discoveries have forced a re-evaluation of
are still capable of stratifying risk for these patients.12-15 The diagnostic definitions and criteria. We are now firmly within
cIMPACT-NOW Working Committee 1 evaluated the liter- the molecular era in which integrated diagnoses are for-
ature to determine whether there was sufficient evidence to mulated by directly incorporating ancillary test results.
define molecular genetic or other criteria that could reliably Neuropathologists, molecular pathologists, and clinicians
stratify risk among patients with IDH mutant diffuse astro- now work closely together to provide optimal clinical care by
cytic gliomas or could identify those tumors that would utilizing molecular biomarkers for diagnostic and treatment
behave most aggressively, with a clinical course corre- purposes.
sponding to WHO grade 4. They concluded that the evi-
THE ROLE OF SURGICAL DEVICES IN THE TREATMENT OF
dence in the literature was strong for an association between
MALIGNANT GLIOMAS
homozygous deletion of CDKN2A/B and shorter survival for
patients with IDH mutant astrocytomas, corresponding to The standard of care for newly diagnosed malignant gliomas
WHO grade 4 clinical behavior. They recommended that involves maximal safe surgical resection of most lesions
CDKN2A/B homozygous deletion be considered a criterion prior to treatment with chemotherapy and radiation.19 The
for astrocytoma, IDH mutant, grade 4, along with necrosis purpose of surgery is to establish a diagnosis and cyto
and microvascular hyperplasia.16 reduce the tumor. Multiple studies have demonstrated
important survival benefits related to a greater extent of
IDH wild-type astrocytomas resection (EOR) for patients with low-grade and high-grade
Multiple studies have indicated that a substantial subset of gliomas.20-23 However, EOR must be balanced against the
histologic grade 2 and 3 IDH wild-type astrocytomas that risk of causing neurologic deficits, especially with tumors
occur in adults has an aggressive clinical course, with near eloquent regions. For gliomas, this delicate balance is
overall patient survival equal to or only slightly longer than made more difficult by the lack of clear visual distinction
patients with IDH wild-type GBM, WHO grade 4.9,10,17 The between normal and tumor-infiltrated tissue.
cIMPACT-NOW Working Committee 1 evaluated the liter- The technological advancements in brain tumor surgery
ature to determine whether there was sufficient evidence to over the past 2 decades generally serve one of two pur-
define molecular genetic criteria that could reliably identify poses: to improve recognition of tumor-infiltrated tissue,
histologic grade 2 and 3 IDH wild-type that would behave allowing greater EOR, or to improve recognition of eloquent
with a clinical course corresponding to WHO grade 4. They tissue to reduce the risk of neurologic deficits (Table 1).
concluded that those IDH wild-type diffuse astrocytic glio- Together, these technologies allow aggressive, but safe,
mas of histologic grade 2 or 3 that carry EGFR amplification, resections.
+7/ 10 or TERT promoter mutation have clinical outcomes
similar to patients with IDH wild-type GBM and, therefore, Surgical Navigation
should be considered grade 4. They reached consensus on The most important technological advancement in modern
the designation “diffuse astrocytic glioma, IDH wild-type, brain tumor surgery is the use of frameless stereotactic
with molecular features of GBM, WHO grade 4” as the most navigation. 24 Localization of structures within the brain
appropriate terminology.18 requires correlation of preoperative imaging with
TABLE 1. Impact of Technology on Surgical Outcome for Brain Tumors

Surgical Impact
Technology Reduce Neurologic Morbidity Improve EOR Improve Efficacy of Medical Therapy
Stereotactic navigation X X
Functional imaging (fMRI, DTI, MEG) X
Direct cortical stimulation X
Intraoperative imaging (iMRI, iUS) X
Fluorescent tumor dyes X
Intraoperative histopathology (Raman spectroscopy) X
Laser ablation X X
Abbreviations: DTI, diffusion tensor imaging; EOR, extent of resection; fMRI, functional MRI; iMRI, intraoperative MRI; iUS, intraoperative ultrasound; MEG,
magnetoencephalography.

Hu et al
intraoperative spatial anatomy. Errors in localization can randomized trial.31 However, the high cost of intraoperative
lead to unnecessary brain injury and incomplete identifi- MRI is not justifiable for most hospitals. Intraoperative ul-
cation/resection of tumor. Frame-based stereotactic local- trasound is a relatively inexpensive real-time imaging ad-
ization has been used in neurosurgical procedures for more juvant, but it has more limited resolution.32 Intraoperative CT
than a century. The introduction of modern computers and has limited value for identifying tumors during surgery, but it
optical trackers has allowed a frameless approach where can be used to achieve very high accuracy for stereotactic
any instrument can be tracked and its position mapped onto registration.33
preoperative imaging.25 However, the accuracy of naviga- In addition to augmenting imaging data, new technology
tion is only as good as the registration between the pre- has improved the surgeon’s direct visualization of tissue.
operative imaging and the patient’s anatomy.26 Navigation Tools for enhanced surgical vision, including robot-driven
systems cannot account for shifts in brain position relative to microscopes, two-dimensional/three-dimensional endo-
the fixed anatomy of the skull when cerebral spinal fluid scopes, and three-dimensional exoscopes, have entered
(CSF) is drained and tissue is removed. Therefore, this routine clinical practice. Although these tools allow im-
technology becomes increasingly inaccurate as the re- proved visualization at high-powered magnification, the
section progresses. visual distinction between tumor and normal tissue remains
Technology to Improve Identification of Functional Tissue challenging. To address this, tumor-specific fluorescent
Preoperative localization Numerous preoperative imaging dyes have been developed that capitalize on the metabolic
studies can be used to identify eloquent function in the differences between tumor and surrounding tissue. For
cortex and subcortical tracts, including functional MRI, high-grade gliomas, the use of 5-aminolevulinic acid has
diffusion tensor imaging, high-density electroencephalog- been shown to double the rate of achieved gross total
raphy, magnetoencephalography, and preoperative brain resection in a phase III randomized trial.34 However, lower-
stimulation using transcranial magnetic stimulation.27 grade tumors are less metabolically active, and 5-
Functional localization data can be superimposed over aminolevulinic acid has not been correlated with tumor
anatomic imaging in frameless stereotactic systems identification or improvement in EOR for low-grade glio-
to display enriched information during surgery. Each im- mas.35 Finally, the introduction of rapid histopathologic
aging technique has its own limitations in spatial resolution, analysis, such as the use of Raman spectroscopy coupled
which are magnified by the registration error and brain shift. with artificial intelligence analysis, allows rapid assessment
of tumor margin biopsies in the operating room without
Intraoperative localization Because of the limitations in- reliance on a pathologist.36 This approach is still limited by
herent in stereotactic navigation, the gold standard for the need to process tissue removed from the operative field
functional anatomic localization remains direct cortical/ and does not allow identification of tumor in situ.
subcortical stimulation. Numerous paradigms exist for di-
rect cortical/subcortical stimulation through handheld Technology to Improve the Biologic Response to
probes or grids placed over the cortical surface. Studies Systemic Therapy
comparing the accuracy of various modalities of pre- Although a greater extent of surgical resection has been
operative functional localization (functional MRI, magne- shown to incrementally improve survival, gliomas are not
toencephalography, transcranial magnetic stimulation) with fundamentally a surgically curable disease. Therefore,
direct cortical/subcortical stimulation clearly demonstrate marginal advancements in the ability to safely resect more
localization errors of 4 mm to 10 mm compared with tumors are not likely to result in additional substantial in-
intraoperative stimulation.28-30 creases in survival. Neurosurgical oncology needs new
Technology to Improve Identification of Tumor technologies that impact the underlying biology of the
The same limitations in stereotactic navigation that impact disease. Of all new instrumentation introduced over the past
the accuracy of functional tissue localization also influence decade, one technology stands out in this regard: laser
the ability to identify tumor. To achieve more complete interstitial thermal therapy (LITT).
resections, numerous technologies have been used to LITT, also known as stereotactic laser ablation, uses
enhance tumor identification, particularly at the lesion a stereotactically implanted fiber-optic probe to ablate
margin. The use of intraoperative imaging gives real-time tumors with a near-infrared wavelength laser while moni-
information about residual tumor and can be used to update toring thermal injury in real-time using MRI. This minimally
the stereotactic navigation. The value of intraoperative im- invasive approach offers an alternative method of cytor-
aging is only as good as the imaging quality. Intraoperative eduction with equivalent safety to open craniotomy and
MRI is the most useful, giving surgeons the ability to achieve decreased recovery time.37,38 The greatest drawback is
a complete resection in 96% of cases compared with 68% that, by ablating tissue without resecting it, mass-related
of cases using standard navigation in a prospective symptoms are not addressed as with resection.39 In addition

to direct tumor cytotoxicity, LITT has secondary effects on planning.43-49 At the same time, completely nonenhancing
blood-brain barrier (BBB) permeability and local in- gliomas are classically considered low grade when, in fact,
flammatory response. Thermal tissue damage results in approximately one-third of these tumors are high grade.50-52
coagulative necrosis within the primary zone of injury, re- Surgical targeting becomes problematic in these cases,
leasing potential tumor antigens and inflammatory cyto- because high-grade and low-grade components can coexist
kines.40 In the surrounding tissue, transient disruption of the within a single tumor but are indistinguishable on con-
BBB and improved penetration of systemic chemotherapy ventional MRI, with a reported 30% incidence of sampling
have been demonstrated.41 The proinflammatory vascular error and misdiagnosis (i.e., undergrading). 53 MRI
and cytotoxic changes produced by LITT are believed to enhancement also fails to distinguish nontumoral post-
be similar to those produced by ionizing radiation.42 The treatment–related inflammation (e.g., pseudoprogression,
combination of LITT with early delivery of cytotoxic che- radiation necrosis) from tumor recurrence for image-based
motherapy or systemic immunotherapy may enhance de- response assessment.54,55 Although nontumoral-related
livery to the tumor microenvironment. As such, LITT may changes represent a positive response to treatment and
offer a biologic advantage to resection alone, although a good prognosis, their identical appearance to tumor
further studies are necessary. There are two ongoing studies regrowth can misguide treatment decisions.54,55 Intrale-
of LITT for high-grade gliomas in combination with immune sional heterogeneity and admixture between tumor and
checkpoint inhibition (NCT03277638, NCT03341806). post-treatment changes can also impact surgical targeting
In summary, advances in surgical technology serve an when histologic confirmation of recurrent disease is
important role to protect neurologic function while maxi- needed.56,57
mizing EOR, which is associated with improved survival. Advanced Techniques in Physiologic Imaging and
Although these tools are certainly more than just toys, we Tumor Modeling
have reached the limit of efficacy for EOR. New technology
The past decade has seen increasing assimilation of ad-
must focus on alerting the underlying biology of the tumor
vanced imaging techniques that provide diverse and re-
environment to enhance the efficacy of adjuvant therapies.
gionally specific information on brain tumor physiology. This
CONSISTENCY AND INNOVATION IN BRAIN TUMOR IMAGING includes insight into regional differences in bulk water
For more than four decades, conventional contrast- movement on diffusion-weighted imaging,58 white matter
enhanced MRI has guided initial diagnosis, targeted integrity on diffusion tensor imaging,58,59 and microvessel
treatment planning, and response assessment for central morphology on dynamic susceptibility contrast perfusion
nervous system tumors like high-grade glioma. Classically, MRI (DSC-MRI).60,61 In particular, DSC-MRI measures of
tumor burden has been defined by MRI enhancement on relative cerebral blood volume (rCBV) have become an
T1-weighted images, which signifies extravasation of almost indispensable tool in neuro-oncologic imaging, given
gadolinium-based contrast agent through regions of dis- the highly angiogenic nature of high-grade gliomas and their
rupted BBB. For surgical planning, this enhancement helps propensity to form large tortuous microvessels that associate
to localize biopsy targets for histologic confirmation and with high rCBV values.62 This elevated rCBV helps to dif-
defines the anatomic extent of disease for maximal safe ferentiate high-grade gliomas from relatively nonangiogenic
resection. Areas of MRI enhancement also typically receive low-grade gliomas, as well as nontumoral tissues, such as
the highest prescribed radiation treatment dose, and they pseudoprogression and radiation necrosis, which all exhibit
define progressive/recurrent tumor burden on post- low rCBV values. Different studies have demonstrated how
treatment response assessment. Yet, this conventional rCBV values can distinguish high-grade components from
paradigm has started to evolve over recent years, as we low-grade components within nonenhancing gliomas,50,53
better understand the biologic complexities and internal as well as help to differentiate tumor recurrence from
heterogeneity of these tumors, as well as the clinical limi- nontumoral post-treatment effects.57,63-67
tations (and pitfalls) of relying solely on MRI enhancement. Methodology of Correlative Studies Can Impact the
Limitations of Conventional MRI Guidelines for Prospective Diagnosis
MRI enhancement generally underestimates true tumor Clinical guidelines for imaging diagnosis are typically driven
burden. Many enhancing high-grade gliomas, such as by correlative studies comparing imaging measurements
GBMs, also contain substantial portions of nonenhancing with clinical benchmarks (e.g., histologic diagnosis). Yet,
tumor (with intact BBB) that are poorly visualized on T1- methodologic differences between published studies can
weighted postcontrast images.43-49 These nonenhancing impact the quality of correlation and the consistency of
components are often indistinguishable from surrounding recommendations. This is well illustrated in the context of
edematous nontumoral brain parenchyma, which presents DSC-MRI perfusion guidelines for response assessment in
challenges for surgical targeting and dosimetric radiation GBMs. A recent meta-analysis has highlighted substantial

Hu et al
variability in recommended rCBV thresholds across the algorithms, which then predict genetic aberrations on new
many published studies using rCBV to distinguish tumor unseen cases, using the MRI images alone. Historically,
from post-treatment effects.66 This can be attributed, at least most published radiogenomics studies have used non-
in part, to differences in DSC-MRI methodology, which has localizing correlations, which assume homogeneous ex-
motivated efforts to develop consensus recommendations pression of these genetic aberrations but remain incapable
for standardization.68 However, at the same time, differ- of resolving intratumoral heterogeneity. 72,78-85 Recent
ences in the correlative analysis itself (and how the “gold studies using image-localized biopsies, spatially matched
standard” reference benchmark is defined) can sub- MRI, and genetic data inputs have shown the feasibility of
stantially contribute to discrepancies in recommended using MRI features and machine learning–based algorithms
guidelines. This is particularly true in the context of intra- to resolve the genetically distinct subpopulations throughout
tumoral heterogeneity. In particular, the post-treatment individual GBM tumors, including the nonenhancing
tumor bed can contain variable histologic admixture that components that are typically left unresected.71,72 None-
influences the histologic quantity of recurrent tumor relative theless, clinical assimilation of radiogenomics models will
to post-treatment effects. This can degrade imaging-tissue require continued work to address issues such as re-
correlations if “nonlocalizing” biopsies are not spatially producibility of model outputs and biologic interpretability,86
aligned with the imaging measurement locations. This can as well as an understanding of the predictive uncertainty
result in wide overlap in rCBV values, with poor distinction associated with individual model predictions.87
between tumor and post-treatment effect.68,69 By compar- TECHNOLOGY IN THE NEURO-ONCOLOGY CLINIC
ison, use of image-localized biopsies and spatially matched
localized rCBV measurements can help to resolve intra- Tumor Treating Fields
tumoral heterogeneity and isolate more histologically pure Electrical activity plays a key role in many essential cellular
biopsy samples. This has helped to identify highly accurate processes, including cell division. Living cells contain
rCBV thresholds that can differentiate tumor from post- multiple charged or polar constituents, including the mitotic
treatment effect with greater than 95% accuracy, at the spindle apparatus, that can be affected and disrupted by
resolution of a single voxel.56,57,65 Similarly, image-localized externally applied electric fields, leading to mitotic catas-
biopsy methods have also proved instrumental for other trophe and cell death.88,89 TTFields, or Optune, is a device
correlative studies resolving intratumoral heterogeneity, that produces low-intensity intermediate-frequency
including prediction of tumor cell density and extent,44,49,70 (100–300 kHz) alternating electrical fields that are inten-
molecular and genetic heterogeneity,71,72 and regional ded to disrupt cell division and inhibit tumor growth. The
variability in histopathologic features.73,74 optimal field frequency to target a specific tumor histology is
dependent on cell size and shape; the optimal frequency for
Radiogenomics and Precision Medicine GBM was found to be 200 kHz with a field strength of at least
Precision medicine offers the potential to optimize in- 1 Vcm.88,89 Preclinical and animal experiments demon-
dividualized drug regimens based on the genetic aberra- strated profound growth inhibition in dividing cells, in part
tions and unique sensitivities of each patient’s tumor.75 through cell cycle arrest and apoptotic cell death. Other
These therapies represent the primary lines of defense in potential antitumor mechanisms of TTFields include effects
combating residual unresected tumor, particularly for ag- on membrane integrity, cell migration, BBB integrity, an-
gressive cancers like GBM, which are rarely amenable to giogenesis, and local immune activation.90,91
complete surgical resection. A fundamental challenge for The approved TTFields device used in patients consists of
precision medicine lies in resolving intratumoral genetic a voltage generator attached to two pairs of transducer
heterogeneity. Specifically, each patient’s GBM tumor ac- arrays, each of which contains nine insulated electrodes,
tually consists of multiple genetically distinct sub- that are applied in a roughly orthogonal pattern on a pa-
populations, such that the genetic drug targets from one tient’s shaved scalp.92 A proprietary treatment-planning
biopsy location may not accurately reflect those from other software (NovoTAL) produces an individualized array ap-
parts of the same tumor.76 This heterogeneity and associ- plication template, based on head size and shape, as well as
ated sampling errors create uncertainty in the true genetic on tumor size and location, that is designed to maximize
drug targets that are left behind in the residual unresected tumor exposure to the generated alternating electrical
tumor.59,71,77 Ironically, these “unknown” targets are the fields.93 Following early-phase safety studies, the device was
most critical in deciding appropriate adjuvant therapy.45,46,72 initially evaluated in a phase III trial (EF-11) in patients with
Radiogenomics offers a promising and clinically feasible recurrent GBM.94
solution to bridge the challenges of intratumoral hetero- The EF-11 trial assessed the efficacy of TTFields in 237
geneity. These models integrate noninvasive MRI images patients with recurrent GBM who were randomly assigned in
with genetic profiles as data inputs to train machine-learning a 1:1 fashion to receive TTFields versus physician’s choice

chemotherapy. Patients could have had any number of prior least 90% of the time experienced a median survival of
therapies for recurrence, and the primary study endpoint about 25 months (almost 29 months since diagnosis), and
was overall survival (OS). Although not powered as a non- their 5-year survival was 29.3% versus only 4.5% for pa-
inferiority trial, the study did not reveal any significant dif- tients treated with standard chemoradiation therapy alone.
ferences between the two groups, with a median OS of Ongoing or planned evaluations of TTFields include com-
6.6 months versus 6.0 months (hazard ratio [HR], 0.86; p = binatorial strategies in patients with GBM, in addition to
.27) in the TTFields and best-choice chemotherapy arms, clinical trials evaluating efficacy in patients with malignant
respectively. Patients who were randomly selected to re- mesothelioma, as well as pancreatic, ovarian, and non–
ceive TTFields, experienced improved quality of life, as small cell lung cancers.97
measured by improved cognitive and emotional functioning,
Next-Generation Sequencing and Beyond
and they experienced fewer severe adverse events com-
pared with the standard treatment group. Over the past decade, the use of advanced technologies,
such as NGS, has changed the way in which primary brain
The EF-14 study was the pivotal clinical trial that led to the
tumors are classified and introduced the concept of an
approval of TTFields for use in adults with newly diagnosed
integrated diagnosis.98 Incorporation of genomic data into
GBM; it enrolled 695 patients in North America, Europe,
the classification of brain tumors changed the way in which
Korea, and Israel who had completed concurrent standard
we use data to guide discussions with patients regarding
radiation and temozolomide.95 Patients were randomly se-
prognosis, treatment options, and enrollment on clinical
lected to receive, in a 2:1 ratio, TTFields plus monthly
trials. In the coming decade, we can expect that additional
maintenance temozolomide or temozolomide alone, and
data will be gathered from tumor tissues, using multiplat-
patients whose clinical status worsened or who developed
form technologies to support clinical decision making. This
disease progression during concurrent therapy were in-
multi- “omics” or integrative “omics” approach will expand
eligible. TTFields was initiated at least 4 weeks, but not more
the use of DNA-based NGS testing (genomics), including
than 7 weeks, from the last day of radiation, and this therapy
analysis of circulating tumor DNA from CSF or plasma, RNA
was continued for at least the 6-month duration of adjuvant
sequencing (transcriptomics), and genome-wide methyla-
temozolomide. Patients who were randomly assigned to
tion profiling (epigenomics).
TTFields were recommended to use the device for a mini-
mum of 18 hours a day. The median progression-free DNA-based NGS has been the most widely adopted
survival from randomization was 6.7 months versus 4 methodology and can take on several forms, including
months (HR, 0.63; p , .001), and the median OS was 20.9 targeted panels, whole-exome sequencing, and whole-
months versus 16.0 months (HR, 0.6; p , .001) in the genome sequencing. Although the number of genes and
TTFields and temozolomide-alone groups, respectively. An regions tested will vary between these approaches, the
exploratory subgroup analysis did not reveal any differences results generated from tissue-based NGS testing will con-
in these results based on age, performance status, EOR, or tinue to provide the genomic landscape for individual pa-
methylguanine methyltransferase (MGMT) promoter tient’s tumors and serve as the basis for prognosis (e.g.,
methylation status. The median duration of TTFields IDH1 R132H), lineage classification (oligodendroglial vs.
treatment was 8.2 months, and 51% of patients continued astrocytic), and new therapeutic approaches. Tumor tissue
on this treatment after their first progression, as allowed in has been required to perform NGS testing for patients with
the protocol. Notably, 13% of TTFields patients were alive at glioma; however, several recent studies have shown great
5 years versus 5% of patients in the control arm. The only potential in adapting the technology to circulating tumor
significantly increased adverse event on the TTFields arm DNA from CSF,99-101 which may allow for genomic-based
was mild to moderate skin toxicity, which occurred in ap- tumor lineage classification in the presurgical setting and
proximately 52% of patients who wore the device. Ap- provide genomic data for patients with recurrent disease,
proximately 75% of patients who were assigned to TTFields many of whom will not undergo a second neurosurgical
used the device for at least 18 hours per day (monthly procedure. It may also provide a more sensitive mechanism
average for the first 6 months of the study). Despite these to track tumor evolution while patients are undergoing
positive results, this open-label study has been criticized for treatment, assess response to treatment, and help to dis-
not using a placebo or sham device in the control arm. tinguish between disease progression and
Post hoc analysis of the EF-14 phase III data suggested pseudoprogression.100
a dose effect of TTFields exposure, with device compliance Transcriptomic analysis through RNA sequencing adds
of at least 50% leading to improved progression-free survival another layer of data that is focused on expression of genes
and OS compared with the standard arm.96 There was also in the tumor and microenvironment. This approach provides
evidence of a dose response with increasing use of the an opportunity to understand how tumor cells aberrantly
device. Patients whose device was in place and in use for at activate specific signaling pathways, which may highlight

Hu et al
tumor vulnerabilities, as well as mechanisms of reagents. Although current MGMT assays evaluate only a few
sistance. 102 Transcriptomics can also aid in mapping CpG sites within the MGMT promoter, methylation arrays or
intratumoral and peritumoral immune infiltrates.103,104 NGS-based methylation sequencing methods can provide
When paired with spatial mapping, this technology will be a genome-wide view of the methylome far beyond traditional
able to characterize subtypes of immune cells present in the MGMT promoter testing. Genome-wide methylation signa-
tumor microenvironment.103,105 As the value of immuno- tures have been shown to support brain tumor lineage
therapy in gliomas continues to be investigated, integration classification, as well as distinguish subtypes within dis-
of biomarkers linked to immune signatures may be im- eases, such as medulloblastoma.106,107
portant in identifying which patients may benefit from such In summary, technologic advancements are providing op-
therapeutic approaches. portunities to fully characterize individual tumors in
Epigenomics has a long history within neuro-oncology, given a comprehensive manner. These data will aid in diagnosis,
that MGMT promoter methylation testing is routinely per- provide opportunities for new therapeutic strategies, and
formed to identify patients who are more or less likely to offer robust mechanisms for identifying and potentially
respond to standard of care treatment with alkylating treating progressive disease.

1
Neuroradiology Section, Department of Radiology, Mayo Clinic, Phoenix, Glenn J. Lesser, MD, FACP, Wake Forest Baptist Comprehensive Cancer
AZ Center, Medical Center Boulevard, Winston-Salem, NC 27157-1082;
2
Department of Pathology, Northwestern University Feinberg School of email: [email protected].
Medicine, Chicago, IL
3
Department of Neurologic Surgery, UC Davis Comprehensive Cancer
Center, Sacramento, CA AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
4
Foundation Medicine, Inc., Morrisville, NC AND DATA AVAILABILITY STATEMENT
5
Comprehensive Cancer Center, Wake Forest Baptist Health, Winston- Disclosures provided by the authors and data availability statement (if
Salem, NC applicable) are available with this article at DOI https://doi.org/10.1200/
6
Department of Pathology, Wake Forest School of Medicine, Winston- EDBK_280955.
Salem, NC
REFERENCES
1. Louis DN, Ohgaki H, Wiestler OD, et al. WHO Classification of Tumours of the Central Nervous System, Revised 4th ed. Lyon: International Agency for Research
on Cancer; 2016.
2. van den Bent MJ. Interobserver variation of the histopathological diagnosis in clinical trials on glioma: a clinician’s perspective. Acta Neuropathol. 2010;
120:297-304.
3. Parsons DW, Jones S, Zhang X, et al. An integrated genomic analysis of human glioblastoma multiforme. Science. 2008;321:1807-1812.
4. Yan H, Parsons DW, Jin G, et al. IDH1 and IDH2 mutations in gliomas. N Engl J Med. 2009;360:765-773.
5. Wu G, Broniscer A, McEachron TA, et al; St. Jude Children’s Research Hospital–Washington University Pediatric Cancer Genome Project. Somatic histone H3
alterations in pediatric diffuse intrinsic pontine gliomas and non-brainstem glioblastomas. Nat Genet. 2012;44:251-253.
6. Schwartzentruber J, Korshunov A, Liu XY, et al. Driver mutations in histone H3.3 and chromatin remodelling genes in paediatric glioblastoma. Nature. 2012;
482:226-231.
7. Mackay A, Burford A, Carvalho D, et al. Integrated molecular meta-analysis of 1,000 pediatric high-grade and diffuse intrinsic pontine glioma. Cancer Cell. 2017;
32:520-537.e5.
8. Solomon DA, Wood MD, Tihan T, et al. Diffuse midline gliomas with histone H3-K27M mutation: a series of 47 cases assessing the spectrum of morphologic
variation and associated genetic alterations. Brain Pathol. 2016;26:569-580.
9. Brat DJ, Verhaak RG, Aldape KD, et al; Cancer Genome Atlas Research Network. Comprehensive, integrative genomic analysis of diffuse lower-grade gliomas.
N Engl J Med. 2015;372:2481-2498.
10. Eckel-Passow JE, Lachance DH, Molinaro AM, et al. Glioma groups based on 1p/19q, IDH, and TERT promoter mutations in tumors. N Engl J Med. 2015;
372:2499-2508.
11. Louis DN, Aldape K, Brat DJ, et al. Announcing cIMPACT-NOW: the Consortium to Inform Molecular and Practical Approaches to CNS Tumor Taxonomy. Acta
Neuropathol. 2017;133:1-3.
12. Olar A, Wani KM, Alfaro-Munoz KD, et al. IDH mutation status and role of WHO grade and mitotic index in overall survival in grade II-III diffuse gliomas. Acta
Neuropathol. 2015;129:585-596.

13. Reuss DE, Mamatjan Y, Schrimpf D, et al. IDH mutant diffuse and anaplastic astrocytomas have similar age at presentation and little difference in survival:
a grading problem for WHO. Acta Neuropathol. 2015;129:867-873.
14. Shirahata M, Ono T, Stichel D, et al. Novel, improved grading system(s) for IDH-mutant astrocytic gliomas. Acta Neuropathol. 2018;136:153-166.
15. Yang RR, Shi Z-F, Zhang Z-Y, et al. IDH mutant lower grade (WHO Grades II/III) astrocytomas can be stratified for risk by CDKN2A, CDK4 and PDGFRA copy
number alterations. Brain Pathol. Epub 2019 Nov 16.
16. Brat DJ, Aldape K, Colman H, et al. cIMPACT-NOW update 5: recommended grading criteria and terminologies for IDH-mutant astrocytomas. Acta Neuropathol.
2020;139:603-608.
17. Ceccarelli M, Barthel FP, Malta TM, et al; TCGA Research Network. Molecular profiling reveals biologically discrete subsets and pathways of progression in
diffuse glioma. Cell. 2016;164:550-563.
18. Brat DJ, Aldape K, Colman H, et al. cIMPACT-NOW update 3: recommended diagnostic criteria for “Diffuse astrocytic glioma, IDH-wildtype, with molecular
features of glioblastoma, WHO grade IV”. Acta Neuropathol. 2018;136:805-810.
19. Stupp R, Mason WP, van den Bent MJ, et al; National Cancer Institute of Canada Clinical Trials Group. Radiotherapy plus concomitant and adjuvant
temozolomide for glioblastoma. N Engl J Med. 2005;352:987-996.
20. Jakola AS, Myrmel KS, Kloster R, et al. Comparison of a strategy favoring early surgical resection vs a strategy favoring watchful waiting in low-grade gliomas.
JAMA. 2012;308:1881-1888.
21. Kavouridis VK, Boaro A, Dorr J, et al. Contemporary assessment of extent of resection in molecularly defined categories of diffuse low-grade glioma: a volumetric
analysis. J Neurosurg. Epub 2019 Oct 25.
22. Molinaro AM, Hervey-Jumper S, Morshed RA, et al. Association of Maximal Extent of Resection of Contrast-Enhanced and Non-Contrast-Enhanced Tumor With
Survival Within Molecular Subgroups of Patients With Newly Diagnosed Glioblastoma. JAMA Oncol. Epub 2020 Feb 6.
23. Sanai N, Polley MY, McDermott MW, et al. An extent of resection threshold for newly diagnosed glioblastomas. J Neurosurg. 2011;115:3-8.
24. Willems PW, Taphoorn MJ, Burger H, et al. Effectiveness of neuronavigation in resecting solitary intracerebral contrast-enhancing tumors: a randomized
controlled trial. J Neurosurg. 2006;104:360-368.
25. Willems PW, van der Sprenkel JW, Tulleken CA, et al. Neuronavigation and surgery of intracerebral tumours. J Neurol. 2006;253:1123-1136.
26. Steinmeier R, Rachinger J, Kaus M, et al. Factors influencing the application accuracy of neuronavigation systems. Stereotact Funct Neurosurg. 2000;
75:188-202.
27. Kreidenhuber R, De Tiège X, Rampp S. Presurgical functional cortical mapping using electromagnetic source imaging. Front Neurol. 2019;10:628.
28. Ellis DG, White ML, Hayasaka S, et al. Accuracy analysis of fMRI and MEG activations determined by intraoperative mapping. Neurosurg Focus. 2020;48:E13.
29. Giussani C, Roux FE, Ojemann J, et al. Is preoperative functional magnetic resonance imaging reliable for language areas mapping in brain tumor surgery?
Review of language functional magnetic resonance imaging and direct cortical stimulation correlation studies. Neurosurgery. 2010;66:113-120.
30. Ille S, Sollmann N, Hauck T, et al. Combined noninvasive language mapping by navigated transcranial magnetic stimulation and functional MRI and its
comparison with direct cortical stimulation. J Neurosurg. 2015;123:212-225.
31. Senft C, Bink A, Franz K, et al. Intraoperative MRI guidance and extent of resection in glioma surgery: a randomised, controlled trial. Lancet Oncol. 2011;
12:997-1003.
32. Sastry R, Bi WL, Pieper S, et al. Applications of ultrasound in the resection of brain tumors. J Neuroimaging. 2017;27:5-15.
33. Carl B, Bopp M, Saß B, et al. Reliable navigation registration in cranial and spine surgery based on intraoperative computed tomography. Neurosurg Focus.
2019;47:E11.
34. Stummer W, Pichlmeier U, Meinel T, et al; ALA-Glioma Study Group. Fluorescence-guided surgery with 5-aminolevulinic acid for resection of malignant glioma:
a randomised controlled multicentre phase III trial. Lancet Oncol. 2006;7:392-401.
35. Almekkawi AK, El Ahmadieh TY, Wu EM, et al. The Use of 5-Aminolevulinic Acid in Low-Grade Glioma Resection: A Systematic Review. Oper Neurosurg
(Hagerstown). Epub 2019 Dec 12.
36. Hollon TC, Pandian B, Adapa AR, et al. Near real-time intraoperative brain tumor diagnosis using stimulated Raman histology and deep neural networks. Nat
Med. 2020;26:52-58.
37. Ivan ME, Mohammadi AM, De Deugd N, et al. Laser ablation of newly diagnosed malignant gliomas: a meta-analysis. Neurosurgery. 2016;79(suppl 1):S17-S23.
38. Shah AH, Semonche A, Eichberg DG, et al. The Role of Laser Interstitial Thermal Therapy in Surgical Neuro-Oncology: Series of 100 Consecutive Patients.
Neurosurgery. Epub 2019 Nov 19.
39. Shah AH, Burks JD, Buttrick SS, et al. Laser interstitial thermal therapy as a primary treatment for deep inaccessible gliomas. Neurosurgery. 2019;84:768-777.
40. Jermakowicz WJ, Mahavadi AK, Cajigas I, et al. Predictive modeling of brain tumor laser ablation dynamics. J Neurooncol. 2019;144:193-203.
41. Leuthardt EC, Duan C, Kim MJ, et al. Hyperthermic laser ablation of recurrent glioblastoma leads to temporary disruption of the peritumoral blood brain barrier.
PLoS One. 2016;11:e0148613.
42. Ahluwalia M, Barnett GH, Deng D, et al. Laser ablation after stereotactic radiosurgery: a multicenter prospective study in patients with metastatic brain tumors
and radiation necrosis. J Neurosurg. 2018;130:804-811.
43. Sarkaria JN, Hu LS, Parney IF, et al. Is the blood-brain barrier really disrupted in all glioblastomas? A critical assessment of existing clinical data. Neuro-oncol.
2018;20:184-191.

Hu et al
44. Pafundi DH, Laack NN, Youland RS, et al. Biopsy validation of 18F-DOPA PET and biodistribution in gliomas for neurosurgical planning and radiotherapy target
delineation: results of a prospective pilot study. Neuro-oncol. 2013;15:1058-1067.
45. Price SJ, Jena R, Burnet NG, et al. Improved delineation of glioma margins and regions of infiltration with the use of diffusion tensor imaging: an image-guided
biopsy study. AJNR Am J Neuroradiol. 2006;27:1969-1974.
46. Berens ME, Giese A. “...those left behind.” Biology and oncology of invasive glioma cells. Neoplasia. 1999;1:208-219.
47. Baldock AL, Ahn S, Rockne R, et al. Patient-specific metrics of invasiveness reveal significant prognostic benefit of resection in a predictable subset of gliomas.
PLoS One. 2014;9:e99057.
48. Corwin D, Holdsworth C, Rockne RC, et al. Toward patient-specific, biologically optimized radiation therapy plans for the treatment of glioblastoma. PLoS One.
2013;8:e79115.
49. Hu LS, Ning S, Eschbacher JM, et al. Multi-parametric MRI and texture analysis to visualize spatial histologic heterogeneity and tumor extent in glioblastoma.
PLoS One. 2015;10:e0141506.
50. Scott JN, Brasher PM, Sevick RJ, et al. How often are nonenhancing supratentorial gliomas malignant? a population study. Neurology. 2002;59:947-949.
51. Maia AC Jr., Malheiros SM, da Rocha AJ, et al. Stereotactic biopsy guidance in adults with supratentorial nonenhancing gliomas: role of perfusion-weighted
magnetic resonance imaging. J Neurosurg. 2004;101:970-976.
52. Barker FG II, Chang SM, Huhn SL, et al. Age and the risk of anaplasia in magnetic resonance-nonenhancing supratentorial cerebral tumors. Cancer. 1997;
80:936-941.
53. Chaskis C, Stadnik T, Michotte A, et al. Prognostic value of perfusion-weighted imaging in brain glioma: a prospective study. Acta Neurochir (Wien). 2006;
148:277-285, discussion 285.
54. Brandes AA, Tosoni A, Spagnolli F, et al. Disease progression or pseudoprogression after concomitant radiochemotherapy treatment: pitfalls in neurooncology.
Neuro-oncol. 2008;10:361-367.
55. Clarke JL, Chang S. Pseudoprogression and pseudoresponse: challenges in brain tumor imaging. Curr Neurol Neurosci Rep. 2009;9:241-246.
56. Hu LS, Eschbacher JM, Heiserman JE, et al. Reevaluating the imaging definition of tumor progression: perfusion MRI quantifies recurrent glioblastoma tumor
fraction, pseudoprogression, and radiation necrosis to predict survival. Neuro-oncol. 2012;14:919-930.
57. Prah MA, Al-Gizawiy MM, Mueller WM, et al. Spatial discrimination of glioblastoma and treatment effect with histologically-validated perfusion and diffusion
magnetic resonance imaging metrics. J Neurooncol. 2018;136:13-21.
58. Ellingson BM, Malkin MG, Rand SD, et al. Validation of functional diffusion maps (fDMs) as a biomarker for human glioma cellularity. J Magn Reson Imaging.
2010;31:538-548.
59. Sottoriva A, Spiteri I, Piccirillo SG, et al. Intratumor heterogeneity in human glioblastoma reflects cancer evolutionary dynamics. Proc Natl Acad Sci USA. 2013;
110:4009-4014.
60. Stadlbauer A, Ganslandt O, Buslei R, et al. Gliomas: histopathologic evaluation of changes in directionality and magnitude of water diffusion at diffusion-tensor
MR imaging. Radiology. 2006;240:803-810.
61. LaViolette PS, Mickevicius NJ, Cochran EJ, et al. Precise ex vivo histological validation of heightened cellularity and diffusion-restricted necrosis in regions of
dark apparent diffusion coefficient in 7 cases of high-grade glioma. Neuro-oncol. 2014;16:1599-1606.
62. Maia AC Jr., Malheiros SM, da Rocha AJ, et al. MR cerebral blood volume maps correlated with vascular endothelial growth factor expression and tumor grade in
nonenhancing gliomas. AJNR Am J Neuroradiol. 2005;26:777-783.
63. Shiroishi MS, Boxerman JL, Pope WB. Physiologic MRI for assessment of response to therapy and prognosis in glioblastoma. Neuro-oncol. 2016;18:467-478.
64. Barajas RF Jr., Chang JS, Segal MR, et al. Differentiation of recurrent glioblastoma multiforme from radiation necrosis after external beam radiation therapy with
dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging. Radiology. 2009;253:486-496.
65. Hu LS, Baxter LC, Smith KA, et al. Relative cerebral blood volume values to differentiate high-grade glioma recurrence from posttreatment radiation effect: direct
correlation between image-guided tissue histopathology and localized dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging mea-
surements. AJNR Am J Neuroradiol. 2009;30:552-558.
66. Patel P, Baradaran H, Delgado D, et al. MR perfusion-weighted imaging in the evaluation of high-grade gliomas after treatment: a systematic review and meta-
analysis. Neuro-oncol. 2017;19:118-127.
67. Fatterpekar GM, Galheigo D, Narayana A, et al. Treatment-related change versus tumor recurrence in high-grade gliomas: a diagnostic conundrum--use of
dynamic susceptibility contrast-enhanced (DSC) perfusion MRI. AJR Am J Roentgenol. 2012;198:19-26.
68. Welker K, Boxerman J, Kalnin A, et al; American Society of Functional Neuroradiology MR Perfusion Standards and Practice Subcommittee of the ASFNR
Clinical Practice Committee. ASFNR recommendations for clinical performance of MR dynamic susceptibility contrast perfusion imaging of the brain. AJNR Am
J Neuroradiol. 2015;36:E41-E51.
69. Sugahara T, Korogi Y, Tomiguchi S, et al. Posttherapeutic intraaxial brain tumor: the value of perfusion-sensitive contrast-enhanced MR imaging for dif-
ferentiating tumor recurrence from nonneoplastic contrast-enhancing tissue. AJNR Am J Neuroradiol. 2000;21:901-909.
70. Hu LS, Yoon H, Eschbacher JM, et al. Accurate patient-specific machine learning models of glioblastoma invasion using transfer learning. AJNR Am
J Neuroradiol. 2019;40:418-425.
71. Hu LS, Ning S, Eschbacher JM, et al. Radiogenomics to characterize regional genetic heterogeneity in glioblastoma. Neuro-oncol. 2017;19:128-137.

72. Barajas RF Jr., Hodgson JG, Chang JS, et al. Glioblastoma multiforme regional genetic and cellular expression patterns: influence on anatomic and physiologic
MR imaging. Radiology. 2010;254:564-576.
73. Hu LS, Eschbacher JM, Dueck AC, et al. Correlations between perfusion MR imaging cerebral blood volume, microvessel quantification, and clinical outcome
using stereotactic analysis in recurrent high-grade glioma. AJNR Am J Neuroradiol. 2012;33:69-76.
74. Barajas RF Jr., Phillips JJ, Parvataneni R, et al. Regional variation in histopathologic features of tumor specimens from treatment-naive glioblastoma correlates
with anatomic and physiologic MR Imaging. Neuro-oncol. 2012;14:942-954.
75. Bai RY, Staedtke V, Riggins GJ. Molecular targeting of glioblastoma: drug discovery and therapies. Trends Mol Med. 2011;17:301-312.
76. Ene CI, Fine HA. Many tumors in one: a daunting therapeutic prospect. Cancer Cell. 2011;20:695-697.
77. Marusyk A, Almendro V, Polyak K. Intra-tumour heterogeneity: a looking glass for cancer? Nat Rev Cancer. 2012;12:323-334.
78. Gutman DA, Cooper LA, Hwang SN, et al. MR imaging predictors of molecular profile and survival: multi-institutional study of the TCGA glioblastoma data set.
Radiology. 2013;267:560-569.
79. Jain R, Poisson LM, Gutman D, et al. Outcome prediction in patients with glioblastoma by using imaging, clinical, and genomic biomarkers: focus on the
nonenhancing component of the tumor. Radiology. 2014;272:484-493.
80. Itakura H, Achrol AS, Mitchell LA, et al. Magnetic resonance image features identify glioblastoma phenotypic subtypes with distinct molecular pathway activities.
Sci Transl Med. 2015;7:303ra138.
81. Tykocinski ES, Grant RA, Kapoor GS, et al. Use of magnetic perfusion-weighted imaging to determine epidermal growth factor receptor variant III expression in
glioblastoma. Neuro-oncol. 2012;14:613-623.
82. Pope WB, Chen JH, Dong J, et al. Relationship between gene expression and enhancement in glioblastoma multiforme: exploratory DNA microarray analysis.
Radiology. 2008;249:268-277.
83. Gupta A, Young RJ, Shah AD, et al. Pretreatment dynamic susceptibility contrast MRI perfusion in glioblastoma: prediction of eGFR gene amplification. Clin
Neuroradiol. 2015;25:143-150.
84. Ryoo I, Choi SH, Kim JH, et al. Cerebral blood volume calculated by dynamic susceptibility contrast-enhanced perfusion MR imaging: preliminary correlation
study with glioblastoma genetic profiles. PLoS One. 2013;8:e71704.
85. Aghi M, Gaviani P, Henson JW, et al. Magnetic resonance imaging characteristics predict epidermal growth factor receptor amplification status in glioblastoma.
Clin Cancer Res. 2005;11:8600-8605.
86. Gaw N, Hawkins-Daarud A, Hu LS, et al. Integration of machine learning and mechanistic models accurately predicts variation in cell density of glioblastoma
using multiparametric MRI. Sci Rep. 2019;9:10063.
87. Hawkins-Daarud A, Johnston SK, Swanson KR. Quantifying uncertainty and robustness in a biomathematical model-based patient-specific response metric for
glioblastoma. JCO Clin Cancer Inform. 2019;3:1-8.
88. Kirson ED, Dbalý V, Tovarys F, et al. Alternating electric fields arrest cell proliferation in animal tumor models and human brain tumors. Proc Natl Acad Sci USA.
2007;104:10152-10157.
89. Kirson ED, Gurvich Z, Schneiderman R, et al. Disruption of cancer cell replication by alternating electric fields. Cancer Res. 2004;64:3288-3295.
90. Kim EH, Song HS, Yoo SH, et al. Tumor treating fields inhibit glioblastoma cell migration, invasion and angiogenesis. Oncotarget. 2016;7:65125-65136.
91. Chang E, Patel CB, Pohling C, et al. Tumor treating fields increases membrane permeability in glioblastoma cells. Cell Death Discov. 2018;4:113.
92. Trusheim J, Dunbar E, Battiste J, et al. A state-of-the-art review and guidelines for tumor treating fields treatment planning and patient follow-up in glioblastoma.
CNS Oncol. 2017;6:29-43.
93. Chaudhry A, Benson L, Varshaver M, et al. NovoTTFTM-100A System (Tumor Treating Fields) transducer array layout planning for glioblastoma: a NovoTALTM
system user study. World J Surg Oncol. 2015;13:316.
94. Stupp R, Wong ET, Kanner AA, et al. NovoTTF-100A versus physician’s choice chemotherapy in recurrent glioblastoma: a randomised phase III trial of a novel
treatment modality. Eur J Cancer. 2012;48:2192-2202.
95. Stupp R, Taillibert S, Kanner A, et al. Effect of tumor-treating fields plus maintenance temozolomide vs maintenance temozolomide alone on survival in patients
with glioblastoma: a randomized clinical trial. JAMA. 2017;318:2306-2316.
96. Toms SA, Kim CY, Nicholas G, et al. Increased compliance with tumor treating fields therapy is prognostic for improved survival in the treatment of glioblastoma:
a subgroup analysis of the EF-14 phase III trial. J Neurooncol. 2019;141:467-473.
97. Benson L. Tumor treating fields technology: alternating electric field therapy for the treatment of solid tumors. Semin Oncol Nurs. 2018;34:137-150.
98. Louis DN, Perry A, Reifenberger G, et al. The 2016 World Health Organization Classification of Tumors of the Central Nervous System: a summary. Acta
Neuropathol. 2016;131:803-820.
99. De Mattos-Arruda L, Mayor R, Ng CKY, et al. Cerebrospinal fluid-derived circulating tumour DNA better represents the genomic alterations of brain tumours than
plasma. Nat Commun. 2015;6:8839.
100. Miller AM, Shah RH, Pentsova EI, et al. Tracking tumour evolution in glioma through liquid biopsies of cerebrospinal fluid. Nature. 2019;565:654-658.
101. Wang Y, Springer S, Zhang M, et al. Detection of tumor-derived DNA in cerebrospinal fluid of patients with primary tumors of the brain and spinal cord. Proc Natl
Acad Sci USA. 2015;112:9704-9709.
102. Rogawski DS, Vitanza NA, Gauthier AC, et al. Integrating RNA sequencing into neuro-oncology practice. Transl Res. 2017;189:93-104.

Hu et al
103. Saltz J, Gupta R, Hou L, et al. Spatial organization and molecular correlation of tumor-infiltrating lymphocytes using deep learning on pathology images. Cell Rep.
2018;23:181-193.e7.
104. Pinton L, Masetto E, Vettore M, et al. The immune suppressive microenvironment of human gliomas depends on the accumulation of bone marrow-derived
macrophages in the center of the lesion. J Immunother Cancer. 2019;7:58.
105. Heindl A, Nawaz S, Yuan Y. Mapping spatial heterogeneity in the tumor microenvironment: a new era for digital pathology. Lab Invest. 2015;95:377-384.
106. Korshunov A, Sahm F, Zheludkova O, et al. DNA methylation profiling is a method of choice for molecular verification of pediatric WNT-activated medul-
loblastomas. Neuro-oncol. 2019;21:214-221.
107. Mathios D, Hwang T, Xia Y, et al. Genome-wide investigation of intragenic DNA methylation identifies ZMIZ1 gene as a prognostic marker in glioblastoma and
multiple cancer types. Int J Cancer. 2019;145:3425-3435.

Developmental
Therapeutics—
Immunotherapy

Best Foot Forward: Neoadjuvant Systemic

Therapy as Standard of Care in Triple-Negative
and HER2-Positive Breast Cancer
Priyanka Sharma, MD1; Roisin M. Connolly, MBBCh, MD2; Evanthia T. Roussos Torres, MD, PhD3; and Alastair Thompson, MD4
overview
Neoadjuvant systemic treatment of early-stage breast cancer has been used to improve resectability and
reduce the extent of breast and axillary surgery. More recently, several other merits of neoadjuvant systemic
treatment have emerged, including the ability to tailor clinically available adjuvant systemic therapy options
based on pathologic response and to serve as a platform for early assessment of novel agents and response
biomarkers and as an avenue for treatment optimization investigations (local and systemic therapy escalation
and de-escalation trials guided by pathologic response). Attainment of a pathologic complete response (pCR)
is associated with excellent long-term outcomes; conversely, the presence of residual disease is associated
with a high risk of recurrence for patients with HER2-positive breast cancer and triple-negative breast cancer
(TNBC). Treatment strategies in early-stage HER2-positive breast cancer include regimens incorporating
trastuzumab, pertuzumab, ado-trastuzumab emtansine, and neratinib, resulting in high pCR rates and overall
excellent long-term outcomes. Currently available cytotoxic regimens yield pCR for 35% to 55% of patients
with TNBC, and immune checkpoint inhibition is showing early promise for this subtype. New drug and
predictive biomarker evaluations in the neoadjuvant setting aim to develop optimal treatment strategies for the
individual patient, with the ultimate goal of maximizing efficacy and minimizing toxicity. Research efforts
involving novel agents are being undertaken to address the high risk of recurrence for patients with residual
disease. Omission of breast surgery following neoadjuvant chemotherapy requires further development of
imaging and biopsy techniques to accurately assess the extent of residual disease before clinical application.
INTRODUCTION U.S. Food and Drug Administration (FDA) also sup-

Neoadjuvant treatment of breast cancer with chemo- ports the use of pCR as an endpoint for clinical testing
therapy, targeted therapy, and/or endocrine therapy of neoadjuvant treatment for patients with HER2-
has long been used to downstage disease, improve positive breast cancer and TNBC.6,7 Cytotoxic and tar-
geted systemic regimens result in pCR for up to 50% to
resectability, and potentially reduce the extent of breast
60% of patients with HER2-positve disease and 35% to
and axillary surgery.1,2 More recently, several other
55% of patients with TNBC.8-12 After neoadjuvant
merits of neoadjuvant systemic treatment have been
systemic therapy (NAST), surgical resection of the
recognized, specifically for HER2-positive and triple-
residual primary breast tumor and surgical staging of
negative breast cancer (TNBC). These include the
the axillary lymph nodes, followed by adjuvant radiation
ability to tailor clinically available adjuvant systemic
therapy (when indicated), is the standard of care.13
therapy options based on pathologic response and to
Author affiliations
serve as a platform for early assessment of novel THE CHANGING LANDSCAPE OF NEOADJUVANT
and support agents and response biomarkers (imaging, blood, THERAPY AND ITS IMPACT ON ADJUVANT THERAPY
information (if tissue-based) and as an avenue for treatment opti- FOR TNBC
applicable) appear mization investigations (local and systemic therapy
at the end of this
TNBC, which is defined by the lack of expression of
escalation and de-escalation trials guided by patho- estrogen receptor (ER) and progesterone receptor and
article.
Accepted on March
logic response). The most widely accepted definition of the absence of HER2 overexpression and/or gene
30, 2020 and pathologic complete response (pCR) includes absence amplification, accounts for 15% of all breast cancers
published at of residual invasive disease in the breast and the in the United States and is associated with poor long-
ascopubs.org on sampled axillary nodes (ypT0/is, ypN0). Patients who term outcomes compared with other breast cancer
April 21, 2020,
achieve a pCR have lower rates of systemic and local subtypes.14-20 Because of its molecular heterogeneity
10.1200/EDBK_ recurrence, and pCR predicts for excellent survival, and lack of defined targets, advancements that have
281381 regardless of subtype.3-5 Regulatory guidance from the changed the outlook of other breast cancer subtypes

Sharma et al
evaluated primarily in neoadjuvant trials (Table 1). Three

randomized trials have shown that adding carboplatin to
a backbone of neoadjuvant anthracycline/taxane regi-
• Complete pathologic response to neoadjuvant mens improves pCR in early-stage TNBC (absolute pCR
chemotherapy (plus HER2 targeted therapy
improvement of 10%–16%).10,27,31
when indicated) is associated with excellent
long-term outcomes in breast cancer. Although the addition of carboplatin to anthracycline/taxane
• Pathologic response guides adjuvant systemic chemotherapy results in higher pCR rates, this benefit may
therapy decisions for patients with HER2- come at the cost of increased toxicity and compromised
positive breast cancer and TNBC. ability to deliver all doses of weekly paclitaxel. Long-term
• Research efforts with novel agents are ongoing benefits of incorporating neoadjuvant carboplatin when
to address the high risk of recurrence for pa- added to anthracycline/taxane regimens remain unclear.
tients with residual disease following neo- Although the GeparSixto study reported an improvement in
adjuvant systemic therapy. disease-free survival (DFS) for patients who received neo-
• New drug and predictive biomarker evaluations adjuvant carboplatin (86% vs. 76%; HR, 0.56; p = .0325),
in the neoadjuvant setting aim to develop an there was no difference in 3-year event-free survival (EFS) in
optimal treatment strategy for individual pa- the CALGB 40603 trial (71% vs. 76%; HR, 0.84; p = .36).
tients, with the ultimate goal of maximizing ef- Notably, neither of these two trials were powered for EFS or
ficacy and minimizing toxicity associated with overall survival (OS) endpoints, and long-term outcomes
systemic therapy. from the BrighTNess trial have yet to be reported.
• Omission of surgery following neoadjuvant
chemotherapy requires further development of Anthracycline-sparing regimens Carboplatin also has re-
imaging and biopsy techniques to accurately cently been studied in combination with taxane (without
assess the extent of residual disease before anthracycline) in neoadjuvant regimens, with encouraging
clinical application. pCR rates ranging from 45% to 55% with 12 to 18 weeks of
treatment and a favorable toxicity profile (Table 1).11,32
Patients who achieved pCR with carboplatin/taxane regi-
(like HER2-directed therapy and antiestrogen therapy) have mens demonstrated an excellent 3-year recurrence-free
proved elusive for TNBC. Cytotoxic chemotherapy, which survival rate of 90% without adjuvant anthracyclines.32,33
reduces the risk of recurrence and death, remains the Single-agent paclitaxel or cisplatin treatment of 12 weeks
standard of care for patients with stage I (T . 1 cm) to stage has yielded a low pCR rate of , 15% in unselected TNBC.34
III TNBC and typically includes anthracycline/cyclophos- Neoadjuvant anthracycline-containing and anthracycline-
phamide plus taxane-based regimens.13,21 However, de- sparing regimens have not been directly compared in large
spite the administration of anthracycline plus taxane-based trials. A recent randomized phase II study demonstrated that
systemic chemotherapy, approximately 20% to 40% of pa- 18 weeks of carboplatin/docetaxel regimen yielded pCR rates
tients with early-stage TNBC develop metastatic disease,22-24 similar to the four-drug regimen of paclitaxel/carboplatin →
highlighting the need for improved therapeutic approaches for AC (52% vs. 54%) but with a more favorable toxicity profile.35
this subtype.
In the adjuvant setting, the ABC trials compared six cycles of
Neo/adjuvant Chemotherapy Regimens taxane/carboplatin (75 mg/m2 of docetaxel plus 600 mg/m2
In recent years, chemotherapy for early-stage TNBC has of cyclophosphamide) with various taxane/AC regimens
been increasingly applied in the neoadjuvant setting owing among 2,125 patients and demonstrated a statistically
to the reasons highlighted above. Meta-analyses of pro- significant improvement in 4-year invasive DFS with taxane/
spective trials have demonstrated that achievement of pCR AC compared with taxane/carboplatin (90.7% vs. 88.2%;
with NAST in TNBC is associated with excellent prognosis, p = .04) among patients with HER2-negative breast can-
with an approximately 90% likelihood of cure, and the cer.36 Subgroup analyses suggested that the benefit of
presence of residual disease predicts for a high (30%–40%) taxane/AC over taxane/carboplatin was more evident in
risk of recurrence.16,25,26 Standard doxorubicin/cyclophos- patients with TNBC and for those with node-positive dis-
phamide (AC) followed by taxane regimens yields pCR for ease, thus establishing taxane/AC as a standard adjuvant
30% to 44% of patients with TNBC.10,27,28 regimen for most patients with TNBC. The absolute 4-year
invasive DFS difference between taxane/AC and taxane/
Role of platinum agents Like anthracyclines, platinum carboplatin for patients with node-negative TNBC was
agents damage DNA and have shown synergistic activity modest (2.5%); thus, a nonanthracycline regimen may be
when given in combination with taxanes in preclinical used for patients with low disease burden and a contrain-
models.29,30 The efficacy of carboplatin in TNBC has been dication to anthracycline.

Neoadjuvant Systemic Therapy for Triple-Negative and HER2-Positive Breast Cancer
TABLE 1. Neoadjuvant Clinical Trials Incorporating Platinum in Triple-Negative Breast Cancer

pCRa (%)
Study Design Chemotherapy Regimen N Control Platinum

Anthracycline-Containing Regimens
GeparSixto31 Randomized 80 mg/m2 of paclitaxel weekly + 315 43 53
phase II 20 mg/m2 of NPLD weekly +
15 mg/kg of bevacizumab every
3 weeks 6 carboplatin AUC 1.5–2
weekly 18 weeks
Alliance/CALGB 4060310 Randomized 80 mg/m2 of paclitaxel weekly 12 6 433 41 54
phase II (2 2 carboplatin AUC 6 every 3 weeks
factorial design) 4 → AC every 2 weeks 4 6
10 mg/kg of bevacizumab every
2 weeks 9 cycles
BrighTNess27 Randomized Weekly paclitaxel 6 carboplatin AUC 634 31 57 (carboplatin)
phase III (3-arm) 6 6 50 mg of veliparib 12 weeks, 53 (carboplatin +
followed by AC every 2–3 weeks veliparib)
4 cycles
I-SPY 232 Randomized 80 mg/m2 of paclitaxel weekly 12 6 60 26 (estimated) 51 (estimated)
phase II carboplatin AUC 6 every 3 weeks 4
and 50 mg of veliparib twice a day
orally → AC every 2 weeks 4 cycles
NeoSTOP35 Randomized 80 mg/m2 of paclitaxel weekly 12 + 100 — 54 (arm A)
phase II carboplatin AUC 6 every 3 weeks
4 → AC every 2 weeks
4 (arm A)
Carboplatin AUC 6 + 75 mg/m2 of 52 (arm B)
docetaxel every 3 weeks 6 cycles
(arm B)
Anthracycline-Free Regimens
WSG-ADAPT32 Randomized 125 mg/m2 of nab-paclitaxel weekly + 336 29 45
phase II carboplatin AUC 2 or 1,000 mg/m2 of
gemcitabine on days 1 and 8 every
3 weeks 4 cycles
Sharma et al11 Single arm Carboplatin AUC 6 + 75 mg/m2 of 190 — 55
docetaxel every 3 weeks 6 cycles
TBCRC 03034 Randomized 80 mg/m2 of paclitaxel 12 weeks or 140 12 15
phase II 75 mg/m2 cisplatin every 3 weeks 4
Abbreviations: pCR, pathologic complete response; NPLD, nonpegylated liposomal doxorubicin; AUC, area under the curve; AC, doxorubicin /cyclophosphamide.
a
pCR is defined as ypT0/isN0.
Role of Germline BRCA Mutation Status in Neoadjuvant cyclophosphamide causes DNA crosslinks leading to
Chemotherapy Regimen Selection double-stranded breaks).
Germline (g) BRCA1/2 mutations are detected in 10% to A recently reported phase II neoadjuvant trial (TBCRC031)
20% of patients with early-stage TNBC; however, homol- for patients with gBRCA-associated HER2-negative breast
ogous recombination deficiency mediated by other mech- cancer reported similar pCR rates with cisplatin (18%) and
anisms has been demonstrated in 50% to 60% of AC (26%) among all patients and also among the 76 pa-
TNBC.37,38 Platinum compounds generate double-stranded tients with TNBC (23% vs. 29%).39 In comparison, neo-
breaks that are preferentially repaired by BRCA1- and adjuvant trials of longer chemotherapy regimens (20–24
BRCA2-mediated homologous recombination repair. It is weeks) that included anthracycline and taxane (with or
also noteworthy that commonly used neo/adjuvant che- without carboplatin) reported pCR rates of 40% to 66% for
motherapeutics for breast cancer include DNA-damaging gBRCA mutation carriers. Thus, for the majority of patients
drugs (anthracyclines induce double-stranded breaks, and with gBRCA-associated TNBC, 12 weeks of either AC or

Sharma et al
platinum monotherapy would not be considered adequate oncologists should consider these trials in appropriate
neo/adjuvant therapy, although the TBCRC031 results clinical settings.
suggest that platinum is a reasonable substitute for those
with a contraindication to anthracycline. Predictive Biomarkers of Response to NAST in TNBC
Subgroup analysis of trials has shed light on the topic of The availability of predictive markers to identify patients with
preferential carboplatin benefit for patients with BRCA wild- TNBC who are likely to have excellent pathologic response
type TNBC. Post hoc analysis of GeparSixto demonstrated with NAST will immensely aid treatment personalization.
that BRCA mutation carriers experienced a high pCR rate Multiple studies have demonstrated the prognostic value
(66.7%) with the anthracycline/taxane/bevacizumab regi- of tumor-infiltrating lymphocytes (TILs) in TNBC.41-44 The
men and that this rate was not increased further by the association between increasing TILs in pretreatment tumor
addition of carboplatin.40 In contrast, the addition of car- tissue and higher pCR has been observed with different
boplatin increased pCR from 36% to 55% for patients with NAST chemotherapy regimens across several trials and
BRCA wild-type TNBC. In the BrighTNess trial, carboplatin appears to be independent of the type or duration of
was associated with a larger absolute pCR improvement NAST.12,32,41,45,46 A recent pooled analysis of nine adjuvant
among patients with BRCA wild-type disease (absolute clinical trials (including 2,148 patients with TNBC) con-
increase of 30%) compared with gBRCA mutation carriers firmed the favorable prognostic role of TILs for patients with
(absolute increase of 9%).27 early-stage TNBC.43 TILs can be easily assessed on stan-
dard hematoxylin and eosin slides; with the availability of
Taken together, the findings from the TBCRC031, Gepar-
standardized methodology for TIL scoring and reporting, this
Sixto, and BrighTNess trials suggest that the substantial
is a very attractive candidate for patient selection for sys-
homologous recombination deficiency induced by gBRCA1/
temic therapy optimization (de-escalation or escalation)
2 mutation in TNBC may be therapeutically responsive to
approaches. Transcriptional profiling has identified several
most types of DNA-damaging chemotherapy, including AC,
potential multigene expression signatures that can predict
and that this responsiveness does not improve further by
response to anthracycline with or without taxane chemo-
addition of a platinum. However, the homologous re-
therapy in TNBC.47-50 A homologous recombination de-
combination deficiency brought about by mechanisms
ficiency genomic instability assay has also been shown to be
other than gBRCA mutation may not be as robust and could
associated with NAST response, although it does not pre-
respond to targeting by platinum agents, even in the
dict preferential response to specific chemotherapy
presence of anthracyclines. Further translational research is
regimens.34,51 No biomarkers are yet routinely available in
needed to understand mechanisms of homologous re-
the clinical setting to identify patients with TNBC who are
combination deficiency in BRCA wild-type TNBC and its
likely to have excellent pathologic response and outcome
relationship to platinum response.
with NAST, although measures to identify such markers are
In summary, multiagent neoadjuvant chemotherapy yields being pursued.
pCR for 30% to 55% of patients with TNBC. Even though
long-term gains are not clear, the individual patient benefit Emerging Role of Immune Checkpoint Inhibitor Therapy in
the Neoadjuvant Setting
from attainment of pCR may justify inclusion of neoadjuvant
platinum in the treatment armamentarium for patients with Given the success of immune checkpoint inhibitor therapy in
locally advanced TNBC. The benefit of neoadjuvant car- metastatic TNBC, several ongoing phase III trials are evalu-
boplatin (when used as an additive approach to anthra- ating the role of immune checkpoint blockade in TNBC when
cycline) seems to be more robust in BRCA wild-type TNBC. given concurrently with neoadjuvant chemotherapy.52,53
The decision to incorporate platinum into neoadjuvant Results from the KEYNOTE-522 study recently demon-
treatment of TNBC should be individualized, weighing the strated that the addition of pembrolizumab (an anti–PD-1
potential benefit against the likelihood of increased toxicity monoclonal antibody) to carboplatin/paclitaxel followed by
and treatment delays. The availability of biomarkers that can anthracycline-based chemotherapy significantly improved
identify a subset of patients most likely to benefit from pCR from 51.2% to 64.8% (p = .00055) for patients with
carboplatin will immensely simplify this decision; this topic is stage II or III TNBC.54,55 Interestingly, in the KEYNOTE-522
the focus of several ongoing and planned translational trial, greater benefit of pembrolizumab was noted for pa-
studies. Neoadjuvant carboplatin/taxane regimens provide tients with higher disease burden. The greatest magnitude
pCR rates comparable to the anthracycline/taxane/carbo- of benefit for addition of pembrolizumab was seen for pa-
platin regimens and should be considered for patients with tients with stage III disease (absolute pCR increase = 25%),
a contraindication to anthracyclines. Several ongoing ran- with lower absolute benefit for those with stage II disease
domized phase III trials are evaluating the definitive role of (absolute pCR increase = 7%–11%). More than 80% of
platinum in early-stage TNBC (Table 2), and patients and patients in the KEYNOTE-522 trial were considered to have

TABLE 2. Summary of Ongoing Phase III Neo/adjuvant Trials in Triple-Negative Breast Cancer
Trial Details Phase NCT Number
A Randomized Phase III Trial to Evaluate Efficacy and Safety of MK-3475 (Pembrolizumab) as Adjuvant Therapy for TNBC With III NCT02954874
1 cm Residual Invasive Cancer or Positive Lymph Nodes (ypN1mi, ypN1-3) After Neoadjuvant Chemotherapy (S1418/
BR006)
A Randomized Double-Blind Phase III Clinical Trial of Neoadjuvant Chemotherapy With Atezolizumab or Placebo in Patients III NCT03281954
With TNBC Followed by Adjuvant Atezolizumab or Placebo (NSABP B-59/GBG 96-GeparDouze)
Atezolizumab in Combination With Adjuvant Anthracycline/Taxane-Based Chemotherapy vs. Chemotherapy Alone in Patients III NCT03498716
With Operable Triple-Negative Breast Cancer (IMpassion030)
PARP Inhibitors and/or Platinum Agents
Adjuvant AC Followed by Paclitaxel 6 Carboplatin in Early-Stage TNBC (NRG-BR003) III NCT02488967
Adjuvant Treatment of EC Followed by Weekly Taxane 6 Carboplatin in Early-Stage TNBC (TCTN) III NCT02455141
Doxorubicin and Cyclophosphamide Followed by Taxane 6 Carboplatin as (Neo)adjuvant Therapy in Early-Stage TNBC III NCT02441933
(PEARLY)
Adjuvant Platinum vs. Capecitabine in Stage II-III TNBC Patients Who Have Basal-Like Residual Disease After Neoadjuvant III NCT02445391
Taxane 6 Anthracycline Chemotherapy (ECOG-ACRIN 1131)
Adjuvant Olaparib in Patients With Germline BRCA-Associated HER2-Negative Early-Stage Breast Cancer (OlympiA) III NCT02032823
Neoadjuvant Carboplatin and Paclitaxel Plus Olaparib in TNBC and/or Germline BRCA-Mutated HER2-Negative Breast Cancer II/III NCT03150576
(PARTNER)
Abbreviations: TNBC, triple-negative breast cancer; AC, doxorubicin/cyclophosphamide; EC, epirubicin/cyclophosphamide.
PD-L1–positive disease (based on a combined positive of these factors could be contributing to the different results
score 1 by the PD-L1 IHC 22C3 pharmDx assay; Agilent noted in these studies. Several ongoing phase III trials continue
Technologies, Santa Clara, CA); however, the benefit of to address the role of immune checkpoint inhibitors in early-
pembrolizumab was noted regardless of PD-L1 expression. stage TNBC (Table 2).
Patients with PD-L1–positive status achieved a high pCR
rate in both arms (pCR of 55% with chemotherapy and 68% Adjuvant Systemic Therapy in the Setting of
with pembrolizumab). The KEYNOTE-522 trial included Residual Disease
adjuvant pembrolizumab or placebo for up to 1 year for all Because patients with residual disease after neoadjuvant
patients, with EFS as a coprimary endpoint. At an early chemotherapy are at a high risk of recurrence, several
report with a median follow-up of 15.5 months, separation of systemic therapy approaches are being explored in this
the EFS curves was reported (EFS of 91% with pem- setting. The CREATE-X (Capecitabine for Residual Cancer
brolizumab compared with 85% with placebo; HR, 0.63), as Adjuvant Therapy) trial demonstrated the benefit of
although mature follow-up data are pending. capecitabine in this setting. CREATE-X randomly assigned
Another phase III trial (NeoTRIP) with a primary endpoint of 910 patients with residual HER2-negative invasive breast
EFS evaluated the addition of an anti–PD-L1 antibody, cancer after NAST (containing anthracycline, taxane, or
atezolizumab, to a 24-week carboplatin/nab-paclitaxel both) to receive standard postsurgical treatment with or
chemotherapy backbone for patients with locally ad- without capecitabine (1,250 mg/m2 twice daily on days
vanced TNBC.56 In NeoTRIP, the pCR rates were not 1–14 for six to eight cycles).57 Improvement in DFS and OS
significantly different between the two study arms (43.5% was noted with capecitabine, with the most pronounced
with atezolizumab vs. 40.8% with chemotherapy) in the impact for patients with TNBC, who accounted for 30% of
overall study population or among the 42% of patients the study population. Among patients with TNBC, the 5-year
with PD-L1–positive tumors. In the phase II GeparNuevo DFS was 69.8% for the capecitabine group compared with
study, the addition of durvalumab (PD-L1 inhibitor) to 56.1% for the control group (HR, 0.58), and OS was 78.8%
nab-paclitaxel followed by epirubicin/cyclophosphamide versus 70.3% (HR, 0.52). This dose schedule of capeci-
numerically increased the pCR rate from 44% to 53%, tabine was associated with expected toxicity, requiring dose
although this difference was not statistically significant. reductions and discontinuation for 37% to 25% of patients,
Compared with KEYNOTE-522, NeoTRIP and GeparNuevo respectively. In a recently presented meta-analysis of neo/
were much smaller studies and used different chemotherapy adjuvant trials involving more than 15,000 patients, the
backbones and immune checkpoint inhibitors (PD-L1 anti- benefit of capecitabine was noted in trials when capecita-
body vs. PD-1 antibody) and different schedules. One or many bine was added to (and not substituted for) standard

Sharma et al
TABLE 3. Proposed Neo/adjuvant Management for HER2-Positive Breast Cancer

Extended
Pathologic Adjuvant
Clinical Stage Initial Therapy Stage Adjuvant Therapy Therapy
T1N0 Surgery N0 T1: paclitaxel + H, then 1 year of H
T2: standard chemotherapy + H,
then 1 year of H
N1 Standard chemotherapy + H (6 P) Consider
then 1 year of H (6 P) neratinib if
no prior P
‡ T2 and/or ‡ N1 Neoadjuvant pCR cN0: 1 year of H
chemotherapy +
cN1: 1 year of H (6 P)
HP → surgery
No pCR T-DM1
Abbreviations: H, trastuzumab; P, pertuzumab; pCR, pathologic complete response; T-DM1, ado-trastuzumab emtansine.
chemotherapy, and it was primarily restricted to the TNBC approval in the late 1990s for patients with advanced HER2-
subtype, for which an 18% improvement in DFS and a 22% positive disease66; after four randomized adjuvant trials
improvement in OS was noted.58 Both National Compre- (NSABP B-31, NCCTG N9831, HERA, and BCIRG 006),
hensive Cancer Network and St. Gallen guidelines recom- trastuzumab was subsequently approved for patients with
mend consideration of adjuvant capecitabine in the setting early-stage disease.67-70 OS and DFS were improved with
of residual disease following neoadjuvant taxane/alkylator added trastuzumab compared with chemotherapy alone in
and anthracycline chemotherapy.13,59 Capecitabine is a meta-analysis of eight trials investigating the addition of
broadly available; given the poor prognosis of TNBC with trastuzumab to chemotherapy.71 Based on this success in
residual disease, individualized discussion with patients the adjuvant setting, neoadjuvant trials that combined
on benefit versus toxicity is warranted. trastuzumab with chemotherapy were designed. These
trials reported higher pCR rates, a surrogate for EFS,72 or
Several ongoing trials are exploring other approaches for
recurrence-free survival5,73,74 with the addition of trastu-
patients with TNBC who have residual disease. SWOG
zumab to chemotherapy versus chemotherapy alone. The
1418/BR006 (NCT02954874) is assessing the efficacy of
NOAH trial, for example, concluded that neoadjuvant
adjuvant pembrolizumab compared with observation for
trastuzumab significantly improved both pCR and EFS
patients with TNBC who have residual disease after neo-
compared with chemotherapy alone.75 These results led
adjuvant chemotherapy. EA1131 (NCT02445391) is com-
the way for a new paradigm in managing HER2-positive
paring adjuvant capecitabine versus platinum agents for
breast cancer.
patients with TNBC.
In an effort to further improve outcomes for patients, ad-
NEOADJUVANT AND ADJUVANT OPTIONS FOR HIGH-RISK ditional novel HER2-directed therapies have been de-
HER2-POSITIVE BREAST CANCER veloped. Pertuzumab (Perjeta; Genentech) targets the
HER2 is amplified or overexpressed in approximately 20% extracellular domain II of HER2, which results in ligand-
of all breast cancers.60 Over the last few decades, therapies dependent heterodimerization between HER2 and HER3.76
directed against HER2 have transformed the treatment Preclinical models found that the drug, when combined with
paradigm for this subtype of breast cancer such that long- trastuzumab (dual HER2-blockade), elicited complete
term outcomes are excellent for the majority of patients with blockade of signaling through the EGFR pathway77,78 and
this disease.61 HER2-directed therapies target members of overcame resistance to single-agent trastuzumab.79-81 After
the EGFR family, including HER2, and have become the initial success and regulatory approval in the metastatic
backbone of standard treatment regimens for patients with setting,76 pertuzumab was combined with chemotherapy
HER2-positive breast cancer. Table 3 outlines a proposed and trastuzumab in the neoadjuvant setting.
treatment paradigm for these patients, with the supporting
In the NeoSphere trial, patients treated with pertuzumab/
trials discussed in this section.
trastuzumab and docetaxel achieved a pCR rate of ap-
Trastuzumab (Herceptin; Genentech, South San Francisco, proximately 45% compared with patients who received
CA) was the first HER2-directed therapy approved in breast trastuzumab/docetaxel (29%). A high pCR rate of 63% was
cancer. This monoclonal antibody targets the extracellular observed in those with hormone receptor–negative disease
domain of HER2, preventing ligand-independent HER2 treated with pertuzumab/trastuzumab/docetaxel.82 Im-
signaling.62-65 Trastuzumab received initial regulatory pressive pCR rates were also observed in the TRYPHAENA

study when pertuzumab was added to a chemotherapy/ III KATHERINE trial recently reported a benefit in invasive
trastuzumab backbone.83 These results led to accelerated DFS with 14 cycles of T-DM1 in the post-neoadjuvant setting
FDA approval of pertuzumab in 2013 for neoadjuvant use compared with trastuzumab for patients with residual tumor
for patients with locally advanced or higher-risk early-stage (no pCR) in the breast or axilla following neoadjuvant
disease (patients with tumors . 2 cm or node positive). therapy with HER2-directed therapy and chemotherapy
Reassuringly, the combination of trastuzumab and pertu- (taxane with or without anthracycline). The HR for invasive
zumab was not found to be associated with increased disease or death was 0.5 (95% CI, 0.39–0.64; p , .001)
cardiac toxicity.84,85 Neutropenia, rash, and diarrhea are when comparing the T-DM1 group with the trastuzumab
more frequent with the addition of pertuzumab to group. Although there were more adverse events associated
chemotherapy-containing regimens but are manageable with T-DM1, the clinical benefit was felt to outweigh po-
with appropriate supportive care.86 Treatment regimens that tential risks, and regulatory approval was granted.
add pertuzumab to trastuzumab and chemotherapy are Finally, neratinib was approved in 2017 for use in the ex-
now considered standard of care internationally for patients tended adjuvant setting following the completion of 1 year of
with high-risk features who are candidates for neoadjuvant adjuvant trastuzumab. This agent is an irreversible tyrosine
therapy, pending drug availability. kinase inhibitor of EGFR, HER2, and HER4 and was also
Full FDA approval of pertuzumab in the early-stage setting recently approved by the FDA for use in the metastatic
was dependent on its success in ongoing, large, randomized setting in combination with capecitabine.92 The ExteNET
phase III trials; FDA approval ultimately was obtained in trial reported that 1 year of oral neratinib, administered
2017 on the basis of results of the APHINITY trial.87 within 2 years of completion of adjuvant trastuzumab, im-
APHINITY enrolled 4,800 patients with high-risk stage I proved 5-year DFS compared with placebo (90.2% vs.
to III HER2-positive breast cancer and randomly assigned 87.7%).93 Interestingly, there seemed to be a greater benefit
them to either standard chemotherapy (nonanthracycline or for patients with hormone receptor–positive breast cancer
anthracycline based) concurrent with pertuzumab/trastu- versus hormone receptor–negative disease, which may
zumab or standard chemotherapy concurrent with trastu- relate to bidirectional cross-talk between ER and HER2
zumab/placebo. A modest absolute improvement in 3-year receptor signaling.94 The most notable adverse event was
invasive DFS was observed when comparing the pertuzu- diarrhea, with 55% of patients experiencing grade 1 to 2,
mab arm with the standard arm (94.1% vs. 93.2%, re- 40% experiencing grade 3, and , 1% of patients experi-
spectively). Patients who appeared to benefit most were encing grade 4 diarrhea, largely occurring within the first
those with node-positive disease (92.0% vs. 90.2%) com- cycle of treatment. As such, diarrhea prophylaxis is rec-
pared with the node-negative population (97.5% vs. 98.4%), ommended from initiation of therapy, and this treatment is
further supporting the use of dual HER2-directed therapy for usually preferred for patients with high-risk disease (i.e., lymph
higher-risk patients. A second interim analysis reported that for node–positive disease). In addition, no prior pertuzumab (ei-
the node-positive population, 6-year follow-up of invasive DFS ther neoadjuvant or adjuvant) was received by enrolled pa-
was 87.9% for the pertuzumab-treated group versus 83.4% for tients, so it is unclear whether there is additional benefit for
the placebo-treated group (4.5% difference; 95% CI, 1.9%– these patients.
7.1%), with no benefit observed for the node-negative pop-
ulation.88 After 74.1 months of median follow-up, the 6-year OS Treatment Optimization in HER2-Positive Breast Cancer
was 94.8% for the pertuzumab group versus 93.9% for the With the rapid pace of development of HER2-directed
placebo group, a 0.9% difference (95% CI, 0.5%–2.2%) therapies, treatment strategies in early breast cancer
with an HR of 0.85 (95% CI, 0.67–1.07), but this was not have escalated to include regimens incorporating tras-
statistically significant. International guidelines consider tuzumab, pertuzumab, T-DM1, and neratinib, as de-
adjuvant pertuzumab as a treatment option for patients scribed above. These developments have resulted in high
with high-risk HER2-positive breast cancer (e.g., those pCR rates and overall excellent long-term outcomes.
with node-positive or hormone receptor–negative breast Following the success of the KATHERINE trial, however,
cancer).13,89 there is increasing interest in using the post-neoadjuvant
When considering escalation of treatment, the oncology space as a platform for new drug and biomarker devel-
community is considering post-neoadjuvant residual dis- opment.95 Investigation in this “higher-risk” space may
ease as an ideal setting in which to maximize efficacy include agents with recent success in the advanced
without overtreating the wider patient population.90 Ado- breast cancer setting.
trastuzumab emtansine (T-DM1; Kadcyla; Genentech) is an The addition of tucatinib, an oral tyrosine kinase inhibitor
antibody drug conjugate that exploits the HER2-directed that is highly selective for the HER2 kinase domain, to
activity of trastuzumab for targeted delivery of the antimitotic trastuzumab and capecitabine (phase III HER2CLIMB
chemotherapeutic drug emtansine (i.e., DM1).91 The phase study) has resulted in progression-free survival and OS

Sharma et al
advantages in advanced HER2-positive breast cancer As data continue to emerge regarding the optimal man-
previously treated with pertuzumab and T-DM1.96 Trastu- agement of early-stage HER2-positive breast cancer, robust
zumab deruxtecan (Enhertu, T-DXd, previously known as predictive biomarker development is essential. Tissue-
DS-8201; Daiichi Sankyo, Basking Ridge, NJ), an antibody based biomarkers, such as the HER2-enriched intrinsic
drug conjugate, has also demonstrated impressive early subtype, have been associated with a higher likelihood of
clinical activity in the phase II DESTINY-Breast 01 trial for pCR following combination of chemotherapy with HER2-
patients whose disease has progressed after T-DM1 in the directed therapies.113 These findings were reported in ret-
advanced breast cancer setting.97 Last, in the age of im- rospective analyses of the CHER-LOB,114 NeoALTTO,115
munotherapy, combinations of immune checkpoint agents CALGB40601,116 and NOAH117 trials. HER2-enriched
with trastuzumab have been shown to be synergistic in subtypes and ERBB2 mRNA might also predict pCR to
animal models of breast cancer,98,99 and early results of chemotherapy-free approaches, such as those in-
clinical trials with patients with metastatic disease are corporating lapatinib and trastuzumab.118 In addition, the
promising.100,101 It is likely that exciting new drugs such as presence of blood-based biomarkers, such as circulating
these will be incorporated early in the disease stage, but tumor DNA detection, prior to neoadjuvant HER2-
careful selection of a higher-risk population will be of utmost directed therapy has been associated with decreased
importance to avoid unwanted toxicities when possible. We pCR rates.119 Finally, imaging-based biomarkers are also
would anticipate that the role of so-called “liquid biopsies” under investigation as predictors of response, including
the use of fluorodeoxyglucose PET in the NeoALTTO,120
will be considered in such study designs in order to identify
TBCRC026,105 and PHERGain121 clinical trials.
the presence or emergence of minimal residual disease,
escalating therapy in those at the highest risk of relapse.102 Ultimately, novel study designs and incorporation of pre-
dictive biomarker investigation will aid in treatment decision-
Approaches to streamline the use of therapeutic combi-
making for patients with HER2-positive breast cancer. This
nations, minimize toxicities, and decrease treatment du-
approach will ensure that the optimal treatment strategy is
ration are worth highlighting. For example, the single-arm
chosen for a particular patient, aiming to maximize efficacy
APT trial suggests that patients with small, node-negative,
and minimize toxicity for the individual.
HER2-positive breast cancers have excellent long-term
outcomes with the use of adjuvant paclitaxel and trastu- LESS CAN BE MORE: SURGICAL CONSIDERATIONS IN A
zumab (Table 3), avoiding combination chemotherapy.103 NEOADJUVANT WORLD
The follow-on ATEMPT trial is investigating T-DM1 mono- Following NAST, consideration must be given to locore-
therapy compared with paclitaxel/trastuzumab for the ad- gional therapy, and particularly to what surgery may be
juvant treatment of stage I HER2-positive breast cancer, required for the breast and axilla. Historically, neoadjuvant
with encouraging results reported at the 2019 San Antonio therapy has been used to render inoperable disease op-
Breast Cancer Symposium.104 Along these lines, the single- erable, to reduce the extent of breast surgery from mas-
arm, multicenter CompassHER2 trial has recently activated tectomy to breast conservation approaches, and, most
and is investigating 3-year recurrence-free survival among recently, to consider whether less axillary resection may be
patients with stage II/III, HER2-positive breast cancer performed, particularly using sampling and/or targeted
treated with preoperative taxane, trastuzumab, and pertu- axillary approaches. This move has been to reduce the
zumab (NCT04266249). impact of surgery and surgical/radiation therapy morbidity
Whether a subgroup of patients could be treated with HER2- while securing excellent long-term outcomes.
directed therapy alone is also under investigation. Sub- A controversial element of the treatment of patients un-
stantial pCR rates have been observed with pertuzumab and dergoing neoadjuvant therapy is whether all patients require
trastuzumab (without chemotherapy) in the NeoSphere82 surgery. In the context of systemic therapy that downstages
and TBCRC026105 trials and with trastuzumab and lapatinib or eliminates malignant disease, omission of one or more
in the SOLTI-PAMELA and TBCRC-006/023 trials.106,107 local therapeutic modalities can be considered. Avoiding
Results from NeoSphere, for example, reported a 27% surgery for responsive disease has become attractive for
pCR rate following treatment with pertuzumab/trastuzumab certain cancer sites, such as anorectal malignancy.122 In
alone for patients with ER-negative HER2-positive breast this setting, establishing locoregional control with a combi-
cancer. For patients with ER-positive disease, ongoing trials nation of drug therapy and radiation, but avoiding surgery,
are investigating the role of dual HER2-directed therapy has resulted in sustained excellent outcomes, including
when combined with CDK4/6 inhibition in an effort to in- maintained organ function.123,124 Although the concept of
crease pCR rates further in this subgroup.108-112 Innovative surgical de-escalation is established for some cancers, there
clinical trial design will be required to move these interesting has been a cautious approach for breast cancer, given the
results forward and will require careful patient selection. potential for excellent long-term outcomes with conventional

treatment. Nevertheless, given the exceptional responses of analysis of 98 evaluable patients demonstrated that 36 had
certain breast cancer subtypes to NAST, the need for either invasive or ductal carcinoma in situ residual disease,
surgical resection of the original primary site following which surpassed the required number of non-pCRs for the
a complete response to NAST has long been questioned.125 primary analysis (the original trial was for 175 women). The
The omission of surgery after NAST in breast cancer may be NPV of 77.5% (95% CI, 66.8%–86.1%) and sensitivity of
desirable in terms of reducing the often-underestimated 50% (95% CI, 32.9%–67.1%) led to discontinuation of the
postoperative pain and distress.126,127 However, omission trial. Comparison of the findings from biopsy versus final
of surgery requires ascertainment that patients have no surgery considering purely residual invasive disease came
remaining disease in the breast without surgical resection, close to the original stated aim of an NPV of 90% or greater,
hence not compromising locoregional disease control. with an NPV of 89.2% (CI, 80.4%–94.9%) and sensitivity of
60.9% (CI, 38.5%–80.3%). Of interest, in the HER2-
Imaging of the Breast positive subset (44 patients), an NPV of 89.5% (CI, 75.2%–
Pre- and post-NAST imaging are both essential in a trial of 97.1%) and a sensitivity of 60% (CI, 26.2%–87.8%) were
surgical omission and serve different purposes. Prior to identified. The study concluded, based on this interim and
NAST, imaging evaluation of the tumor bed and regional now final analysis, that omitting surgery based on the criteria
nodal basins should include mammography, breast ultra- of clinical response and negative tumor bed biopsies was not
sound, and, perhaps, breast MRI with marker placement achieved.132
prior to commencing therapy. Imaging findings suspicious Other large studies to examine diagnosing complete re-
for multicentric or multifocal disease require evaluation with sponse in the breast by biopsy following NAST have been
core needle biopsy and marker placement prior to NAST. conducted and were also reported at the 2019 San Antonio
Post-NAST imaging is required to identify patients with clear Breast Cancer Symposium. The multicenter German Breast
evidence of residual disease, as they would not be candi- Group RESPONDER trial133 used minimally invasive, image-
dates for omission of surgery. Comparison of pre- and post- guided, vacuum-assisted core biopsy (ultrasound guided,
NAST imaging may also be useful for subsequent breast large bore, median of 7 biopsies) to detect residual tumor
radiation therapy planning if surgery is not performed. after NAST. The primary endpoint was a false-negative rate
However, data suggest that imaging alone is likely in- (FNR) of 10% or less, aiming for 595 patients, but again the
sufficient to identify patients with no residual disease after trial was stopped for futility after a planned interim analysis
NAST. For example, the NRG BR005 trial mandates tri-
of 398 women. There were false-negative biopsies for 37
modality imaging (mammography, ultrasound, and MRI)
women, with an FNR of 17.8% (95% CI, 12.8%–23.7%).
following neoadjuvant chemotherapy to assess the tumor
One-third had residual ductal carcinoma in situ alone (also
bed; assessment of these data may identify criteria pre-
a finding in the NRG BR005 trial), 54% had a less than
dictive of pCR that could significantly improve sensitivity and
5 mm minimal residual tumor, and 76% had residual in-
specificity.128,129 Further refinements of imaging, such as
vasive cancer cell content of 10% or less; each, in theory, is
shearwave ultrasound130 or MRI texture analyses,131 al-
amenable to radiotherapy treatment.133 The use of seven-
though currently research tools, may in the future help
gauge needles (41 patients) had an FNR of 0%, and those
better predict which patients will go on to attain a pCR.
with no tumor on vacuum-assisted biopsy (of whatever
Given the international interest in the concept of de- gauge) and no tumor visible in breast imaging after NAST
escalation of surgery following successful NAST, particu- had an FNR of 6.2% (95% CI, 3.4%–10.5%; 398 patients).
larly for HER2-positive breast cancer or TNBC, several trials
Other combined data from the United States, Korea, and the
of differing design have been conducted or are underway. In
United Kingdom134 on 159 women of all receptor subtypes
the United States, three national, multicenter clinical trials
demonstrated an FNR of 11.8% (CI, 0%–23.5%) and NPV
aim to define the accuracy of imaging, tissue sampling, and
of 93.1% (CI, 83.9%–100%). Planned subgroup analysis
liquid biopsy in assessing residual burden of disease post-
concluded that a standardized image-guided protocol of six
NAST and to explore patient and health care professionals’
or more vacuum-assisted biopsies of a residual imaging
views regarding nonsurgical management in this setting.
abnormality of 2 cm or less might predict residual cancer
NRG BR005 (NCT03188393) was a feasibility study to
with an FNR of 5% or less.
assess whether image-directed core needle biopsies of the
tumor bed can accurately predict a pCR for patients with In terms of other approaches, the PREDICT DNA trial
clinical/radiologic response after NAST (by mammography, (Pathologic Response & Evaluation in Circulating Tumor
ultrasound, and MRI).132 NRG BR005 sought to demon- DNA; NCT02743910) is prospectively measuring residual
strate a negative predictive value (NPV) of 90% or greater for microscopic tumor burden by assessing circulating plasma
core needle biopsy following NAST. Including ER-positive as tumor DNA in blood samples from 229 patients with HER2-
well as HER2-positive breast cancer and TNBC, the interim positive breast cancer or TNBC to establish whether the

Sharma et al
absence of plasma tumor DNA after NAST can predict pCR. positive nodes. In these studies, there was no difference in
NRG CC006, to complement NRG BR005, will explore DFS or local recurrence, respectively, between those un-
attitudes and perceptions regarding a nonsurgical approach dergoing axillary radiotherapy versus axillary lymph node
to breast cancer therapy and identify potential barriers and dissection (ALND) in the AMAROS trial or no ALND versus
facilitators of accrual to a future randomized trial of this ALND in the Z0011 trial. Importantly, however, the AMAROS
approach. The mixed-methods design of NRG CC006 and Z0011 patient populations received adjuvant systemic
comprises a qualitative exploratory study of the percep- therapy, not NAST. Although radiation therapy could be
tions and attitudes of physicians and patients registered sufficient to control disease for patients who achieve a breast
in NRG BR005, which will be followed by a survey of pCR, residual disease after NAST is likely chemotherapy
medical oncologists, surgeons, and radiation oncologists. resistant, for which patients will have already received most
of their main systemic agents. This issue of whether ra-
Axillary Management
diotherapy may be sufficient to control axillary disease rather
Appropriate management of the axilla is crucial to the than requiring ALND in this setting is the primary question in
success of a surgical de-escalation strategy. The optimal the Alliance A011202 trial.
approach differs according to the initial size and nodal
Currently, surgical staging of the axilla is recommended for
status of the primary tumor. Currently, axillary nodal man-
patients with initial cN1 disease139,140 even if an excellent
agement for patients treated with NAST is based on pre-
response to NAST is detected in the breast and the axilla.
treatment clinical evaluation and imaging. Patients with
NRG BR005 will collect data and images from preoperative
clinically and radiographically normal nodes or negative
axillary nodal staging and will provide additional information
core needle biopsy are considered cN0. Patients with cT1-2
on whether a negative imaging study of the axilla following
cN0 TNBC or HER2-positive disease who achieve a pCR in
NAST is sufficient to determine the status of patients who
the breast after NAST have a residual positive axillary node
originally had cN1 disease. Should imaging in this setting
rate of less than 2%.135,136 Thus, it might be reasonable to
prove to be a robust method to assess the axilla after NAST,
consider avoiding surgical staging of the axilla altogether in
it could support the option of omitting axillary surgical
this group, provided that these patients can be reliably
staging in those who achieve a breast pCR in the setting of
identified prior to surgery.
a clinical trial, given the high likelihood of response in the
Patients with biopsy-proven axillary disease at diagnosis axillary nodes.135,136,141 Although omitting axillary staging in
(cN1 disease) may convert to ypN0 after NAST. On axillary this setting would avoid the morbidity associated with
ultrasound after NAST, longer short-axis diameter, longer sentinel lymph node biopsy, there remains potential for
long-axis diameter, increased cortical thickness, and ab- higher axillary recurrence rates. In addition, the loss of
sence of a fatty hilum are significantly associated with information regarding residual disease in the axilla after
persistent malignancy. However, in the Z1071 trial, 56% of NAST may compromise systemic management decisions.
patients with sonographically normal lymph nodes after In the KATHERINE and CREATE-X trials, patients with re-
NAST still had residual nodal disease.137 Hence, the tar- sidual HER2-positive breast cancer and TNBC, respectively,
geted axillary node approach is under development for have been shown to benefit from additional systemic
those with cN1 disease at diagnosis whose disease retreatment.57,142 Patients with residual disease after NAST
sponds well to NAST.138 Placement of a marker in the in- could also participate in clinical trials testing novel agents.
volved node at diagnosis (just like in the original primary)
For patients with clinically N2 disease (palpable matted
with resection of that node at the time of sentinel lymph
nodes) at diagnosis, ALND after NAST remains the standard
node surgery to examine the efficacy of the NAST may
of care; such patients are less likely to achieve a pCR in the
confirm conversion of node-positive to node-negative axil-
axillary nodes,136 so they would not be candidates for a trial
lary disease, noting that the marked node may not be one of
omitting axillary surgery, even if they demonstrate an ex-
the mapped sentinel lymph nodes for approximately 25% of
cellent response to NAST.
patients.138
Among patients with cT1-2 N1 TNBC or HER2-positive CONCLUSION
disease who achieve a pCR in the breast after NAST, Complete pathologic response to neoadjuvant chemother-
11% to 14% have residual positive nodal disease.135,136 De- apy (with HER2-directed therapy when appropriate) is as-
escalation of axillary surgery following NAST for such pa- sociated with excellent long-term outcomes in HER2-
tients may be informed to some extent by data from the positive breast cancer and TNBC. For patients with early-
AMAROS and Z0011 trials. In the AMAROS trial, 33% of stage TNBC and HER2-positive breast cancer, once the
patients with a positive sentinel node(s) had additional nodal decision to give chemotherapy has been made (according
involvement, and 8% had four or more additional positive to clinical characteristics), administration of this therapy in
nodes. In the Z0011 trial, 13.7% had four or more additional the neoadjuvant setting represents the new standard for

most patients. Pathologic response guides adjuvant sys- aiming to maximize efficacy and minimize toxicity for the
temic therapy decisions for patients with HER2-positive individual. Research efforts with novel agents are ongoing to
breast cancer and TNBC. Novel study designs and in- address the high risk of recurrence for patients with residual
corporation of predictive biomarker investigation will aid in disease following NAST. Omission of surgery following
treatment decision-making for patients with HER2-positive neoadjuvant chemotherapy requires further development of
breast cancer and TNBC. This approach will ensure that the imaging and biopsy techniques to accurately assess the
optimal treatment strategy is chosen for a particular patient, extent of residual disease before clinical application.

1
Division of Medical Oncology, Department of Internal Medicine, Priyanka Sharma, MD, Division of Medical Oncology, Department of
University of Kansas Medical Center, Westwood, KS Internal Medicine, University of Kansas Medical Center, 2330 Shawnee
2
Cancer Research, University College Cork, Cork, Ireland Mission Pkwy., Ste. 210, MS 5003, Westwood, KS 66205; email:
3
Norris Comprehensive Cancer Center, University of Southern California, [email protected].
Los Angeles, CA
4
Department of Surgery, Dan L. Duncan Comprehensive Cancer Center,
Baylor College of Medicine, Houston, TX
EDBK_281381.
REFERENCES
1. Fisher B, Brown A, Mamounas E, et al. Effect of preoperative chemotherapy on local-regional disease in women with operable breast cancer: findings from
National Surgical Adjuvant Breast and Bowel Project B-18. J Clin Oncol. 1997;15:2483-2493.
2. Connor CS, Kimler BF, Mammen JMV, et al. Impact of neoadjuvant chemotherapy on axillary nodal involvement in patients with clinically node negative triple
negative breast cancer. J Surg Oncol. 2015;111:198-202.
3. Mamounas EP, Anderson SJ, Dignam JJ, et al. Predictors of locoregional recurrence after neoadjuvant chemotherapy: results from combined analysis of
National Surgical Adjuvant Breast and Bowel Project B-18 and B-27. J Clin Oncol. 2012;30:3960-3966.
4. Symmans WF, Wei C, Gould R, et al. Long-term prognostic risk after neoadjuvant chemotherapy associated with residual cancer burden and breast cancer
subtype. J Clin Oncol. 2017;35:1049-1060.
5. Cortazar P, Zhang L, Untch M, et al. Pathological complete response and long-term clinical benefit in breast cancer: the CTNeoBC pooled analysis. Lancet.
2014;384:164-172.
6. Center for Drug Evaluation and Research. Guidance for Industry: Pathological Complete Response in Neoadjuvant Treatment of High-Risk Early-Stage Breast
Cancer: Use as an Endpoint to Support Accelerated Approval. Silver Spring, MD: U.S. Food and Drug Administration; 2014.
7. European Medicines Agency. The Role of the Pathological Complete Response as an Endpoint in Neoadjuvant Breast Cancer Studies. Amsterdam, Neth-
erlands: European Medicines Agency; 2014.
8. Hurvitz SA, Martin M, Jung KH, et al. Neoadjuvant trastuzumab emtansine and pertuzumab in human epidermal growth factor receptor 2-positive breast
cancer: three-year outcomes from the phase III KRISTINE study. J Clin Oncol. 2019;37:2206-2216.
9. van Ramshorst MS, van der Voort A, van Werkhoven ED, et al; Dutch Breast Cancer Research Group (BOOG). Neoadjuvant chemotherapy with or without
anthracyclines in the presence of dual HER2 blockade for HER2-positive breast cancer (TRAIN-2): a multicentre, open-label, randomised, phase 3 trial. Lancet
Oncol. 2018;19:1630-1640.
10. Sikov WM, Berry DA, Perou CM, et al. Impact of the addition of carboplatin and/or bevacizumab to neoadjuvant once-per-week paclitaxel followed by dose-
dense doxorubicin and cyclophosphamide on pathologic complete response rates in stage II to III triple-negative breast cancer: CALGB 40603 (Alliance). J Clin
Oncol. 2015;33:13-21.
11. Sharma P, López-Tarruella S, Garcı́a-Saenz JA, et al. Efficacy of neoadjuvant carboplatin plus docetaxel in triple-negative breast cancer: combined analysis of
two cohorts. Clin Cancer Res. 2017;23:649-657.
12. Loibl S, Untch M, Burchardi N, et al. A randomised phase II study investigating durvalumab in addition to an anthracycline taxane-based neoadjuvant therapy in
early triple-negative breast cancer: clinical results and biomarker analysis of GeparNuevo study. Ann Oncol. 2019;30:1279-1288.
13. National Comprehensive Cancer Network. NCCN clinical practice guidelines in oncology: breast cancer 2019. https://www.nccn.org/professionals/physician_
gls/pdf/breast.pdf. Accessed December 4, 2019.
14. Carey LA, Dees EC, Sawyer L, et al. The triple negative paradox: primary tumor chemosensitivity of breast cancer subtypes. Clin Cancer Res. 2007;
13:2329-2334.
15. Dent R, Trudeau M, Pritchard KI, et al. Triple-negative breast cancer: clinical features and patterns of recurrence. Clin Cancer Res. 2007;13:4429-4434.

Sharma et al
16. Liedtke C, Mazouni C, Hess KR, et al. Response to neoadjuvant therapy and long-term survival in patients with triple-negative breast cancer. J Clin Oncol. 2008;
26:1275-1281.
17. Bauer KR, Brown M, Cress RD, et al. Descriptive analysis of estrogen receptor (ER)-negative, progesterone receptor (PR)-negative, and HER2-negative invasive
breast cancer, the so-called triple-negative phenotype: a population-based study from the California Cancer Registry. Cancer. 2007;109:1721-1728.
18. Hammond ME, Hayes DF, Wolff AC, et al. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for immu-
nohistochemical testing of estrogen and progesterone receptors in breast cancer. J Oncol Pract. 2010;6:195-197.
19. Wolff AC, Hammond MEH, Hicks DG, et al; College of American Pathologists. Recommendations for human epidermal growth factor receptor 2 testing in breast
cancer: American Society of Clinical Oncology/College of American Pathologists clinical practice guideline update. J Clin Oncol. 2013;31:3997-4013.
20. Kohler BA, Sherman RL, Howlader N, et al. Annual report to the nation on the status of cancer, 1975-2011, featuring incidence of breast cancer subtypes by
race/ethnicity, poverty, and state. J Natl Cancer Inst. 2015;107:djv048.
21. Denduluri N, Somerfield MR, Eisen A, et al. Selection of optimal adjuvant chemotherapy regimens for human epidermal growth factor receptor 2 (HER2)-
negative and adjuvant targeted therapy for HER2-positive breast cancers: an American Society of Clinical Oncology guideline adaptation of the Cancer Care
Ontario clinical practice guideline. J Clin Oncol. 2016;34:2416-2427.
22. Haffty BG, Yang Q, Reiss M, et al. Locoregional relapse and distant metastasis in conservatively managed triple negative early-stage breast cancer. J Clin Oncol.
2006;24:5652-5657.
23. Nielsen TO, Hsu FD, Jensen K, et al. Immunohistochemical and clinical characterization of the basal-like subtype of invasive breast carcinoma. Clin Cancer Res.
2004;10:5367-5374.
24. Tan DS, Marchió C, Jones RL, et al. Triple negative breast cancer: molecular profiling and prognostic impact in adjuvant anthracycline-treated patients. Breast
25. Cortazar P, Geyer CE Jr Pathological complete response in neoadjuvant treatment of breast cancer. Ann Surg Oncol. 2015;22:1441-1446.
26. Spring L, Fell G, Arfe A, et al. Pathological complete response after neoadjuvant chemotherapy and impact on breast cancer recurrence and mortality, stratified
by breast cancer subtypes and adjuvant chemotherapy usage: individual patient-level meta-analyses of over 27,000 patients. Cancer Res. 2019;79 (suppl;
abstr GS2-03).
27. Loibl S, O’Shaughnessy J, Untch M, et al. Addition of the PARP inhibitor veliparib plus carboplatin or carboplatin alone to standard neoadjuvant chemotherapy in
triple-negative breast cancer (BrighTNess): a randomised, phase 3 trial. Lancet Oncol. 2018;19:497-509.
28. Arun B, Bayraktar S, Liu DD, et al. Response to neoadjuvant systemic therapy for breast cancer in BRCA mutation carriers and noncarriers: a single-institution
experience. J Clin Oncol. 2011;29:3739-3746.
29. Engblom P, Rantanen V, Kulmala J, et al. Additive and supra-additive cytotoxicity of cisplatin-taxane combinations in ovarian carcinoma cell lines. Br J Cancer.
1999;79:286-292.
30. Delbaldo C, Michiels S, Syz N, et al. Benefits of adding a drug to a single-agent or a 2-agent chemotherapy regimen in advanced non–small cell lung cancer:
a meta-analysis. JAMA. 2004;292:470-484.
31. von Minckwitz G, Schneeweiss A, Loibl S, et al. Neoadjuvant carboplatin in patients with triple-negative and HER2-positive early breast cancer (GeparSixto; GBG
66): a randomised phase 2 trial. Lancet Oncol. 2014;15:747-756.
32. Gluz O, Nitz U, Liedtke C, et al. No survival benefit of chemotherapy escalation in patients with pCR and “high-immune” triple-negative early breast cancer in the
neoadjuvant WSG-ADAPT-TN trial. Cancer Res. 2019;79 (suppl; abstr GS5-06).
33. Sharma P, López-Tarruella S, Garcia-Saenz JA, et al. Pathological response and survival in triple-negative breast cancer following neoadjuvant carboplatin plus
docetaxel. Clin Cancer Res. 2018;24:5820-5829.
34. Mayer EL, Abramson VG, Jankowitz RC, et al. TBCRC 030: a randomized phase II study of preoperative cisplatin versus paclitaxel in TNBC—evaluating the
homologous recombination deficiency (HRD) biomarker. J Clin Oncol. 2019;37 (suppl; abstr 507).
35. Sharma P, Kimler BF, O’Dea A, et al. Results of randomized phase II trial of neoadjuvant carboplatin plus docetaxel or carboplatin plus paclitaxel followed by AC
in stage I-III triple-negative breast cancer (NCT02413320). J Clin Oncol. 2019;37 (suppl; abstr 516).
36. Blum JL, Flynn PJ, Yothers G, et al. Anthracyclines in early breast cancer: the ABC trials-USOR 06-090, NSABP B-46-I/USOR 07132, and NSABP B-49 (NRG
Oncology). J Clin Oncol. 2017;35:2647-2655.
37. Sharma P, Barlow WE, Godwin AK, et al. Impact of homologous recombination deficiency biomarkers on outcomes in patients with triple-negative breast cancer
treated with adjuvant doxorubicin and cyclophosphamide (SWOG S9313). Ann Oncol. 2018;29:654-660.
38. Tutt A, Tovey H, Cheang MCU, et al. Carboplatin in BRCA1/2-mutated and triple-negative breast cancer BRCAness subgroups: the TNT trial. Nat Med. 2018;
24:628-637.
39. Tung N, Arun B, Hacker MR, et al. TBCRC 031: randomized phase II study of neoadjuvant cisplatin versus doxorubicin-cyclophosphamide in germline BRCA
carriers with HER2-negative breast cancer (the INFORM trial). J Clin Oncol. Epub 2020 Feb 25.
40. Hahnen E, Lederer B, Hauke J, et al. Germline mutation status, pathological complete response, and disease-free survival in triple-negative breast cancer:
secondary analysis of the GeparSixto randomized clinical trial. JAMA Oncol. 2017;3:1378-1385.
41. Denkert C, von Minckwitz G, Brase JC, et al. Tumor-infiltrating lymphocytes and response to neoadjuvant chemotherapy with or without carboplatin in human
epidermal growth factor receptor 2-positive and triple-negative primary breast cancers. J Clin Oncol. 2015;33:983-991.
42. Savas P, Salgado R, Denkert C, et al. Clinical relevance of host immunity in breast cancer: from TILs to the clinic. Nat Rev Clin Oncol. 2016;13:228-241.

43. Loi S, Drubay D, Adams S, et al. Tumor-infiltrating lymphocytes and prognosis: a pooled individual patient analysis of early-stage triple-negative breast cancers.
J Clin Oncol. 2019;37:559-569.
44. Luen SJ, Salgado R, Dieci MV, et al. Prognostic implications of residual disease tumor-infiltrating lymphocytes and residual cancer burden in triple-negative
breast cancer patients after neoadjuvant chemotherapy. Ann Oncol. 2019;30:236-242.
45. Denkert C, Loibl S, Noske A, et al. Tumor-associated lymphocytes as an independent predictor of response to neoadjuvant chemotherapy in breast cancer.
J Clin Oncol. 2010;28:105-113.
46. Denkert C, von Minckwitz G, Darb-Esfahani S, et al. Tumour-infiltrating lymphocytes and prognosis in different subtypes of breast cancer: a pooled analysis of
3771 patients treated with neoadjuvant therapy. Lancet Oncol. 2018;19:40-50.
47. Ayers M, Symmans WF, Stec J, et al. Gene expression profiles predict complete pathologic response to neoadjuvant paclitaxel and fluorouracil, doxorubicin, and
cyclophosphamide chemotherapy in breast cancer. J Clin Oncol. 2004;22:2284-2293.
48. Gianni L, Zambetti M, Clark K, et al. Gene expression profiles in paraffin-embedded core biopsy tissue predict response to chemotherapy in women with locally
advanced breast cancer. J Clin Oncol. 2005;23:7265-7277.
49. Hatzis C, Pusztai L, Valero V, et al. A genomic predictor of response and survival following taxane-anthracycline chemotherapy for invasive breast cancer. JAMA.
2011;305:1873-1881.
50. Sharma P, Barlow WE, Godwin AK, et al. Validation of the DNA damage immune response signature in patients with triple-negative breast cancer from the SWOG
9313c trial. J Clin Oncol. 2019;37:3484-3492.
51. Telli ML, Metzger O, Timms K, et al. Evaluation of homologous recombination deficiency (HRD) status with pathological response to carboplatin +/ veliparib in
BrighTNess, a randomized phase 3 study in early stage TNBC. J Clin Oncol. 2018;36 (suppl; abstr 519).
J Med. 2018;379:2108-2121.
53. Merck & Co. Merck’s KEYTRUDA (pembrolizumab) in combination with chemotherapy met primary endpoint of progression-free survival (PFS) as first-line
treatment for metastatic triple-negative breast cancer (mTNBC) [press release]. https://investors.merck.com/news/press-release-details/2020/Mercks-
KEYTRUDA-pembrolizumab-in-Combination-with-Chemotherapy-Met-Primary-Endpoint-of-Progression-Free-Survival-PFS-as-First-Line-Treatment-for-
Metastatic-Triple-Negative-Breast-Cancer-mTNBC/default.aspx. Accessed March 2, 2020.
54. Schmid P, Cortes J, Pusztai L, et al; KEYNOTE-522 Investigators. Pembrolizumab for early triple-negative breast cancer. N Engl J Med. 2020;382:810-821.
55. Schmid P, Park YH, Ferreira M, et al. KEYNOTE-522 study of neoadjuvant pembrolizumab + chemotherapy vs placebo + chemotherapy, followed by adjuvant
pembrolizumab vs placebo for early triple-negative breast cancer: pathologic complete response in key subgroups and by treatment exposure, residual cancer
burden, and breast-conserving surgery. Presented at: San Antonio Breast Cancer Symposium. San Antonio, TX; 2019. Abstract GS3-03.
56. Gianni L, Huang C-S, Egle D, et al. Pathologic complete response (pCR) to neoadjuvant treatment with or without atezolizumab in triple negative, early high-risk
and locally advanced breast cancer. NeoTRIPaPDL1 Michelangelo randomized study. Cancer Res. 2020;80 (suppl; abstr GS3-04).
57. Masuda N, Lee S-J, Ohtani S, et al. Adjuvant capecitabine for breast cancer after preoperative chemotherapy. N Engl J Med. 2017;376:2147-2159.
58. van Mackelenbergh M, Seither F, Möbus V, et al. Effects of capecitabine as part of neo-/adjuvant chemotherapy. A meta-analysis of individual patient data from
12 randomized trials including 15,457 patients. Cancer Res. 2020;80 (abstr GS1-07).
59. Burstein HJ, Curigliano G, Loibl S, et al; Members of the St. Gallen International Consensus Panel on the Primary Therapy of Early Breast Cancer 2019.
Estimating the benefits of therapy for early-stage breast cancer: the St. Gallen International Consensus Guidelines for the primary therapy of early breast cancer
2019. Ann Oncol. 2019;30:1541-1557.
60. Ménard S, Pupa SM, Campiglio M, et al. Biologic and therapeutic role of HER2 in cancer. Oncogene. 2003;22:6570-6578.
61. Slamon DJ, Clark GM, Wong SG, et al. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science. 1987;
235:177-182.
62. Cuello M, Ettenberg SA, Clark AS, et al. Down-regulation of the erbB-2 receptor by trastuzumab (herceptin) enhances tumor necrosis factor-related apoptosis-
inducing ligand-mediated apoptosis in breast and ovarian cancer cell lines that overexpress erbB-2. Cancer Res. 2001;61:4892-4900.
63. Molina MA, Codony-Servat J, Albanell J, et al. Trastuzumab (herceptin), a humanized anti-Her2 receptor monoclonal antibody, inhibits basal and activated Her2
ectodomain cleavage in breast cancer cells. Cancer Res. 2001;61:4744-4749.
64. Junttila TT, Akita RW, Parsons K, et al. Ligand-independent HER2/HER3/PI3K complex is disrupted by trastuzumab and is effectively inhibited by the PI3K
inhibitor GDC-0941. Cancer Cell. 2009;15:429-440.
65. Arnould L, Gelly M, Penault-Llorca F, et al. Trastuzumab-based treatment of HER2-positive breast cancer: an antibody-dependent cellular cytotoxicity
mechanism? Br J Cancer. 2006;94:259-267.
66. Slamon DJ, Leyland-Jones B, Shak S, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses
HER2. N Engl J Med. 2001;344:783-792.
67. Ahmed S, Sami A, Xiang J. HER2-directed therapy: current treatment options for HER2-positive breast cancer. Breast Cancer. 2015;22:101-116.
68. Slamon D, Eiermann W, Robert N, et al; Breast Cancer International Research Group. Adjuvant trastuzumab in HER2-positive breast cancer. N Engl J Med.
2011;365:1273-1283.
69. Perez EA, Suman VJ, Davidson NE, et al. Sequential versus concurrent trastuzumab in adjuvant chemotherapy for breast cancer. J Clin Oncol. 2011;
29:4491-4497.

Sharma et al
70. Gianni L, Dafni U, Gelber RD, et al; Herceptin Adjuvant (HERA) Trial Study Team. Treatment with trastuzumab for 1 year after adjuvant chemotherapy in patients
with HER2-positive early breast cancer: a 4-year follow-up of a randomised controlled trial. Lancet Oncol. 2011;12:236-244.
71. Moja L, Tagliabue L, Balduzzi S, et al. Trastuzumab containing regimens for early breast cancer. Cochrane Database Syst Rev. 2012;(4):CD006243.
72. McAndrew N, DeMichele A. neoadjuvant chemotherapy considerations in triple-negative breast cancer. J Target Ther Cancer. 2018;7:52-69.
73. Untch M, Rezai M, Loibl S, et al. Neoadjuvant treatment with trastuzumab in HER2-positive breast cancer: results from the GeparQuattro study. J Clin Oncol.
2010;28:2024-2031.
74. Buzdar AU, Ibrahim NK, Francis D, et al. Significantly higher pathologic complete remission rate after neoadjuvant therapy with trastuzumab, paclitaxel, and
epirubicin chemotherapy: results of a randomized trial in human epidermal growth factor receptor 2-positive operable breast cancer. J Clin Oncol. 2005;
23:3676-3685.
75. Gianni L, Eiermann W, Semiglazov V, et al. Neoadjuvant and adjuvant trastuzumab in patients with HER2-positive locally advanced breast cancer (NOAH):
follow-up of a randomised controlled superiority trial with a parallel HER2-negative cohort. Lancet Oncol. 2014;15:640-647.
76. Baselga J, Cortés J, Kim S-B, et al; CLEOPATRA Study Group. Pertuzumab plus trastuzumab plus docetaxel for metastatic breast cancer. N Engl J Med. 2012;
366:109-119.
77. Nahta R, Hung M-C, Esteva FJ. The HER-2-targeting antibodies trastuzumab and pertuzumab synergistically inhibit the survival of breast cancer cells. Cancer
Res. 2004;64:2343-2346.
78. Scheuer W, Friess T, Burtscher H, et al. Strongly enhanced antitumor activity of trastuzumab and pertuzumab combination treatment on HER2-positive human
xenograft tumor models. Cancer Res. 2009;69:9330-9336.
79. Sergina NV, Rausch M, Wang D, et al. Escape from HER-family tyrosine kinase inhibitor therapy by the kinase-inactive HER3. Nature. 2007;445:437-441.
80. Nagata Y, Lan K-H, Zhou X, et al. PTEN activation contributes to tumor inhibition by trastuzumab, and loss of PTEN predicts trastuzumab resistance in patients.
Cancer Cell. 2004;6:117-127.
81. Berns K, Horlings HM, Hennessy BT, et al. A functional genetic approach identifies the PI3K pathway as a major determinant of trastuzumab resistance in breast
cancer. Cancer Cell. 2007;12:395-402.
82. Gianni L, Pienkowski T, Im Y-H, et al. Efficacy and safety of neoadjuvant pertuzumab and trastuzumab in women with locally advanced, inflammatory, or early
HER2-positive breast cancer (NeoSphere): a randomised multicentre, open-label, phase 2 trial. Lancet Oncol. 2012;13:25-32.
83. Schneeweiss A, Chia S, Hickish T, et al. Pertuzumab plus trastuzumab in combination with standard neoadjuvant anthracycline-containing and anthracycline-
free chemotherapy regimens in patients with HER2-positive early breast cancer: a randomized phase II cardiac safety study (TRYPHAENA). Ann Oncol. 2013;
24:2278-2284.
84. Yu AF, Singh JC, Wang R, et al. Cardiac safety of dual anti-HER2 therapy in the neoadjuvant setting for treatment of HER2-positive breast cancer. Oncologist.
2017;22:642-647.
85. Rahmani H, Shahriary A, Sheikhi MA, et al. Applications of cardiotoxicity in breast cancer: a meta-analysis. Panminerva Med. 2017;59:90-96.
86. van Ramshorst MS, van Werkhoven E, Honkoop AH, et al; Dutch Breast Cancer Research Group (BOOG). Toxicity of dual HER2-blockade with pertuzumab
added to anthracycline versus non-anthracycline containing chemotherapy as neoadjuvant treatment in HER2-positive breast cancer: the TRAIN-2 study.
Breast. 2016;29:153-159.
87. von Minckwitz G, Procter M, de Azambuja E, et al; APHINITY Steering Committee and Investigators. Adjuvant pertuzumab and trastuzumab in early HER2-
88. Piccart M, Procter M, Fumagalli D, et al. Interim overall survival analysis of APHINITY (BIG 4-11): a randomized multicenter, double-blind, placebo-controlled
trial comparing chemotherapy plus trastuzumab plus pertuzumab versus chemotherapy plus trastuzumab plus placebo as adjuvant therapy in patients with
operable HER2-positive early breast cancer. Cancer Res. 2020;80 (abstr GS1-04).
89. European Society for Medical Oncology. NICE recommends pertuzumab for the neoadjuvant treatment of HER2-positive breast cancer. www.esmo.org/
oncology-news/NICE-Recommends-Pertuzumab-for-the-Neoadjuvant-Treatment-of-HER2-positive-Breast-Cancer. Accessed March 2, 2020.
90. Roussos Torres ET, Connolly RM. Optimizing HER2-directed therapy in early-stage breast cancer. Curr Breast Cancer Rep. 2018;10:262-273.
91. Ballantyne A, Dhillon S. Trastuzumab emtansine: first global approval. Drugs. 2013;73:755-765.
92. U.S. Food and Drug Administration. FDA approves neratinib for metastatic HER2-positive breast cancer. www.fda.gov/drugs/resources-information-approved-
drugs/fda-approves-neratinib-metastatic-her2-positive-breast-cancer. Accessed March 2, 2020.
94. Arpino G, Wiechmann L, Osborne CK, et al. Crosstalk between the estrogen receptor and the HER tyrosine kinase receptor family: molecular mechanism and
clinical implications for endocrine therapy resistance. Endocr Rev. 2008;29:217-233.
95. Prowell TM, Beaver JA, Pazdur R. Residual disease after neoadjuvant therapy - developing drugs for high-risk early breast cancer. N Engl J Med. 2019;
380:612-615.
96. Murthy RK, Loi S, Okines A, et al. Tucatinib, trastuzumab, and capecitabine for HER2-positive metastatic breast cancer. N Engl J Med. 2020;382:597-609.
J Med. 2020;382:610-621.

98. Stagg J, Loi S, Divisekera U, et al. Anti-ErbB-2 mAb therapy requires type I and II interferons and synergizes with anti–PD-1 or anti-CD137 mAb therapy. Proc
Natl Acad Sci USA. 2011;108:7142-7147.
99. Christmas BJ, Rafie CI, Hopkins AC, et al. Entinostat converts immune-resistant breast and pancreatic cancers into checkpoint-responsive tumors by
reprogramming tumor-infiltrating MDSCs. Cancer Immunol Res. 2018;6:1561-1577.
100. Dirix LY, Takacs I, Jerusalem G, et al. Avelumab, an anti-PD-L1 antibody, in patients with locally advanced or metastatic breast cancer: a phase 1b JAVELIN solid
tumor study. Breast Cancer Res Treat. 2018;167:671-686.
101. Loi S, Giobbe-Hurder A, Gombos A, et al. Phase Ib/II study evaluating safety and efficacy of pembrolizumab and trastuzumab in patients with trastuzumab-
resistant HER2-positive metastatic breast cancer: results from the PANACEA (IBCSG 45-13/BIG 4-13/KEYNOTE-014) study. Cancer Res. 2018;78 (abstr GS2-
06).
102. Coakley M, Garcia-Murillas I, Turner NC. Molecular residual disease and adjuvant trial design in solid tumors. Clin Cancer Res. 2019;25:6026-6034.
103. Tolaney SM, Guo H, Pernas S, et al. Seven-year follow-up analysis of adjuvant paclitaxel and trastuzumab trial for node-negative, human epidermal growth factor
receptor 2-positive breast cancer. J Clin Oncol. 2019;37:1868-1875.
104. Tolaney SM, Trippa L, Barry W, et al. TBCRC 033: a randomized phase II study of adjuvant trastuzumab emtansine (T-DM1) vs paclitaxel (T) in combination with
trastuzumab (H) for stage I HER2-positive breast cancer (BC) (ATEMPT). Cancer Res. 2020;80 (abstr GS1-05).
105. Connolly RM, Leal JP, Solnes L, et al. TBCRC026: phase II clinical trial assessing the correlation of standardized uptake value (SUV) on positron emission
tomography (PET) with pathological complete response (pCR) to pertuzumab and trastuzumab in patients with primary operable HER2-positive breast cancer.
106. Llombart-Cussac A, Cortés J, Paré L, et al. HER2-enriched subtype as a predictor of pathological complete response following trastuzumab and lapatinib without
chemotherapy in early-stage HER2-positive breast cancer (PAMELA): an open-label, single-group, multicentre, phase 2 trial. Lancet Oncol. 2017;18:545-554.
107. Rimawi MF, Mayer IA, Forero A, et al. Multicenter phase II study of neoadjuvant lapatinib and trastuzumab with hormonal therapy and without chemotherapy in
patients with human epidermal growth factor receptor 2-overexpressing breast cancer: TBCRC 006. J Clin Oncol. 2013;31:1726-1731.
108. Goel S, Pernas S, Tan-Wasielewski Z, et al. Ribociclib plus trastuzumab in advanced HER2-positive breast cancer: results of a phase 1b/2 trial. Clin Breast
Cancer. 2019;19:399-404.
109. Gianni L, Bisagni G, Colleoni M, et al. Neoadjuvant treatment with trastuzumab and pertuzumab plus palbociclib and fulvestrant in HER2-positive, ER-positive
breast cancer (NA-PHER2): an exploratory, open-label, phase 2 study. Lancet Oncol. 2018;19:249-256.
110. NCT02448420. Study of Palbociclib and Trastuzumab With Endocrine Therapy in HER2-positive Metastatic Breast Cancer. https://clinicaltrials.gov/ct2/show/
NCT02448420. Accessed March 2, 2020.
111. NCT03530696. T-DM1 and Palbociclib for Metastatic HER2 Breast Cancer (T-DM1). https://clinicaltrials.gov/ct2/show/NCT03530696. Accessed March 2,
2020.
112. NCT02947685. Randomized, Open Label, Clinical Study of the Targeted Therapy, Palbociclib, to Treat Metastatic Breast Cancer (PATINA). https://clinicaltrials.
gov/ct2/show/NCT02947685. Accessed March 2, 2020.
113. Prat A, Pascual T, Adamo B. Intrinsic molecular subtypes of HER2+ breast cancer. Oncotarget. 2017;8:73362-73363.
114. Dieci MV, Prat A, Tagliafico E, et al. Integrated evaluation of PAM50 subtypes and immune modulation of pCR in HER2-positive breast cancer patients treated
with chemotherapy and HER2-targeted agents in the CherLOB trial. Ann Oncol. 2016;27:1867-1873.
115. Fumagalli D, Venet D, Ignatiadis M, et al. RNA sequencing to predict response to neoadjuvant anti-HER2 therapy: a secondary analysis of the NeoALTTO
randomized clinical trial. JAMA Oncol. 2017;3:227-234.
116. Carey LA, Berry DA, Cirrincione CT, et al. Molecular heterogeneity and response to neoadjuvant human epidermal growth factor receptor 2 targeting in CALGB
40601, a randomized phase III trial of paclitaxel plus trastuzumab with or without lapatinib. J Clin Oncol. 2016;34:542-549.
117. Prat A, Bianchini G, Thomas M, et al. Research-based PAM50 subtype predictor identifies higher responses and improved survival outcomes in HER2-positive
breast cancer in the NOAH study. Clin Cancer Res. 2014;20:511-521.
118. Prat A, Angelis CD, Pascual Ts, et al. HER2-enriched subtype and ERBB2 mRNA as predictors of pathological complete response following trastuzumab and
lapatinib without chemotherapy in early-stage HER2-positive breast cancer: a combined analysis of TBCRC006/023 and PAMELA trials. J Clin Oncol. 2018;36
(suppl; abstr 509).
119. Rothé F, Silva MJ, Venet D, et al. Circulating tumor DNA in HER2-amplified breast cancer: a translational research substudy of the NeoALTTO phase III trial. Clin
Cancer Res. 2019;25:3581-3588.
120. Di Cosimo S, Campbell C, Azim HA Jr., et al. The use of breast imaging for predicting response to neoadjuvant lapatinib, trastuzumab and their combination in
HER2-positive breast cancer: results from Neo-ALTTO. Eur J Cancer. 2018;89:42-48.
121. NCT03161353. Chemotherapy-Free Trastuzumab and Pertuzumab in HER2-Positive Breast Cancer: FDG-PET Response-adapted Strategy. (PHERGain).
https://clinicaltrials.gov/ct2/show/NCT03161353. Accessed March 2, 2020.
122. Bentzen AG, Guren MG, Wanderås EH, et al. Chemoradiotherapy of anal carcinoma: survival and recurrence in an unselected national cohort. Int J Radiat Oncol
Biol Phys. 2012;83:e173-e180.
123. Habr-Gama A, Lynn PB, Jorge JMN, et al. Impact of organ-preserving strategies on anorectal function in patients with distal rectal cancer following neoadjuvant
chemoradiation. Dis Colon Rectum. 2016;59:264-269.
124. Smith FM, Cresswell K, Myint AS, et al. Is “watch-and-wait” after chemoradiotherapy safe in patients with rectal cancer? BMJ. 2018;363:k4472.

Sharma et al
125. Ring A, Webb A, Ashley S, et al. Is surgery necessary after complete clinical remission following neoadjuvant chemotherapy for early breast cancer? J Clin Oncol.
2003;21:4540-4545.
126. Bruce J, Thornton AJ, Scott NW, et al. Chronic preoperative pain and psychological robustness predict acute postoperative pain outcomes after surgery for
breast cancer. Br J Cancer. 2012;107:937-946.
127. Bruce J, Thornton AJ, Powell R, et al; Recovery Study Group. Psychological, surgical, and sociodemographic predictors of pain outcomes after breast cancer
surgery: a population-based cohort study. Pain. 2014;155:232-243.
128. De Los Santos JF, Cantor A, Amos KD, et al. Magnetic resonance imaging as a predictor of pathologic response in patients treated with neoadjuvant systemic
treatment for operable breast cancer. Translational Breast Cancer Research Consortium trial 017. Cancer. 2013;119:1776-1783.
129. Marinovich ML, Houssami N, Macaskill P, et al. Meta-analysis of magnetic resonance imaging in detecting residual breast cancer after neoadjuvant therapy.
J Natl Cancer Inst. 2013;105:321-333.
130. Evans A, Whelehan P, Thompson A, et al. Prediction of pathological complete response to neoadjuvant chemotherapy for primary breast cancer comparing
interim ultrasound, shear wave elastography and MRI. Ultraschall Med. 2018;39:422-431.
131. Henderson S, Purdie C, Michie C, et al. Interim heterogeneity changes measured using entropy texture features on T2-weighted MRI at 3.0 T are associated with
pathological response to neoadjuvant chemotherapy in primary breast cancer. Eur Radiol. 2017;27:4602-4611.
132. Basik M, Costantino J, De Los Santos J, et al. Phase II trial assessing accuracy of tumor bed biopsies in predicting pathologic response in patients with clinical/
radiological complete response after neoadjuvant chemotherapy in order to explore the feasibility of breast-conserving surgery without surgery: NRG Oncology
BR005. Cancer Res. 2019;79 (abstr OT1-09-01).
133. Heil J, Pfob A, Sinn H-PP, et al. Diagnosing residual disease and pathologic complete response after neoadjuvant chemotherapy in breast cancer patients by
image-guided vacuum-assisted breast biopsy: results of a prospective multicenter trial. Cancer Res.2020;80 (abstr GS5-03).
134. Tasoulis MK, Lee H-B, Yang W, et al. Accuracy of post-neoadjuvant chemotherapy image-guided breast biopsy to predict the presence of residual cancer:
a multi-institutional pooled analysis. Cancer Res. 2020;80 (abstr GS5-04).
135. Barron AU, Hoskin TL, Day CN, et al. Association of low nodal positivity rate among patients with ERBB2-positive or triple-negative breast cancer and breast
pathologic complete response to neoadjuvant chemotherapy. JAMA Surg. 2018;153:1120-1126.
136. Tadros AB, Yang WT, Krishnamurthy S, et al. Identification of patients with documented pathologic complete response in the breast after neoadjuvant
chemotherapy for omission of axillary surgery. JAMA Surg. 2017;152:665-670.
137. Le-Petross HT, McCall LM, Hunt KK, et al. Axillary ultrasound identifies residual nodal disease after chemotherapy: results from the American College of
Surgeons Oncology Group Z1071 Trial (Alliance). AJR Am J Roentgenol. 2018;210:669-676.
138. Caudle AS, Yang WT, Krishnamurthy S, et al. Improved axillary evaluation following neoadjuvant therapy for patients with node-positive breast cancer using
selective evaluation of clipped nodes: implementation of targeted axillary dissection. J Clin Oncol. 2016;34:1072-1078.
139. Boughey JC, Suman VJ, Mittendorf EA, et al; Alliance for Clinical Trials in Oncology. Sentinel lymph node surgery after neoadjuvant chemotherapy in patients
with node-positive breast cancer: the ACOSOG Z1071 (Alliance) clinical trial. JAMA. 2013;310:1455-1461.
140. Kuehn T, Bauerfeind I, Fehm T, et al. Sentinel-lymph-node biopsy in patients with breast cancer before and after neoadjuvant chemotherapy (SENTINA):
a prospective, multicentre cohort study. Lancet Oncol. 2013;14:609-618.
141. Fayanju OM, Ren Y, Thomas SM, et al. The clinical significance of breast-only and node-only pathologic complete response (pCR) after neoadjuvant
chemotherapy (NACT): a review of 20,000 breast cancer patients in the National Cancer Data Base (NCDB). Ann Surg. 2018;268:591-601.
142. von Minckwitz G, Huang C-S, Mano MS, et al; KATHERINE Investigators. Trastuzumab emtansine for residual invasive HER2-positive breast cancer. N Engl
J Med. 2019;380:617-628.
143. Rugo HS, Olopade OI, DeMichele A, et al; I-SPY 2 Investigators. Adaptive randomization of veliparib-carboplatin treatment in breast cancer. N Engl J Med. 2016;
375:23-34.

Beyond Tumor PD-L1: Emerging Genomic

Biomarkers for Checkpoint Inhibitor
Immunotherapy
Galina G. Lagos, MD1; Benjamin Izar, MD, PhD1; and Naiyer A. Rizvi, MD1
overview
Despite the success of immune checkpoint blockade as a strategy for activating an antitumor immune re-
sponse and promoting cancer regression, only a subset of patients have durable clinical benefit. Efforts are
ongoing to identify robust biomarkers that can effectively predict treatment response to immune checkpoint
inhibitors (ICIs). Although PD-L1 expression is useful for stratifying patients, it is an imperfect tool. Com-
prehensive next-generation sequencing platforms that are readily used in clinical practice to identify a tumor’s
potentially actionable genetic alterations also reveal tumor genomic features, including tumor mutation
burden (TMB), that may impact the response to ICIs. High TMB enhances tumor immunogenicity through
increased numbers of tumor neoantigens that may promote an immune response. Defective DNA repair,
leading to microsatellite instability, is an endogenous mechanism for increased tumor TMB that augments
response to anti–PD-1 blockade. Alternatively, DNA damage from exogenous factors is responsible for high
TMB seen in melanoma, lung cancer, and urothelial carcinoma, among tumor subtypes with higher response
rates to ICIs. In this review, we summarize data supporting the use of TMB as a biomarker as well as its known
limitations. We also highlight specific tumor suppressor genes and oncogenes that are under investigation as
biomarkers for ICI response and resistance. Efforts are ongoing to delineate which genomic tumor charac-
teristics can eventually be utilized in clinical practice to ascertain the benefit of ICIs for an individual patient.
INTRODUCTION genomic amplification of encoding gene regions (e.g.,

Although the advent of immunotherapy has revolu- Hodgkin lymphoma) or induced dynamically by
tionized cancer treatment across a range of malig- adaptive responses in the microenvironment, chiefly by
nancies, biomarkers predictive of treatment response interferon gamma secreted by T cells.1-3 Emerging data
are an area of ongoing investigation. In cancer de- also demonstrate expression of PD-L1 on various
velopment, tumor cells evolve to evade immune rec- subsets of tumor-infiltrating immune cells. The initial
ognition and destruction. One mechanism of immune phase I trial of the monoclonal PD-1 checkpoint in-
escape is through the modulation of immune check- hibitor nivolumab showed an association of tumor cell
points, including CTLA-4 and PD-L1, which regulate PD-L1 expression with treatment response, which was
the priming phase and effector phase of T-cell acti- subsequently supported by results from trials across
vation, respectively. Antibodies inhibiting these im- various tumor types and ICIs.4-10 However, other large
mune checkpoints promote antitumor immunity and studies did not consistently demonstrate this
produce durable cancer regression. Since the initial association.11-14 In part, this is due to the variable
approval of ipilimumab, an anti–CTLA-4 antibody, for definitions of “PD-L1 positivity,” measurement of PD-
Author affiliations
the treatment of melanoma in 2011, seven different L1 on tumor versus immune cells versus both, lack of
and support harmonization of reagents and IHC conditions, and
information (if
immune checkpoint inhibitors (ICIs) have been ap-
proved for the treatment of 15 different tumor types. clinical sampling bias (tumor center vs. tumor/normal
applicable) appear
at the end of this Despite the success of this treatment approach, ICIs boarder). The variability in PD-L1 assessment is re-
article. are only effective in a subset of cancer types and in only flected in the nine U.S. Food and Drug Administration
Accepted on March a fraction of patients, highlighting the need for robust (FDA) approvals for ICIs that are tied to three different
24, 2020 and
biomarkers to stratify patients to ICIs or other therapies. companion diagnostic tests with PD-L1 thresholds that
published at vary across tumor type and indication.15
ascopubs.org on
Cancer cell–autonomous PD-L1 protein expression
April 21, 2020:
measured by immunohistochemistry (IHC) is the most Nevertheless, PD-L1 expression alone is insufficient for
10.1200/EDBK_ established biomarker for ICI response in clinical predicting treatment response. In an analysis sum-
289967 practice. Expression of PD-L1 may be defined by marizing results from anti–PD-1/PD-L1 antibody trials,

Lagos, Izar, and Rizvi
druggable oncogene targets and are becoming part of the

standard of care workup of patients with cancer. In this
review, we outline how NGS data could also be used to
• Comprehensive next-generation sequencing inform immunotherapy sensitivity and resistance. We
platforms provide tumor genomic data that may
highlight the aspects of these assays that are most relevant
have a role in predicting response to immune
to immunotherapy decision-making based on currently
checkpoint blockade.
available data.
• Tumor mutational burden has emerged as
a biomarker of immunotherapy response across HIGH TMB RELATED TO DNA REPAIR DEFICIENCY
tumor types independent of PD-L1 expression. Increased TMB can be a function of impaired DNA repair.
• High tumor mutational burden can occur as Targeted NGS platforms assess for the presence of
a function of endogenous mechanisms microsatellite instability (MSI) within a tumor, a signature
(i.e., impaired DNA repair) or exogenous factors of deficient mismatch repair (dMMR). dMMR can occur
(i.e., DNA damage from carcinogen exposure). through inherited germline defects in mismatch repair
• The applicability of tumor mutational burden in genes, including MLH1, MSH2, MSH6, or PMS2. Alterna-
clinical practice is limited by the lack of stan- tively, somatic mutations or epigenetic silencing can in-
dardized tumor mutational burden thresholds activate both alleles, generating sporadic dMMR tumors.
and variability in quantification methods. These mutations result in MSI and in the accumulation of
• Several oncogenes and tumor suppressor frameshifts mutations and single-nucleotide variants that
genes have been identified as potential pre- produce immunogenic neoantigens, which correlates with
dictors of response to immune checkpoint high TMB. Furthermore, MSI high (MSI-H) tumors express
blockade and are an area of ongoing elevated levels of PD-L1, particularly on tumor-infiltrating
investigation. immune cells. These findings provide the putative biologic
basis for the high anti–PD-1 response seen in dMMR/MSI-H
patients with “PD-L1–positive” tumors had an overall re- tumors.20,21
sponse rate of 48%, whereas 15% of patients responded An ongoing phase II study demonstrated the utility of MMR/
despite having “PD-L1–negative” tumors.16 Barring sam- MSI status as a predictive marker in patients with treatment-
pling biases associated with any clinical testing, it has refractory dMMR cancers treated with the PD-1 inhibitor
become evident that more comprehensive biomarkers are pembrolizumab. The trial enrolled 40 patients with co-
necessary to encapsulate the variables associated with lorectal cancer and 46 patients with noncolorectal cancer,
treatment response, including neoantigen expression and representing 12 tumor types. Fifty-three percent of patients
heterogeneity, proficiency to present neoantigens, T-cell achieved an objective radiographic response, including
recognition, and capacity for asymmetric division and dif- 21% with a complete response. Two-year progression-free
ferentiation. By studying tumor tissue and blood from af- survival (PFS) and overall survival (OS) rates were 53% and
fected patients with response or resistance to ICIs, we have 64%, respectively.22 On whole-exome sequencing (WES),
recently gained additional insights into putative and clini- dMMR tumors had an average of 1,782 mutations per tu-
cally relevant biomarkers. mor, with 578 predicted to result in a neoantigen, under-
Tumor mutational burden (TMB) has emerged as a bio- scoring the immunogenicity of these tumors.23 Given these
marker predictive of response to immunotherapy, in- impressive results, the FDA granted the first approval for
dependent of PD-L1.17 Increased TMB can be driven by a tumor-agnostic therapy to pembrolizumab for use in MSI-
endogenous (impaired DNA repair) or exogenous factors H/dMMR cancers after progression on first-line treatment.24
(ultraviolet light, tobacco carcinogen exposure, chronic viral In systemic analyses, MSI is most prevalent in endometrial
infection). TMB is defined as the number of nonsynonymous (approximately 30%), gastric (approximately 20%), and
coding mutations per tumor genomic region analyzed. TMB colorectal (approximately 15%) cancers but is also seen at
correlates with the number of tumor-specific neoantigens, new a lower proportion in other diverse tumor types. dMMR
epitopes predicted to be presented via major histocompatibility tumors represent approximately 20,000 annual stage IV
complex (MHC) class I proteins, and capable of engaging the diagnoses in the United States.22 Therefore, MMR/MSI
T-cell receptor to elicit an adaptive immune response.18,19 status in treatment refractory disease can greatly affect
Through increased neoantigen expression, increased TMB treatment approach.
is thought to enhance tumor immunogenicity and the response Additional DNA repair alterations associated with high TMB
to ICIs. and ICI response include loss-of-function variants in the
Precision medicine analyses incorporate comprehensive exonuclease domains of DNA polymerase " and δ (POLE
next-generation sequencing (NGS) platforms to identify and POLD1, respectively). Although the presence of these

Genomic Biomarkers of Immune Checkpoint Inhibitor Response
mutations has been associated with response to ICIs,25,26 including the sequencing methodology, length of genome
the majority of the mutations detected by NGS panels are analyzed, types of mutations included in the TMB calculation,
variants of unknown significance and must be interpreted cutoff for variant allele frequencies, and use of matched normal
with caution. tissue to distinguish germline polymorphisms from somatic
HIGH TMB RELATED TO EXOGENOUS FACTORS mutations. This variability contributes the difficulty of stan-
dardizing TMB assessment.42
Early observations showed responses to ICIs in the setting of
DNA damage from exogenous factors in melanoma, non– Accurate TMB calculation depends on the size of genome
small cell lung cancer (NSCLC), and bladder cancer.27 sequenced with targeted gene panels. Whereas WES covers
Snyder et al28 first demonstrated the association between a 30-megabase (Mb) coding region, commercially available
high TMB and response to anti–CTLA-4 blockade in the panels provide 0.5 to 2.8 Mb of genomic coverage. TMB
clinical setting in patients with melanoma. Using a cutoff of calculated from larger gene panels (300 genes) correlates
100 mutations quantified with WES, a high TMB correlated strongly with WES-based calculation. Targeted sequencing
with sustained clinical benefit. Rizvi et al29 observed for samples with low to intermediate TMB becomes less
a similar benefit of anti–PD-1 blockade with pembrolizumab reliable with less than 0.5 Mb of genomic coverage.39,43
in patients with NSCLC with high TMB based on longer Furthermore, distinguishing between somatic and germline
durable clinical benefit, higher objective response rates mutations is an important factor in TMB calculation. MSK-
(ORRs), and longer PFS. Subsequent retrospective analyses IMPACT, like WES, uses matched germline blood samples
of cohort studies and clinical trials in melanoma, NSCLC, to exclude germline variants. Other platforms, including
urothelial carcinoma, head and neck cancer, and small cell FoundationOne, use databases of known germline poly-
lung cancer suggested an association between TMB and morphisms and a somatic-germline/zygosity algorithm to
improved clinical outcomes with anti–CTLA-4, anti–PD-1, filter out potential germline variants.42 The importance of
and/or anti–PD-L1 blockade.30-33 a germline control is amplified by increased understanding
of clonal hematopoiesis of indeterminant significance, in
Additional studies support the predictive potential of TMB which hematopoietic cells harbor mutations frequently at-
across tumor types. Yarchoan et al34 evaluated the re- tributed to tumor in bulk-tumor sequencing, such as TP53,
lationship between median TMB specific to tumor type and TET2, and DNMT1 mutations.44
the ORR to anti–PD-1/PD-L1 blockade among 27 different
malignancies using pooled data from published studies. A Given the variability in TMB quantification with different
strong association was seen, with a correlation coefficient of sequencing panels and the diversity in TMB across tumor
0.74 (p , .001). Several studies using patient-level data types, ascertaining TMB thresholds predictive of ICI re-
supported the usefulness of TMB across cancer types.35,36 sponse has been a challenge. Aggregating data from
The largest multicancer analysis of tumor TMB and ICI multiple studies, the TMB threshold for ICI benefit on the
response compiled data collected from the MSK-IMPACT basis of WES in NSCLC, melanoma, and urothelial carci-
panel performed as part of clinical care. A total of 1,662 noma is approximately 200 missense mutations, with
patients treated with at least one dose of any ICI were in- a range of 100 to 248 mutations as potential cutoffs.45 Most
cluded. High TMB (top 20% of each histology) was asso- data on TMB from targeted NGS panels and immuno-
ciated with improved OS (HR, 0.61; p = 1.3 10 7).37 therapy response are derived from FoundationOne CDx
Collectively, these data highlight the clinical utility of TMB testing in NSCLC studies. Proposed TMB thresholds pre-
yet underscore the difficulty in defining high TMB, given the dictive of response in these studies are in the range of 10 or
variability in TMB across cancer histologies and method- greater to 17.1 mutations (mut)/Mb.46 A receiver operating
ologies for TMB calculation. Given that these results were characteristic analysis from the phase II CheckMate 568
derived from retrospective analyses, additional prospective trial of nivolumab with ipilimumab as first-line treatment in
data are needed to confirm their clinical relevance. NSCLC proposed a cutoff of 10 or more mut/Mb to define
TMB QUANTIFICATION AND THRESHOLDS FROM the high-TMB population, with an area under the curve of
TARGETED NGS 0.73, which was used in the design of the phase III
CheckMate 227 trial.47 Notably, the official FoundationOne
Initial studies used WES to calculate TMB, but this approach CDx testing report defines high TMB as 20 or more mut/Mb,
is impractical in the clinical setting due to high costs, long intermediate as 6 to 19 mut/Mb, and low as 5 or fewer mut/
turnaround time, and analytical complexity. Alternatively, Mb. A proposed TMB cutoff with MSK-IMPACT in NSCLC is
empiric and in silico analyses demonstrated that TMB as- 7.4 or more mut/Mb (area under the curve, 0.601).30
sessment from targeted gene sequencing panels, including
FoundationOne and MSK-IMPACT, correlates well with These challenges support efforts attempting to establish
WES-derived TMB and clinical response to ICIs.30,38-41 harmonized metrics for TMB evaluation for clinical decision-
Notably, there is variability in the targeted NGS platforms, making. Spearheaded by the cancer advocacy group

Friends of Cancer Research and the German Quality As- high TMB and PD-L1 of 50% or greater correlated with
surance Initiative Pathology, efforts are underway to stan- a higher response rate of 75% with nivolumab.11 Similarly, in
dardize TMB calculations and reporting.48 In addition, some the IMvigor211 phase III trial of atezolizumab versus che-
studies have already demonstrated the potential for bio- motherapy in platinum-treated urothelial carcinoma, TMB,
informatics approaches for standardizing TMB measurements.49 evaluated with FoundationOne NGS, was an independent
BLOOD TMB AS AN ALTERNATIVE TO TISSUE TMB predictor for OS in exploratory analyses at a TMB threshold
of 9.65 mut/Mb (HR, 0.68). Patients with high TMB and PD-
Tissue TMB (tTMB) has several limitations as a predictive L1 expression had even better survival outcomes.31 These
biomarker. Measuring tTMB requires an invasive pro- trials suggest that both TMB and PD-L1 can be valuable for
cedure, sufficient quantity and quality of tissue, and predicting responders to ICIs.
a turnaround time of 2 to 3 weeks.50 In clinical trials, more
than 60% of patients have TMB-evaluable tissue LIMITATIONS OF TMB
samples.11,17,31 Blood-based assays for analyzing TMB offer
Other data have highlighted that TMB is an imperfect
several advantages, including ease of testing, readily
predictive tool. Checkmate 227 and NEPTUNE are large
available diagnostic material, and less susceptibility to
trials that incorporated prospective TMB assessment in their
sampling bias associated with single-site tissue biopsies.
study design. Prior retrospective data suggested the utility of
Several technologies have been developed to assess blood
TMB in predicting response to the combination of PD-1 plus
TMB (bTMB) from plasma cell-free DNA. This approach has
CTLA-4 blockade.10 Checkmate 227, a phase III trial in
primarily been used in studies of patients with NSCLC.
patients with advanced NSCLC, included a coprimary
Gandara et al50 assessed bTMB on plasma samples from endpoint of PFS with nivolumab with ipilimumab versus
patients with advanced NSCLC treated in the POPLAR and chemotherapy among patients with a TMB of 10 or more
OAK trials of atezolizumab. bTMB was predictive of PFS in mut/Mb. The median PFS with immunotherapy was sig-
patients who received atezolizumab monotherapy using nificantly longer in the high TMB group (HR, 0.58; p ,
a bTMB cutoff of 16 mut/Mb or greater. A positive, although .001). However, in exploratory analyses, the HR for OS with
imperfect, correlation between tTMB and bTMB was seen nivolumab with ipilimumab versus chemotherapy was
(Spearman’s ρ, 0.64; 95% CI, 0.56–0.71). Possible ex- comparable in the TMB less than 10 and 10 or greater mut/
planations for the discrepancy included differences in the Mb groups.17,53 In the phase III NEPTUNE trial of durva-
types of mutation detected by the assays and clinical lumab with tremelimumab versus platinum-based chemo-
sampling bias of the heterogeneous tumor tissue. A similar therapy in stage IV NSCLC, the primary analysis population
degree of concordance between tTMB testing with Foun- was patients with bTMB of 20 mut/Mb or more. The trial did
dationOne and bTMB with GuardantOMNI was noted in the not meet its primary endpoint of improved OS with ICIs in the
phase III MYSTIC trial of durvalumab and tremelimumab in high TMB population.54
advanced NSCLC (Spearman’s ρ, 0.6; Pearson’s r, 0.7).
This trial also demonstrated an improved response to TMB is not a useful marker in all cancer types. In some
durvalumab with tremelimumab versus chemotherapy histologies, substantial responses to ICIs are observed de-
among patients with bTMB levels of 16 mut/Mb or greater spite a low TMB. In Merkel cell carcinoma, there is
(HR, 0.62; 95% CI, 0.45–0.86).51 Given these results, there a bimodal distribution of TMB. The etiology for cancer
is great interest in implementing bTMB in clinical practice. development in the TMB-low cohort is primarily Merkel
cell polyomavirus (MCPyV). Meanwhile, TMB-high patients
TMB AND PD-L1 EXPRESSION AS INDEPENDENT express an ultraviolet damage mutational signature. Both
BIOMARKERS MCPyV-positive and MCPyV-negative patients respond to
TMB is not a substitute biomarker for PD-L1 assessment; anti–PD-L1 blockade, with response rates of 62% and 44%,
rather, it should be used in combination with PD-L1 ex- respectively, in the first-line setting. The biologic basis for
pression. Several studies have demonstrated that TMB and the robust response in the MCPyV group is still under in-
PD-L1 are independent predictive variables of ICI response. vestigation.55 In renal cell carcinoma (RCC), the response
However, a composite of TMB and PD-L1 enriches for rate to ICIs is higher than predicted on the basis of TMB
greater benefit from ICIs.11,17,30,46,52 In the phase III alone. Among 31 patients with metastatic RCC treated with
Checkmate 026 trial of first-line nivolumab versus che- ICIs, the duration of treatment with immunotherapy was
motherapy among patients with NSCLC and PD-L1 ex- similar between patients with high and low TMB, and the
pression of 1% or greater, patients with a high TMB, defined median TMB did not differ between responders and non-
as 243 or more missense mutations by WES (107 patients), responders.56 It is hypothesized that the high proportion of
had higher response rates and PFS with nivolumab com- insertion-deletion (indel) mutations in RCC can create
pared with chemotherapy. There was no association be- a novel open reading frame, generating high-affinity neo-
tween TMB and PD-L1 expression (Pearson’s r, 0.059), but antigens that increase tumor immunogenicity.57 Another

proposed mechanism is through increased expression of infiltration and function. Although the exact mechanisms
human endogenous retroviruses, remnants of viral genetic of T-cell exclusion in this context are under investigation,
material that are integrated into the genome from prior together these findings suggest that STK11 mutations
exogenous retroviral infections, which correlates with local promote a cold tumor immune microenvironment that
immune checkpoint activation. In small cohorts of patients contributes to immunotherapy resistance.63,64
with RCC treated with anti–PD-1 therapies, elevated ex-
KEAP1
pression of a set of endogenous retroviruses was associated
with responsiveness to ICIs.58,59 ICI resistance can also occur through mutations in KEAP1.
KEAP1 is a protein involved in the regulation of the oxidative
Together these findings suggest that TMB measurement is stress response pathway through binding of NRF2. KEAP1
a useful biomarker across multiple tumor types in the mutations lead to the release and constitutive activation of
prediction of ICI response, although it is neither necessary NRF2, promoting cellular resistance to oxidative stress, pro-
nor sufficient. Efforts are ongoing to elucidate the optimal liferation, and metabolic reprogramming.65 Across cancer
use of methods to determine and use TMB in clinical types, the prevalence of KEAP1 mutations is 2.7%, and it is
practice. 15% to 18% in NSCLC. In addition, KEAP1 mutations often co-
ONCOGENE AND TUMOR SUPPRESSOR GENE INFLUENCE ON occur with STK11 mutations in NSCLC, and the co-mutations
IMMUNE CHECKPOINT INHIBITOR RESPONSE correlate with worse clinical outcomes.66 In tumors sequenced
with MSK-IMPACT, patients with KEAP1 mutations had sig-
Analysis of genes implicated, to varying degrees, as predictors
nificantly shorter OS (39 vs. 109 months; p , .0001) and an
of ICI response are also included in standard targeted NGS
inferior OS of 10 versus 20 months (p , .0029) among those
panels. These genetic modifications are an area of ongoing
treated with ICIs.67 Similar outcomes were seen in exploratory
investigation, and our current understanding of their role in ICI
analyses from the MYSTIC trial in advanced NSCLC, for
response is primarily based on retrospectives studies. In the
which OS was shorter for patients with KEAP1 mutation
future, they may have a role in patient management.60
regardless of whether they received chemotherapy, dur-
STK11 valumab, or durvalumab with tremelimumab.68 KEAP1-
mutant tumors have significantly lower CD8+ T-cell, neu-
Serine/threonine kinase 11 (STK11), also known as liver
trophil, and dendritic cell infiltration.67 This cold tumor
kinase B1, is a tumor-suppressor gene that modulates cell
microenvironment may in part explain the resistance of
proliferation, energy metabolism, and polarity through phos-
these tumors to ICIs.
phorylation of adenosine monophosphate–activated protein
kinase and other adenosine monophosphate–activated pro- PTEN
tein kinase–related kinases. Germline mutations in STK11
PTEN, another tumor suppressor protein, also appears to
result in Peutz Jeghers syndrome, which is associated with an
modulate the ICI response. As a lipid phosphatase, PTEN is
increased risk of gastrointestinal, pancreatic, cervical, ovarian,
the natural regulator of signaling via the PI3K pathway,
and breast cancers.61 Somatic mutations are most prevalent
a central cancer pathway involved in cell survival, pro-
in lung adenocarcinomas, in which STK11 inactivation occurs
liferation, metabolism, and differentiation. Inactivation or
in 20% to 30% of tumors, more commonly among KRAS
loss of PTEN drives oncogenesis across multiple tumor
mutants. In patients with stage IV NSCLC, co-occurring
types. Genomic sequencing data from the Cancer Genome
mutations in KRAS and STK11 confer a worse prognosis
Atlas reveal that endometrial cancer (66.4%; n = 530) and
according to increased frequency of systemic and intracranial
glioblastoma multiforme (34.9%; n = 393) have the highest
metastases and a trend toward decreased survival.62
percentage of PTEN mutations and homozygous loss. In
Furthermore, STK11 inactivation has surfaced as a major addition, 10% to 15% of cervical, lung squamous cell,
driver of immune escape and resistance to immunotherapy melanoma, gastric, and colorectal cancers have PTEN
in KRAS-mutant lung adenocarcinoma. Skoulidis et al63 genetic aberrations. Peng et al69 demonstrated that PTEN
reported an ORR of 7.4% with anti–PD-1 blockade in loss in melanoma was associated with an immunosup-
KRAS/STK11-mutant tumors compared with 28.6% in pressive phenotype. Silencing PTEN expression in in vitro
KRAS-mutant/STK11–wild type tumors. A mouse model of and in vivo impaired T-cell–mediated antitumor activity. In
STK11 inactivation validated the observation of primary clinical samples, patients with melanoma whose tumors
resistance to PD-1 blockade. Several mechanisms likely lacked PTEN expression by immunohistochemistry had
underlie this blunted response. Regardless of intermediate inferior responses to anti–PD-1 therapy, and these PTEN-
or high TMB status, STK11-mutant tumors have low PD-L1 deficient tumors had decreased CD8+ T-cell tumor in-
expression. In mouse models and human tissue, STK11- filtration. The role of PTEN in modulating responses to ICIs
deficient tumors express increased inflammatory cytokines and the underlying mechanisms are currently being
that recruit suppressive myeloid cells and inhibit T-cell explored.

Beta Catenin melanoma, NSCLC, and MSI-H colorectal cancer.78-80

Upregulation of WNT-beta (β) catenin signaling also has Furthermore, pre-existing B2M deficiency may be associ-
been associated with T-cell exclusion and immunotherapy ated with worse survival.81 However, emerging data in pa-
resistance. WNT signaling has a pivotal role in embryonic tients with melanoma with B2M deficiency suggest that
development, including cell fate specification, differentia- patients whose tumors express MHC class II may still re-
tion, and migration as well as adult tissue homeostasis and spond to anti–CTLA-4 therapy.82 Deep-tissue immune
regeneration. Given its multitude of functions, dysregulation profiling elucidating the association of MHC class I and class
of the pathway is frequently involved in tumorigenesis and II expression with various immune infiltrates may provide
cancer metastasis.70 further insight into the underlying mechanisms of response
to ICIs in B2M-proficient and -deficient tumors.
Beta catenin signaling can be upregulated through gain-of-
function mutations in the β catenin gene, CTNNB1, or PD-L1 Amplification
through loss-of-function mutations in negative regulators, PD-L1 gene amplification emerged as a marker for ICI re-
such as APC and AXIN. The role of mutations in the regulators sponse on the basis of experiences in the treatment of
of the WNT pathway in tumorigenesis is best characterized in lymphoma. Patients with refractory classic Hodgkin lym-
colorectal cancer, but CTNNB1 alterations also are common phoma have high response rates of 65% to 87% to single-
in endometrial cancer, hepatobiliary cancer, and mela- agent ICIs. Amplification of the chromosome 9p24.1 locus
noma.71 Gene expression profiling of cutaneous melanomas in Reed-Sternberg cells augments the expression of PD-L1,
by Spranger et al72 suggested preferential activation of the PD-L2, and JAK2, which at least in part accounts for the
WNT-β catenin pathway in non–T-cell inflamed tumors. vulnerability of classic Hodgkin lymphoma to PD-1
Preclinical models suggest a causal relationship between β blockade. 83,84 Although 97% of patients with classic
catenin activation and minimal T-cell infiltration as well as Hodgkin lymphoma have alterations in PD-L1 and PD-
resistance to ICIs. A proposed mechanism is upregulation of L2 loci, only 0.7% of 118,187 sequenced solid tumor
a transcriptional repressor, reduced secretion of the che- samples had PD-L1 amplification with FoundationOne
mokine CCL4, and impaired recruitment of CD103+ dendritic testing.1 PD-L1 amplification was most prevalent in mixed
cells, leading to failed priming and recruitment of CD8+ T cells hepatocellular cholangiocarcinoma (10.5%; 19 tumors),
to the tumor. Additionally, β catenin inhibition combined with bladder squamous cell carcinoma (7.5%; 40 tumors), renal
anti–PD-1/CTLA-4 blockade in preclinical models sensitizes sarcomatoid carcinoma (6.1%; 66 tumors), nasopharyngeal
noninflamed tumors to potentiate tumor growth inhibition carcinoma (5.1%; 99 tumors), and thyroid anaplastic car-
across a range of tumor types, including melanoma, neuro- cinoma (5.1%; 177 tumors). Interestingly, PD-L1 amplifi-
blastoma, mammary carcinoma, and renal adenocarci- cation does not always correlate with high PD-L1 expression
noma.73 Furthermore, in the Cancer Genome Atlas data on IHC, and 85% of PD-L1–amplified tumors have low to
set, 90% of non–T-cell-inflamed tumors demonstrated intermediate TMB. Of the PD-L1–amplified patients, nine
higher β catenin activity, in part through DNA alterations who had available clinical data were treated with an ICI, with
in WNT-related genes.74 a response rate of 66.7%.85 PD-L1 amplification is a rare
event in solid tumors, and reports of ICI treatment are limited
Beta-2 Microglobulin
to a small cohort, highlighting the need for additional pro-
Beta-2 microglobulin (B2M) combines with the heavy chain spective studies to validate this is a biomarker.
to create an MHC class I complex, which presents peptides
to CD8+ T cells. Loss of B2M results in the inability to present CONCLUSION
antigens via the MHC class I complex and has long been In summary, efforts are underway to identify biomarkers that
recognized as a mechanism of immune evasion.75 The can reliably predict which patients will respond to ICIs.
prevalence of B2M mutations is best described in colorectal Although PD-L1 expression by IHC is an established bio-
cancer, with somatic mutations identified in approximately marker with FDA-approved companion diagnostic testing, it
30% of dMMR colorectal cancers but in less than 2% of is not sufficient to stratify patients to or withhold patients
preserved MMR colorectal cancers. The presence of coding from potentially beneficial therapy. Building on experiences
microsatellites within the B2M gene makes it susceptible to with PD-L1, other tumor characteristics, including readily
mutations in dMMR cancer.76 B2M mutations are less testable genomic features, as summarized in Table 1, may
common in other cancer types, but alterations in B2M help select patients for treatment with ICIs. Some of these
expression can also occur through loss of heterozygosity, features are well established, including MSI, which has an
which has a high prevalence in bladder cancer (44%) and FDA-approved indication for anti–PD-L1 blockade, and
a moderate prevalence in melanoma (16%).77 Mutations TMB, for which levels correlate with ICI response across
and loss of heterozygosity of B2M are associated with ac- many tumor types. Other features, including specific
quired resistance to anti–PD-1 therapy in patients with genetic mutations, such as STK11, KEAP1, and PTEN,

TABLE 1. Characteristics of Commercially Available Comprehensive Genomic Profiling Platforms Specifying the Inclusion of Features That May Be Relevant to ICI Response
Genomic
Signature Somatic Gene Variants of Interest
No. of Matched Neo-
Targeted Sample Normal MSI PD-L1 Antigen HLA
Test Name Company Genes Type Blood TMB Status STK11 KEAP1 PTEN CTNNB1 APC AXIN1 AXIN2 B2M Amplification Prediction Typing
FoundationOne Foundation 324 FFPE X X X X X X X X X
CDx Medicine
MSK-IMPACT MSKCC 468 FFPE X X X X X X X X X X X X
Tempus xT Tempus 648 FFPE X X X X X X X X X X X X X X
OmniSeq OmniSeq 144 FFPE X X X X X X X
Advance
TruSight Illumina 523 FFPE X X X X X X X X X X
Oncology 500
CANCERPLEX KEW 435 FFPE X X X X X X X X X X
Guardant360 Guardant 73 Blood X X X X
FoundationOne Foundation 70 Blood X X X X X X
Liquid Medicine
Abbreviations: ICI, immune checkpoint inhibitor; TMB, tumor mutational burden; MSI, microsatellite instability; HLA, human leukocyte antigen; FFPE, formalin fixed, paraffin embedded; MSKCC,
Memorial Sloan Kettering Cancer Center.

2020 ASCO EDUCATIONAL BOOK | asco.org/edbook
e53
comprise an area of ongoing investigation. . Currently, the into clinical practice. Furthermore, broad application of
presence of these mutations does not influence treatment our evolving understanding of tumor immune biology will
decisions, but, as our understanding of their immuno- yield novel biomarkers and will improve our clinical
modulating effects develops, they may be incorporated decision-making.
AFFILIATION AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

1
Columbia University Medical Center, New York, NY AND DATA AVAILABILITY STATEMENT
EDBK_289967.
Naiyer A. Rizvi, Columbia University Medical Center, 177 Fort Washington
Ave., New York, NY 10032; email: [email protected].
REFERENCES
1. Roemer MG, Advani RH, Ligon AH, et al. PD-L1 and PD-L2 genetic alterations define classical Hodgkin lymphoma and predict outcome. J Clin Oncol. 2016;
34:2690-2697.
2. Topalian SL, Taube JM, Anders RA, et al. Mechanism-driven biomarkers to guide immune checkpoint blockade in cancer therapy. Nat Rev Cancer. 2016;
16:275-287.
3. Taube JM, Anders RA, Young GD, et al. Colocalization of inflammatory response with B7-h1 expression in human melanocytic lesions supports an adaptive
resistance mechanism of immune escape. Sci Transl Med. 2012;4:127ra37.
4. Brahmer JR, Drake CG, Wollner I, et al. Phase I study of single-agent anti–programmed death-1 (MDX-1106) in refractory solid tumors: safety, clinical activity,
pharmacodynamics, and immunologic correlates. J Clin Oncol. 2010;28:3167-3175.
5. Topalian SL, Hodi FS, Brahmer JR, et al. Safety, activity, and immune correlates of anti–PD-1 antibody in cancer. N Engl J Med. 2012;366:2443-2454.
6. Garon EB, Rizvi NA, Hui R, et al; KEYNOTE-001 Investigators. Pembrolizumab for the treatment of non–small cell lung cancer. N Engl J Med. 2015;
372:2018-2028.
7. Reck M, Rodrı́guez-Abreu D, Robinson AG, et al; KEYNOTE-024 Investigators. Pembrolizumab versus chemotherapy for PD-L1–positive non–small cell lung
8. Ferris RL, Blumenschein G Jr., Fayette J, et al. Nivolumab for recurrent squamous cell carcinoma of the head and neck. N Engl J Med. 2016;375:1856-1867.
9. Rosenberg JE, Hoffman-Censits J, Powles T, et al. Atezolizumab in patients with locally advanced and metastatic urothelial carcinoma who have progressed
following treatment with platinum-based chemotherapy: a single-arm, multicentre, phase 2 trial. Lancet. 2016;387:1909-1920.
10. Hellmann MD, Nathanson T, Rizvi H, et al. Genomic features of response to combination immunotherapy in patients with advanced non–small cell lung cancer.
Cancer Cell. 2018;33:843-852.e4.
11. Carbone DP, Reck M, Paz-Ares L, et al; CheckMate 026 Investigators. First-line nivolumab in stage IV or recurrent non–small cell lung cancer. N Engl J Med.
2017;376:2415-2426.
12. Bellmunt J, de Wit R, Vaughn DJ, et al; KEYNOTE-045 Investigators. Pembrolizumab as second-line therapy for advanced urothelial carcinoma. N Engl J Med.
2017;376:1015-1026.
13. Brahmer J, Reckamp KL, Baas P, et al. Nivolumab versus docetaxel in advanced squamous cell non–small cell lung cancer. N Engl J Med. 2015;373:123-135.
14. Motzer RJ, Escudier B, McDermott DF, et al; CheckMate 025 Investigators. Nivolumab versus everolimus in advanced renal-cell carcinoma. N Engl J Med. 2015;
373:1803-1813.
15. Davis AA, Patel VG. The role of PD-L1 expression as a predictive biomarker: an analysis of all US Food and Drug Administration (FDA) approvals of immune
checkpoint inhibitors. J Immunother Cancer. 2019;7:278.
16. Sunshine J, Taube JM. PD-1/PD-L1 inhibitors. Curr Opin Pharmacol. 2015;23:32-38.
17. Hellmann MD, Ciuleanu TE, Pluzanski A, et al. Nivolumab plus ipilimumab in lung cancer with a high tumor mutational burden. N Engl J Med. 2018;
378:2093-2104.
18. Rooney MS, Shukla SA, Wu CJ, et al. Molecular and genetic properties of tumors associated with local immune cytolytic activity. Cell. 2015;160:48-61.
19. Van Allen EM, Miao D, Schilling B, et al. Genomic correlates of response to CTLA-4 blockade in metastatic melanoma. Science. 2015;350:207-211.
20. Dudley JC, Lin MT, Le DT, et al. Microsatellite instability as a biomarker for PD-1 blockade. Clin Cancer Res. 2016;22:813-820.
21. Germano G, Amirouchene-Angelozzi N, Rospo G, et al. The clinical impact of the genomic landscape of mismatch repair-deficient cancers. Cancer Discov. 2018;
8:1518-1528.

2019;25:3753-3758.
25. Mehnert JM, Panda A, Zhong H, et al. Immune activation and response to pembrolizumab in POLE-mutant endometrial cancer. J Clin Invest. 2016;
126:2334-2340.
26. Santin AD, Bellone S, Buza N, et al. Regression of chemotherapy-resistant polymerase " (POLE) ultra-mutated and msh6 hyper-mutated endometrial tumors with
nivolumab. Clin Cancer Res. 2016;22:5682-5687.
27. Alexandrov LB, Nik-Zainal S, Wedge DC, et al; ICGC PedBrain. Signatures of mutational processes in human cancer [published correction appears in Nature.
2013;502:258]. Nature. 2013;500:415-421.
28. Snyder A, Makarov V, Merghoub T, et al. Genetic basis for clinical response to CTLA-4 blockade in melanoma. N Engl J Med. 2014;371:2189-2199.
29. Rizvi NA, Hellmann MD, Snyder A, et al. Mutational landscape determines sensitivity to PD-1 blockade in non–small cell lung cancer. Science. 2015;
348:124-128.
30. Rizvi H, Sanchez-Vega F, La K, et al. Molecular determinants of response to anti–programmed cell death (PD)-1 and anti–programmed death-ligand 1 (PD-L1)
blockade in patients with non–small cell lung cancer profiled with targeted next-generation sequencing. J Clin Oncol. 2018;36:633-641.
31. Powles T, Durán I, van der Heijden MS, et al. Atezolizumab versus chemotherapy in patients with platinum-treated locally advanced or metastatic urothelial
carcinoma (IMvigor211): a multicentre, open-label, phase 3 randomised controlled trial. Lancet. 2018;391:748-757.
32. Seiwert TY, Haddad R, Bauml J, et al. Biomarkers predictive of response to pembrolizumab in head and neck cancer (HNSCC). Cancer Res. 2018;78 (suppl;
abstr LB-339).
33. Hellmann MD, Callahan MK, Awad MM, et al. Tumor mutational burden and efficacy of nivolumab monotherapy and in combination with ipilimumab in small-cell
lung cancer [published correction appears in Cancer Cell. 2019:35;329]. Cancer Cell. 2018;33:853-861.e4.
34. Yarchoan M, Hopkins A, Jaffee EM. Tumor mutational burden and response rate to PD-1 inhibition. N Engl J Med. 2017;377:2500-2501.
35. Cristescu R, Mogg R, Ayers M, et al. Pan-tumor genomic biomarkers for PD-1 checkpoint blockade-based immunotherapy. Science. 2018;362:eaar3593.
36. Miao D, Margolis CA, Vokes NI, et al. Genomic correlates of response to immune checkpoint blockade in microsatellite-stable solid tumors. Nat Genet. 2018;
50:1271-1281.
37. Samstein RM, Lee CH, Shoushtari AN, et al. Tumor mutational load predicts survival after immunotherapy across multiple cancer types. Nat Genet. 2019;
51:202-206.
38. Szustakowski JD, Green G, Geese WJ, et al. Evaluation of tumor mutation burden as a biomarker for immune checkpoint inhibitor efficacy: a calibration study of
whole exome sequencing with FoundationOne®. Cancer Res. 2018;78 (suppl; abstr 5528).
39. Chalmers ZR, Connelly CF, Fabrizio D, et al. Analysis of 100,000 human cancer genomes reveals the landscape of tumor mutational burden. Genome Med. 2017;
9:34.
40. Johnson DB, Frampton GM, Rioth MJ, et al. Targeted next generation sequencing identifies markers of response to PD-1 blockade. Cancer Immunol Res. 2016;
4:959-967.
41. Zehir A, Benayed R, Shah RH, et al. Mutational landscape of metastatic cancer revealed from prospective clinical sequencing of 10,000 patients [published
correction appears in Nat Med. 2017;23:1004]. Nat Med. 2017;23:703-713.
42. Büttner R, Longshore JW, López-Rı́os F, et al. Implementing TMB measurement in clinical practice: considerations on assay requirements. ESMO Open. 2019;
4:e000442.
43. Garofalo A, Sholl L, Reardon B, et al. The impact of tumor profiling approaches and genomic data strategies for cancer precision medicine. Genome Med. 2016;
8:79.
44. Jaiswal S, Fontanillas P, Flannick J, et al. Age-related clonal hematopoiesis associated with adverse outcomes. N Engl J Med. 2014;371:2488-2498.
45. Chan TA, Yarchoan M, Jaffee E, et al. Development of tumor mutation burden as an immunotherapy biomarker: utility for the oncology clinic. Ann Oncol. 2019;
30:44-56.
46. Kowanetz M, Zou W, Shames D, et al. Tumor mutation burden (TMB) is associated with improved efficacy of atezolizumab in 1L and 2L+ NSCLC patients.
J Thorac Oncol. 2017;12 (suppl; abstr OA20.01).
47. Ready N, Hellmann MD, Awad MM, et al. First-line nivolumab plus ipilimumab in advanced non-small-cell lung cancer (CheckMate 568): outcomes by
programmed death ligand 1 and tumor mutational burden as biomarkers. J Clin Oncol. 2019;37:992-1000.
48. Stenzinger A, Allen JD, Maas J, et al. Tumor mutational burden standardization initiatives: recommendations for consistent tumor mutational burden assessment
in clinical samples to guide immunotherapy treatment decisions. Genes Chromosomes Cancer. 2019;58:578-588.
49. Chang H, Sasson A, Srinivasan S, et al. Bioinformatic methods and bridging of assay results for reliable tumor mutational burden assessment in non–small cell
lung cancer. Mol Diagn Ther. 2019;23:507-520.
50. Gandara DR, Paul SM, Kowanetz M, et al. Blood-based tumor mutational burden as a predictor of clinical benefit in non–small cell lung cancer patients treated
with atezolizumab. Nat Med. 2018;24:1441-1448.
51. Peters S, Cho BY, Reinmuth N, et al. Tumor mutational burden (TMB) as a biomarker of survival in metastatic non–small cell lung cancer (mNSCLC): Blood and
tissue TMB analysis from MYSTIC, a Phase III study of first-line durvalumab 6tremelimumab vs chemotherapy. AACR Proc. 2019;79 (suppl; abstr CT074).

52. Chen Y, Liu Q, Chen Z, et al. PD-L1 expression and tumor mutational burden status for prediction of response to chemotherapy and targeted therapy in non–small
cell lung cancer. J Exp Clin Cancer Res. 2019;38:193.
53. Bristol-Myers Squibb. Bristol-Myers Squibb provides update on the ongoing regulatory review of Opdivo plus low-dose Yervoy in first-line lung cancer patients with
tumor mutational burden 10 mut/Mb. https://news.bms.com/press-release/corporatefinancial-news/bristol-myers-squibb-provides-update-ongoing-
regulatory-review. Accessed March 1, 2020.
54. AstraZeneca. Update on the Phase III NEPTUNE trial of Imfinzi plus tremelimumab in Stage IV non–small cell lung cancer. Accessed March 1, 2020.
55. Knepper TC, Montesion M, Russell JS, et al. The genomic landscape of Merkel cell carcinoma and clinicogenomic biomarkers of response to immune checkpoint
inhibitor therapy. Clin Cancer Res. 2019;25:5961-5971.
56. Maia MC, Almeida L, Bergerot PG, et al. Relationship of tumor mutational burden (TMB) to immunotherapy response in metastatic renal cell carcinoma (mRCC).
57. Turajlic S, Litchfield K, Xu H, et al. Insertion-and-deletion-derived tumour-specific neoantigens and the immunogenic phenotype: a pan-cancer analysis. Lancet
Oncol. 2017;18:1009-1021.
58. Panda A, de Cubas AA, Stein M, et al. Endogenous retrovirus expression is associated with response to immune checkpoint blockade in clear cell renal cell
carcinoma. JCI Insight. 2018;3:e121522.
59. Smith CC, Beckermann KE, Bortone DS, et al. Endogenous retroviral signatures predict immunotherapy response in clear cell renal cell carcinoma. J Clin Invest.
2018;128:4804-4820.
60. Wellenstein MD, de Visser KE. Cancer-cell-intrinsic mechanisms shaping the tumor immune landscape. immunity. 2018;48:399-416.
61. Momcilovic M, Shackelford DB. Targeting LKB1 in cancer: exposing and exploiting vulnerabilities. Br J Cancer. 2015;113:574-584.
62. Facchinetti F, Bluthgen MV, Tergemina-Clain G, et al. LKB1/STK11 mutations in non-small cell lung cancer patients: descriptive analysis and prognostic value.
Lung Cancer. 2017;112:62-68.
63. Skoulidis F, Goldberg ME, Greenawalt DM, et al. STK11/LKB1 mutations and PD-1 inhibitor resistance in KRAS-mutant lung adenocarcinoma. Cancer Discov.
2018;8:822-835.
64. Koyama S, Akbay EA, Li YY, et al. STK11/lkb1 deficiency promotes neutrophil recruitment and proinflammatory cytokine production to suppress T-cell activity in
the lung tumor microenvironment. Cancer Res. 2016;76:999-1008.
65. Nadal E, Palmero R, Muñoz-Pinedo C. Mutations in the antioxidant KEAP1/NRF2 pathway define an aggressive subset of NSCLC resistant to conventional
treatments. J Thorac Oncol. 2019;14:1881-1883.
66. Arbour KC, Jordan E, Kim HR, et al. Effects of co-occurring genomic alterations on outcomes in patients with KRAS-mutant non–small cell lung cancer. Clin
Cancer Res. 2018;24:334-340.
67. Chen X, Su C, Ren S, et al. Pan-cancer analysis of KEAP1 mutations as biomarkers for immunotherapy outcomes. Ann Transl Med. 2020;8:141.
68. Rizvi N, Cho BC, Reinmuth N, et al. Mutations associated with sensitivity or resistance to immunotherapy in mNSCLC: analysis from the MYSTIC Trial. J Thorac
Oncol. 2019;14 (suppl; abstr OA04.07).
69. Peng W, Chen JQ, Liu C, et al. Loss of PTEN promotes resistance to T cell-mediated immunotherapy. Cancer Discov. 2016;6:202-216.
70. Logan CY, Nusse R. The Wnt signaling pathway in development and disease. Annu Rev Cell Dev Biol. 2004;20:781-810.
71. Kim S, Jeong S. Mutation hotspots in the β-catenin gene: lessons from the human cancer genome databases. Mol Cells. 2019;42:8-16.
72. Spranger S, Bao R, Gajewski TF. Melanoma-intrinsic β-catenin signalling prevents anti-tumour immunity. Nature. 2015;523:231-235.
73. Ganesh S, Shui X, Craig KP, et al. RNAi-mediated β-catenin inhibition promotes T cell infiltration and antitumor activity in combination with immune checkpoint
blockade. Mol Ther. 2018;26:2567-2579.
74. Luke JJ, Bao R, Sweis RF, et al. WNT/β-catenin pathway activation correlates with immune exclusion across human cancers. Clin Cancer Res. 2019;
25:3074-3083.
75. Restifo NP, Marincola FM, Kawakami Y, et al. Loss of functional beta 2-microglobulin in metastatic melanomas from five patients receiving immunotherapy. J Natl
Cancer Inst. 1996;88:100-108.
76. Barrow P, Richman SD, Wallace AJ, et al. Confirmation that somatic mutations of beta-2 microglobulin correlate with a lack of recurrence in a subset of stage II
mismatch repair deficient colorectal cancers from the QUASAR trial. Histopathology. 2019;75:236-246.
77. Maleno I, Aptsiauri N, Cabrera T, et al. Frequent loss of heterozygosity in the β2-microglobulin region of chromosome 15 in primary human tumors. Im-
munogenetics. 2011;63:65-71.
78. Zaretsky JM, Garcia-Diaz A, Shin DS, et al. Mutations associated with acquired resistance to PD-1 blockade in melanoma. N Engl J Med. 2016;375:819-829.
79. Pereira C, Gimenez-Xavier P, Pros E, et al. Genomic profiling of patient-derived xenografts for lung cancer identifies B2M inactivation impairing immunor-
ecognition. Clin Cancer Res. 2017;23:3203-3213.
80. Yeon Yeon S, Jung SH, Jo YS, et al. Immune checkpoint blockade resistance-related B2M hotspot mutations in microsatellite-unstable colorectal carcinoma.
Pathol Res Pract. 2019;215:209-214.
81. Sade-Feldman M, Jiao YJ, Chen JH, et al. Resistance to checkpoint blockade therapy through inactivation of antigen presentation. Nat Commun. 2017;8:1136.
82. Rodig SJ, Gusenleitner D, Jackson DG, et al. MHC proteins confer differential sensitivity to CTLA-4 and PD-1 blockade in untreated metastatic melanoma. Sci
Transl Med. 2018;10:10eaar3342.

83. Chen R, Zinzani PL, Fanale MA, et al; KEYNOTE-087. Phase II study of the efficacy and safety of pembrolizumab for relapsed/refractory classic Hodgkin
lymphoma. J Clin Oncol. 2017;35:2125-2132.
84. Ansell SM, Lesokhin AM, Borrello I, et al. PD-1 blockade with nivolumab in relapsed or refractory Hodgkin’s lymphoma. N Engl J Med. 2015;372:311-319.
85. Goodman AM, Piccioni D, Kato S, et al. Prevalence of PD-L1 amplification and preliminary response to immune checkpoint blockade in solid tumors. JAMA
Oncol. 2018;4:1237-1244.

Biomarkers in Precision Cancer Immunotherapy:

Promise and Challenges
William B. McKean, MD, PhD1; Justin C. Moser, MD2; David Rimm, MD, PhD3; and Siwen Hu-Lieskovan, MD, PhD1
overview
The rapid expansion of modern cancer immunotherapeutics has led to a dramatic improvement in patient
survival and sustained remission for otherwise refractory malignancies. However, a significant limitation
behind these current treatment modalities is an irregularity in clinical response, which is especially pro-
nounced among checkpoint inhibition. This unpredictability leads to significant side effects, financial costs,
and health care burden, with unsatisfactory clinical benefit in the majority of treated patients. Additionally,
although ongoing studies and trials investigate the use of multiple biomarkers predictive of patient response or
harm, none of these are comprehensive in predicting potential benefit. This unmet need for validated bio-
markers is largely secondary to a prohibitive complexity within tumor parenchyma and microenvironment,
dynamic clonal and proteomic changes to therapy, heterogenous host immune defects, and varied stan-
dardization among sample preparation and reporting. Herein, we discuss current advantages of predictive
biomarkers, as well as limitations in their clinical use and application. We also review future directions, ideal
characteristics, and trial design needed for proper precision immuno-oncology and biomarker development.
INTRODUCTION Adoptive cell therapies have also shown sustained

Immuno-oncology has witnessed a revolution of basic treatment effects among otherwise refractory and
science advances and novel clinical therapies over the heavily pretreated patients. Since 2017, FDA approval
past decade. Advanced malignancies, such as stage IV of tisagenlecleucel for B-cell precursor ALL and axi-
melanoma, lung cancer, and relapsed/refractory acute cabtagene ciloleucel for diffuse large B-cell lymphoma
lymphoblastic leukemia (ALL), have shown dramatic has demonstrated impressive CR rates (approximately
improvements in overall survival (OS), progression-free 90% and 50%, respectively).3 In many cases, there is
survival (PFS), and complete response (CR) with the also evidence of chimeric transgene sequences per-
introduction of therapies such as immune checkpoint sisting within peripheral blood for years after initial
inhibition (ICI), bispecific antibodies, and chimeric treatment, suggestive of a more permanent cancer
antigen receptor (CAR) T cells. “immune surveillance.”4,5 Such promising features of
CAR T-cell therapy have inspired the rapid expansion
Anti–CTLA-4 therapy was originally approved by the of clinical studies in solid and hematologic malig-
U.S. Food and Drug Administration (FDA) in 2011 for nancies alike, with a global increase to 364 active trials
advanced melanoma following multiple clinical ob- by early 2019.6
servations that it prolonged OS.1 Since that time, the
repertoire of checkpoint inhibitors and their onco- Despite these successes, the majority of patients who
logic targets have expanded dramatically. With the qualify for immune therapy derive no benefit, as a re-
incorporation of PD-1 and PD-L1 targets, six additional sult of primary or secondary resistance to therapy.
antibodies have been approved for use against a variety Such treatment failures occur even in ICI patient co-
Author affiliations of malignancies, from cervical cancer to advanced horts, despite respective prescreening with biomarkers
and support gastroesophageal carcinomas. Recent estimates sug- such as PD-L1 tumor proportion scores or combined
information (if
gest that patient candidacy for ICI has increased from positive scores. Other predictive models such as tumor
applicable) appear mutation burden (TMB), although validated in several
at the end of this 1.54% in 2011 to 43.63% in 2018, and total re-
sponders have increased from less than 0.14% to clinical trials,7-10 show limitations under a variety of
article.
12.46%.2 An especially attractive feature of ICI has circumstances, including prior immunotherapy,11-14
Accepted on April
24, 2020 and been a durable response in subsets of patients, evi- mutational variation (including copy number alter-
published at denced by a plateau of the treatment “tails” in Kaplan- ations and indels),15 or concomitant chemotherapy.
ascopubs.org on
Meier survivorship curves. Rather than ultimately Importantly, although more predictive models of ICI
doi.org/10.1200/ descending to baseline, these curves flatten before the response have recently been identified, broader ap-
EDBK_280571 end of surveillance. plication is incomplete. Meta-analysis and systematic

McKean et al
downregulate priming and activity of cytotoxic responses.

Following initial cloning and characterization of the key
components,80 the PD-1/PD-L1 checkpoint was specifically
• Understanding advances in current biomarker shown to be manipulated by malignant cells to avoid host
design helps to guide appropriate application of
immunity.81 Concurrently, data were published reporting
immunotherapy.
that anti–CTLA-4 therapy promotes tumor shrinkage and
• Tumoral complexity, clonal evolution, and secondary immunity in preclinical models of sarcoma and
heterogenous host immune defects limit the colorectal adenocarcinoma.82 ICI has since become the
utility and predictability of standard biomarker
most prolific approved cancer immunotherapy, highlighted
options.
by Dr. Tasuku Honjo and Dr. James Allison receiving the
• Ideal biomarker design will require a multifac- Nobel Prize in Physiology or Medicine 2018 for their pio-
eted and largely individualized design to guide neering contributions to the field.
therapy selection and duration.
PD-L1
• Further validation of novel biomarkers through
adaptive clinical trials will require more com- PD-L1 is the only biomarker validated for its predictive utility
prehensive data collection before and during in prospective clinical trials and approved by the FDA as
treatment. a companion diagnostic prior to ICI therapy. Anti–PD-1 was
approved in metastatic non–small cell lung cancer (NSCLC)
review show similar correlation between patient response to as first-line monotherapy for patients with partial or com-
PD-1/PD-L1 therapy and TMB, PD-L1 immunohistochem- plete surface PD-L1 staining (tumor proportion scores) on
istry (IHC), or gene expression profiles (GEPs).16 However, 1% or more of malignant cells.19 Related immunohisto-
improved sensitivity, specificity, positive predictive value, chemical tests have been developed that measure PD-L1+
and positive likelihood ratio are seen with multiplex IHC tumor and immune cells relative to total tumor population
(mIHC)/immunofluorescence (IF). Such advances require (combined positive score). This assay is required for
further clinical use and development. anti–PD-1 approval in a variety of other indications, in-
cluding first-line treatment for advanced head and neck
Another complication of modern immunotherapy is the squamous cell carcinoma (HNSCC),20 second-line treat-
development of unpredictable and unique toxicities. Immune- ment for esophageal squamous cell carcinoma,21,22 third-
related adverse events (irAEs), which are considered sec- line treatment for recurrent gastric/esophagogastric junction
ondary to off-target inflammation induced by ICI therapy, can adenocarcinoma,23 second-line treatment for progressive
have a widely variable onset and affect a diversity of tissues or cervical cancer,24 and bacillus Calmette-Guérin–refractory
organ systems. Similarly, patients receiving CAR T-cell therapy high-risk nonmuscle invasive bladder cancer. These and
must be closely monitored for symptoms related to immune other studies9,25-28 highlight the versatility of PD-L1 as
effector cell–associated neurotoxicity syndrome or cytokine a positive prognostic correlate for OS across a wide spec-
release syndrome (CRS). trum of malignancies.
Discrepancies in patient treatment responses and devel- Despite their broad utility, programmed death ligand levels
opment of unpredictable toxicities highlight a special need remain an incomplete test for universal prescription of
for predictive biomarkers in immuno-oncology to further checkpoint inhibitors. Multiple studies have shown an ab-
select appropriate patient treatment cohorts. 17,18 Un- sence of association between PD-L1 expression and OS in
fortunately, the current spectrum for such biomarkers is ICI therapy.7,13,83,84 Still other cohorts have patients with
extremely limited, especially those established in large minimal or absent PD-L1 tumor expression that derive
prospective clinical trials. Additionally, substantial variation sustained responses to checkpoint inhibition. A recent
between individual patient tumors, their microenvironment, retrospective analysis of clinical trials (2011–2019) prompting
and treatment history highlights the need for more per- FDA approval of checkpoint inhibitor regimens identified
sonalized diagnostics and therapy. This review describes PD-L1 as a predictive biomarker in only 28.9% of cases.85
recent advances and established biomarkers in cancer This is attributable, in part, to dynamic host immunity; rapid
immunotherapy (Table 1) while exploring limitations behind induction of PD-L1 levels following adaptive responses to
current predictive tests and models. We propose potential therapy may not be representative of baseline patient test-
trial design and ideal biomarker use to overcome these ing. Other potential explanations for these discrepancies are
limitations in search of more precision medicine. discussed in later sections.
IMMUNE CHECKPOINT INHIBITION BIOMARKERS Tumor Mutation Burden and Microsatellite Instability
Early discoveries in the 1990s demonstrated that specific The number of genetic alterations within a tumor genome is
T-cell protein receptors—PD-1 and CTLA-4—function to considered correlative with mutant protein burden and,

Biomarkers in Precision Cancer Immunotherapy
TABLE 1. Characteristics and Clinical Correlates of Biomarkers in Cancer Immunotherapy

Biomarker Immunotherapy Malignancy Example Tests Clinical Utility Tissue Source
PD-L1 Checkpoint inhibition Multiple TPS,19 CPS,20–24 IC expression25 CR/PR26 Tumor, infiltrating
21,23–25,27,28 lymphocytes/
ORR
macrophages
DCB/DOR21,23
PFS28
OS19,20,28
TMB Checkpoint inhibition Multiple ΔTMB 12
CR/PR13,25,29 Tumor
9,30–32
ORR
DCB/
DOR30,33,34
PFS9,30,31,34
OS12,13,31,34
MMR/MSI Checkpoint inhibition Multiple N/A ORR35 Tumor
35
PFS
OS35
36
Aneuploidy Checkpoint inhibition Multiple SCNA level CR/PR37 Tumor
36
OS
TIL Checkpoint inhibition Multiple Immunoscore,38–40 CTLA-4hiPD-1hi,41 CR/ Infiltrating
TRM,42,43 PD-1T,44 4PD1hi,45 PR12,41,44,47,48 lymphocytes
CYT12,46
ORR49,50
PFS41,50
OS42,44,48
51–55 56
GEP Checkpoint inhibition Multiple IFN-γ signature, IMPRES, CR/PR52,53,55–58 Tumor, infiltrating
TIDE,57 immunophenoscore58 54 lymphocytes
ORR
PFS53–57
OS51,54–57
mIHC/IF Checkpoint inhibition Melanoma, MCC, N/A CR/PR47,59,60 Tumor, infiltrating
NSCLC 50 lymphocytes
ORR
DCB/DOR61
PFS50,59,61
OS59,61
62 89
PET/CT imaging Checkpoint inhibition Multiple Immune-PET (i.e., ZR- CR/PR63,64 N/A
desferrioxamine), 18F-FDG62,63 64
PFS
OS64
Circulating CAR Adoptive cell therapy B-cell precursor N/A ORR65 Peripheral blood
T cells ALL, DLBCL
DFS66
Peripheral Checkpoint inhibition, Multiple Treg,67 CD27+CD45RO CD8+,68 CR/PR67,68 Peripheral blood
lymphocytes adoptive cell therapy, RLC69
OS69
T-cell engagers

McKean et al
TABLE 1. Characteristics and Clinical Correlates of Biomarkers in Cancer Immunotherapy (Continued)

Biomarker Immunotherapy Malignancy Example Tests Clinical Utility Tissue Source
70–77
Proinflammatory Checkpoint inhibition, Melanoma, B-cell IFN-γ, IL-13, MIP-1α, IL-6, IL-8, Toxicity Peripheral blood
cytokines adoptive cell therapy precursor ALL, sCD25, IL-17, sCD163, CXCL5
CLL, NHL
Autoantibodies Checkpoint inhibition Multiple Anti-thyroglobulin, anti-GAD65, anti- Toxicity78,79 Peripheral blood
IA2, anti-insulin, anti-ZnT8
Abbreviations: TPS, tumor proportion score; CPS, combined positive score; IC, infiltrating immune cell; CR, complete response; PR, partial response; ORR,
objective or overall response rate; DCB, durable clinical benefit; DOR, duration of response; PFS, progression-free survival; OS, overall survival; TMB, tumor
mutation burden; ΔTMB, change in TMB; MMR/MSI, mismatch repair/microsatellite instable; N/A, not applicable; SCNA, somatic copy number alteration;
TIL, tumor-infiltrating lymphocyte; TRM, tissue-resident memory T cell; PD-1T, intratumoral CD8+ T cell PD-1 elevation; 4PD1hi, unconventional PD-
1–expressing CD4+FOXP3 subpopulation; CYT, cytolytic score; IMPRES, immune-predictive score; GEP, gene expression profile; IFN-γ, interferon gamma;
TIDE, tumor immune dysfunction and exclusion; mIHC, multiplex immunohistochemical; IF, immunofluorescence; MCC, Merkel cell carcinoma; NSCLC,
non–small cell lung cancer; FDG, fluorodeoxyglucose; CAR, chimeric antigen receptor; ALL, acute lymphocytic leukemia; DLBCL, diffuse large B-cell
lymphoma; DFS, disease-free survival; Treg, regulatory T cell; RLC, relative lymphocyte count; CLL, chronic lymphocytic leukemia; NHL, non-Hodgkin
lymphoma; IL-13, interleukin-13; MIP-1α, macrophage inflammatory protein 1α; IL-6, interleukin-6; IL-8, interleukin-8; sCD25, soluble interleukin-2
receptor; IL-17, interleukin 17; sCD163, soluble cluster of differentiation 163; CXCL5, CXC motif chemokine 5; GAD65, glutamic acid decarboxylase 65; IA2,
α-islet antigen 2; ZnT8, zinc transporter 8.
thus, host immunity against “non-self” neoantigens. Non- MMR deficiency independently correlates with ICI efficacy.
synonymous single nucleotide variants, collectively known Tumors that demonstrate deficiency in MMR (and, thus, are
as TMB, are the most commonly associated mutational high in MSI) have improved response to checkpoint in-
variant with ICI therapy response. Initial studies evaluating hibition. These data were first demonstrated in phase II
cohorts of patients with melanoma treated with ipilimumab29,33 studies evaluating anti–PD-1 therapy among patients with
and patients with NSCLC on pembrolizumab30 demonstrated progressive and heavily pretreated metastatic colorectal
higher levels of nonsynonymous single nucleotide variants carcinoma.35 Objective response rate (ORR) to pembrolizumab
among patients with improved clinical response and PFS. among MSI-high cohorts increased to 40% (compared with
Since these data, positive associations with checkpoint in- 0% in MSI-stable), and median PFS and OS were not
hibitor response and TMB have been demonstrated in several reached. Subsequent experiments among multiple tumor
other malignancies, including small cell lung cancer,31 uro- types (including prostate, pancreas, thyroid, neuroendo-
thelial carcinoma,25 and HPV HNSCC.13 TMB may have even crine, endometrial, gastroesophageal, and biliary) repro-
broader predictive applications across a range of tumor types. duced response and survival benefits among patients with
For example, a meta-analysis assessing objective response MMR deficiency (radiographic response, 53%; CR, 21%;
rate to ICI therapy across 27 cancer subtypes demon- median PFS and OS not reached).87 Accelerated approval for
strated positive correlation in linear regression with logarithmic pembrolizumab as second-line therapy for MMR-deficient
measurements of TMB.32 Indeed, a very high response rate to solid malignancies was granted by the FDA consequent to
anti–PD-1 therapy was seen in a cohort of patients with these findings.
desmoplastic melanoma, one of the tumor types that has the Although TMB appears promising, there are substantial
highest TMB.86 More recently, high nonsynonymous somatic concerns related to its clinical implementation. Among the
TMB assessed by next-generation sequencing correlated more important of these is the lack of assay standardization.
positively with OS among a large cohort of patients with ad- Although a number of standardization efforts have begun,
vanced or metastatic cancers (including NSCLC, melanoma, none have yet been published. Standardization is further
HNSCC, renal cell carcinoma, and bladder cancer; 1,662 complicated by different tests using different criteria. For
patients).34 This study suggests an important feature of TMB as instance, one of the more popular assays, the Foundation
a biomarker: even relative increases among histologic sub- Medicine test, uses synonymous and nonsynonymous
types with canonically low mutational burden correlate with mutations, in conflict with the original concept of neoantigen
immunotherapy response. In these data, the top 20% of so- detection (by definition, synonymous mutations do not
matic nonsynonymous single nucleotide variants relative to generate neoantigens). An equally significant issue is the
specific cancer histologies was predictive of improved OS definition of the TMB cut-point. For example, although some
within each group. groups used 10 mutations per megabase pair, others used
Mismatch repair (MMR) status is an underlying genetic different values. Perhaps the best analysis of this issue is
process that contributes to neoantigen formation and TMB. illustrated by Samstein et al34; the hazard ratio (HR) for OS
The presence of microsatellite instability (MSI) secondary to following ICI is plotted against TMB cut-points among

multiple advanced cancers. This curve shows that the HR PD-1 levels.44 Importantly, increased PD-1T CD8+ T-cell
declines as mutations per megabase pair increase, reaching levels strongly improved OS and response to nivolumab
a plateau at about 23 to 25 mutations per megabase pair. in this cohort. Similarly, among patients with metastatic
These data suggest that the cut-point for clinical testing melanoma, increased TIL expression of PD-1 and CTLA-4
should be more than 20 mutations per megabase pair; (CTLA-4hiPD-1hi) correlates with improved response and
however, the disadvantage of this threshold is that only PFS after treatment with anti–PD-1 therapy.41 Conversely,
a small percentage of any population exceeds that level preclinical studies suggest that more chronic stimulation of
of TMB. PD-1 surface expression can terminally exhaust CD8+
T cells through regulatory T-cell (Treg) recruitment or epi-
Tumor-Infiltrating Lymphocytes
genetic or transcriptional modulation,96-100 blunting their
Checkpoint inhibition unleashes cytotoxic T-cell activity response to ICI. Additionally, unconventional PD-1–expressing
through direct blockade of CTLA-4 or PD-1 checkpoint CD4+FOXP3 subpopulations are proportional to tumor bur-
signaling. Within tumoral stroma and parenchyma, this den among patients with melanoma or NSCLC and correlate
serves to reactivate otherwise exhausted effector T-cell inversely with OS among cohorts with NSCLC cohorts.45 Un-
populations made quiescent by antihost countermea- usually, CTLA-4 inhibition is shown to increase unconven-
sures. As a correlative, the density and location of T lym- tional PD-1–expressing CD4+FOXP3 subpopulation cell
phocytes within the tumor and its microenvironment could counts within tumors and peripheral blood, whereas PD-1
predict ICI response. For example, an “immune-inflamed” inhibition has the opposite effect.
signature is characterized by infiltration of CD8+ and CD4+
Other TIL immunophenotypes have been proposed as
T cells within tumoral parenchyma and in close proximity to
predictive biomarkers for ICI patient response. For example,
malignant cells.88 This phenotype is associated with im-
not all subsets of effector T cells subjected to prolonged
proved responses and OS among patients treated with
stimuli lose responsiveness to checkpoint inhibition. CD8+
checkpoint inhibition for a variety of malignancies.47,49
T cells coexpressing T cell factor 1 and PD-1, which are
Among cohorts with early-stage colorectal carcinoma,
present during chronic infections, are expanded intra-
quantification of cytotoxic and memory T-cell populations
tumorally and peripherally by checkpoint inhibition and
within tumoral cores and invasive margins (Immunoscore)
correlate positively with immunotherapy response.101-103 In
was more prognostic of survival and disease recurrence
addition, tissue-resident memory T cells—another recently
than MSI status or TNM staging.38,39,89,90 Similar tumor
identified population—have been linked with immune re-
profiling with Immunoscore is being evaluated in patients
sponse and prognosis among patients with solid tumors
with melanoma or NSCLC treated with ICI.40,91,92 In-
(e.g., breast cancer and NSCLC).42,43 Expression of integrins
terestingly, studies in melanoma suggest that the tumor-
such as CD103 and CD49a is thought to provide tissue-
infiltrating lymphocyte (TIL) density at the invasive margin
resident memory T cells with the ability to persist in non-
alone is most correlative with checkpoint inhibitor (anti–PD-
lymphoid tissues (including tumor parenchyma), whereas
1) response.47,50
increased levels of granzyme B, tumor necrosis factor-α,
By extension, tumors that lack parenchymal invasion of T interferon-γ (IFN-γ), PD-1, and CTLA-4 make them espe-
lymphocytes are resistant to ICI therapy. This can result from cially cytotoxic upon checkpoint inhibition.104,105 Even
T-cell exclusion to the surrounding stroma via WNT/β- simply measuring expression of cytolytic activity among
catenin or transforming growth factor-β signaling (“immune effector T cells may serve as an effective biomarker in ICI.
excluded”)48,88,93-95 or a complete lack of effector T-cell Among cohorts with progressive melanoma, the geometric
infiltration in the parenchyma or stroma (“immune des- mean of granzyme A and perforin 1 (cytolytic score) was
ert”). In broader terms, immune-excluded tumors, although shown to better predict response to nivolumab therapy than
sensitized to host immunity, actively evade detection with TIL density,12 and it correlates strongly with TMB.46
countermeasures that include active expression of PD-1 Importantly, although pathologic analysis of TILs can be
and CTLA-4. Alternatively, immune-desert tumors are com- predictive of response to immunotherapy, tissue biopsy is
pletely agnostic to host immune response. not possible in all patients. As a consequence, less invasive
In addition to density and intratumoral location, immu- biomarkers are needed. A significant advancement in pre-
nophenotype is a level of TIL diversity that can be in- cision medicine and biomarker classification is the use of
dependently associated with patient ICI response. For advanced imaging to characterize tumors.62-64 Of note,
instance, PD-1 expression among TILs correlates with a novel technique for detecting whole-body CD8+ T-cell
a variety of outcomes. In patients with stage IV NSCLC, subsets using 89ZR-desferrioxamine-labeled cys-diabodies
increased cell surface expression of PD-1 among intra- (immune-PET) has been validated in anti–PD-L1 preclinical
tumoral CD8+ T cells (PD-1T) disposes to improved tumor models and is currently undergoing a phase II clinical
reactivity compared with normal, absent, or virally induced trial.106

McKean et al
Gene Expression Profiles among immune-activating and immune-inhibitory states. For

Gene expression signatures is a rapidly expanding and instance, close analysis between spontaneously regressing or
comprehensive tool for evaluation of tumoral response to ICI. high-risk progressing neuroblastomas using RNA-sequencing
Primarily used in analysis of immunologic transcriptomic data led to the development of the immune-predictive score.56
patterns, multiple high-throughput genomic tests have been This utilizes 15 pairwise comparisons between immune
developed to predict sensitivity or resistance to checkpoint checkpoint expression profiles (including immunoinhibitory
inhibition. Among these signatures is the assessment of and immunostimulatory genes) and accurately predicts
expression patterns relating to IFN-γ and responsive genes. neuroblastoma response with an AUC of 0.9. Because of
possible correlations between immune regression in neuro-
For instance, in the POPLAR trial, pretreatment effector T-cell
blastoma and melanoma, immune-predictive score was
transcription patterns were analyzed in tumor samples of
validated among 10 independent cohorts of patients with
patients with progressive NSCLC receiving atezolizumab or
melanoma who were treated with ICI (AUC, 0.83); it also
docetaxel.51 These patterns incorporated expression levels of
correlated with PFS and OS. Similarly, transcriptomic patterns
CD8A, GZMA, GZMB, IFN-γ, EOMES, CXC motif chemokine
of tumoral immune escape have been used to predict patient
9 (CXCL9), CXCL10, and TBX21. OS was significantly im-
response to checkpoint inhibition. The tumor immune dys-
proved in patients receiving atezolizumab with high levels of
function and exclusion score uses Cox proportional hazards
this IFN-γ signature (HR, 0.43). Interestingly, PD-1/PD-L1
to compare gene expression signatures in tumor or immu-
pathway components were also assessed (including B7.1
nomodulatory cells with TIL levels.57 In cohorts with mela-
and PD-L2) and correlated with improved OS (HR, 0.46,
noma or NSCLC, this algorithm was more predictive of ICI
0.43, 0.45, and 0.39, respectively). In addition, subsequent
response than PD-L1 expression, IFN-γ signature, or TMB. It
data analyzing 18 genes (also largely IFN-γ–responsive tar-
also correlated strongly with PFS and OS among patients with
gets) among nine malignancies demonstrated a moderate
melanoma. Furthermore, the immunophenoscore has been
association with patient response to pembrolizumab.52 Im-
validated in two independent cohorts of patients with mel-
portantly, this “T-cell inflamed” GEP was prospectively vali-
anoma as being markedly superior (AUC, 1.0) in predicting
dated within the KEYNOTE-012 trial among patients with
response to checkpoint inhibition compared with TMB, cy-
HNSCC and was comparable to PD-L1 IHC predictive utility
tolytic score, or expression of CTLA-4, PD-L1, or PD-1.58 The
(area under the curve [AUC], 0.75 vs. 0.65).53 Furthermore,
score was derived using machine learning and incorporated
a smaller IFN-γ messenger RNA transcription pattern (in- metagene analysis unique to tumoral immune infiltrates with
corporating CD274, LAG3, CXCL9, and IFN-γ) was recently tumor genetic heterogeneity and clonality.
validated in a nonrandomized phase Ib/II clinical trial among
patients with NSCLC or urothelial cancer.54 Treatment with Multiplex immunohistochemistry and immunofluorescence
durvalumab among patients with elevated signature levels A unique approach to pathologic analysis of tumoral het-
was associated with higher overall response and longer erogeneity and treatment response is mIHC/IF. Multiple
median PFS and OS, also independent of PD-L1 IHC. In analytes within histologic preparations can be visualized
a retrospective collection of ICI-treated patients with mela- simultaneously with this technique and provide quantitative
noma, even single-gene (PD-L1) messenger RNA expression data about cellular spatial relationships. Earlier referenced
was associated with response.55 studies used this approach to determine the composition
and location of TILs among cohorts with melanoma dem-
An important concept within GEPs is that the predictive utility onstrating ICI response.47 Subsequent research has con-
of such algorithms may be dependent on individual therapy firmed the spatial density of PD-1 and PD-L1 within tumor
plans. For instance, recent studies evaluating the efficacy of parenchyma as a predictive biomarker. For instance, within
antihuman glucocorticoid-induced tumor necrosis factor re- a small discovery cohort of patients with metastatic mela-
ceptor antibodies among advanced solid malignancies failed noma, PD-1/PD-L1 proximity or coexpression of the major
to demonstrate predictability of IFN-γ GEPs to treatment.107 histocompatibility complex-II (HLA-DR) and indoleamine
This suggests that IFN-γ signaling and transcriptomic patterns 2,3-dioxygenase 1 was highly correlative with anti–PD-1
may correlate only with response to therapy of directly related response.59 Validation within a larger cohort demonstrated
targets (i.e., downstream PD-1/L1 inhibition). Unrelated im- that increases in these signatures were predictive of im-
mune pathways involving targets such as tumor necrosis provement in PFS (HR, 0.36; p = .0004) and OS (HR, 0.39;
factor-α or matrix metalloproteinase-9 may instead require p = .0011). Similar studies assessing PD-1 and PD-L1
separate and individualized gene expression assays. marker proximity in Merkel cell carcinoma reported that
One way to improve the accuracy of immune biomarkers, and they correlate positively with pembrolizumab response.60
to potentially make them applicable to more than one type of Another recent study used quantitative IF to assay TIL
therapy, is to develop comprehensive multifactor prediction activation and proliferation among patients with NSCLC.61
tools. One method is to evaluate the transcriptome profiles Increased coexpression of CD3, Ki-67, and granzyme B

among cytotoxic T cells was prognostic of improved survival elevation of peripheral CD27+CD45RO CD8+ T cells was
in ICI-naive patients. Interestingly, a “dormant” multiplex IF highly sensitive and specific for CR to tisagenlecleucel.68
signature demonstrating high CD3, but low granzyme B and Similarly, a subpopulation of CAR T-cell product (CD27+PD-
Ki-67, was significantly associated with durable clinical 1 CD8+) was highly predictive of treatment response (AUC,
benefit (86% rate), improved PFS, and improved OS fol- 0.92). Also, in a small study evaluating anti-CD19 CAR T-cell
lowing checkpoint inhibition. therapy among patients with B-cell leukemia or lymphoma,
These and other data clearly demonstrate the advantages of elevated pretreatment levels of interleukin (IL)-12 and den-
a multiplexed system over more traditional single-analyte dritic cell lysosome-associated membrane glycoprotein in
IHC/IF. As mentioned previously, systematic review and peripheral blood served as good predictors of positive re-
meta-analysis showed a high correlation between mIHC/IF sponse. Additionally, TNA-related apoptosis-inducing ligand
and patient response to anti–PD-1/L1 therapy (AUC, 0.79), and Fas ligand directly correlated with patient survival,
while outperforming biomarkers lsuch as PD-L1, TMB, and whereas IL-6, IL-8, PD-L1, PD-L2, and NAP3 were inversely
GEP (AUC, 0.65, 0.69, and 0.65, respectively).16 related.110 Another valuable metric for CAR T-cell therapy is
TMB and neoantigen formation. For instance, mutational
CHIMERIC ANTIGEN RECEPTOR T-CELL AND T-CELL load in related adoptive cell therapies holds promising pre-
ENGAGER BIOMARKERS dictive utility. In an early clinical phase trial among patients
A subset of adoptive cell therapy, CAR T-cell regimens with advanced and refractory melanoma, higher levels of
function by collecting autologous or allogeneic peripheral somatic mutations correlated with improved response, PFS,
T cells and selectively expressing synthetic receptors and OS to TIL transfer and IL-2.111 And several studies across
against tumoral antigens. Patients subsequently receive various cancer subtypes demonstrate improved efficacy of
infusions of these modified lymphocytes, which directly adoptive cell therapy when mutation-specific T-cell pop-
target cancerous cells and, ideally, maintain an active CAR ulations are expanded.112 Though these data do not directly
T-cell population for sustained surveillance. A major advent utilize therapy with CARs, clear practical extensions exist. A
in the treatment of relapsed and refractory hematologic broader mutational landscape and more diverse neoantigens
malignancies was the FDA approval of the anti-CD19 CAR increase the repertoire of putative CAR targets. This improves
T-cell products tisagenlecleucel and axicabtagene cil- the likelihood of developing truly tumor-specific CAR antigen
oleucel for the treatment of pediatric/adolescent and young combinations and overcoming a limitation behind the ma-
adult B-cell precursor ALL and adult large B-cell lymphoma, jority of current regimens: host tissue target expression and
respectively.65,108 The indications for tisagenlecleucel were unwanted toxicity.
quickly expanded to include cohorts with adult large B-cell Therapeutic targeting of effector T cells and innate immune
lymphoma.109 Multiple preclinical models and active clinical cells to tumor antigens can also be enhanced with the use of
trials are evaluating the efficacy of alternative synthetic bispecific monoclonal antibodies. Modified proteins with
receptors in diverse malignancies, including chronic lym- distinct antigen-binding fragments or fusion products be-
phocytic leukemia, multiple myeloma, mesothelioma, NSCLC, tween single-chain variable fragments serve to bridge im-
breast cancer, and glioblastoma, among others. mune cell receptors and cytotoxic activity with malignant
Although not as extensive or validated as biomarkers in cells.113 Despite rapid growth in the discovery and devel-
checkpoint inhibition, several quantitative patient measure- opment of these immunomodulatory formats, only one is
ments can predict response in CAR T-cell therapy. A major globally approved as a cancer therapeutic.114 Initial devel-
determinant of treatment success appears to be expansion opment of the trispecific antibody catumaxomab allowed for
and persistence of CAR T-cell populations following infusion. bridging among carcinoma-upregulated epithelial cell ad-
For example, among a small cohort of patients with advanced hesion molecules, CD3 receptors on T cells, and Fc-γ re-
B-cell ALL, anti-CD19 CAR T-cell therapy was associated with ceptors on accessory immune cells. This was approved in
an impressive CR (86%, negative minimal residual disease) by 2009 by the European Medicines Agency for the man-
day 28 postinfusion. However, loss of circulating CAR T cells agement of malignant ascites, although it has since been
by host CD8+ activity against the transgene product promoted removed from the market because of manufacturer in-
relapse among a subset of these responders.66 Correspond- solvency.115 Subsequently, the bispecific T-cell engager
ingly, in the ZUMA-1 trial, axicabtagene ciloleucel expansion antibody blinatumomab, targeting CD3 and CD19, was
was significantly associated with improved objective response approved by the FDA following the BLAST trial for use in
(peak factor 4.0); persistent serum levels were also demon- relapsed/refractory B-cell precursor ALL.116
strated in three patients with sustained CR at 24 months.65 Post hoc analysis of catumaxomab phase II/III testing
Immunophenotypes also serve as biomarkers for CAR T-cell demonstrated that increased pretherapy levels of peripheral
efficacy. For example, among high-risk and heavily treated blood relative lymphocytes (. 13%) improved mean OS
patients with chronic lymphocytic leukemia, pretherapy flow among patients treated with intraperitoneal catumaxomab

McKean et al
(131 days; p = .0072).69 Tregs are another biomarker that lower baseline levels of IL-6, the risk of ipilimumab toxicity
has been shown to correlate with bispecific-antibody re- was increased (odds ratio, 2.84; p = .007).73 Similarly,
sponse. Among a cohort of patients with B-cell precursor decreased pretreatment levels of circulating IL-6, IL-8,
ALL receiving blinatumomab, Treg levels in the peripheral soluble IL-2 receptor (sCD25), and IL-17 among cohorts
blood were increased among nonresponders (10.25%) with locally advanced or metastatic melanoma are associ-
compared with responders (4.82%).67 ated with ipilimumab-induced colitis.74,75 Interestingly, it
may be the net increase in these proinflammatory cytokines
BIOMARKERS AND ADVERSE EVENTS
from ICI therapy that determines immune toxicity. In two
The advent of modern cancer immunotherapeutics has been studies of melanoma cohorts treated with nivolumab, on-
accompanied by challenging side effect profiles requiring therapy increases in IL-676 or in soluble cluster of differ-
prompt and unique management. The clinical and financial entiation 163 (sCD163) and CXCL577 were predictive of
burden from these toxicities is significant. In ipilimumab psoriasiform dermatitis or diverse irAEs, respectively.
monotherapy alone, irAEs are seen in more than 70% of
Furthermore, the presence of autoantibodies may predict
patients.117,118 Combination checkpoint inhibition increases
development of endocrine-specific toxicity following check-
the incidence and onset of these irAEs; pooled analysis of
point inhibition. In a multivariate analysis of patients with
ipilimumab and nivolumab treatment among cohorts with
advanced solid tumors treated with nivolumab, pretreatment
melanoma suggests that nearly all patients (94%) experience
elevation of serum antithyroglobulin antibodies was signifi-
some degree of toxicity, with up to 55% of them being high
cantly associated with subsequent autoimmune thyroid
grade (Common Terminology Criteria for Adverse Events
dysfunction (odds ratio, 26.5; 95% CI, 8.18–85.8).78 Among
grades 3-4).119 Among patients with B-cell precursor ALL or
a separate cohort diagnosed with a variety of solid tumors,
diffuse large B-cell lymphoma treated with CAR T-cell therapy,
the presence of one or more diabetes autoantibodies (against
the development of CRS may increase hospitalization costs by
glutamic acid decarboxylase 65, islet antigen 2, insulin, islet
more than $50,000.120 Unfortunately, our understanding of
cells, or zinc transporter 8) prior to ICI therapy hastened the
biomarkers that are predictive for immunotherapy toxicity is
development of clinical diabetes.79
underdeveloped and warrants further investigation.121
PITFALLS OF CURRENT BIOMARKER USE
Cytokine Release Syndrome and Immune Effector
Cell–Associated Neurotoxicity Syndrome Biomarkers Although there is clear prognostic and predictive benefit
behind biomarkers of immunotherapy response and tox-
Predictive algorithms for the development of CAR T-cell
icity, significant limitations prohibit more widespread benefit
toxicity have been proposed in several studies.122 Among
among patients (Fig. 1). Dramatic variation within individual
cohorts of pediatric and adult patients with B-cell precursor
tumor parenchyma, chronologic change, microenviron-
ALL, forward-selected logistic regression or decision tree
ment, and sample preparation are each a potential source
models using combinations of cytokines accurately predicted
behind failure of treatment selection and response.
the development of severe (grades 4-5) CRS. 70 When
combining IFN-γ, IL-13, and macrophage inflammatory Intratumoral Heterogeneity
protein 1α in pediatric patients, sensitivity and specificity
Depending on the degree of diversity, subclonal populations
reached 100% and 96%, respectively. In a phase I/II clinical
within a tumor may not be adequately sampled with biopsy
trial of adult patients with advanced non-Hodgkin lymphoma,
alone (Fig. 1A). As such, preliminary analysis and biomarker
chronic lymphocytic leukemia, or B-cell precursor ALL,
testing from limited tissue may not be representative of
preinfusion increases in proteins associated with capillary
actual response to immunotherapy. Recent data published
leak (angiopoietin ratios and von Willebrand factor) correlated
on patients with NSCLC or melanoma suggest that subclonal
with increased severity of subsequent CRS.71 In this same
neoantigen heterogeneity is enriched in patients with poor
trial, classification-tree modeling demonstrated that elevated
response to ICI.123 This neoantigen diversity is more rep-
serum monocyte chemoattractant protein-1 levels ( 1343.5
resentative of patient outcomes than total neoantigen
pg/mL) and development of a fever 38.9°C or greater within
burden alone, and it can be combined with TMB to better
36 hours of CAR T-cell infusion were highly sensitive (100%)
stratify predictive data of median OS.
and specific (95%) for subsequent CRS (grade 4-5). Com-
bining elevated IL-6 concentrations ( 16 pg/mL) with this Mutational variety within a tumor may also limit the predictive
algorithm was also predictive of severe (grade 4-5) immune yield of certain biomarkers (Fig. 1D). For instance, although
effector cell–associated neurotoxicity syndrome.72 TMB shows broad correlation with patient survivorship and
therapy response in ICI, certain tumors with low TMB also
Immune-Related Adverse Event Biomarkers show increased responsiveness to checkpoint inhibition. This
Cytokines appear to be important markers of irAE risk as may be due to alternative mutations such as indels (seen at
well. For instance, among patients with melanoma and high frequency in renal cell carcinoma).15 Indeed, positive

FIGURE 1. Sources of Biomarker

Error
(A) Intratumoral (and intraper-
sonal) cellular heterogeneity re-
mains a significant limitation in
biomarker validity. Further dy-
namic alterations in clonal com-
position under the pressure of
time (1) and therapy (2) prohibit
pretreatment biomarker accu-
racy. (B) Patient host immunity
and surrounding tumor micro-
environment remain highly in-
dividualized and responsive to
progressive cytokine (1) and/or
treatment (2) exposures. These
multiple variables abrogate the
accuracy of single biomarker
tests. (C) HLA allelic polymorphism
and TCR selection generate sig-
nificant variety among patient
antigen processing and pre-
sentation. (D) Post-translational
protein modifications or novel
mutations such as indels and
copy number alterations gen-
erate neoantigens that may re-
main undetected by traditional
biomarkers.
Abbreviations: TMB, tumor mu-
tation burden; TAM, tumor-
associated macrophage; TIL,
tumor-infiltrating lymphocyte;
Treg, regulatory T cell; MDSC, myeloid-derived suppressor cell; O-GlcNAc, O-linked β-N-acetylglucosamine; TCR, T-cell receptor; MHC,
major histocompatibility complex.
Created with Biorender.
correlations with indel burden in cohorts with melanoma are predictors of patient response to immunotherapy (Fig. 1B).
associated with improved response to ICI therapy. Somatic For example, in several studies incorporating multiple
copy number alterations are another mutation type not cancer types, PD-L1 expression on tumor-infiltrating im-
currently assessed by TMB assays that have been positively mune cells (macrophages, dendritic cells, or T lympho-
associated with immune infiltration and survival advantage cytes) was correlative with ICI patient response, whereas
among cohorts with melanoma.36 Also, there is evidence that PD-L1 tumoral expression alone was not.25,48,49,126 Addi-
some neoantigens are generated by post-translational tionally, there is a growing body of evidence that tumor-
modifications, and these would not be detected by any associated macrophages may abrogate anti–PD-1 response
nucleic acid assay.124,125 in patient cohorts with advanced melanoma.12,127,128 Al-
ternatively, another unique cell population, myeloid-derived
Tumor Microenvironment suppressor cells, is inversely correlated with anti–CTLA-4
The complex interplay between tumor cells and their mi- response in patients with melanoma.129-132 Similarly, failure
croenvironment plays a significant role in patient response of adoptive T-cell therapy in solid tumors can be associated
to immunomodulatory agents. The proliferation or retardation directly with the proportion of myeloid-derived suppressor
of malignant tissue relies on multiple surrounding host cells; natural killer targeting of these cells improves results in
factors such as cytotoxicity of infiltrating lymphocytes, im- neuroblastoma xenograft models.133 These and other find-
munosuppression from tumor-associated macrophages, ings make it increasingly clear that nontumoral cells within
and paracrine signaling from supportive fibroblasts. As the microenvironment can affect biomarker utility and may
such, biomarkers of tumor cells alone may be insufficient serve as an underutilized predictive resource themselves.

McKean et al
Interpatient Variation and Host Immunity chemotherapy and immune therapy. With the generation of
Similar to intratumoral subclonal diversity, significant variation resistant subclones, the predictive power of static biomarkers
exists within histologic cancer types among separate patients. becomes more limited. For example, in patients with advanced
For instance, among patients with metastatic melanoma and melanoma, pretreatment TMB is positively associated with
naive to checkpoint inhibition, PD-L1 was homogenously anti–PD-1 survivorship only within therapy-naive cohorts.12
positive (defined as . 1% tumor expression) in 26% of the However, quantitative change in TMB during anti–PD-1
cohort and negative in 22% of subjects.134 Importantly, 52% of therapy strongly correlates with response and OS, even
patients showed intrapersonal PD-L1 heterogeneity among within patients previously treated with ICI. Similarly, on-therapy
primary melanoma sites, locoregional disease, and distant changes in T-cell diversity correlate with anti–PD-1 treatment
metastases, highlighting yet another limitation of current testing. outcomes. Among independent cohorts of patients with
melanoma naive to checkpoint inhibition, a dynamic loss in
Another level of this interpatient heterogeneity is the striking pretherapy TCR variation (and subsequent oligoclonal ex-
complexity within host antigen processing and presentation pansion) correlates positively with nivolumab response.12,142
(Fig. 1C). Significant polymorphism among HLA alleles—as However, this relationship is reversed among patients with
well as the extensive process behind cognate T-cell receptor progressive melanoma on anti–CTLA-4 treatment. Here,
(TCR) expression, recombination, and selection—generates nivolumab response correlates with increasing richness of the
remarkable individuality between patient immune systems. complementarity-determining region 3 on TCRs.12 Such
This variability, in turn, can influence patient outcomes to dramatic chronologic shifts among tumor and immune cells
checkpoint inhibition. For example, recent analysis of multiple require more frequent biomarker surveillance. Clearly, bi-
cancers among two cohorts demonstrated that maximal HLA opsies cannot be limited to single or pretreatment sampling.
loci heterozygosity improved OS after ICI therapy.135 Similarly,
loss of HLA-I heterozygosity correlated with reduced survival Sample Collection and Evaluation
after ICI therapy (HR, 1.60; p = .05) and was most pro- A major restriction in current biomarker use is the significant
nounced among patients with low TMB (HR, 3.68; p = .0006). variability in preanalytical processing and subsequent clinical
Within this same study, distinct supertype alleles had dis- interpretation. Depending on the laboratory test used, several
parate associations with patient outcomes after ICI. HLA-B44 independent factors are prone to error.143 For example,
expression was seen in patients with extended survival (HR, 0. background signal and morphologic clarity of IHC/IF can be
61; p = .01), whereas the presence of HLA-B62 had the influenced by sample fixation time, quality of embedding, or
opposite correlation (HR, 2.29; p = .0007). Extensive work has time from extraction to preservation. Similarly, the quality of
demonstrated that aberrant tumoral regulation or mutation of biomarkers in peripheral blood can be altered by cryopres-
HLA-I genes contributes to immune evasion and poorer ervation temperature, repeated freeze-thaw cycles, antico-
clinical response.136,137 By extension, pretherapy abrogation agulant type, and processing delay. Although not always
of major histocompatibility complex class I components is controllable, the ramification of such variables in immuno-
predictive of ipilimumab resistance, whereas response to therapy is clear. In PD-L1 IHC alone, numerous diagnostic
nivolumab may require major histocompatibility complex assays are available but are not always interchangeable; thus,
class II expression and IFN-γ activation.138 they can yield disparate results, even among case-matched
In addition, systemic TCR sequence diversity may be related samples.144 Furthermore, analytical interpretation of pro-
to ICI response. For instance, in two separate studies eval- cessed samples can show significant variation. Although
uating small cohorts of patients with metastatic pancreatic good reproducibility is observed among tumor cells, pa-
adenocarcinoma139 or urothelial carcinoma,140 the presence thologists have demonstrated poor concordance in PD-L1
of pretherapy peripheral T-cell clonality predicted poorer OS scoring of immune cells in NSCLC samples.145 Tighter control
on ipilimumab and poorer PFS and OS on atezolizumab. The of preanalytical processing and greater concordance among
relationship between TIL TCR variability and ICI response is data reporting are needed for further biomarker development
less clear; some studies suggest pretherapy clonality as and validation in clinical trials.146
a marker for anti–PD-1 response,37,47 whereas another study IDEAL AND FUTURISTIC BIOMARKER USE
correlates post-therapy clones with positive NSCLC outcomes
on nivolumab.141 The variance in these findings highlights the Biomarkers will be the future foundation of treating all
complexity of interactions between host immunity and the cancers with immunotherapy. Given the complexity of the
tumoral milieu. Such complexity significantly limits the power immune system and its regulation, it is unlikely that any single
and interpretation of canonical biomarkers. component biomarker will be predictive enough for all pa-
tients or all types of immunotherapies. Therefore, ideal bio-
Chronologic Change markers will need to be multifaceted and comprehensive
Dynamic change within tumoral subpopulations can oc- and be able to guide clinicians about whether the patient
cur over time, especially under the selective pressure of will respond to immunotherapy, as well as which specific

immunotherapeutic agent(s) (e.g., checkpoint inhibitors, OPTIMAL CLINICAL TRIAL DESIGN FOR BIOMARKER UTILITY
immunostimulatory therapies, cellular therapies) are needed Prospective data are required for proper biomarker devel-
(Table 2). If no such biomarker is available, early on- opment. Therefore, all future clinical trials must be designed
treatment biomarkers that confirm that a patient is (or is to incorporate biomarkers to some degree. At a minimum,
not) responding to immunotherapy will be needed. This type pretreatment and on-treatment information (e.g., tissue or
of response detection will allow physicians to identify patients blood samples, raw data from images, or clinical de-
who are not benefiting from a specific immunotherapy early mographics) must be collected and stored for potential
enough in their course so that treatment can be switched/ exploratory biomarker analyses. Biopsy and blood collection
escalated prior to complications that are associated with upon progression should also be considered and can be
tumor progression. Ideally, such biomarkers would be non- critical to elucidate resistance mechanisms and guide the
invasive; therefore, the development of noninvasive assess- next line of therapy. This is especially important for early-
ments of changes in the tumor microenvironment will be phase trials of novel immunotherapies, because the bio-
needed.147 Examples of these include PET imaging of TILs or marker(s) for response and/or resistance will likely be
immune checkpoints, as well as systemic cytokine changes. therapy specific. Being able to identify potential biomarkers
TABLE 2. Design and Advantages of Ideal Biomarkers in Immunotherapy

Noninvasive Better sample accessibility using peripheral blood or imaging allows for closer monitoring of therapy
response or the development of treatment resistance
Time Sensitive Reliable and accessible preliminary testing, combined with regular and consistent on-therapy
sampling, could provide actionable results of chronologic tumoral changes
Multifactorial and Because of the complexity of interactions among tumor parenchyma, microenvironment, and host
Comprehensive immune response, it is unlikely that single analytes or tumor-specific values alone will be sufficient for
predictive capacity in immunotherapy
Therapy Specific Within precision medicine, comprehensive biomarkers could predict ideal initial immunotherapy class
and subtype (i.e., anti–PD-1/L1 vs. anti–CTLA-4) and direct subsequent changes should resistance
develop
Resistance and Toxicity Adaptive clinical trials and individualized patient therapy will require predictors of adverse events and
treatment refractoriness to appropriately prioritize therapeutic options

McKean et al
for novel therapies early in development will allow the ex- improve the proportion of patients who are likely to respond
pedited validation and clinical use of these biomarkers in to therapy, there is a significant need for biomarkers with
late-phase trials. Although understanding who will respond stronger predictive abilities. Given the complexity of the
to a specific treatment is of high interest, equal interest and immune system, single-marker biomarkers will likely be in-
effort should be placed into evaluating biomarkers of sufficient to adequately predict who will and will not respond
nonresponse or resistance. Indeed, understanding these to cancer immunotherapy. One potential way to mitigate this
negative metrics may identify the best therapies to use in is to develop biomarkers that allow for the early assessment of
specific patient populations. Once these biomarkers of immune activation within the tumor microenvironment (e.g.,
response and resistance are further developed, adaptive noninvasive assessment of TILs, alternative checkpoints)
clinical trials will be needed. Here, on-treatment biomarkers rather than pretreatment predictive biomarkers. These types
of early response or resistance can identify patients who do of early response assessment biomarkers would allow pa-
not benefit from a particular therapy but instead require tients to initiate single-agent treatments with low side effect
treatment changes or dose escalation. Another large hurdle profiles and subsequently identify patients who need esca-
in immunotherapy development is the heterogeneity of lations/changes in therapy prior to clinical progression and
resistance mechanisms between patients. Many trials have deterioration. Given the complexity of host immunity, another
negative results, not because the agents are not active, but
potential mechanism for developing immune biomarkers is to
because the signal is diluted. Using biomarkers to separate
use multimarker assessments. However, given the multitude
patients into different groups of similar resistance mecha-
of novel immunotherapies currently under investigation (e.g.,
nisms can be a solution to focus drug-development efforts.
targeting alternative checkpoints, the adenosine pathway,
CONCLUSIONS immunostimulatory molecules, cytokines), it is likely that no
Cancer immunotherapy, primarily ICI, can provide durable, single biomarker, or even biomarker signatures, will be
long-lasting responses for some patients with cancer; predictive for each type of cancer immunotherapy. Therefore,
however, these therapies are currently effective in only comprehensive immune profiling of individual tumors will
a small minority. Although currently available biomarkers likely be needed to develop predictive biomarkers that can
(primarily PD-L1 and IFN-γ gene signature panels) can accurately direct the selection of patient immunotherapy.

1
Huntsman Cancer Institute, University of Utah, Salt Lake City, UT Siwen Hu-Lieskovan, MD, PhD, Huntsman Cancer Institute, University of
2
HonorHealth Research Institute, Scottsdale, AZ Utah, 2000 Circle of Hope Drive, HCI-RS-2703, Salt Lake City, UT
3
Department of Pathology, Yale University School of Medicine, New 84112; email: [email protected].
Haven, CT
EDBK_280571.
REFERENCES
1. Schadendorf D, Hodi FS, Robert C, et al. Pooled analysis of long-term survival data from phase II and phase III trials of ipilimumab in unresectable or metastatic
melanoma. J Clin Oncol. 2015;33:1889-1894.
2. Haslam A, Prasad V. Estimation of the percentage of US patients with cancer who are eligible for and respond to checkpoint inhibitor immunotherapy drugs.
JAMA Netw Open. 2019;2:e192535.
3. Cheng J, Zhao L, Zhang Y, et al. Understanding the mechanisms of resistance to CAR T-cell therapy in malignancies. Front Oncol. 2019;9:1237.
4. Mueller KT, Maude SL, Porter DL, et al. Cellular kinetics of CTL019 in relapsed/refractory B-cell acute lymphoblastic leukemia and chronic lymphocytic
leukemia. Blood. 2017;130:2317-2325.
5. Mueller KT, Waldron E, Grupp SA, et al. Clinical pharmacology of tisagenlecleucel in B-cell acute lymphoblastic leukemia. Clin Cancer Res. 2018;
24:6175-6184.
6. Xin Yu J, Hubbard-Lucey VM, Tang J. The global pipeline of cell therapies for cancer. Nat Rev Drug Discov. 2019;18:821-822.
7. Carbone DP, Reck M, Paz-Ares L, et al; CheckMate 026 Investigators. First-line nivolumab in stage IV or recurrent non-small-cell lung cancer. N Engl J Med.
2017;376:2415-2426.

378:2093-2104.
9. Hellmann MD, Nathanson T, Rizvi H, et al. Genomic features of response to combination immunotherapy in patients with advanced non-small-cell lung cancer.
Cancer Cell. 2018;33:843-852.e4.
10. Peters S, Cho BC, Reinmuth N, et al. Tumor mutational burden (TMB) as a biomarker of survival in metastatic non-small cell lung cancer (mNSCLC): blood and
tissue TMB analysis from MYSTIC, a Phase III study of first-line durvalumab 6 tremelimumab vs. chemotherapy. Presented at: American Association for Cancer
Research Annual Meeting 2019. Atlanta, GA; 2019.
11. McDermott DF, Huseni MA, Atkins MB, et al. Clinical activity and molecular correlates of response to atezolizumab alone or in combination with bevacizumab
versus sunitinib in renal cell carcinoma. Nat Med. 2018;24:749-757.
12. Riaz N, Havel JJ, Makarov V, et al. Tumor and microenvironment evolution during immunotherapy with nivolumab. Cell. 2017;171:934-949.e16.
13. Hanna GJ, Lizotte P, Cavanaugh M, et al. Frameshift events predict anti-PD-1/L1 response in head and neck cancer. JCI Insight. 2018;3:e98811.
14. Miao D, Margolis CA, Gao W, et al. Genomic correlates of response to immune checkpoint therapies in clear cell renal cell carcinoma. Science. 2018;
359:801-806.
15. Turajlic S, Litchfield K, Xu H, et al. Insertion-and-deletion-derived tumour-specific neoantigens and the immunogenic phenotype: a pan-cancer analysis. Lancet
Oncol. 2017;18:1009-1021.
16. Lu S, Stein JE, Rimm DL, et al. Comparison of biomarker modalities for predicting response to PD-1/PD-L1 checkpoint blockade: a systematic review and meta-
analysis. JAMA Oncol. 2019;5:1195-1204.
17. Havel JJ, Chowell D, Chan TA. The evolving landscape of biomarkers for checkpoint inhibitor immunotherapy. Nat Rev Cancer. 2019;19:133-150.
16:275-287.
19. Mok TSK, Wu YL, Kudaba I, et al; KEYNOTE-042 Investigators. Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally
advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomised, open-label, controlled, phase 3 trial. Lancet. 2019;393:1819-1830.
20. Burtness B, Harrington KJ, Greil R, et al; KEYNOTE-048 Investigators. Pembrolizumab alone or with chemotherapy versus cetuximab with chemotherapy for
recurrent or metastatic squamous cell carcinoma of the head and neck (KEYNOTE-048): a randomised, open-label, phase 3 study. Lancet. 2019;
394:1915-1928.
21. Shah MA, Kojima T, Hochhauser D, et al. Efficacy and safety of pembrolizumab for heavily pretreated patients with advanced, metastatic adenocarcinoma or
squamous cell carcinoma of the esophagus: the phase 2 KEYNOTE-180 Study. JAMA Oncol. 2019;5:546-550.
22. Metges J, et al. The phase 3 KEYNOTE-181 study: pembrolizumab versus chemotherapy as second-line therapy for advanced esophageal cancer. Ann Oncol.
2019;30(suppl 4):iv130.
23. Fuchs CS, Doi T, Jang RW, et al. Safety and efficacy of pembrolizumab monotherapy in patients with previously treated advanced gastric and gastroesophageal
junction cancer: phase 2 clinical KEYNOTE-059 Trial. JAMA Oncol. 2018;4:e180013.
24. Chung HC, Ros W, Delord JP, et al. Efficacy and safety of pembrolizumab in previously treated advanced cervical cancer: results from the phase II KEYNOTE-
158 Study. J Clin Oncol. 2019;37:1470-1478.
26. Brahmer JR, Drake CG, Wollner I, et al. Phase I study of single-agent anti-programmed death-1 (MDX-1106) in refractory solid tumors: safety, clinical activity,
27. Topalian SL, Hodi FS, Brahmer JR, et al. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N Engl J Med. 2012;366:2443-2454.
28. Daud AI, Wolchok JD, Robert C, et al. Programmed death-ligand 1 expression and response to the anti-programmed death 1 antibody pembrolizumab in
melanoma. J Clin Oncol. 2016;34:4102-4109.
29. Van Allen EM, Miao D, Schilling B, et al. Genomic correlates of response to CTLA-4 blockade in metastatic melanoma. Science. 2015;350:207-211.
30. Rizvi NA, Hellmann MD, Snyder A, et al. Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancer.
Science. 2015;348:124-128.
31. Hellmann MD, Callahan MK, Awad MM, et al. Tumor mutational burden and efficacy of nivolumab monotherapy and in combination with ipilimumab in small-
cell lung cancer. Cancer Cell. 2018;33:853-861.e4.
34. Samstein RM, Lee CH, Shoushtari AN, et al. Tumor mutational load predicts survival after immunotherapy across multiple cancer types. Nat Genet. 2019;
51:202-206.
36. Davoli T, Uno H, Wooten EC, et al. Tumor aneuploidy correlates with markers of immune evasion and with reduced response to immunotherapy. Science. 2017;
355:eaaf8399.

McKean et al
37. Roh W, Chen PL, Reuben A, et al. Integrated molecular analysis of tumor biopsies on sequential CTLA-4 and PD-1 blockade reveals markers of response and
resistance. Sci Transl Med. 2017;9:eaah3560.
38. Mlecnik B, Bindea G, Angell HK, et al. Integrative analyses of colorectal cancer show immunoscore is a stronger predictor of patient survival than microsatellite
instability. Immunity. 2016;44:698-711.
39. Pagès F, Mlecnik B, Marliot F, et al. International validation of the consensus Immunoscore for the classification of colon cancer: a prognostic and accuracy
study. Lancet. 2018;391:2128-2139.
40. Paulsen EE, Kilvaer TK, Khanehkenari MR, et al. Assessing PDL-1 and PD-1 in non-small cell lung cancer: a novel immunoscore approach. Clin Lung Cancer.
2017;18:220-233.e8.
41. Daud AI, Loo K, Pauli ML, et al. Tumor immune profiling predicts response to anti-PD-1 therapy in human melanoma. J Clin Invest. 2016;126:3447-3452.
42. Ganesan AP, Clarke J, Wood O, et al. Tissue-resident memory features are linked to the magnitude of cytotoxic T cell responses in human lung cancer. Nat
Immunol. 2017;18:940-950.
43. Savas P, Virassamy B, Ye C, et al; Kathleen Cuningham Foundation Consortium for Research into Familial Breast Cancer (kConFab). Single-cell profiling of
breast cancer T cells reveals a tissue-resident memory subset associated with improved prognosis. Nat Med. 2018;24:986-993.
44. Thommen DS, Koelzer VH, Herzig P, et al. A transcriptionally and functionally distinct PD-1+ CD8+ T cell pool with predictive potential in non-small-cell lung
cancer treated with PD-1 blockade. Nat Med. 2018;24:994-1004.
45. Zappasodi R, Budhu S, Hellmann MD, et al. Non-conventional inhibitory CD4+Foxp3 PD-1hi T cells as a biomarker of immune checkpoint blockade activity.
Cancer Cell. 2018;33:1017-1032.e7.
47. Tumeh PC, Harview CL, Yearley JH, et al. PD-1 blockade induces responses by inhibiting adaptive immune resistance. Nature. 2014;515:568-571.
48. Mariathasan S, Turley SJ, Nickles D, et al. TGFβ attenuates tumour response to PD-L1 blockade by contributing to exclusion of T cells. Nature. 2018;
554:544-548.
49. Herbst RS, Soria JC, Kowanetz M, et al. Predictive correlates of response to the anti-PD-L1 antibody MPDL3280A in cancer patients. Nature. 2014;
515:563-567.
50. Wong PF, Wei W, Smithy JW, et al. Multiplex quantitative analysis of tumor-infiltrating lymphocytes and immunotherapy outcome in metastatic melanoma. Clin
Cancer Res. 2019;25:2442-2449.
51. Fehrenbacher L, Spira A, Ballinger M, et al; POPLAR Study Group. Atezolizumab versus docetaxel for patients with previously treated non-small-cell lung cancer
(POPLAR): a multicentre, open-label, phase 2 randomised controlled trial. Lancet. 2016;387:1837-1846.
52. Ayers M, Lunceford J, Nebozhyn M, et al. IFN-γ-related mRNA profile predicts clinical response to PD-1 blockade. J Clin Invest. 2017;127:2930-2940.
53. Seiwert TY, Burtness B, Mehra R, et al. Safety and clinical activity of pembrolizumab for treatment of recurrent or metastatic squamous cell carcinoma of the
head and neck (KEYNOTE-012): an open-label, multicentre, phase 1b trial. Lancet Oncol. 2016;17:956-965.
54. Higgs BW, Morehouse CA, Streicher K, et al. Interferon gamma messenger RNA signature in tumor biopsies predicts outcomes in patients with non-small cell
lung carcinoma or urothelial cancer treated with durvalumab. Clin Cancer Res. 2018;24:3857-3866.
55. Gupta S, McCann L, Chan YGY, et al. Closed system RT-qPCR as a potential companion diagnostic test for immunotherapy outcome in metastatic melanoma.
J Immunother Cancer. 2019;7:254.
56. Auslander N, Zhang G, Lee JS, et al. Robust prediction of response to immune checkpoint blockade therapy in metastatic melanoma. Nat Med. 2018;
24:1545-1549.
57. Jiang P, Gu S, Pan D, et al. Signatures of T cell dysfunction and exclusion predict cancer immunotherapy response. Nat Med. 2018;24:1550-1558.
58. Charoentong P, Finotello F, Angelova M, et al. Pan-cancer immunogenomic analyses reveal genotype-immunophenotype relationships and predictors of
response to checkpoint blockade. Cell Reports. 2017;18:248-262.
59. Johnson DB, Bordeaux J, Kim JY, et al. Quantitative spatial profiling of PD-1/PD-L1 interaction and HLA-DR/IDO-1 predicts improved outcomes of anti-PD-1
therapies in metastatic melanoma. Clin Cancer Res. 2018;24:5250-5260.
60. Giraldo NA, Nguyen P, Engle EL, et al. Multidimensional, quantitative assessment of PD-1/PD-L1 expression in patients with Merkel cell carcinoma and
association with response to pembrolizumab. J Immunother Cancer. 2018;6:99.
61. Gettinger SN, Choi J, Mani N, et al. A dormant TIL phenotype defines non-small cell lung carcinomas sensitive to immune checkpoint blockers. Nat Commun.
2018;9:3196.
62. van de Donk PP, Kist de Ruijter L, Lub-de Hooge MN, et al. Molecular imaging biomarkers for immune checkpoint inhibitor therapy. Theranostics. 2020;
10:1708-1718.
63. Kaira K, Higuchi T, Naruse I, et al. Metabolic activity by 18F-FDG-PET/CT is predictive of early response after nivolumab in previously treated NSCLC. Eur J Nucl
Med Mol Imaging. 2018;45:56-66.
64. Bensch F, van der Veen EL, Lub-de Hooge MN, et al. 89Zr-atezolizumab imaging as a non-invasive approach to assess clinical response to PD-L1 blockade in
cancer. Nat Med. 2018;24:1852-1858.
65. Locke FL, Ghobadi A, Jacobson CA, et al. Long-term safety and activity of axicabtagene ciloleucel in refractory large B-cell lymphoma (ZUMA-1): a single-arm,
multicentre, phase 1-2 trial. Lancet Oncol. 2019;20:31-42.

66. Turtle CJ, Hanafi LA, Berger C, et al. CD19 CAR-T cells of defined CD4+:CD8+ composition in adult B cell ALL patients. J Clin Invest. 2016;126:2123-2138.
67. Duell J, Dittrich M, Bedke T, et al. Frequency of regulatory T cells determines the outcome of the T-cell-engaging antibody blinatumomab in patients with
B-precursor ALL. Leukemia. 2017;31:2181-2190.
68. Fraietta JA, Lacey SF, Orlando EJ, et al. Determinants of response and resistance to CD19 chimeric antigen receptor (CAR) T cell therapy of chronic lymphocytic
leukemia. Nat Med. 2018;24:563-571.
69. Heiss MM, Ströhlein MA, Bokemeyer C, et al. The role of relative lymphocyte count as a biomarker for the effect of catumaxomab on survival in malignant ascites
patients: results from a phase II/III study. Clin Cancer Res. 2014;20:3348-3357.
70. Teachey DT, Lacey SF, Shaw PA, et al. Identification of predictive biomarkers for cytokine release syndrome after chimeric antigen receptor T-cell therapy for
acute lymphoblastic leukemia. Cancer Discov. 2016;6:664-679.
71. Hay KA, Hanafi LA, Li D, et al. Kinetics and biomarkers of severe cytokine release syndrome after CD19 chimeric antigen receptor-modified T-cell therapy.
Blood. 2017;130:2295-2306.
72. Gust J, Hay KA, Hanafi LA, et al. Endothelial activation and blood-brain barrier disruption in neurotoxicity after adoptive immunotherapy with CD19 CAR-T cells.
Cancer Discov. 2017;7:1404-1419.
73. Valpione S, Pasquali S, Campana LG, et al. Sex and interleukin-6 are prognostic factors for autoimmune toxicity following treatment with anti-CTLA4 blockade.
J Transl Med. 2018;16:94.
74. Chaput N, Lepage P, Coutzac C, et al. Baseline gut microbiota predicts clinical response and colitis in metastatic melanoma patients treated with ipilimumab.
Ann Oncol. 2017;28:1368-1379.
75. Tarhini AA, Zahoor H, Lin Y, et al. Baseline circulating IL-17 predicts toxicity while TGF-β1 and IL-10 are prognostic of relapse in ipilimumab neoadjuvant therapy
of melanoma. J Immunother Cancer. 2015;3:39.
76. Tanaka R, Okiyama N, Okune M, et al. Serum level of interleukin-6 is increased in nivolumab-associated psoriasiform dermatitis and tumor necrosis factor-α is
a biomarker of nivolumab recativity. J Dermatol Sci. 2017;86:71-73.
77. Fujimura T, Sato Y, Tanita K, et al. Serum levels of soluble CD163 and CXCL5 may be predictive markers for immune-related adverse events in patients with
advanced melanoma treated with nivolumab: a pilot study. Oncotarget. 2018;9:15542-15551.
78. Kimbara S, Fujiwara Y, Iwama S, et al. Association of antithyroglobulin antibodies with the development of thyroid dysfunction induced by nivolumab. Cancer
Sci. 2018;109:3583-3590.
79. Stamatouli AM, Quandt Z, Perdigoto AL, et al. Collateral damage: insulin-dependent diabetes induced with checkpoint inhibitors. Diabetes. 2018;
67:1471-1480.
80. Ishida Y, Agata Y, Shibahara K, et al. Induced expression of PD-1, a novel member of the immunoglobulin gene superfamily, upon programmed cell death.
EMBO J. 1992;11:3887-3895.
81. Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer. 2012;12:252-264.
82. Leach DR, Krummel MF, Allison JP. Enhancement of antitumor immunity by CTLA-4 blockade. Science. 1996;271:1734-1736.
83. Borghaei H, Paz-Ares L, Horn L, et al. Nivolumab versus docetaxel in advanced nonsquamous non-small-cell lung cancer. N Engl J Med. 2015;373:1627-1639.
84. Motzer RJ, Escudier B, McDermott DF, et al; CheckMate 025 Investigators. Nivolumab versus everolimus in advanced renal-cell carcinoma. N Engl J Med.
2015;373:1803-1813.
85. Davis AA, Patel VG. The role of PD-L1 expression as a predictive biomarker: an analysis of all US Food and Drug Administration (FDA) approvals of immune
86. Eroglu Z, Zaretsky JM, Hu-Lieskovan S, et al. High response rate to PD-1 blockade in desmoplastic melanomas. Nature. 2018;553:347-350.
88. Chen DS, Mellman I. Elements of cancer immunity and the cancer-immune set point. Nature. 2017;541:321-330.
89. Galon J, Costes A, Sanchez-Cabo F, et al. Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science. 2006;
313:1960-1964.
90. Fridman WH, Pagès F, Sautès-Fridman C, et al. The immune contexture in human tumours: impact on clinical outcome. Nat Rev Cancer. 2012;12:298-306.
91. Melanoma and immunotherapy bridge 2015: Naples, Italy. 1-5 December 2015. J Transl Med. 2016;14:65.
92. Galon J, Fox BA, Bifulco CB, et al. Immunoscore and immunoprofiling in cancer: an update from the melanoma and immunotherapy bridge 2015. J Transl Med.
2016;14:273.
93. Salmon H, Franciszkiewicz K, Damotte D, et al. Matrix architecture defines the preferential localization and migration of T cells into the stroma of human lung
tumors. J Clin Invest. 2012;122:899-910.
94. Spranger S, Bao R, Gajewski TF. Melanoma-intrinsic β-catenin signalling prevents anti-tumour immunity. Nature. 2015;523:231-235.
95. Tauriello DVF, Palomo-Ponce S, Stork D, et al. TGFβ drives immune evasion in genetically reconstituted colon cancer metastasis. Nature. 2018;554:538-543.
96. Blackburn SD, Shin H, Freeman GJ, et al. Selective expansion of a subset of exhausted CD8 T cells by alphaPD-L1 blockade. Proc Natl Acad Sci USA. 2008;
105:15016-15021.
97. Paley MA, Kroy DC, Odorizzi PM, et al. Progenitor and terminal subsets of CD8+ T cells cooperate to contain chronic viral infection. Science. 2012;
338:1220-1225.

McKean et al
98. Schietinger A, Philip M, Krisnawan VE, et al. Tumor-specific T cell dysfunction is a dynamic antigen-driven differentiation program initiated early during
tumorigenesis. Immunity. 2016;45:389-401.
99. Sen DR, Kaminski J, Barnitz RA, et al. The epigenetic landscape of T cell exhaustion. Science. 2016;354:1165-1169.
100. Pauken KE, Sammons MA, Odorizzi PM, et al. Epigenetic stability of exhausted T cells limits durability of reinvigoration by PD-1 blockade. Science. 2016;
354:1160-1165.
101. Utzschneider DT, Charmoy M, Chennupati V, et al. T cell factor 1-expressing memory-like CD8(+) T cells sustain the immune response to chronic viral
infections. Immunity. 2016;45:415-427.
102. Im SJ, Hashimoto M, Gerner MY, et al. Defining CD8+ T cells that provide the proliferative burst after PD-1 therapy. Nature. 2016;537:417-421.
103. Siddiqui I, Schaeuble K, Chennupati V, et al. Intratumoral Tcf1+PD-1+CD8+ T cells with stem-like properties promote tumor control in response to vaccination
and checkpoint blockade immunotherapy. Immunity. 2019;50:195-211.e10.
104. Mami-Chouaib F, Blanc C, Corgnac S, et al. Resident memory T cells, critical components in tumor immunology. J Immunother Cancer. 2018;6:87.
105. Amsen D, van Gisbergen KPJM, Hombrink P, et al. Tissue-resident memory T cells at the center of immunity to solid tumors. Nat Immunol. 2018;19:538-546.
106. Tavaré R, Escuin-Ordinas H, Mok S, et al. An effective immuno-PET imaging method to monitor CD8-dependent responses to immunotherapy. Cancer Res.
2016;76:73-82.
107. Papadopoulos KP, Autio KA, Golan T, et al. Phase 1 study of MK-4166, an anti-human glucocorticoid-induced tumor necrosis factor receptor (GITR) antibody,
as monotherapy or with pembrolizumab in patients with advanced solid tumors. J Clin Oncol. 2019;37:15s (suppl: abstr 9509).
109. Schuster SJ, Bishop MR, Tam CS, et al; JULIET Investigators. Tisagenlecleucel in adult relapsed or refractory diffuse large B-cell lymphoma. N Engl J Med.
2019;380:45-56.
110. Enblad G, Karlsson H, Gammelgård G, et al. A phase I/IIa trial using CD19-targeted third-generation CAR T cells for lymphoma and leukemia. Clin Cancer Res.
2018;24:6185-6194.
111. Lauss M, Donia M, Harbst K, et al. Mutational and putative neoantigen load predict clinical benefit of adoptive T cell therapy in melanoma. Nat Commun. 2017;
8:1738.
112. Ott PA, Dotti G, Yee C, et al. An update on adoptive T-cell therapy and neoantigen vaccines. Am Soc Clin Oncol Educ Book. 2019;39:e70-e78.
113. Duell J, Lammers PE, Djuretic I, et al. Bispecific antibodies in the treatment of hematologic malignancies. Clin Pharmacol Ther. 2019;106:781-791.
114. Velasquez MP, Bonifant CL, Gottschalk S. Redirecting T cells to hematological malignancies with bispecific antibodies. Blood. 2018;131:30-38.
115. Heiss MM, Murawa P, Koralewski P, et al. The trifunctional antibody catumaxomab for the treatment of malignant ascites due to epithelial cancer: results of
a prospective randomized phase II/III trial. Int J Cancer. 2010;127:2209-2221.
116. Gökbuget N, Dombret H, Bonifacio M, et al. Blinatumomab for minimal residual disease in adults with B-cell precursor acute lymphoblastic leukemia. Blood.
2018;131:1522-1531.
117. Weber JS, Kähler KC, Hauschild A. Management of immune-related adverse events and kinetics of response with ipilimumab. J Clin Oncol. 2012;
30:2691-2697.
118. Bertrand A, Kostine M, Barnetche T, et al. Immune related adverse events associated with anti-CTLA-4 antibodies: systematic review and meta-analysis. BMC
Med. 2015;13:211.
119. Sznol M, Ferrucci PF, Hogg D, et al. Pooled analysis safety profile of nivolumab and ipilimumab combination therapy in patients with advanced melanoma. J Clin
Oncol. 2017;35:3815-3822.
120. Hernandez I, Prasad V, Gellad WF. Total costs of chimeric antigen receptor T-cell immunotherapy. JAMA Oncol. 2018;4:994-996.
121. Nakamura Y. Biomarkers for immune checkpoint inhibitor-mediated tumor response and adverse events. Front Med (Lausanne). 2019;6:119.
122. Wang Z, Han W. Biomarkers of cytokine release syndrome and neurotoxicity related to CAR-T cell therapy. Biomark Res. 2018;6:4.
123. McGranahan N, Furness AJ, Rosenthal R, et al. Clonal neoantigens elicit T cell immunoreactivity and sensitivity to immune checkpoint blockade. Science.
2016;351:1463-1469.
124. Dalet A, Robbins PF, Stroobant V, et al. An antigenic peptide produced by reverse splicing and double asparagine deamidation. Proc Natl Acad Sci USA. 2011;
108:E323-E331.
125. Cobbold M, De La Peña H, Norris A, et al. MHC class I-associated phosphopeptides are the targets of memory-like immunity in leukemia. Sci Transl Med. 2013;
5:203ra125.
126. Liu Y, Zugazagoitia J, Ahmed FS, et al. Immune cell PD-L1 colocalizes with macrophages and is associated with outcome in PD-1 pathway blockade therapy.
Clin Cancer Res. 2020;26:970-977.
127. Arlauckas SP, Garris CS, Kohler RH, et al. In vivo imaging reveals a tumor-associated macrophage-mediated resistance pathway in anti-PD-1 therapy. Sci Transl
Med. 2017;9:eaal3604.
128. Neubert NJ, Schmittnaegel M, Bordry N, et al. T cell-induced CSF1 promotes melanoma resistance to PD1 blockade. Sci Transl Med. 2018;10:eaan3311.
129. Meyer C, Cagnon L, Costa-Nunes CM, et al. Frequencies of circulating MDSC correlate with clinical outcome of melanoma patients treated with ipilimumab.
Cancer Immunol Immunother. 2014;63:247-257.

130. Gebhardt C, Sevko A, Jiang H, et al. Myeloid cells and related chronic inflammatory factors as novel predictive markers in melanoma treatment with ipilimumab.
Clin Cancer Res. 2015;21:5453-5459.
131. Sade-Feldman M, Kanterman J, Klieger Y, et al. Clinical significance of circulating CD33+CD11b+HLA-D myeloid cells in patients with stage IV melanoma
treated with ipilimumab. Clin Cancer Res. 2016;22:5661-5672.
132. Martens A, Wistuba-Hamprecht K, Geukes Foppen M, et al. Baseline peripheral blood biomarkers associated with clinical outcome of advanced melanoma
patients treated with ipilimumab. Clin Cancer Res. 2016;22:2908-2918.
133. Parihar R, Rivas C, Huynh M, et al. NK cells expressing a chimeric activating receptor eliminate MDSCs and rescue impaired CAR-T cell activity against solid
tumors. Cancer Immunol Res. 2019;7:363-375.
134. Madore J, Vilain RE, Menzies AM, et al. PD-L1 expression in melanoma shows marked heterogeneity within and between patients: implications for anti-PD-1/
PD-L1 clinical trials. Pigment Cell Melanoma Res. 2015;28:245-253.
135. Chowell D, Morris LGT, Grigg CM, et al. Patient HLA class I genotype influences cancer response to checkpoint blockade immunotherapy. Science. 2018;
359:582-587.
136. Garrido C, Paco L, Romero I, et al. MHC class I molecules act as tumor suppressor genes regulating the cell cycle gene expression, invasion and intrinsic
tumorigenicity of melanoma cells. Carcinogenesis. 2012;33:687-693.
137. Garrido F, Ruiz-Cabello F, Aptsiauri N. Rejection versus escape: the tumor MHC dilemma. Cancer Immunol Immunother. 2017;66:259-271.
138. Rodig SJ, Gusenleitner D, Jackson DG, et al. MHC proteins confer differential sensitivity to CTLA-4 and PD-1 blockade in untreated metastatic melanoma. Sci
Transl Med. 2018;10:eaar3342.
139. Hopkins AC, Yarchoan M, Durham JN, et al. T cell receptor repertoire features associated with survival in immunotherapy-treated pancreatic ductal ade-
nocarcinoma. JCI Insight. 2018;3:e122092.
140. Snyder A, Nathanson T, Funt SA, et al. Contribution of systemic and somatic factors to clinical response and resistance to PD-L1 blockade in urothelial cancer:
an exploratory multi-omic analysis. PLoS Med. 2017;14:e1002309.
141. Forde PM, Chaft JE, Smith KN, et al. Neoadjuvant PD-1 blockade in resectable lung cancer. N Engl J Med. 2018;378:1976-1986.
142. Inoue H, Park JH, Kiyotani K, et al. Intratumoral expression levels of PD-L1, GZMA, and HLA-A along with oligoclonal T cell expansion associate with response to
nivolumab in metastatic melanoma. OncoImmunology. 2016;5:e1204507.
143. Agrawal L, Engel KB, Greytak SR, et al. Understanding preanalytical variables and their effects on clinical biomarkers of oncology and immunotherapy. Semin
Cancer Biol. 2018;52:26-38.
144. Masucci GV, Cesano A, Hawtin R, et al. Validation of biomarkers to predict response to immunotherapy in cancer: volume I - pre-analytical and analytical
validation. J Immunother Cancer. 2016;4:76.
145. Rimm DL, Han G, Taube JM, et al. A prospective, multi-institutional, pathologist-based assessment of 4 immunohistochemistry assays for PD-L1 expression in
non-small cell lung cancer. JAMA Oncol. 2017;3:1051-1058.
146. McShane LM. In pursuit of greater reproducibility and credibility of early clinical biomarker research. Clin Transl Sci. 2017;10:58-60.
147. Nixon AB, Schalper KA, Jacobs I, et al. Peripheral immune-based biomarkers in cancer immunotherapy: can we realize their predictive potential? J Immunother
Cancer. 2019;7:325.

Leveraging Patient-Derived Models for

Immunotherapy Research
Katerina Politi, PhD1
overview
As cancer immunotherapies become mainstream for the treatment of many different cancer types and the
numbers of new agents continue to increase, the need for experimental models is also rising. An approach to
develop and study models for immune-oncology that has garnered intense interest in recent years is that of
using patient-derived models. Patient-derived models can recapitulate many of the features and heterogeneity
of the corresponding human tumors. Historically these models have been used to study cancer cell–intrinsic
properties of tumors and drugs that target tumor cells directly. In recent years, increasing recognition of the
role immune cells play in cancer and how these represent good therapeutic targets has led to efforts to
optimize and use patient-derived models for cancer immunotherapy studies. Patient-derived models are now
being used to study the properties of cancer cells that modulate their ability to respond to immune stimulation.
Further efforts are underway to use and develop patient-derived models that incorporate human immune cells
in vitro and in vivo (humanized mice) so that cancer cell–immune cell interactions can be studied in the
context of cancer immunotherapies. As these models are further refined, leveraging patient-derived models for
cancer immunotherapy research can provide insight into mechanisms of sensitivity and resistance to cancer
immunotherapies, uncover new targets, reveal how specific agents work, and be used to evaluate the an-
titumor efficacy of therapeutic regimens.
INTRODUCTION immune system that is a critically important feature of

Immunotherapeutic approaches to treat patients with models for immunotherapy research and allows for
many different cancer types are increasingly becoming studies of both the innate and adaptive immune sys-
mainstream. Moreover, the number of new immuno- tems. Genetically engineered mouse models, which
therapies and immunotherapy combinations under also have an intact immune system, are increasingly
clinical development and investigation is also rising.1 being used, given that tumors in these mice develop
Additionally, the recent success and renewed interest autochtonously in a tissue and are subject to immune
in immunotherapies has prompted an expansion of pressures that occur during tumor development.
research that is uncovering new targets and the de- However, many genetically engineered cancer models
velopment of new immunotherapy drugs.2 These re- harbor very few mutations beyond the ones that have
markable advances also raise a number of questions been introduced genetically and thus are poorly anti-
related to how and when to use specific immunogenic (for example, lung cancer models4,5). This rep-
therapies and what are the underlying mechanisms resents a challenge in cancer immunology research,
of sensitivity and resistance to the agents. Although especially for studies of adaptive immunity. Efforts to
addressing these questions directly in patients through optimize genetically engineered mouse models to
clinical trials and studies of patient specimens is on- harbor tumors with a higher mutation load are ongoing
going, the sheer combinatorial complexity of all the and will likely represent valuable tools as studies of
Author affiliations
and support possible therapies, indications, and analyses render cancer immunotherapy progress.
information (if this task extremely challenging. Experimental models An additional area of rapid model development has
applicable) appear can play an important role in shedding light on new
at the end of this
come from work to adapt patient-derived models for
targets, providing mechanistic insights into the biology cancer immunology studies. Patient-derived cancer
article.
of the disease under investigation, and contributing to cell lines or xenografts mirror the variability and genetic
Accepted on May 11,
2020 and published prioritizing therapeutic avenues to be studied. landscape of human tumors and thus are an incredible
at ascopubs.org on Historically, the use of mouse cancer cell lines resource for cancer research.6 However, their use in
May 22, 2020:
transplanted into syngeneic hosts has played—and cancer immunology has been hampered by the re-
10.1200/EDBK_ continues to play—a pivotal role in cancer immunology quirement for a functional immune system necessary
280579 research.3 In these models, the hosts have an intact for research in this field. In recent years, investigators

Patient-Derived Models for Immunotherapy Research
and transcriptomic profiling data can be performed on these

high–tumor purity frozen specimens (rather than more
heterogeneous formalin-fixed paraffin-embedded speci-
• Patient-derived models recapitulate several of mens), yielding high-quality data that can be used to vali-
the features of the original patient tumor
date findings from the patient specimen and also provide
specimen.
a more detailed picture of the -omic landscape of the tu-
• Patient-derived models can be used to assess mor.8 Patient-derived models are also amenable to genetic
the ability of a tumor to respond to IFNγ, which modification and can be processed for analysis of single-cell
can reveal potential defects in this pathway and
transcriptomics and epigenomics to gain further insight into
downstream targets such as antigen pre-
sentation that have been linked to immune the biology of the tumor. Given the increasing appreciation
checkpoint inhibitor resistance. for the importance of tumor cell–intrinsic properties in
modulating the responsiveness to immunotherapies, un-
• Several approaches have been developed to
derstanding the biology of the cancer cells is critically
incorporate human immune cells into mice that
harbor patient-derived xenografts, and these important.
can be used to study the efficacy of cancer A unique application of patient-derived models for immu-
immunotherapies. notherapy research comes from the fact that these can be
• Three-dimensional cultures of patient-derived used to measure dynamic changes in a tumor in response to
tumor and immune cells are emerging that an external stimulus (Fig. 1). Of particular relevance in the
could provide a rapid platform with which to context of cancer immunology, for example, is that the
evaluate cancer immunotherapies. ability of a patient-derived model to respond to IFN stim-
ulation can be measured. In melanoma, where patient-
have developed approaches to use the patient-derived derived cell lines can readily be generated from tumors,
models to understand tumor cell–intrinsic processes im- a panel of primary cell lines was used to assess their ability to
portant for cancer immunology and identified clever strat- respond to IFN gamma (IFNγ) stimulation by measuring the
egies to incorporate human immune cells into the models change in cell surface PD-L1 by flow cytometry.9 PD-L1 is
both in mice and in three-dimensional culture systems. a target of IFNγ, and the expectation is that its levels should
Here, we review these approaches and discuss their ap- increase on treatment of the cell lines with the cytokine. In
plications for cancer immunotherapy studies with an em- this study, although the baseline levels of PD-L1 were similar
phasis on studies of solid tumors. across most of the lines, there were several patterns of PD-
L1 expression observed upon IFNγ treatment. Although, in
PATIENT-DERIVED MODELS TO STUDY RESPONSIVENESS TO most cell lines, PD-L1 expression increased on treatment, in
IMMUNE STIMULATION approximately 8% of the lines, PD-L1 levels either did not
It is increasingly becoming clear that several properties of increase or increased in a very limited way. In the lines that
cancer cells and signaling mechanisms within them are did not show upregulation of PD-L1 on IFNγ treatment, the
important determinants of sensitivity and resistance to authors then identified cell lines with mutations in the JAK1
immunotherapies, with most studies focused on T cell– and JAK2 adapter proteins that transduce the IFNγ signal.
directed therapies to date.7 These include tumor antige- As a result of these studies, these adapter proteins have
nicity, the status of antigen presentation, and interferon been linked to primary and acquired resistance to immune
(IFN) signaling pathways and other intracellular signaling checkpoint inhibitors.9,10 Importantly, some of the cell lines
networks that influence interactions of cancer cells with identified using this functional IFNγ assay did not have
immune cells. Some of these lend themselves to direct obvious genetic alterations in the IFNγ pathway, despite
analysis in patient samples. For example, the tumor mu- exhibiting very low induction of PD-L1, suggesting that these
tation burden (a proxy for tumor antigenicity) and the may represent tumors with other unknown defects in the
identification of mutations in specific pathways that mod- pathway that warrant further investigation. In lung cancer,
ulate immune responses can be identified by direct se- patient-derived xenografts of tumors resistant to immune
quencing of the patient specimen. Similarly, transcriptional checkpoint inhibitors were used to identify tumors with
and immune profiling can be used to study active pathways defects in HLA I antigen presentation (antigen presentation
and cellular components present in the patient specimen. In genes are induced by IFNγ treatment) by treating the tumors
addition to studying the patient specimen directly, analysis with IFNγ and using flow cytometry to quantify the levels of
of patient-derived models can be advantageous. Indeed, cell surface HLA I and β2-microglobulin.11,12 This work shed
patient-derived models frequently will have a higher tumor light on how tumors can have defects in the IFNγ-sensing
purity than their corresponding patient specimens, and mechanisms, not all of which can be explained through
frozen specimens are likely to be available. Thus, genomic genetic alterations in known IFNγ/antigen presentation

Politi
pathway components, similar to the observations in mela- the use of mouse strains lacking mouse antigen pre-
noma. Collectively, these studies highlight the utility of as- sentation molecules such as β2-microglobulin or major
says to determine the ability of tumors to effectively respond histocompatibility complex I/II.18,19 An additional experi-
to immune stimulation, which could provide insights into mental variable in these studies is whether autologous or
resistance to immune checkpoint inhibitors and other allogeneic immune cells are used. Using autologous, rather
immunotherapies. than allogeneic (even if HLA matched to the tumor), im-
HUMANIZED MICE mune cells has the advantage of ensuring a perfect match
between the tumor cells and immune cells and eliminates
Approaches to Develop Humanized Mice potential confounding factors that could arise from using
Throughout the years, a variety of mice have been cells from different individuals. CD34+ hematopoietic stem
developed—with varying degrees of immunodeficiency—in cells (HSCs) can also be used to reconstitute the human
which to engraft human cancer cells. Models that lack immune system in immunodeficient mice.20 These can be
T cells, B cells, and natural killer (NK) cells optimally derived from a range of sources including, for example, cord
support engraftment of primary human tumor and immune blood, bone marrow, and fetal liver. Engraftment of HSCs
cells and are most frequently used for these studies, such is longer lasting than that of PBMCs, and multiple human
as NSG (NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ), NOG (NOD-Cg- immune cell lineages are observed in the models including
Prkdcscid Il2rgtm1Sug/JicTac), and BRG (BALB/c.Cg-Rag2 B and T lymphocytes, myeloid cells, and NK cells. However,
tm1Fwa
Il2rgtm1Sug/JicTac) models.13-16 These severely im- the immune cells derived from HSCs in these models have
munocompromised mice can be engrafted with primary varying degrees of maturation defects and functional im-
human tumors to establish patient-derived xenografts and pairment, which can thus be a confounding issue.
then reconstituted with human immune cells in several In recent years, there have been additional advances in the
ways. T cells are the major immune cell type present after development of humanized mice for cancer immunotherapy
transfer of peripheral blood mononuclear cells (PBMCs) into research. In particular, these efforts have focused on
mice, where they persist for 4–6 weeks. Alternatively, T modifying the mouse host genetically to introduce alter-
lymphocytes can be expanded from the PBMCs or from the ations that better support the engraftment and function of
tumor and transferred into the mouse harboring the xe- specific human immune cells like myeloid, NK, and B cells,
nograft. A general limitation of this methodology, however, is which exhibit defects in many immunodeficient mouse
that human immune cells will recognize mouse cells as models.21 This is especially important given the well-
foreign and lead to graft-versus-host disease.16-18 Therefore, documented differences between the mouse and human
experiments can only be performed within the window of immune systems.22,23 One such approach has come from
time before the emergence of graft-versus-host disease. genetically engineering severely immunocompromised
Potential ways of avoiding graft-versus-host disease include mice to express specific human cytokines and growth
factors, in varying combinations, like granulocyte macro-
phage colony-stimulating factor, interleukin 3 (IL-3), stem
cell factor (which is a KIT ligand), thrombopoietin, and the
signal regulatory protein alpha. Depending on the genetic
modifications present, different subsets of immune cells
are affected. Thus, in mice engineered to express gran-
ulocyte macrophage colony-stimulating factor and IL-3,
increased macrophage and dendritic cell differentiation is
observed,24-27 and models expressing human IL-2 or IL-15
exhibit improved development and persistence of NK
cells.28-30 Additional modifications to further humanize
mice can be achieved by generating mice that express spe-
cific human HLA I or II alleles, facilitating analysis of antigen-
specific responses and improving B- and T-lymphocyte
functionality.31-34
FIGURE 1. Patient-Derived Models to Study Responsiveness to Interferon
Stimulation Applications of Humanized Mice for
Two- and three-dimensional patient-derived culture systems and xe-
Cancer Immunotherapy
nografts can be used to assess the ability of a tumor to respond to in-
terferon stimulation and upregulate interferon-responsive pathways Immune checkpoint inhibitors and other immunotherapies
such as HLA I and II antigen presentation and PD-L1. Created with Humanized mice reconstituted with PBMCs have been
BioRender.com. used to test the antitumor efficacy of immunotherapeutics.35

For example, in a xenograft model derived from a patient and hematologic cancers mostly with established cell line
with gastric cancer, in which autologous PBMCs were xenografts and in some cases patient-derived xenografts.35
transferred, treatment with urelumab and nivolumab re- These models can be useful to establish and optimize the
strained tumor growth, and treated tumors were found to conditions required for the antitumor efficacy of adoptive
have increased tumor-infiltrating lymphocytes (TILs) and cell therapies. For example, in studies of TIL therapies,
decreased regulatory T cells.36 Similarly, a carcinoem- adoptive transfer of autologous T cells into mice implanted
bryonic antigen T-cell bispecific antibody was tested on with patient-derived melanoma xenografts has been shown
carcinoembryonic antigen–expressing colorectal tumor to lead to responses that mirror those observed in the
cells implanted subcutaneously into NOG mice and in corresponding patients to TIL therapy and that these re-
which human allogeneic PBMCs had been transferred.37 sponses required the presence of human IL-2.41 Further-
These approaches all rely on the transfer of peripheral more, the animal models can be used to identify and
immune cells into the mouse, which could have properties study the antitumor reactivity of specific subsets of tumor-
that are different from native TILs. An alternative is to im- infiltrating T cells.42 Similar approaches can be used to
plant a tumor directly into the mouse and evaluate the study NK cell therapies. Indeed, NK cells transferred into
effects of immunotherapies immediately after implantation mice expressing human IL-15 were shown to exhibit anti-
when TILs are still present. This method was applied to tumor effects, highlighting the potential of using these
surgical lung cancers implanted into mice and then treated models to study NK cell therapies.30
with anti–PD-1, where the treatment was shown to boost the
Mouse models harboring human-derived tumors can also
levels of granzyme B and Ki-67 in TILs.38 Although this
be used to study chimeric antigen receptor (CAR) T cells.
approach is not optimal for studying the antitumor efficacy
For example, they can be used to establish the validity of
of the therapies, it can be used to study qualitative changes
a specific target. Indeed, autologous CAR T cells targeting
in tumor and immune cells after treatment.
mesothelin were shown to lead to the regression of meso-
Response to anti–PD-1 therapy has also been studied in thelioma in vivo,43 highlighting the promise of mesothelin-
PDX models reconstituted with partially HLA-matched targeted CARs for this disease. These models can also
CD34+ HSCs in multiple tumor types. 39 In these ex- provide insights into mechanisms of resistance to CAR
periments, CD8 T cell–dependent responses to anti–PD- T-cell therapies. In a recent study using a leukemia model,
1 therapy were observed in NSG mice humanized with trogocytosis of the target antigen by CAR T cells was shown
CD34+ cells but not standard NSG mice. Interestingly, to lead to therapeutic resistance.44 Finally, these mice can
CD34+ cells from different donors can give rise to different also be used to develop approaches to counter the toxicities
results, indicating that the properties of individual donor associated with CAR T-cell therapies.45
pools of cells can affect the outcomes of these studies. A
In summary, there are multiple ways in which humanized
similar model has also been used to study the antitumor
mice and patient-derived models can be applied to studying
effects of a histone deacetylase inhibitor and anti–PD-1 in
triple-negative breast cancer xenografts.40 specific immunotherapies (Fig. 2). Moreover, the variety of
options available to humanize mice provides investigators
Adoptive cell therapies Immunodeficient mice are being with the possibility of selecting the optimal model for their
used to study adoptive cell therapies across various solid specific studies.
FIGURE 2. Humanized Mice Ap-

plied to Cancer Immunology
Research
Immunodeficient mice harbor-
ing xenografts can be recon-
stituted with immune cells using
PBMCs, HSCs, or specific sub-
sets of expanded and/or engi-
neered lymphocytes or NK cells.
On reconstitution, the activity of
immunotherapeutic agents can
be studied. Created with Bio-
Render.com.
Abbreviations: PBMC, periph-
eral blood mononuclear cell;
HSC, hematopoietic stem cell; NK, natural killer.

Politi
THREE-DIMENSIONAL CULTURE MODELS 7 days led to induction of markers of T-cell activation like
Tremendous advances have been made in recent years on IFNγ, PRF1, and GZMB and, in some cases, expansion of
the establishment of three-dimensional culture systems to TILs and cancer cell death. A major advantage of this
propagate and study patient-derived models in a manner system is that it maintains the stromal elements present
that better recapitulates the heterogeneity and intercellular from the tumor and not just from the periphery.
interactions present in tumors. These models can largely LIMITATIONS
be subdivided into two categories either (1) based on the Work with patient-derived models is complex and entails
reconstitution of epithelial cancer-derived models with multiple regulatory requirements and a robust infrastructure
autologous peripheral immune cells or (2) based on ap- to ensure that one can successfully obtain the specimens,
proaches that rely on tumor-infiltrating immune cells generate the models, and perform adequate quality control
present in the tumor and ex vivo tumor models. In an before experimental studies.49 Such a process can be
example of the former, mismatch-deficient colorectal prohibitively costly. Moreover, depending on the type of
cancer and non–small cell lung cancer patient-derived cancer, the efficiency with which patient-derived models
organoids (PDOs), coculture of the PDOs, and autolo- can be established is variable and, in some cases, can be
gous PBMCs (in media containing an anti–PD-1 anti- quite low. This can lead to the generation of small numbers
body) led to an increase in IFNγ secretion and induction of patient-derived models and over-representation of tumors
of the degranulation marker CD107a on CD8+ cells in with characteristics that allow them to grow in culture or in
approximately 50% and 30% of cases, respectively. 46 a mouse. An additional challenge comes from work with
Moreover, in a subset of the models, coculture of T cells humanized mice, where ideally, autologous cells are used to
with the PDOs led to T cell–dependent apoptosis of the reconstitute the immune system requiring sampling of blood
tumor organoids but not nontumor organoids. This type of (often recurrently), mobilized HSCs, or bone marrow from
approach has the advantage that tumor organoids can the patient.
be established independently from the immune cell
There are two additional considerations when working with
cultures, both of which can be frozen and thawed as
patient-derived models. First, most often these models are
needed. However, a limitation is that these studies rely
established in a microenvironment (whether in culture or in
on peripheral immune cells rather than tumor-infiltrating
mice) that does not mimic the one of the cancer in the
immune cells, which could have different properties,
patient. In particular, in mice, many patient-derived xeno-
and that these generally lack other stromal components grafts are generated by implanting the tumor subcutaneously.
of the tumor microenvironment, which could also con- Efforts to implant the tumors orthotopically may be valuable to
tribute to modulating tumor–immune cell interactions overcome this issue. Similarly, in culture, it is possible to
and responses to immunotherapies. supplement the tissue culture media with key growth
Several methods have been used to culture patient-derived factors and molecules present in the corresponding mi-
tumors with their native stromal cells including using croenvironment from which the tumor was derived. Sec-
a microfluidic device to culture patient-derived organotypic ond, in many of the humanized mouse models and three-
tumor spheroids or the use of an air–liquid interface. dimensional culture systems, cancer cells and immune
Patient-derived organotypic tumor spheroids were gener- cells are acutely exposed to each other. Because this is not
ated from patient-derived specimens of several types of what occurs in patients, how accurately the changes ob-
cancer, including melanoma, Merkel cell carcinoma, and served in both cell types recapitulate what occurs in a tumor in
thyroid cancer. These models were used to evaluate the patients is unclear. Complementing such studies with direct
consequences of anti–PD-1 treatment and revealed mod- analysis of patient specimens and studies in autochthonous
ulation of specific cytokines in response to treatment.47 genetically engineered or carcinogen models will be
important.
A promising approach to culture epithelial cancer cells and
their native stroma comes from the use of an air–liquid CONCLUSIONS
interface to grow the organoids. Using the air–liquid in- As immunotherapy comes of age in the oncology clinic,
terface, Neal and colleagues successfully established PDOs patient-derived models of cancer are increasingly playing
from a wide range of tumor types including colon, lung, a role in cancer immunotherapy research. There has been
pancreas, and kidney cancer.48 In these models, native significant progress in the development of approaches to
stromal elements including fibroblasts and immune cells generate and analyze patient-derived models of cancer and
were preserved for 1–2 months, although these did pro- more recently to incorporate immune cells to allow for
gressively decrease over time. Treatment of PDOs, estab- cancer immunology studies. Together with the novel tools
lished from non–small cell lung cancer, renal cell carcinoma, available now, such as analysis of single cells and genome
and melanoma, with the anti–PD-1 antibody nivolumab for editing technologies, and new drugs to target components

of the immune system, patient-derived models represent underlie sensitivity and resistance to cancer immunother-
a valuable tool that can provide insight into mechanisms that apies and contribute to unveiling new targets.
AFFILIATION of Medicine, 333 Cedar St., SHM-I 234D, New Haven, CT 06510;
1
Departments of Pathology and Internal Medicine (Section of Medical Twitter: @politikaterina; email: [email protected].
Oncology), Yale Cancer Center, Yale School of Medicine, New Haven, CT
CORRESPONDING AUTHOR Disclosures provided by the authors and data availability statement (if
Katerina Politi, PhD, Departments of Pathology and Internal Medicine applicable) are available with this article at DOI https://doi.org/10.1200/
(Section of Medical Oncology), Yale Cancer Center, Yale University School EDBK_280579.
REFERENCES
1. Tang J, Yu JX, Hubbard-Lucey VM, et al. Trial watch: the clinical trial landscape for PD1/PDL1 immune checkpoint inhibitors. Nat Rev Drug Discov. 2018;
17:854-855.
2. Xin Yu J, Hubbard-Lucey VM, Tang J. Immuno-oncology drug development goes global. Nat Rev Drug Discov. 2019;18:899-900.
3. Olson B, Li Y, Lin Y, et al. Mouse models for cancer immunotherapy research. Cancer Discov. 2018;8:1358-1365.
4. McFadden DG, Politi K, Bhutkar A, et al. Mutational landscape of EGFR-, MYC-, and Kras-driven genetically engineered mouse models of lung adenocarcinoma.
Proc Natl Acad Sci USA. 2016;113:E6409-E6417.
5. Westcott PM, Halliwill KD, To MD, et al. The mutational landscapes of genetic and chemical models of Kras-driven lung cancer. Nature. 2015;517:489-492.
6. Byrne A, Alférez D, Amant F, et al. Interrogating open issues in cancer precision medicine with patient-derived xenografts. Nat Rev Cancer. 2017;17:254-268.
7. Kalbasi A, Ribas A. Tumour-intrinsic resistance to immune checkpoint blockade. Nat Rev Immunol. 2020;20:25-39.
8. Pereira C, Gimenez-Xavier P, Pros E, et al. Genomic profiling of patient-derived xenografts for lung cancer identifies B2M inactivation impairing immunor-
ecognition. Clin Cancer Res. 2017;23:3203-3213.
9. Shin DS, Zaretsky JM, Escuin-Ordinas H, et al. Primary resistance to PD-1 blockade mediated by JAK1/2 mutations. Cancer Discov. 2017;7:188-201.
10. Zaretsky JM, Garcia-Diaz A, Shin DS, et al. Mutations associated with acquired resistance to PD-1 blockade in melanoma. N Engl J Med. 2016;375:819-829.
11. Gettinger S, Choi J, Hastings K, et al. Impaired HLA class I antigen processing and presentation as a mechanism of acquired resistance to immune checkpoint
inhibitors in lung cancer. Cancer Discov. 2017;7:1420-1435.
12. Cardenas JJ, Robles-Oteiza C, Politi K. Assessment of IFNγ responsiveness in patient-derived xenografts. Methods Enzymol. 2020;631:415-427.
13. Ishikawa F, Yasukawa M, Lyons B, et al. Development of functional human blood and immune systems in NOD/SCID/IL2 receptor gamma chain(null) mice.
Blood. 2005;106:1565-1573.
14. Shultz LD, Lyons BL, Burzenski LM, et al. Human lymphoid and myeloid cell development in NOD/LtSz-scid IL2R gamma null mice engrafted with mobilized
human hemopoietic stem cells. J Immunol. 2005;174:6477-6489.
15. Shultz LD, Ishikawa F, Greiner DL. Humanized mice in translational biomedical research. Nat Rev Immunol. 2007;7:118-130.
16. Ito M, Hiramatsu H, Kobayashi K, et al. NOD/SCID/gamma(c)(null) mouse: an excellent recipient mouse model for engraftment of human cells. Blood. 2002;
100:3175-3182.
17. Mosier DE, Gulizia RJ, Baird SM, et al. Transfer of a functional human immune system to mice with severe combined immunodeficiency. Nature. 1988;
335:256-259.
18. King MA, Covassin L, Brehm MA, et al. Human peripheral blood leucocyte non-obese diabetic-severe combined immunodeficiency interleukin-2 receptor
gamma chain gene mouse model of xenogeneic graft-versus-host-like disease and the role of host major histocompatibility complex. Clin Exp Immunol. 2009;
157:104-118.
19. Yaguchi T, Kobayashi A, Inozume T, et al. Human PBMC-transferred murine MHC class I/II-deficient NOG mice enable long-term evaluation of human immune
responses. Cell Mol Immunol. 2018;15:953-962.
20. Yu CI, Marches F, Wu TC, et al. Techniques for the generation of humanized mouse models for immuno-oncology. Methods Enzymol. 2020;636:351-368.
21. Rongvaux A, Takizawa H, Strowig T, et al. Human hemato-lymphoid system mice: current use and future potential for medicine. Annu Rev Immunol. 2013;
31:635-674.
22. Hagai T, Chen X, Miragaia RJ, et al. Gene expression variability across cells and species shapes innate immunity. Nature. 2018;563:197-202.
23. Mestas J, Hughes CC. Of mice and not men: differences between mouse and human immunology. J Immunol. 2004;172:2731-2738.
24. Billerbeck E, Barry WT, Mu K, et al. Development of human CD4+FoxP3+ regulatory T cells in human stem cell factor-, granulocyte-macrophage colony-
stimulating factor-, and interleukin-3-expressing NOD-SCID IL2Rγ(null) humanized mice. Blood. 2011;117:3076-3086.

Politi
25. Rongvaux A, Willinger T, Martinek J, et al. Development and function of human innate immune cells in a humanized mouse model. Nat Biotechnol. 2014;
32:364-372.
26. Ito R, Takahashi T, Katano I, et al. Establishment of a human allergy model using human IL-3/GM-CSF-transgenic NOG mice. J Immunol. 2013;191:2890-2899.
27. Willinger T, Rongvaux A, Takizawa H, et al. Human IL-3/GM-CSF knock-in mice support human alveolar macrophage development and human immune
responses in the lung. Proc Natl Acad Sci USA. 2011;108:2390-2395.
28. Katano I, Takahashi T, Ito R, et al. Predominant development of mature and functional human NK cells in a novel human IL-2-producing transgenic NOG mouse.
J Immunol. 2015;194:3513-3525.
29. Katano I, Nishime C, Ito R, et al. Long-term maintenance of peripheral blood derived human NK cells in a novel human IL-15- transgenic NOG mouse. Sci Rep.
2017;7:17230.
30. Herndler-Brandstetter D, Shan L, Yao Y, et al. Humanized mouse model supports development, function, and tissue residency of human natural killer cells. Proc
Natl Acad Sci USA. 2017;114:E9626-E9634.
31. Shultz LD, Saito Y, Najima Y, et al. Generation of functional human T-cell subsets with HLA-restricted immune responses in HLA class I expressing NOD/SCID/
IL2r gamma(null) humanized mice. Proc Natl Acad Sci USA. 2010;107:13022-13027.
32. Danner R, Chaudhari SN, Rosenberger J, et al. Expression of HLA class II molecules in humanized NOD.Rag1KO.IL2RgcKO mice is critical for development and
function of human T and B cells. PLoS One. 2011;6:e19826.
33. Suzuki M, Takahashi T, Katano I, et al. Induction of human humoral immune responses in a novel HLA-DR-expressing transgenic NOD/Shi-scid/γcnull mouse. Int
Immunol. 2012;24:243-252.
34. Zeng Y, Liu B, Rubio MT, et al. Creation of an immunodeficient HLA-transgenic mouse (HUMAMICE) and functional validation of human immunity after transfer
of HLA-matched human cells. PLoS One. 2017;12:e0173754.
35. De La Rochere P, Guil-Luna S, Decaudin D, et al. Humanized mice for the study of immuno-oncology. Trends Immunol. 2018;39:748-763.
36. Sanmamed MF, Rodriguez I, Schalper KA, et al. Nivolumab and urelumab enhance antitumor activity of human T lymphocytes engrafted in Rag2-/-IL2Rγnull
immunodeficient mice. Cancer Res. 2015;75:3466-3478.
37. Bacac M, Fauti T, Sam J, et al. A novel carcinoembryonic antigen T-cell bispecific antibody (CEA TCB) for the treatment of solid tumors. Clin Cancer Res. 2016;
22:3286-3297.
38. Gettinger SN, Choi J, Mani N, et al. A dormant TIL phenotype defines non-small cell lung carcinomas sensitive to immune checkpoint blockers. Nat Commun.
2018;9:3196.
39. Wang M, Yao LC, Cheng M, et al. Humanized mice in studying efficacy and mechanisms of PD-1-targeted cancer immunotherapy. FASEB J. 2018;
32:1537-1549.
40. Capasso A, Lang J, Pitts TM, et al. Characterization of immune responses to anti-PD-1 mono and combination immunotherapy in hematopoietic humanized mice
implanted with tumor xenografts. J Immunother Cancer. 2019;7:37.
41. Jespersen H, Lindberg MF, Donia M, et al. Clinical responses to adoptive T-cell transfer can be modeled in an autologous immune-humanized mouse model. Nat
Commun. 2017;8:707.
42. Ye Q, Song DG, Poussin M, et al. CD137 accurately identifies and enriches for naturally occurring tumor-reactive T cells in tumor. Clin Cancer Res. 2014;
20:44-55.
43. Zhao Y, Moon E, Carpenito C, et al. Multiple injections of electroporated autologous T cells expressing a chimeric antigen receptor mediate regression of human
disseminated tumor. Cancer Res. 2010;70:9053-9061.
44. Hamieh M, Dobrin A, Cabriolu A, et al. CAR T cell trogocytosis and cooperative killing regulate tumour antigen escape. Nature. 2019;568:112-116.
45. Kim MY, Yu KR, Kenderian SS, et al. Genetic inactivation of CD33 in hematopoietic stem cells to enable CAR T cell immunotherapy for acute myeloid leukemia.
Cell. 2018;173:1439-1453.
46. Dijkstra K, Cattaneo C, Weeber F, et al. Generation of Tumor-Reactive T Cells by Co-culture of Peripheral Blood Lymphocytes and Tumor Organoids. Cell. 2018;
174:1586-1598.
47. Jenkins RW, Aref AR, Lizotte PH, et al. Ex vivo profiling of PD-1 blockade using organotypic tumor spheroids. Cancer Discov. 2018;8:196-215.
48. Neal JT, Li X, Zhu J, et al. Organoid modeling of the tumor immune microenvironment. Cell. 2018;175:1972-1988.
49. Meehan TF, Conte N, Goldstein T, et al. PDX-MI: minimal information for patient-derived tumor xenograft models. Cancer Res. 2017;77:e62-e66.

DEVELOPMENTAL
THERAPEUTICS—
MOLECULARLY TARGETED
AGENTS AND TUMOR
BIOLOGY

Antibody-Drug Conjugates: Patient and

Treatment Selection
Shalini Makawita, MD1 and Funda Meric-Bernstam, MD1,2
overview
Antibody-drug conjugates (ADCs) are a promising drug platform designed to enhance the therapeutic index
and minimize the toxicity of anticancer agents. ADCs have experienced substantial progress and technological
growth over the past decades; however, several challenges to patient selection and treatment remain. Methods
to optimally capture all patients who may benefit from a particular ADC are still largely unknown. Although
target antigen expression remains a biomarker for patient selection, the impact of intratumor heterogeneity on
antigen expression, as well as the dynamic changes in expression with treatment and disease progression, are
important considerations in patient selection. Better understanding of these factors, as well as minimum
levels of target antigen expression required to achieve therapeutic efficacy, will enable further optimization of
selection strategies. Other important considerations include understanding mechanisms of primary and
acquired resistance to ADCs. Ongoing efforts in the design of its constituent parts to possess the intrinsic
ability to overcome these mechanisms, including use of the “bystander effect” to enhance efficacy in het-
erogeneous or low target antigen-expressing tumors, as well as modulation of the chemical and immuno-
phenotypic properties of antibodies and linker molecules to improve payload sensitivity and therapeutic
efficacy, are under way. These strategies may also lead to improved safety profiles. Similarly, combination
strategies using ADCs with other cytotoxic or immunomodulatory agents are also under development. Great
strides have been made in ADC technology. With further refinements, this therapeutic modality has the
potential to make an important clinical impact on a wider range of tumor types.
INTRODUCTION because they represented a therapeutic modality with

Antibody-drug conjugates are a complex drug-delivery less toxicity in comparison with standard chemother-
system designed to enhance toxicity to target cells apy agents to the normal tissues of the organism.5,6
and minimize off-target effects. They are the “homing To enhance their therapeutic efficacy, conjugation to
missiles” of anticancer therapy and consist of three various anticancer effector molecules was postulated.
critical elements: (1) an antibody with specificity for an Antibodies have since been linked to numerous types
antigen, ideally expressed solely on tumor cells, (2) of cytotoxic agents, as well as to immunotoxins derived
from plant and bacterial protein toxins and radio-
a cytotoxic agent (or payload) that is designed to cause
pharmaceuticals.7 However, initial attempts were not
destruction of the target cell once internalized and
without substantial unanticipated toxicities from off-
released, and (3) a linker molecule that attaches the
target effects. In the early 1990s, conjugation of an
cytotoxic agent to the antibody.1-3 Once in the systemic
immunoglobulin G1 antibody targeting the LewisY an-
circulation, the ADC would make its way to its cellular
tigen, coupled via a cleavable linker to doxorubicin
target, resulting in internalization and trafficking of
at a drug/antibody ratio (DAR) of approximately 4,
the complex, most commonly via the endolysosomal
Author affiliations resulted in the development of BR-96 doxorubicin.8
and support
degradation pathway. The mechanism by which the
Preclinical data in murine models of breast cancer had
information (if cytotoxic payload is released is dependent on the
shown substantial antitumor effects of BR-96 doxo-
applicable) appear chemistry of the linker molecule used. Once released,
at the end of this
rubicin resulting in complete responses (CRs) and
the payload is free to cause cellular destruction via its
article. sustained cures; however, this had not translated into
prespecified pathway. Multiple molecules of the cy-
Accepted on successes in human clinical trials. Initial human
totoxic agent can be conjugated to each antibody for
February 28, 2020 studies had shed light on unexpected on-target effects
and published at delivery, thereby increasing the payload. The mech-
leading to substantial gastrointestinal toxicities, in-
ascopubs.org on anism of ADCs has been reviewed in great detail
cluding nausea, vomiting, and gastric ulceration and
March 26, 2020: previously.1-5
bleeding, caused by expression of LewisY antigen in
10.1200/EDBK_ In the late 20th century, development of monoclonal gastric mucosal cells in humans.9 Randomized phase
280775 antibodies for the treatment of cancer garnered interest, II data comparing BR-96 doxorubicin (700 mg/m2)

Makawita and Meric-Bernstam
with bendamustine and rituximab (2019)17; enfortumab

vedotin-ejfv for locally advanced or metastatic bladder
cancer after progression on PD-1 or PD-L1 and platinum-
• ADCs are a drug platform consisting of an an- based chemotherapy (2019) 18; and fam-trastuzumab
tibody, a cytotoxic payload, and a linker mole-
deruxtecan-nxki (DS-8201) for HER2-positive breast can-
cule designed to enhance the therapeutic index
cer (2019).19 In this article, we review potential strategies
and minimize toxicity of anticancer agents.
for improvement of patient selection, the use of histology as
• Optimal patient selection strategies will require a surrogate for target expression, the impact of intratumor
consideration of intratumor heterogeneity of
heterogeneity on patient selection, overcoming mecha-
target antigen expression and dynamic changes
nisms of resistance, and other factors that are considered
in expression of target antigen with treatment
and disease progression. in the design of ADCs, such as toxicity and safety profile
(Fig. 1).
• Optimizing thresholds of target antigen ex-
pression that demonstrate clinical therapeutic
PATIENT-SELECTION STRATEGIES
benefit will improve applicability of ADCs in the
clinical area and improve patient selection. Selection Based on Tissue Expression of Target Antigen
• Utilization of bystander effect and appropriate Intuitively, target antigen expression seems to be the most
pairing of the cytotoxic payload, modulation of appropriate biomarker for patient selection and, certainly,
chemical properties of antibodies and linker ADC efficacy has been shown to correlate with the level of
molecules, as well as combination therapies are target antigen expression in some studies. T-DM1 offers an
being explored to overcome mechanisms of
example of an ADC whereby tissue expression of the target
resistance and improve efficacy and safety
antigen correlates with clinical efficacy. T-DM1 consists of
profiles of ADCs.
the monoclonal antibody trastuzumab, which targets HER2
found on the cell surface of approximately 20% of breast
cancers, as well as several other cancer types, such as
with doxorubicin (75 mg/m2) in metastatic breast cancer
uterine serous, gastric, and gall bladder carcinoma.20,21 Its
had shown superior efficacy and tolerability with doxoru-
cytotoxic partner DM1 (emtansine) is a maytansine de-
bicin compared with the ADC.10 Such initial studies have led
rivative and antimicrotubule agent. DM1 is linked to tras-
to technologic advancements and the development of
tuzumab via a stable covalently conjugated thioether
current-generation ADCs with improved therapeutic indices
linker.22 T-DM1 is approved by the FDA for metastatic
and more manageable safety profiles; however the optimal
HER2-amplified breast cancer in the second line after
design for an ADC to maximally harness efficacy for target
progression on HER2-directed therapy, as well as in the
cell destruction, while maintaining a favorable pharmaco-
adjuvant setting in patients with breast cancer who do not
kinetic and toxicity profile, remains an ongoing challenge in
achieve a pathologic CR (pCR) following neoadjuvant
the field.
HER2-directed therapy, as per the EMELIA and KATHER-
More than 100 ADCs have been evaluated in the investi- INE trials, respectively.15,23 It has been well established
gational setting; however, only a handful have been ap- through in vitro and in vivo studies that T-DM1 efficacy
proved by the U.S. Food and Drug Administration (FDA) correlates with levels of HER2 expression. For instance, in
for clinical use.5 This includes gemtuzumab ozogamicin, uterine sarcoma cell lines expressing high levels of HER2,
an ADC targeting CD33 linked to a calicheamicin agent, the median cytotoxicity of T-DM1 was 61% 6 5.3% ver-
which was the first ADC approved by the FDA in 2001 sus 3.02% 6 0.98% in low expressors.24 In patients with
via accelerated review for acute myelogenous leukemia.11 metastatic breast cancer, median disease-free progression
Concerns about veno-occlusive disease and the inability has been shown to correlate with tissue messenger RNA
to meet efficacy cutoffs in postmarketing trials resulted in expression of HER2 greater than or equal to the mean (HR,
its withdrawal in 2010.12 It was reapproved in 2017 after 0.39; 95% CI, 0.18–0.85) compared with expression below
further analyses showed benefit.13 Other FDA-approved the mean (HR, 0.85; 95% CI, 0.44–1.67).25 This was also
ADCs include brentuximab vedotin (BV) for refractory noted in the phase II study of T-DM1 in metastatic breast
Hodgkin lymphoma (HL) and anaplastic large cell lym- cancer with objective response rates (ORRs) of 36% versus
phoma (ALCL; 2011)14; trastuzumab emtansine (T-DM1) for 28% in patients with levels of HER2 greater than or equal to
metastatic HER2-amplified breast cancer (2013)15; inotu- the mean versus below the mean, as assessed using
zumab ozogamicin for relapsed/refractory B-cell acute reverse-transcriptase polymerase chain reaction. Of 112
lymphoblastic leukemia (2017)16; polatuzumab vedotin- treated patients in this study, the overall ORR was 29% via
piiq, an anti-CD79b–targeting ADC approved via accelera- central review and 33.8% in the 74 patients who had HER2
ted review for relapsed/refractory diffuse large B-cell lymphoma positivity with an immunohistochemistry (IHC) score of 3+ or

Antibody Drug Conjugates: Patient and Treatment Selection
FIGURE 1. Points of Consideration in Patient

Selection and Treatment
Abbreviation: ADCs, antibody-drug conjugates.
fluorescence in situ hybridization (FISH) compared with using tissue microarrays and a semiquantitative IHC scoring
4.8% in the 21 patients noted to be HER2 normal.26 system found that approximately 83% of bladder cancers,
Similarly, sacituzumab govitecan is another ADC that is 78% of breast cancers, and 71% of pancreatic cancers
currently in phase III testing for metastatic triple-negative stained overall positive for Nectin-4.31 Staining of 294
breast cancer targeting Trop-2, a glycoprotein that is normal human tissue samples from 36 organ sites had
upregulated in multiple carcinomas.27 The cytotoxic pay- shown a homogenous pattern of weak to moderate staining,
load is SN-38, a topoisomerase I inhibitor with a 100- to predominantly in skin, bladder, salivary gland ducts, breast,
1,000-fold increase in potency in comparison with its esophagus, and stomach. Enfortumab vedotin is an ADC
prodrug irinotecan, and it has poor tolerability when given as that was developed to target Nectin-4. Based on a positive
a standalone systemic therapy. Initial safety and efficacy phase II single-arm study of enfortumab vedotin in locally
evaluations included 69 patients.28 Trop-2 expression was advanced or metastatic urothelial carcinoma after pro-
not part of eligibility criteria; however, it had been assessed gression on prior platinum and anti–PD-1/PD-L1 therapy, it
in post hoc and exploratory analyses. Forty-eight patients was given accelerated approval by the FDA in 2019 for this
had Trop-2 expression evaluated in archived tissue; 88% indication, pending confirmatory phase III testing com-
had moderate (2+) or strong (3+) expression per a preparing the drug with single-agent cytotoxic chemotherapy.18
viously reported IHC scoring system.29 All responders were In the phase II study, enfortumab vedotin had shown
noted to have moderate to strong staining patterns. Those a confirmed ORR of 44% (95% CI, 35.1–53.2) at a median
with weakly positive Trop-2 tissue staining were noted to have follow-up of 10.2 months, with a 12% CR rate. Median
stable disease as best response.28 Sacituzumab govitecan has duration of response was 7.6 months (range, 0.95–11.30+).
been investigated further in a phase I/II trial in patients with Fifty-four percent had grade 3 or greater treatment-related
triple-negative breast cancer who have undergone at least two adverse events, with 12% discontinuing therapy for this
prior lines of therapy, showing an ORR of 33.3% (95% CI, reason. Given the expression of Nectin-4 in skin, rash was
24.6–43.1), although correlation with tissue expression was an anticipated on-target toxicity; it was observed and was
not reported.30 noted to be of low grade and managed medically. One case
of Stevens-Johnson syndrome had been noted. Nectin-4
Histology as Surrogate for Target Expression expression in pretreatment biopsies had been assessed in
Nectin-4 is a cell surface protein that is highly expressed in the study population and scored based on a semiquantitative
multiple cancer types. IHC analysis of 2,394 tissue specimens IHC score (H-score): 0 indicates no expression, and a score of

300 indicates maximum possible expression per assay. The investigators had postulated that targeted anti-HER2
Median H-score was 290 (range, 14–300).18 Correlation of therapy without chemotherapy may not be sufficient to eradicate
response with Nectin-4 expression has been suggested in disease in patients who exhibit substantial intratumor hetero-
patient-derived xenograft models of triple-negative breast geneity in HER2 expression. Patients with early-stage centrally
cancer using a different ADC targeting this protein.32 Bio- confirmed HER2-positive disease underwent six cycles of
marker expression was not used for patient selection in the T-DM1, in combination with pertuzumab in the neoadjuvant
phase II study protocol of enfortumab vedotin, and correlation setting. Intratumor heterogeneity of HER2 is defined based on
of response with Nectin-4 expression was not reported; both genetic and/or regional heterogeneity from two distinct sites
of these are important considerations to evaluate in future (three core biopsies per site). Genetically, the tumors are defined
studies. as heterogeneous if an area of tumor exhibits an HER2/CEP17
ratio 2.0 or greater or gene copy number greater than six by
Clinical Efficacy in Low Expressors of Target Antigen: Are
FISH in more than 5% but less than 50% of tumor cells. Re-
We Missing Patients Who Could Benefit?
gionally, they will be considered heterogenous if at least one of
ADCs have also shown clinical activity in patients with low six tumor areas is negative by FISH (with IHC , 3+). Thus,
levels of target antigen expression. For instance, BV is an a centrally confirmed HER2-positive tumor will be considered
ADC targeting CD30, and it is approved in classic HL and heterogenous if it meets either of the above criteria. pCR rate is
ALCL, both of which have uniform high expression of the primary objective of this study. Preliminary results had
CD30.33 It has also been tested in relapsed/refractory CD30- shown pCR in 49% of enrolled patients. No pCR was seen
positive non-HL, in which clinical activity has been noted at among the patients classified as having HER2 heterogeneity,
a wide range of CD30 expression. In a phase II study of the and exploratory studies have shown higher pCR rates among
drug in this patient population, an ORR of 44% was seen in those with HER2 scores of 3+ versus 2+ (56% vs. 27%; odds
diffuse large B-cell lymphoma, and CD30 surface expres- ratio, 3.4; p = .002), indicating that heterogeneity of target
sion was not noted to be predictive of response.34 A follow- antigen expression is an important factor in patient selection.39
up study of BV in diffuse large B-cell lymphoma, with
Bystander Killing Effect
undetectable CD30 expression based on IHC analysis,
resulted in ORRs in 31% of patients with 12% CRs (6 The bystander killing effect of certain ADCs has been shown
patients). More sensitive computer-assisted digital image to increase their payload sensitivity. The bystander effect
analysis had shown at least 1% CD30 expression via IHC in occurs when the cytotoxic payload diffuses across the
11 of 16 responders, indicating that a minimal threshold cell membrane to surrounding cells, increasing cytotoxicity
expression may be required for antitumor efficacy of BV in irrespective of target antigen expression levels.40 For in-
diffuse large B-cell lymphoma. Another explanation may be stance, DS-8201a, an anti-HER2 ADC coupled to a topo-
an indirect antitumor effect via the “bystander effect,” or isomerase I inhibitor, known as trastuzumab deruxtecan,
cytotoxic effect, in surrounding cells.35,36 BV consists of with high permeability has been shown in vitro and in vivo
a human chimeric immunoglobulin G1 antibody against to demonstrate a bystander effect in adjacent cells not
CD30, coupled via a cleavable linker to monomethyl auri- expressing HER2, whereas T-DM1 has comparatively low
statin E (MMAE), an antimicrotubule agent. MMAE has been permeability and does not have the same bystander effects
shown, in vitro and in vivo, to be able to cross the cell on adjacent cells not expressing HER2.41 Tumors with low
membrane into the surrounding extracellular matrix and HER2 expression (IHC score of 1+ or 2+ with negative in situ
exert an antitumor effect, which may account for its activity hybridization amplification) had shown ORR of 38.5% with
in tumors with heterogeneous or low target antigen DS-8201a in phase I studies.42 Similarly, its efficacy in in situ
expression.37 hybridization nonamplified colorectal cancer cell lines with
varying levels of HER2 protein expression has also been
Intratumor Heterogeneity and Its Impact on ADCs demonstrated.43 The most common side effects of this ADC
Understanding intratumor heterogeneity is important to the include nausea, vomiting, and decreased appetite. Rare
field of precision oncology. Like all cells, cancer cells are grade 5 interstitial lung disease has also been reported,
under strong selection and evolutionary pressures, partic- which may be an on-target effect, because it has also been
ularly in response to cancer therapies, resulting in the noted with T-DM1.44-46 SYD985 [(vic)-trastuzumab duo-
development of subclones and heterogeneity in gene ex- carmazine] is another anti-HER2 agent conjugated to a
pression within tumors.38 As an extension, this can result duocarmycin payload that was also engineered to exhibit
in heterogeneity of target antigen expression. The concept a potent bystander effect; therefore, it is efficacious for
of intratumor heterogeneity is undergoing evaluation in tumors expressing levels of HER2 (IHC 1+/2+ and in situ
a phase II study of patients with HER2-positive breast hybridization negative tumors).45 Use of the bystander effect
cancer as a predictor of response to neoadjuvant T-DM1 may be beneficial for solid tumors with heterogenous ex-
and pertuzumab (ClinicalTrials.gov identifier: NCT02326974).39 pression of target antigen expression, whereby the drug can

exert its effects on adjacent cells not expressing the antigen. generation of biparatopic monoclonal antibodies. Bipar-
For instance, MMAE results in a neutral metabolite that is atopic antibodies recognize the same target antigen but
known to cross cell membranes and exert cytotoxic effects bind two separate and nonoverlapping epitopes of the
on neighboring cells, whereas the use of monomethyl antigen. An anti-HER2 biparatotic ADC called MED14276
auristatin F results in a charged metabolite that does not has recently shown activity in preclinical models of meta-
cross cell membranes. Thus, engineering of ADCs with an static breast cancer with heterogeneity of HER2 expression
enhanced or reduced bystander effect based on hetero- and acquired resistance to T-DM1 therapy.54 MED14276
geneity or level of target antigen expression within the tumor utilizes components of trastuzumab and another human
would be one way in which to modulate therapeutic efficacy, monoclonal anti-HER2 antibody (39S), which bind discrete
regardless of patient selection.47,48 and nonoverlapping epitopes of HER2. This ADC contains
four high-affinity antigen binding sites in each arm, making
Host-Related Factors in Patient Selection it a tetravalent biparatopic monospecific antibody. Addition
Several patient or host-associated factors have also been of an L234F substitution to the hinge region of the antibody
studied in relation to efficacy of ADCs, although they are not reduces off-target toxicity and HER2-independent Fcγ
used for patient selection.49 T-DM1 is dosed based on receptor–mediated uptake of the antibody into normal cells.
weight, with standard dosing of 3.6 mg/m2. In an analysis of The cytotoxic payload is a microtubule polymerization in-
five phase I-III studies including 671 patients, those with hibitor (AZ13599185) attached via a maleimidocaproyl
lower body weight (5th percentile) were noted to have slower linker. Further characterization of the ADC revealed its abil-
clearance of T-DM1 compared with those with higher body ity to inhibit functional dimerization of HER2 via the 39S arm
weight (95th percentile).50 Similarly, those with higher body and improved clustering of receptors on the surface of target
weight had higher volumes of distribution in central com- cells, leading to improved internalization, trafficking, and
partment (Vc). Higher baseline serum HER2 shed extra- degradation.54 Bystander effect was also shown using co-
cellular domain concentration, higher baseline sum of culture cell-killing assay.55 Potency in cell lines with low
longest target lesion dimension, and lower serum albumin expression of HER2 and in triple-negative breast cancer
were also associated with higher drug clearance. How- preclinical models has been demonstrated.55
ever, this was not considered to have a clinically relevant
Bispecific antibodies are another means for increasing thera-
impact.50
peutic efficacy by generating antibodies that recognize
Data-driven pharmacokinetics/pharmacodynamics model- two antigens.56,57 They simultaneously combine the spec-
ing tools are being developed and validated to integrate ificities of two antibodies and can be used for more effective
biomeasures from preclinical data to predict response.51,52 targeting and modulation of multiple pathways to increase
One such model described for inotuzumab ozogamicin, an the potency and specificity to target cancer cells. Some
ADC targeting CD22, used and modeled data from multiple examples of bispecific antibodies include blinatumomab,
preclinical factors, including a characterization of clearance which targets CD3 and CD19 and is approved for use in
of inotuzumab ozogamicin and its payload, binding, in- relapsed/refractory Philadelphia chromosome–negative acute
ternalization, release, efflux from the cell, diffusion of ino- lymphoblastic leukemia,58 and emicizumab, which has
tuzumab ozogamicin into the tumor microenvironment, and shown efficacy in patients with hemophilia A, because it
modeling of tumor growth and inhibition from xenograft recognizes factor IXa and X, enabling colocalization of the
models in mice. Once the relationship between pharma- enzyme-substrate complex that is typically accomplished by
cokinetics/pharmacodynamics and inhibition of tumor growth factor VIII, which is deficient in hemophilia A.59 Combining
as a function of the payload had been characterized, the model bispecific antibody technology with ADCs has increased
was used to predict progression free-survival. The investigators interest in the field. Using a bispecific antibody against
note that pharmacokinetics and efflux of the calicheamicin EGFR and c-MET, coupled with an MMAE cytotoxic pay-
payload are better predictors of outcome than is expression of load, showed increased in vitro sensitivity for cancer cells
CD22 receptor.52 Development of such predictive models is overexpressing EGFR and/or c-MET in comparison with
ongoing; however, they require additional verification and normal tissues.60 Potency of inhibition in cell lines was in-
validation prior to their introduction into the clinical arena for dependent of cell surface expression of EGFR, although
patient selection. a lower affinity for EGFR resulted in less toxicity on normal
keratinocytes.
CONSIDERATIONS FOR PAYLOAD SENSITIVITY
Modulation of Novel Antibodies to Improve Delivery Drug/Antibody Ratio
In patients who express low levels of target antigen, several Other means for improving payload sensitivity include the
means to increase efficacy of ADCs and improve payload DAR, or the number of payload molecules attached to each
sensitivity are under development.53 One method is the antibody.61 The typical range for DARs is 0 to 8. T-DM1 has

an average DAR of 3.5.62 Although ADCs with higher DARs Combination Therapy Strategies
have shown increased efficacy in vitro, a DAR greater than 4 Other novel strategies for increasing payload sensitivity and
has been associated with increased toxicity and poor ag- antitumor efficacy include combination therapies with
gregation and pharmacokinetics in vivo. Previous studies of ADCs. ADCs are being included in combination studies
anti-CD30 binding ADCs, in which drug/payload to mono- with chemotherapy. For instance, in a large, phase III,
clonal antibody molar ratios of 2, 4, and 8 were tested open-label, multicenter study that included 1,334 patients
in vitro, demonstrated decreased half-maximal inhibitory who were randomly assigned to treatment, BV in combi-
concentration requirements with increasing molar ratio. nation with doxorubicin, vinblastine, and dacarbazine was
However, in vivo testing in xenograft models of ALCL had shown to have an incremental absolute benefit of 4.9%
shown equal antitumor efficacy for the 4-M and 8-M DARs (82.1%; 95% CI, 78.8–85.0 vs. 77.2%; 95% CI, 73.7–80.4)
at different dose levels.62 Clearance and volume of distri- over chemotherapy (doxorubicin, bleomycin, vinblastine, and
bution increased with increasing DAR as well, through dacarbazine) for 2-year progression-free survival in stage III or
a mechanism that has not been well characterized. Opti- IV refractory HL.72
mization of DAR is an important consideration in therapeutic
efficacy.61,63 Improved efficacy of ADCs has also been shown in vivo by
combining ADCs with targeted therapy. In melanoma,
increased expression of the receptor tyrosine kinase gene
Overcoming Resistance Mechanisms
AXL has shown resistance to MAPK inhibitors.73,74 AXL-
As with most systemic anticancer therapies, selection 107-MMAE, an anti-AXL antibody conjugated to auristatin
pressures have resulted in the development of resistance E, an antimicrotubule agent, has been shown to have
to ADCs via multiple mechanisms.64,65 One mechanism is antitumor efficacy in patient-derived xenograft models
downregulation of target antigen, as seen with the use of an from multiple cancer types. Combination therapy with
anti-HER2 ADC similar to T-DM1.66 Similarly, increased inhibitors of the MAPK pathway, have shown enhanced
shedding of antigen into the circulation or increased antigen inhibition of tumor growth. Further, BRAF/MEK inhibitors
load reduces ADC effectiveness by consuming the ADC in have been shown to increase AXL expression and, thereby,
circulation, as seen with gemtuzumab ozogamicin.67 ADCs, increase efficacy of the anti-AXL ADC in a melanoma
like many compounds, are also substrates for adenosine patient–derived xenograft model.75 Integrative molecular
triphosphate–binding cassette transporters and multidrug analysis of resistance models of T-DM1 in HER2-amplified
transporters, such as MDR1, which cause efflux of drugs out breast cancer has identified the mitotic kinase Polo-like
of the cell. MDR1 activity and expression have been as- kinase 1 as a mediator of T-DM1 inhibition. Subsequent
sociated with patient outcomes in hematologic malignan- inhibition of Polo-like kinase 1 utilizing a selective inhibitor,
cies treated with gemtuzumab ozogamicin and BV.68 BV volasertib, in combination with T-DM1, has shown syner-
uses MMAE as its cytotoxic payload. Interestingly, on an- gistic growth inhibition in preclinical models of acquired and
alyzing acquired resistance to BV in cell lines from ALCL de novo resistance to T-DM1.76
and HL, the two cell lines were noted to exhibit different
Combination of ADCs with anti-inflammatory agents, such
mechanisms of resistance. Although downregulation of
as tofacitinib, a JAK/STAT inhibitor, has been shown to
target antigen (CD30) expression was noted in the ALCL
increase efficacy of anetumab ravtansine, an antimesothelin
resistant cell line compared with the parental and HL cell
antibody conjugated with a maytansinoid tubulin inhibitor
lines, increased expression of MDR1 was noted in the HL
DM4 in vitro and in vivo in human pancreatic cancer tumor
cell line. The former, although shown in vitro, has not been
models. It is thought to result from enhanced delivery of the
recapitulated in vivo; however, the latter has been noted
ADC to tumor cells via a reduction in inflammatory cytokines
in vitro and in patients with HL.69 Methods to bypass such
and cells in the tumor microenvironment.77
drug resistance are in development. Inhibition of MDR1,
with the addition of cyclosporine A to BV, was shown to The recent development of anti–PD-L1 ADCs, as well as
restore sensitivity to BV in HL.70 Resistance via MDR1 has immunotherapy/ADC combinations, has also been re-
also been noted for the cytotoxic component (DM1) of ported.78 Combining the single-chain variable fragment,
T-DM1, and conjugation of DM1 via a maleimidyl-based which harbors anti–PD-L1 activity, with the DM1 cytotoxic
hydrophilic linker PEG(4)Mal was shown to have improved payload decreased proliferation of PD-L1–positive cells
retention of the drug in cells expressing MDR1 com- in vitro.79 Another reported PD-L1 ADC uses a doxorubicin
pared with nonpolar linker molecules.71 In human xenograft payload linked via a hydrazone linker, enabling dissocia-
models, conjugates with the hydrophilic linker were noted to tion in the tumor microenvironment. Interestingly, this ADC
have improved therapeutic efficacy and antitumor effect in has been developed with the intent of utilizing the doxo-
MDR1-expressing tumors compared with a nonpolar linker rubicin payload to disrupt the tumor microenvironment so
counterpart.71 that the anti–PD-L1 molecule can better penetrate into the

center of the tumor and exert its immunotherapeutic ef- progression and metastasis. The effects of different in-
fects. An increase in T-cell activation and interferon-γ has tervening therapies on expression are also largely unknown.
been noted with the ADC construct.80 The common cytotoxic Target assessment has also been challenging as a result
payloads used in ADCs, such as MMAE and ansamitocin, have of known challenges with IHC assessment/interpretation,
been shown to stimulate migration of CD8+ effector cells into as well as challenges in determining the minimal ex-
the tumor microenvironment. Additionally, increased priming pression required and expression for optimal activity.
of T cells by resident dendritic cells has been noted upon Further, target expression assessments have differed by
exposure to antimicrotubule ADCs.78,79 Following treatment target, different groups assessing intensity, percentage
with T-DM1, tumor-infiltrating lymphocytes were noted to expression, or a combination; there are also variations
be substantially increased in comparison with pretreatment between assessment by localization (total expression vs.
tissues in HER2-amplified/estrogen receptor–positive breast cell surface expression). There remains controversy as to
cancer samples (p = .005). CTLA-4 expression was also noted whether each target expression cutoff should be validated
to be increased in CD4+ and CD8+ cells from posttreatment in each histology to be tested.
samples. Similarly, tumor-associated macrophages had sub-
stantially increased levels of PD-L1 expression. In in vivo Patient selection in ADC trials has been challenging. Many
models, CTLA-1 and PD-1 blockade alone did not result in trials enroll patients in an unselected fashion or enrich for
antitumor efficacy; however, strong antitumor efficacy was tumor types that are thought to express or overexpress the
noted when either of the immune checkpoint inhibitors was target. Selected trials screen for ADC targets, but this is
combined with T-DM1, without adverse toxicity to the mice.81 made more challenging when screening in histologies with
less frequent expression is initiated. There remains a need
OTHER PATIENT CONSIDERATIONS for multiplex testing for multiple targets, either through
Understanding and managing the toxicities of ADCs are multiplex proteomics strategies as these evolve or through
essential to the continued development of these agents and transcriptional profiling, because many ADC targets are
their successful use in patients who have cancer.82 The also overexpressed at the messenger RNA level. We ex-
toxicities of ADCs are related primarily to release of the pect that we will be able to start screening across a variety
cytotoxic payload at off-target sites and unwanted bystander of ADC options to better personalize therapy selection as
effect. The most common classes of cytotoxic agents used integrated multianalyte testing of patients with cancer be-
in ADCs are topoisomerase II inhibitors (anthracyclines), comes more routine.
alkylating agents (calicheamicin), and microtubule-disrupting
CONCLUSIONS
agents (maytansinoids, such as emtansine).4 Ocular tox-
icities are a common off-target effect of antimicrotubule ADCs are a rapidly evolving class of drug in the anticancer
inhibitors (DM4 or monomethyl auristatin F) seen across armamentarium with the potential to harness the cytotoxic
a range of target antigens, resulting in corneal abnormali- nature of antineoplastic agents with activity engineered to
ties, blurred vision, and dry eyes. These adverse events occur at target sites. The optimal ADC design would involve
are typically reversible upon discontinuation of the ADC and careful selection of each of its critical elements and its target
are managed medically. Other cytotoxic agents, such as within the context of the constantly evolving complexities of
MMAEs, can result in myelosuppression and neuropathy. the tumor microenvironment and heterogeneity of target
Reduced differentiation and increased destruction of expression patterns. Intuitively, an ideal target would be one
megakaryocytic precursors have been seen with these with preferential expression or substantial upregulation in
agents, resulting in thrombocytopenia.83,84 More serious tumor tissue in comparison with normal tissues to minimize
organ dysfunction, such as liver dysfunction and veno- off-target toxicities. However, data reviewed here suggest
occlusive disease, have been reported with gemtuzumab that there is clinical efficacy in patients who also have low
ozogamicin and inotuzumab ozogamicin, and pneumonitis target antigen expression, shifting toward a threshold ex-
and pulmonary toxicity have been observed with sacitu- pression that is still unknown for many target antigens over
zumab govitecan. An overview of ADC safety is summarized which ADCs have therapeutic efficacy. Therapeutic efficacy
in a recent meta-analysis of 70 publications.85 and payload sensitivity are also dependent on the mech-
anism of action of the cytotoxic payload, its permeability
REMAINING PATIENT- AND TREATMENT-SELECTION across membranes, and bystander effect on adjacent cells.
CHALLENGES AND FUTURE DIRECTIONS As we learn more about the vast intratumor heterogeneity of
Although there are a growing number of ADC technologies cancer, additional and more sensitive means for patient
in development, several challenges remain. For many tar- selection are needed to choose those who would benefit
gets, there is limited information about their expression optimally from ADC therapy. Additionally, identification of
across a variety of tumor types, as well as intratumoral mechanisms of resistance to ADCs will result in novel
heterogeneity and evolution of expression with tumor methods to circumvent them such as modulation of the

intrinsic structure of antibody molecules and other ADC cytotoxic or immunomodulatory therapies to further en-
components, and their use in combination with other hance the antitumor effect.
AFFILIATIONS Cancer Center, 1400 Holcombe Blvd., Unit 455, Houston, TX 77030;
1
Division of Cancer Medicine, University of Texas MD Anderson Cancer email: [email protected].
Center, Houston, TX
2
Department of Investigational Cancer Therapeutics, University of Texas
MD Anderson Cancer Center, Houston, TX
Funda Meric-Bernstam, MD, Division of Cancer Medicine, Department of
Investigational Cancer Therapeutics, University of Texas MD Anderson
REFERENCES
1. Thomas A, Teicher BA, Hassan R. Antibody-drug conjugates for cancer therapy. Lancet Oncol. 2016;17:e254-e262.
2. Lambert JM, Morris CQ. Antibody-drug conjugates (ADCs) for personalized treatment of solid tumors: a review. Adv Ther. 2017;34:1015-1035.
3. Lambert JM, Berkenblit A. Antibody-drug conjugates for cancer treatment. Annu Rev Med. 2018;69:191-207.
4. Nagayama A, Ellisen LW, Chabner B, et al. Antibody-drug conjugates for the treatment of solid tumors: clinical experience and latest developments. Target Oncol.
2017;12:719-739.
5. Chau CH, Steeg PS, Figg WD. Antibody-drug conjugates for cancer. Lancet. 2019;394:793-804.
6. Strebhardt K, Ullrich A. Paul Ehrlich’s magic bullet concept: 100 years of progress. Nat Rev Cancer. 2008;8:473-480.
7. Tolcher AW. Antibody drug conjugates: lessons from 20 years of clinical experience. Ann Oncol. 2016;27:2168-2172.
8. Trail PA, Willner D, Lasch SJ, et al. Cure of xenografted human carcinomas by BR96-doxorubicin immunoconjugates. Science. 1993;261:212-215.
9. Saleh MN, Sugarman S, Murray J, et al. Phase I trial of the anti-Lewis Y drug immunoconjugate BR96-doxorubicin in patients with Lewis Y-expression epithelial
tumors. J Clin Oncol. 2000;18:2282-2292.
10. Tolcher W, Sugarman S, Gelmon KA, et al. Randomized phase II study of BR96-doxorubicin conjugate in patients with metastatic breast cancer. J Clin Oncol.
1999;17:478-484.
11. Bross PF, Beitz J, Chen G, et al. Approval summary: gemtuzumab ozogamicin in relapsed acute myeloid leukemia. Clin Cancer Res. 2001;7:1490-1496.
12. Petersdorf SH, Kopecky KJ, Slovak M, et al. A phase 3 study of gemtuzumab ozogamicin during induction and postconsolidation therapy in younger patients with
acute myeloid leukemia. Blood. 2013;121:4854-4860.
13. Norsworthy KJ, Ko CW, Lee JE, et al. FDA approval summary: mylotarg for treatment of patients with relapsed or refractory CD33-positive acute myeloid leukemia.
Oncologist. 2018;23:1103-1108.
14. de Claro RA, McGinn K, Kwitkowski V, et al. U.S. Food and Drug Administration approval summary: brentuximab vedotin for the treatment of relapsed Hodgkin
lymphoma or relapsed systemic anaplastic large-cell lymphoma. Clin Cancer Res. 2012;18:5845-5849.
15. Verma S, Miles D, Gianni L, et al; EMILIA Study Group. Trastuzumab emtansine for HER2-positive advanced breast cancer. N Engl J Med. 2012;367:1783-1791.
16. Kantarjian HM, DeAngelo DJ, Stelljes M, et al. Inotuzumab ozogamicin versus standard therapy for acute lymphoblastic leukemia. N Engl J Med. 2016;
375:740-753.
17. Deeks ED. Polatuzumab vedotin: first global approval. Drugs. 2019;79:1467-1475.
18. Rosenberg JE, O’Donnell PH, Balar AV, et al. Pivotal trial of enfortumab vedotin in urothelial carcinoma after platinum and anti-programmed death 1/programmed
death ligand 1 therapy. J Clin Oncol. 2019;37:2592-2600.
19. Modi S, Saura C, Yamashita T, et al. Trastuzumab Deruxtecan in Previously Treated HER2-Positive Breast Cancer. N Engl J Med. 2020;382:610-621.
20. Slamon DJ, Clark GM, Wong SG, et al. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science. 1987;
235:177-182.
21. Oh DY, Bang YJ. HER2-targeted therapies - a role beyond breast cancer. Nat Rev Clin Oncol. 2020;17:33-48.
22. LoRusso PM, Weiss D, Guardino E, et al. Trastuzumab emtansine: a unique antibody-drug conjugate in development for human epidermal growth factor receptor
2-positive cancer. Clin Cancer Res. 2011;17:6437-6447.
2019;380:617-628.

24. English DP, Bellone S, Schwab CL, et al. T-DM1, a novel antibody-drug conjugate, is highly effective against primary HER2 overexpressing uterine serous
carcinoma in vitro and in vivo. Cancer Med. 2014;3:1256-1265.
25. Perez EA, Hurvitz SA, Amler LC, et al. Relationship between HER2 expression and efficacy with first-line trastuzumab emtansine compared with trastuzumab plus
docetaxel in TDM4450g: a randomized phase II study of patients with previously untreated HER2-positive metastatic breast cancer. Breast Cancer Res. 2014;
16:R50.
26. Burris HA III, Rugo HS, Vukelja SJ, et al. Phase II study of the antibody drug conjugate trastuzumab-DM1 for the treatment of human epidermal growth factor
receptor 2 (HER2)-positive breast cancer after prior HER2-directed therapy. J Clin Oncol. 2011;29:398-405.
27. Sahota S, Vahdat LT. Sacituzumab govitecan: an antibody-drug conjugate. Expert Opin Biol Ther. 2017;17:1027-1031.
28. Bardia A, Mayer IA, Diamond JR, et al. Efficacy and safety of anti-Trop-2 antibody drug conjugate sacituzumab govitecan (IMMU-132) in heavily pretreated
patients with metastatic triple-negative breast cancer. J Clin Oncol. 2017;35:2141-2148.
29. Starodub AN, Ocean AJ, Shah MA, et al. First in-human trial of a novel anti-Trop-2 antibody-SN-38 conjugate, sacituzumab govitecan, for the treatment of diverse
metastatic solid tumors. Clin Cancer Res. 2015;21:3870-3878.
30. Bardia A, Mayer IA, Vahdat LT, et al. Sacituzumab govitecan-hziy in refractory metastatic triple-negative breast cancer. N Engl J Med. 2019;380:741-751.
31. Challita-Eid PM, Satpayev D, Yang P, et al. Enfortumab vedotin antibody-drug conjugate targeting nectin-4 is a highly potent therapeutic agent in multiple
preclinical cancer models. Cancer Res. 2016;76:3003-3013.
32. M-Rabet M, Cabaud O, Josselin E, et al. Nectin-4: a new prognostic biomarker for efficient therapeutic targeting of primary and metastatic triple-negative breast
cancer. Ann Oncol. 2017;28:769-776.
33. van der Weyden CA, Pileri SA, Feldman AL, et al. Understanding CD30 biology and therapeutic targeting: a historical perspective providing insight into future
directions. Blood Cancer J. 2017;7:e603.
34. Jacobsen ED, Sharman JP, Oki Y, et al. Brentuximab vedotin demonstrates objective responses in a phase 2 study of relapsed/refractory DLBCL with variable
CD30 expression. Blood. 2015;125:1394-1402.
35. Bartlett NL, Smith MR, Siddiqi T, et al. Brentuximab vedotin activity in diffuse large B-cell lymphoma with CD30 undetectable by visual assessment of
conventional immunohistochemistry. Leuk Lymphoma. 2017;58:1607-1616.
36. Breij EC, de Goeij BE, Verploegen S, et al. An antibody-drug conjugate that targets tissue factor exhibits potent therapeutic activity against a broad range of solid
tumors. Cancer Res. 2014;74:1214-1226.
37. Scott LJ. Brentuximab vedotin: a review in CD30-positive Hodgkin lymphoma. Drugs. 2017;77:435-445.
38. McGranahan N, Swanton C. Biological and therapeutic impact of intratumor heterogeneity in cancer evolution. Cancer Cell. 2015;27:15-26.
39. Filho OM, Viale G, Trippa L, et al. HER2 heterogeneity as a predictor of response to neoadjuvant T-DM1 plus pertuzumab: results from a prospective clinical trial. J
Clin Oncol. 2019;37 (suppl; abstr 502).
40. Parslow AC, Parakh S, Lee FT, et al. Antibody-drug conjugates for cancer therapy. Biomedicines. 2016;4:E14.
41. Pegram MD, Miles D, Tsui CK, et al. HER2-overexpressing/amplified breast cancer as a testing ground for antibody-drug conjugate drug development in solid
tumors. Clin Cancer Res. 2020;26:775-786.
42. Modi S, Tsurutani J, Tamura K, et al. Trastuzumab deruxtecan (DS-8201a) in subjects with HER2-low expressing breast cancer: updated results of a large phase
1 study. Presented at: San Antonio Breast Cancer Symposium; San Antonio, TX; 2018. Abstract P6-17-02.
43. Takegawa N, Tsurutani J, Kawakami H, et al. [fam-] trastuzumab deruxtecan, antitumor activity is dependent on HER2 expression level rather than on HER2
amplification. Int J Cancer. 2019;145:3414-3424.
44. Iwata H, Tamura K, Doi T, et al., Trastuzumab deruxtecan (DS-8201a) in subjects with HER2-expressing solid tumors: longterm results of a large phase 1 study
with multiple expansion cohorts. J Clin Oncol. 2018;36 (suppl; abstr 2501).
45. van der Lee MM, Groothuis PG, Ubink R, et al. The preclinical profile of the duocarmycin-based HER2- targeting ADC SYD985 predicts for clinical benefit in low
HER2-expressing breast cancers. Mol Cancer Ther. 2015;14:692-703.
46. Saura C, Thistelthwaite F, Banerji U, et al. A phase I expansion cohorts study of SYD985 in heavily pretreated patients with HER2-positive or HER2-low metastatic
breast cancer. J Clin Oncol. 2018;36 (suppl; abstr 1014).
47. Xu Z, Guo D, Jiang Z, et al. Novel HER2-targeting antibody-drug conjugates of trastuzumab beyond T-DM1 in breast cancer: trastuzumab deruxtecan (DS-8201a)
and (vic-)trastuzumab duocarmazine (SYD985). Eur J Med Chem. 2019;183:111682.
48. Beck A, Goetsch L, Dumontet C, et al. Strategies and challenges for the next generation of antibody-drug conjugates. Nat Rev Drug Discov. 2017;16:315-337.
49. Lucas AT, Robinson R, Schorzman AN, et al. Pharmacologic considerations in the disposition of antibodies and antibody-drug conjugates in preclinical models
and in patients. Antibodies (Basel). 2019;8:E3.
50. Lu D, Girish S, Gao Y, et al. Population pharmacokinetics of trastuzumab emtansine (T-DM1), a HER2-targeted antibody-drug conjugate, in patients with HER2-
positive metastatic breast cancer: clinical implications of the effect of covariates. Cancer Chemother Pharmacol. 2014;74:399-410.
51. Shah DK, Haddish-Berhane N, Betts A. Bench to bedside translation of antibody drug conjugates using a multiscale mechanistic PK/PD model: a case study with
brentuximab-vedotin. J Pharmacokinet Pharmacodyn. 2012;39:643-659.
52. Betts AM, Haddish-Berhane N, Tolsma J, et al. Preclinical to clinical translation of antibody-drug conjugates using PK/PD modeling: a retrospective analysis of
inotuzumab ozogamicin. AAPS J. 2016;18:1101-1116.

53. Carter PJ, Lazar GA. Next generation antibody drugs: pursuit of the ‘high-hanging fruit’. Nat Rev Drug Discov. 2018;17:197-223.
54. Oganesyan V, Peng L, Bee JS, et al. Structural insights into the mechanism of action of a biparatopic anti-HER2 antibody. J Biol Chem. 2018;293:8439-8448.
55. Li JY, Perry SR, Muniz-Medina V, et al. A biparatopic HER2-targeting antibody-drug conjugate induces tumor regression in primary models refractory to or
ineligible for HER2-targeted therapy. Cancer Cell. 2016;29:117-129.
56. Maruani A. Bispecifics and antibody-drug conjugates: a positive synergy. Drug Discov Today Technol. 2018;30:55-61.
57. Amani N, Dorkoosh FA, Mobedi H. ADCs, as novel revolutionary weapons for providing a step forward in targeted therapy of malignancies. Curr Drug Deliv. 2020;17:23-51.
58. Kantarjian H, Jabbour E, Topp MS. Blinatumomab for acute lymphoblastic leukemia. N Engl J Med. 2017;376:e49.
59. Lenting PJ, Denis CV, Christophe OD. Emicizumab, a bispecific antibody recognizing coagulation factors IX and X: how does it actually compare to factor VIII?
Blood. 2017;130:2463-2468.
60. Sellmann C, Doerner A, Knuehl C, et al. Balancing selectivity and efficacy of bispecific epidermal growth factor receptor (EGFR) c-MET antibodies and
antibody-drug conjugates. J Biol Chem. 2016;291:25106-25119.
61. Dan N, Setua S, Kashyap VK, et al. Antibody-drug conjugates for cancer therapy: chemistry to clinical implications. Pharmaceuticals (Basel). 2018;11:E32.
62. Krop I, Winer EP. Trastuzumab emtansine: a novel antibody-drug conjugate for HER2-positive breast cancer. Clin Cancer Res. 2014;20:15-20.
63. Hamblett KJ, Senter PD, Chace DF, et al. Effects of drug loading on the antitumor activity of a monoclonal antibody drug conjugate. Clin Cancer Res. 2004;
10:7063-7070.
64. Garcı́a-Alonso S, Ocaña A, Pandiella A. Resistance to antibody-drug conjugates. Cancer Res. 2018;78:2159-2165.
65. Loganzo F, Sung M, Gerber HP. Mechanisms of resistance to antibody-drug conjugates. Mol Cancer Ther. 2016;15:2825-2834.
66. Li G, Guo J, Shen BQ, et al. Mechanisms of acquired resistance to trastuzumab emtansine in breast cancer cells. Mol Cancer Ther. 2018;17:1441-1453.
67. van der Velden VH, Boeckx N, Jedema I, et al. High CD33-antigen loads in peripheral blood limit the efficacy of gemtuzumab ozogamicin (Mylotarg) treatment in
acute myeloid leukemia patients. Leukemia. 2004;18:983-988.
68. Matsumoto T, Jimi S, Hara S, et al. Importance of inducible multidrug resistance 1 expression in HL-60 cells resistant to gemtuzumab ozogamicin. Leuk
Lymphoma. 2012;53:1399-1405.
69. Chen R, Hou J, Newman E, et al. CD30 downregulation, MMAE resistance, and MDR1 upregulation are all associated with resistance to brentuximab vedotin. Mol
Cancer Ther. 2015;14:1376-1384.
70. Chen R, Herrera AF, Hou J, et al. Inhibition of MDR1 overcomes resistance to brentuximab vedotin in Hodgkin lymphoma. Clin Cancer Res. 2020;26:1034-1044.
71. Kovtun YV, Audette CA, Mayo MF, et al. Antibody-maytansinoid conjugates designed to bypass multidrug resistance. Cancer Res. 2010;70:2528-2537.
72. Connors JM, Jurczak W, Straus DJ, et al; ECHELON-1 Study Group. Brentuximab vedotin with chemotherapy for stage III or IV Hodgkin’s lymphoma. N Engl
J Med. 2018;378:331-344.
73. Müller J, Krijgsman O, Tsoi J, et al. Low MITF/AXL ratio predicts early resistance to multiple targeted drugs in melanoma. Nat Commun. 2014;5:5712.
74. Konieczkowski DJ, Johannessen CM, Abudayyeh O, et al. A melanoma cell state distinction influences sensitivity to MAPK pathway inhibitors. Cancer Discov.
2014;4:816-827.
75. Boshuizen J, Koopman LA, Krijgsman O, et al. Cooperative targeting of melanoma heterogeneity with an AXL antibody-drug conjugate and BRAF/MEK inhibitors.
Nat Med. 2018;24:203-212.
76. Saatci Ö, Borgoni S, Akbulut Ö, et al. Targeting PLK1 overcomes T-DM1 resistance via CDK1-dependent phosphorylation and inactivation of Bcl-2/xL in HER2-
positive breast cancer. Oncogene. 2018;37:2251-2269.
77. Simon N, Antignani A, Hewitt SM, et al. Tofacitinib enhances delivery of antibody-based therapeutics to tumor cells through modulation of inflammatory cells. JCI
Insight. 2019;4:123281.
78. Gerber HP, Sapra P, Loganzo F, et al. Combining antibody-drug conjugates and immune-mediated cancer therapy: What to expect? Biochem Pharmacol. 2016;102:1-6.
79. Kalim M, Wang S, Liang K, et al. Engineered scPDL1-DM1 drug conjugate with improved in vitro analysis to target PD-L1 positive cancer cells and intracellular
trafficking studies in cancer therapy. Genet Mol Biol. 2020;42:e20180391.
80. Sau S, Petrovici A, Alsaab HO, et al. PDL-1 antibody drug conjugate for selective chemo-guided immune modulation of cancer. Cancers (Basel). 2019;11:E232.
81. Müller P, Kreuzaler M, Khan T, et al. Trastuzumab emtansine (T-DM1) renders HER2+ breast cancer highly susceptible to CTLA-4/PD-1 blockade. Sci Transl
Med. 2015;7:315ra188.
82. Wolska-Washer A, Robak T. Safety and tolerability of antibody-dug conjugates in cancer. Drug Saf. 2019;42:295-314.
83. Hinrichs MJ, Dixit R. Antibody drug conjugates: nonclinical safety considerations. AAPS J. 2015;17:1055-1064.
84. Donaghy H. Effects of antibody, drug and linker on the preclinical and clinical toxicities of antibody-drug conjugates. MAbs. 2016;8:659-671.
85. Masters JC, Nickens DJ, Xuan D, et al. Clinical toxicity of antibody drug conjugates: a meta-analysis of payloads. Invest New Drugs. 2018;36:121-135.

Selective CDK4/6 Inhibitors: Biologic Outcomes,

Determinants of Sensitivity, Mechanisms of
Resistance, Combinatorial Approaches, and
Pharmacodynamic Biomarkers
Erik S. Knudsen, PhD1; Geoffrey I. Shapiro, MD, PhD2; and Khandan Keyomarsi, PhD3
overview
CDK4/6 inhibitors are now part of the standard armamentarium for hormone receptor–positive breast cancer.
In this article, we review the biologic outcomes imposed by these drugs on cancer cells, determinants of
response, mechanisms of intrinsic and acquired resistance, as well as combinatorial approaches emanating
from mechanistic studies that may allow use of these agents to extend beyond breast cancer. In addition, we
will address tumor-, imaging-, and blood-based pharmacodynamic biomarkers that can inform rationally
designed trials as clinical development continues.
CELL CYCLE CONTROL, THE RETINOBLASTOMA phosphorylation. In the absence of phosphorylation,

PATHWAY, AND THE CANONICAL VIEW OF CDK4/ RB, in an activated state, elicits potent transcriptional
6 INHIBITION repression over a gene expression program required for
The mitotic cell cycle underlies the division of all somatic subsequent cell cycle progression.9 This list, which is
cells in the body. Because uncontrolled cell division is ever evolving, includes genes required for DNA syn-
a hallmark of cancer, it has been posited that targeting the thesis, mitosis, and cytokinesis.5 In essence, the RB
cell cycle could represent a universal mechanism for activation state is a bottleneck for the cell cycle; when
cancer therapy.1,2 The signaling pathways that control the RB is activated and unphosphorylated, cells cannot
cell cycle are complex, but most mitogenic and oncogenic progress through the cell cycle because of the absence
signals act by driving cells through the G0/G1 to S-phase of required proteins.
transition (Fig. 1). The key machinery of the cell cycle If such a model is accurate, it follows that pharma-
includes the cyclin-dependent kinases (CDKs) and the cologically inhibiting CDK4/6 could have potent ef-
associated cyclins required for their catalytic activity.3 Cell fects for blocking tumor cell division.3,4 Although the
cycle traversal during mid-G1 is governed by CDK4 or
class of CDK4/6 inhibitors is expanding, palbociclib,
CDK6 kinase complexes that are highly responsive to
ribociclib, and abemaciclib are the only currently
a host of cancer-relevant perturbations.4,5 For exam-
clinically approved agents, all of which are highly
ple, multiple oncogenic factors hijack normal regulation of
selective for CDK4 and CDK6 relative to other CDK
cyclin D1, which promotes CDK4/6 activity.6 In addition,
complexes.3,4 Classically, each of these agents arrests
genetic permutations directly target CDK4/6 activity,
tumor cells in a G1-like state. Concordantly, they
including amplification of genes that encode CDK4, CDK6,
generally elicit a cytostatic, as opposed to cytotoxic,
or cyclin D1, which have been observed in more than 15%
mechanism of action in solid tumor cell lines, animal
of all tumors, as well as deletion, mutation, or methylation of
Author affiliations models, and in patient samples.10-17 Ample pre-
and support CDKN2A, which encodes the endogenous CDK4/6 inhibitor
clinical data indicate that this activity in restraining
information (if p16INK4A.4 Thus, much of the oncogenic activity observed in
applicable) appear
cell division is mediated by the activation of RB.13,18,19
the cell cycle is targeted at net CDK4/6 activity. This is also
at the end of this This model suggests that CDK4/6 inhibitors should
borne out because the retinoblastoma (RB) tumor sup-
article. have potent clinical activity against a wide variety of
pressor is one of the key substrates of CDK4/67 and is also
Accepted on cancers; however, the evaluation of this class of
February 28, 2020 a commonly mutated tumor suppressor in a fashion that is
agents in a multitude of tumor types has demon-
and published at mutually exclusive with CDK4/6 deregulation.8
strated a varying degree of clinical efficacy. 3,20 This
ascopubs.org on May
18, 2020: DOI https://
The canonical view of the RB pathway is presented in may be related to diverse biologic outcomes that may
doi.org/10.1200/ Figure 1, where the activation of CDKs 4 and 6 by occur after CDK4/6 inhibitor–mediated quiescence.
EDBK_281085 multiple mechanisms is required for the initiation of RB Some cell types may more easily adapt than others and

Knudsen, Shapiro, and Keyomarsi
with letrozole alone was similar in unselected patients and

in those whose tumors harbored CCND1 amplification or
CDKN2A loss.31 However, more recent analyses among
• Selective CDK4/6 inhibitors typically lead to G1 a large number of cell lines and preclinical models of varying
arrest and/or senescence against solid tumors
tumor types suggested that tumors with D-cyclin activating
and have demonstrated meaningful clinical and
features (DCAF) might be particularly sensitive to CDK4/6
survival benefit in combination with hormonal
treatment in estrogen receptor–positive breast inhibition, including those with 30 UTR alterations that sta-
cancer. bilize D-cyclin mRNAs and encoded proteins, as can occur in
mantle cell lymphoma (MCL) or endometrial cancer. Other
• Multiple intrinsic and acquired resistance
DCAF features include CCND2 or CCND3 amplification, as
mechanisms have been described, including
loss of RB, elevated CDK6 activity, FGFR in multiple hematopoietic malignancies, or less commonly
pathway activation, and cyclin E-CDK2 among solid tumors.32,33
activation. CELL CYCLE PLASTICITY AND MECHANISMS OF INTRINSIC
• Analysis of matched pre- and post-progression AND ACQUIRED RESISTANCE
biopsies among early and late progressors will
help define biomarkers of intrinsic and acquired Loss of RB has emerged as a mechanism of both intrinsic and
resistance that will ultimately guide develop- acquired resistance. At the genetic level, RB1 loss is rare in
ment of rational combinations relevant to breast advanced ER+ breast cancer, occurring in less than 10% of
cancer and other tumor types. patients. Although a rare event, in one report, in 9 of 338
• Pharmacodynamic markers of selective CDK4/ patients with RB1 loss who were subsequently treated with
6 inhibition include reduced phosphorylation CDK4/6 inhibitors, progression-free survival (PFS) was only
of RB, as well as reduced fluorothymidine- 3.6 months compared with 10.1 months for patients with
positron emission tomography uptake and intact RB1.34 Similarly, RB1 mutations have been reported
reduced serum thymidine kinase 1 activity as in samples obtained from patients after development of
noninvasive markers of G1 arrest that can acquired resistance to CDK4/6 inhibitors.35
confirm target engagement in clinical trials.
However, RB1 mutation or loss does not account for most of
the acquired resistance observed to date in breast cancer; other
mechanisms must govern the evolution to a presumed CDK4-/
readily demonstrate resistance, whereas other cells may 6-independent state. In addition, in many other tumor types, it
convert to a senescent state, resulting in more long-term appears that a CDK4-/6-independent state is either preexisting
disease control.21-23 or evolves rapidly,5,36 thereby limiting clinical efficacy of mon-
otherapy. Work in mouse models has provided insight into
CLINICAL USE OF CDK4/6 INHIBITORS AND BIOMARKERS mechanisms by which CDK4/6 inhibition is bypassed, dem-
OF SENSITIVITY onstrating that cells in the mouse can adapt to the loss or re-
CDK4/6 inhibitors as single agents are particularly effective in duction of CDK4/6 activity so that other cyclin and/or CDK
suppressing the proliferation of luminal-like breast cancer complexes take their place and mediate the phosphorylation of
models.13 In addition, CDK4/6 inhibition can be combined RB, referred to as “cell cycle plasticity.”37,38
effectively with endocrine therapies used in the management
The ability to bypass pharmacologic CDK4/6 inhibition,
of breast cancer and can clearly exert activity in tumor models
whether intrinsic or acquired by cell cycle plasticity, occurs
resistant to endocrine therapy.13,24,25 This preclinical efficacy
through two broad mechanisms. First, high levels of CDK4/
has been validated in the clinic, which led to the U.S. Food
6 and D-cyclin complexes can either titrate the pharma-
and Drug Administration approvals for palbociclib, ribociclib,
cologic inhibitor or escape inhibition and have been de-
and abemaciclib in combination with endocrine therapy in the
scribed in multiple preclinical and clinical settings. For
context of HR+/HER2– breast cancer, 4,20 where they are
example, FAT1, a putative tumor suppressor and a member
highly effective in delaying progression of disease. In
of the cadherin superfamily that interacts with the Hippo
several cases, these treatments have improved overall
signaling pathway, has recently been shown to regulate
survival 26-29 compared with endocrine therapy alone.
the expression of CDK6, and its loss may mediate resis-
Abemaciclib has also demonstrated monotherapy activity.16,30
tance to CDK4/6 inhibitors. Knockout of FAT1 leads to
Despite these striking clinical results, biomarkers predicting the downregulation of the Hippo pathway and over-
activity of CDK4/6 inhibitors have been difficult to identify, expression of CDK6.34 Genetic sequencing of 348 bi-
other than preserved expression of the RB protein, a primary opsies from patients who were subsequently treated
CDK4/6 target. In the PALOMA-1 study, the improvement with CDK4/6 inhibitor–based therapy revealed that FAT1
afforded by combined palbociclib and letrozole compared was mutated in approximately 6% of these patients.

Selective CDK4/6 Inhibition As an Anticancer Strategy
FIGURE 1. Canonical G1/S Cell Cycle Control

Activated retinoblastoma (RB) participates in the repression of an E2F-mediated transcriptional program required for S-phase
progression. Mitogenic signals stimulate the assembly of cyclin D-CDK4/6 complexes that initiate RB phosphorylation, followed
sequentially by cyclin E-CDK2–mediated phosphorylation that ultimately releases phosphorylated RB and allows E2F-mediated
transcription. Cyclin D-CDK4/6 complexes may be inhibited by endogenous inhibitors, such as p16INK4A, or by mimetic drugs,
including palbociclib, ribociclib, and abemaciclib.
Abbreviation: RB, retinoblastoma.
Patients with FAT1 mutation had a PFS of only 2.4 months specimens. 43 FGFR-mediated signaling also played
compared with 10.1 months for patients without FAT1 a role in resistance in KRAS-dependent models.42 In
mutations.34 In preclinical models of acquired CDK4/6 contrast, the role of mutations in other upstream proteins
inhibitor resistance, both CDK6 amplification39 and CDK6 that could affect cyclin D1-CDK4/6 activity is less clear,
overexpression via microRNA-mediated modulation of including ESR1 and PIK3CA mutations. For example, in the
the transforming growth factorbeta (TGF-β) pathway PALOMA-3 study, the phase III clinical trial testing the
have been described; the latter has also been validated combination of palbociclib and fulvestrant, there was no
in samples from patients whose tumors demonstrated correlation between response and PFS and the presence of
CDK4/6 inhibitor resistance. 40 In addition, in KRAS- ESR1 or PIK3CA mutations, which indicated that palbociclib
dependent models of pancreatic and lung cancer, adap- was equally active regardless of mutational status.44,45
tation to CDK4/6 inhibition has resulted in increased Similar data were reported with ribociclib used as first-
functional cyclin-CDK complexes that mediate resistance.41,42 line therapy in hormone receptor–positive breast cancer
(MONALEESA-2).46
Similarly, upstream alterations that likely modulate cyclin
D1-CDK4/6 activity have also been observed among re- Second, in addition to alterations in the expression of the
sistant tumors, including those activating the fibroblast target cyclin-CDK complexes, CDK2 can assume the role of
growth factor receptor (FGFR) signaling pathway. Next- CDK4/6 and therefore render tumor cells resistant.47 Cyclin E,
generation sequencing of circulating tumor DNA from a regulatory subunit of CDK2, is central to the initiation of
patients enrolled in the MONALEESA-2 trial demon- DNA replication at the G1/S checkpoint. Cyclin E-CDK2
strated that patients with FGFR1 amplification exhibited can also phosphorylate RB and can result in the acti-
a shorter PFS compared with patients with wild-type vation of E2F target genes. Because cyclin E-CDK2
FGFR1.43 Similarly, following progression on CDK4/6 in- phosphorylation events are downstream of those me-
hibitors, FGFR1/2 amplification or activating muta- diated by cyclin D-CDK4/6, overexpression of cyclin E
tions were identified in 14 of 34 (41%) postprogression renders inhibition of CDK4/6 ineffective in inducing G1

arrest or subsequent growth inhibition. Hence, tumors are largely prognostic. To better develop biomarkers of re-
with high levels of cyclin E are likely to be intrinsically sistance to CDK4/6 inhibitors, it is critical to first identify if
resistant to CDK4/6 inhibition. Gene expression profil- patients are early or late progressors. To this end, it is reasonable
ing on 302 archival formalin-fixed, paraffin-embedded to define time to progression in groups of patients based on
samples from the PALOMA 3 trial revealed that in patients the historic median PFS of 24.8 months for first-line therapy
treated with combined palbociclib and fulvestrant, those and 9.5 months for second-line therapy. Progressors can
with high cyclin E1 mRNA expression had a median PFS of be subdivided into three groups that likely have distinct
7.6 months compared with those with low cyclin E mRNA, molecular characteristics (Fig. 2): (1) early progressors
who had a PFS of 14.1 months.48 (6 months post-therapy for all patients); (2) intermediate
Cyclin E is also post-translationally modified by neutrophil progressors (6–24 months post-therapy for first-line therapy
elastase–mediated proteolytic cleavage to generate the low patients [Fig. 2A] and 6–9 months post-therapy for second-
molecular–weight isoforms of cyclin E (LMW-E) that are line therapy patients [Fig. 2B]); and (3) late progressors (. 24
detected in many cancer types. 49 LMW-E lacks the months post-therapy for first-line therapy patients [Fig. 2A]
N-terminal cyclin E nuclear localization signal and promotes and . 9 months post-therapy for second-line therapy pa-
its accumulation in the cytoplasm. Compared with full- tients [Fig. 2B]). Early progressors are considered to be in-
length cyclin E, the aberrant localization and unique ste- trinsically resistant to CDK4/6 inhibitors or endocrine
reochemistry of LMW-E dramatically alters the substrate therapy (first or second line) as single agents, whereas
specificity and selectivity of CDK2, increasing tumorigenicity intermediate and late progressors are considered to have
in experimental models. Cytoplasmic LMW-E, which can be acquired resistance to the combination of CDK4/6 in-
assessed by immunohistochemistry, is prognostic of poor hibitors and endocrine therapy. Having distinguished the
survival independent of breast cancer subtype50,51 and also early from the late progressors, the selected biomarkers
predicts lack of response to neoadjuvant chemotherapy in must be assessed in matched pretreatment and post-
breast cancer.52 Immunohistochemical analysis of cyto- progression biopsy material from these patients. By ex-
plasmic cyclin E on archival tumor tissue from 109 hormone amining only the pretreatment samples, the biomarkers
receptor–positive patients with advanced breast cancer who selected will likely identify the intrinsically resistant patient
were treated with palbociclib as first line (+ letrozole) or cohort. However, if postprogression matched samples are
second line (+ fulvestrant) endocrine therapy revealed that also assessed for the same biomarkers, novel insights into
49.5% of these patients were positive for cyclin E, which was the utility of that specific biomarker for potentially pre-
associated with a worse prognosis.53 Specifically, patients dicting acquired resistance to CDK4/6 inhibitor–based
with cytoplasmic cyclin E–positive tumors had a median therapy can be attained. When matched tumor samples
PFS of 13.4 months compared with a PFS of 36.5 months in (n = 25) from patients with breast cancer who progressed
those with cytoplasmic negative tumors. on palbociclib were examined for deregulation of cyclin E,
estrogen receptor, DNA repair, and interleukin-6/STAT3
GUIDELINES FOR BIOMARKER IDENTIFICATION signaling, results revealed that these pathways were all
As evident by these examples, biomarkers of resistance to altered compared with pretreatment tumor samples, which
CDK4/6 inhibitors are heterogeneous in nature and currently suggested that the alterations of these pathways were
have been primarily examined in pretreatment biopsies, which specific to palbociclib resistance.54
FIGURE 2. Analysis of Disease

Progression
Analysis of disease progression
while on treatment in advanced
breast cancer, ER-positive/HER-2
negative patients treated with (A)
first-line (CDK4/6 inhibitors +
aromatase inhibitor) or (B)
second-line (CDK4/6 inhibitors +
fulvestrant) therapy. The pro-
gressors in each treatment group
are divided into intrinsic (early)
or acquired (intermediate + late)
resistance as described in the
text.

TABLE 1. Representative Combinations of CDK4/6 Inhibitors With Hormonal Agents, Signal Transduction Inhibitors, and Chemotherapy
Combination Representative Clinical Studies NCT Identifier
+ Estrogen inhibition (breast cancer) FDA-approved
+ Estrogen inhibition (gyn malignancies) Abemaciclib With Letrozole in Recurrent or Persistent Endometrial NCT03675893
Cancer
Ribociclib and Letrozole Treatment in Ovarian Cancer NCT03673124
Palbociclib Plus Letrozole Treatment After Progression to Second Line NCT03936270
Chemotherapy for Women With ER/PR-positive Ovarian Cancer.
(LACOG1018)
+ Androgen inhibition (prostate cancer) A Study of Abiraterone Acetate Plus Prednisone With or Without NCT03706365
Abemaciclib (LY2835219) in Participants With Prostate Cancer
Enzalutamide With and Without Ribociclib for Metastatic, Castrate- NCT02555189
Resistant, Chemotherapy-Naive Prostate Cancer That Retains RB
Expression
A Phase II Study of Androgen Deprivation Therapy With or Without NCT02059213
Palbociclib in RB-Positive Metastatic Prostate Cancer
+ MTOR inhibitor Ribociclib and Everolimus in Treating Children With Recurrent or NCT03387020
Refractory Malignant Brain Tumors
Phase II Trial of Ribociclib and Everolimus in Advanced Dedifferentiated NCT03114527
Liposarcoma (DDL) and Leiomyosarcoma (LMS)
A Study of LEE011 With Everolimus in Patients With Advanced NCT03070301
Neuroendocrine Tumors
+ MEK inhibitor Study of Safety and Efficacy of Ribociclib and Trametinib in Patients With NCT02703571
Metastatic or Advanced Solid Tumors
A Phase Ib/II Study of LEE011 in Combination With MEK162 in Patients NCT01781572
With NRAS Mutant Melanoma
Study of the CDK4/6 Inhibitor Palbociclib (PD-0332991) in Combination NCT03170206
With the MEK Inhibitor Binimetinib (MEK162) for Patients With
Advanced KRAS Mutant Non-Small Cell Lung Cancer
Binimetinib and Palbociclib or TAS-102 in Treating Patients With KRAS NCT03981614
and NRAS Mutant Metastatic or Unresectable Colorectal Cancer
+ PI3K inhibitor Study of the CDK4/6 Inhibitor Palbociclib (PD-0332991) in Combination NCT03065062
With the PI3K/mTOR Inhibitor Gedatolisib (PF-05212384) for Patients
With Advanced Squamous Cell Lung, Pancreatic, Head & Neck and
Other Solid Tumors
PIPA: Combination of PI3 Kinase Inhibitors and PAlbociclib (PIPA) NCT02389842
+ Tyrosine kinase inhibitor A Study of Abemaciclib in Combination With Sunitinib in Metastatic Renal NCT03905889
Cell Carcinoma
+ ALK inhibitor Study of Safety and Efficacy of LEE011 and Ceritinib in Patients With ALK- NCT02292550
positive Nonsmall Cell Lung Cancer.
+ EGFR inhibitor Study of Safety and Efficacy of LEE011 and Ceritinib in Patients With ALK- NCT02292550
positive Nonsmall Cell Lung Cancer.
Palbociclib and Cetuximab in Metastatic Colorectal Cancer NCT03446157
PD 0332991 and Cetuximab in Patients With Incurable SCCHN NCT02101034
+ Taxane LEE011 (Ribociclib) in Combination With Docetaxel Plus Prednisone in NCT02494921
mCRPC
Dose-Escalation Study of Palbociclib + Nab-Paclitaxel in mPDAC NCT02501902

TABLE 1. Representative Combinations of CDK4/6 Inhibitors With Hormonal Agents, Signal Transduction Inhibitors, and Chemotherapy (Continued)
+ Platinum Palbociclib With Cisplatin or Carboplatin in Advanced Solid Tumors NCT02897375
Ribociclib (Ribociclib (LEE-011)) With Platinum-based Chemotherapy in NCT03056833
Recurrent Platinum Sensitive Ovarian Cancer
Phase II Trial Evaluating the Efficacy of Palbociclib in Combination With NCT03194373
Carboplatin for the Treatment of Unresectable Recurrent or Metastatic
Head and Neck Squamous Cell Carcinoma
+ Anthracycline Ribociclib and Doxorubicin in Treating Patients With Metastatic or NCT03009201
Advanced Soft Tissue Sarcomas That Cannot Be Removed by Surgery
+ Gemcitabine Ribociclib and Gemcitabine Hydrochloride in Treating Patients With NCT02414724
Advanced Solid Tumors or Lymphoma
+ Combination chemotherapy PD-0332991, 5-FU, and Oxaliplatin for Advanced Solid Tumor NCT01522989
Malignancies
Study of Palbociclib Combined With Chemotherapy in Pediatric Patients NCT03709680
With Recurrent/Refractory Solid Tumors
Abbreviation: FDA, U.S. Food and Drug Administration.
Therefore, there is a continued need for development of Examples include the combination of EGFR and CDK4/6
reliable biomarkers to identify patients with resistance to inhibition in EGFR-mutant lung cancer or MEK inhibition with
endocrine therapy and CDK4/6 inhibitors as single agents CDK4/6 inhibition in a RAS-driven tumor.42,56-58 Considerable
and in combination that will be clinically meaningful to guide preclinical data support a wide spectrum of such combination
patient care. In addition, exploring mechanisms of disease therapies that either build off U.S. Food and Drug Administration–
resistance could considerably improve development of approved targeted therapies or represent new combination
novel treatments effective in delaying or reversing resistance approaches to target a specific cancer (Table 1). Although
and ultimately improving survival. many of these strategies are strongly supported by preclinical
data, emerging clinical results remain uneven and will require
COMBINATION THERAPIES TO ENHANCE THE ACTIVITY AND randomized studies to clarify the contribution of CDK4/6
INDICATIONS FOR CDK4/6 INHIBITORS inhibition. In addition, toxicity of certain combinations rep-
How cell cycle plasticity is fundamentally controlled or can resents a substantial challenge.
be predicted is unclear. However, the framework of re- Chemotherapy and CDK4/6 inhibition at face value would
sistance outlined provides a basis for combination therapy, appear to be antagonistic, because most chemotherapy
with the hope of enhancing clinical efficacy of CDK4/6 in- kills tumor cells based on ongoing cell cycle progression
hibitors beyond HR+/HER2 breast cancer. and mitotic division. CDK4/6 inhibition has been shown to
To expand the therapeutic efficacy of CDK4/6 inhibitors, antagonize responses to chemotherapy in RB-expressing
multiple combinatorial approaches have been undertaken cells.59,60 However, most of these studies investigated
and are summarized in Table 1. These efforts largely fall short-term endpoints and did not necessarily model how to
among three different classes of agents: endocrine therapy, best combine the agents for efficacy. A number of recent
kinase inhibitors, and chemotherapy. studies have suggested that CDK4/6 inhibitors could be
used metronomically or as maintenance strategies follow-
Similar to combined CDK4/6 inhibition and endocrine ing different chemotherapy regimens.12,61-64 The nature
therapies in breast cancer, it has been shown that CDK4/ of the positive interaction of chemotherapy and CDK4/6
6 inhibitors will cooperate with androgen-deprivation inhibition is complex. RB regulates the expression of
strategies in preclinical models of prostate cancer55 and many factors associated with DNA repair or recovery from
with estrogen depletion in specific gynecologic malignan- mitotic stress,9 and therefore, if RB is active, it could
cies. Whether there is clinical benefit of the combination of prevent recovery from chemotherapy-mediated damage.
CDK4/6 inhibition and these approaches is being tested in Conversely, it is well known that chemotherapy will elicit
several clinical trials (Table 1). mechanisms to downregulate CDK activity and could there-
In addition to endocrine therapy, a host of kinase inhibitors fore cooperate with CDK4/6 inhibition. Irrespective of these
have been shown to positively interact with CDK4/6 inhibitors. mechanisms, clinical studies with combination strategies
In many such contexts, the kinase inhibitor is targeting an have produced some degree of efficacy, and multiple trials are
oncogenic pathway present in the associated tumor type. ongoing (Table 1).65

NONCANONICAL EFFECTS OF ACTIVATING THE RB-PATHWAY pharmacodynamic markers capable of monitoring target
AND NEW VULNERABILITIES WITH CDK4/6 INHIBITION engagement and/or pathway modulation will be critical.
With the extensive research in the activity of CDK4/6 in- Several candidates have been identified for this purpose,
hibitors, it is not surprising that new discoveries are being including expression of phosphorylated RB (phospho-RB)
made that could be clinically actionable beyond effects on protein, 30 -deoxy-30 [18F]-fluorothymidine (FLT)–positron
the cell cycle.5,21,66 Analysis of gene expression data from emission tomography (PET) and serum thymidine kinase-1
cells and tumors in which the RB pathway is activated by (TK1) activity.
CDK4/6 inhibition revealed the induction of an immunologic Clinical studies support the monitoring of phospho-RB for
gene expression program, including genes involved in anti- evidence of CDK4/6 inhibition. This was initially performed
gen presentation and interferon response, which suggests in a pharmacodynamic study of palbociclib in MCL, which
that the tumor cell may become more susceptible to im- demonstrated an 89% reduction in phospho-RB in on-
munologic surveillance.67-71 Parallel studies demonstrated treatment biopsies taken after 3 weeks of palbociclib
that systemic CDK4/6 inhibition led to changes in T-cell treatment (125 mg daily for 3 of 4 weeks), compared with
activity through modulation of NFAT (nuclear factor of pretreatment, with no corresponding changes in total RB.15
activated T cells), which can drive a more potent anti- There was a concurrent reduction in the proliferative marker
tumor response.68 These results have been extended by Ki-67 that correlated with the degree of phospho-RB re-
illustrating cooperation of CDK4/6 inhibition with the duction and that confirmed phospho-RB as a potential
immune-checkpoint blockade in a variety of tumor cell types. marker for monitoring CDK4/6 inhibitor–mediated anti-
In addition, correlative analysis from tumors treated with proliferative responses. However, further correlative ana-
CDK4/6 inhibitors have indicated enhanced immunologic lyses indicated that reductions in Ki-67 and phospho-RB
infiltrates.67 These findings, in concert with the overall ex- were necessary but not sufficient to predict long-term
citement surrounding immunotherapy, have led to a pro- clinical benefit in heavily pretreated patients with MCL,
liferation of clinical studies combining CDK4/6 inhibitors, likely because of subsequent development of resistance in
alone or in combination with other targeted agents and with a subset of patients that limited the use of phospho-RB
immunotherapy (Table 2). Because many of these studies early after treatment initiation as a predictive biomarker.
have only recently been initiated, the data on clinical efficacy However, the absence of reduced RB phosphorylation
and tolerability are not yet mature; however, current early after the start of treatment may correlate with ultimate
preliminary findings appear promising. 72 lack of response. For example, in a short-term preopera-
tive breast cancer trial with palbociclib as a single agent,
The RB pathway, in addition to controlling cell cycle progres-
nonresponders (measured by change in Ki67) were char-
sion, also plays a role in coordinating different features of
acterized by no change in levels of phospho-RB.78 Similar
metabolism. Evolutionarily, this makes sense because cellular
results were seen in the initial trial of abemaciclib mono-
division is tightly linked to nutrient availability to support the
therapy, in which there was a significant correlation be-
biologic mass of two daughter cells. In the case of CDK4/6
tween clinical efficacy and modulation of phospho-RB in
inhibition, there is a shift toward oxidative73-75 or autophagic
proliferating keratinocytes and tumor biopsies. Here, a 60%
metabolism.53,76 These findings could be leveraged using se-
phospho-RB reduction threshold was able to separate most
lective inhibitors targeting these processes to essentially shift the
patients with stable disease or response from those who
cytostatic state induced by CDK4/6 inhibit into a selective
progressed.16
vulnerability.53 Similarly, because RB activation can limit the
expression of multiple DNA repair–associated genes, treat- The role of phospho-RB as a marker for CDK4/6 inhibi-
ment with CDK4/6 inhibitors could yield a functional state tor activity was also assessed in a randomized phase II
similar to multiple forms of DNA repair deficiency, which study of palbociclib (125 or 100 mg) in combination with
ostensibly can be exploited by chemotherapy or radiation either fulvestrant or tamoxifen in metastatic hormone
therapy. Recent studies have also suggested that CDK4/6 receptor–positive breast cancer (TBCRC 035).79 Findings
inhibitors could yield sufficient deficiency in homologous were in line with those observed in patients with MCL;
recombination-mediated repair to drive cooperative in- there was a considerable decrease in phospho-RB and
teractions with PARP inhibitors.77 Evaluation of these vul- Ki-67 staining observed in both skin and tumor biopsies.
nerabilities are being explored clinically (Table 2). The degree of change was comparable between the 125-
and 100-mg palbociclib dose levels, as was clinical
outcome, which indicated that dose reduction of palbo-
PHARMACODYNAMIC MONITORING OF CDK4/6 INHIBITION ciclib, if required for hematologic toxicity, was unlikely
IN CLINICAL TRIALS to compromise pharmacodynamic effects and efficacy
As clinical trials of CDK4/6 inhibitors study multiple tumor signals. In another phase I clinical trial of palbociclib
types in combination with other agents, the assessment of and the MEK inhibitor PD-0325901 in RAS-mutant solid

TABLE 2. Representative Clinical Trials Combining CDK4/6 Inhibition With Immune Checkpoint Blockade or With Inhibitors of Cellular Metabolism
+ Immune-checkpoint inhibitor (anti–PD-1/anti–PD-L1) Abemaciclib and Pembrolizumab in Locally Advanced Unresectable or NCT03997448
Metastatic Gastroesophageal Adenocarcinoma: Big Ten Cancer
Research Consortium BTCRC-GI18-149
Clinical Trial of Abemaciclib in Combination With Pembrolizumab in NCT03938337
Patients With Metastatic or Recurrent Head and Neck Cancer
Phase Ib Study of TNO155 in Combination With Spartalizumab or NCT04000529
Ribociclib in Selected Malignancies
Avelumab, Cetuximab, and Palbociclib in Recurrent or Metastatic Head NCT03498378
and Neck Squamous Cell Carcinoma
Pilot Study of Pembrolizumab Combined With Pemetrexed or NCT04220892
Abemaciclib for High Grade Glioma
Abemaciclib and Nivolumab for Subjects With Hepatocellular Carcinoma NCT03781960
Ribociclib and Spartalizumab in R/M HNSCC (RISE-HN) NCT04213404
+ Metabolic inhibitors Hydroxychloroquine, Palbociclib, and Letrozole Before Surgery in NCT03774472
Treating Participants With Estrogen Receptor Positive, HER2 Negative
Breast Cancer
A Study of Telaglenastat (CB-839) in Combination With Palbociclib in NCT03965845
Patients With Solid Tumors
tumors that compared continuous and intermittent (3 weeks TK1 detectable in serum. Serum TK1 activity can be measured
of every 4 weeks [3/1 dosing]) palbociclib dosing, serial anal- in patients with cancer using the DiviTum Assay (Biovica,
ysis of skin keratinocytes during the first treatment cycle Uppsala, Sweden). This is an enzyme-linked immunosorbent–
demonstrated that phospho-RB and Ki-67 staining mirrored based assay in which BrdU (5-bromo-20 -deoxyuridine) is
palbociclib exposure. There was a sustained decrease in pa- incorporated into a synthetic DNA strand fixed to the assay
tients who underwent continuous dosing, whereas intermittent plate, with the extent of BrdU incorporation reflecting TK1
dosing resulted in a recovery of phospho-RB staining during the activity present in the serum (Fig. 3). In breast cancer,
off week of palbociclib.80 serum TK1 activity was associated with advanced stage,
higher grade, tumor necrosis, vascularity, and estrogen
The ability to use noninvasive or minimally invasive bio-
markers is preferential to biopsy-based biomarkers, and [18F] receptor and progesterone receptor negativity.82 TK1 activity
FLT identifies cells undergoing active DNA synthesis, which was also elevated in patients with BRCA1/2 mutations
allows for its potential use as a surrogate marker for S-phase compared with wild-type and was an independent predictor
and the quantification of cellular proliferation.81 The MCL of disease recurrence.82 In addition, TK1 activity levels were
study also examined the role of pre- and on-treatment [18F] determined in patients with advanced breast cancer before
FLT-PET as an imaging biomarker to demonstrate the bi- they underwent combination chemotherapy with either
ologic effect of G1 arrest mediated by palbociclib.15 Of 16 epirubicin/paclitaxel or epirubicin/paclitaxel/capecitabine
evaluable patients, there were significant reductions in the using the DiviTum assay. Analysis then identified TK1 ac-
summed [18F] FLT-PET maximal standard uptake value in tivity to be predictive for PFS and overall survival, as well as
most patients, and in patients with PFS longer than 1 year response to treatment.83 Similarly, patients with hormone
(n = 5, including 1 patient with complete response and 2 receptor–positive breast cancer with low baseline levels of
patients with partial responses), there was a more than 70% serum TK1 activity or an early drop in TK1 activity had
reduction in summed [18F] FLT maximal standard uptake improved outcomes in response to endocrine therapy.84
value. However, correlative analysis with clinical response Because TK1 expression is E2F-dependent, CDK4/6 inhibitor
indicated that as with phospho-RB and Ki-67 reductions, the treatment should result in reduced serum TK1 activity levels.
reduction in [18F] FLT uptake was necessary but not sufficient Coupled with its demonstrated prognostic value in early-stage
for predicting long-term survival benefit. breast cancer, serum TK1 activity may act as a potential bio-
TK1 is an enzyme critical for DNA synthesis and is expressed marker for target engagement and clinical benefit following
in all human cells during normal cell division through E2F- CDK4/6 inhibitor treatment. This was examined as part of the
dependent transcription, with small amounts found in serum. NeoPalAna trial14 that examined neoadjuvant palbociclib and
Tumors, because of their higher replication rates compared anastrazole in early-stage hormone receptor–positive breast
with normal tissues, are capable of secreting pathologic levels of cancer, in which serum TK1 activity was serially monitored until

FIGURE 3. Liquid Biopsy for G1 arrest (Biovica DiviTum Assay)

TK phosphorylates the nucleoside analog BrdU (5-bromo-20 -deoxyuridine) to BrdUMP (5-bromo-20 -deoxyuridine
50 -monophosphate), which is further phosphorylated to BrdUTP (5-bromo-20 -deoxyuridine 50 -triphosphate).
BrdUTP is then incorporated into a solid-phase DNA-strand and incorporated BrdU detected using an anti-BrdU
monoclonal antibody conjugated to the signal generating enzyme alkaline phosphatase. The level of BrdU in-
corporated over time is proportional to the level of TK activity in the sample. Reproduced with permission from
Biovica.com/Divitum.
Abbreviation: TK, thymidine kinase.
the time of surgery using the DiviTum assay.85 Patients received tested the activity and safety of single-agent palbociclib
an initial 4 weeks of anastrozole, followed by palbociclib on against palbociclib combined with the endocrine therapy on
cycle 1, day 1 (C1D1) for four 28-day cycles. Surgery occurred which patients had progressed most recently before enroll-
following 3–5 weeks of washout from the last dose of palbo- ment. Although baseline TK1 activity was not prognostic, after
ciclib, except in a small subset of patients who received pal- one cycle of treatment, patients who demonstrated an in-
bociclib (cycle 5) continuously until surgery. Serum TK1 activity crease in TK1 activity had a worse outcome compared with
was determined at baseline, on C1D1, on C1D15, and at time of those in which activity was decreased or stable, with a median
surgery. Despite a considerable drop in tumor Ki-67 with PFS of 3.0 months versus 9.0 months, respectively. TK1
anastrozole monotherapy, there was no statistically significant activity was also assayed at the time of resistance; those with
change in TK1 activity. However, a striking reduction in TK1 activity higher than the median fared worse on poststudy
activity was observed 2 weeks after initiation of palbociclib treatment compared with those with lower activity.86 This
(C1D15), which then rose significantly with palbociclib wash- work suggested that an initial rise in TK1 activity correlated
out. At C1D15, TK1 activity was below the detection limit in with intrinsic drug resistance and that a rise after an initial
more than 90% of patients, indicating a profound effect of reduction correlated with acquired resistance that could be
CDK4/6 inhibition. There was high concordance between stratified prognostically.
changes in serum TK1 and tumor Ki-67 and TK1 mRNA levels Further work will be required to fully define the degree of re-
in the same direction from C1D1 to C1D15 and from C1D15 to duction in TK1 activity necessary to predict response or long-
surgery time points. Similar findings were demonstrated in the term clinical benefit. Identification of this threshold is particularly
phase I study of palbociclib and the MEK inhibitor PD- important in combination therapy studies in which attenuation of
0325901, which compared two different palbociclib dosing CDK4/6 inhibitor dosing may be required. In addition, further
schedules—continuous and 3 weeks of every 4 weeks (3/1 studies with more intensive serial sampling will be required to
dosing).80 Serial measurements of TK activity during the first determine whether a rise in TK1 activity that indicates acquired
treatment cycle demonstrated that serum TK activity mirrored resistance predates resistance defined radiographically. Finally,
palbociclib administration; in patients who received 3/1 pal- to date, most studies using serum TK1 activity have focused on
bociclib dosing serum, TK1 activity decreased for the 3 weeks palbociclib, and additional studies of the effects of abemaciclib
of palbociclib exposure, with subsequent increases during the and ribociclib on this marker are warranted.
off week. In patients treated with continuous palbociclib, TK FUTURE PROSPECTS
activity remained low throughout the cycle. In addition, TK
With intense ongoing research, it is expected that more
activity profiles also reflected those seen for phospho-RB and
therapeutic combinatorial strategies will emerge from
Ki-67 staining in keratinocytes.
the preclinical arena. Rationally developed clinical trials
These preliminary results highlight the potential for serum that incorporate biomarkers for patient selection, assess
TK1 activity to act as a noninvasive biomarker for CDK4/6 tumors at the time of resistance, and incorporate phar-
inhibitor target engagement. This was further examined in 46 macodynamic endpoints will be crucial to fully leverage the
patients enrolled in the TREnd study, a phase II trial that promise of targeting the RB pathway in cancers.


1
Center for Personalized Medicine and Department of Molecular and Geoffrey I. Shapiro, MD, PhD, Dana-Farber Cancer Institute, Mayer 446,
Cellular Biology, Roswell Park Cancer Institute, Buffalo, NY 450 Brookline Avenue, Boston, MA 02215; email: geoffrey_shapiro@
2
Early Drug Development Center, Department of Medical Oncology, Dana- dfci.harvard.edu.
Farber Cancer Institute and Department of Medicine, Brigham and
Women’s Hospital and Harvard Medical School, Boston, MA
3
Department of Experimental Radiation Oncology, The University of Texas
MD Anderson Cancer Center, Houston, TX
EDBK_281085.
REFERENCES
1. Sherr CJ. Cancer cell cycles. Science. 1996;274:1672-1677.
2. Shapiro GI. Cyclin-dependent kinase pathways as targets for cancer treatment. J Clin Oncol. 2006;24:1770-1783.
3. Asghar U, Witkiewicz AK, Turner NC, et al. The history and future of targeting cyclin-dependent kinases in cancer therapy. Nat Rev Drug Discov. 2015;
14:130-146.
4. Sherr CJ, Beach D, Shapiro GI. Targeting CDK4 and CDK6: from discovery to therapy. Cancer Discov. 2016;6:353-367.
5. Knudsen ES, Pruitt SC, Hershberger PA, et al. Cell cycle and beyond: exploiting new RB1 controlled mechanisms for cancer therapy. Trends Cancer. 2019;
5:308-324.
6. Diehl JA. Cycling to cancer with cyclin D1. Cancer Biol Ther. 2002;1:226-231.
7. Matsushime H, Ewen ME, Strom DK, et al. Identification and properties of an atypical catalytic subunit (p34PSK-J3/cdk4) for mammalian D type G1 cyclins. Cell.
1992;71:323-334.
8. Palmero I, Peters G. Perturbation of cell cycle regulators in human cancer. Cancer Surv. 1996;27:351-367.
9. Cam H, Dynlacht BD. Emerging roles for E2F: beyond the G1/S transition and DNA replication. Cancer Cell. 2003;3:311-316.
10. Toogood PL, Harvey PJ, Repine JT, et al. Discovery of a potent and selective inhibitor of cyclin-dependent kinase 4/6. J Med Chem. 2005;
48:2388-2406.
11. Fry DW, Harvey PJ, Keller PR, et al. Specific inhibition of cyclin-dependent kinase 4/6 by PD 0332991 and associated antitumor activity in human tumor
xenografts. Mol Cancer Ther. 2004;3:1427-1438.
12. Gelbert LM, Cai S, Lin X, et al. Preclinical characterization of the CDK4/6 inhibitor LY2835219: in-vivo cell cycle-dependent/independent anti-tumor activities
alone/in combination with gemcitabine. Invest New Drugs. 2014;32:825-837.
13. Finn RS, Dering J, Conklin D, et al. PD 0332991, a selective cyclin D kinase 4/6 inhibitor, preferentially inhibits proliferation of luminal estrogen receptor-positive
human breast cancer cell lines in vitro. Breast Cancer Res. 2009;11:R77.
14. Ma CX, Gao F, Luo J, et al. NeoPalAna: neoadjuvant palbociclib, a cyclin-dependent kinase 4/6 inhibitor, and anastrozole for clinical stage 2 or 3 estrogen
receptor-positive breast cancer. Clin Cancer Res. 2017;23:4055-4065.
15. Leonard JP, LaCasce AS, Smith MR, et al. Selective CDK4/6 inhibition with tumor responses by PD0332991 in patients with mantle cell lymphoma. Blood. 2012;
119:4597-4607.
16. Patnaik A, Rosen LS, Tolaney SM, et al. Efficacy and safety of abemaciclib, an inhibitor of CDK4 and CDK6, for patients with breast cancer, non-small cell lung
cancer, and other solid tumors. Cancer Discov. 2016;6:740-753.
17. Infante JR, Cassier PA, Gerecitano JF, et al. A phase I study of the cyclin-dependent kinase 4/6 inhibitor ribociclib (LEE011) in patients with advanced solid
tumors and lymphomas. Clin Cancer Res. 2016;22:5696-5705.
18. Konecny GE, Winterhoff B, Kolarova T, et al. Expression of p16 and retinoblastoma determines response to CDK4/6 inhibition in ovarian cancer. Clin Cancer Res.
2011;17:1591-1602.
19. Dean JL, Thangavel C, McClendon AK, et al. Therapeutic CDK4/6 inhibition in breast cancer: key mechanisms of response and failure. Oncogene. 2010;
29:4018-4032.
20. O’Leary B, Finn RS, Turner NC. Treating cancer with selective CDK4/6 inhibitors. Nat Rev Clin Oncol. 2016;13:417-430.
21. Klein ME, Kovatcheva M, Davis LE, et al. CDK4/6 inhibitors: the mechanism of action may not be as simple as once thought. Cancer Cell. 2018;34:9-20.
22. Kovatcheva M, Liao W, Klein ME, et al. ATRX is a regulator of therapy induced senescence in human cells. Nat Commun. 2017;8:386.
23. Kovatcheva M, Liu DD, Dickson MA, et al. MDM2 turnover and expression of ATRX determine the choice between quiescence and senescence in response to
CDK4 inhibition. Oncotarget. 2015;6:8226-8243.

24. Thangavel C, Dean JL, Ertel A, et al. Therapeutically activating RB: reestablishing cell cycle control in endocrine therapy-resistant breast cancer. Endocr Relat
Cancer. 2011;18:333-345.
25. Miller TW, Balko JM, Fox EM, et al. ERα-dependent E2F transcription can mediate resistance to estrogen deprivation in human breast cancer. Cancer Discov.
2011;1:338-351.
26. Turner NC, Slamon DJ, Ro J, et al. Overall survival with palbociclib and fulvestrant in advanced breast cancer. N Engl J Med. 2018;379:1926-1936.
27. Sledge GW, Jr., Toi M, Neven P, et al. The effect of abemaciclib plus fulvestrant on overall survival in hormone receptor-positive, ERBB2-negative breast cancer
that progressed on endocrine therapy-MONARCH 2: a randomized clinical trial. JAMA Oncol. 2019;6:116-124.
28. Im SA, Lu YS, Bardia A, et al. Overall survival with ribociclib plus endocrine therapy in breast cancer. N Engl J Med. 2019;381:307-316.
29. Slamon DJ, Neven P, Chia S, et al. LBA7_PR - Overall survival results of the Phase III MONALEESA-3 trial of postmenopausal patients with hormone receptor-
positive human epidermal growth factor 2-negative advanced breast cancer treated with fulvestrant 6 ribociclib. Ann Oncol. 2019;30 (Suppl_5):856-857.
30. Dickler MN, Tolaney SM, Rugo HS, et al. MONARCH 1, a phase II study of abemaciclib, a CDK4 and CDK6 inhibitor, as a single agent, in patients with refractory
HR+/HER2- metastatic breast cancer [published correction appears in Clin Cancer Res. 2018;24:5485]. Clin Cancer Res. 2017;23:5218-5224.
31. Finn RS, Crown JP, Lang I, et al. The cyclin-dependent kinase 4/6 inhibitor palbociclib in combination with letrozole versus letrozole alone as first-line treatment of
oestrogen receptor-positive, HER2-negative, advanced breast cancer (PALOMA-1/TRIO-18): a randomised phase 2 study. Lancet Oncol. 2015;16:25-35.
32. Shapiro GI. Genomic biomarkers predicting response to selective CDK4/6 inhibition: progress in an elusive search. Cancer Cell. 2017;32:721-723.
33. Gong X, Litchfield LM, Webster Y, et al. Genomic aberrations that activate D-type cyclins are associated with enhanced sensitivity to the CDK4 and CDK6 inhibitor
abemaciclib. Cancer Cell. 2017;32:761-776.e6.
34. Li Z, Razavi P, Li Q, et al. Loss of the FAT1 tumor suppressor promotes resistance to CDK4/6 inhibitors via the Hippo pathway. Cancer Cell. 2018;34:893-905.e8.
35. Condorelli R, Spring L, O’Shaughnessy J, et al. Polyclonal RB1 mutations and acquired resistance to CDK 4/6 inhibitors in patients with metastatic breast cancer.
Ann Oncol. 2018;29:640-645.
36. Knudsen ES, Witkiewicz AK. The strange case of CDK4/6 inhibitors: mechanisms, resistance, and combination strategies. Trends Cancer. 2017;3:39-55.
37. Malumbres M, Sotillo R, Santamarı́a D, et al. Mammalian cells cycle without the D-type cyclin-dependent kinases Cdk4 and Cdk6. Cell. 2004;118:493-504.
38. Kozar K, Sicinski P. Cell cycle progression without cyclin D-CDK4 and cyclin D-CDK6 complexes. Cell Cycle. 2005;4:388-391.
39. Yang C, Li Z, Bhatt T, et al. Acquired CDK6 amplification promotes breast cancer resistance to CDK4/6 inhibitors and loss of ER signaling and dependence.
Oncogene. 2017;36:2255-2264.
40. Cornell L, Wander SA, Visal T, et al. MicroRNA-mediated suppression of the TGF-beta pathway confers transmissible and reversible CDK4/6 inhibitor resistance.
Cell Rep. 2019;26:2667-2680.e7.
41. Knudsen ES, Kumarasamy V, Ruiz A, et al. Cell cycle plasticity driven by MTOR signaling: integral resistance to CDK4/6 inhibition in patient-derived models of
pancreatic cancer. Oncogene. 2019;38:3355-3370.
42. Haines E, Chen T, Kommajosyula N, et al. Palbociclib resistance confers dependence on an FGFR-MAP kinase-mTOR-driven pathway in KRAS-mutant non-
small cell lung cancer. Oncotarget. 2018;9:31572-31589.
43. Formisano L, Lu Y, Servetto A, et al. Aberrant FGFR signaling mediates resistance to CDK4/6 inhibitors in ER+ breast cancer. Nat Commun. 2019;10:1373.
44. Cristofanilli M, Turner NC, Bondarenko I, et al. Fulvestrant plus palbociclib versus fulvestrant plus placebo for treatment of hormone-receptor-positive, HER2-
negative metastatic breast cancer that progressed on previous endocrine therapy (PALOMA-3): final analysis of the multicentre, double-blind, phase 3
randomised controlled trial [published correction appears in Lancet Oncol. 2016;e136; e270]. Lancet Oncol. 2016;17:425-439.
45. Fribbens C, O’Leary B, Kilburn L, et al. Plasma ESR1 mutations and the treatment of estrogen receptor-positive advanced breast cancer. J Clin Oncol. 2016;
34:2961-2968.
46. Hortobagyi GN, Stemmer SM, Burris HA, et al. Ribociclib as first-line therapy for HR-positive, advanced breast cancer. N Engl J Med. 2016;375:1738-1748.
47. Herrera-Abreu MT, Palafox M, Asghar U, et al. Early adaptation and acquired resistance to CDK4/6 inhibition in estrogen receptor-positive breast cancer. Cancer
Res. 2016;76:2301-2313.
48. Turner NC, Liu Y, Zhu Z, et al. Cyclin E1 expression and palbociclib efficacy in previously treated hormone receptor-positive metastatic breast cancer. J Clin
Oncol. 2019;37:1169-1178.
49. Caruso JA, Duong MT, Carey JPW, et al. Low-molecular-weight cyclin E in human cancer: cellular consequences and opportunities for targeted therapies. Cancer
Res. 2018;78:5481-5491.
50. Hunt KK, Karakas C, Ha MJ, et al. Cytoplasmic cyclin E predicts recurrence in patients with breast cancer. Clin Cancer Res. 2017;23:2991-3002.
51. Karakas C, Biernacka A, Bui T, et al. Cytoplasmic cyclin E and phospho-cyclin-dependent kinase 2 are biomarkers of aggressive breast cancer. Am J Pathol.
2016;186:1900-1912.
52. Karakas C, Francis AM, Ha MJ, et al. Cytoplasmic cyclin E expression predicts for response to neoadjuvant chemotherapy in breast cancer. Ann Surg. Epub 2019 Aug 20.
53. Vijayaraghavan S, Karakas C, Doostan I, et al. CDK4/6 and autophagy inhibitors synergistically induce senescence in Rb positive cytoplasmic cyclin E negative
cancers. Nat Commun. 2017;8:15916.
54. Kettner NM, Vijayaraghavan S, Durak MG, et al. Combined inhibition of STAT3 and DNA repair in palbociclib-resistant ER-positive breast cancer. Clin Cancer Res.
2019;25:3996-4013.
55. Comstock CE, Augello MA, Goodwin JF, et al. Targeting cell cycle and hormone receptor pathways in cancer. Oncogene. 2013;32:5481-5491.

56. Franco J, Witkiewicz AK, Knudsen ES. CDK4/6 inhibitors have potent activity in combination with pathway selective therapeutic agents in models of pancreatic
cancer. Oncotarget. 2014;5:6512-6525.
57. Teh JLF, Cheng PF, Purwin TJ, et al. In Vivo E2F reporting reveals efficacious schedules of MEK1/2-CDK4/6 targeting and mTOR-S6 resistance mechanisms
[published correction appears in Cancer Discov. 2018;8:1654]. Cancer Discov. 2018;8:568-581.
58. Vora SR, Juric D, Kim N, et al. CDK 4/6 inhibitors sensitize PIK3CA mutant breast cancer to PI3K inhibitors. Cancer Cell. 2014;26:136-149.
59. Roberts PJ, Bisi JE, Strum JC, et al. Multiple roles of cyclin-dependent kinase 4/6 inhibitors in cancer therapy. J Natl Cancer Inst. 2012;104:476-487.
60. McClendon AK, Dean JL, Rivadeneira DB, et al. CDK4/6 inhibition antagonizes the cytotoxic response to anthracycline therapy. Cell Cycle. 2012;11:2747-2755.
61. Chou A, Froio D, Nagrial AM, et al.; Australian Pancreatic Cancer Genome Initiative (APGI). Tailored first-line and second-line CDK4-targeting treatment
combinations in mouse models of pancreatic cancer. Gut. 2018;67:2142-2155.
62. Kumarasamy V, Ruiz A, Nambiar R, et al. Chemotherapy impacts on the cellular response to CDK4/6 inhibition: distinct mechanisms of interaction and efficacy in
models of pancreatic cancer. Oncogene. 2019;39:1831-1845.
63. Witkiewicz AK, Cox D, Knudsen ES. CDK4/6 inhibition provides a potent adjunct to Her2-targeted therapies in preclinical breast cancer models. Genes Cancer. 2014;5:261-272.
64. Cao J, Zhu Z, Wang H, et al. Combining CDK4/6 inhibition with taxanes enhances anti-tumor efficacy by sustained impairment of pRB-E2F pathways in squamous
cell lung cancer. Oncogene. 2019;38:4125-4141.
65. Clark AS, McAndrew NP, Troxel A, et al. Combination paclitaxel and palbociclib: results of a phase I trial in advanced breast cancer. Clin Cancer Res. 2019;
25:2072-2079.
66. Chaikovsky AC, Sage J. Beyond the cell cycle: enhancing the immune surveillance of tumors via CDK4/6 inhibition. Mol Cancer Res. 2018;16:1454-1457.
67. Goel S, DeCristo MJ, Watt AC, et al. CDK4/6 inhibition triggers anti-tumour immunity. Nature. 2017;548:471-475.
68. Deng J, Wang ES, Jenkins RW, et al. CDK4/6 inhibition augments antitumor immunity by enhancing T-cell activation. Cancer Discov. 2018;8:216-233.
69. Schaer DA, Beckmann RP, Dempsey JA, et al. The CDK4/6 inhibitor abemaciclib induces a T cell inflamed tumor microenvironment and enhances the efficacy of
PD-L1 checkpoint blockade. Cell Rep. 2018;22:2978-2994.
70. Jerby-Arnon L, Shah P, Cuoco MS, et al. A cancer cell program promotes T cell exclusion and resistance to checkpoint blockade. Cell. 2018;175:984-997.e24.
71. Zhang J, Bu X, Wang H, et al. Cyclin D-CDK4 kinase destabilizes PD-L1 via cullin 3-SPOP to control cancer immune surveillance [published correction appears in
Nature 2019;571:E10]. Nature. 2018;553:91-95.
72. Rugo HS, Kabos P, Dickler MN, et al. Abstract P1-09-01: A phase 1b study of abemaciclib plus pembrolizumab for patients with hormone receptor-positive
(HR+), human epidermal growth factor receptor 2-negative (HER2-) metastatic breast cancer (MBC). Cancer Res. 2018;78:P1-09-01.
73. Franco J, Balaji U, Freinkman E, et al. Metabolic reprogramming of pancreatic cancer mediated by CDK4/6 inhibition elicits unique vulnerabilities. Cell Rep.
2016;14:979-990.
74. Wang H, Nicolay BN, Chick JM, et al. The metabolic function of cyclin D3-CDK6 kinase in cancer cell survival. Nature. 2017;546:426-430.
75. Tarrado-Castellarnau M, de Atauri P, Tarragó-Celada J, et al. De novo MYC addiction as an adaptive response of cancer cells to CDK4/6 inhibition. Mol Syst Biol.
2017;13:940.
76. Booth L, Roberts JL, Rais R, et al. Palbociclib augments neratinib killing of tumor cells that is further enhanced by HDAC inhibition. Cancer Biol Ther. 2019;20:157-168.
77. Yi J, Liu C, Tao Z, et al. MYC status as a determinant of synergistic response to olaparib and palbociclib in ovarian cancer. EBioMedicine. 2019;43:225-237.
78. Arnedos M, Bayar MA, Cheaib B, et al. Modulation of Rb phosphorylation and antiproliferative response to palbociclib: the Preoperative-Palbociclib (POP)
randomized clinical trial. Ann Oncol. 2018;29:1755-1762.
79. Rugo HS, Mayer EL, Storniolo AM, et al. Abstract PD2-12: palbociclib in combination with fulvestrant or tamoxifen as treatment for hormone receptor positive
(HR+) metastatic breast cancer (MBC) with prior chemotherapy for advanced disease (TBCRC 035) A phase II study with pharmacodynamics markers. Cancer
Res. 2019;79:PD2-PD12.
80. Shapiro GI, Hilton J, Gandi L, et al. Abstract CT046: phase I dose escalation study of the CDK4/6 inhibitor palbociclib in combination with the MEK inhibitor PD-
0325901 in patients with RAS mutant solid tumors. Cancer Res. 2017;77:CT046.
81. Grierson JR, Schwartz JL, Muzi M, et al. Metabolism of 30 -deoxy-30 -[F-18]fluorothymidine in proliferating A549 cells: validations for positron emission to-
mography. Nucl Med Biol. 2004;31:829-837.
82. Nisman B, Allweis T, Kaduri L, et al. Serum thymidine kinase 1 activity in breast cancer. Cancer Biomark. 2010;7:65-72.
83. Bjöhle J, Bergqvist J, Gronowitz JS, et al. Serum thymidine kinase activity compared with CA 15-3 in locally advanced and metastatic breast cancer within
a randomized trial. Breast Cancer Res Treat. 2013;139:751-758.
84. Bonechi M, Galardi F, Biagioni C, et al. Plasma thymidine kinase-1 activity predicts outcome in patients with hormone receptor positive and HER2 negative
metastatic breast cancer treated with endocrine therapy. Oncotarget. 2018;9:16389-16399.
85. Bagegni N, Thomas S, Liu N, et al. Serum thymidine kinase 1 activity as a pharmacodynamic marker of cyclin-dependent kinase 4/6 inhibition in patients with
early-stage breast cancer receiving neoadjuvant palbociclib. Breast Cancer Res. 2017;19:123.
86. McCartney A, Bonechi M, De Luca F, et al. Plasma thymidine kinase activity as a biomarker in patients with luminal metastatic breast cancer treated with
palbociclib within the TREnd trial. Clin Cancer Res. Epub 2020 Jan 14.

The Evolution of Antibody-Drug Conjugates: A

Positive Inflexion Point
Anthony W. Tolcher, MD, FRCPC, FACP, FASCO1,2
overview
In 2019, an important inflection point occurred when the U.S. Food and Drug Administration approved three
new antibody-drug conjugates (ADCs) for the treatment of malignancies, including urothelial cancer
(enfortumab vedotin-ejfv), diffuse large B-cell lymphoma (polatuzumab vedotin-piiq), and HER2 breast
cancer (fam-trastuzumab deruxtecan-nxki), and expanded the indication for ado-trastuzumab emtansine to
early breast cancer. This near doubling in the number of approved ADCs within 1 year validates the ADC
platform and represents a successful evolution over the past 30 years. ADCs were born in an era when systemic
therapy for cancer was largely cytotoxic chemotherapy. Many of the investigational cytotoxic agents were
determined to be too toxic for oral and intravenous use. The agents were especially potent, with inhibitory
concentrations that inhibited 50% of cells in the nanomolar and picomolar range but had poor therapeutic
indexes when administered systemically. Now, over the last 30 years, we have seen an evolution of the many
aspects of this complex platform with better antigen target selection, more sophisticated chemistry for the
linkers, a growing diversity of payloads from cytotoxic chemotherapy to targeted therapies and immunosti-
mulants, and, with the recent series of regulatory approvals, a buoyed sense of optimism for the technology.
Nonetheless, we have not fully realized the full potential of this platform. In this review, the many components
of ADCs will be discussed, the difficulties encountered will be highlighted, the innovative strategies that are
being used to improve them will be assessed, and the direction that the field is going will be considered.
A COMPLEX THERAPEUTIC PLATFORM CD19) in which bispecific antibodies and CAR T-cell
ADCs are, arguably, the most complex biochemical therapy have demonstrated that although the target
platform for systemic delivery of antineoplastic medi- antigen is expressed on both malignant and normal
cines. The four key and distinct components of ADCs cells, the ablation of the lineage will not lead to im-
include the target antigen, the antibody construct, the portant long-term complications.1-3 In contrast, for
linker technology for the payload, and the payload solid tumors, most target antigens are tumor-
itself. There are a multitude of additional complexities associated rather than tumor-specific, and hence,
within each component that belie the difficulties en- delivery of the payload to normal tissues is one of the
countered and the limited successes found over the factors that determines the therapeutic index of
years, despite several hundred molecules having en- ADCs, both in preclinical and clinical toxicology.
tered clinical studies. To better understand this, each Therefore, lineage ablation is not an option. This,
component will be presented and the successes, the along with various widely disparate rates of in-
failures, and the opportunities for innovation will be ternalization and antigen processing, antigen recy-
discussed. cling, preferential degradation in the lysosome, and
access of the antibody to the antigen in the vascu-
Author affiliations TARGET ANTIGEN SELECTION lature, all add to difficulties of replicating the pre-
and support
information (if In an ideal situation, the target antigen would only be clinical activity in clinical trials, and the often over-
applicable) appear expressed and of sufficient density on tumor cells, be promising, curative results observed in preclinical
at the end of this rapidly internalized upon ADC binding, and efficiently xenograft experiments.
article.
trafficked to the endosome and the lysosome for re- It is critical to consider that in the patient, the distri-
Accepted on
lease of the payload. Optimally, this would lead to cell bution of the target antigen in tumors, as determined by
February 21, 2020
and published at death in only the cells that express the target antigen, currently available immunohistochemical assays, are
ascopubs.org on reflecting the strategy of “lineage ablation.” often quite different from preclinical models. Most
March 31, 2020:
Unfortunately, few, if any solid tumor antigens meet preclinical host animals do not express the target
10.1200/EDBK_ these criteria. In hematologic malignancies, this may antigen, so the delivery of the ADC to the implanted
281103 be achievable with lineage-specific antigens (e.g., tumor is not confounded by biodistribution of the ADC

Anthony W. Tolcher
predicted from the species used in preclinical toxicology

experiments, as described previously.
• ADCs have become mainstream therapies for If Target Antigens Are Expressed in Normal Tissues, Can
several cancer indications. ADCs Become Prodrugs?
• Complexities arise from the four key compo- Because ADCs were intended to improve the TI, and as
nents, which include the target antigen, the described previously, because most targets are tumor-
antibody, the chemical linker, and the payload. associated antigens rather than tumor-specific antigens,
• Adverse events can be caused by on-target, off- several strategies have been developed to mask the antigen-
tumor binding to normal tissues and also by off- binding portion of the antibody so that selective binding
target, off-tumor biodistribution to normal occurs in tumor cells and not in normal tissues. The basis for
tissues. this is protein or peptide masks that are released by tumor-
• Developmental work seeks to broaden the specific proteases, thereby activating the ADC. These
payloads from cytotoxic agents to targeted and tumor-activated prodrug-like approaches have demon-
immune stimulants. strated improvement in therapeutic indexes in preclinical
models and have been seen in early clinical trials. Some of
these technologies, such as Probody (CytomX; South San
to normal host tissues. Furthermore, in patients, the ex- Francisco, California), Safebody (Adagene; Suzhou, China),
pression of target antigen on the membrane surface of and coiled coil mask (Seattle Genetics; Seattle, Washington)
tumor cells has important characteristics that can affect have entered the clinic and will, no doubt, evolve and im-
delivery and the binding of the ADC. These include relative prove over the coming years.8
membranous and cytosolic expression, the orientation and
polarity of the target antigen, and disparate expression on The Tumor or the Tumor’s Neighborhood?
apical, basal lateral, or circumferential cellular surfaces, An innovative strategy in the development of ADCs is the
depending upon the antigen. Moreover, there exists selection of antigens that are not only expressed by some
a noteworthy heterogeneity of target antigen expression tumors but can be expressed on normal cells within the
among adjacent tumor cells, and this can meaningfully tumor microenvironment and/or tumor stromal cells. This
affect antitumor activity. The assay selection and cutoff approach has already been studied using an ADC targeted
values for scoring potentially sensitive tumor cells is a sci- to LRRC15, a cell surface protein upregulated by trans-
ence unto itself and requires sophisticated input from pa- forming growth factor beta and expressed on stromal cells
thologists experienced in this field before it can be scaled up only in the case of some breast, lung, pancreatic, head,
for clinical trials or validated as a companion diagnostic. For and neck cancers, as well as found on some sarcomas.9,10
most tumor target antigens that are targets for ADCs, the Antitumor activity has been observed using ABBV-085 in
antigen must be present for antitumor activity, but ex- early clinical studies, as a monomethyl auristatin E
pression alone does not predict antitumor activity. This has (MMAE) payload.10
been observed in countless preclinical models and clinical
Alternatively, targeting the tumor microenvironment of
studies. As discussed in the following, the selection of the
“cold tumors” may represent an ideal situation for delivery
payload is inextricably related to the tumor indications that
of payloads that permit trafficking of immune cells to the
express the target antigen.
tumor. Experimental evidence using chemokines, stimu-
Although the basis for ADCs being developed as a platform lator of interferon gene (STING) agonists, and toll-like
was to improve the therapeutic index of toxic payloads, even receptor agonists as payloads suggests this may be pos-
low expression of the target antigen on normal cells has sible. Clinical studies of some of these agents will enter the
important safety considerations that can negatively affect clinic within the year.11-13
the TI. Clinical examples in which target antigen selection
did not have sufficient tumor specificity, and hence, had Failure (Maybe) Is Merely a Step to Success
unexpected on-target, off-tumor toxicity, include the targets As a result of the complexity of this platform, failure of
(and the respective ADC): Lewis Y (BR96 doxorubicin); CA a particular ADC may have little to do with the selection of
IX (BAY794620); and CD44v6 (bivatuzumab mertansine). the target antigen–linker stability, off-target toxicity, and
In each of these aforementioned examples, normal tissue fucosylation of the antibody that results in altered bio-
expression led to hematemesis, fatal gastrointestinal events, distribution. Commercial considerations have had consid-
or fatal exfoliative dermatitis, respectively.4-7 The conse- erable effects on clinical development. Examples of target
quences that resulted from the selection of the target an- antigens that underwent clinical development that failed
tigen should not be underestimated, nor can they always be and then underwent further refinement and an evolution of

Evolution of ADC
the ADC, the antibody, or both include HER2, 5T4, and antibody therapeutics.14 However, for ADCs, the ADCC
NaPi2b. and complement activation are superfluous to the primary
role of the antibody as a delivery vehicle of the payload
HER2 was the first solid tumor ADC approved by regulatory
and may actually reduce the therapeutic index. The dose-
authorities, and despite many attempts to improve this
limiting toxicity of ado-trastuzuamb emtansine was transient
molecule (most of which used antimicrotubule payloads
grade 3 thrombocytopenia. Detailed investigation of this toxicity
and pyrrolbenzodiazepines), it failed to achieve safety and
demonstrated this was, at least in part, a HER2-independent
antitumor results above those achieved by ado-trastuzumab
mechanism mediated by a FcγRIIa-dependent uptake of
emtansine. However, recently, trastuzumab deruxtecan
the maytansine payload by the megakaryocytes. This uptake
demonstrated successful clinical benefit using a topoisom-
resulted in diminished megakaryocyte maturation and di-
erase I inhibitor payload in contrast to the many payloads
minished platelet production.15
used previously. Similarly, XMT1536 (Mersana; Cam-
bridge, Massachusetts), a NaPi2b that targets ADC, has Many cells within the reticuloendothelial system express
shown promising antitumor activity using the fleximer FcγRs and uptake leads to some of the toxicities ob-
linker technology. The fleximer linker technology used in served, including myelosuppression, hyposplenism,
XMT1536 has a greater drug-to-antibody ratio (DAR) of 14 Kupfer cell depletion, and liver toxicities. To reduce the
compared with the usual DAR of 2-4, and despite sharing biodistribution of ADCs to normal cells and tissues that
the same antigen (NaPi2b) and similar payload (auristatin) express FcγRs (neutrophils, macrophages, eosinophils,
as lifastuzumab vedotin (DAR 4), the latter was dis- megakaryocytes, and some B cells), there are currently
continued from clinical development because of a disap- several strategies to lower the level of afucosylation from
pointing rate of clinical benefit. the manufacture of antibodies from CHO cells (e.g., in-
crease fucose at the Fc region), and thereby reduce the
To conclude, the target antigen and the many complexities
affinity of IgG1 binding to members of the FcγR family to
involved with expression, selectivity, polarity and orientation,
improve the therapeutic index. This increased custom-
and internalization are important both for antitumor activity
ization of the antibody for ADC design should be en-
and tolerability, but by itself, it does not guarantee success.
couraged as the field evolves.
THE ANTIBODY: ONE SIZE DOES NOT FIT ALL The rate of internalization of the antibody upon binding of
The most common monoclonal antibodies used in ADC the ADC to the target antigen is also a key component for
development belong to the immunoglobulin (Ig)-G1 subtype activity. The internalization kinetics are critical to target
and are based on the experience of many decades of selection. Slow or no internalization rates will affect the
successful naked (non-ADC) therapeutic antibody devel- potency of ADCs, and these antigen targets are best suited
opment. However, it remains to be determined whether, at for radio-immunoconjugates in which proximity to the path
this time, we have yet optimized the antibody component for of radioactive decay, rather than internalization, is a re-
ADC delivery, and whether we have sufficiently explored this quirement. There are current developmental strategies to
aspect. enhance internalization of antibodies bound to the target
membrane antigen. This includes the polyclonal antibody
A Delivery Vehicle That Needs to Be Optimized for ADCs
approach from Symphogen (Copenhagen, Denmark) and
Most of the existing therapeutic IgG1 antibodies used in the use of biparatopic antibodies from Zymeworks (Van-
cancer medicine have been designed to have antibody- couver, BC, Canada). This latter technology uses a bispe-
dependent cellular cytotoxicity (ADCC). Human IgG1 has cific antibody that binds to one arm (a single chain variable
a glycoprotein bearing two N-linked biantennary complex fragment to extracellular domain 4), whereas another Fab
oligosaccharides bound to the antibody constant region fragment binds to extracellular domain 2. Preclinical evi-
(Fc), in which most of the oligosaccharides are core dence demonstrates that this biparatopic approach leads to
fucosylated. Fucosylation creates steric hinderance that greater internalization of the antibody antigen complex. The
reduces binding to the Fc gamma receptor family (FcγRs). strategy for ADC is to use this molecule ZW25 with an
The Fc region creates the desired effector function of ADCC auristatin payload. This agent, ZW49, entered the clinic
and complement-mediated cytotoxicity through the in- in 2019.
teraction of the Fc of the antibody with the FcγRs and
mediates, at least in part, the anticancer activity observed Location Matters: Site-Specific Conjugation to the
with rituximab, trastuzuamb, cetuximab, and other ap- Antibody and Steric Hinderance
proved cancer therapeutic antibodies. Bioengineering of Experimental evidence suggests that site-specific conju-
Chinese hamster ovary (CHO) cells to reduce fucosylation gation of the payload to the antibody can improve the
(afucosylation) in the manufacturing of therapeutic anti- stability of the ADC, improve the pharmacokinetics, and
bodies has been used to enhance ADCC activity of “naked” perhaps improve upon tolerability in preclinical models. In

Anthony W. Tolcher
truth, early ADCs often had a spectrum of DAR with a me- preclinical and early clinical studies, with measurements of
dian and/or mean of three to four. However, the outliers on all three components intact: conjugate, naked antibody, and
either side of the range have been associated, in part, with payload, if possible. Rapid loss of the payload with elimi-
either pharmacologic instability and/or toxicity.16 To address nation of the ADC compared with the naked antibody should
this, a number of site-specific conjugation bioengineering lead to re-engineering of the linker and stop advancement to
techniques have been developed to create ADC with fixed the clinic.
sites for the linker payload as well as ensuring a more The design and extent of the cytotoxic payload retention
homogenous DAR in the preparation.17-19 Many of the newer within the tumor cell after internalization of the ADC is an
ADCs entering the clinic use many methodologies to ac- important but controversial discussion. Two themes exist: 1)
complish this, and time will tell whether this improves increase intracellular retention to permit greater cytotoxicity
clinical outcome. in antigen-bearing cells; or 2) accept efflux for the bystander
Beyond Antibodies effects on adjacent cells that may or may not express the
The typical molecular weight of an IgG antibody is ap- target antigen. Merits to both strategies are obvious, but the
proximately 150 kDa. This size, and the natural forces of apparent and overriding risk of designing an ADC with drug
oncotic pressure within a tumor, have the potential to limit efflux is the loss of potency in those antigen-expressing
antibody penetration into a tumor, and thereby reduce the tumor cells.27 Furthermore, it is not clear that the bystander
delivery of the payload. If we accept that ADCs are fun- effect of drug efflux reaches sufficient payload concentra-
damentally a tumor-antigen targeting delivery system, one tions in the adjacent extracellular space to be pharmaco-
can then move beyond antibodies if novel tumor antigen logically relevant.28-34 Newer proprietary linker technologies
targeting is possible. Considerable work is being pursued to seek to modify the linker to improve retention of the payload
be “antibody killers” in this field. Two new protein platforms within the cell after uptake. Following lysosomal degrada-
are in early development and include the centyrins (Aro tion, the polymer-payload complex retains hydrophilic
Biotherapeutics, Philadelphia, Pennsylvania) and Alphalex properties that reduce transit of the payload out of the tumor
(Cybrexa Therapeutics; New Haven, Connecticut) plat- cell while maintaining the cytotoxic potential. Although this
forms. These smaller molecular weight proteins can be may reduce bystander effects, the concept of one-way
linked to payloads and specifically targeted to cell surface delivery is intriguing.
antigens. The pharmacologic properties can be enhanced, THE PAYLOAD EVOLVES (AT LAST!)
at least in the case of the centyrins linked to albumin, to
ensure prolonged exposure. Preclinical evidence suggests In past reviews, I proposed a principle that the target be
that upon binding to the target antigen, these smaller protein matched to the indication and to match the indication to the
drug conjugates may be more readily taken into the early payload.35 This was based on the finding that most ADCs
endosome and trafficked to the cytosol compared with had payloads that had antimicrotubule agents with a mi-
antibodies. Time will tell whether these approaches succeed crotubule destabilizing mechanism of action. Historically,
antibody-based approaches, but early preclinical data are these antimicrotubule destabilizing agents had a limited
encouraging. spectrum of activity as single agents when given as sys-
temically for cancer. Despite more than 100 clinical trials,
THE LINKER: MORE THAN JUST CHEMISTRY few of these antimicrotubule ADCs had robust single-agent
Perhaps the most complex aspect of all ADC development is activity, with only one at the time meeting regulatory
the chemistry of the linker and the disposition of the payload approval.36-42 The end result was ample evidence of dis-
upon internalization. In an ideal situation, the linker is strong appointing antitumor activity, with few if any complete re-
enough to withstand degradation in the plasma and in- sponses in solid tumors—this despite promising preclinical
terstitial space, and yet vulnerable enough for degradation data, including cures in tumor-bearing animals with these
and release of the payload in the endosome/lysosome. antimicrotubule payloads.35,43
There are several excellent reviews of the complexity of The spectrum of cytotoxic payloads beyond antimicrotubule
linker chemistry.20-24 agents has finally broadened beyond maytansines and
The consequence of an unstable linker is obvious. The auristatins. With the approval of trastuzumab deruxtecan,
cytotoxic comes off the ADC and the toxicity of the free a highly potent topoisomerase inhibitor that could not be
payload may be observed. This was a concern early in the easily delivered intravenously, the field has evolved to other
course of ADC development, and with cantuzumab mer- classes of cytotoxic agents, and the movement to re-examine
tansine, neuropathy from the DM1 payload was evident.25,26 what constitutes a payload has occurred.44,45 This allows the
Conversely, stable covalent linkers may not release suffi- 5 decades of experience with cytotoxic chemotherapy and
cient payload and be relatively inert in vivo. Validation of the permits to potentially enlist once abandoned cytotoxic drugs,
chemistry should occur in the pharmacokinetic analysis in like exatecan.44 Approved cytotoxic chemotherapy payloads

Evolution of ADC
TABLE 1. Currently Approved Antibody-Drug Conjugates With Target and Payload

Name Target Indication(s) Payload Year FDA First Approved
Gemtuzumab ozogamicine CD33 AML Calicheamicin 2000
Ado-trastuzumab emtansine HER2 HER2+ MBC Maytansine 2013
HER2+ adjuvant
Brentuximab vedotin CD30 Hodgkin lymphoma Auristatin MMAE 2011
Anaplastic large cell lymphoma
First-line Hodgkin disease
Mycosis fungoides
Inotuzumab ozogamicin CD22 B-cell lymphoma Calicheamicin 2017
Polatuzumab vedotin CD79b Refractory DLBCL Auristatin MMAE 2019
Enfortumab vedotin Nectin-4 Refractory urothelial Auristatin MMAE 2019
Trastuzumab deruxtecan HER2 Refractory HER2 MBC Exatecan 2019
Abbreviations: FDA, U.S. Food and Drug Administration; AML, acute myeloid leukemia; MBC, metastatic breast cancer; MMAE, monomethyl auristantin E;
DLBCL, diffuse large B-cell lymphoma.
that span many classes and targets are summarized in systemic payload degradation. Pyrrolbenzodiazepines are
Table 1.46-52 perhaps the most potent of the current cytotoxic payloads.53
However, rovalpituzumab tesirine, a promising ADC that
However, one downside of the strategy to select ever more
targeted DLL3, had cumulative thrombocytopenia that both
potent cytotoxic payloads is the potential for toxicities be-
decreased the frequency of administration and the number
cause of even a small amount of free payload. This idea, that
of cycles that could be administered safely, which might
potency can be increased to the point that a single molecule
have contributed to the failure of this agent.54,55 Perhaps the
delivered will kill the tumor cell, or “one shot one kill," is
next evolution of payloads should be focused on selecting
attractive in theory but has hazards from the aforementioned
less potent but more targeted agents designed for specific
indications.
TABLE 2. Novel Non-Antimicrotubule Payloads by Mechanism of Action
Targeted Payloads to Match ADC Targeted Delivery
Mechanism of
Payload Action Reference If the initial premise of ADC delivery was to deliver agents
Pyrrolbenzodiazepine DNA minor groove Mantaj et al, 53 that had potent activity but could not be delivered sys-
binding Rudin et al, 54 temically because of the therapeutic index, then the
Morgensztem spectrum of potential payloads could be large. As dem-
et al55 onstrated in clinical trials, many targeted agents (e.g., ty-
49
Indolinobenzodiazepine Minor groove/ Singh et al rosine kinase inhibitors and apoptotic pathway targeting
alkylating agent agents) and immune stimulants (e.g., toll-like receptor
SN38 Topoisomerase 1 Starodub et al50 agonists and STING agonists) can have low therapeutic
inhibitor indexes. ADC delivery may therefore be an opportunity to
Exatecan Topoisomerase 1 Modi et al45 deliver these payloads to the tumor or tumor microenvi-
inhibitor ronment with a reduction of either off-target or on-target off-
Duocarmycin DNA minor groove Black et al, 46 tumor toxicity.
binding Dokter et al, 47
Elgersma et al48 Many targeted therapies have systemic and normal tissue
52
adverse events that limit the dose that can be administered.
PNU (nemrubicin DNA binding Stefan et al
analog) The therapeutic index issue is at the heart of the value
proposition of cytotoxic ADCs and can be, by inference,
BCL-Xl inhibitor Apoptotic pathway NCT0359505956
translated to antitumor effectiveness of targeted agents. A
Chemokine Inflammatory cell Mohit et al11 targeted agent as a payload has recently entered the clinic in
invasion
early phase I clinical studies. ABBV-155 is a monoclonal
STING agonist Immune stimulant Centinbas et al12 antibody directed to B7H3 with an apoptotic pathway tar-
Toll-like receptor agonist Immune stimulant Moyes et al13 geting agent.56
(innate)
Furthermore, one can speculate that the failure of MEK
Abbreviation: STING, stimulator of interferon genes. and ERK inhibitors in KRAS tumors was due, at least in part,

Anthony W. Tolcher
to the inability to deliver sufficient drug concentrations tumor cell or to antigens on stromal cells in the tumor
selectively to the tumor before normal tissue toxicity became microenvironment. This strategy to turn tumors that are
dose limiting. Also, some combinations with profound considered immunologically “cold” into “hot” may be viable
synergistic activity in preclinical models were not feasible in using ADCs in a field where either systemic administration or
the clinical setting because of synergistic normal tissue intratumoral injections have, so far, been disappointing.
toxicity. A good example is that the combination of MEK and
A select number of these promising payloads that have been
AKT/PI3K inhibitors in clinical trials was disappointing as
approved or are entering clinical trials are either targeted
a result of normal tissue toxicity and the need for reduced
therapies or immune stimulants and are summarized in
doses of both agents.57-59 This raises an interesting op-
Table 2.
portunity for the ADC platform to deliver one or both of these
agents specifically to the tumor with reduced toxicity and
enhanced tumor activity. THE PLATFORM HAS EVOLVED
Immune-based therapy has been added as a major and This is a time for optimism. ADCs, long viewed by many as
widely used component of cancer therapy. Several lines of not living up to their promise, have finally become a stan-
evidence suggest the antitumor activity of ADC therapy is, in dard of treatment of several malignancies. Because the path
part, attributable to immune-mediated cell death.60 Further to curative therapy has been the combination of multiple
overlap of ADC technology with immunotherapy comes from agents with nonoverlapping toxicities and mechanism of
the addition of immunostimulant payloads such as che- action, ADCs could be pushed to the forefront of treatment
mokines, STING, and toll-like receptor agonists to ADCs.12,13 lines, and certainly, combinations of ADCs are possible.
As mentioned previously, many of these agents cannot be Recent success will surely accelerate the science, improve
safely administered systemically, but by using a delivery the bioengineering, and benefit our patients in need of
vehicle, they can be delivered to either antigens on the innovative therapy.

1
New Experimental Therapeutics (NEXT), San Antonio, TX AND DATA AVAILABILITY STATEMENT
2
Texas Oncology, San Antonio, TX Disclosures provided by the authors and data availability statement (if
EDBK_281103.
Anthony W. Tolcher, MD, 2829 Babcock Rd., Suite 300, San Antonio, TX
78229; email: [email protected].
REFERENCES
1. Przepiorka D, Ko CW, Deisseroth A, et al. FDA approval: blinatumomab. Clin Cancer Res. 2015;21:4035-4039.
2. von Stackelberg A, Locatelli F, Zugmaier G, et al. Phase I/phase II study of blinatumomab in pediatric patients with relapsed/refractory acute lymphoblastic
leukemia. J Clin Oncol. 2016;34:4381-4389.
3. Braendstrup P, Levine BL, Ruella M. The long road to the first FDA-approved gene therapy: chimeric antigen receptor T cells targeting CD19. Cytotherapy. 2020;
22:57-69.
4. Bayer Health Care. Clinical Study Report. PH-37705/12671. http://trialfinder.bayerscheringpharma.de/html/pdf/12671_Study_Synopsis_CTP.pdf. Accessed on
June 16, 2016.
5. Saleh MN, Sugarman S, Murray J, et al. Phase I trial of the anti-Lewis Y drug immunoconjugate BR96-doxorubicin in patients with Lewis Y-expressing epithelial
tumors. J Clin Oncol. 2000;18:2282-2292.
6. Tolcher AW, Sugarman S, Gelmon KA, et al. Randomized phase II study of BR96-doxorubicin conjugate in patients with metastatic breast cancer. J Clin Oncol.
1999;17:478-484.
7. Tijink BM, Buter J, de Bree R, et al. A phase I dose escalation study with anti-CD44v6 bivatuzumab mertansine in patients with incurable squamous cell
carcinoma of the head and neck or esophagus. Clin Cancer Res. 2006;12:6064-6072.
8. Desnoyers LR, Vasiljeva O, Richardson JH, et al. Tumor-specific activation of an EGFR-targeting probody enhances therapeutic index. Sci Transl Med. 2013;
5:207ra144.
9. Purcell JW, Tanlimco SG, Hickson J, et al. LRRC15 is a novel mesenchymal protein and stromal target for antibody-drug conjugates. Cancer Res. 2018;
78:4059-4072.
10. Demetri GD, Luke JJ, Hollebecque A, et al. First-in-human phase 1 study of ABBV-085, an antibody-drug conjugate (ADC) targeting LRRC15, in sarcomas and
other advanced solid tumors. J Clin Oncol. 2019;37:3004.

Evolution of ADC
11. Mohit E, Rafati S. Chemokine-based immunotherapy: delivery systems and combination therapies. Immunotherapy. 2012;4:807-840.
12. Centinbas NC, Catcott KC, Avocetien K, et al. Tumor targeting of a STING agonist with an antibody-drug conjugate elicits potent antitumor responses. Presented
at: SITC. National Harbor, MD; 2019. Abstract 695.
13. Moyes K, Brender T, Smith SW, et al. Abstract 3271: a systemically administered, conditionally active TLR8 agonist for the treatment of HER2-expressing tumors.
Cancer Res. 2019;79 (13 Suppl):Abstract nr 3271.
14. Mori K, Iida S, Yamane-Ohnuki N, et al. Non-fucosylated therapeutic antibodies: the next generation of therapeutic antibodies. Cytotechnology. 2007;
55:109-114.
15. Uppal H, Doudement E, Mahapatra K, et al. Potential mechanisms for thrombocytopenia development with trastuzumab emtansine (T-DM1). Clin Cancer Res.
2015;21:123-133.
16. Strop P, Delaria K, Foletti D, et al. Site-specific conjugation improves therapeutic index of antibody drug conjugates with high drug loading. Nat Biotechnol. 2015;
33:694-696.
17. Axup JY, Bajjuri KM, Ritland M, et al. Synthesis of site-specific antibody-drug conjugates using unnatural amino acids. Proc Natl Acad Sci USA. 2012;
109:16101-16106.
18. Beerli RR, Hell T, Merkel AS, et al. Sortase enzyme-mediated generation of site-specifically conjugated antibody drug conjugates with high in vitro and in vivo
potency. PLoS One. 2015;10:e0131177.
19. Zhou Q, Stefano JE, Manning C, et al. Site-specific antibody-drug conjugation through glycoengineering. Bioconjug Chem. 2014;25:510-520.
21. Jain N, Smith SW, Ghone S, et al. Current ADC linker chemistry. Pharm Res. 2015;32:3526-3540.
22. Panowski S, Bhakta S, Raab H, et al. Site-specific antibody drug conjugates for cancer therapy. MAbs. 2014;6:34-45.
23. Ross PL, Wolfe JL. Physical and chemical stability of antibody drug conjugates: current status. J Pharm Sci. 2016;105:391-397.
24. Tsuchikama K, An Z. Antibody-drug conjugates: recent advances in conjugation and linker chemistries. Protein Cell. 2018;9:33-46.
25. Tolcher AW, Ochoa L, Hammond LA, et al. Cantuzumab mertansine, a maytansinoid immunoconjugate directed to the CanAg antigen: a phase I, pharma-
cokinetic, and biologic correlative study. J Clin Oncol. 2003;21:211-222.
26. Rodon J, Garrison M, Hammond LA, et al. Cantuzumab mertansine in a three-times a week schedule: a phase I and pharmacokinetic study. Cancer Chemother
Pharmacol. 2008;62:911-919.
27. Zhang D, Yu SF, Khojasteh SC, et al. Intratumoral payload concentration correlates with the activity of antibody-drug conjugates. Mol Cancer Ther. 2018;
17:677-685.
28. Byun JH, Jung IH. Modeling to capture bystander-killing effect by released payload in target positive tumor cells. BMC Cancer. 2019;19:194.
29. Li F, Emmerton KK, Jonas M, et al. Intracellular released payload influences potency and bystander-killing effects of antibody-drug conjugates in preclinical
models. Cancer Res. 2016;76:2710-2719.
30. Singh AP, Shah DK. Measurement and mathematical characterization of cell-level pharmacokinetics of antibody-drug conjugates: a case study with trastuzumab-
vc-MMAE. Drug Metab Dispos. 2017;45:1120-1132.
31. Singh AP, Shah DKA. A “dual” cell-level systems PK-PD model to characterize the bystander effect of ADC. J Pharm Sci. 2019;108:2465-2475.
32. Singh AP, Sharma S, Shah DK. Quantitative characterization of in vitro bystander effect of antibody-drug conjugates. J Pharmacokinet Pharmacodyn. 2016;
43:567-582.
33. Staudacher AH, Brown MP. Antibody drug conjugates and bystander killing: is antigen-dependent internalisation required? Br J Cancer. 2017;117:1736-1742.
34. Vasalou C, Helmlinger G, Gomes B. A mechanistic tumor penetration model to guide antibody drug conjugate design. PLoS One. 2015;10:e0118977.
36. Amiri-Kordestani L, Blumenthal GM, Xu QC, et al. FDA approval: ado-trastuzumab emtansine for the treatment of patients with HER2-positive metastatic breast
cancer. Clin Cancer Res. 2014;20:4436-4441.
37. Krop IE, Kim SB, Martin AG, et al. Trastuzumab emtansine versus treatment of physician’s choice in patients with previously treated HER2-positive metastatic
breast cancer (TH3RESA): final overall survival results from a randomised open-label phase 3 trial. Lancet Oncol. 2017;18:743-754.
38. Krop IE, LoRusso P, Miller KD, et al. A phase II study of trastuzumab emtansine in patients with human epidermal growth factor receptor 2-positive
metastatic breast cancer who were previously treated with trastuzumab, lapatinib, an anthracycline, a taxane, and capecitabine. J Clin Oncol. 2012;
30:3234-3241.
39. Cabanillas F, Bodey GP, Burgess MA, et al. Results of a phase II study of maytansine in patients with breast carcinoma and melanoma. Cancer Treat Rep. 1979;
63:507-509.
40. Cabanillas F, Rodriguez V, Hall SW, et al. Phase I study of maytansine using a 3-day schedule. Cancer Treat Rep. 1978;62:425-428.
41. Eagan RT, Creagan ET, Ingle JN, et al. Phase II evaluation of maytansine in patients with metastatic lung cancer. Cancer Treat Rep. 1978;62:1577-1579.
42. Eagan RT, Ingle JN, Rubin J, et al. Early clinical study of an intermittent schedule for maytansine (NSC-153858): brief communication. J Natl Cancer Inst. 1978;
60:93-96.
43. Ricart AD, Tolcher AW. Technology insight: cytotoxic drug immunoconjugates for cancer therapy. Nat Clin Pract Oncol. 2007;4:245-255.

Anthony W. Tolcher
44. Garrison MA, Hammond LA, Geyer CE Jr., et al. A phase I and pharmocokinetic study of exatecan mesylate administered as a protracted 21-day infusion in
patients with advanced solid malignancies. Clin Cancer Res. 2003;9:2527-2537.
45. Modi S, Saura C, Yamashita T, et al. Trastuzumab deruxtecan in previously treated HER2-positive breast cancer. N Engl J Med. 2020;382:610-621.
46. Black J, Menderes G, Bellone S, et al. SYD985, a novel duocarmycin-based HER2-targeting antibody-drug conjugate, shows antitumor activity in uterine serous
carcinoma with HER2/Neu expression. Mol Cancer Ther. 2016;15:1900-1909.
47. Dokter W, Ubink R, van der Lee M, et al. Preclinical profile of the HER2-targeting ADC SYD983/SYD985: introduction of a new duocarmycin-based linker-drug
platform. Mol Cancer Ther. 2014;13:2618-2629.
48. Elgersma RC, Coumans RG, Huijbregts T, et al. Design, synthesis, and evaluation of linker-duocarmycin payloads: toward selection of HER2-targeting antibody-
drug conjugate SYD985. Mol Pharm. 2015;12:1813-1835.
49. Singh R, Reid EE, Harris L, et al. Antibody-drug conjugates with indolinobenzodiazepine dimer payloads: DNA-binding mechanism of indolinobenzodiazepine
dimer catabolites in target cancer cells. Mol Pharm. 2020;17:50-58.
50. Starodub AN, Ocean AJ, Shah MA, et al. First-in-human trial of a novel anti-trop-2 antibody-SN-38 conjugate, sacituzumab govitecan, for the treatment of diverse
metastatic solid tumors. Clin Cancer Res. 2015;21:3870-3878.
51. Hellmann I, Waldmeier L, Bannwarth-Escher MC, et al. Novel antibody drug conjugates targeting tumor-associated receptor tyrosine kinase ROR2 by functional
screening of fully human antibody libraries using transpo-mAb display on progenitor B cells. Front Immunol. 2018;9:2490.
52. Stefan N, Gébleux R, Waldmeier L, et al. Highly potent, anthracycline-based antibody-drug conjugates generated by enzymatic, site-specific conjugation. Mol
Cancer Ther. 2017;16:879-892.
53. Mantaj J, Jackson PJ, Rahman KM, et al. From anthramycin to pyrrolobenzodiazepine (PBD)-containing antibody-drug conjugates (ADCs). Angew Chem Int Ed
Engl. 2017;56:462-488.
54. Rudin CM, Pietanza MC, Bauer TM, et al; SCRX16-001 Investigators. Rovalpituzumab tesirine, a DLL3-targeted antibody-drug conjugate, in recurrent small-cell
lung cancer: a first-in-human, first-in-class, open-label, phase 1 study. Lancet Oncol. 2017;18:42-51.
55. Morgensztern D, Besse B, Greillier L, et al. Efficacy and safety of rovalpituzumab tesirine in third-line and beyond patients with DLL3-expressing, relapsed/
refractory small-cell lung cancer: results from the phase II TRINITY study. Clin Cancer Res. 2019;25:6958-6966.
56. NCT03595059. A phase I study of ABBV 155 alone or in combination with taxane therapy in adults with relapsed and/or refractory solid tumors. www.clinicaltrials.
gov/ct2/show/NCT03595059. Accessed February 10, 2020.
57. Tolcher AW, Patnaik A, Papadopoulos KP, et al. Phase I study of the MEK inhibitor trametinib in combination with the AKT inhibitor afuresertib in patients with
solid tumors and multiple myeloma. Cancer Chemother Pharmacol. 2015;75:183-189.
58. Yap TA, Yan L, Patnaik A, et al. Interrogating two schedules of the AKT inhibitor MK-2206 in patients with advanced solid tumors incorporating novel
pharmacodynamic and functional imaging biomarkers. Clin Cancer Res. 2014;20:5672-5685.
59. Tolcher AW, Bendell JC, Papadopoulos KP, et al. A phase IB trial of the oral MEK inhibitor trametinib (GSK1120212) in combination with everolimus in patients
with advanced solid tumors. Ann Oncol. 2015;26:58-64.
60. Gerber HP, Sapra P, Loganzo F, et al. Combining antibody-drug conjugates and immune-mediated cancer therapy: what to expect? Biochem Pharmacol. 2016;
102:1-6.

The Resurgence of Antibody Drug Conjugates in

Cancer Therapeutics: Novel Targets and Payloads
Valentina Boni, MD, PhD1,2; Manish R. Sharma, MD3; and Amita Patnaik, MD, FRCPC4
overview
Antibody drug conjugates (ADCs) are an emerging class of therapeutics that consist of a cytotoxic agent linked
covalently to an antibody, which is directed toward a specific cell surface target expressed by tumor cells and/
or the microenvironment. ADCs leverage the specificity of the antibody such that it functions as a carrier to
deliver the cytotoxic payload into the tumor. Four parameters are considered critical for this class of complex
engineered therapeutics: target selection, antibody, cytotoxic payload, as well as conjugation and linker
technology. The development of this class of drugs has proven more complex than expected. Several
challenges have arisen, including a lack of true tumor antigen specificity, early release of the cytotoxic payload
into the bloodstream due to linker instability, and low potency of the payload, resulting in either greater toxicity
or lack of improved efficacy compared with unconjugated cytotoxics. The approval of trastuzumab emtansine
in 2013 for HER2-positive breast cancer served as a proof of concept that ADCs have therapeutic application
in solid tumors. Two novel ADCs have recently been approved: trastuzumab deruxtecan for HER2-positive
breast cancer and enfortumab vedotin for locally advanced or metastatic urothelial cancer. Trastuzumab
deruxtecan is distinguished by a unique biochemical structure with a novel cytotoxic payload, deruxtecan—a
highly potent, topoisomerase I inhibitor. Enfortumab vedotin is directed toward nectin-4 and represents an
example of successful and strategic target selection. This review focuses on the concepts underlying the
choice of suitable targets and novel payloads, discusses specific examples of ADCs in preclinical and clinical
development, and provides future directions related to this unique class of therapeutics.
INTRODUCTION trial.2 Although it was voluntarily withdrawn from the

Chemotherapy is limited by a narrow therapeutic index market in 2010 after concerns regarding its safety and
because the dose necessary to kill cancer cells is also efficacy were raised by subsequent trials, it was ap-
toxic to normal cells. ADCs are a class of therapeutics proved again in 2017 after new data emerged sup-
in oncology that seek to increase the therapeutic index porting a lower dose and “fractionated” schedule.3
by “targeting” cancer cells with monoclonal antibodies Since 2011, seven additional ADCs have been gran-
(mAbs) that bind to antigens on cancer cells and are ted FDA approval, including three with accelerated
conjugated by chemical linkers to a cytotoxic payload. approval in 2019. Table 1 summarizes the eight ADCs
This strategy allows specific delivery of the cytotoxic that have garnered FDA approval to date.2-11
agent to the tumor site while minimizing the exposure One strategy for ADCs to overcome resistance to
to normal tissues. The primary mechanism of action is chemotherapy and/or biologic therapy is by selecting
internalization of the ADC, with release of the payload novel targets. The targets of ADCs have conventionally
into the cytoplasm after cleavage of the linker.1 A been antigens that have high levels of expression on
secondary mechanism of action is the “bystander ef- cancer cells and low or absent expression on normal
Author affiliations
and support
fect,” whereby the payload or payload-linker perme- cells. A classic example is HER2, which is highly
information (if ates nearby cells, including those that may not express expressed in many breast cancers but minimally
applicable) appear the target antigen.1 expressed in normal tissues. The approval of trastu-
at the end of this
article. Gemtuzumab ozogamicin, an anti-CD33 targeting zumab emtansine in 2013 was highly noteworthy
Accepted on March ADC, was the first of its class to receive approval by the because it demonstrated efficacy in patients with
20, 2020 and U.S. Food and Drug Administration (FDA) in 2000 for HER2-positive breast cancer refractory to trastuzumab,
published at the treatment of CD33-expressing acute myeloid leu- establishing for the first time the proof of concept that
ascopubs.org on kemia. Conjugated to calicheamicin, a naturally oc- an ADC may be efficacious even after the antibody
April 21, 2020:
curring antitumor antibiotic causing double-strand itself has ceased to provide benefit.6 A more recent
10.1200/EDBK_ DNA breaks, gemtuzumab ozogomicin was granted example is enfortumab vedotin, which has a novel
281107 accelerated approval on the basis of a single pivotal target called nectin-4.10 The differential expression of

Novel Targets and Payloads in Antibody Drug Conjugates
reported that approximately 80 ADCs are in clinical de-

velopment, with nearly 600 trials ongoing.13 In this review,
we focus on novel targets and novel payloads that are being
• There are now eight FDA-approved ADCs for the used in ADCs that are in both preclinical and clinical
treatment of cancer, including three with
development.
accelerated approval within the last year.
• Recently approved ADCs, as well as many of NOVEL TARGETS FOR ANTIBODY DRUG CONJUGATES IN
those in development, seek to overcome re- SOLID TUMORS
sistance and increase the therapeutic index by Target selection is one of the most critical issues in ADC
using novel targets and/or novel payloads. development. To facilitate intracellular delivery of the cy-
• Novel targets for ADCs include, for example, totoxic payload, an ideal target should be highly, homo-
those in the TME and those that are involved in geneously, and exclusively expressed on the surface of
immune evasion. tumor cells; have minimal or no antigen shedding within the
• Novel payloads for ADCs include, for example, circulation to prevent off-tumor antibody binding; and have
topoisomerase inhibitors and those that en- internalization capability to facilitate intracellular delivery of
hance the antitumor immune response. the cytotoxic payload.14 The threshold for antigen expres-
sion on tumor cells is not well established given the pres-
ence of many other critical factors, including internalization
nectin-4, also known as poliovirus receptor–related protein rate, antibody binding affinity, cytotoxic payload potency,
4, in bladder cancer specimens compared with healthy and linker stability.15 The selectivity of the target and its
tissues led to its development as a target for enfortumab relative distribution in tumor versus healthy tissues are key
vedotin in urothelial cancer.12 Many other antigens over- factors in determining activity and toxicity, which play
expressed on cancer cells are being targeted by ADCs in a major role in defining therapeutic index of the ADC.16
development, with notable examples including EGFR,
Antigens demonstrating high expression within tumor and
HER3, and TROP-2. In contrast to ADCs targeting antigens
low or absent expression in healthy tissues represent an
on cancer cells, many of the ADCs in development are
ideal target for optimal drug delivery while avoiding toxic-
targeting antigens in the tumor microenvironment (TME),
ities.14 Most commonly, however, tumor antigens are not
including the stroma, the vasculature, and immune cells.
specific. They may be highly expressed on the tumor but
Notable examples of such targets include CD25 (on regu-
also present in healthy tissues (tumor-associated antigens),
latory T cells) and B7-H3 (on tumor cells and tumor vascular
resulting in off-tumor distribution of the ADC and not only
endothelial cells).
the occurrence of on-target off-tumor toxicities but also
Another strategy for ADCs to overcome resistance to che- reduced efficacy due to a lower amount of intratumoral drug
motherapy and/or biologic therapy is by using novel pay- delivery.17-19 A low or moderate expression of the target
loads. The payloads of ADCs have conventionally been antigen in the tumor requires high payload potency to be
cytotoxic drugs that either damage DNA or block mitosis by active.20,21 Antigens, which are ubiquitously expressed in
inhibiting microtubule polymerization/depolymerization. normal epithelium and also overexpressed in carcinomas,
Two recently approved therapies have used unique pay- have not previously been considered as a viable target
loads that differ from those used historically. Trastuzumab for drug development because of toxicities related to off-
deruxtecan, bearing a topoisomerase I inhibitor as a pay- tumor ADC distribution. To expand the universe of targets
load, remarkably demonstrated a 61% response rate and previously considered “undruggable” due to their presence
a median response duration of 14.8 months in patients with on healthy tissues, new engineered antibodies are cur-
HER2-positive breast cancer previously treated with tras- rently under development, including probody drug
tuzumab emtansine.11 Because the target (HER2) is conjugates,22,23 which are designed with a masked antigen
identical to trastuzumab emtansine, the efficacy of trastu- binding site and are preferentially activated by TME pro-
zumab deruxtecan is attributable to the novel payload. teases removing their masks and activating them to bind
Moxetumumab pasudotox, having a payload consisting of tumor target with minimal binding in the inactive/masked
Pseudomonas exotoxin A, demonstrated a high rate of state in nonmalignant tissue.24
durable complete responses in refractory hairy cell leuke-
The majority of the selected targets for ADCs in development
mia.8 Other novel payloads currently in development in-
are localized on the surface of cancer cells and/or cancer
clude novel DNA damaging agents, Amanitins, inducers of
stem cells. 25 Targeting the TME, including antigens
apoptosis, and immunomodulatory agents.
expressed on endothelial cells, fibroblasts, and immune
ADCs represent a rapidly evolving area of drug develop- cells, remains a novel and interesting approach.14,25,26 The
ment and hold significant promise. A recent review types of malignancies covered by target expression, the

Boni, Sharma, and Patnaik
TABLE 1. FDA-Approved Antibody Drug Conjugates for the Treatment of Cancer (as of March 15, 2020)
Year
Name Approved Indication Target Payload (Mechanism) Refs
2,3
Gemtuzumab 2000 and Acute myeloid leukemia CD33 Calicheamicin (DNA cleavage)
ozogamicin 2017
4,5
Brentuximab vedotin 2011 Hodgkin lymphoma, anaplastic large cell lymphoma CD30 Auristatin (microtubule inhibitor)
6
Trastuzumab 2013 HER2-positive breast cancer (after prior anti-HER2 HER2 Maytansine (microtubule inhibitor)
emtansine therapy)
7
Inotuzumab 2017 B-cell precursor acute lymphoblastic leukemia CD22 Calicheamicin (DNA cleavage)
ozogamicin
Moxetumomab 2018 Hairy cell leukemia CD22 Pseudomonas exotoxin A (induction of 8
pasudotox apoptosis)
9
Polatuzumab vedotin* 2019 Diffuse large B-cell lymphoma CD79b Auristatin (microtubule inhibitor)
10
Enfortumab vedotin* 2019 Urothelial cancer Nectin- Auristatin (microtubule inhibitor)
4
11
Trastuzumab 2019 HER2-positive breast cancer (after two or more lines of HER2 Deruxtecan (topoisomerase inhibitor)
deruxtecan* anti-HER2 therapy)
Abbreviation: FDA, U.S. Food and Drug Administration.

*Granted accelerated approval.
recycling capability of antigens, localization of the target HER3

(tumor vs. stroma), and functional role are important fea- Although HER3 remains an established target for cancer, it
tures to consider for target selection.21,27,28 Target antigens is a novel consideration in the development of ADCs.106
for ADCs in clinical and preclinical development are listed in Belonging to the HER/EGFR family, HER3 lacks intrinsic
Table 2. kinase activity; however, after binding with its ligand
The impact of biologic function as it pertains to target se- (heregulin or neuregulin), it is able to transduce signals
lection is not completely understood. At least in theory, the through the downstream PI3K/AKT pathway by hetero-
dimerization with other HER family members, such as
target could merely serve as an anchor for the binding of
EGFR and HER2.107,108 The role of HER3 has been high-
antibody with resultant delivery of the payload into the cell;
lighted in a number of HER family–driven cancers. In HER2-
therefore, it is not considered essential for a putative target
amplified breast cancer, HER3 expression is required for
to have a role in tumor growth. 102 However, successful tumorigenic growth and is involved in resistance to HER2-
strategies for ADCs developed in solid tumors include directed therapies.109 HER3 signaling plays a crucial role in
HER2-targeting antibodies where HER2 is a well-recognized EGFR tyrosine kinase inhibitor–resistant non–small cell lung
oncogenic driver serving as the target for therapeutic naked cancers (NSCLCs) as well as in cetuximab resistance me-
antibodies.6,11,103,104 Therefore, ADCs targeting antigens diated by heregulin in colorectal cancer (CRC).110,111 Sev-
that promote cancer growth have an advantage because eral mAbs targeting HER3 have been developed with only
they are capable of eliciting antigen-mediated antitumor limited antitumor activity, and, to date, there are no ap-
activity in addition to the cytotoxic activity arising from the proved HER3-targeting anticancer therapies.112 As such,
conjugated drug. ADCs are now being evaluated as a novel strategy for the
treatment of HER3-expressing tumors.
Target expression on tumor cells has been used for selecting
patients most likely to derive benefit from ADCs; however, U3-1402 is a first-in-class anti-HER3–targeting ADC com-
paradoxically, in many instances, they have proven in- posed of the fully human mAb (patritumab, U3-1287) di-
sufficient as predictive biomarkers.13,105 A better un- rected against the extracellular domain of HER3, covalently
derstanding of cancer dependency on target inhibition and conjugated, via cysteine residues, through a tumor selective
the sensitivity of tumor cells to ADC warheads might help in cleavable linker, to DX-8951 derivative (DXd), a highly
potent topoisomerase I inhibitor.113,114 U3-1402 uses the
the selection of the most optimal patient population and
same linker-payload system of trastuzumab deruxtecan,
improve the likelihood of success.
exhibits a high drug-to-antibody ratio (DAR) of 7:8 and has
The following examples highlight some aspects of target shown bystander effect due to its highly membrane-
antigen characteristics as well as complexities involved in permeable payload.42 Interestingly, U3-1402 was effi-
selection of a suitable target for an ADC. ciently internalized into cells at an average rate of over

TABLE 2. Target Antigens for Antibody Drug Conjugates in Clinical and Preclinical Development
Target ADC Ab Linker Payload Disease Indication Development Sponsor References
Well-Established Targets Associated With Driver Oncogenes

29
HER2 [Vic-]Trastuzumab IgG1 trastuzumab Protease-cleavable Seco-DUBA mBC HER2+ FIH (NCT02277717) Synthon
duocarmazine (SYD985) dipeptide valine-citrulline completed, phase III Biopharmaceuticals BV
(NCT03262935), phase
Ib (NCT04235101),
phase II (NCT04205630)
30
ZW49 ZW25 bispecific biparatopic ZymeLink N-acyl sulfonamide High and low HER2- Phase I (NCT03821233) Zymeworks
auristatin expressing ASTs
31
MEDI4276 Bispecific Ab: domain IV Cleavable linker Tubulysin HER2-expressing ASTs Phase I (NCT02576548) MedImmune
and II of HER2 completed
32
A166 HER2 mAB N/A N/A HER2-expressing or Phase I/II (NCT03602079) Klus Pharma
amplified ASTs
33
SBT6050 HER2 mAb IgG1 N/A TLR8 agonist Moderate or high HER2- Preclinic Silverback
expressing ASTs
34
EGFR ABBV-321 High affinity-matured Ab Undisclosed PBD dimer Overexpressing EGFR ASTs Phase I (NCT03234712) AbbVie
(ABT-806 AM1)
35
AVID100 MAB100 Stable linker, SMCC DM1 ASTs resistant to currently Phase I/IIa (NCT03094169) Formation Biologics Forbius
approved EGFR therapy,
TNBC, HNSCC, NSCLC
36
EGFRvIII AMG595 Human IgG Noncleavable MCC DM1 GBM Discontinued Amgen
37
cMET Telisotumab vedotin ABT-700 Valine-citrulline (vc) peptide MMAE Solid tumors Phase I (NCT01472016) AbbVie
completed
38
FGFR-2 Aprotumab ixadotin Human IgG Noncleavable Auristatin W FGFR2-expressing solid Phase I discontinued (poor Bayer
(BAY1187982) tumors tolerance)
39
FGFR-3 LY3076226 N/A N/A DM4 FGFR3-expressing solid Phase I discontinued Eli Lilly
tumors
Targeting Antigens Overexpressed in Cancer Cells

40
AXL Enapotamab vedotin IgG1 Protease-cleavable linker MMAE Ovarian, cervical, Phase I/II (NCT02988817) Genmab
endometrial, NSCLC,
thyroid cancer,
melanoma, and sarcoma
41
ADCT-601 Fully Hz Ab Valine-alanine cleavable PBD Advanced solid tumors Phase I (NCT03700294) ADC Therapeutics
42–48
HER3 U3-1402 Patritumab Peptide DXd DX 8951 (exetecan HER3+ BC, NSCLC Phase I/II (NCT02980341), Daiichi Sankyo
derivate)-topoisomerase phase I (NCT03260491)

inhibitor
49
CD166 CTMX-2009 PROBODY Therapeutic N/A DM4 CD166+ BC, ovarian, Phase I/II (NCT03149549) CytomX
HNSCC, NSCLC

50
CEACAM5 SAR408701 N-sucinimidyl4-(2- DM4 CEACAM+ CRC, non-SQ Phase I/II (NCT02187848) Sanofi
pyridyldithio) butyrate NSCLC, SCLC, gastric

51
IMMU-130 Labetuzumab Hz Ab SPDB SN-38 CRC Phase I/II (NCT01605318) Immunomedics
govitecan completed
52–54
GPNMB Glembatumumab vedotin Human IgG2 Valine-citrulline enzyme- MMAE TNBC, SQ NSCLC, Phase II completed Celldex
cleavable linker melanoma, uveal (NCT01997333), phase I/
melanoma, osteosarcoma II (NCT02713828), phase
II (NCT02363283), phase
II (NCT02302339), phase
II (NCT02487979)
discontinued

e61
TABLE 2. Target Antigens for Antibody Drug Conjugates in Clinical and Preclinical Development (Continued)
e62
55,56
Mesothelin Anetumab ravtansine Human IgG1 A reducible SPDB linker DM4 Mesothelin + malignant Phase I/II (NCT03126630), Bayer
[N-succinimidyl 4-(2- mesothelioma, advanced phase II (NCT03102320,
pyridyldithio)butanoate] solid tumors, ovarian, NCT03587311), phase I/
on an average of three pancreatic II (NCT03816358)
lysyl
57
BMS-986148 Human IgG1 Not disclosed Tubulysin Mesothelin + malignant Phase I (NCT02341625) BMS
mesothelioma, NSCLC,
ovarian
58
LIV1A Ladiratuzumab vedotin Hz IgG1 Protease cleavable MMAE Advanced solid tumors, Phase I/II NCT04032704, Seattle Genetcis
TNBC NCT03310957
Targeting Antigens Overexpressed in Cancer Cells

59,60
TF Tisotumab Vedotin Humab IgG1 A cleavable MMAE Cervical carcinoma, solid Phase I (NCT02552121), Genmab and Seattle
maleimidocaproyl-valyl- tumors, platinum phase I/II Genetics
citrullinylp- resistance, ovarian (NCT03786081), phase II
aminobenzyloxycarbonyl carcinoma (NCT03485209), phase II
(mc-val-cit-PABC) type (NCT03657043)
linker on an average of 3-
4 cysteinyl
61
CD71 (trasferrin receptor) CX-2029 PROBODY Therapeutic N/A MMAE NSCLC, HNSCC, relapsed PROCLAIM-CX-2029 Cytomx/AbbVie
DLBCL (NCT03543813)

62
CD228 SGN-CD228A Hz IgG1, hL49 PEGylate glucuronide MMAE Advanced solid tumors FIH phase I Seattle Genetics
including cutaneous (NCT04042480)
melanoma, pleural
mesothelioma, BC,
NSCLC, CRC, and PDAC
63,64
FRα Mirvetuximab soravtansine Humanized antibody N-Succinimidyl 4-(2- DM4 Ovarian, fallopian tube, Phase I (NCT03552471), ImmunoGen
M9346A pyridyldithio)-2- primary peritoneal phase I (NCT03045393),
sulfobutanoate linker carcinoma phase III
(sulfo-SPDB) (NCT04209855), phase II
(NCT03832361), phase I
(NCT02996825), phase I
(NCT03835819)
65
NaPi2b XMT-1536 Hz Ab Protease-cleavable linker MMAF PROC, mNSCLC FIH (NCT03319628) Mersana
66–69
TROP-2 Sacituzumab hRS7 IgG1κ Cleavable CL2A linker SN-38 TNBC, HR+ BC, basket Phase I/II (NCT01631552), MedImmune
phase II (NCT04230109),
phase Ib/II
(NCT04039230), phase
III (NCT02574455),
phase III
(NCT03901339), phase II
(NCT03992131)

SKB-264 N/A N/A N/A PROC, gastric and Phase FIH trial Klus Pharma N/A
pancreatic cancer, TNBC, (NCT04152499)

UC
70
BAT8003 Glycoengineered by Uncleavable Mayntasine ASTs Phase I FIH Biothera
afucosylation of the Fc (NCT03884517)
region
71
PF-06664178 Hz IgG1 AcLys-VC AUR0101 ASTs Phase I prematurely Pfizer
discontinued
(NCT02122146)
72
DS-1062 Hz Ab Cleavable tetrapeptide- DX-8951 NSCLC FIH NCT03401385 Daiichi Sankyo
based linker

73
PSMA PSMA ADC Human Ab Native cysteine residues, VC MMAE Prostatic carcinoma, GBM Phase I (NCT01414283), Progenics
protease cleavable linker phase II (NCT01695044),
phase II (NCT01856933)
74
MLN2704 Human Ab Lysine residues, SPP DM1 Prostatic carcinoma Phase I completed Millennium
disulfide cleavable linker (NCT00052000), phase I/
II completed
(NCT00070837)
75,76
CD70 SGN-CD70A Human Ab VA linker PBD Renal cell carcinoma and Phase I completed Seattle Genetics
non-Hodgkin lymphoma (NCT02216890)/
discontinued
77
SGN-75 Hz Ab Native cysteine residues, MMAF Renal cell carcinoma and Phase I completed Seattle Genetics
MC noncleavable non-Hodgkin lymphoma (NCT01015911)
72
MDX-1203 mAb Native cysteine residues, VC Duocarmycin (DUO) Renal cell carcinoma and Phase I completed BMS
proteasecleavable non-Hodgkin lymphoma (NCT00944905)
78
STEAP1 RG7450, DSTP3086S, Hz IgG1 Native cysteine residues, MMAE mCRPC Discontinued 2017 Genentech
vandortuzumab vedotin MC-vc-PAB
79
P Cadherin PCA062 IgG1, p Cadherin 3 Native lysine residues, DM1 P-cadherin + tumors Phase I completed Novartis
SMCC noncleavable including TNBC, (NCT02375958)
thioether esophageal, and HNSCC
80
SLITRK6 AGS15E Hu IgG2 Native cysteines residues, MMAE Urothelial carcinoma Phase I completed Agensys
VC proteasecleavable (NCT01963052)
linker
81
LAMP1 SAR428926 IgG1 Cleavable linker SPDB DM4 Advanced solid tumors Phase I completed Sanofi
(NCT02575781)
82
CA9 BAY79-4620 IgG1 Valine–citrulline (vc) MMAE Advanced solid tumors Discontinued Bayer
(NCT01028755)
83
GPR20 DS-6157 GPR20 Ab GGFG protease-cleavable Deruxtecan GIST (highly specific of Preclinic Daiichi Sankyo
linker interstitial cells of Cajal)
84
CLDN18.2 CLDN18.2 CD3 IgG2 bispecific or diabody Valine-citrulline cleavable Auristatin Gastric and pancreatic Preclinic Ganymed Pharmaceutical,
targeting CLDN18.2 and linker tumors Astellas
CD3
Target Antigens in the Tumor Microenvironment

85
Leucine-rich repeat ABBV-085 Hz IgG1 Cleavable MMAE ASTs, undifferentiated FIH (NCT02565758) AbbVie
containing 15 (LRRC15) maleimidocaproyl-valyl- pleomorfic sarcoma, completed
citrullinyl-p- HNSCC, BC
aminobenzyloxycarbonyl
(mc-val-cit-PABC)
86
FAP expressing CAF OMTX705 Hz Ab vcPABA linker Novel cytolysin NSCLC Preclinic Oncomatryx Biopharma
87
ANTXR1 m825-MMAE Dipeptide linker (cleavable MMAE TME: tumor-associated Preclinic N/A
by cathepsin B) fibroblasts, pericytes, and
endothelium

88
TM4SF1 TM4SF1-ADC mAb IgG2 Maleimidocaproyl linker Auristatin cytotoxic agent Tumor cells and vasculature Preclinic Angiex
LP2
89,90

CD25 ADCT-301 (camidanlumab IgG1 Protease cleavable PBD AST with Treg cells CD25+, Phase I (NCT03621982), ADC Therapeutics
tesirine) hematologic tumors phase II (NCT04052997),
CD25+ Phase I completed
(NCT02588092)
91
CD205 MEN1309/OBT076 Hz IgG1 Native lysine residues, DM4 CD205+ AST non-Hodgkin FIH (NCT03403725), phase Menarini Ricerche/Oxford
SPDB disulfide cleavable lymphoma I (NCT04064359) BioTherapeutics
linker
92
B7-H3 DS-7300 Not disclosed Tetrapeptide-based linker DXd AST Phase I/II (NCT04145622) Daiichi
93
MGC018 Hz Ab Valine-citrulline cleavable Seco-DUBA AST Phase I/II (NCT03729596) Macrogenics
peptide

e63
e64
94
HLA-DR IMMU-140 IMMU-114, Hz IgG4 Not disclosed SN-38 Hematologic tumors and Preclinic MedImmune
melanoma
Targeting Stem Cells

95
Delta-like 1 homolog protein ADCT-701 Hz IgG1 GlycoconnectTM technology PBD dimer cytotoxin Liver cancer and other ASTs Preclinic ADC Therapeutics
(DLK-1) to PL1601, which SG3199
contains a valine-alanine
cleavable linker
96
Delta-like ligand 3 (DLL3) Rovalpituzumab Tesirine Hz Polyethylene glycol spacer- PBD dimer SCLC Phase III (NCT03033511) AbbVie
maleimide linker discontinued
97
Ephrin-A4 (EFNA4) PF-06647263 Hz IgG1 AcButDMH-N Calicheamicin-γ1 ASTs Phase I (NCT02078752) Pfizer
completed
98
PTK7 PF-06647020 Hz Ab Cleavable valine-citrulline MMAE TNBC, ovarian, NSCLC Phase I (NCT02222922) Pfizer
linker completed, phase I
(NCT03243331)
99
ROR1 NBE-002 N/A Site-specific SMAC- PNU-159682 TNBC and other solid Preclinic NBE Therapeutics
conjugation (anthracycline derivative) tumors
100
5T4 PF-06263507 Hz IgG1 Noncleavable MMAE ASTs unselected for 5T4 Phase I (NCT01891669) Pfizer
maleimidocaproyl expression completed
101
KAAG1 ADCT-901 Hz Ab GlycoconnectTM technology PBD dimer cytotoxin Ovarian, TNBC Preclinic ADC Therapeutics

SG3199
Source: ClinicalTrials.gov, accessed February 18, 2020.

Abbreviations: ADC, antibody drug conjugate; Ab, antibody; seco-DUBA, seco-duocarmycin–hydroxybenzamide–azaindole; mBC, metastatic breast cancer; FIH, first in human; AST, advanced
solid tumors; mAb, monoclonal antibody; N/A, not available; PBD, pyrrolobenzodiazepine; SMCC, succinimidyl trans-4-(maleimidylmethyl) cyclohexane-1-carboxylate; TNBC, triple-negative breast
cancer; HNSCC, head and neck squamous cell carcinoma; NSCLC, non–small cell lung cancer; MCC, maleimidomethyl cyclohexane-1-carbozylate; GBM, glioblastoma; MMAE, monomethyl
auristatin E; BC, breast cancer; CRC, colorectal cancer; non-SQ, nonsquamous; SCLC, small cell lung cancer; SQ, squamous; SPDB, N-succinimidyl-4-(2-pyridyldithio)butyrate; BMS, Britsol-
Myers Squibb; TF, tissue factor; DLBCL, diffuse large B-cell lymphoma; PDAC, pancreatic ductal adenocarcinoma; MMAF, monomethyl auristatin F; PROC, platinum-resistant ovarian cancer;
mNSCLC, metastatic NSCLC; HR, hormone receptor; UC, urothelial carcinoma; mCRPC, metastatic castration-resistant prostate cancer; GIST, gastrointestinal stromal tumor; TME, tumor
microenvironment; Treg, regulatory T cell.

50%.43 This internalization rate was higher than that of anti- also reported promising activity of U3-1402 in CRC, thus
HER2–targeting mAbs, even though the expression level of warranting additional clinical investigation.47 Furthermore,
cell surface HER3 was lower than that of HER2.42 In- there is a strong rationale for exploring U3-1402 in com-
triguingly, several molecules have been found to influence bination with EGFR and HER2-directed therapies given the
HER3 internalization.115,116 Preliminary results of a phase I/II role of HER3 in mediating resistance. Simultaneously tar-
trial (NCT02980341) have shown that U3-1402 can be geting several members of the HER family may influence
safely administered with promising antitumor efficacy in receptor trafficking and the ability to deliver the payload
heavily treated HER3-overexpressing (HER3 score of tumor more effectively into the tumor, and more complete in-
cells assessed by immunohistochemistry, 2+/3+) metastatic hibition of HER signaling increases the likelihood of anti-
breast cancer.44 The confirmed overall response rate was tumor efficacy. In addition, preclinical data have shown that
40% at a dosage of 4.8 mg/kg and 60.0% at a dosage of U3-1402 can elicit potent antitumor immunity even in the
6.4 mg/kg. Responses were seen independent of molecular setting of tumors insensitive to PD-1 and PD-L1 immune
subtype, with patients with HER2-amplified breast cancer checkpoint inhibitors and that its efficacy is more pro-
representing only 16.7% of the treated population.45 U3-1402 nounced in the presence of PD-1 inhibition, suggesting that
was also tested in patients with metastatic EGFR-mutated, U3-1402 sensitizes insensitive tumors to PD-1 blockade
tyrosine kinase inhibitor–resistant NSCLC (NCT03260491), and has synergistic effects.48 Thus, combination with im-
showing preliminary encouraging antitumor activity and a mune checkpoint inhibitors is also of interest.
manageable safety profile.46 An additional cohort of patients
with metastatic squamous or nonsquamous NSCLC (without Adapting Antibody Drug Conjugate Technology to Antigen
EGFR mutation) and progression after chemotherapy and Trafficking and Recycling
anti–PD-L1 regimens has been recently added to explore the Melanotransferrin (CD228/MFI2/MELTF) is a cell-surfaced
efficacy of U3-1402 at the recommended dosage. glycoprotein belonging to the transferrin family of iron-
Further clinical investigation of U3-1402 is warranted, along binding proteins.118 MELTF is an oncofetal antigen, with
with biomarkers for optimal patient selection. HER3 is a low level of expression in normal tissue and fetal ex-
known to be modulated by a number of different mecha- pression in colon and umbilical cord but is highly expressed
nisms that may dynamically change its expression and lo- in multiple tumor types, including melanoma, mesotheli-
calization and ultimately influence the efficacy of targeted oma, pancreatic, HER2-negative breast cancer, NSCLC,
therapeutics such as ADCs.117 Achieving a greater un- and CRC.119,120 Although the exact mechanisms of its bi-
derstanding of the factors that influence the biology of HER3 ologic function in cancer remain to be elucidated, emerg-
expression as well as its recycling is critical to selecting the ing data indicate that MELTF is involved in angiogenesis,
most optimal patient population. Preclinical studies have cellular proliferation, and tumorigenesis.121,122 Preclinical
TABLE 3. Investigational Antibody Drug Conjugates Using Novel* Payloads

Drug(s) Sponsor Target(s) Payload (Mechanism) Reference(s)
141
Sacituzumab govitecan Immunomedics Trop-2 SN-38 (topoisomerase inhibitor)
29
Trastuzumab Synthon HER2 Duocarmazine (DNA alkylating agent)
duocarmazine Biopharmaceuticals
41,89,142
ADCT-301 ADC Therapeutics CD25 Tesirine (DNA cross-linking agent)
ADCT-402 CD19 NCT03698552
ADCT-601 AXL
ADCT-602 CD22
143,144
HDP-101 Heidelberg Pharma B-cell maturation antigen Amanitin (RNA polymerase inhibitor)
(CD269)
LMB-100 National Cancer Institute Mesothelin Pseudomonas exotoxin A (induces 145,146
apoptosis)
ABBV-155 AbbVie B7-H3 Bcl-2 inhibitor (induces apoptosis) NCT03595059
33
SBT6050 Silverback Therapeutics HER2 TLR8 agonist
BDC-1001 Bolt Biotherapeutics HER2 TLR7/8 agonist NCT04278144
NJH395 Novartis HER2 TLR7/8 agonist NCT03696771
*Novel is defined as anything other than calicheamicin and microtubule targeting agents.

studies demonstrate that the use of a CD228-targeting mAb mesenchymal stromal cells, pericytes, occasional adipo-
results in a decreased ability of melanoma cells to metas- cytes, immune cells (T and B lymphocytes, natural killer
tasize to the brain.123 Given its high distribution in tumor cells, and tumor-associated macrophages), as well as
versus healthy tissue, its wide expression across a broad lymphatic and vascular networks.125 The tumor TME plays
range of malignancies, and its functional role, MELTF is an an important role in tumor growth and metastasis and thus
appealing target for ADCs. SGN-CD228A is a humanized serves as a promising target for anticancer therapies.126,127
anti-CD228 mAb (hL49) to which eight molecules of Such an approach may yield several advantages: it may be
monomethyl auristin E (MMAE), a potent microtubule- applicable in many solid tumors; nonmalignant cells around
disrupting cytotoxic drug, have been conjugated via a β- the tumor are less likely to undergo somatic mutations and
glucuronidase–cleavable linker, which incorporates a therefore are less prone to developing drug resistance; and
polyethylene glycol side chain and self-stabilizing maleimide there is potential for synergism between cytotoxic com-
to achieve homogeneous conjugation.62 The ADC was pounds targeting tumor cells and vascular targeting drugs
further optimized by changing the linker from a dipeptide to as well as between immune checkpoint inhibitors and drugs
a β-glucuronide. Interestingly, it was demonstrated in vitro targeting immune suppressive cells, thus potentiating the
that the cytotoxicity of the ADC was increased by changing antitumor effects of the immune system. In addition, anti-
only the linker. The use of an optimized linker led to faster gens within the TME may have greater accessibility to the
cleavage of the payload and greater activity in low–CD228- circulation, which can facilitate binding to the ADC. As with
expressing cells due to the unique trafficking and recycling antigens expressed on malignant cells, the target should
of CD228 that ultimately increased the internalization rate of be highly expressed within the TME with minimal or ab-
the payload.124 In contrast to the first-generation ADC sent expression in healthy tissues.128 Blockade of immune
(vedotin), SGN-CD228A possesses a more stable linker, checkpoint proteins using anti–PD-1 and PD-L1 antibodies
homogenous drug loading, and an increased DAR (from has clearly emerged as a groundbreaking strategy for an-
four mixed to eight molecules of MMAE), resulting in im- ticancer therapy; however, innate and acquired mecha-
proved pharmacokinetic qualities and increased drug ac- nisms of resistance ultimately limit its therapeutic
cumulation in cells as well as enhancement of its preclinical benefit.129-131 To overcome these limitations, an intriguing
antitumor activity. Preclinical studies also highlighted an strategy is the use of ADCs to target immune suppressive
important finding that cell line sensitivity may ultimately be cells. CD25 (interleukin-2 receptor) is an antigen expressed
a reflection of MMAE sensitivity, with higher intracellular in regulatory T cells, or Tregs, which infiltrate the local TME,
drug concentrations failing to yield more benefit beyond and has been found to be overexpressed in both Hodgkin
a certain threshold.62 In addition, when compared with other lymphoma and non-Hodgkin lymphoma.132 ADCT-301 is
CD228 antibodies with identical binding affinities, hL49 composed of an antibody targeting CD25 linked to a pyr-
demonstrated superiority, in part due to its pH sensitivity in rolobenzodiazepine (PBD)-based toxin.89 Based on the
antigen binding. SGN-CD228A thus represents a uniquely classic concept of targeting tumor cells expressing CD25,
engineered construct, resulting from a preclinical effort to such as lymphoma cells, ADCT-301 has been tested in
optimize the ADC in all its components, including antibody, Hodgkin lymphoma and non-Hodgkin lymphoma, with
linker, and payload, while taking into consideration antigen evidence of preliminary antitumor activity.90 Given its dual
trafficking and recycling. In vivo xenograft models con- mechanism in targeting immune suppressive cells in the
firmed high antitumor activity of the ADC in melanoma, TME and CD25-positive malignant cells, ADCT-301 may
NSCLC, and triple-negative breast cancer (TNBC). The warrant future investigation in several solid tumors.133 A
overall response rate achieved in TNBC animal models was phase Ib study (NCT03621982) is currently ongoing. How-
60%; however, the extent of CD228 expression on cells did ever, the transient expression of CD25 in T cells may pose
not correlate with response, highlighting the need to better a challenge in terms of optimal timing and dosing of treat-
understand other contributing factors, such as TME, MMAE ment.134 The use of pharmacodynamic biomarkers, including
sensitivity, and tumor characteristics that contribute to assessment of tumor-infiltrating Treg cells, may be a con-
ADC activity.62 A phase I trial (NCT04042480) exploring sideration when predicting the efficacy of Treg depletion and
the safety, tolerability, and antitumor activity of SGN- its relationship to antitumor immunity.
CD228A is currently ongoing in selected advanced solid
tumors, including cutaneous melanoma, pleural mesothe- Additional examples of ADCs designed to modulate the
lioma, breast cancer, NSCLC, CRC, and pancreatic cancer. immune system include targeting B7-H3 (CD276), an im-
mune checkpoint from the B7 family, that is overexpressed
Targeting the Tumor Microenvironment Using Antibody in lung, breast, renal cell, and prostate cancers.135 Its in-
Drug Conjugates creased expression in tumors has been correlated with
Tumors are surrounded by an extracellular matrix that in- adverse outcomes.136 The biologic role of B7-H3 is not fully
cludes several nonmalignant cells, such as fibroblasts, elucidated; however, recently it was found to inhibit T-cell

proliferation.137 The molecule has also been linked to the calicheamicin and microtubule-targeting agents. These
decrease in secretion of interferon-γ, tumor necrosis factor- novel payloads can be divided into four main categories:
α, and other cytokines that allow immune escape.138,139 novel DNA-damaging agents, amanitins, inducers of apo-
Based on its overexpression and role in immune evasion, ptosis, and immunomodulatory agents. The investigational
B7-H3 appears to be a compelling target for ADCs, and ADCs using novel payloads are summarized in Table 3, and
currently two new ADCs targeting B7-H3 are in early clin- the FDA-approved therapies with novel payloads are in-
ical development: DS-7300 is a humanized anti–B7-H3 cluded in Table 1.
mAb attached to DXd by a tetrapeptide-based linker, and
MGC018 (hMG.Ab.02–DUBA) is a humanized antibody DNA-Damaging Agents
directed against B7-H3 linked to a potent DNA alkylating Topoisomerase inhibitors, DNA alkylating agents, and DNA
payload, a synthetic duocarmycin analog prodrug known cross-linking agents are all examples of payloads that
as seco-duocarmycin-hydroxybenzamide-azaindole (seco- damage DNA in a manner distinct from calicheamicin.
DUBA), via a valine-citrulline cleavable peptide linker.92,93 Trastuzumab deruxtecan was granted accelerated approval
Both ADCs are very promising and represent an innovative by the FDA in December 2019 for patients with unresect-
way of attacking tumor cells directly via ADC internalization able or metastatic HER2-positive breast cancer who have
into the cell but also by blocking immune suppressive received two or more prior anti-HER2–based regimens in
signaling. These agents not only have the potential to en- the metastatic setting.11 Although the target (HER2) is
hance the efficacy of immune checkpoint inhibitors by established, the payload, a topoisomerase I inhibitor, is
directly promoting T-cell proliferation but also deliver che- a unique exatecan derivative (DXd). In patient-derived xe-
motherapy into the tumor, thus increasing immunogenic nograft (PDX) models of breast cancer, trastuzumab der-
cell death. As such, future development in combination with uxtecan demonstrated activity in HER2-positive models
anti–PD-1/PD-L1 therapies is warranted. resistant to trastuzumab emtansine and in models with low
HER2 expression.20 The preclinical activity observed in low
ADCs targeting PD-L1 represent an innovative therapeutic
HER2-expressing models is likely due to the bystander
approach, taking advantage of immune stimulation coupled
killing effect that results from the highly membrane-
with targeted delivery of a cytotoxic payload. Novel agents
permeable payload.147 In a mouse model, the drug en-
belonging to this class include PD-L1-Dox, an ADC com-
hances antitumor immunity, and combination therapy
posed of an approved antibody targeting PD-L1 conjugated
with an anti–PD-1 antibody was more effective than mon-
to doxorubicin via a hydrazone linker containing a poly-
otherapy.148 According to ClinicalTrials.gov (as of February
ethylene glycol spacer.140 The antibody binds to PD-L1 on
23, 2020), there are eight ongoing studies with trastuzumab
the surface of the cancer cell, resulting in inhibition of PD-
deruxtecan either as monotherapy or in combination with
L1, thus promoting the activation of T cells. Because PD-L1
anti–PD-1 antibodies, including studies with HER2-
is not internalized via endocytosis, the cytotoxic payload
low–expressing metastatic breast cancer, urothelial can-
(Dox) is released outside the cells, disrupting the extra-
cer, and NSCLC. It should be noted that grade 5 interstitial
cellular environment of the tumor and increasing the
lung disease was observed in 2.2% of patients treated in the
penetration of the PD-L1 antibody into the tumor core.
pivotal trial for HER2-positive breast cancer.11 It will be
Building upon the classic concept of an ADC designed to
important to determine if a similar rate of grade 5 intersti-
deliver chemotherapy into the cell as well as the in-
tial lung disease is observed in additional studies. Another
ternalization of antigens, this ADC has been designed to
ADC with a topoisomerase inhibitor payload is sacituzumab
stimulate the immune system by PD-L1 inhibition while
govitecan, which is awaiting FDA review of a resubmitted
simultaneously delivering doxorubicin, a well-known anti-
biologics license application. Sacituzumab targets Trop-
cancer agent, directly into the TME, ultimately resulting in
2, a cell-surface glycoprotein overexpressed in a broad
a synergistic outcome of chemotherapeutic and immu-
range of epithelial cancers, and the SN-38 payload is the
nostimulatory effects.
active metabolite of irinotecan. Sacituzumab govitecan
NOVEL PAYLOADS FOR ANTIBODY DRUG CONJUGATES demonstrated efficacy with a 33% response rate (median
duration, 7.7 months) in heavily pretreated patients with
The majority of approved ADCs, and many of those in
TNBC. Similar to irinotecan, the dose-limiting toxicity
development, use calicheamicin or microtubule-targeting
was myelosuppression, including a 9.3% rate of febrile
agents, such as maytansinoids or auristatin, as the payload.
neutropenia.141
Many of the ADCs recently developed and currently in
development are attempting to overcome resistance to Trastuzumab duocarmazine is another example of an ADC
previous therapies and expand to new indications by us- with a novel DNA damaging agent, with the duocarmycin
ing novel payloads. For the purpose of this review, “novel” payload functioning as a DNA alkylating agent. Results of
payloads are any payloads that are distinct from a phase I trial were recently reported, and the drug was

noted to have a “manageable safety profile,” with 71% of Immunomodulatory

patients having at least one ocular adverse event (all grades) An exciting new class of ADCs are those that use immu-
and 7% having grade 3 ocular adverse events. In dose- nomodulatory payloads because this mechanism of action
expansion cohorts, responses were observed in multiple is truly novel compared with all prior ADCs on the market
HER2-positive tumors, including HER2-positive breast and under investigation. These noncytotoxic payloads ac-
cancer resistant to trastuzumab emtansine, as well as in tivate myeloid antigen-presenting cells, thereby stimulating
HER2-low–expressing breast cancer.29 Finally, ADC Ther- the immune system with a mechanism that is distinct from
apeutics is developing ADCs with various targets (CD25, checkpoint blockade. SBT6050 is an ADC that targets
CD19, AXL, CD22) for hematologic malignancies and solid HER2 with a payload that is a toll-like receptor 8 (TLR8)
tumors that use the payload tesirine, a synthetic PBD dimer agonist.33 The Fc domain of the antibody engages Fcγ
that causes DNA cross-linking.41,89,142 receptors on the surface of myeloid cells, and subsequent
Amanitins agonism of TLR8 in these cells induces direct macrophage
killing of tumor cells, repolarizes suppressive myeloid cells
Amanitins are small bicyclic peptides naturally found in
to a proimmunogenic phenotype, and induces dendritic
poisonous mushrooms. Amanitins exert their toxic effect by
cells to promote tumor-specific cytotoxic T lymphocyte re-
inhibiting RNA polymerase II, thereby inhibiting transcrip-
sponses.33 Preclinical studies in mice demonstrated robust
tion.143 This mechanism of action is unique among oncology
activity in models that are resistant to checkpoint blockade,
drugs and portends an unacceptable therapeutic index
without any evidence of systemic toxicity resembling cyto-
unless delivered in the context of an ADC. HDP-101 is an
kine release syndrome.33 A phase I trial with SBT6050 is
ADC targeting B-cell maturation antigen (CD269) with an
expected to be launched in 2020. Two very similar ADCs
amanitin payload. Preclinical evaluation of HDP-101 dem-
targeting HER2 have already entered phase I trials. The first,
onstrated promising activity in myeloma cell lines and
BDC-1001, has a TLR7/8 agonist payload and is being
subcutaneous myeloma mouse xenograft models, with
studied in a phase I trial both as monotherapy and in
evidence of tolerability and a wide therapeutic index in
combination with pembrolizumab in HER2-positive ad-
nonhuman primates. HDP-101 has not yet entered clinical
vanced solid tumors (NCT04278144). The second, NJH395,
testing; however, a phase I trial in multiple myeloma is
also has a TLR7/8 agonist payload and is being evaluated in a
expected soon.143,144
phase I trial in nonbreast HER2-positive advanced solid
Inducers of Apoptosis tumors (NCT03696771).
Immunotoxins and Bcl-2 inhibitors are examples of pay-
loads that induce apoptosis. Moxetumomab pasudotox was CONCLUSIONS AND FUTURE DEVELOPMENTS
FDA approved in 2018 for hairy cell leukemia on the basis of The regulatory approval in 2019 of both trastuzumab der-
a single trial showing a high rate of durable complete re- uxtecan and enfortumab vedotin-ejfv clearly confirmed the
sponses in this rare disease, with small but substantial risks potential of antibody-directed delivery of cytotoxics in solid
of capillary leak syndrome and hemolytic uremic syn- tumors. As the field continues to evolve, the selection of
drome.8 The payload, Pseudomonas exotoxin A, induces suitable ADC targets and the identification of a target
apoptosis by inhibiting elongation factor-2. LMB-100, an population remain critical challenges. Taking into account
ADC targeting mesothelin, was engineered to have a the unique mechanism of action of ADCs, matching the
Pseudomonas exotoxin A payload with reduced immu- optimal tumor and indication with the ideal target and
nogenicity and toxicity compared with its predecessors.145 payload in a well-selected population is crucial. Target
A number of monotherapy and combination studies with expression and its biology, including selectivity, heteroge-
LMB-100, all sponsored by the National Cancer Institute, neity, internalization rate, and functional role, are important
are ongoing or have been recently completed. The first parameters that determine the efficacy, therapeutic win-
study with published results was LMB-100 in combination dow, and toxicity profile of ADCs.149 Efforts to further opti-
with nab-paclitaxel in advanced pancreatic cancer, which mize ADCs using engineered antibodies, innovative linkers,
was poorly tolerated due to capillary leak syndrome.146 conjugation methods, and novel payloads are in progress.150
Another ADC with a payload that induces apoptosis is The key role of the payload has been clearly demonstrated
ABBV-155. ABBV-155 targets B7-H3 and has clezutoclax by the development of trastuzumab deruxtecan, the re-
as its payload, which induces apoptosis by inhibiting Bcl- cently approved ADC targeting HER2 that is able to rescue
XL (NCT03595059). ABBV-155 is currently being studied trastuzumab emtansine–resistant tumors using a novel
in a phase I trial as monotherapy and in combination with warhead. High-potency payloads having an innovative
docetaxel or paclitaxel, with expansion cohorts planned mechanism of action are warranted to further expand the
for small cell lung cancer, NSCLC, and breast cancer use of ADCs in the clinical setting. Recent advances in linker
(NCT03595059). and conjugation technologies permit improved stability of

the payload and more optimal pharmacokinetic properties presented on the cell surface in a complex with major
of ADCs. Increasing the DAR and homogenous conjuga- histocompatibility complex class I molecules, to form the
tion are crucial properties for the next generation of ADCs peptide/major histocompatibility complex, which can be
as well as thoughtful selection of the antibody used as specifically recognized by TCRs.152 TCR-mimic ADCs ex-
a carrier for delivery of the payload. High specificity, ad- hibit favorable in vitro and in vivo antitumor activity, making
equate affinity, and receptor-mediated internalization are
this approach very attractive.153
among the key factors to be considered when selecting
an antibody. The next generation of clinical-stage ADCs in- Several studies have demonstrated that ADCs may increase
cludes the use of biparatopic mAbs and small molecule–drug sensitivity to anti–PD-1/PD-L1 therapies given their ability
conjugates.151 Smaller carriers than antibodies may increase to increase immunogenic cell death. Immunomodulatory
the internalization rate, the loading capability, penetration into payloads and targeting the TME represent a unique op-
tissues, and pharmacokinetic properties of the drug, resulting portunity to heighten the immune response and to exert
in a wider therapeutic index.13 ZD49 (developed by Zyme- greater control over the tumor, thereby providing a strong
works, Vancouver, British Columbia, Canada), an ADC cur- rationale for combining these agents with immune check-
rently in development, combines ZW25, a biparatropic, point inhibitors.
HER2-targeting antibody binding the same subdomains as
trastuzumab (ECD4) and pertuzumab (ECD2), to a novel Despite the many challenges experienced to date in ADC
auristatin payload. Preclinical data demonstrate an increased design and development, the future of this class of thera-
binding and internalization rate in cancer cell lines with peutics is very promising. Clinical development strategies
a range of low to high HER2 expression when compared with will require a greater emphasis on basic research to im-
trastuzumab, thus improving the amount of payload delivered prove the biologic properties of ADCs, rigorous translational
into tumor cells and its cytotoxic activity.30 studies to select the most optimal patient population, and
To further expand the landscape of novel targets, T-cell thoughtfully designed clinical trials that can potentially
receptor (TCR) mimetic ADCs are on the horizon with the accelerate their approval. Taken together, these efforts will
unique capability of targeting the TCR and potentially in- contribute to an increased likelihood of success in the
tracellular proteins that are degraded, processed, and development of ADCs.

1
START (South Texas Accelerated Research Therapeutics), Madrid, Spain Amita Patnaik, MD, START (South Texas Accelerated Research
2
Centro Integral Oncológico Clara Campal, Hospital Universitario HM Therapeutics), 4383 Medical Drive, Suite 4021, San Antonio, TX 78229;
Sanchinarro, Madrid, Spain Twitter: @AmitaPatnaikMD; email: [email protected].
3
START (South Texas Accelerated Research Therapeutics), Grand Rapids,
MI
4
START (South Texas Accelerated Research Therapeutics), San Antonio,
TX
EDBK_281107.
References
1. Abdollahpour-Alitappeh M, Lotfinia M, Gharibi T, et al. Antibody-drug conjugates (ADCs) for cancer therapy: Strategies, challenges, and successes. J Cell
Physiol. 2019;234:5628-5642.
2. Sievers EL, Larson RA, Stadtmauer EA, et al; Mylotarg Study Group. Efficacy and safety of gemtuzumab ozogamicin in patients with CD33-positive acute myeloid
leukemia in first relapse. J Clin Oncol. 2001;19:3244-3254.
3. Norsworthy KJ, Ko CW, Lee JE, et al. FDA approval summary: mylotarg for treatment of patients with relapsed or refractory CD33-positive acute myeloid
leukemia. Oncologist. 2018;23:1103-1108.
4. Pro B, Advani R, Brice P, et al. Brentuximab vedotin (SGN-35) in patients with relapsed or refractory systemic anaplastic large-cell lymphoma: results of a phase
II study. J Clin Oncol. 2012;30:2190-2196.
5. Younes A, Gopal AK, Smith SE, et al. Results of a pivotal phase II study of brentuximab vedotin for patients with relapsed or refractory Hodgkin’s lymphoma. J Clin
Oncol. 2012;30:2183-2189.
367:1783-1791.

375:740-753.
8. Kreitman RJ, Dearden C, Zinzani PL, et al. Moxetumomab pasudotox in relapsed/refractory hairy cell leukemia. Leukemia. 2018;32:1768-1777.
9. Sehn LH, Herrera AF, Flowers CR, et al. Polatuzumab vedotin in relapsed or refractory diffuse large B-cell lymphoma. J Clin Oncol. 2020;38:155-165.
10. Rosenberg JE, O’Donnell PH, Balar AV, et al. Pivotal trial of enfortumab vedotin in urothelial carcinoma after platinum and anti-programmed death 1/pro-
grammed death ligand 1 therapy. J Clin Oncol. 2019;37:2592-2600.
J Med. 2020;382:610-621.
12. Challita-Eid PM, Satpayev D, Yang P, et al. Enfortumab vedotin antibody-drug conjugate targeting nectin-4 is a highly potent therapeutic agent in multiple
preclinical cancer models. Cancer Res. 2016;76:3003-3013.
13. Coats S, Williams M, Kebble B, et al. Antibody-drug conjugates: future directions in clinical and translational strategies to improve the therapeutic index. Clin
Cancer Res. 2019;25:5441-5448.
14. Damelin M, Zhong W, Myers J, et al. Evolving strategies for target selection for antibody-drug conjugates. Pharm Res. 2015;32:3494-3507.
15. Lucas AT, Price LSL, Schorzman AN, et al. Factors affecting the pharmacology of antibody-drug conjugates. Antibodies (Basel). 2018;7:10.
16. Bander NH. Antibody-drug conjugate target selection: critical factors. Methods Mol Biol. 2013;1045:29-40.
18. Tijink BM, Buter J, de Bree R, et al. A phase I dose escalation study with anti-CD44v6 bivatuzumab mertansine in patients with incurable squamous cell
carcinoma of the head and neck or esophagus. Clin Cancer Res. 2006;12:6064-6072.
19. Deslandes A. Comparative clinical pharmacokinetics of antibody-drug conjugates in first-in-human Phase 1 studies. MAbs. 2014;6:859-870.
20. Ogitani Y, Aida T, Hagihara K, et al. DS-8201a, a novel HER2-targeting ADC with a novel DNA topoisomerase I inhibitor, demonstrates a promising antitumor
efficacy with differentiation from T-DM1. Clin Cancer Res. 2016;22:5097-5108.
21. Perez HL, Cardarelli PM, Deshpande S, et al. Antibody-drug conjugates: current status and future directions. Drug Discov Today. 2014;19:869-881.
22. Autio KA, Boni V, Humphrey RW, et al. Probody therapeutics: an emerging class of therapies designed to enhance on-target effects with reduced off-tumor
toxicity for use in immuno-oncology. Clin Cancer Res. 2020;26:984-998.
23. Polu KR, Lowman HB. Probody therapeutics for targeting antibodies to diseased tissue. Expert Opin Biol Ther. 2014;14:1049-1053.
24. Desnoyers LR, Vasiljeva O, Richardson JH, et al. Tumor-specific activation of an EGFR-targeting probody enhances therapeutic index. Sci Transl Med. 2013;
5:207ra144.
25. Teicher BA. Antibody-drug conjugate targets. Curr Cancer Drug Targets. 2009;9:982-1004.
26. Gerber HP, Senter PD, Grewal IS. Antibody drug-conjugates targeting the tumor vasculature: current and future developments. MAbs. 2009;1:247-253.
28. Khongorzul P, Ling CJ, Khan FU, et al. Antibody-drug conjugates: a comprehensive review. Mol Cancer Res. 2020;18:3-19.
29. Banerji U, van Herpen CML, Saura C, et al. Trastuzumab duocarmazine in locally advanced and metastatic solid tumours and HER2-expressing breast cancer:
a phase 1 dose-escalation and dose-expansion study. Lancet Oncol. 2019;20:1124-1135.
30. Hamblett K, Hammond PW, Barnscher SD, et al. ZW49, a HER2 targeted biparatopic antibody drug conjugate for the treatment of HER2 expressing cancers.
Cancer Res. 2019;79(suppl; abstr 3914).
31. Pegram M, Hamilton E, Tan AR, et al. Phase 1 study of bispecific HER2 antibody-drug conjugate MEDI4276 in patients with advanced HER2-positive breast or
gastric cancer. Ann Oncol. 2018;29(suppl; abstr 470).
32. Rinnerthaler G, Gampenrieder SP, Greil R. HER2 directed antibody-drug-conjugates beyond T-DM1 in breast cancer. Int J Mol Sci. 2019;20:1115.
33. Moyes K, Brender T, Smith SW, et al. A systemically administered, conditionally active TLR8 agonist for the treatment of HER2-expressing tumors. Cancer Res.
2019;79(suppl; abstr 3271).
34. Chia PL, Cao D, Gan HK, et al. ABT-806 derived antibody drug conjugates (ADCs) inhibit growth of malignant mesothelioma in-vivo. J Clin Oncol. 2019;37:15s
(suppl; abstr e14677).
35. Thwaites MJ, Figueredo R, Tremblay Get al. AVID100 is an anti-EGFR ADC that promotes DM1-meditated cytotoxicity on cancer cells but not on normal cells.
36. Rosenthal M, Curry R, Reardon DA, et al. Safety, tolerability, and pharmacokinetics of anti-EGFRvIII antibody-drug conjugate AMG 595 in patients with recurrent
malignant glioma expressing EGFRvIII. Cancer Chemother Pharmacol. 2019;84:327-336.
37. Camidge DR, Barlesi F, Goldman JW, et al. Results of the phase 1b study of ABBV-399 (telisotuzumab vedotin; teliso-v) in combination with erlotinib in patients
with c-Met+ non-small cell lung cancer by EGFR mutation status. J Clin Oncol. 2019;37:15s (suppl; abstr 3011).
38. Kim SB, Meric-Bernstam F, Kalyan A, et al. First-in-human phase I STUDY of aprutumab ixadotin, a fibroblast growth factor receptor 2 antibody-drug conjugate
(BAY 1187982) in patients with advanced cancer. Target Oncol. 2019;14:591-601.
39. Surguladze D, Pennello A, Ren X, et al. LY3076226, a novel anti-FGFR3 antibody drug conjugate exhibits potent and durable anti-tumor activity in tumor models
harboring FGFR3 mutations or fusions. Cancer Res. 2019;79(suppl; abstr 4835).

40. Koopman LA, Terp MG, Zom GG, et al. Enapotamab vedotin, an AXL-specific antibody-drug conjugate, shows preclinical antitumor activity in non-small cell lung
cancer. JCI Insight. 2019;4:e128199.
41. Zammarchi F, Havenith K, Chivers S, et al. Preclinical activity of ADCT-601, a novel pyrrolobenzodiazepine (PBD) dimer-based antibody-drug conjugate (ADC)
targeting AXL-expressing tumors. Cancer Res. 2018;78(suppl; abstr 2792A).
42. Hashimoto Y, Koyama K, Kamai Y, et al. A novel HER3-targeting antibody-drug conjugate, U3-1402, exhibits potent therapeutic efficacy through the delivery of
cytotoxic payload by efficient internalization. Clin Cancer Res. 2019;25:7151-7161.
43. Koyama K, Hashimoto Y, Kamai Y, et al. U3-1402, a novel HER3-targeting antibody-drug conjugate, exhibits its antitumor activity through increased payload
intracellular delivery via highly efficient drug internalization. Cancer Res. 2019;79(suppl; abstr LB-275).
44. Masuda N, Yonemori K, Takahashi Set al. Single agent activity of U3-1402, a HER3-targeting antibody-drug conjugate, in HER3-overexpressing metastatic
breast cancer: updated results of a phase 1/2 trial. Cancer Res. 2019;79(suppl; abstr PD1-03).
45. Yonemori K, Masuda N, Takahashi S, et al. Single agent activity of U3-1402, a HER3-targeting antibody-drug conjugate, in HER3-overexpressing metastatic
breast cancer: updated results from a phase I/II trial. Ann Oncol. 2019;30(suppl; abstr PD1-03).
46. Janne PA, Yu HA, Johnson ML, et al. Safety and preliminary antitumor activity of U3-1402: a HER3-targeted antibody drug conjugate in EGFR TKI-resistant,
EGFRm NSCLC. J Clin Oncol. 2019;37:15s (suppl; abstr 9010).
47. Koganemaru S, Kuboki Y, Koga Y, et al. U3-1402, a novel HER3-targeting antibody-drug conjugate, for the treatment of colorectal cancer. Mol Cancer Ther.
2019;18:2043-2050.
48. Haratani K, Yonesaka K, Takamura S, et al. U3-1402 sensitizes HER3-expressing tumors to PD-1 blockade by immune activation. J Clin Invest. 2020;
130:374-388.
49. Meric-Bernstam F, Boni V, Spira AI, et al. Preliminary results of PROCLAIM-CX-2009, a first-in-human, dose-finding study of the Probody drug conjugate
CX-2009 in patients with advanced solid tumors. Cancer Res. 2019;79(suppl; abstr LB-185).
50. Gazzah A, Cousin S, Boni V, et al. First-in-human phase 1 study of the antibody-drug conjugate (ADC) SAR408701 in advanced solid tumors: dose-expansion
cohort of patients (pts) with non-squamous non-small cell lung cancer (NSQ NSCLC). J Clin Oncol. 2019;37:15s (suppl; abstr 9072).
51. Dotan E, Cohen SJ, Starodub AN, et al. Phase I/II trial of labetuzumab govitecan (anti-CEACAM5/SN-38 antibody-drug conjugate) in patients with refractory or
relapsing metastatic colorectal cancer. J Clin Oncol. 2017;35:3338-3346.
52. Rose AAN, Biondini M, Curiel R, et al. Targeting GPNMB with glembatumumab vedotin: current developments and future opportunities for the treatment of
cancer. Pharmacol Ther. 2017;179:127-141.
53. Roth M, Barris DM, Piperdi S, et al. Targeting glycoprotein NMB with antibody-drug conjugate, glembatumumab vedotin, for the treatment of osteosarcoma.
Pediatr Blood Cancer. 2016;63:32-38.
54. Yardley DA, Weaver R, Melisko ME, et al. EMERGE: a randomized phase II study of the antibody-drug conjugate glembatumumab vedotin in advanced
glycoprotein NMB-expressing breast cancer. J Clin Oncol. 2015;33:1609-1619.
55. Golfier S, Kopitz C, Kahnert A, et al. Anetumab ravtansine: a novel mesothelin-targeting antibody-drug conjugate cures tumors with heterogeneous target
expression favored by bystander effect. Mol Cancer Ther. 2014;13:1537-1548.
56. Quanz M, Hagemann UB, Zitzmann-Kolbe S, et al. Anetumab ravtansine inhibits tumor growth and shows additive effect in combination with targeted agents
and chemotherapy in mesothelin-expressing human ovarian cancer models. Oncotarget. 2018;9:34103-34121.
57. Clarke J, Chu S-C, Siu LL, et al. BMS-986148, an anti-mesothelin antibody-drug conjugate (ADC), alone or in combination with nivolumab demonstrates clinical
activity in patients with select advanced solid tumors. Mol Cancer Ther. 2019;18(suppl; abstr B057).
58. Han HS, Alemany CA, Brown-Glaberman UA, et al. SGNLVA-002: Single-arm, open label phase Ib/II study of ladiratuzumab vedotin (LV) in combination with
pembrolizumab for first-line treatment of patients with unresectable locally advanced or metastatic triple-negative breast cancer. J Clin Oncol. 2019;37:15s
(suppl; abstr TPS1110).
59. de Bono JS, Concin N, Hong DS, et al. Tisotumab vedotin in patients with advanced or metastatic solid tumours (InnovaTV 201): a first-in-human, multicentre,
phase 1-2 trial. Lancet Oncol. 2019;20:383-393.
60. Hong DS, Concin N, Vergote I, et al. Tisotumab vedotin in previously treated recurrent or metastatic cervical cancer. Clin Cancer Res. 2020;26:1220-1228.
61. Singh S, DuPage A, Weaver AY, et al. Development of a probody drug conjugate (PDC) targeting CD71 for the treatment of solid tumors and lymphomas. Cancer
Res. 2016;76(suppl; abstr 2975).
62. Sandall SL, Mason M, Olson D, et al. SGN-CD228A: a novel humanized anti-CD228 antibody-drug conjugate for the treatment of solid tumors. Cancer Res.
2019;79(suppl; abstr 2688).
63. Moore KN, O’Malley DM, Vergote I, et al. Safety and activity findings from a phase 1b escalation study of mirvetuximab soravtansine, a folate receptor alpha
(FRα)-targeting antibody-drug conjugate (ADC), in combination with carboplatin in patients with platinum-sensitive ovarian cancer. Gynecol Oncol. 2018;
151:46-52.
64. Moore KN, Vergote I, Oaknin A, et al. FORWARD I: a Phase III study of mirvetuximab soravtansine versus chemotherapy in platinum-resistant ovarian cancer.
Future Oncol. 2018;14:1669-1678.
65. Tolcher AW, Ulahannan SV, Papadopoulos KP, et al. Phase 1 dose escalation study of XMT-1536, a novel NaPi2b-targeting antibody-drug conjugate (ADC), in
patients (pts) with solid tumors likely to express NaPi2b. J Clin Oncol. 2019;37:15s (suppl; abstr 3010).
66. Bardia A, Mayer IA, Kalinsky K. Sacituzumab govitecan-hziy in triple-negative breast cancer: reply. N Engl J Med. 2019;380:2382.

67. Fenn KM, Kalinsky K. Sacituzumab govitecan: antibody-drug conjugate in triple-negative breast cancer and other solid tumors. Drugs Today (Barc). 2019;
55:575-585.
68. Goldenberg DM, Sharkey RM. Antibody-drug conjugates targeting TROP-2 and incorporating SN-38: a case study of anti-TROP-2 sacituzumab govitecan.
MAbs. 2019;11:987-995.
69. Weiss J, Glode A, Messersmith WA, et al. Sacituzumab govitecan: breakthrough targeted therapy for triple-negative breast cancer. Expert Rev Anticancer Ther.
2019;19:673-679.
70. Tang W, Huang X, Ou Z, et al. BAT8003, a potent anti-Trop-2 antibody-drug conjugate, for the treatment of triple negative breast cancer. Cancer Res. 2019;
79(suppl; abstr P6-20-16).
71. King GT, Eaton KD, Beagle BR, et al. A phase 1, dose-escalation study of PF-06664178, an anti-Trop-2/Aur0101 antibody-drug conjugate in patients with
advanced or metastatic solid tumors. Invest New Drugs. 2018;36:836-847.
72. Sands JM, Shimizu T, Garon EB, et al. First-in-human phase 1 study of DS-1062a in patients with advanced solid tumors. J Clin Oncol. 2019;37:15s (suppl;
abstr 9051).
73. Stone L. PSMA ADC shows promise in advanced disease. Nat Rev Urol. 2019;16:206.
74. Milowsky MI, Galsky MD, Morris MJet al. Phase 1/2 multiple ascending dose trial of the prostate-specific membrane antigen-targeted antibody drug conjugate
MLN2704 in metastatic castration-resistant prostate cancer. Urol Oncol. 2016;34:530.e515–530.e521.
75. Pal SK, Forero-Torres A, Thompson JA, et al. A phase 1 trial of SGN-CD70A in patients with CD70-positive, metastatic renal cell carcinoma. Cancer. 2019;
125:1124-1132.
76. Phillips T, Barr PM, Park SI, et al. A phase 1 trial of SGN-CD70A in patients with CD70-positive diffuse large B cell lymphoma and mantle cell lymphoma. Invest
New Drugs. 2019;37:297-306.
77. Tannir NM, Forero-Torres A, Ramchandren R, et al. Phase I dose-escalation study of SGN-75 in patients with CD70-positive relapsed/refractory non-Hodgkin
lymphoma or metastatic renal cell carcinoma. Invest New Drugs. 2014;32:1246-1257.
78. Moreaux J, Kassambara A, Hose D, et al. STEAP1 is overexpressed in cancers: a promising therapeutic target. Biochem Biophys Res Commun. 2012;
429:148-155.
79. Leal AD, Krishnamurthy A, Head L, et al. Antibody drug conjugates under investigation in phase I and phase II clinical trials for gastrointestinal cancer. Expert
Opin Investig Drugs. 2018;27:901-916.
80. Morrison K, Challita-Eid PM, Raitano A, et al. Development of ASG-15ME, a novel antibody-drug conjugate targeting SLITRK6, a new urothelial cancer
biomarker. Mol Cancer Ther. 2016;15:1301-1310.
81. Campos MP, Konecny GE. The target invites a foe: antibody-drug conjugates in gynecologic oncology. Curr Opin Obstet Gynecol. 2018;30:44-50.
82. Petrul HM, Schatz CA, Kopitz CC, et al. Therapeutic mechanism and efficacy of the antibody-drug conjugate BAY 79-4620 targeting human carbonic anhydrase
9. Mol Cancer Ther. 2012;11:340-349.
83. Hase M, Yokomizo T, Shimizu T, et al. Characterization of an orphan G protein-coupled receptor, GPR20, that constitutively activates Gi proteins. J Biol Chem.
2008;283:12747-12755.
84. Zhu G, Foletti D, Liu X, et al. Targeting CLDN18.2 by CD3 bispecific and ADC modalities for the treatments of gastric and pancreatic cancer. Sci Rep. 2019;
9:8420.
85. Demetri GD, Luke J, Hollebecque A, et al. First-in-human phase 1 study of ABBV-085, an antibody-drug conjugate (ADC) targeting LRRC15, in sarcomas and
other advanced solid tumors. J Clin Oncol. 2019;37:15s (suppl; abstr 3004).
86. Fabre M, Ferrer C, Domı́nguez-Hormaetxe S, et al. 1200PTumour stroma targeting and modulation by OMTX705 ADC, a novel and potent immunotherapeutic
treatment of solid tumours. Ann Oncol. 2019;30(suppl; abstr v490).
87. Szot C, Saha S, Zhang XM, et al. Tumor stroma-targeted antibody-drug conjugate triggers localized anticancer drug release. J Clin Invest. 2018;128:2927-2943.
88. Visintin A, Knowlton K, Tyminski E, et al. Novel anti-TM4SF1 antibody-drug conjugates with activity against tumor cells and tumor vasculature. Mol Cancer Ther.
2015;14:1868-1876.
89. Flynn MJ, Zammarchi F, Tyrer PC, et al. ADCT-301, a pyrrolobenzodiazepine (PBD) dimer-containing antibody-drug conjugate (ADC) targeting CD25-
expressing hematological malignancies. Mol Cancer Ther. 2016;15:2709-2721.
90. Horwitz SM, Fanale MA, Spira AU, et al. Interim data from the first clinical study of ADCT-301, a novel pyrrolobenzodiazapine-based antibody drug conjugate, in
relapsed/refractory Hodgkin/non-Hodgkin lymphoma. Hematol Oncol. 2017;35:1270-1271.
91. Merlino G, Fiascarelli A, Bigioni M, et al. MEN1309/OBT076, a first-in-class antibody-drug conjugate (ADC) targeting CD205 in solid tumors. Mol Cancer Ther.
2019;18:1533-1543.
92. Michiko Yamato JH, Hattori C, Maejima T, et al. DS-7300a, a novel B7-H3-targeting ADC with a DNA topoisomerase I inhibitor, shows potent anti-tumor effects.
Paper presented at The 23rd JFCR-International Symposium on Cancer Chemotherapy; December 2018; Tokyo, Japan.
93. Son T, Scribner JA, Hooley J, et al. Preclinical development of MGC018, a duocarmycin-based antibody-drug conjugate targeting B7-H3 for solid cancer.
94. Cardillo TM, Govindan SV, Zalath MB, et al. IMMU-140, a novel SN-38 antibody-drug conjugate targeting HLA-DR, mediates dual cytotoxic effects in he-
matologic cancers and malignant melanoma. Mol Cancer Ther. 2018;17:150-160.

95. Zammarchi F, Havenith K, Chivers S, et al. ADCT-701, a novel pyrrolobenzodiazepine (PBD) dimer-based antibody-drug conjugate (ADC) targeting DLK1-
expressing tumors. Cancer Res. 2018;78(suppl; abstr 744).
96. Morgensztern D, Besse B, Greillier L, et al. Efficacy and safety of rovalpituzumab tesirine in third-line and beyond patients with DLL3-expressing, relapsed/
refractory small-cell lung cancer: results from the phase II TRINITY study. Clin Cancer Res. 2019;25:6958-6966.
97. Garrido-Laguna I, Krop IE, Burris Het al. A phase I study of PF-06647263, a novel EFNA4-ADC, in patients with metastatic triple negative breast cancer. J Clin
Oncol. 2017;35:15s (suppl; abstr 2511).
98. Sachdev JC, Maitland ML, Sharma MR, et al. PF-06647020 (PF-7020), an antibody-drug conjugate (ADC) targeting protein tyrosine kinase 7 (PTK7), in patients
(pts) with advanced solid tumors: results of a phase I dose escalation and expansion study. J Clin Oncol. 2018;36:15s (suppl; abstr 5565).
99. Beerli RR, Waldmeier L, Gébleux R, et al. NBE-002, an anthracycline-based immune-stimulatory antibody drug conjugate (iADC) targeting ROR1 for the
treatment of triple-negative breast cancer. Cancer Res. 2019;79(suppl; abstr LB-197).
100. Shapiro GI, Vaishampayan UN, LoRusso P, et al. First-in-human trial of an anti-5T4 antibody-monomethylauristatin conjugate, PF-06263507, in patients with
advanced solid tumors. Invest New Drugs. 2017;35:315-323.
101. Zammarchi F, Bertelli F, Havenith K, et al. Pre-clinical characterization of 3A4-PL1601, a novel pyrrolobenzodiazepine (PBD) dimer-based antibody-drug
conjugate (ADC) directed against KAAG1-expressing malignancies. Cancer Res. 2019;79 (suppl; abstr 234).
102. Nejadmoghaddam MR, Minai-Tehrani A, Ghahremanzadeh R, et al. Antibody-drug conjugates: possibilities and challenges. Avicenna J Med Biotechnol. 2019;
11:3-23.
103. Baselga J. Treatment of HER2-overexpressing breast cancer. Ann Oncol. 2010;21(Suppl 7):vii36-vii40.
104. Cortés J, Fumoleau P, Bianchi GV, et al. Pertuzumab monotherapy after trastuzumab-based treatment and subsequent reintroduction of trastuzumab: activity
and tolerability in patients with advanced human epidermal growth factor receptor 2-positive breast cancer. J Clin Oncol. 2012;30:1594-1600.
105. Peters S, Stahel R, Bubendorf L, et al. Trastuzumab emtansine (T-DM1) in patients with previously treated HER2-overexpressing metastatic non-small cell lung
cancer: efficacy, safety, and biomarkers. Clin Cancer Res. 2019;25:64-72.
106. Liu X, Liu S, Lyu H, et al. Development of effective therapeutics targeting HER3 for cancer treatment. Biol Proced Online. 2019;21:5.
107. Citri A, Skaria KB, Yarden Y. The deaf and the dumb: the biology of ErbB-2 and ErbB-3. Exp Cell Res. 2003;284:54-65.
108. Olayioye MA, Neve RM, Lane HA, et al. The ErbB signaling network: receptor heterodimerization in development and cancer. EMBO J. 2000;19:3159-3167.
109. Holbro T, Beerli RR, Maurer F, et al. The ErbB2/ErbB3 heterodimer functions as an oncogenic unit: ErbB2 requires ErbB3 to drive breast tumor cell proliferation.
Proc Natl Acad Sci USA. 2003;100:8933-8938.
110. Engelman JA, Jänne PA, Mermel C, et al. ErbB-3 mediates phosphoinositide 3-kinase activity in gefitinib-sensitive non-small cell lung cancer cell lines. Proc Natl
Acad Sci USA. 2005;102:3788-3793.
111. Kawakami H, Okamoto I, Yonesaka K, et al. The anti-HER3 antibody patritumab abrogates cetuximab resistance mediated by heregulin in colorectal cancer
cells. Oncotarget. 2014;5:11847-11856.
112. Zhang N, Chang Y, Rios A, et al. HER3/ErbB3, an emerging cancer therapeutic target. Acta Biochim Biophys Sin (Shanghai). 2016;48:39-48.
113. Ogitani Y, Abe Y, Iguchi T, et al. Wide application of a novel topoisomerase I inhibitor-based drug conjugation technology. Bioorg Med Chem Lett. 2016;
26:5069-5072.
114. Nakada T, Masuda T, Naito H, et al. Novel antibody drug conjugates containing exatecan derivative-based cytotoxic payloads. Bioorg Med Chem Lett. 2016;
26:1542-1545.
115. Huang Z, Choi BK, Mujoo K, et al. The E3 ubiquitin ligase NEDD4 negatively regulates HER3/ErbB3 level and signaling. Oncogene. 2015;34:1105-1115.
116. Szymanska M, Fosdahl AM, Raiborg C, et al. Interaction with epsin 1 regulates the constitutive clathrin-dependent internalization of ErbB3. Biochim Biophys
Acta. 2016;1863:1179-1188.
117. Collins DM, Bossenmaier B, Kollmorgen G, et al. Acquired resistance to antibody-drug conjugates. Cancers (Basel). 2019;11:394.
118. Suryo Rahmanto Y, Dunn LL, Richardson DR. The melanoma tumor antigen, melanotransferrin (p97): a 25-year hallmark: from iron metabolism to tu-
morigenesis. Oncogene. 2007;26:6113-6124.
119. Duś-Szachniewicz K, Ostasiewicz P, Woźniak M, et al. Pattern of melanotransferrin expression in human colorectal tissues: an immunohistochemical study on
potential clinical application. Anticancer Res. 2015;35:6551-6561.
120. Rose TM, Plowman GD, Teplow DB, et al. Primary structure of the human melanoma-associated antigen p97 (melanotransferrin) deduced from the mRNA
sequence. Proc Natl Acad Sci USA. 1986;83:1261-1265.
121. Sala R, Jefferies WA, Walker B, et al. The human melanoma associated protein melanotransferrin promotes endothelial cell migration and angiogenesis in vivo.
Eur J Cell Biol. 2002;81:599-607.
122. Dunn LL, Sekyere EO, Suryo Rahmanto Y, et al. The function of melanotransferrin: a role in melanoma cell proliferation and tumorigenesis. Carcinogenesis.
2006;27:2157-2169.
123. Rolland Y, Demeule M, Fenart L, et al. Inhibition of melanoma brain metastasis by targeting melanotransferrin at the cell surface. Pigment Cell Melanoma Res.
2009;22:86-98.
124. Burke PJ, Hamilton JZ, Jeffrey SC, et al. Optimization of a PEGylated glucuronide-monomethylauristatin E linker for antibody-drug conjugates. Mol Cancer Ther.
2017;16:116-123.

125. Hui L, Chen Y. Tumor microenvironment: sanctuary of the devil. Cancer Lett. 2015;368:7-13.
126. Quail DF, Joyce JA. Microenvironmental regulation of tumor progression and metastasis. Nat Med. 2013;19:1423-1437.
127. Wang M, Zhao J, Zhang L, et al. Role of tumor microenvironment in tumorigenesis. J Cancer. 2017;8:761-773.
128. Mathur R, Weiner GJ. Picking the optimal target for antibody-drug conjugates. Am Soc Clin Oncol Educ Book. 2013;33:103-107.
130. Zou W, Wolchok JD, Chen L. PD-L1 (B7-H1) and PD-1 pathway blockade for cancer therapy: mechanisms, response biomarkers, and combinations. Sci Transl
Med. 2016;8:328rv4.
131. Nowicki TS, Hu-Lieskovan S, Ribas A. Mechanisms of resistance to PD-1 and PD-L1 blockade. Cancer J. 2018;24:47-53.
132. Sakaguchi S, Sakaguchi N, Asano M, et al. Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor alpha-chains (CD25):
breakdown of a single mechanism of self-tolerance causes various autoimmune diseases. J Immunol. 1995;155:1151-1164.
133. Arce Vargas F, Furness AJS, Solomon I, et al; Lung TRACERx Consortium. Fc-optimized anti-CD25 depletes tumor-infiltrating regulatory T cells and synergizes
with PD-1 blockade to eradicate established tumors. Immunity. 2017;46:577-586.
134. Flynn MJ, Hartley JA. The emerging role of anti-CD25 directed therapies as both immune modulators and targeted agents in cancer. Br J Haematol. 2017;
179:20-35.
135. Ye Z, Zheng Z, Li X, et al. B7-H3 overexpression predicts poor survival of cancer patients: a meta-analysis. Cell Physiol Biochem. 2016;39:1568-1580.
136. Zhang X, Fang C, Zhang G, et al. Prognostic value of B7-H3 expression in patients with solid tumors: a meta-analysis. Oncotarget. 2017;8:93156-93167.
137. Castellanos JR, Purvis IJ, Labak CM, et al. B7-H3 role in the immune landscape of cancer. Am J Clin Exp Immunol. 2017;6:66-75.
138. Chapoval AI, Ni J, Lau JS, et al. B7-H3: a costimulatory molecule for T cell activation and IFN-gamma production. Nat Immunol. 2001;2:269-274.
139. Dong H, Strome SE, Salomao DR, et al. Tumor-associated B7-H1 promotes T-cell apoptosis: a potential mechanism of immune evasion. Nat Med. 2002;
8:793-800.
140. Sau S, Petrovici A, Alsaab HO, et al. PDL-1 antibody drug conjugate for selective chemo-guided immune modulation of cancer. Cancers (Basel). 2019;11:232.
141. Bardia A, Mayer IA, Vahdat LT, et al. Sacituzumab govitecan-hziy in refractory metastatic triple-negative breast cancer. N Engl J Med. 2019;380:741-751.
142. Kahl BS, Hamadani M, Radford J, et al. A phase I study of ADCT-402 (loncastuximab tesirine), a novel pyrrolobenzodiazepine-based antibody-drug conjugate,
in relapsed/refractory B-cell non-Hodgkin lymphoma. Clin Cancer Res. 2019;25:6986-6994.
143. Pahl A, Lutz C, Hechler T. Amanitins and their development as a payload for antibody-drug conjugates. Drug Discov Today Technol. 2018;30:85-89.
144. Pahl A, Ko J, Breunig C, et al. HDP-101: preclinical evaluation of a novel anti-BCMA antibody drug conjugates in multiple myeloma. J Clin Oncol. 2018;36:15s
(suppl; abstr e14527).
145. Bauss F, Lechmann M, Krippendorff BF, et al. Characterization of a re-engineered, mesothelin-targeted Pseudomonas exotoxin fusion protein for lung cancer
therapy. Mol Oncol. 2016;10:1317-1329.
146. Alewine C, Ahmad M, Peer CJ, et al. Phase I/II study of the mesothelin-targeted immunotoxin LMB-100 with nab-paclitaxel for patients with advanced pancreatic
adenocarcinoma. Clin Cancer Res. 2020;26:828-836.
147. Ogitani Y, Hagihara K, Oitate M, et al. Bystander killing effect of DS-8201a, a novel anti-human epidermal growth factor receptor 2 antibody-drug conjugate, in
tumors with human epidermal growth factor receptor 2 heterogeneity. Cancer Sci. 2016;107:1039-1046.
148. Iwata TN, Ishii C, Ishida S, et al. A HER2-targeting antibody-drug conjugate, trastuzumab deruxtecan (DS-8201a), enhances antitumor immunity in a mouse
model. Mol Cancer Ther. 2018;17:1494-1503.
149. Diamantis N, Banerji U. Antibody-drug conjugates: an emerging class of cancer treatment. Br J Cancer. 2016;114:362-367.
150. Birrer MJ, Moore KN, Betella I, et al. Antibody-drug conjugate-based therapeutics: state of the science. J Natl Cancer Inst. 2019;111:538-549.
151. Chari RV. Expanding the reach of antibody-drug conjugates. ACS Med Chem Lett. 2016;7:974-976.
152. Shen Y, Li YM, Zhou JJ, et al. The antitumor activity of TCR-mimic antibody-drug conjugates (TCRm-ADCs) targeting the intracellular Wilms Tumor 1 (WT1)
oncoprotein. Int J Mol Sci. 2019;20:3912.
153. Lowe DB, Bivens CK, Mobley AS, et al. TCR-like antibody drug conjugates mediate killing of tumor cells with low peptide/HLA targets. MAbs. 2017;9:603-614.

Strategic Combinations to Prevent and Overcome

Resistance to Targeted Therapies in Oncology
Ozge Gumusay, MD1,2; Pietro Paolo Vitiello, MD3,4; Chiara Wabl, BA1; Ryan B. Corcoran, MD, PhD5; Alberto Bardelli, PhD3,6; and
Hope S. Rugo, MD1
overview
Recent advances in the understanding of underlying molecular signaling mechanisms of cancer susceptibility
and progression have led to an increase in the use of targeted therapies for cancer treatment. Despite im-
provements in survival with new treatment options in oncology, resistance to therapy is a major obstacle to the
long-term effectiveness of targeted agents in metastatic cancer treatment, culminating in insensitivity to
treatment and tumor outgrowth. Adaptive resistance can play an important role in primary and upfront re-
sistance to therapy as well as in secondary or acquired resistance. By focusing on colorectal and breast tumors,
we discuss how therapeutic combinations based on specific drivers of tumor biology can be used to overcome
resistance. We present how monitoring tumor dynamics over time may allow early adaptation of treatment.
Breast cancer is the most common malignancy in women worldwide, and the majority of these cancers are
sensitive to endocrine therapy (ET) blocking the production of or response to estrogen. However, primary and
acquired resistance limits efficacy. Recent combinations of agents targeted to pathways that drive tumor
growth resistance with ET have resulted in remarkable improvements in disease response and control, im-
proving survival in some settings. In this review, we summarize adaptive resistance mechanisms, approaches
to combination strategies, and dynamic tumor monitoring to improve efficacy and overcome resistance. We
provide examples of combination therapy to enhance the efficacy of targeted therapies in breast and colorectal
tumors.
INTRODUCTION focus on preventing or delaying the onset of resistance.

Despite the ever-increasing understanding of cancer However, this approach is not always feasible because
genetics and the availability of novel anticancer drugs different genetic events can cause resistance to the
(there were 10 new U.S. Food and Drug Administration same therapy and because the therapeutic interval of
[FDA] approvals in 20191), the onset of resistance to multiple combinations is usually narrow and clinically
therapy is a constant in cancer treatment, with very few challenging.9 Therefore, an alternative and powerful
exceptions, and represents the main limiting factor to approach consists of monitoring tumor dynamics over
achieving cure in patients with metastatic cancer.2 As time to promptly adapt the treatment strategy. How-
ever, this requires a strong diagnostic tool (e.g., liquid
for targeted therapies, generated in the long wake of
biopsy) to monitor the continuous evolution of cancer.10
the “oncogene addiction” hypothesis, the fact that
In this article, we discuss mechanisms of adaptive re-
blocking a single driver gene would eradicate the
sistance, potential strategies to overcome resistance,
disease was proven wrong almost immediately after
and real-time monitoring. We also discuss combina-
the approval of the first targeted agents.3,4 Nonethe-
tion therapy to improve outcome in hormone receptor
less, these therapies constitute an important added
(HR)–positive (HR+) breast cancer as an example of
Author affiliations
value in the clinical management of solid and he-
this approach.
and support matologic malignancies by significantly increasing
information (if patient survival and reducing the typical burden of OVERCOMING ADAPTIVE RESISTANCE WITH
applicable) appear chemotherapy-induced side effects.5-7 THERAPEUTIC COMBINATIONS
at the end of this
article. Today, more than 20 years since the approval of the Classically, targeted therapies are most effective when
Accepted on April first targeted agents, a large body of evidence has been they inhibit key signals from an altered oncogene to
13, 2020 and generated, and the onset of resistance is recognized to which the cancer cell is addicted. Data have suggested
published at be a multifactorial process involving several biologic that high-level or near-complete suppression of path-
ascopubs.org on
determinants converging toward a common outcome: way signaling is critical for effective tumor response.11
doi.org/10.1200/ insensitivity to treatment and tumor outgrowth (Fig. 1).8 Tumors have evolved a number of mechanisms to
EDBK_280845 For this reason, the ideal therapeutic approach should maintain key survival pathway signals in the presence

Strategic Combinations of Targeted Therapies in Oncology
cascade.16 BRAFV600 mutations are present in approxi-

mately half of all patients with melanoma, and studies of
BRAF inhibitors conducted more than a decade ago
• Combination therapies in BRAFV600 melanoma exhibited impressive response rates of approximately 50%
and CRC have transformed the treatment of
or greater in this population. 17 Analysis of paired pre-
these diseases and provide evidence that
treatment and on-treatment tumor biopsies obtained from
combinations of targeted therapies can over-
come mechanisms of adaptive resistance and patients initiating BRAF inhibitor therapy suggested that
improve clinical outcome. near-complete MAPK pathway suppression of approxi-
mately 80% or more was required for clinical response.11
• Potential strategies to block adaptive feedback
resistance include RTK, SHP2, or downstream However, preclinical studies revealed that gradual MAPK
(MEK or ERK) MAPK pathway inhibition, which, reactivation could be observed over time in some BRAFV600
when given in combination with the original melanoma models, providing evidence that adaptive
inhibitor such as BRAF, suppress pathway reactivation of MAPK signaling could limit the degree of
reactivation. pathway suppression and contribute to therapeutic re-
• ctDNA has the potential to anticipate clinical sistance.18 Notably, dual MAPK pathway blockade with
progression and identify actionable new targets, combinations of BRAF and MEK inhibitors demonstrated
providing an opportunity to improve outcome. increased ability to maintain suppression of MAPK reac-
• Pathway-independent mechanisms of re- tivation over time. Consistent with these data, clinical trials of
sistance to targeted therapy, such as apoptosis BRAF and MEK inhibitor combinations showed improved
evasion, have led to the development of rational efficacy over BRAF inhibitions alone, with improvement in
combinations, such as targeted agents given response rate, progression-free survival (PFS), and overall
with inhibitors of antiapoptotic proteins. survival (OS).19-21 Three such combinations of BRAF and
• There are multiple mechanisms of endocrine MEK inhibitors—dabrafenib plus trametinib, vemurafenib
resistance in HR+ breast cancer, and rational plus cobimetinib, and encorafenib plus binimetinib—are
combinations of ET with specific targeted now FDA approved for the treatment of BRAFV600 mela-
agents have already revolutionalized care of the noma. These therapies transformed the treatment of this
most common subset in this common malig- disease but also provided key initial clinical proof-of-
nancy, with many new combinations under concept that combinations of targeted therapies could
investigation. overcome mechanisms of adaptive resistance and im-
prove clinical outcome.
of a targeted inhibitor, which sometimes involve genetic
OVERCOMING ADAPTIVE RESISTANCE IN BRAFV600
changes, such as alterations of the drug target itself or al-
COLORECTAL CANCER
teration of other genes in the signaling pathway that “by-
pass” the effects of the inhibitor (Fig. 1A, #1 and #4).12 Subsequently, studies in BRAFV600 colorectal cancer (CRC),
Tumors can also maintain critical survival signals through which accounts for approximately 8% to 10% of all CRC,22,23
nongenetic mechanisms, and one of the most notable deepened our mechanistic understanding of adaptive re-
mechanisms involves adaptive resistance to reactivate key sistance and demonstrated that the potential for adaptive
signaling pathways (Fig. 1A, #3). These changes are often resistance to the same therapy may differ across tumor types
mediated through feedback loops that lead to reactivation of despite the presence of the same oncogenic target alteration.
the original signaling pathway or activation of parallel sur- The first evidence of tumor type–specific differences came
vival pathways.13-15 Adaptive resistance can play key roles in from clinical studies in which BRAF inhibitors, despite en-
primary or upfront resistance to therapy as well as in sec- gendering response rates of approximately 50% or more in
ondary or acquired resistance. BRAFV600 melanoma, showed negligible response rates of
approximately 5% in patients with CRC harboring the same
EARLY LESSONS IN ADAPTIVE RESISTANCE FROM BRAFV600 mutation.24
BRAFV600 MELANOMA
Preclinical studies modeling the differences between sig-
One of the most notable and well-studied mechanisms of naling response to BRAF inhibition in BRAFV600 melanoma
adaptive resistance involves feedback reactivation of the versus CRC models helped to determine the molecular
RAS-MAPK pathway. Our understanding of the complex drivers of this differential sensitivity but also helped de-
mechanisms of adaptive feedback reactivation of MAPK lineate the key molecular mechanisms of adaptive reac-
signaling came from studies in cancers harboring BRAFV600 tivation of MAPK inhibition. Indeed, when models of
mutations, which lead to constitutive activation of down- BRAFV600 melanoma and CRC were treated with BRAF
stream MAPK signaling through the RAF-MEK-ERK kinase inhibitor in vitro, complete suppression of MAPK signaling

Gumusay et al
FIGURE 1. Mechanisms of Re-

sistance to Targeted Therapy Suit-
able for Combination Treatments
(A) Pathway-dependent mech-
anisms: (1) activation of down-
stream effectors, (2) activation of
parallel signaling pathways, (3)
feedback reactivation of on-
cogenic pathway, and (4) tar-
get modification. (B) Pathway-
independent mechanisms: (1)
chromatin remodeling, (2) tran-
scriptional and translational re-
adaptation, (3) oxidative stress,
(4) apoptosis evasion, and (5)
tumor microenvironment stimuli.
was observed out to 48 hours in melanoma models as little effect, combinations of BRAF and EGFR inhibitors led
measured by levels of ERK phosphorylation.13 However, in to improved MAPK pathway suppression and to tumor re-
BRAFV600 CRC models, inhibition of MAPK signaling was gressions in BRAFV600 CRC xenografts, suggesting that
transient over the first few hours, followed by rapid and targeting adaptive feedback might be an effective clinical
substantial reactivation of MAPK signaling despite the con- strategy.13,27
tinued presence and/or refreshment of the drug. This Based on these data, several clinical trials were initiated
reactivation was accompanied by increased phosphorylation evaluating combinations of BRAF and EGFR inhibitors with
and activation of upstream MAPK signaling kinases, in- or without MEK inhibitors in BRAFV600 CRC. “Doublet”
cluding MEK and wild-type RAF (CRAF), and by feedback combinations of BRAF and EGFR inhibitors led to increased
reactivation of RAS activity, as measured by increased GTP- response rates of 10% to 20%, and “triplet” combinations of
bound RAS. The MAPK pathway is known to drive tran- BRAF, EGFR, and MEK inhibitors have produced response
scription of several key negative feedback genes, including rates of up to 21% to 26%.29,30 Importantly, a recent ran-
the sprouty (SPRY) and dual-specificity phosphatase domized phase III trial evaluating the triplet combination of
genes.25,26 In particular, evidence suggests that SPRY genes encorafenib, cetuximab, and binimetinib and the doublet
can inhibit the ability of receptor tyrosine kinases (RTKs) to combination of encorafenib and cetuximab compared with
activate RAS.25 Shortly after BRAF inhibitor treatment, a de- standard chemotherapy in second- or third-line BRAFV600
crease in SPRY gene expression was observed in BRAFV600 CRC showed a statistically significant improved OS of
CRC cells,13 suggesting that a loss of negative feedback could 9 months and 8.4 months versus 5.4 months for standard
be facilitated by RTK-mediated reactivation of MAPK sig- therapy.29 On the basis of these results, on April 8,
naling through wild-type RAS and RAF proteins (Fig. 2). 2020, the FDA approved encorafenib in combination
with cetuximab for the treatment of adult patients with
Mechanistic studies and unbiased functional genomic
BRAFV600E metastatic CRC after a prior therapy. 31 Al-
screens supported that RTKs, and specifically EGFR,
though further improvements in therapeutic efficacy are
were key to driving MAPK reactivation after BRAF inhibition needed, these data highlight how combination therapies to
in BRAFV600 CRC (although in some cases, RTKs other overcome adaptive resistance can lead to improved clinical
than EGFR were able to contribute to adaptive feedback outcome.
reactivation).13,27,28 Indeed, co-inhibition of EGFR in com-
bination with BRAF was able to reduce EGFR-mediated ADAPTIVE RESISTANCE TO OTHER RAS-MAPK
reactivation of RAS and maintain long-term suppression of PATHWAY INHIBITORS
MAPK signaling in most BRAFV600 CRC models.13 Fur- This similar mechanism of adaptive feedback resistance
thermore, whereas BRAF or EGFR inhibitors alone showed driven by pathway reactivation through loss of negative

FIGURE 2. Adaptive Feedback Resistance to RAS-MAPK Inhibitors

(Left) In the absence of an inhibitor, MAPK pathway activation is driven primarily by the altered oncogene (i.e., KRAS or BRAF), and ERK activity drives
transcription of negative feedback signals, which suppress RTK-mediated signaling input to RAS. (Center) In the presence of a RAS-MAPK inhibitor
(i.e., KRASG12C, BRAF, or MEK inhibitor), ERK-dependent negative feedback is lost, leading to RTK-mediated reactivation of RAS-MAPK signaling. (Right)
Potential strategies to block adaptive feedback resistance, including RTK, SHP2, or downstream (MEK or ERK) MAPK pathway inhibition. Abbreviations:
RTK, receptor tyrosine kinase; WT, wild type.
feedback is not exclusive to BRAF inhibitors, and studies RTK-mediated feedback reactivation of RAS-MAPK sig-
have suggested that this same mechanism may drive re- naling occurs rapidly (within 24–72 hours) after KRASG12C
sistance to other inhibitors of the RAS-MAPK pathway. inhibitor treatment. 37 KRAS G12C inhibitor combinations
MEK inhibitors have been evaluated as downstream in- (such as combinations with RTK inhibitors) can block
hibitors of the RAS-MAPK pathway, both alone and feedback reactivation, leading to improved efficacy in vitro
in combination with other agents, in multiple settings, and in vivo, and several clinical trials of KRASG12C inhibitor
including cancers with KRAS mutations. However, reac- combinations are currently underway.16,37,39
tivation of MAPK signaling can be observed within 24–48 COMBINATION STRATEGIES TO OVERCOME
hours after MEK inhibitor treatment in multiple KRAS- ADAPTIVE RESISTANCE
mutant models.14 Overall, several studies suggest that this
RTK Inhibitor Combinations
pathway reactivation is also due to loss of ERK-dependent
negative feedback signals and is mediated through RTK- Because adaptive RAS-MAPK feedback reactivation is
dependent reactivation of MAPK signaling.14,32-36 Thus, thought to be mediated by RTKs in the setting of loss of ERK-
adaptive feedback resistance to MEK inhibitors could be dependent negative feedback signals, combinations of
one explanation for why MEK inhibitors and associated MAPK inhibitors are an attractive strategy, especially in light
combinations have failed to produce clinical efficacy in of the clinical success of RTK co-inhibition with BRAF and
EGFR inhibitors in BRAFV600 CRC. Specific roles for EGFR
KRAS-mutant cancers.
(or HER family) and FGFR signaling in driving adaptive
Similarly, recent clinical trials have demonstrated the po- resistance to RAS-MAPK pathway inhibitors have been
tential benefit of direct, mutant-specific KRAS inhibitors proposed, and several strategies are currently being eval-
designed to specifically target KRASG12C mutations, but uated in clinical trials.35-37 However, one potential liability of
preclinical and clinical evidence suggests that adap- this strategy is that feedback may not be mediated con-
tive resistance may be limiting the effectiveness of these sistently by a single RTK either between different patients
agents.16,37-39 Indeed, several recent studies suggest that with the same cancer type or within an individual patient’s

Gumusay et al
cancer. Therefore, it may be difficult to develop an RTK distinct resistance mechanisms can arise in each metastatic
inhibitor combination strategy that is universally effective. site and in different regions within the same lesion.45-47
SHP2 Inhibitor Combinations Liquid biopsy constitutes a noninvasive methodology to

monitor tumor evolution through the study of cirulating
One potential solution to this issue would be to target tumor DNA (ctDNA), which faithfully captures clonal dy-
a common signaling node that links signaling from multiple namics under treatment with targeted agents, anticipating
RTKs to downstream effectors, such as RAS. Recently, radiologic and clinical progression by several months.48 This
SHP2 has been proposed as a convergent signaling node technique has proven superiority over tissue biopsy for
that is critical for mediating signaling from multiple RTKs detecting tumor heterogeneity and acquired resistance in
to RAS and the RAS-MAPK pathway.40 Preclinical studies gastrointestinal tumors,49 but there is strong evidence of its
have suggested that SHP2 inhibitors can block adaptive utility also in lung cancer50,51 and other solid tumors.52,53
feedback reactivation in response to RAS-MAPK inhibitors,
and several studies are underway evaluating SHP2 in- Because multiple resistance mechanisms can exist in
hibitors in doublet combination with MEK inhibitors, ERK a single patient, comprehensive genomic panels are de-
inhibitors, or KRASG12C inhibitors.16,32-34,37,39 sirable to identify all the potential actionable new targets.54,55
Here, we describe the most common mechanisms of ac-
Downstream MAPK Inhibitor Combinations quired resistance in parallel with the best characterized
In addition to targeting feedback reactivation at an upstream combination strategies.
point, combinations with downstream MAPK inhibitors,
PATHWAY-DEPENDENT RESISTANCE MECHANISMS:
such as MEK or ERK inhibitors, have been studied exten-
CO-TARGETING OF MULTIPLE NODES IN
sively, especially in light of the improved efficacy of BRAF
CONVERGING PATHWAYS
inhibitor combinations with MEK inhibitors in BRAFV600
melanoma. Some studies have suggested that ERK in- Driver mutations are usually truncal events and, as such,
hibitors may be better able to suppress MAPK pathway generate a strong addiction of cancer cells to the pathway
reactivation due to increased upstream signaling relative to activated by the driver oncogene56; in other cases, as for
MEK inhibitors, but this has yet to be validated in clinical EGFR in CRC, the reliance upon a specific pathway is not
trials.41,42 Combination strategies of BRAF inhibitors with caused by an oncogenic event but is rather intrinsic in the
ERK inhibitors and KRASG12C inhibitors with MEK or ERK biology of the tissue of origin, albeit amplified in the ma-
inhibitors are currently in development or underway. lignant cells.57 It is not surprising, then, that the most fre-
quent and well-characterized genetic events leading to
REAL-TIME MONITORING OF THERAPEUTIC RESISTANCE resistance concern the reactivation of the very same
Most targeted agents are exploited in molecularly defined pathway inhibited by the therapy, through several recog-
subgroups of patients, such as erbB2-overexpressing breast nized mechanisms, as shown in Fig. 1A.12
cancer or EGFR-mutated non–small cell lung cancer Activation of Downstream Effectors
(NSCLC), used as positive predictive factors for response to
KRAS-activating mutations represent the most frequent
anti-HER2 or anti-EGFR monoclonal therapies, respectively.
oncogenic alterations in cancer and an established
These biomarkers, which represent the presence or ab-
mechanism of both intrinsic and acquired resistance to anti-
sence of specific cellular dependencies toward oncogenic
EGFR agents in CRC, but activating mutations or amplifi-
pathways, are usually determined on tissue specimens from
cation of other components of the MAPK pathway may also
primary or metastatic cancer sites. However, even if the
result in acquired resistance to anti-EGFR monoclonal
predictivity of these molecular markers is well defined in
antibodies.58-60 KRAS mutations have been considered
clinical trials, spatial and temporal tumor heterogeneity
undruggable until a very recent past, whereas today two
constitute both a strong limitation of tissue analysis and
different selective KRAS G12C-allosteric inhibitors are in
a key determinant of resistance to therapy.43,44
clinical development for cancers harboring this specific
Acquired resistance to targeted therapies often involves mutation.61
novel genetic events and/or the selection of pre-existing
subclones with a genetic makeup that confers insensitivity Activation of Parallel Pathways
to the agents used. Early detection of these genetic events Survival signals are mediated by a multitude of signaling
may dictate a shift in the treatment strategy before re- pathways, and the interference with one can be compen-
sistance is clinically objective. However, in the case of sated through a crosstalk with another, as in the case of the
metastatic disease, genetic analysis of tumor tissue biopsy RAS-MAPK and phosphatidylinositol 30 -kinase (PI3K)-
during treatment does not reflect the full set of genomic mTOR pathways.62-64 However, the clinical development of
events that are taking place at every cancer lesion because MAPK pathway and PI3K-inhibitor combinations has been

hampered by unsatisfactory preclinical and clinical results groups have investigated the cellular dynamics that underlie
and limited tolerability.65-67 Several RTKs can be hyper- this phenomenon. Evidence has accumulated over the
activated in response to inhibition of EGFR in lung and colon existence of a drug-tolerant cell population that is able to
cancer: MET amplification,68,69 HER2 amplification,70,71 and survive the exposure of targeted therapy, the so-called
AXL overexpression72,73 have been implicated in primary or “persister cells,” representing a reservoir from which ge-
acquired resistance to EGFR blockade. Clinical validation of netically divergent drug-resistant derivatives eventually
co-targeting of different receptors has already provided new emerge.96 These cell populations, evidenced in different
therapeutic options, as in the case of HER2-amplified cancer lineages, including lung and CRCs and melanoma,
CRC,74,75 although the clinical investigation of combina- are characterized by a reversible state of insensitivity to the
tions containing MET or AXL inhibitors is still ongoing.76,77 oncogene-directed targeted therapy.97-99 Recently, we have
Pathway Feedback Reactivation demonstrated in CRC that oncogenic EGFR/BRAF inhibition
induces adaptive mutability, providing a potential mecha-
The oncogenic signal that is blocked by a targeted therapy nism for the enhanced genomic instability that generates
can also be restored by feedback loops, as in the case of the diversity seen in resistant clones after an initial response
MAPK pathway effector inhibition.78 This mechanism is re- to the treatment.100,101 Even if the mechanisms leading to
sponsible for acquired resistance to BRAF inhibitors in the establishment of these persister cells are still unclear,
BRAF-mutant melanoma and CRC.27,79,80 In melanoma, these most probably are not genetically determined and
MAPK reactivation occurs by bypassing BRAF activation,81,82 involve the activation of a stress response that contrasts
and the combination of a BRAF and a MEK inhibitor is able to the cellular deprivation from oncogenic signaling. In this
prevent it and represents an established treatment of BRAF-
framework, co-targeting of the oncogenic pathway and the
mutant melanoma.20 On the other hand, in BRAF-mutant
consequential pathway-independent cellular response may
CRC, the feedback loop elicited by BRAF blockade involves
constitute a valid approach to prevent or overcome acquired
EGFR reactivation,27 and the combined inhibition of EGFR
resistance (Fig. 1B).
and BRAF has recently shown substantial clinical activity that
has been incorporated in clinical guidelines.29,30,83 In- Several reports have identified, in chromatin remodeling
terestingly, co-targeting of EGFR and MEK in CRC has been and epigenetic changes, a causative role in the drug-
shown to intercept different resistance mechanisms to anti- tolerant phenotype, showing the potential utility of com-
EGFR monoclonal antibodies that converge on MAPK bining EGFR-TKI with histone deacetylase (HDAC) or
reactivation and may prevent or even overcome the onset of KDM5A inhibitors in EGFR-mutated NSCLC to overcome
resistance to anti-EGFR agents.84-86 acquired resistance.97 Other researchers have reported how
blocking transcriptional or translational adaptive responses
Target Modification
using a cyclin-dependent kinase (CDK) 7/12 inhibitor or
Target mutation is a frequent event associated with acquired a eukaryotic initiation factor 4A inhibitor, respectively, may
resistance to tyrosine kinase inhibitors caused by the hamper the onset of persister cells.99,102 Moreover, the
structural change of the drug-binding site in the oncogene. adaptive stress responses generated in drug-tolerant cells
This mechanism has been described for several different create potential liabilities in the oxidative stress response
oncogenic RTKs or fusion proteins, including EGFR, ALK, that may be exploited for therapeutic combinations.103,104
NTRK, and BCR-ABL.87-90 In these cases, switching to
Another pathway-independent mechanism of resistance to
another agent characterized by a different binding profile is
targeted therapy consists of apoptosis evasion, achieved by
the most used approach and underlines the importance of
monitoring genetic evolution of the disease.91 an imbalance between pro- and antiapoptotic proteins.67,105,106
This knowledge has led to the development of rational
Mutations of the extracellular domain of EGFR, such as combinations of targeted therapy with inhibitors of anti-
S492R, constitute 1% to 2% of the cases of acquired re- apoptotic proteins, which in some cases are already in
sistance to cetuximab and can be overcome by using a mix clinical testing, such as the combination of osimertinib
of different epitope-binding anti-EGFR antibodies such as and navitoclax after progression to EGFR-TKI in EGFR-
Sym004 or MM151.92-95 mutant NSCLC (NCT02520778) or the combination of
PATHWAY-INDEPENDENT RESISTANCE MECHANISMS: trametinib and navitoclax in chemo-refractory RAS-
CO-TARGETING OF THE ONCOGENIC PATHWAY AND ITS mutated cancers (NCT02079740).107,108
STRESS RESPONSE Finally, the tumor microenvironment has recently emerged
The acquisition of resistance to therapy constitutes a mul- as a novel factor in the acquisition of resistance to targeted
tifactorial process that culminates in the selection of one therapy by providing growth factors and decreasing drug
or more clones with decreased sensitivity to the treatment and immune cell penetration.109 Although research is at an
and, eventually, increased replicative fitness. Many research early stage in this field, the preclinical rationale to combine

Gumusay et al
targeted therapy with tumor microenvironment–directed

agents is strong and technically feasible.110
COMBINATIONS TO ENHANCE THE EFFICACY OF

HORMONAL THERAPY
Approximately 70% of all invasive breast cancers express
HRs.111 ET, including selective estrogen receptor (ER)
modulators (tamoxifen), aromatase inhibitors (letrozole,
anastrozole, exemestane), and selective ER downregulators
(fulvestrant), remain the mainstay of treatment of HR+/
HER2– advanced disease. However, effectiveness is
limited by intrinsic and acquired endocrine resistance.
Oncogenic drivers and resistance pathways include
mutations in the ESR1 gene, crosstalk between the ER and
growth factor signaling, activation of the PI3K-Akt-mTOR
pathway, dysregulation of cell cycle progression, epigenetic
modifications by HDAC, and interactions with the tumor
microenvironment (Fig. 3).112
Based on these specific mechanisms, combining ET with
targeted agents including CDK4/6 inhibitors (CDK4/6i),
selective inhibitors of PI3Kα, and mTOR inhibitors has
resulted in improved outcome. Additional agents, such as
AKT and HDAC inhibitors, are being actively investigated. In
this section, we will summarize the rationale and clinical
data behind various targeted-agent combinations.
Established Combination Therapy

CDK4/6 inhibitors ET can reduce cellular proliferation, in-
duce apoptosis, and reduce growth rate as a result of cell
cycle arrest in the G1 phase.116-118 The G1 to S phase
transition is regulated by CDK4/6, which is activated by
binding to D-type cyclins, leading to phosphorylation of the
retinoblastoma susceptibility (RB1) gene product.119 Se-
lective inhibition of CDK4/6 inhibited proliferation of luminal
HR+ human breast cancer cell lines in vitro,120 leading to
the phase II PALOMA-1 trial, which demonstrated marked
improvement in PFS with the addition of the CDK4/6i
palbociclib to letrozole compared with letrozole alone as first-
line therapy for postmenopausal women with HR+ metastatic
breast cancer (MBC).121 Based on these data, the FDA
granted accelerated approval for palbociclib as first-line
treatment HR+ MBC in 2015.122
FIGURE 3. (Continued). pathway (c), dysregulation of cell cycle

progression via CDK4/6 (d), angiogenesis and its VEGFR signaling
pathway (e; adapted from Young et al114), interactions with the
tumor microenvironment (f), and epigenetic modifications by HDAC
(g). For PI3K pathway activation, the pink color codes activating
FIGURE 3. Mechanisms of Endocrine Resistance signals and green codes suppression signals (adapted from Rugo
Oncogenic drivers and important mechanisms of endocrine re- et al115). Abbreviations: ER, estrogen receptor; PR, progesterone
sistance include ER/PR mutations and alterations in expression (a; receptor; RTK, receptor tyrosine kinase; PI3K, phosphatidylinositol
adapted from Protein Data Bank113), crosstalk between ER and 30 -kinase; CDK4/6, cyclin-dependent kinase 4/6; HDAC, histone
receptor tyrosine kinases (b), activation of the PI3K-Akt-mTOR deacetylase.

Since that time, a number of phase III trials have demon- implicated pathways, including amplification of cyclin E and
strated marked improvements in outcome in combination loss of Rb.148-151 Numerous trials are also evaluating CDK4/
with ET (Table 1), leading to regulatory approval of three oral 6i as therapy for high-risk HR+ BC in the adjuvant setting.
CDK4/6i (palbociclib, ribociclib, and abemaciclib) for the PALLAS (NCT02513394)152 and MonarchE (NCT03155997)
treatment of MBC.123 The addition of CDK4/6i to ET sig- are evaluating palbociclib (PALLAS) or abemaciclib (Mon-
nificantly increased PFS in all settings and improved OS in archE) for 2 years in addition to adjuvant ET; NATALEE
several studies (Table 1). (NCT03701334) extends the duration of ribociclib with ET to
All three CDK4/6i demonstrated improvement in PFS when 3 years. PENELOPE-B (NCT01864746) is evaluating pal-
combined with ET in the first-line metastatic setting, with bociclib in addition to standard ET for 1 year in patients with
remarkably similar hazard ratios. All but one trial evaluated residual disease after neoadjuvant chemotherapy.
nonsteroidal AIs and excluded patients on chemical ovarian The theory that continued CDK4/6 inhibition may be ef-
suppression (Table 1). OS data remain immature. MON- fective post-progression on CDK4/6i therapy is being eval-
ALEESA-7 is the only trial that focused on premenopausal uated in ongoing studies such as PACE and MAINTAIN. The
women on chemical ovarian suppression.133,134 The addi- phase II PACE trial is evaluating the activity of fulvestrant
tion of ribociclib to first-line nonsteroidal AI or tamoxifen alone; fulvestrant and palbociclib; or fulvestrant, palboci-
significantly improved PFS, with hazard ratios similar to what clib, and avelumab combined, in patients with HR+ MBC
was seen for postmenopausal women, and improved OS in whose disease has previously stopped responding to prior
a protocol-specified interim analysis, with the median OS in palbociclib and ET (NCT03147287). The randomized
the combination arm not yet reached. MAINTAIN phase II trial is investigating fulvestrant alone or
Several trials evaluated CDK4/6i with fulvestrant in the fulvestrant with ribociclib in patients whose disease has
second-line or greater setting, with relapse on or after progressed on an AI plus a CDK4/6i, including either pal-
progression on nonsteroidal AIs. PFS was also significantly bociclib or ribociclib (NCT02632045).
prolonged, with similar hazard ratios across trials (Table 1).
MONALEESA-3 included first-line patients as well.131,132 CDK4/6 Inhibitors in HER2+ Breast Cancer
Interestingly, PFS for the fulvestrant control arm was Fifteen percent to 20% of all breast cancers are HER2+, and
shorter in the palbocicib trial, PALOMA-3, compared with roughly half of these tumors are also HR+ .153 The cyclin D1/
MONALEESA-3 and MONARCH-2, the trials evaluating CDK4/6/pRb axis is activated by HER2 ligand interaction
ribociclib and abemaciclib.126,127,131,136 PALOMA-3 allowed with the PI3K-Akt pathway, and downstream activation
prior chemotherapy, which was given in about one-third of of cyclin D1 can induce resistance to trastuzumab and
the trial population.126-128 This suggests that prior chemo- other HER2-targeted treatments.154,155 Previous preclinical
therapy results in greater resistance to ET or that prior studies have suggested that inhibition of CDK4/6i enhances
chemotherapy identifies a group of patients with more the activity of HER2-targeted agents by re-sensitizing re-
endocrine-resistant disease. OS was improved in both trials sistant tumors to HER2 blockade.156,157 The phase II
not allowing prior chemotherapy (ribociclib and abemaci- monarcHER trial randomly assigned 237 pretreated post-
clib) but was only improved in the endocrine-sensitive menopausal women with pre-treated HR+, HER2+ MBC to
subset of PALOMA-2.124,125 receive the triplet of abemaciclib, trastuzumab, and ful-
Abemaciclib has also been evaluated as a single agent, at vestrant; abemaciclib and trastuzumab; or trastuzumab and
a higher dose than that used in combination with ET, and in chemotherapy. The primary endpoint—PFS comparing the
patients who have received up to two lines of prior triplet versus trastuzumab and chemotherapy—was im-
chemotherapy.135-137 Although these data showed clinically proved with the triplet with a hazard ratio of 0.673, repre-
important single-agent activity, combination therapy has senting an absolute benefit of 2.6 months (p = .0506), along
largely replaced the late-line, single-agent approach. with more than a doubling of response.158 In the neo-
adjuvant setting, the NA-PHER2 phase II trial treated 30
CDK4/6i are generally well tolerated. The most common grade patients with early-stage HR+, HER2+ breast cancer with
3 or greater toxicity for palbociclib and ribociclib is neutropenia the combination of trastuzumab, pertuzumab, palbociclib,
without an increase in febrile neutropenia,126,129-131,133 and, and fulvestrant (NCT02530424). Change in Ki67 was the
for abemaciclib, it is diarrhea, although neutropenia also primary endpoint in this single-arm trial; in 22 evaluable
occurs.135,136,138,139 Additional unique toxicities occur with patients, there was a significant decrease. A clinical re-
specific agents, which may require special monitoring. sponse was seen in 29 patients, and eight (27%) had
Based on these data, CDK4/6i combined with ET is now the a pathologic complete response in breast and axillary
standard of care in the first- or second-line setting in HR+/ nodes.159 This chemotherapy-free strategy could be a pos-
HER2 MBC. The molecular mechanisms of resistance sible treatment option in the neoadjuvant setting for elderly
to CDK4/6i are under active investigation, with a number of patients or for those with poor performance status, to avoid

TABLE 1. Phase II/III Trials of Targeted Agents to Enhance Efficacy of ET
e300
Targeted
Agent Number of
Class Agent Trial Randomization Phase Line of Therapy Therapy Patients PFS (mos), HR OS (mos), HR
121
CDK4/6i Palbociclib PALOMA-1 Yes 1:1 II First Palbo/ LET vs. 165 20.2 vs. 10.2; HR, 0.48 37.5 vs. 34.5; HR, 0.89 (NS)
LET
PALOMA-2124,125 Yes 2:1 III First Palbo/LET vs. 666 27.6 vs. 14.5; HR, 0.58 NA
PBO/LET
PALOMA-3126-128 Yes 2:1 III Prior ET; 1 prior Palbo/FUL vs. 521 9.5 vs. 4.6; HR, 0.46 34.9 vs. 28.0; HR, 0.81
chemotherapy* PBO/FUL
Ribociclib MONALEESA-2129,130 Yes 1:1 III First Ribo/LET vs. 668 25.3 vs. 16.0; HR, 0.56 NR vs.
PBO/LET 33.0; HR, 0.74
MONALEESA-3131,132 Yes 2:1 III First or Ribo/FUL vs. 726 20.5 vs. 12.8; HR, 0.59 NR vs.
second; no PBO/FUL 40.0; HR, 0.72
chemotherapy
MONALEESA-7133,134 Yes 1:1 III First; 1 prior Ribo/ET vs. 672 23.8 vs. 13; HR, 0.55 NR vs.

chemotherapy* PBO/ET** 40.9; HR, 0.71
Abemaciclib MONARCH-1135 No II Prior ET; 1-2 Abema 184 6 22.3
lines of
chemotherapy
MONARCH-2136,137 Yes 2:1 III Prior ET; no Abema/FUL vs. 669 16.4 vs. 9.3; HR, 0.55 46.7 vs. 37.3; HR, 0.75
chemotherapy PBO/FUL
MONARCH-3138,139 Yes 2:1 III First Abema/ANA or 493 28.1 vs. 14.7; HR, 0.54 NA
Gumusay et al
LET vs. PBO/

ANA or LET
PI3K/AKT Everolimus TAMRAD140 Yes 1:1 II Prior ET EVE/TAM vs. TAM 111 8.6 vs. 4.5; HR, 0.54 NA
141,142
BOLERO-2 Yes 2:1 III Prior ET EVE/EXE vs. PBO/ 724 7.8 vs. 3.2; HR, 0.45 31.0 vs. 26.6; HR, 0.89
EXE
Alpelisib SOLAR-1143 Yes 1:1 III Prior ET ALP/FUL vs. PBO/ 341 11.0 vs. 5.7; HR, 0.65 NA
FUL
Capivasertib FAKTION144 Yes 1:1 II Prior ET FUL/Capi vs. 140 10.3 vs. 4.8; HR, 0.58 26.0 vs. 20.0; HR, 0.59
PBO/Capi

Anti-VEGF Bevacizumab LEA145 Yes 1:1 III First BEV/LET or FUL 374 19.3 vs. 14.4; HR, 0.83 (NS) 52.1 vs. 51.8; HR, 0.87 (NS)
vs. LET or FUL

CALGB 40503146 Yes 1:1 III First, 1 prior BEV/LET vs. LET 343 20.2 vs. 15.6; HR, 0.75 47.2 vs. 43.9; HR, 0.87 (NS)
chemotherapy*
HDACi Entinostat ENCORE 301147 Yes 1:1 II Prior ET ENT/EXE vs. PBO/ 130 4.3 vs. 2.3; HR, 0.73 28.1 vs. 19.8; HR, 0.59
EXE
Abbreviations: ET, endocrine therapy; PFS, progression-free survival; OS, overall survival; CDK4/6i, cyclin-dependent kinase 4/6 inhibitors; palbo, palbociclib; LET, letrozole; HR, hazard ratio; NS,
not significant; PBO, placebo; NA, not available; FUL, fulvestrant; ribo, ribociclib; NR, not reached; abema, abemaciclib; ANA, anastrozole; PI3K/AKT, phosphatidylinositol 30 -kinase-Akt; EVE,
everolimus; TAM, tamoxifen; EXE, exemestane; ALP, alpelisib; Capi, capivasertib; BEV, bevacizumab; HDACi, histone deacetylase inhibitor; ENT, entinostat.
*One prior line of chemotherapy in the advanced setting allowed.
**Plus gonadotropin-releasing hormone therapy for chemical ovarian suppression.
chemotherapy-related toxicities. Clearly, additional data are delay resistance to ET,150 TRINITI-1 evaluated triplet ther-
needed before this can be incorporated into clinical apy with ribociclib, low-dose everolimus, and exemestane in
practice to understand the value and risk of adding the patients with ET-refractory HR+/HER2 MBC and disease
CDK4/6i to this regimen. A number of clinical trials are progression on CDK4/6i. Clinical benefit at 24 weeks was
evaluating the impact of adding CDK4/6i to HER2-targeted seen in 41%, with toxicity including neutropenia and sto-
therapy in patients with HR+/HER2+ disease. PATINA, matitis.167 The combination of alpelisib with ribociclib and
a phase III ongoing trial in patients with HR+/HER2+ MBC, ET was associated with substantial toxicity. Further studies
is evaluating trastuzumab with or without pertuzumab in of triplet therapy are ongoing with AKT inhibitors (see AKT
combination with ET (AI or fulvestrant) with or without Inhibitors section).
palbociclib following induction chemotherapy in the first-
PIK3CA Studies investigating pan-class I PI3K inhibitors
line setting (NCT02947685). One area of interest in ad-
(buparlisib and pictilisib) resulted in disappointing efficacy
dition to overall disease control is the ability of CDK4/6i in
as well as toxicity.168-170 Taselisib, a beta-sparing PI3K in-
HER2+ disease to cross the blood–brain barrier, potentially
hibitor, was limited by gastrointestinal toxicity, although
offering some therapeutic advantage for patients with or at
antitumor activity was seen in patients whose tumors had
risk for brain metastases. Ongoing trials will further elu-
PIK3CA mutations.171
cidate the role of CDK4/6i in the treatment of HER2+ breast
cancer. In contrast, the alpha-specific PI3K inhibitor alpelisib
demonstrated clinical efficacy with fulvestrant in patients
Despite effective therapy in the first- or second-line settings,
with PIK3CA-mutated tumors in a phase Ib trial,172 which
disease progression invariably occurs. A number of targeted
was confirmed in the phase III SOLAR-1 trial (Table 1) and
agents have been explored in this setting, with encouraging
led to regulatory approval of the combination therapy and
results. Selected data are summarized in the following
use of ctDNA to detect PIK3CA mutations. In SOLAR-1,
sections.160
patients with HR+ PIK3CA-mutated MBC progressing on
Targeting the PI3K/AKT/mTOR pathway The PI3K/Akt/mTOR AIs who were randomly assigned to receive fulvestrant plus
pathway has been shown to play a role in endocrine re- alpelisib had significantly longer PFS compared with those
sistance. Approximately 40% of patients with HR+/HER2 receiving fulvestrant and placebo.143 The most frequent
disease have PIK3CA mutations that induce hyperactivation grade III/IV adverse events were hyperglycemia, rash, and
of the alpha isoform of PI3K.161 The PI3K/AKT and the diarrhea,173 which can generally be controlled with early
MAPK pathways, activated by estrogen activity, down- intervention or prophylaxis. Few patients in SOLAR-1 re-
regulate ER and PR cell surface expression.162 Increased ceived prior CDK4/6i. BYLieve is an ongoing trial to assess
PI3K/AKT signaling can promote HR deactivation and tumor the efficacy of alpelisib in patients with PIK3CA mutations
growth in the absence of endocrine signaling and directly after treatment with a CDK4/6i (NCT03056755); preliminary
phosphorylate ER, resulting in estrogen-independent acti- data are encouraging.174 Several additional PIK3CA in-
vation of HRs, providing a secondary mechanism of ET hibitors are in clinical trials.
resistance.163 Agents targeting the PI3K pathway have been
evaluated in clinical trials, as outlined in the PIK3CA section AKT Inhibitors
(Fig. 3). Capivasertib is a potent selective oral inhibitor of all three
isoforms of the serine/threonine kinase AKT.175 The phase II
mTOR inhibitors Several studies have revealed that in-
FAKTION trial evaluated capivasertib plus fulvestrant
hibition of the mTOR pathway can improve the duration of
compared with placebo plus fulvestrant in patients with
response to ET. Based on encouraging data with the mTOR
progression on AIs (Table 1).144 Median PFS was signifi-
inhibitor everolimus combined with ET in the neoadjuvant
cantly prolonged with addition of capivasertib to fulvestrant,
setting,164 everolimus was evaluated in combination with
with no impact from PIK3CA mutation status. The most
exemestane in patients with ET-resistant MBC in the BO-
common grade 3/4 toxicities were rash and diarrhea, with
LERO-2 trial.141 PFS was significantly longer in patients
hyperglycemia seen in only 4%. CAPitello-291 is a phase III
treated with everolimus and exemestane compared with
trial designed to confirm these results (NCT04305496), and
exemestane alone (Table 1).142 The most common toxic-
IPATunity150 is evaluating another AKT inhibitor, ipatasertib,
ity was stomatitis, with grade 3 stomatitis seen in 8%. A
with palbociclib and fulvestrant (NCT04060862).
subsequent study showed a marked decrease in stomatitis
with preventive steroid mouthwash.165 Two additional phase Others
II trials showed improved PFS with everolimus combined Targeting VEGF Angiogenesis and its VEGFR signaling
with tamoxifen140 or fulvestrant compared with ET alone.166 pathway are potential mechanisms of resistance to ET.176
Based on preclinical data suggesting that the combination Therefore, targeting VEGF and the VEGFR pathway was
of agents targeting the PIK3CA pathway with CDK4/6i could hypothesized to be a way to reverse ET resistance.

Gumusay et al
Two phase III randomized trials (LEA and Cancer and bevacizumab, and abemaciclib (NCT03280563). Based
Leukemia Group B [CALGB 40503]) compared ET with ET on strong preclinical data, studies have evaluated the
plus bevacizumab (ET-B). In the LEA trial, the addition of combination of checkpoint inhibition, hormone therapy,
bevacizumab numerically improved PFS, but this difference and CDK4/6i; however, a recent trial suggested enhanced
was not statistically significant.145 In contrast, in CALGB immune toxicity with this therapy. The ongoing PACE trial
40503, the addition of bevacizumab to ET resulted in (described in the CDK4/6 Inhibitors section) includes an
a substantial improvement in PFS in HR+ MBC.146 A pooled arm with triplet therapy.
analysis of the two trials reported improved PFS, objective
response rate, and clinical benefit rate with bevacizumab.177 CONCLUSION
The incidence of grade 3/4 adverse events was higher in the Resistance to therapy remains a challenge in the treatment
ET-B arm, including hypertension and proteinuria. Although of both early- and late-stage malignancies. Combinations of
the efficacy data were intriguing, the role of bevacizumab in targeted agents that block more than one driver of disease
treating breast cancer remains unclear. have the potential to overcome adaptive resistance. Co-
HDAC inhibitors Epigenetic modifications in breast cancer targeting the oncogenic pathway and the subsequent
lead to resistance to ET that could potentially be modulated pathway-independent cellular response may be an effective
with HDAC inhibitors.178 Previous in vitro studies showed approach to prevent or overcome acquired resistance. Al-
that HDAC inhibitors enhance the activity of and restore though drug combinations have increased survival, further
sensitivity to ET in ER+ cell lines.179,180 Entinostat is an oral predictive markers for response to therapy must be de-
synthetic benzamide derivative with a long half-life.181 The veloped. One such tool involves liquid biopsy, which can
ENCORE 301 trial was a phase II trial that randomly monitor tumor evolution through the study of ctDNA and
assigned patients with MBC progressing on AI therapy to capture tumor and environmental changes before clinical
receive entinostat or placebo with exemestane.147 Entinostat evidence of disease progression.
did not improve PFS but significantly improved OS, an In HR+ breast cancer, understanding pathways of acquired
exploratory secondary endpoint. Toxicity included neu- and primary endocrine resistance has led to approvals of
tropenia and fatigue. The phase III trial E2112 randomly hormone therapy in combination with targeted agents, such
assigned patients with HR+ MBC with disease progression as CDK4/6i, that have markedly improved outcome. The
after a nonsteroidal AI to receive entinostat or placebo with challenge is to identify markers of sensitivity and resistance
exemestane; results are expected in 2021. that will help direct therapy and understand how these
Immunotherapy There is tremendous interest in immu- agents are best used in early-stage disease.
notherapy in breast cancer, with studies focusing on
triple-negative and HER2+ disease based on increased ACKNOWLEDGMENT
expression of PD-L1 in these disease subsets. Single- O. Gumusay and P. P. Vitiello contributed equally to this
agent checkpoint inhibitor therapy had limited efficacy article. R.B.C is supported by NIH/NCI Gastrointestinal
in HR+ breast cancer,182 spurring interest in combi- Cancer SPORE P50 CA127003, R01CA208437, and
nation studies designed to increase tumor immunoge- U54CA224068. A.B is supported by AIRC IG n.21923;
nicity and host immune response. MORPHEUS is an AIRC-CRUK-FC AECC Accelerator Award contract 22795;
ongoing phase I/II trial evaluating the safety and efficacy Fondazione AIRC under 5 per Mille 2018 ID. 21091 pro-
of fulvestrant and immunotherapy combined with multiple gram Alberto Fondazione Piemontese per la Ricerca sul
agents. Experimental arms include various combinations Cancro-ONLUS 5 per mille 2014 e 2015 ministero della
of drugs such as atezolizumab, ipatasertib, entinostat, Salute; and Ministero Salute, RC 2019.

1
UCSF Helen Diller Family Comprehensive Cancer Center, San Francisco, Hope S. Rugo, MD, UCSF Helen Diller Family Comprehensive Cancer
CA Center, 1600 Divisadero Ave., Second Floor, Box 1710, San Francisco,
2
Department of Internal Medicine, Division of Medical Oncology, CA 94115; Twitter: @hoperugo; email: [email protected].
Gaziosmanpasa University Faculty of Medicine, Tokat, Turkey
3
Department of Oncology, University of Torino, Candiolo (TO), Italy
4
Dipartimento di Medicina di Precisione, Unità di Oncologia Medica, AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Università degli Studi della Campania Luigi Vanvitelli, Italy AND DATA AVAILABILITY STATEMENT
5
Massachusetts General Hospital Cancer Center, Boston, MA Disclosures provided by the authors and data availability statement (if
6
Candiolo Cancer Institute, Candiolo (TO), Italy applicable) are available with this article at DOI https://doi.org/10.1200/
EDBK_280845.

REFERENCES
1. U.S. Food and Drug Administration. Advancing health through innovation: 2018 new drug therapy approvals. https://www.fda.gov/drugs/new-drugs-fda-cders-
new-molecular-entities-and-new-therapeutic-biological-products/new-drug-therapy-approvals-2019. Accessed March 18, 2020.
2. Vasan N, Baselga J, Hyman DM. A view on drug resistance in cancer. Nature. 2019;575:299-309.
3. Weinstein IB. Cancer. Addiction to oncogenes--the Achilles heal of cancer. Science. 2002;297:63-64.
4. Sawyers CL. Shifting paradigms: the seeds of oncogene addiction. Nat Med. 2009;15:1158-1161.
5. Soria J-C, Blay JY, Spano JP, et al. Added value of molecular targeted agents in oncology. Ann Oncol. 2011;22:1703-1716.
6. Camidge DR. Targeted therapy vs chemotherapy: which has had more impact on survival in lung cancer? Does targeted therapy make patients live longer? Hard
to prove, but impossible to ignore. Clin Adv Hematol Oncol. 2014;12:763-766.
7. Pegram MD, Pietras R, Bajamonde A, et al. Targeted therapy: wave of the future. J Clin Oncol. 2005;23:1776-1781.
8. Konieczkowski DJ, Johannessen CM, Garraway LA. A convergence-based framework for cancer drug resistance. Cancer Cell. 2018;33:801-815.
9. Lopez JS, Banerji U. Combine and conquer: challenges for targeted therapy combinations in early phase trials. Nat Rev Clin Oncol. 2017;14:57-66.
10. Crowley E, Di Nicolantonio F, Loupakis F, et al. Liquid biopsy: monitoring cancer-genetics in the blood. Nat Rev Clin Oncol. 2013;10:472-484.
11. Bollag G, Hirth P, Tsai J, et al. Clinical efficacy of a RAF inhibitor needs broad target blockade in BRAF-mutant melanoma. Nature. 2010;467:596-599.
12. Garraway LA, Jänne PA. Circumventing cancer drug resistance in the era of personalized medicine. Cancer Discov. 2012;2:214-226.
13. Corcoran RB, Ebi H, Turke AB, et al. EGFR-mediated re-activation of MAPK signaling contributes to insensitivity of BRAF mutant colorectal cancers to RAF
inhibition with vemurafenib. Cancer Discov. 2012;2:227-235.
14. Lito P, Saborowski A, Yue J, et al. Disruption of CRAF-mediated MEK activation is required for effective MEK inhibition in KRAS mutant tumors. Cancer Cell.
2014;25:697-710.
15. Bosch A, Li Z, Bergamaschi A, et al. PI3K inhibition results in enhanced estrogen receptor function and dependence in hormone receptor-positive breast
cancer. Sci Transl Med. 2015;7:283ra51.
16. Canon J, Rex K, Saiki AY, et al. The clinical KRAS(G12C) inhibitor AMG 510 drives anti-tumour immunity. Nature. 2019;575:217-223.
17. Flaherty KT, Puzanov I, Kim KB, et al. Inhibition of mutated, activated BRAF in metastatic melanoma. N Engl J Med. 2010;363:809-819.
18. Smalley KS, Flaherty KT. Integrating BRAF/MEK inhibitors into combination therapy for melanoma. Br J Cancer. 2009;100:431-435.
19. Flaherty KT, Infante JR, Daud A, et al. Combined BRAF and MEK inhibition in melanoma with BRAF V600 mutations. N Engl J Med. 2012;367:1694-1703.
20. Larkin J, Ascierto PA, Dréno B, et al. Combined vemurafenib and cobimetinib in BRAF-mutated melanoma. N Engl J Med. 2014;371:1867-1876.
21. Dummer R, Ascierto PA, Gogas HJ, et al. Encorafenib plus binimetinib versus vemurafenib or encorafenib in patients with BRAF-mutant melanoma (CO-
LUMBUS): a multicentre, open-label, randomised phase 3 trial. Lancet Oncol. 2018;19:603-615.
22. Davies H, Bignell GR, Cox C, et al. Mutations of the BRAF gene in human cancer. Nature. 2002;417:949-954.
23. Richman SD, Seymour MT, Chambers P, et al. KRAS and BRAF mutations in advanced colorectal cancer are associated with poor prognosis but do not preclude
benefit from oxaliplatin or irinotecan: results from the MRC FOCUS trial. J Clin Oncol. 2009;27:5931-5937.
24. Kopetz S, Desai J, Chan E, et al. Phase II pilot study of vemurafenib in patients with metastatic BRAF-mutated colorectal cancer. J Clin Oncol. 2015;
33:4032-4038.
25. Mason JM, Morrison DJ, Basson MA, et al. Sprouty proteins: multifaceted negative-feedback regulators of receptor tyrosine kinase signaling. Trends Cell Biol.
2006;16:45-54.
26. Montero-Conde C, Ruiz-Llorente S, Dominguez JM, et al. Relief of feedback inhibition of HER3 transcription by RAF and MEK inhibitors attenuates their
antitumor effects in BRAF-mutant thyroid carcinomas. Cancer Discov. 2013;3:520-533.
27. Prahallad A, Sun C, Huang S, et al. Unresponsiveness of colon cancer to BRAF(V600E) inhibition through feedback activation of EGFR. Nature. 2012;
483:100-103.
28. Whittaker SR, Cowley GS, Wagner S, et al. Combined Pan-RAF and MEK inhibition overcomes multiple resistance mechanisms to selective RAF inhibitors. Mol
Cancer Ther. 2015;14:2700-2711.
29. Kopetz S, Grothey A, Yaeger R, et al. Encorafenib, binimetinib, and cetuximab in BRAF V600E-mutated colorectal cancer. N Engl J Med. 2019;381:1632-1643.
30. Corcoran RB, André T, Atreya CE, et al. Combined BRAF, EGFR, and MEK inhibition in patients with BRAFV600E-mutant colorectal cancer. Cancer Discov. 2018;
8:428-443.
31. U.S. Food and Drug Administration. FDA approves encorafenib in combination with cetuximab for metastatic colorectal cancer with a BRAF V600E mutation.
https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-encorafenib-combination-cetuximab-metastatic-colorectal-cancer-braf-
v600e-mutation. Accessed April 22, 2020.
32. Ruess DA, Heynen GJ, Ciecielski KJ, et al. Mutant KRAS-driven cancers depend on PTPN11/SHP2 phosphatase. Nat Med. 2018;24:954-960.
33. Fedele C, Ran H, Diskin B, et al. SHP2 inhibition prevents adaptive resistance to MEK inhibitors in multiple cancer models. Cancer Discov. 2018;8:1237-1249.
34. Mainardi S, Mulero-Sánchez A, Prahallad A, et al. SHP2 is required for growth of KRAS-mutant non-small-cell lung cancer in vivo. Nat Med. 2018;24:961-967.

Gumusay et al
35. Sun C, Hobor S, Bertotti A, et al. Intrinsic resistance to MEK inhibition in KRAS mutant lung and colon cancer through transcriptional induction of ERBB3. Cell
Rep. 2014;7:86-93.
36. Manchado E, Weissmueller S, Morris JP IV, et al. A combinatorial strategy for treating KRAS-mutant lung cancer. Nature. 2016;534:647-651.
37. Ryan MB, Fece de la Cruz F, Phat S, et al. Vertical pathway inhibition overcomes adaptive feedback resistance to KRASG12C inhibition. Clin Cancer Res. 2020;
26:1633-1643.
38. Xue JY, Zhao Y, Aronowitz J, et al. Rapid non-uniform adaptation to conformation-specific KRAS(G12C) inhibition. Nature. 2020;577:421-425.
39. Hallin J, Engstrom LD, Hargis L, et al. The KRASG12C inhibitor MRTX849 provides insight toward therapeutic susceptibility of KRAS-mutant cancers in mouse
models and patients. Cancer Discov. 2020;10:54-71.
40. Chen YN, LaMarche MJ, Chan HM, et al. Allosteric inhibition of SHP2 phosphatase inhibits cancers driven by receptor tyrosine kinases. Nature. 2016;
535:148-152.
41. Ahronian LG, Sennott EM, Van Allen EM, et al. Clinical acquired resistance to RAF inhibitor combinations in BRAF-mutant colorectal cancer through MAPK
pathway alterations. Cancer Discov. 2015;5:358-367.
42. Hazar-Rethinam M, Kleyman M, Han GC, et al. Convergent therapeutic strategies to overcome the heterogeneity of acquired resistance in BRAFV600E
colorectal cancer. Cancer Discov. 2018;8:417-427.
43. Dagogo-Jack I, Shaw AT. Tumour heterogeneity and resistance to cancer therapies. Nat Rev Clin Oncol. 2018;15:81-94.
44. Alizadeh AA, Aranda V, Bardelli A, et al. Toward understanding and exploiting tumor heterogeneity. Nat Med. 2015;21:846-853.
45. Russo M, Siravegna G, Blaszkowsky LS, et al. Tumor heterogeneity and lesion-specific response to targeted therapy in colorectal cancer. Cancer Discov. 2016;
6:147-153.
46. Gerlinger M, Rowan AJ, Horswell S, et al. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med. 2012;
366:883-892.
47. Chabon JJ, Simmons AD, Lovejoy AF, et al. Circulating tumour DNA profiling reveals heterogeneity of EGFR inhibitor resistance mechanisms in lung cancer
patients. Nat Commun. 2016;7:11815.
48. Siravegna G, Mussolin B, Buscarino M, et al. Clonal evolution and resistance to EGFR blockade in the blood of colorectal cancer patients. Nat Med. 2015;
21:795-801.
49. Parikh AR, Leshchiner I, Elagina L, et al. Liquid versus tissue biopsy for detecting acquired resistance and tumor heterogeneity in gastrointestinal cancers. Nat
Med. 2019;25:1415-1421.
50. Blakely CM, Watkins TBK, Wu W, et al. Evolution and clinical impact of co-occurring genetic alterations in advanced-stage EGFR-mutant lung cancers. Nat
Genet. 2017;49:1693-1704.
51. Iwama E, Sakai K, Hidaka N, et al. Longitudinal monitoring of somatic genetic alterations in circulating cell-free DNA during treatment with epidermal growth
factor receptor-tyrosine kinase inhibitors. Cancer. 2020;126:219-227.
52. Ku S-Y, Gleave ME, Beltran H. Towards precision oncology in advanced prostate cancer. Nat Rev Urol. 2019;16:645-654.
53. Esposito A, Bardelli A, Criscitiello C, et al. Monitoring tumor-derived cell-free DNA in patients with solid tumors: clinical perspectives and research opportunities.
Cancer Treat Rev. 2014;40:648-655.
54. Malone ER, Oliva M, Sabatini PJB, et al. Molecular profiling for precision cancer therapies. Genome Med. 2020;12:8.
55. Michels S, Heydt C, van Veggel B, et al. Genomic profiling identifies outcome-relevant mechanisms of innate and acquired resistance to third-generation
epidermal growth factor receptor tyrosine kinase inhibitor therapy in lung cancer. Precision Oncol. 2019;3:1-14.
56. Levine AJ, Jenkins NA, Copeland NG. The roles of initiating truncal mutations in human cancers: the order of mutations and tumor cell type matters. Cancer Cell.
2019;35:10-15.
57. Ciardiello F, Tortora G. EGFR antagonists in cancer treatment. N Engl J Med. 2008;358:1160-1174.
58. De Roock W, Claes B, Bernasconi D, et al. Effects of KRAS, BRAF, NRAS, and PIK3CA mutations on the efficacy of cetuximab plus chemotherapy in
chemotherapy-refractory metastatic colorectal cancer: a retrospective consortium analysis. Lancet Oncol. 2010;11:753-762.
59. Misale S, Yaeger R, Hobor S, et al. Emergence of KRAS mutations and acquired resistance to anti-EGFR therapy in colorectal cancer. Nature. 2012;
486:532-536.
60. Misale S, Di Nicolantonio F, Sartore-Bianchi A, et al. Resistance to anti-EGFR therapy in colorectal cancer: from heterogeneity to convergent evolution. Cancer
Discov. 2014;4:1269-1280.
61. Nagasaka M, Li Y, Sukari A, et al. KRAS G12C Game of Thrones, which direct KRAS inhibitor will claim the iron throne? Cancer Treat Rev. 2020;84:101974.
62. Mendoza MC, Er EE, Blenis J. The Ras-ERK and PI3K-mTOR pathways: cross-talk and compensation. Trends Biochem Sci. 2011;36:320-328.
63. Sartore-Bianchi A, Martini M, Molinari F, et al. PIK3CA mutations in colorectal cancer are associated with clinical resistance to EGFR-targeted monoclonal
antibodies. Cancer Res. 2009;69:1851-1857.
64. Vitiello PP, Cardone C, Martini G, et al. Receptor tyrosine kinase-dependent PI3K activation is an escape mechanism to vertical suppression of the EGFR/RAS/
MAPK pathway in KRAS-mutated human colorectal cancer cell lines. J Exp Clin Cancer Res. 2019;38:41.
65. Migliardi G, Sassi F, Torti D, et al. Inhibition of MEK and PI3K/mTOR suppresses tumor growth but does not cause tumor regression in patient-derived xenografts
of RAS-mutant colorectal carcinomas. Clin Cancer Res. 2012;18:2515-2525.

66. Shapiro GI, LoRusso P, Kwak E, et al. Phase Ib study of the MEK inhibitor cobimetinib (GDC-0973) in combination with the PI3K inhibitor pictilisib (GDC-0941) in
patients with advanced solid tumors. Invest New Drugs. 2020;38:419-432.
67. Clarke PA, Roe T, Swabey K, et al. Dissecting mechanisms of resistance to targeted drug combination therapy in human colorectal cancer. Oncogene. 2019;
38:5076-5090.
68. Bardelli A, Corso S, Bertotti A, et al. Amplification of the MET receptor drives resistance to anti-EGFR therapies in colorectal cancer. Cancer Discov. 2013;
3:658-673.
69. Engelman JA, Zejnullahu K, Mitsudomi T, et al. MET amplification leads to gefitinib resistance in lung cancer by activating ERBB3 signaling. Science. 2007;
316:1039-1043.
70. Bertotti A, Migliardi G, Galimi F, et al. A molecularly annotated platform of patient-derived xenografts (“xenopatients”) identifies HER2 as an effective therapeutic
target in cetuximab-resistant colorectal cancer. Cancer Discov. 2011;1:508-523.
71. Takezawa K, Pirazzoli V, Arcila ME, et al. HER2 amplification: a potential mechanism of acquired resistance to EGFR inhibition in EGFR-mutant lung cancers
that lack the second-site EGFRT790M mutation. Cancer Discov. 2012;2:922-933.
72. Postel-Vinay S, Ashworth A. AXL and acquired resistance to EGFR inhibitors. Nat Genet. 2012;44:835-836.
73. Martinelli E, Martini G, Cardone C, et al. AXL is an oncotarget in human colorectal cancer. Oncotarget. 2015;6:23281-23296.
74. Sartore-Bianchi A, Trusolino L, Martino C, et al. Dual-targeted therapy with trastuzumab and lapatinib in treatment-refractory, KRAS codon 12/13 wild-type,
HER2-positive metastatic colorectal cancer (HERACLES): a proof-of-concept, multicentre, open-label, phase 2 trial. Lancet Oncol. 2016;17:738-746.
75. Meric-Bernstam F, Hurwitz H, Raghav KPS, et al. Pertuzumab plus trastuzumab for HER2-amplified metastatic colorectal cancer (MyPathway): an updated
report from a multicentre, open-label, phase 2a, multiple basket study. Lancet Oncol. 2019;20:518-530.
76. Rimassa L, Bozzarelli S, Pietrantonio F, et al. Phase II study of tivantinib and cetuximab in patients with KRAS wild-type metastatic colorectal cancer with
acquired resistance to EGFR inhibitors and emergence of MET overexpression: lesson learned for future trials with EGFR/MET dual inhibition. Clin Colorectal
Cancer. 2019;18:125-132.e2.
77. Leonetti A, Sharma S, Minari R, et al. Resistance mechanisms to osimertinib in EGFR-mutated non-small cell lung cancer. Br J Cancer. 2019;121:725-737.
78. Kochańczyk M, Kocieniewski P, Kozłowska E, et al. Relaxation oscillations and hierarchy of feedbacks in MAPK signaling. Sci Rep. 2017;7:38244.
79. Doudican NA, Orlow SJ. Inhibition of the CRAF/prohibitin interaction reverses CRAF-dependent resistance to vemurafenib. Oncogene. 2017;36:423-428.
80. Lito P, Rosen N, Solit DB. Tumor adaptation and resistance to RAF inhibitors. Nat Med. 2013;19:1401-1409.
81. Johannessen CM, Boehm JS, Kim SY, et al. COT drives resistance to RAF inhibition through MAP kinase pathway reactivation. Nature. 2010;468:968-972.
82. Nazarian R, Shi H, Wang Q, et al. Melanomas acquire resistance to B-RAF(V600E) inhibition by RTK or N-RAS upregulation. Nature. 2010;468:973-977.
83. Hong DS, Morris VK, El Osta B, et al. Phase IB study of vemurafenib in combination with irinotecan and cetuximab in patients with metastatic colorectal cancer
with BRAFV600E mutation. Cancer Discov. 2016;6:1352-1365.
84. Misale S, Arena S, Lamba S, et al. Blockade of EGFR and MEK intercepts heterogeneous mechanisms of acquired resistance to anti-EGFR therapies in
colorectal cancer. Sci Transl Med. 2014;6:224ra26.
85. Troiani T, Napolitano S, Vitagliano D, et al. Primary and acquired resistance of colorectal cancer cells to anti-EGFR antibodies converge on MEK/ERK pathway
activation and can be overcome by combined MEK/EGFR inhibition. Clin Cancer Res. 2014;20:3775-3786.
86. Misale S, Bozic I, Tong J, et al. Vertical suppression of the EGFR pathway prevents onset of resistance in colorectal cancers. Nat Commun. 2015;6:8305.
87. Pao W, Miller VA, Politi KA, et al. Acquired resistance of lung adenocarcinomas to gefitinib or erlotinib is associated with a second mutation in the EGFR kinase
domain. PLoS Med. 2005;2:e73.
88. Russo M, Misale S, Wei G, et al. Acquired resistance to the TRK inhibitor entrectinib in colorectal cancer. Cancer Discov. 2016;6:36-44.
89. Wang H-Y, Ho C-C, Shih J-Y. Multiple acquired resistance mutations of the ALK tyrosine kinase domain after sequential use of ALK inhibitors. J Thorac Oncol.
2017;12:e49-e51.
90. O’Hare T, Eide CA, Deininger MW. Bcr-Abl kinase domain mutations, drug resistance, and the road to a cure for chronic myeloid leukemia. Blood. 2007;
110:2242-2249.
91. Pennell NA, Arcila ME, Gandara DR, et al. Biomarker testing for patients with advanced non–small cell lung cancer: real-world issues and tough choices. Am
Soc Clin Oncol Educ Book. 2019;39:531-542.
92. Arena S, Bellosillo B, Siravegna G, et al. Emergence of multiple EGFR extracellular mutations during cetuximab treatment in colorectal cancer. Clin Cancer Res.
2015;21:2157-2166.
93. Sánchez-Martı́n FJ, Bellosillo B, Gelabert-Baldrich M, et al. The first-in-class anti-EGFR antibody mixture Sym004 overcomes cetuximab resistance mediated by
EGFR extracellular domain mutations in colorectal cancer. Clin Cancer Res. 2016;22:3260-3267.
94. Arena S, Siravegna G, Mussolin B, et al. MM-151 overcomes acquired resistance to cetuximab and panitumumab in colorectal cancers harboring EGFR
extracellular domain mutations. Sci Transl Med. 2016;8:324ra14.
95. Montagut C, Argilés G, Ciardiello F, et al. Efficacy of Sym004 in patients with metastatic colorectal cancer with acquired resistance to anti-EGFR therapy and
molecularly selected by circulating tumor DNA analyses: a phase 2 randomized clinical trial. JAMA Oncol. 2018;4:e175245.
96. Oxnard GR. The cellular origins of drug resistance in cancer. Nat Med. 2016;22:232-234.
97. Sharma SV, Lee DY, Li B, et al. A chromatin-mediated reversible drug-tolerant state in cancer cell subpopulations. Cell. 2010;141:69-80.

Gumusay et al
98. Ramirez M, Rajaram S, Steininger RJ, et al. Diverse drug-resistance mechanisms can emerge from drug-tolerant cancer persister cells. Nat Commun. 2016;
7:10690.
99. Shen S, Faouzi S, Bastide A, et al. An epitranscriptomic mechanism underlies selective mRNA translation remodelling in melanoma persister cells. Nat
Commun. 2019;10:5713.
100. Russo M, Crisafulli G, Sogari A, et al. Adaptive mutability of colorectal cancers in response to targeted therapies. Science. 2019;366:1473-1480.
101. Hata AN, Niederst MJ, Archibald HL, et al. Tumor cells can follow distinct evolutionary paths to become resistant to epidermal growth factor receptor inhibition.
Nat Med. 2016;22:262-269.
102. Rusan M, Li K, Li Y, et al. Suppression of adaptive responses to targeted cancer therapy by transcriptional repression. Cancer Discov. 2018;8:59-73.
103. Hangauer MJ, Viswanathan VS, Ryan MJ, et al. Drug-tolerant persister cancer cells are vulnerable to GPX4 inhibition. Nature. 2017;551:247-250.
104. Lorenzato A, Magrı̀ A, Matafora V, et al. Vitamin C restricts the emergence of acquired resistance to EGFR-targeted therapies in colorectal cancer. Cancers
(Basel). 2020;12:685.
105. Hata AN, Engelman JA, Faber AC. The BCL2 family: key mediators of the apoptotic response to targeted anticancer therapeutics. Cancer Discov. 2015;
5:475-487.
106. Montero J, Letai A. Why do BCL-2 inhibitors work and where should we use them in the clinic? Cell Death Differ. 2018;25:56-64.
107. Corcoran RB, Cheng KA, Hata AN, et al. Synthetic lethal interaction of combined BCL-XL and MEK inhibition promotes tumor regressions in KRAS mutant
cancer models. Cancer Cell. 2013;23:121-128.
108. Gibson CJ, Davids MS. BCL-2 antagonism to target the intrinsic mitochondrial pathway of apoptosis. Clin Cancer Res. 2015;21:5021-5029.
109. Sun Y. Tumor microenvironment and cancer therapy resistance. Cancer Lett. 2016;380:205-215.
110. Valkenburg KC, de Groot AE, Pienta KJ. Targeting the tumour stroma to improve cancer therapy. Nat Rev Clin Oncol. 2018;15:366-381.
111. Harvey JM, Clark GM, Osborne CK, et al. Estrogen receptor status by immunohistochemistry is superior to the ligand-binding assay for predicting response to
adjuvant endocrine therapy in breast cancer. J Clin Oncol. 1999;17:1474-1481.
112. Musgrove EA, Sutherland RL. Biological determinants of endocrine resistance in breast cancer. Nat Rev Cancer. 2009;9:631-643.
113. Protein Data Bank. Molecule of the month. https://pdb101.rcsb.org/motm/45. Accessed March 8, 2020.
114. Young E, Miele L, Tucker KB, et al. SU11248, a selective tyrosine kinases inhibitor suppresses breast tumor angiogenesis and growth via targeting both tumor
vasculature and breast cancer cells. Cancer Biol Ther. 2010;10:703-711.
115. Rugo HS, Vidula N, Ma C. Improving response to hormone therapy in breast cancer: new targets, new therapeutic options. Am Soc Clin Oncol Educ Book. 2016;
35:e40-e54.
116. Sherr CJ. Cancer cell cycles. Science. 1996;274:1672-1677.
117. Sherr CJ, Roberts JM. CDK inhibitors: positive and negative regulators of G1-phase progression. Genes Dev. 1999;13:1501-1512.
118. van den Heuvel S, Harlow E. Distinct roles for cyclin-dependent kinases in cell cycle control. Science. 1993;262:2050-2054.
119. Weinberg RA. The retinoblastoma protein and cell cycle control. Cell. 1995;81:323-330.
120. Finn RS, Dering J, Conklin D, et al. PD 0332991, a selective cyclin D kinase 4/6 inhibitor, preferentially inhibits proliferation of luminal estrogen receptor-positive
human breast cancer cell lines in vitro. Breast Cancer Res. 2009;11:R77.
121. Finn RS, Crown JP, Lang I, et al. The cyclin-dependent kinase 4/6 inhibitor palbociclib in combination with letrozole versus letrozole alone as first-line treatment
of oestrogen receptor-positive, HER2-negative, advanced breast cancer (PALOMA-1/TRIO-18): a randomised phase 2 study. Lancet Oncol. 2015;16:25-35.
122. Pfizer. Pfizer Receives U.S. FDA Accelerated Approval of IBRANCE (palbociclib). https://www.pfizer.com/news/press-release/press-release-detail/pfizer_
receives_u_s_fda_accelerated_approval_of_ibrance_palbociclib. Accessed May 8, 2020.
123. Hart CD, Migliaccio I, Malorni L, et al. Challenges in the management of advanced, ER-positive, HER2-negative breast cancer. Nat Rev Clin Oncol. 2015;
12:541-552.
124. Finn RS, Martin M, Rugo HS, et al. PALOMA-2: primary results from a phase III trial of palbociclib (P) with letrozole (L) compared with letrozole alone in
postmenopausal women with ER+/HER2– advanced breast cancer (ABC). J Clin Oncol. 2016;34:15s (suppl; abstr 507).
125. Rugo HS, Finn RS, Diéras V, et al. Palbociclib plus letrozole as first-line therapy in estrogen receptor-positive/human epidermal growth factor receptor 2-negative
advanced breast cancer with extended follow-up. Breast Cancer Res Treat. 2019;174:719-729.
126. Turner NC, Ro J, André F, et al; PALOMA3 Study Group. Palbociclib in hormone-receptor–positive advanced breast cancer. N Engl J Med. 2015;373:209-219.
127. Cristofanilli M, Turner NC, Bondarenko I, et al. Fulvestrant plus palbociclib versus fulvestrant plus placebo for treatment of hormone-receptor-positive, HER2-
negative metastatic breast cancer that progressed on previous endocrine therapy (PALOMA-3): final analysis of the multicentre, double-blind, phase 3
randomised controlled trial. Lancet Oncol. 2016;17:425-439.
128. Turner NC, Slamon DJ, Ro J, et al. Overall survival with palbociclib and fulvestrant in advanced breast cancer. N Engl J Med. 2018;379:1926-1936.
129. Hortobagyi GN, Stemmer SM, Burris HA, et al. Ribociclib as first-line therapy for HR-positive, advanced breast cancer. N Engl J Med. 2016;375:1738-1748.
130. Hortobagyi GN, Stemmer SM, Burris HA, et al. Updated results from MONALEESA-2, a phase III trial of first-line ribociclib plus letrozole versus placebo plus
letrozole in hormone receptor-positive, HER2-negative advanced breast cancer. Ann Oncol. 2018;29:1541-1547.

131. Slamon DJ, Neven P, Chia S, et al. Phase III randomized study of ribociclib and fulvestrant in hormone receptor-positive, human epidermal growth factor
receptor 2-negative advanced breast cancer: MONALEESA-3. J Clin Oncol. 2018;36:2465-2472.
132. Slamon DJ, Neven P, Chia S, et al. Overall survival with ribociclib plus fulvestrant in advanced breast cancer. N Engl J Med. 2020;382:514-524.
133. Tripathy D, Im S-A, Colleoni M, et al. Ribociclib plus endocrine therapy for premenopausal women with hormone-receptor-positive, advanced breast cancer
(MONALEESA-7): a randomised phase 3 trial. Lancet Oncol. 2018;19:904-915.
134. Im S-A, Lu Y-S, Bardia A, et al. Overall survival with ribociclib plus endocrine therapy in breast cancer. N Engl J Med. 2019;381:307-316.
135. Dickler MN, Tolaney SM, Rugo HS, et al. MONARCH 1, a phase II study of abemaciclib, a CDK4 and CDK6 inhibitor, as a single agent, in patients with refractory
HR+/HER2 metastatic breast cancer. Clin Cancer Res. 2017;23:5218-5224.
136. Sledge GW Jr., Toi M, Neven P, et al. MONARCH 2: abemaciclib in combination with fulvestrant in women with HR+/HER2 advanced breast cancer who had
progressed while receiving endocrine therapy. J Clin Oncol. 2017;35:2875-2884.
137. Sledge G, Toi M, Neven P, et al. MONARCH 2: overall survival of abemaciclib plus fulvestrant in patients with HR+, HER2-advanced breast cancer. Ann Oncol.
2019;30:v856.
138. Goetz MP, Toi M, Campone M, et al. MONARCH 3: abemaciclib as initial therapy for advanced breast cancer. J Clin Oncol. 2017;35:3638-3646.
139. Johnston S, Martin M, Di Leo A, et al. MONARCH 3 final PFS: a randomized study of abemaciclib as initial therapy for advanced breast cancer. NPJ Breast
Cancer. 2019;5:5.
140. Bachelot T, Bourgier C, Cropet C, et al. Randomized phase II trial of everolimus in combination with tamoxifen in patients with hormone receptor-positive, human
epidermal growth factor receptor 2-negative metastatic breast cancer with prior exposure to aromatase inhibitors: a GINECO study. J Clin Oncol. 2012;
30:2718-2724.
141. Piccart M, Hortobagyi GN, Campone M, et al. Everolimus plus exemestane for hormone-receptor-positive, human epidermal growth factor receptor-2-negative
advanced breast cancer: overall survival results from BOLERO-2. Ann Oncol. 2014;25:2357-2362.
142. Yardley DA, Noguchi S, Pritchard KI, et al. Everolimus plus exemestane in postmenopausal patients with HR(+) breast cancer: BOLERO-2 final progression-free
survival analysis. Adv Ther. 2013;30:870-884.
143. André F, Ciruelos E, Rubovszky G, et al; SOLAR-1 Study Group. Alpelisib for PIK3CA-mutated, hormone receptor–positive advanced breast cancer. N Engl
J Med. 2019;380:1929-1940.
144. Jones RH, Casbard A, Carucci M, et al. Fulvestrant plus capivasertib versus placebo after relapse or progression on an aromatase inhibitor in metastatic,
oestrogen receptor-positive breast cancer (FAKTION): a multicentre, randomised, controlled, phase 2 trial. Lancet Oncol. 2020;21:345-357.
145. Martı́n M, Loibl S, von Minckwitz G, et al. Phase III trial evaluating the addition of bevacizumab to endocrine therapy as first-line treatment for advanced breast
cancer: the letrozole/fulvestrant and avastin (LEA) study. J Clin Oncol. 2015;33:1045-1052.
146. Dickler MN, Barry WT, Cirrincione CT, et al. Phase III trial evaluating letrozole as first-line endocrine therapy with or without bevacizumab for the treatment of
postmenopausal women with hormone receptor-positive advanced-stage breast cancer: CALGB 40503 (Alliance). J Clin Oncol. 2016;34:2602-2609.
147. Yardley DA, Ismail-Khan RR, Melichar B, et al. Randomized phase II, double-blind, placebo-controlled study of exemestane with or without entinostat in
postmenopausal women with locally recurrent or metastatic estrogen receptor-positive breast cancer progressing on treatment with a nonsteroidal aromatase
inhibitor. J Clin Oncol. 2013;31:2128-2135.
148. Taylor-Harding B, Aspuria P-J, Agadjanian H, et al. Cyclin E1 and RTK/RAS signaling drive CDK inhibitor resistance via activation of E2F and ETS. Oncotarget.
2015;6:696-714.
149. Caldon CE, Sergio CM, Kang J, et al. Cyclin E2 overexpression is associated with endocrine resistance but not insensitivity to CDK2 inhibition in human breast
cancer cells. Mol Cancer Ther. 2012;11:1488-1499.
150. Herrera-Abreu MT, Palafox M, Asghar U, et al. Early adaptation and acquired resistance to CDK4/6 inhibition in estrogen receptor-positive breast cancer. Cancer
Res. 2016;76:2301-2313.
151. Bollard J, Miguela V, Ruiz de Galarreta M, et al. Palbociclib (PD-0332991), a selective CDK4/6 inhibitor, restricts tumour growth in preclinical models of
hepatocellular carcinoma. Gut. 2017;66:1286-1296.
152. Mayer E, Demichele A, Pfeiler G, et al. 215TiPPALLAS: PALbociclib CoLlaborative adjuvant study: a randomized phase 3 trial of palbociclib with standard
adjuvant endocrine therapy versus standard adjuvant endocrine therapy alone for HR+/HER2-early breast cancer. Ann Oncol. 2017;28(suppl; abstr OT3-05-
08).
153. Burstein HJ. The distinctive nature of HER2-positive breast cancers. N Engl J Med. 2005;352:1652-1654.
154. Garrido-Castro AC, Goel S. CDK4/6 Inhibition in Breast Cancer: Mechanisms of Response and Treatment Failure. Curr Breast Cancer Rep. 2017;9:26-33.
155. O’Sullivan CC, Suman VJ, Goetz MP. The emerging role of CDK4/6i in HER2-positive breast cancer. Ther Adv Med Oncol. 2019;11:1758835919887665.
156. Witkiewicz AK, Cox D, Knudsen ES. CDK4/6 inhibition provides a potent adjunct to Her2-targeted therapies in preclinical breast cancer models. Genes Cancer.
2014;5:261-72.
157. Goel S, Wang Q, Watt AC, et al. Overcoming therapeutic resistance in HER2-positive breast cancers with CDK4/6 inhibitors. Cancer Cell. 2016;29:255-269.
158. Tolaney SM, Wardley AM, Zambelli S, et al. MonarcHER: A randomized phase II study of abemaciclib plus trastuzumab with or without fulvestrant versus
trastuzumab plus standard-of care chemotherapy in women with HR+, HER2+ advanced breast cancer. Ann Oncol. 2019:30(suppl_5):v851-v934.
159. Gianni L, Bisagni G, Colleoni M, et al. Neoadjuvant treatment with trastuzumab and pertuzumab plus palbociclib and fulvestrant in HER2-positive, ER-positive
breast cancer (NA-PHER2): an exploratory, open-label, phase 2 study. Lancet Oncol. 2018;19:249-256.

Gumusay et al
160. Sammons S, Shastry M, Dent S, et al. Practical treatment strategies and future directions after progression while receiving CDK4/6 inhibition and endocrine
therapy in advanced HR+/HER2- breast cancer. Clin Breast Cancer. 2020;20:1-11.
161. Mollon L, Aguilar A, Anderson E, et al. A systematic literature review of the prevalence of PIK3CA mutations and mutation hotspots in HR+/HER2-metastatic
breast cancer. Cancer Res. 2018;78(suppl; abstr 1207).
162. Miller TW, Hennessy BT, González-Angulo AM, et al. Hyperactivation of phosphatidylinositol-3 kinase promotes escape from hormone dependence in estrogen
receptor-positive human breast cancer. J Clin Invest. 2010;120:2406-2413.
163. Ciruelos Gil EM. Targeting the PI3K/AKT/mTOR pathway in estrogen receptor-positive breast cancer. Cancer Treat Rev. 2014;40:862-871.
164. Baselga J, Semiglazov V, van Dam P, et al. Phase II randomized study of neoadjuvant everolimus plus letrozole compared with placebo plus letrozole in patients
with estrogen receptor-positive breast cancer. J Clin Oncol. 2009;27:2630-2637.
165. Rugo HS, Seneviratne L, Beck JT, et al. Prevention of everolimus-related stomatitis in women with hormone receptor-positive, HER2-negative metastatic breast
cancer using dexamethasone mouthwash (SWISH): a single-arm, phase 2 trial. Lancet Oncol. 2017;18:654-662.
166. Kornblum N, Zhao F, Manola J, et al. Randomized phase II trial of fulvestrant plus everolimus or placebo in postmenopausal women with hormone receptor-
positive, human epidermal growth factor receptor 2-negative metastatic breast cancer resistant to aromatase inhibitor therapy: results of PrE0102. J Clin Oncol.
2018;36:1556-1563.
167. Bardia A, Hurvitz SA, DeMichele A, et al. Triplet therapy (continuous ribociclib, everolimus, exemestane) in HR+/HER2 advanced breast cancer post-
progression on a CDK4/6 inhibitor (TRINITI-1): Efficacy, safety, and biomarker results. J Clin Oncol. 2019;37:15s (suppl; abstr 1016).
168. Baselga J, Im S-A, Iwata H, et al. Buparlisib plus fulvestrant versus placebo plus fulvestrant in postmenopausal, hormone receptor-positive, HER2-negative,
advanced breast cancer (BELLE-2): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol. 2017;18:904-916.
169. Campone M, Im S-A, Iwata H, et al. Buparlisib plus fulvestrant versus placebo plus fulvestrant for postmenopausal, hormone receptor-positive, human
epidermal growth factor receptor 2-negative, advanced breast cancer: Overall survival results from BELLE-2. Eur J Cancer. 2018;103:147-154.
170. Di Leo A, Johnston S, Lee KS, et al. Buparlisib plus fulvestrant in postmenopausal women with hormone-receptor-positive, HER2-negative, advanced breast
cancer progressing on or after mTOR inhibition (BELLE-3): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol. 2018;19:87-100.
171. Baselga J, Dent SF, Cortés J, et al. Phase III study of taselisib (GDC-0032) + fulvestrant (FULV) v FULV in patients (pts) with estrogen receptor (ER)-positive,
PIK3CA-mutant (MUT), locally advanced or metastatic breast cancer (MBC): Primary analysis from SANDPIPER. J Clin Oncol. 2018;36:15s (suppl; abstr
LBA1006).
172. Mayer IA, Abramson VG, Formisano L, et al. A phase Ib study of alpelisib (BYL719), a PI3Kα-specific inhibitor, with letrozole in ER+/HER2 metastatic breast
173. Rugo H, André F, Yamashita T, et al. Alpelisib (ALP)+ fulvestrant (FUL) for patients with hormone receptor–positive (HR+), HER2 advanced breast cancer
(ABC): Management and time course of key adverse events of special interest (AESIs) in SOLAR-1. Ann Oncol. 2019;30:v112-v113.
174. Rugo HS, Borrego MR, Chia SKL, et al. Alpelisib (ALP) + endocrine therapy (ET) in patients (pts) with PIK3CA-mutated hormone receptor-positive (HR+),
human epidermal growth factor-2-negative (HER2-) advanced breast cancer (ABC): First interim BYLieve study results. J Clin Oncol. 2019;37:15s (suppl; abstr
1040).
175. Davies BR, Greenwood H, Dudley P, et al. Preclinical pharmacology of AZD5363, an inhibitor of AKT: pharmacodynamics, antitumor activity, and correlation of
monotherapy activity with genetic background. Mol Cancer Ther. 2012;11:873-887.
176. Qu Z, Van Ginkel S, Roy AM, et al. Vascular endothelial growth factor reduces tamoxifen efficacy and promotes metastatic colonization and desmoplasia in
breast tumors. Cancer Res. 2008;68:6232-6240.
177. Martı́n M, Loibl S, Hyslop T, et al; Alliance for Clinical Trials in Oncology (Alliance). Evaluating the addition of bevacizumab to endocrine therapy as first-line
treatment for hormone receptor-positive metastatic breast cancer: a pooled analysis from the LEA (GEICAM/2006-11_GBG51) and CALGB 40503 (Alliance)
trials. Eur J Cancer. 2019;117:91-98.
178. Connolly R, Stearns V. Epigenetics as a therapeutic target in breast cancer. J Mammary Gland Biol Neoplasia. 2012;17:191-204.
179. Thomas S, Thurn KT, Biçaku E, et al. Addition of a histone deacetylase inhibitor redirects tamoxifen-treated breast cancer cells into apoptosis, which is opposed
by the induction of autophagy. Breast Cancer Res Treat. 2011;130:437-447.
180. Hodges-Gallagher L, Valentine CD, Bader SE, et al. Inhibition of histone deacetylase enhances the anti-proliferative action of antiestrogens on breast cancer cells
and blocks tamoxifen-induced proliferation of uterine cells. Breast Cancer Res Treat. 2007;105:297-309.
181. Knipstein J, Gore L. Entinostat for treatment of solid tumors and hematologic malignancies. Expert Opin Investig Drugs. 2011;20:1455-1467.
182. Rugo HS, Delord J-P, Im S-A, et al. Safety and antitumor activity of pembrolizumab in patients with estrogen receptor–positive/human epidermal growth factor
receptor 2–negative advanced breast cancer. Clin Cancer Res. 2018;24:2804-2811.

invited article Molecular Profiling in Drug Development: Paving

a Way Forward
Suzanne F. Jones, PharmD1 and Andrew J. McKenzie, PhD1
overview
As researchers learn more about tumor biology and the molecular mechanisms involved in tumorigenesis,
metastasis, and tumor evolution, clinical trials are growing more complex and patient selection for clinical
trials is becoming more specific. Rather than exploit certain phenotypic characteristics of tumor cells (e.g.,
rapid cell division and uncontrolled cell growth), pharmaceuticals targeting the genotypic causes of tu-
morigenesis are emerging. The sequencing of the human genome, advances in chemical techniques, and
increased efficiency in drug target identification have changed the way drugs are developed. Now, more
precise drugs targeting specific mutations within individual genes are being used to treat narrow patient
populations harboring these specific driver mutations, often with greater efficacy and lower toxicity than
traditional chemotherapeutic agents. This precision in drug development relies not only on the ability to design
exquisitely specific pharmaceuticals but also to identify (with the same level of precision) the patients who are
most likely to respond to those therapies. Robust screening techniques and adequate molecular oncology
education are required to match the appropriate patient to precision therapies, and these same screening
techniques provide the data necessary to advance to the next generation of drug development.
INTRODUCTION numerous additional targets for cancer treatment, and

Historically, cancer treatment relied on the use of advances in chemical techniques and increased effi-
cytotoxic chemotherapy agents to inhibit cell division ciency in drug target identification have yielded nu-
and destroy cancer cells. Unfortunately, these agents merous potential targeted agents, including agents
targeted rapid cell division in both tumor tissue and directed at targets that were felt to be potentially
normal tissue, resulting in substantial treatment- undruggable, such as KRAS.1,15 Currently, there are multiple
related toxicities. Increased understanding of the novel therapeutics focused on KRAS in clinical development
molecular events involved in tumorigenesis and me- with various mechanisms of action and more are on the
tastases has led to the identification of potentially horizon, all relying on the ability to accurately identify and
druggable targets in numerous disease types.1-7 Tar- accrue patients harboring KRAS mutations.15-17 As oncology
geted anticancer agents may allow greater antitumor therapeutics evolve, so too are clinical trials and the treat-
efficacy and lower toxicity than traditional chemo- ment of patients with cancer. Medical oncologists must
therapeutic agents. Trastuzumab is an example of connect complex(ngs molecular characteristics with
a monoclonal antibody targeting HER2-positive breast exquisitely specific targeted therapeutics.
cancer cells that has revolutionized the treatment of
patients with HER2-positive breast cancer.8,9 Small CHANGING PATIENT TREATMENT LANDSCAPE
molecules have also been used to successfully target The successful U.S. Food and Drug Administration
tumor-specific pathways. Imatinib, an oral targeted (FDA) approval of targeted therapies has spurred an
Author affiliations therapy that inhibits BCR-ABL, c-KIT, and PDGFRA increase in obtaining molecular information on patient
and support tyrosine kinases, was initially approved for the treat- tumors. In non–small cell lung cancer (NSCLC), there
information (if
applicable) appear
ment of chronic myelogenous leukemia, a disease that are five genes for which mutation testing is recom-
at the end of this is characterized by the presence of a BCR-ABL fusion mended by the National Comprehensive Cancer
article. gene (Philadelphia chromosome).10-12 Both trastuzu- Network (NCCN; EGFR, ALK, ROS1, BRAF, and
Accepted on May 8, mab and imatinib also received subsequent approvals NTRK) and the guidelines “strongly advise” that this
2020 and published based on the identification of target mutations in other testing be done as part of broader molecular profiling
at ascopubs.org on
diseases, such as HER2 in gastric cancer and c-KIT in with the goal of “identifying rare driver mutations for
doi.org/10.1200/ gastrointestinal stromal tumors.13,14 The sequencing which effective drugs may already be available.”18
EDBK_100024 of the human genome has led to the identification of Similarly, colorectal cancer has seven genes for which

Jones and McKenzie
community oncology practices to inform treatment de-

cisions for patients with NSCLC.25-27 These findings suggest
that there is a shift in standard-of-care practices where
• NGS vendors differ in the genes covered, broad-based NGS is being routinely incorporated into
analytes being analyzed, and technology plat-
treatment planning, even in nonacademic, community-
forms being utilized. Understanding the limi-
based medical oncology practices. Although there are still
tations of the NGS test being ordered is critical
to understanding the results. controversies regarding which patients should receive
broad-based molecular profiling, recent clinical trial data
• Targeted therapies have been approved for
suggest that blood-based NGS testing for patients with
numerous tumor types, and clinical research is
NSCLC in the frontline setting can detect targetable mu-
seeking to expand the use of targeted therapies.
Performing NGS on various tumor types may tations that influence treatment.28,29 This shift in the treat-
uncover potential treatment options in both ment of patients with NSCLC indicates that more patients
standard of care and clinical trial settings. will have baseline NGS testing and can be monitored for the
acquisition of resistance mutations—even resistance to
• Multidisciplinary molecular tumor boards guide
treatment decisions and lead to better patient traditional chemotherapy or chemotherapy/immunotherapy
outcomes. combinations when targetable mutations are not detected.
As the prevalence of testing increases, so do the commercial
offerings of NGS testing. Each NGS vendor offers a slightly
different panel and can use different analytes (RNA, DNA,
mutation analysis is recommended by NCCN guidelines (NRAS/ or protein) from different sources (tissue, blood, saliva, or
HRAS/KRAS, BRAF, HER2, and NTRK1/2/3).19 Selpercatinib urine). A recent review discusses several popular com-
and capmatinib (oral RET and MET inhibitors, respectively) were mercial NGS vendors, and they are summarized in the
also recently granted FDA priority review for new drug ap- Table 1.30 Medical oncologists’ understanding of the dif-
plication for the treatment of advanced RET fusion– ferent testing modalities is critical for optimal utilization of
positive and MET exon 14 skipping–positive NSCLC.20,21 NGS testing. For example, not all commercial NGS tests
Although the NCCN does not recommend specific testing cover NTRK gene fusions equally, and important treatment
modalities in colorectal cancer, the guidelines specify that decisions can be altered based on the limitations of a test
all RAS genes, NTRK, BRAF, and HER2 can be assayed using and not the biology of a patient’s tumor.31 A recent study
next-generation sequencing (NGS). Furthermore, nivolumab demonstrated that there is wide discordance between two
monotherapy and combination therapy with ipilimumab are commercial NGS vendors that could lead to changes in
approved for the treatment of microsatellite instability–high/ treatment options.32 Without knowing the limitations of and
deficient mismatch repair metastatic colorectal cancer that has differences between the tests, a clinician may deem a pa-
progressed on standard therapy.22,23 tient ineligible for a certain therapy or clinical trial when, in
Molecular analysis of all tumor types is increasingly im- actuality, the genetic abnormality was not adequately
portant as the frequency of tumor-agnostic FDA approvals tested. It is not surprising, therefore, that recent studies have
continues to grow (e.g., pembrolizumab for microsatellite shown that medical oncologists are still uncomfortable
instability–high/DNA damage repair–deficient tumors, lar- interpreting results from NGS vendors and a lack of genomic
otrectinib and entrectinib for NTRK fusion–positive tumors). confidence remains.24,33,34
We analyzed NGS ordering patterns in our network of ap- Clinical research will continue to be democratized to pa-
proximately 400 community-based medical oncologists tients within their communities, and the shift of clinical
between 2012 and 2019 and found that NGS ordering has research into community practices means that general
increased nearly 150% across an array of tumor types. medical oncologists require more specialized knowledge of
These data support those reported by the National Survey molecular oncology to support the evolving landscape of
of Precision Medicine in Cancer Treatment, where survey drug development.
results showed that more than 75% of oncologists were
utilizing NGS tests in their practice in 2017.24 Moreover, CHANGING CLINICAL TRIAL LANDSCAPE
health systems are increasingly integrating NGS testing as Rapid advances in medical oncology have spurred un-
a part of standard of care. Legacy Health System, In- precedented growth in oncology-specific FDA approvals.
termountain Healthcare, Swedish Cancer Center, and the From 1998 to 2019, oncology-specific FDA approvals
Sarah Cannon Research Institute are nonacademic medi- jumped from eight to 38, with 13 of those 38 in 2019 re-
cal centers that have integrated NGS into routine clinical quiring some kind of molecular stratification. One of the 38
practice, and a recent retrospective analysis of the Flatiron approvals in 2019 was a tumor-agnostic approval (entrec-
Health Database revealed broad utilization of NGS in tinib for NTRK gene fusion–positive solid tumors), indicating

Utilizing NGS in the Development of Novel Therapeutics
that some molecular drivers of tumorigenesis are common Although many large academic medical centers have long
across tumor types.49 Both clinical trial design and the histories in oncology drug development, nonacademic
regulatory approval process have dramatically changed community-based hospitals and medical oncology practices
during this period as well. To meet the demands of rapidly have also been at the forefront of drug development. One
moving science, the FDA has established four expedited such community-based clinical research organization,
programs for serious conditions to accelerate drug devel- Sarah Cannon Research Institute, has been involved in
opment: fast track, breakthrough therapy, accelerated oncology drug development for more than 25 years, makes
approval, and priority review.50 The FDA approval of pem- clinical research available to community-based medical
brolizumab for refractory metastatic melanoma capitalized on oncology practices, and offers a broad menu of clinical trials
three of these programs and relied solely on the data from that span the cycle of drug development (phases I–IV).
KEYNOTE-001, a single clinical trial that made use of large, Analysis of the clinical trial menu and clinical research
dose-expansion cohorts in both disease and molecularly patterns in an institution like Sarah Cannon gives insight into
targeted patient populations.51 The time from initial in- the broad availability of clinical research to patients in their
vestigational new drug submission to first FDA approval was own communities.
approximately 4 years, significantly shorter than the 10 to As evidence of the emergence of targeted therapy clinical
15 years required for traditional FDA approval.52 This ushered trials, a breakdown of Sarah Cannon’s early-phase clinical
in a new era of “single-arm” approvals that take advantage trial menu shows that 33.6% of all recruiting interventional
of expansion arms in indications where early results are clinical trials at Sarah Cannon in 2019 were targeted therapy
promising and there is an unmet medical need. studies. An additional 52.9% were immunotherapy clinical
TABLE 1. Next-Generation Sequencing Vendors

Tumor/ Genes on FDA
Vendor Test Name (Reference) Biopsy Type Analyte Normal Panel TMB MSI Approval
Foundation Medicine FoundationOne CDx35 Tissue DNA No 324 Yes Yes Yes
FoundationOne Plasma DNA No 70 No Yes No
Liquid36
FoundationOne Heme37 Whole blood, bone marrow DNA/RNA No 426 Yes Yes No
aspirate, or tissue
Memorial Sloan Kettering MSK-IMPACT38 Tissue DNA Yes 468 Yes Yes Yes
Cancer Center
Caris Caris Molecular Tissue DNA/RNA/ No 442 Yes Yes No
Intelligence39 protein*
Guardant Guardant 36040 Plasma DNA No 74 No Yes No
41
ParadigmDx PCDx Tissue DNA/RNA/ No 234 Yes Yes No
protein*
Tempus Tempus |xT42 Tissue DNA/RNA/ Yes 648 Yes Yes No
protein*
Tempus |xF Plasma DNA No 105 No No No
Life Technologies OncomineDx Target Tissue DNA/RNA No 23 No No Yes
Test43
OmniSeq OmniSeq Advance44 Tissue DNA/RNA/ No 144 Yes Yes No
protein*
OmniSeq Tissue DNA/RNA/ No 144 No No No
Comprehensive45 protein*
PathGroup SmartGenomics46 Tissue DNA/RNA/ No 160 Yes Yes No
protein*
NeoGenomics NeoType Discovery47 DNA/RNA/ No 323 Yes Yes No
protein
Illumina Trusight Oncology Tissue DNA/RNA No 523 Yes Yes No
50048
Abbreviations: TMB, tumor mutational burden; MSI, microsatellite instability; FDA, U.S. Food and Drug Administration.
*Protein is assayed using immunohistochemistry; vendors vary in the proteins assayed.

Jones and McKenzie
trials in which biomarkers such as microsatellite instability, a result of a deeper understanding of the biology un-
tumor mutational burden, and PD-L1 status might be used derpinning tumorigenesis and the availability of both clinical
for patient recruitment (Fig. 1A). Moreover, an analysis of and research assays that aim to uncover novel molecular
clinical trial enrollment at Sarah Cannon’s Nashville Drug findings in clinical trial patient populations. A snapshot of
Development Unit showed that of the 570 patients with solid the Sarah Cannon early-phase clinical trial menu shows that
tumors enrolled in clinical trials in 2019, 41% were enrolled only 10% of trials do not require any type of tissue biopsy
in targeted therapy trials (Fig. 1B). Together, the menu and (either fresh or archived) for enrollment (Fig. 1C). Thirty-six
recruitment data reveal that targeted therapies and mo- percent of the current early-phase trials require archived
lecularly motivated clinical trials dominate the current tissue or a fresh baseline biopsy, 49% require a fresh biopsy
clinical trial landscape. at baseline and/or after enrollment, and 5% allow tumor
Clinical trial complexity has increased and now involves biopsies to be optional (Fig. 1C). As a result, more than 450
large biomarker cohorts, pretreatment and postprogression fresh biopsies were collected for patients enrolled in early-
biopsy requirements, and intricate trial design to recruit phase trials in 2019. Obtaining additional NGS testing
molecularly subtyped patients. These complexities are for freshly obtained tumor biopsies provides invaluable
FIGURE 1. Sarah Cannon Clinical Trial Program

(A) Sarah Cannon’s clinical trial menu categorized by primary study drug. (B) Patient enrollment in Sarah Cannon clinical trials. (C) Clinical trial biopsy
requirements. (D) Illustration of traditional and emerging drug development techniques. Traditional drug development refers to nonselective recruitment to
clinical trials (inclusion of HRD, BRCA1/2, and WT) and emerging drug development refers to selective clinical trial recruitment (exclusion of WT). (E)
Summary of clinical trials with either explicit, implied, or general molecular criteria. (F) Percent increase in commercial next-generation sequencing test
orders as a percentage of 2012 orders.
Abbreviations: ADC, antibody drug conjugate; Chemo, cytotoxic chemotherapy; Immuno, immunotherapy; Targeted, targeted therapy; PARPi, PARP
inhibitor; HRD, homologous recombination deficient; WT, wild type.

information that may help guide treatment decisions for the menu demonstrates, there is no indication that biomarker
individual patient, improve trial accrual for targeted clinical discovery or the implementation of molecularly motivated
trials, and contribute to cancer research overall as we try to clinical trials is slowing. Innovative strategies to educate and
further understand cancer biology and drug resistance. inform research clinicians on new therapeutics in devel-
In early drug development trials, safety, rather than efficacy, opment, novel drug mechanisms of action, and emerging
is the primary endpoint. However, in an effort to increase the testing technologies are necessary to continue driving drug
likelihood of clinical response, pharmaceutical companies discovery. Moreover, interrogating population-scale data,
are becoming more specific about the patient populations both clinical and molecular, is essential to continue laying
allowed to enroll, even in early-phase trials. Instead of en- the groundwork for future drug development.
rolling patients from the “ocean” of potential clinical trial Molecular tumor boards (MTBs) are an invaluable tool that
patients, pharmaceutical companies are trying to “fish in the many academic institutions use to educate clinicians re-
right pond” of molecularly subtyped patients at the earliest garding molecular testing. Recent publications shown in
stages of drug development (Fig. 1D). Table 2 highlight the impact of MTBs on patient care.53-61
There are now three distinct types of molecular inclusion The University of California San Diego Moores Cancer
criteria that research clinicians must be aware of: (1) ex- Center60 was one of the first institutions to report the effect of
plicit, (2) implied, and (3) general. Explicit molecular criteria MTB on patient care. Although the patient population was
are those explicitly stated in the inclusion/exclusion criteria small (34 patients), the authors reported that 12 patients
of a clinical trial protocol. There are some early-phase (35%) followed MTB recommendations, and 11 of those 12
clinical trials in which molecular alterations are required (or achieved a partial response. Similarly, Memorial Sloan
forbidden) for enrollment, although this is much more Kettering Cancer Center57 reported a small study of 39
common in later phases of clinical development. Implied pediatric patients in which 61% of patients followed the
molecular criteria are based on the study drug’s mechanism treatment plans recommended by the pediatric MTB. These
of action. For example, a clinical trial of a novel PARP in- encouraging results were followed by other institutions that
hibitor may not require a patient to harbor a BRCA1/2 have since adopted an array of MTB models, including
mutation or homologous recombination deficiency, but a pa- institute-wide, pediatric-specific, or disease-specific MTBs
tient with a BRCA1/2 or homologous recombination–deficient (Table 2). In larger studies, anywhere from 8% to 22% of
mutation may be more likely to respond to that therapy than MTB recommendations were followed, indicating institution-
a patient without those mutations. If clinicians are using specific differences in both MTB process and physician
molecular testing as part of standard of care and have access engagement with MTBs. Although the outcomes of different
to clinical trials, then prioritizing the enrollment of patients with MTBs vary, all studies are consistent in that approximately
a BRCA1/2 mutation in PARP inhibitor clinical trials is in the 80% of all NGS reports have at least one actionable genetic
best interest of both the patient and the clinical trial. General mutation and that access to clinical research programs en-
inclusion criteria refer to the most common “all-comer” in- hances the effectiveness of MTB programs.
clusion/exclusion criteria, where neither the inclusion/exclu- MTBs are designed to bring together experts in basic bi-
sion criteria nor the mechanism of action of the drug have clear ology, pharmacology, molecular pathology, and clinical care
molecular targets. For example, a new cytotoxic chemotherapy to address the complexities of NGS reports and provide
combination or epigenetic-modifying drug (e.g., histone scientific rationale for why certain therapies may (or may
deacetylase inhibitors) is not necessarily dependent on not) be beneficial given a patient’s molecular profile. Un-
a molecular alteration to predict efficacy; therefore, a molecular fortunately, the larger MTB studies showed a relatively low
finding should not affect the appropriateness of a clinical trial adoption of MTB recommendations into clinical care, sig-
investigating these types of drugs. Figure 1E depicts the
nifying an inability to obtain recommended therapies, rec-
breakdown of the current clinical trial menu across Sarah
ommendations made when a patient is not in need of
Cannon (early and late phase) based on the three distinct types
a therapy change, or the unavailability of clinical trial slots
of molecular inclusion criteria. Similarly, many immunotherapy
when the patient does need a therapy change. These
clinical trials are looking to identify potential biomarkers of
challenges are magnified in nonacademic medical centers
response, as PD-L1 status and tumor mutational burden
where care may be less streamlined and general medical
scores are not universal predictors of response like EGFR
oncologists not affiliated with academic medical centers do
mutations are for anti-EGFR therapies in lung cancer.
not have access to MTBs within their institutions.
CONNECTING THE DOTS AND MOVING FORWARD In an effort to expand access to MTBs, nationwide MTB
Biomarkers are being discovered at an unprecedented rate, programs and virtual MTBs have been initiated.66-68 These
and general medical oncologists are unable to keep up with studies indicate that broad access to MTB services in-
this rapidly evolving field. As the Sarah Cannon clinical trial creases patient access to targeted therapy, increases

Jones and McKenzie
TABLE 2. Molecular Tumor Boards

Number of Reference
Institution MTB Name Patients Tumor Types Outcome (Year)
Cincinnati Children’s MGTB 40 Breast Qualitative study of MTB McGowan et al
Hospital Medical participants’ views on MTB; 18% (2016)58
Center of patients subsequently enrolled
in a clinical trial
Dartmouth-Hitchcock MTB 35 Lung, colorectal, breast, brain, sarcoma, 26.7% of patients accepted MTB- Tafe et al
Medical Center skin, other recommended targeted therapy (2015)59
Moffitt Cancer Center Clinical 58 Sarcoma, lung, breast, ALL, brain, Qualitative documentation of Knepper et al
and Research Genomics melanoma, Merkel cell, AML, experience with key lessons (2017)53
Institute Action cholangiocarcinoma, thyroid, ovarian/ learned
Committee AML, mantle cell, head and neck,
adrenocortical, DLBCL, basal cell,
thymic, chordoma, thyroid/GIST,
colorectal, cutaneous T-cell lymphoma
University of California, MTB 34 Breast, gastrointestinal, head and neck, 35% of patients (12) followed MTB- Schwaederle
San Diego Moores lung, other recommended targeted therapy; et al (2014)60
Cancer Center 11 of those 12 achieved a partial
response
Memorial Sloan PMTB 39 AML, glioma, neuroblastoma, 61% of patients followed PMTB Ortiz et al
Kettering Cancer rhabdomyosarcoma, sarcoma, other recommendations (2016)57
Center solid tumors, ALL, lymphoma
Sidney Kimmel GAITWAY 155 Breast, lung, head and neck, 18% of patients accepted MTB- Dalton et al
Comprehensive MTB neuroendocrine, salivary gland, recommended therapy (7.1% (2017)62
Cancer Center at glioblastoma, cancer of unknown received off-label therapy; 8.3%
Johns Hopkins primary, hepatobiliary, pancreas, were enrolled in a targeted clinical
cholangiocarcinoma, sarcoma, trial)
endometrial, prostate, stomach, colon,
ovary, other
University of Alabama at MTB 191 Solid tumors 7.8% of patients received targeted Harada et al
Birmingham therapy based on MTB (2017)54
Comprehensive recommendations
Cancer Center
Institut Curie MTB 736 Solid tumors 10% of patients enrolled in a clinical Basse et al
trial with a matched therapy (2018)63
Sparrow Herbert- MTB 54 Gynecologic, breast, NSCLC, colorectal 22% of patients received Trivedi et al
Herman Cancer cancer, other genomically matched therapy as (2019)64
Center per MTB recommendation
Sarah Cannon Research Genomics 895 Solid tumors 20% of patients were referred for Moore et al
Institute UK/ Review clinical trial screening; 7% of (2019)65
University College Board patients received trial therapy
London
Abbreviations: MTB, molecular tumor board; MGTB, multidisciplinary genomic tumor board; ALL, acute lymphocytic leukemia; AML, acute myeloid
leukemia; DLBCL, diffuse large B-cell lymphoma; GIST, gastrointestinal stromal tumor; PMTB, pediatric molecular tumor board; GAITWAY, Genetic Alteration
in Tumors With Actionable Yields; NSCLC, non–small cell lung cancer.
compliance with NCCN guidelines for targeted therapies, access to clinical research—tools that most general medical
and increases enrollment in targeted therapy clinical trials. oncologists still do not have.34,69
ASCO’s TAPUR (Targeted Agent and Profiling Utilization To solve some of these issues, Sarah Cannon has imple-
Registry) clinical trial offers MTB support for cases sub- mented Molecular Oncology Support Services (MOSS) with
mitted to the study and is available to both academic the mission to provide molecular education to community-
and nonacademic medical centers. However, each of the based physicians and research staff to increase clinical trial
models discussed requires oncologists to submit cases for enrollment, molecular test ordering, and appropriate ther-
review, which requires a basic understanding of molecular apy selection. MOSS services include clinic-specific MTBs,
profiling results, the targeted therapy landscape, and the Genospace web-based clinical trial matching platform,

and MolecularHELP email and phone lines. When a typical data to pave the way for future drug development. Indeed,
case is submitted, it is reviewed and results are returned to recent publications of other such databases like the Amer-
treating oncologists in 24 to 48 hours. Unique from other ican Association for Cancer Research Project GENIE, Me-
similar forums, Sarah Cannon MTBs are deployed at indi- morial Sloan Kettering Cancer Center’s MSK-IMPACT, and
vidual clinics (with between 1 and 11 affiliated clinicians), Tempus’s NGS library exemplify their utility and the need to
and all molecular test orders are reviewed as soon as the create and maintain the critical infrastructure required to
results are returned to the clinician, without the necessity of house such data.70-73
a clinician submitting the case for review. This provides
proactive, real-time molecular analysis by a team of research CONCLUSION
scientists, clinical pharmacologists, genetic counselors, and The changing clinical trial, patient treatment, and drug
disease-specific clinical leaders as a part of routine care. Also development landscapes are resulting in quicker drug
unique is the Genospace clinical trial matching platform, approvals and more complex eligibility criteria for clinical
wherein research staff within the Sarah Cannon Network trial enrollment. This increased complexity has been
manage patient eligibility for molecularly motivated clinical magnified by an increasingly crowded molecular di-
trials. Genospace surfaces potential matches based on agnostics landscape. Substantial challenges persist with
proprietary clinicogenomic matching algorithms and allows regard to integrating NGS in drug development. First, tissue
for secure communication between MOSS members and scarcity remains a large barrier to NGS adoption and the
clinic staff regarding potential eligibility. Knowing that not all increased need for on-study biopsies exacerbates the
clinicians can attend MTBs or are willing to log in to an scarcity challenge. Second, the majority of NGS testing for
additional web-based portal, MolecularHELP inboxes have patients with cancer has been performed in academic
been initiated to answer questions submitted by email or medical centers. As that paradigm shifts and more
phone. The remit of MOSS team members is to provide community-based medical oncology practices integrate
mutation-level analysis on potential clinical trial eligibility, NGS as part of standard of care, more robust education
standard-of-care therapies, contraindications, and prognos- programs are required to ensure that community medical
tic (including potential germline mutations) indications. oncologists are equipped to act on findings from NGS re-
Providing MOSS services at Sarah Cannon has correlated ports. Finally, as NGS testing and clinical research are in-
with a notable year-over-year increase in molecular test tegrated into the community setting, there will be a need to
ordering (Fig. 1E). Importantly, molecular tests ordered aggregate and share data from these centers. Challenges
within the Sarah Cannon Network are ingested and har- will remain in harmonizing the genomic results obtained
monized into Genospace’s Population Analytics platform using different sequencing platforms, but efforts like Sarah
where we perform clinicogenomic analyses. These analyses Cannon’s Genospace are aiming to solve these complex
inform clinical trial feasibility, coinhibitory drug development issues and provide harmonized data sets from community
strategies, and the exploration of potential mechanisms of medical oncology practices across the United States. These
resistance. Our partnered medical oncologists are con- challenges notwithstanding, identifying molecular drivers of
tributing to research by ordering molecular profiling tests, cancers and appropriately populating clinical trials will lead
even if the tests do not reveal immediately actionable results. to the more efficient approval of effective anticancer ther-
This connection of community-based medical oncologists apeutics. The wide adoption of NGS, integration of MOSS,
into a research program provides an unprecedented ability to and the evolving clinical trial landscape are all converging to
capture, track, and analyze real-world clinical and genomic pave a new way forward for drug development.

1
Sarah Cannon Research Institute, Nashville, TN AND DATA AVAILABILITY STATEMENT
EDBK_100024.
Andrew J. McKenzie, PhD, Sarah Cannon Research Institute, 1100 Dr.
Martin Luther King Jr. Blvd., Suite 800, Nashville, TN 37203; email:
[email protected].

Jones and McKenzie
REFERENCES
1. Siu LL, Conley BA, Boerner S, et al. Next-generation sequencing to guide clinical trials. Clin Cancer Res. 2015;21:4536-4544.
2. Copur MS, Jonglertham P. Alpelisib for PIK3CA-mutated advanced breast cancer. N Engl J Med. 2019;381:686-687.
3. Larkin J, Chiarion-Sileni V, Gonzalez R, et al. Combined nivolumab and ipilimumab or monotherapy in untreated melanoma. N Engl J Med. 2015;373:23-34.
4. Kotecha RR, Motzer RJ, Voss MH. Towards individualized therapy for metastatic renal cell carcinoma. Nat Rev Clin Oncol. 2019;16:621-633.
5. Camidge DR, Doebele RC, Kerr KM. Comparing and contrasting predictive biomarkers for immunotherapy and targeted therapy of NSCLC. Nat Rev Clin Oncol.
2019;16:341-355.
6. Cocco E, Scaltriti M, Drilon A. NTRK fusion-positive cancers and TRK inhibitor therapy. Nat Rev Clin Oncol. 2018;15:731-747.
7. Llovet JM, Montal R, Sia D, et al. Molecular therapies and precision medicine for hepatocellular carcinoma. Nat Rev Clin Oncol. 2018;15:599-616.
8. Romond EH, Perez EA, Bryant J, et al. Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med. 2005;
353:1673-1684.
9. Piccart-Gebhart MJ, Procter M, Leyland-Jones B, et al; Herceptin Adjuvant (HERA) Trial Study Team. Trastuzumab after adjuvant chemotherapy in HER2-
10. Hochhaus A, Larson RA, Guilhot F, et al; IRIS Investigators. Long-term outcomes of imatinib treatment for chronic myeloid leukemia. N Engl J Med. 2017;
376:917-927.
11. Druker BJ, Guilhot F, O’Brien SG, et al; IRIS Investigators. Five-year follow-up of patients receiving imatinib for chronic myeloid leukemia. N Engl J Med. 2006;
355:2408-2417.
12. O’Brien SG, Guilhot F, Larson RA, et al; IRIS Investigators. Imatinib compared with interferon and low-dose cytarabine for newly diagnosed chronic-phase chronic
myeloid leukemia. N Engl J Med. 2003;348:994-1004.
13. Bang Y-J, Van Cutsem E, Feyereislova A, et al; ToGA Trial Investigators. Trastuzumab in combination with chemotherapy versus chemotherapy alone for
treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet. 2010;
376:687-697.
14. Demetri GD, von Mehren M, Blanke CD, et al. Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. N Engl J Med. 2002;
347:472-480.
15. Kessler D, Gmachl M, Mantoulidis A, et al. Drugging an undruggable pocket on KRAS. Proc Natl Acad Sci U S A. 2019;116:15823-15829.
16. Hallin J, Engstrom LD, Hargis L, et al. The KRASG12C inhibitor MRTX849 provides insight toward therapeutic susceptibility of KRAS-mutant cancers in mouse
models and patients. Cancer Discov. 2020;10:54-71.
17. Liu P, Wang Y, Li X. Targeting the untargetable KRAS in cancer therapy. Acta Pharm Sin B. 2019;9:871-879.
18. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology: Non-Small Cell Lung Cancer. Version 4.2020. https://www.nccn.org/
professionals/physician_gls/pdf/nscl.pdf. Accessed May 20, 2020.
19. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology. Colon Cancer. Version 3.2020. https://www.nccn.org/professionals/
physician_gls/pdf/colon.pdf. Accessed May 20, 2020.
20. Novartis. Novartis investigational lung cancer therapy capmatinib (INC280) granted FDA Breakthrough Therapy designation for patients with MET-mutated
advanced non-small cell lung cancer. www.novartis.com/news/media-releases/novartis-investigational-lung-cancer-therapy-capmatinib-inc280-granted-fda-
breakthrough-therapy-designation-patients-met-mutated-advanced-non-small-cell-lung. Accessed March 13, 2020.
21. European Society for Medical Oncology. LOXO-292 demonstrates promising anti-tumour activity. www.esmo.org/oncology-news/loxo-292-demonstrates-
promising-anti-tumour-activity-in-ret-altered-thyroid-cancer. Accessed March 13, 2020.
22. Overman MJ, Lonardi S, Wong KYM, et al. Durable clinical benefit with nivolumab plus ipilimumab in DNA mismatch repair-deficient/microsatellite instability-
high metastatic colorectal cancer. J Clin Oncol. 2018;36:773-779.
23. Overman MJ, McDermott R, Leach JL, et al. Nivolumab in patients with metastatic DNA mismatch repair-deficient or microsatellite instability-high colorectal
cancer (CheckMate 142): an open-label, multicentre, phase 2 study. Lancet Oncol. 2017;18:1182-1191.
24. Freedman AN, Klabunde CN, Wiant K, et al. Use of next-generation sequencing tests to guide cancer treatment: results from a nationally representative survey of
oncologists in the United States. JCO Precis Oncol. 2018;2:1-13.
25. Akkari Y, Smith T, Westfall J, et al. Implementation of cancer next-generation sequencing testing in a community hospital. Cold Spring Harb Mol Case Stud. 2019;
5:a003707.
26. Haslem DS, Van Norman SB, Fulde G, et al. A retrospective analysis of precision medicine outcomes in patients with advanced cancer reveals improved
progression-free survival without increased health care costs. J Oncol Pract. 2017;13:e108-e119.
27. Haslem DS, Chakravarty I, Fulde G, et al. Precision oncology in advanced cancer patients improves overall survival with lower weekly healthcare costs.
Oncotarget. 2018;9:12316-12322.
28. Leighl NB, Page RD, Raymond VM, et al. Clinical utility of comprehensive cell-free DNA analysis to identify genomic biomarkers in patients with newly diagnosed
metastatic non–small cell lung cancer. Clin Cancer Res. 2019;25:4691-4700.
29. Aggarwal C, Thompson JC, Black TA, et al. Clinical implications of plasma-based genotyping with the delivery of personalized therapy in metastatic non-small cell
lung cancer. JAMA Oncol. 2019;5:173-180.

30. Nagahashi M, Shimada Y, Ichikawa H, et al. Next generation sequencing-based gene panel tests for the management of solid tumors. Cancer Sci. 2019;
110:6-15.
31. Solomon JP, Benayed R, Hechtman JF, et al. Identifying patients with NTRK fusion cancer. Ann Oncol. 2019;30(suppl 8):viii16-viii22.
32. Kuderer NM, Burton KA, Blau S, et al. Comparison of 2 commercially available next-generation sequencing platforms in oncology. JAMA Oncol. 2017;3:996-998.
33. Chow-White P, Ha D, Laskin J. Knowledge, attitudes, and values among physicians working with clinical genomics: a survey of medical oncologists. Hum Resour
Health. 2017;15:42.
34. Ha VTD, Frizzo-Barker J, Chow-White P. Adopting clinical genomics: a systematic review of genomic literacy among physicians in cancer care. BMC Med
Genomics. 2018;11:18.
35. Foundation Medicine. FoundationOne CDx technical specifications. https://assets.ctfassets.net/vhribv12lmne/4ZHUEfEiI8iOCk2Q6saGcU/

11dd3b532e30c34f56cb8e9b4a896783/F1CDx_TechSpecs_10-06_digital.pdf. Accessed April 9, 2020.
36. Foundation Medicine. FoundationOne Liquid technical specifications. https://assets.ctfassets.net/vhribv12lmne/3SPYAcbGdqAeMsOqMyKUog/
4e0d771e88afc920dc1a6f0515e2ff83/F1L_TechnicalInformation_10.pdf. Accessed April 9, 2020.
37. Foundation Medicine. FoundationOne Heme technical specifications. https://assets.ctfassets.net/vhribv12lmne/zBxaQC12cScqgsEk8seMO/
c32a7d1adf083cb0f5d0c0b2439fdb87/F1H_Technical_Information.pdf. Accessed April 9, 2020.
38. Zehir A, Benayed R, Shah RH, et al. Mutational landscape of metastatic cancer revealed from prospective clinical sequencing of 10,000 patients. Nat Med. 2017;
23:703.
39. Caris. Caris Molecular Intelligence. https://www.carismolecularintelligence.com/wp-content/uploads/2017/03/Profile-Menu-Brochure.pdf. Accessed May 18,
2020.
40. Guardant. Guardant 360. www.guardant360.com. Accessed April 9, 2020.

41. ParadigmDx. PCDx. www.paradigmdx.com/wp-content/uploads/2019/07/technical-document-2-22-19-for-Oncologist.pdf. Accessed April 9, 2020.
42. Tempus. Next generation sequencing. www.tempus.com/genomic-sequencing/. Accessed April 9, 2020.
43. Life Technologies. OncomineDx Target Test. www.thermofisher.com/order/catalog/product/A32451. Accessed April 9, 2020.
44. OmniSeq. OmniSeq Advance. www.omniseq.com/omniseq-advance-assay/. Accessed April 9, 2020.
45. OmniSeq. OmniSeq Comprehensive. www.omniseq.com/comprehensive/. Accessed April 9, 2020.

46. PathGroup. SmartGenomics. www.pathgroup.com/oncology/smartgenomics/. Accessed April 9, 2020.
47. NeoGenomics. NeoType Discovery. https://neogenomics.com/test-menu/neotype-discovery-profile-solid-tumors. Accessed April 9, 2020.
48. Illumina. Trusight Oncology 500. www.illumina.com/products/by-type/clinical-research-products/trusight-oncology-500.html. Accessed April 9, 2020.

49. U.S. Food and Drug Administration. Hematology/oncology (cancer) approvals and safety notifications. www.fda.gov/drugs/resources-information-approved-
drugs/hematologyoncology-cancer-approvals-safety-notifications. Accessed March 13, 2020.
50. U.S. Food and Drug Administration Center for Drug Evaluation and Research. Guidance for Industry: Expedited Programs for Serious Conditions—Drugs and
Biologics. www.fda.gov/regulatory-information/search-fda-guidance-documents/expedited-programs-serious-conditions-drugs-and-biologics. Accessed April 9,
2020.
51. Chuk MK, Chang JT, Theoret MR, et al. FDA approval summary: accelerated approval of pembrolizumab for second-line treatment of metastatic melanoma. Clin
Cancer Res. 2017;23:5666-5670.
52. Kang SP, Gergich K, Lubiniecki GM, et al. Pembrolizumab KEYNOTE-001: an adaptive study leading to accelerated approval for two indications and a companion
diagnostic. Ann Oncol. 2017;28:1388-1398.
53. Knepper TC, Bell GC, Hicks JK, et al. Key lessons learned from Moffitt’s molecular tumor board: the Clinical Genomics Action Committee experience. Oncologist.
2017;22:144-151.
54. Harada S, Arend R, Dai Q, et al. Implementation and utilization of the molecular tumor board to guide precision medicine. Oncotarget. 2017;8:57845-57854.
55. van der Velden DL, van Herpen CML, van Laarhoven HWM, et al. Molecular tumor boards: current practice and future needs. Ann Oncol. 2017;28:3070-3075.
56. Radovich M, Kiel PJ, Nance SM, et al. Clinical benefit of a precision medicine based approach for guiding treatment of refractory cancers. Oncotarget. 2016;
7:56491-56500.
57. Ortiz MV, Kobos R, Walsh M, et al. Integrating genomics into clinical pediatric oncology using the molecular tumor board at the Memorial Sloan Kettering Cancer
Center. Pediatr Blood Cancer. 2016;63:1368-1374.
58. McGowan ML, Ponsaran RS, Silverman P, et al. “A rising tide lifts all boats”: establishing a multidisciplinary genomic tumor board for breast cancer patients with
advanced disease. BMC Med Genomics. 2016;9:71.
59. Tafe LJ, Gorlov IP, de Abreu FB, et al. Implementation of a molecular tumor board: the impact on treatment decisions for 35 patients evaluated at Dartmouth-
Hitchcock Medical Center. Oncologist. 2015;20:1011-1018.
60. Schwaederle M, Parker BA, Schwab RB, et al. Molecular tumor board: the University of California-San Diego Moores Cancer Center experience. Oncologist. 2014;
19:631-636.
61. Rieke DT, Lamping M, Schuh M, et al. Comparison of treatment recommendations by molecular tumor boards worldwide. JCO Precis Oncol. 2018;2:1-14.

Jones and McKenzie
62. Dalton WB, Forde PM, Kang H, et al. Personalized medicine in the oncology clinic: implementation and outcomes of the Johns Hopkins molecular tumor board.
JCO Precis Oncol. 2017;1:1-9.
63. Basse C, Morel C, Alt M, et al. Relevance of a molecular tumour board (MTB) for patients’ enrolment in clinical trials: experience of the Institut Curie. ESMO Open.
2018;3:e000339.
64. Trivedi H, Acharya D, Chamarthy U, et al. Implementation and outcomes of a molecular tumor board at Herbert-Herman Cancer Center, Sparrow Hospital. Acta
Med Acad. 2019;48:105-115.
65. Moore DA, Kushnir M, Mak G, et al. Prospective analysis of 895 patients on a UK genomics review board. ESMO Open. 2019;4:e000469.
66. Mangat PK, Halabi S, Bruinooge SS, et al. Rationale and design of the Targeted Agent and Profiling Utilization Registry (TAPUR) study. JCO Precis Oncol. 2018;
2:1-14.
67. Pishvaian MJ, Blais EM, Bender RJ, et al. A virtual molecular tumor board to improve efficiency and scalability of delivering precision oncology to physicians and
their patients. JAMIA Open. 2019;2:505-515.
68. Marshall CL, Petersen NJ, Naik AD, et al. Implementation of a regional virtual tumor board: a prospective study evaluating feasibility and provider acceptance.
Telemed J E Health. 2014;20:705-711.
69. Gray SW, Hicks-Courant K, Cronin A, et al. Physicians’ attitudes about multiplex tumor genomic testing. J Clin Oncol. 2014;32:1317-1323.
70. AACR Project GENIE Consortium. AACR Project GENIE: powering precision medicine through an international consortium. Cancer Discov. 2017;7:818-831.
71. Beaubier N, Bontrager M, Huether R, et al. Integrated genomic profiling expands clinical options for patients with cancer. Nat Biotechnol. 2019;37:1351-1360.
72. Siu LL, Lawler M, Haussler D, et al. Facilitating a culture of responsible and effective sharing of cancer genome data. Nat Med. 2016;22:464-471.
73. Jensen MA, Ferretti V, Grossman RL, et al. The NCI Genomic Data Commons as an engine for precision medicine. Blood. 2017;130:453-459.

GASTROINTESTINAL
CANCER—COLORECTAL
AND ANAL

Controversies in Rectal Cancer Treatment

and Management
Weijing Sun, MD, FACP1; Raed Al-Rajabi, MD1; Rodrigo O. Perez, MD, PhD2; Saquib Abbasi, MD1; Ryan Ash, MD3; and
Angelita Habr-Gama, MD, PhD2
overview
Incorporation of new treatment modalities has significantly increased the complexity of the treatment and
management of rectal cancer, including perioperative therapy for local advanced disease and organ pres-
ervation for those with response to the preoperative treatment. This review may help practitioners better
understand the rationale and selection.
INTRODUCTION by total mesorectal excision and postoperative ad-

Colorectal cancer is still a top threat to human life juvant chemotherapy (Fig. 1).4-13 This standard ther-
worldwide. The global burden of colorectal cancer is apeutic approach reduced the local recurrence rate to
expected to increase by 60% to more than 2.2 million 5% to 10%, with improvement in the overall survival
new cases and 1.1 million deaths by 2030.1 Globally, (OS) rate as well. However, increasing the cure rate,
colorectal cancer incidence and mortality rates vary especially disease-specific survival, while preserving
widely, with distinct gradients across human devel- organ function (e.g., rectal sphincter preservation, sexual
opment index (HDI) levels. Increasing incidence and function), improving patients’ quality of life, and avoiding
mortality have been observed in rapidly transitioning over-treatment and unnecessary invasive procedures are
countries (e.g., medium- and high-HDI countries in- the main challenges we are facing. Many investigations
cluding those in the Baltics, as well as Russia, China, are focusing on these tasks, such as total neoadjuvant
and Brazil); increasing incidence, with concomitant therapy (TNT) and watch and wait for patients who have
decreases in mortality, have been seen in very high– achieved complete clinical response (cCR) by planning
HDI countries including Canada, the United Kingdom, surgical resection as a salvage method for those with local
Denmark, and Singapore; and decreasing incidence recurrence instead of routinely after nCRT is finished.
and mortality have been observed in a number of the Emerging research is looking at decreasing the dosage or
highest-HDI countries including the United States, length of radiation, even omitting radiation for selected
Japan, and France.2 In the United States, an estimated patient populations. One important factor is that ad-
104,610 cases of colon cancer and 43,340 cases of vances in techniques and methods have also made these
rectal cancer will be diagnosed in 2020, and a total of investigations possible, reliable, and reproducible, with
53,200 deaths from colorectal cancers (approximately the diagnosis and assessment of treatment responses
one in three deaths from rectal cancer).3 Although the more precise and accurate.
overall incidence of colorectal cancer has been trending
down since 1980 because of the decrease in risk factor ROLE OF RADIOLOGY IN THE MULTIDISCIPLINARY
exposure and increasing rates of screening in adults MANAGEMENT OF RECTAL CANCER
Author affiliations
age 55 and older, the incidence has increased by 2% Diagnostic Staging: Local Versus Locally Advanced
and support annually among adults younger than age 55. More Transrectal ultrasound provides detailed spatial reso-
information (if importantly, the mortality rate has increased by 1% lution, which can provide an accurate depth of in-
applicable) appear annually among the population younger than age 55
at the end of this
vasion. This is specifically beneficial in differentiating
in the United States. between T2 and T3 disease, which may be difficult to
article.
Accepted on Worldwide, after decades of research and effort, the distinguish on MRI. One major limitation of transrectal
February 24, 2020 multidisciplinary approach for patients with locally ultrasound is that it is operator dependent. Ultrasound
and published at advanced rectal cancer has been standardized. The also has a limited field of view and cannot visualize all
ascopubs.org on
most accepted standard treatment for locally advanced regional rectal nodal stations.14,15 High-field–strength
April 2, 2020:
(T3/T4 or node-positive rectal adenocarcinoma) is a fluo- pelvic MRI with 1.5- or 3-Tesla magnets acquiring
10.1200/EDBK_ rouracil agent (5-fluorouracil [5-FU] or capecitabine)– multiplanar, thin section (3 mm or less), non–fat-
279871 based neoadjuvant chemoradiotherapy (nCRT) followed saturated T2-weighted images has become routine

a better quality of life. Adequately evaluating response to

neoadjuvant chemotherapy is crucial to create an in-
dividualized treatment plan so health care providers can
• It is important to understand the balance of avoid overtreatment, limit the side effects from neoadjuvant
cancer risk and organ preservation for patients
chemotherapy, and decide on the optimal time for surgery.
with locally advanced rectal cancer.
Currently, anatomic and functional T2-weighted and diffuse-
• Multidisciplinary collaboration is needed for weighted MRI followed by rectal examination and endoscopy
decision making, optimizing multidisciplinary is the optimal strategy to evaluate good response to
management to minimize long-term morbidity
chemoradiation.24-26 Emerging technologies looking at tumor
(e.g., options of “standard” vs. TNT, non-
volume reduction and tumor volume reduction rate evaluated
operative management watch and wait
strategy). by 3D-CUBE could predict neoadjuvant chemotherapy effi-
cacy.27 Radiomics is the automatic extraction of numerous
• Precise image evaluation of chemotherapy and
quantitative imaging features from comprehensive quantifi-
chemoradiation treatment efforts is crucial in
cation of tumor phenotypes. Prediction models based on MRI
decisionmaking.
radiomics can noninvasively predict tumor response after
chemoradiation, identifying patients who are eligible for
in preoperative evaluation for initial local rectal tumor staging, organ-preserving treatment.28
especially high-resolution oblique axial images. The use of Neoadjuvant Rectal Score
intravenous contrast MRI is controversial in its utility in initial
The use of surrogate endpoints enables clinicians and re-
staging, including detecting depth of invasion and nodal
searchers to use clinical and pathologic data to predict long-
metastasis. Diffusion-weighted images are often acquired
term outcomes such as OS and disease-free survival. One
and are especially important for post-therapeutic evaluation.
surrogate is the neoadjuvant rectal (NAR) score.29 The score
MRI provides excellent soft tissue evaluation, including
ranges from 0 to 100, with higher scores indicating a worse
evaluating the muscularis propria, mesorectum, and meso-
prognosis. This range is divided into three subgroups: low,
rectal fascia, which helps form the circumferential resection
intermediate, and high NAR categories. These categories
margin (CRM). The CRM is predictive for treatment and
were significantly associated with OS (p , .0001), with 5-
prognosis of rectal cancer according to the eighth edition of
year OS values of 92%, 89%, and 68%, respectively. This
the American Joint Commission on Cancer Cancer Staging
approach uses the Valentini nomogram for predicting local
Manual. CRM invasion is an independent risk factor for local
recurrence and distant metastasis and OS for patients with
tumor recurrence and a low survival rate.16,17 MRI is also able
locally advanced rectal cancer.30
to identify nodal involvement in all rectal nodal stations in the
lower abdomen and pelvis by evaluating the size, morphol-
½5pN 3ðcT pT Þ þ 122
ogy, and signal of the node (Figs. 2 and 3). Features such as NAR ¼
9:61
signal inhomogeneity can predict nodal metastasis.18 MRI
has the unique ability to evaluate vascular involvement by
where cT = clinical tumor stage, NAR = neoadjuvant rectal
demonstrating altered vascular enhancement or signal that
score, pN = pathologic lymph node stage, and pT =
can indicate extramural vascular invasion, which is another
pathologic tumor stage.
poor prognostic indicator.19,20
ROLE OF IMAGING IN GUIDING SURGICAL DECISIONS
EVALUATION OF RESPONSE AND GUIDANCE
OF MANAGEMENT The MERCURY trial showed that extension of the tumor
within 1 mm of the CRM on imaging was associated with
Post–Neoadjuvant Treatment Assessment 92% accuracy in predicting margin involvement at the time
One of five patients with locally advanced rectal cancer is of surgery.31 The accurate detection of clinical complete
insensitive to neoadjuvant therapy, experiencing recurrence pathologic response is a surrogate marker for pCR. Com-
or developing distant metastasis after or even during ther- bining 18F-FDG PET/CT, diffuse weighted imaging, and T2-
apy.21 The response rates in rectal cancers are heteroge- weighted MRI volumetry obtained before CRT and before
neous, and approximately 20% of patients will achieve surgery may help physicians select patients with rectal
a pathologic complete response (pCR).22 This led to the idea cancer for organ preservation.32 Long-term MRI charac-
of a watch-and-wait approach for patients with a complete teristics of complete responders who undergo a watch-and-
pathologic response. Habr-Gama et al23 found that OS and wait approach as an alternative to surgery can play a critical
disease-free survival at 10 years were 97.7% and 84%, role in following these patients. The European Society of
respectively. With organ preservation, the patient is likely to Gastrointestinal and Abdominal Radiology initiated an ex-
develop fewer treatment-related symptoms and maintain pert consensus meeting on MRI. Lambregts et al33 looked at

Sun et al
FIGURE 1. Chronology of Locally

Advanced Rectal Cancer (LARC)
Management
Abbreviations: NSABP, National
Surgial Adjuvant Breast and Bowel
Project; GITSG, Gastrointestional
Tumor Study Group; CRT, che-
moradiotherapy; NCCTG, North
Central Cancer Treatment Group;
XRT, radiation therapy; NIH, Na-
tional Institutes of Health; TME,
total mesorectal excision; EORTC, European Oraganisation for the Research and Treatment of Cancer; TNT, total neoadjuvant therapy.
1,509 T2-weighted MRI scans of 164 patients who sus- Traditionally, TNT was not used because it would delay
tained complete response and were undergoing watch and potentially curative surgery and increase the difficulty of
wait; the investigators evaluated the tumor bed by MRI. Two surgery with prolonged gaps after radiation. Biologically,
independent reviewers classified the findings on MRI to no concerns have been raised about potentially hastening
fibrosis, minimal fibrosis, full-thickness fibrosis, or irregular tumor cell repopulation by activating mitogenic signaling
fibrosis; and 93% of patients’ morphology remained com- transduction and preferential selection of radiation-resistant
pletely stable during the follow-up, and in 7% a minor in- clones, resulting in reduced radiosensitivity.37 Other con-
crease or decrease of fibrosis was observed. Most complete cerns include the lack of clear benefits in improving local
responders showed fibrosis, although the majority of pa- control and survival outcomes, based on histologic data on
tients at 64% showed minimal fibrosis and 10% of the cases the utility of neoadjuvant chemotherapy in other tumors
had no fibrosis. All rectal wall morphology established on including lung, head and neck, and anal cancer. Evidence
restaging MRI typically remained unchanged during long- from recent investigations suggests that neoadjuvant
term MRI follow-up. Maxime et al34 evaluated 880 rectal chemotherapy has the potential to maximize compliance
cancer complete responders who were identified at 47 without compromising a patient’s ability to undergo planned
participating institutions and followed up for 3.3 years. The CRT or increasing the risk of surgical complications.
2-year cumulative incidence of local regrowth was 25.2%,
and 88% of all local regrowth was diagnosed in the first
2 years. Furthermore, 97% of local regrowth was located in
the bowel wall. Distant metastases were diagnosed in 8%.
The 5-year OS rate was 85%.
TOTAL NEOADJUVANT THERAPY

The standard treatment plan has shown pCR rates of
approximately 20%, and local recurrence seems to be less
of a concern; however, the estimated 5-year distant
metastatic/recurrent disease rate is 35%, making it the
leading cause of cancer death in rectal cancer.11-13 The
main cause of no clear benefit in improving OS or cancer-
specific survival is poor patient compliance with post-
operative adjuvant chemotherapy, a serious concern in
clinical practice.35,36
To improve OS, especially disease-specific survival, with
delivery of systemic chemotherapy to treat occult micro-
metastases early and increase treatment compliance, com-
plete systemic chemotherapy and chemoradiation before
surgery were proposed, a method known as TNT.37 There are FIGURE 2. High-Resolution T2-Weighted Images
two main modes of TNT based on the treatment sequencing Oblique (A) and sagittal (B) images demonstrating a stage T2 low rectal
and incorporation of systemic chemotherapy and chemo- tumor (arrows) without extension through the muscularis propria.
radiation: delivering systemic chemotherapy after nCRT, or Oblique (C) and coronal (D) images demonstrating a stage T3 low rectal
consolidation, and delivering systemic chemotherapy before tumor (arrows) with extension through the muscularis propria but not to
nCRT, or induction. the mesorectal fascia.

Garcia-Aguilar et al 38 conducted a prospective non-

randomized multicenter phase II study to evaluate the effect
of adding various cycles of systemic chemotherapy modi-
fied FOLFOX6 (mFOLFOX6) between standard chemo-
radiation (fluorouracil 225 mg/m2/day by continuous infusion
throughout radiotherapy, and 4.0 Gy in 25 fractions, 5 days
per week for 5 weeks, followed by a minimum boost of 5.4 Gy)
and surgery. This study showed that the pCR rate increased
with the chemoradiation-to-surgery interval and with an in-
creased number of cycles of mFOLFOX6 in different groups
(25% in two cycles, 30% in four cycles, and 38% in six cycles
compared with 18% in the group without FOLFOX; p =
.0036). There was no increased surgical complications or
surgical technical difficulty after surgery was delayed for up to
20 weeks. With no increased disease progression or distant
metastasis, the study demonstrated that consolidation che-
motherapy with FOLFOX is safe. By the time the article was
published, long-term oncologic outcomes were not available.
Treatment compliance with neoadjuvant chemotherapy was
77%, which is much better than the 40% to 60% for the
adjuvant chemotherapy.35,36 Authors also found that adding
mFOLFOX6 after chemoradiation and before total mesorectal
excision has the potential to increase the proportion of pa-
tients eligible for less invasive treatment strategies. However,
the relative contribution of chemotherapy and waiting period
to the rate of pCR is unclear. The ongoing phase III ran- FIGURE 3. High-Resolution T2-Weighted Images
domized trial RAPIDO, comparing short-course radiotherapy Oblique (A) and sagittal (B) high-resolution T2-weighted images dem-
followed by capecitabine and oxaliplatin chemotherapy be- onstrating a stage T4 low rectal tumor (arrows) with multifocal extension
fore surgery to standard long-course nCRT, may further through the mesorectal fascia with invasion of multiple surrounding
confirm these findings and resolve ambiguity.39 A large phase structures. Sagittal (C) SSFSE T2-weighted and sagittal (D) high-resolution
III trial by Polish investigators evaluated the local and long- T2-weighted images demonstrating a high rectal tumor with extramural
term outcome benefits of consolidation chemotherapy and vascular invasion (arrow).
short-course radiation.40 In this trial, patients with fixed cT3 or
cT4 cancer were randomly assigned to receive either 5 course radiotherapy followed by consolidation chemother-
5 Gy and three cycles of FOLFOX4 or 50.4 Gy in 28 fractions apy than the long-course CRT group. The authors hypothesize
combined with two 5-day cycles of bolus 5-FU 325 mg/m2/ that irradiation may activate antitumor immune responses
day and leucovorin 20 mg/m2/day during the first and fifth during the period before surgery.Recently published long-
week of irradiation along with five infusions of oxaliplatin term follow-up data showed that the 7-year OS rate was not
50 mg/m2 once weekly. There were no differences in local significant (HR, 0.90; 95% CI, 0.70–1.15; p = .38).41 The
efficacy between 5 5 Gy with consolidation chemotherapy median follow-up was 7.0 years. However, the difference in
and long-course chemoradiation. The R0 resection pCR rates early OS favoring short-course therapy was observed again
were 77% versus 71% (p = .07) and 16% vs. 12% (p = .17), (9% at 3 years [95% CI, 0.5%–17%]). There was no dif-
respectively. Interestingly, an increased OS (3-year OS and ference in disease-free survival (HR, 0.95; 95% CI, 0.75–
disease-free survival rates of 73% vs. 65%; p = .046) and 1.19; p = .65) at 8 years (43% vs. 41%). These studies
lower acute toxicity (p = .006) favor the 5 5 Gy schedule provided evidence that consolidation chemotherapy can be
with consolidation chemotherapy, with the rates of distant a safe and feasible option in the multimodal approach.
metastasis and local recurrence being similar in the two
The approach of delivering systemic chemotherapy before
groups.
neoadjuvant therapy and surgical resection as induction
The intervals between the start of irradiation and surgery therapy is particularly appealing for locally advanced dis-
were similar between the two groups at approximately 12 eases with a high risk of distant metastasis, such as ex-
weeks. No differences in disease-free survival, local failure tramural venous invasion and lateral pelvic lymph node
rate, or distant metastases were reported; however, better metastasis. The potential benefits of upfront systemic
treatment compliance was found in the patients given short- chemotherapy for primary rectal cancer are that the

Sun et al
chemotherapeutic agents are delivered to the primary tumor recurrence leading to salvage surgery with no oncologic
directly without disrupting vasculature with radiation or compromise.53
surgery. Therefore, the effectiveness of chemotherapy can
RATIONALE FOR ORGAN PRESERVATION IN RECTAL CANCER
be maximized. Meanwhile, upfront full doses of systemic
chemotherapy may also eliminate possible occult distal There has been an increase in popularity of organ-preserving
metastases. Although there are no available data from large strategies for the treatment of rectal cancer in the last few
randomized phase III studies, information from trials with years. The main reasons for avoiding a proctectomy are its
induction systemic chemotherapy showed increased pCR postoperative morbidity, including long-term urinary, sexual,
rates, excellent treatment compliance without adverse effect and fecal continence dysfunction in addition to the need for
that would delay treatment, and early identification of temporary or permanent stomas associated with the pro-
nonresponders.37,42-44 The CONTRE trial, a phase II single- cedure. Postoperative mortality can also be substantial,
arm study, included patients with T3 to T4 or N1 to N2 rectal depending on medical comorbidities and patient age.54 In
cancer who received eight cycles of mFOLFOX6 followed by select patients with evidence of complete primary tumor
capecitabine 825 mg/m twice daily concurrent with 50.4 Gy regression, alternative surgical and even nonsurgical ap-
radiation therapy followed by surgery performed 6 to proaches have been suggested in order to mitigate potential
10 weeks later.45 The study demonstrated 35% pCR, which morbidity and mortality.55
is similar to the result of a consolidation chemotherapy The observation that rectal cancers can develop tumor
trial.38 The compliance rate was more than 90%, and the R0 regression with reduction in primary tumor size (downsiz-
resection rate was 100%. The Spanish GCR-3 phase II trial ing), depth of tumor penetration, and even potential nodal
randomly assigned patients with distal or middle third, T3 to sterilization (downstaging) sets the stage for organ-preserving
T4 or N+ rectal adenocarcinoma to receive standard pre- alternatives including local excision of small and superficial
operative CRT followed by surgery and four cycles of post- residual tumors.56 Furthermore, regression of the primary
operative adjuvant capecitabine and oxaliplatin (CAPOX) or tumor can result in complete disappearance of the tumor in
four cycles of CAPOX followed by CRT and surgery.46 With the resected specimen (pCR) for some patients. For a subset
a median follow-up of 69.5 months, there was no difference of these patients, complete disappearance of the primary
in 5-year disease-free survival and OS. However, the in- tumor is already clinically detected before surgical resection
duction CAPOX arm showed lower toxicity and better com- and referred to as a cCR.57 The patients with complete tumor
pliance. It appears that six to eight cycles of induction regression of their primary rectal cancers in response to nCRT
chemotherapy are needed to achieve the optimal pCR of are the ideal patients for organ-preserving strategies including
35% or above. So far, no extra benefit from adding target- no immediate surgical resection of the area harboring the
originated agents (cetuximab, bevacizumab, or aflibercept) to original cancer.58 To even consider these approaches, co-
induction systemic chemotherapy (either FOLFOX or CAPOX) lorectal surgeons must consider several aspects of the dis-
has been shown.47-49 ease, patients, and treatment modalities that may be relevant
SURGICAL THERAPY OF RECTAL CANCER during clinical decision-making.
The incorporation of new treatment modalities has in- Assessment of Tumor Response
creased the complexity of rectal cancer management.50
In considering patients for an organ-preserving strategy, an
Although definitive surgery is the standard of care for
important issue is identifying patients who have significant
early-stage rectal cancer, there have been advances in
tumor regression with nCRT. The approach to assessing
surgical technique and increased interest in the different
tumor response is critical for identifying these patients.
approaches to total mesorectal excision. These include
However, assessment of tumor response may be chal-
standard laparoscopic, robotic, and transanal total meso-
lenging because of numerous uncertainties including op-
rectal excision. Additionally, the optimal management after
timal timing and clinical and radiologic tools utilized for this
neoadjuvant therapy and assessment of tumor response is
purpose.
unclear. nCRT can lead to tumor regression and, in a subset
of patients, a complete pathologic response. This can be as Assessment of tumor response is also recommended even if
high as 42% of patients.51 The assessment of tumor response an organ-preserving strategy is not being considered as it is
after nCRT and before radical surgery can potentially be used part of surgical planning. After nCRT, even when a radical
to identify a subset of patients who attain a cCR. These resection is planned, the surgeon needs to consider that the
patients can potentially be treated nonoperatively with strict tumor may have considerably changed. The knowledge of
follow-up (watch and wait), thus avoiding unnecessary this “new anatomy” prior to surgical intervention may allow
postoperative morbidity with good long-term oncologic the surgeon to optimize the intraoperative surgical strategy
outcomes and excellent functional results.52 Close surveil- and to know in advance what challenges will be anticipated
lance of these patients may allow early detection of local during the procedure.59 Therefore, the reassessment of

tumor response should be performed for all patients. [NEW In the presence of a cCR by DRE and proctoscopy, en-
SECTION: "Definition and Criteria for a Complete Clinical doscopic biopsies are not recommended. Even in the setting
Response" set as H2] The basic premise for organ pres- of incomplete clinical response, endoscopic biopsy results
ervation with nonoperative management (watch and wait) is should be interpreted with caution. Among patients with
the achievement of a cCR. The strict criteria that defines a significant response, the negative predictive values of
a cCR includes information provided by three assessment these endoscopic biopsies have been reported to be con-
modalities: clinical (provided by digital rectal examination sistently low.61 Therefore, a negative biopsy in the setting of
[DRE]), endoscopic, and radiologic.60 Only patients fulfilling incomplete clinical response does not rule out microscopic
criteria based on these three assessment modalities are residual cancer.
considered for watch and wait. Absence of features con-
Timing
sistent with a cCR in any of the assessment modalities
should raise the suspicion for residual disease and, in this The time required to achieve a cCR varies from patient to
case, surgical management is usually recommended. patient as tumors take variable times to completely regress
after nCRT. In most cases, a cCR is only achieved after
Clinical and Endoscopic Assessment 16 weeks or longer after nCRT. As such, patients will likely
Clinical assessment is one of the most important tools used undergo multiple assessments of tumor response prior to
to evaluate tumor response. Patients with tumor regression being definitively categorized as a cCR. However, the ma-
will commonly experience relief of their symptoms. DRE is jority of the tumor response should already be observed
also an irreplaceable tool for the evaluation of response. The during or within 6 weeks from nCRT. Only patients with
stringent criteria for cCR include the absence of any ir- a major or significant tumor response (“near-complete”
regularity, mass, ulceration, or stenosis during the DRE. The responses) at the first response assessment should be
surface has to be regular and smooth.57 considered as potential candidates for achieving all strict
criteria of a cCR.62,63 Patients are reassessed at 6-8 week
Endoscopic evaluation of the area harboring the original
intervals. Those that develop further response and achieve
tumor is the remaining key component of this clinical as-
all strict criteria of cCR are entered the watch and wait
sessment. It is important to look for any irregularity or su-
program. The vast majority of patients achieve a cCR within
perficial ulcers missed during the DRE. A flat white scar and
34 weeks of nCRT completion. For patients whose disease
telangiectasia are common endoscopic findings among
fails to achieve all strict criteria for cCR, radical surgery is
patients with a cCR. Although flexible scopes may provide
recommended (Fig. 4).60
photographic documentation of endoscopic response, rigid
proctoscopy may suffice for the majority of patients.57 Radiologic Assessment
MRI should be routinely used to assess response after
nCRT, preferably via magnetic resonance tumor regression
grade estimation.64 Patients who achieve no evidence of
disease on clinical and endoscopic examination are only
considered a true complete responder if they show a low–
signal intensity area replacing the area of the previous tu-
mor. The presence of mixed signal intensity in the area of
the previous cancer should raise suspicion of an incomplete
clinical response. In addition to the rectal wall, the meso-
rectum is also at risk for the presence of residual cancer
despite complete primary regression. Therefore, MRI
should also provide the colorectal surgeon with information
regarding possible mesorectal (or even lateral node) in-
volvement regardless of primary tumor response.
Complete Response: Watch-and-Wait Strategy
When a nonoperative strategy for cCR in rectal cancer is
considered, an intensive follow-up is certainly needed.
Patients should be encouraged to adhere to this strict follow-
FIGURE 4. Endoscopic View After Chemotherapy and Chemoradiation up program to allow early recognition of any local or sys-
Endoscopic view of an area consistent with a complete clinical response temic recurrence and thereby increase the chance of
showing whitening of the mucosal scar (line) and multiple areas with a successful salvage treatment. After initial assessment of
submucosal telangiectasia (arrows) response confirming a cCR, visits should be performed

Sun et al
every 1 two 2 months during the first year, every 3 months status on local regrowth or recurrence appears to be limited to
during the second year, and every 6 months thereafter. the first 2 years after the achievement of cCR. Based on
DRE, proctoscopy, and carcinoembryonic antigen level conditional survival estimates, the risk of a local regrowth or
determination are recommended for all visits. Timing for recurrence once patients have sustained a cCR for more than
radiologic assessment during follow-up has not yet been 2 years appears to be independent of baseline T status.74 The
standardized. Routine MRI for the assessment of the International Watch & Wait Database is a web-based data-
rectal wall, mesorectum, and pelvic nodes every 6 months base, initiated in April 2015 with 47 participating institutions
for the first 2 years and yearly thereafter has been our from 15 countries, that contains patient information including
practice.55 tumor characteristics, type of neoadjuvant chemotherapy,
imaging results at diagnosis, and reassessment after treat-
Outcomes ment, as well as disease recurrence, salvage therapy, and
Patients managed nonoperatively under the watch-and- survival. Among 880 patients included with cCR,75 88% of all
wait strategy after a cCR following nCRT were originally local regrowth was diagnosed in the first 2 years, with 97% of
reported to have similar long-term oncological outcomes regrowth located in the bowel wall. The 5-year OS was 85%
to patients with pCR after radical surgery. 58 Of 265 (95% CI, 80.9%–87.7%), with disease-free survival of 94%
patients who received nCRT fluorouracil and leucovorin (95% CI, 91%–96%). Distant metastatic disease was di-
with concurrent radiotherapy for resectable low rectal agnosed in 8% of patients.
adenocarcinoma, those who had a normal DRE, proc-
Close follow-up may allow early detection of regrowths, leading
toscopy with no residual ulcer, and negative biopsy
to similar oncological outcomes for patients with incomplete
results and no evidence of disease on radiographic
clinical response 8 to 12 weeks after CRT completion.76
imaging underwent a watch-and-wait strategy after re-
However, at least two independent retrospective series
ceiving chemoradiation with 5-FU, leucovorin, and 50.
have reported that survival among patients who develop
4 Gy radiation. Of these, 26.8% achieved cCR; another
local regrowth or recurrence is inferior to that of patients
subset of 8.3% who did not achieve cCR subsequently
who sustain a cCR.77,78 In fact, local recurrences (late and
underwent surgical resection and had pCR. The 5-year OS in
early regrowths) are usually amenable to salvage therapies,
the watch-and-wait group was 100%, and disease-free
often allowing sphincter preservation and associated with
survival was 92%. Additional retrospective studies reported
excellent long-term local disease control.53 However, these
by others have consistently shown similar oncological out-
patients seem to be at higher risk for the development of
comes between these subgroups of patients.2265-70 These
systemic recurrences.77
findings further support the idea that patients with cCR
may be spared the surgical morbidity and mortality of Considering that the rate of complete clinical or path-
radical surgery with no oncological compromise.54 In ad- ologic response was historically less than 30% of pa-
dition, functional outcomes of patients managed non- tients across most of the studies, one could assume that
operatively appear to be better than those of patients this treatment strategy could benefit only a limited
treated with radical surgery or with other organ-preserving proportion of patients with rectal cancer. However, the
strategies (e.g., transanal local excision).2252 observation of increased rates of complete response
(clinical or pathologic) to regimens with consolidation
Local recurrences after this treatment strategy are still a con-
chemotherapy and with the inclusion of earlier stages of
cern and may develop at any time during follow-up. The
disease (cT2N0 or candidates for ultra-low resections or
majority of local recurrences appears to develop within the
abdominoperineal resections) may result in nearly 50%
first 12 months of follow-up and may represent limitations in
who ultimately avoid surgical resection.79,80 This finding has
the precise identification of microscopic residual disease
been further confirmed in a prospective trial including patients
among apparent complete clinical responders. For these
with T2 and T3 rectal cancers managed by CRT and an ad-
reasons, these “early recurrences” developing within the
ditional endorectal high-dose brachytherapy boost (total 65 Gy)
initial 12 months of follow-up have been called “early
that showed a 58% cCR rate at 2 years of follow-up without
regrowths” instead.53,70,71 Data from systematic reviews
surgical resection.52
have suggested that the 3-year cumulative risk of local
regrowth or recurrence after initially achieving a cCR is about Finally, in the era of evidence-based medicine, no ran-
25%.72 One study using individual participant data from domized prospective trial has definitively demonstrated the
multiple published series suggested that baseline T status oncological equivalence of watch and wait and radical
was the only relevant risk factor for a local regrowth or re- surgery in the setting of cCR after nCRT.81 Even though such
currence.73 Apparently, for each increase in baseline T status a trial is not likely to be performed, a recent study using
(T2/T3 or T4) there is an estimated increase of 10% in the risk a propensity-score matched cohort analysis comparing
of local regrowth or recurrence. Curiously, the influence of T watch and wait and radical surgery has been designed to

demonstrate noninferiority of the watch and wait ap- after response to neoadjuvant CRT. Patients who develop
proach. Curiously, however, the comparison between groups complete tumor regression with no clinical, endoscopic, or
demonstrated a slight superiority of the nonoperative man- radiologic evidence of residual cancer may be offered no
agement of these patients in terms of survival and a clear immediate surgery and enrolled in a strict surveillance pro-
benefit in colostomy-free survival even when accounting for gram (watch and wait) with excellent functional and ac-
the development of local recurrences.70 ceptable oncological outcomes.
In conclusion, organ preservation in the management of
rectal cancer has become a valid option for select patients

1
University of Kansas Medical Center, Department of Internal Medicine, Weijing Sun, MD, FACP, University of Kansas Medical Center, Department
Medical Oncology Division, Westwood, KS of Radiology, 3901 Rainbow Blvd, MS 4032, Kansas City, KS 66160;
2
Instituto Angelita & Joaquim Gama, São Paulo, Brazil email: [email protected].
3
University of Kansas Medical Center, Department of Radiology, Kansas
City, KS
EDBK_279871.
REFERENCES
1. Ferlay J, Soerjomataram I, Ervik M, et al. GLOBOCAN 2012 v1.0, Cancer Incidence and Mortality Worldwide: IARC Cancer Base No. 11. Lyon, France:
International Agency for Research on Cancer; 2013.
2. Arnold M, Sierra MS, Laversanne M, et al. Global patterns and trends in colorectal cancer incidence and mortality. Gut. 2017;66:683-691.
3. American Cancer Society. Cancer facts & figures 2020. https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/annual-cancer-
facts-and-figures/2020/cancer-facts-and-figures-2020.pdf.
4. Fisher B, Wolmark N, Rockette H, et al. Postoperative adjuvant chemotherapy or radiation therapy for rectal cancer: results from NSABP protocol R-01. J Natl
Cancer Inst. 1988;80:21-29.
5. Thomas PR, Lindblad AS. Adjuvant postoperative radiotherapy and chemotherapy in rectal carcinoma: a review of the Gastrointestinal Tumor Study Group
experience. Radiother Oncol. 1988;13:245-252.
6. Krook JE, Moertel CG, Gunderson LL, et al. Effective surgical adjuvant therapy for high-risk rectal carcinoma. N Engl J Med. 1991;324:709-715.
7. NIH Consensus Conference. Adjuvant therapy for patients with colon and rectal cancer. JAMA. 1990;264:1444-1450.
8. Swedish Rectal Cancer Trial. Improved survival with preoperative radiotherapy in resectable rectal cancer. N Engl J Med. 1997;336:980-987.
9. Sauer R, Liersch T, Merkel S, et al. Preoperative versus postoperative chemoradiotherapy for locally advanced rectal cancer: results of the German CAO/ARO/AIO-
94 randomized phase III trial after a median follow-up of 11 years. J Clin Oncol. 2012;30:1926-1933.
10. van Gijn W, Marijnen CA, Nagtegaal ID, et al; Dutch Colorectal Cancer Group. Preoperative radiotherapy combined with total mesorectal excision for resectable
rectal cancer: 12-year follow-up of the multicentre, randomised controlled TME trial. Lancet Oncol. 2011;12:575-582.
11. Påhlman L, Glimelius B. Pre- or postoperative radiotherapy in rectal and rectosigmoid carcinoma. Report from a randomized multicenter trial. Ann Surg. 1990;
211:187-195.
12. Kapiteijn E, Marijnen CA, Nagtegaal ID, et al; Dutch Colorectal Cancer Group. Preoperative radiotherapy combined with total mesorectal excision for resectable
rectal cancer. N Engl J Med. 2001;345:638-646.
13. Guillem JG, Chessin DB, Cohen AM, et al. Long-term oncologic outcome following preoperative combined modality therapy and total mesorectal excision of locally
advanced rectal cancer. Ann Surg. 2005;241:829-836, NaN-838.
14. Solomon MJ, McLeod RS. Endoluminal transrectal ultrasonography: accuracy, reliability, and validity. Dis Colon Rectum. 1993;36:200.
15. Puli SR, et al. Accuracy of endoscopic ultrasound to diagnose nodal invasion by rectal cancer: a meta-analysis and systemic review. Ann Surg Oncology. 2009;
16:1255.
16. Amin MB, Edge S, Greene F, (eds), et al. AJCC Cancer Staging Manual, 8th ed. New York, NY: Springer International Publishing; 2017.
17. Monson JR, Weiser MR, Buie WD, et al; Standards Practice Task Force of the American Society of Colon and Rectal Surgeons. Practice parameters for the
management of rectal cancer (revised). Dis Colon Rectum. 2013;56:535-550.
18. Brown G, Richards CJ, Bourne MW, et al. Morphologic predictors of lymph node status in rectal cancer with use of high-spatial-resolution MR imaging with
histopathologic comparison. Radiology. 2003;227:371-377.

Sun et al
19. Smith NJ, Barbachano Y, Norman AR, et al. Prognostic significance of magnetic resonance imaging-detected extramural vascular invasion in rectal cancer. Br
J Surg. 2008;95:229-236.
20. Kim TH, Woo S, Han S, et al. The diagnostic performance of MRI for detection of extramural venous invasion in colorectal cancer: a systematic review and meta-
analysis of the literature. AJR Am J Roentgenol. 2019;213:575-585.
21. Edge SB, Compton CC. The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging 226 manual and the future of TNM. Ann Surg Oncol.
2010;17:1471-1474.
22. Maas M, Beets-Tan RGH, Lambregts DMJ, et al. Wait-and-see policy for clinical complete responders after chemoradiation for rectal cancer. J Clin Oncol. 2011;
29:4633-4640.
23. Habr-Gama A, Perez RO, Nadalin W, et al. Operative versus nonoperative treatment for stage 0 distal rectal cancer following chemoradiation therapy: long-term
results. Ann Surg. 2004;240:711-717–718.
24. Weiser MR, Gollub MJ, Saltz LB. Assessment of clinical complete response after chemoradiation for rectal cancer with digital rectal examination, endoscopy, and
MRI. Ann Surg Oncol. 2015;22:3769-3771.
25. Fusco R, Petrillo M, Granata V, et al. Magnetic resonance imaging evaluation in neoadjuvant therapy of locally advanced rectal cancer: a systematic review. Radiol
Oncol. 2017;51:252-262.
26. Joye I, Deroose CM, Vandecaveye V, et al. The role of diffusion-weighted MRI and (18)F-FDG PET/CT in the prediction of pathologic complete response after
radiochemotherapy for rectal cancer: a systematic review. Radiother Oncol. 2014;113:158-165.
27. Wang X-H, Liu Z-J, Xu J-B, et al. Baseline and early 3D-CUBE volume 2 reconstruction of locally advanced rectal 3 cancer to predict tumor response after 4
neoadjuvant chemotherapy. J XRay Sci Technol. doi:10.3233/XST-190594
28. Bulens P, Couwenberg A, Intven M, et al. Predicting the tumor response to chemoradiotherapy for rectal cancer: model development and external validation using
MRI radiomics. Radiother Oncol. 2020;142:246-252.
29. George TJ Jr., Allegra CJ, Yothers G. Neoadjuvant rectal (NAR) score: a new surrogate endpoint in rectal cancer clinical trials. Curr Colorectal Cancer Rep. 2015;
11:275-280.
30. Valentini V, van Stiphout RG, Lammering G, et al. Nomograms for predicting local recurrence, distant metastases, and overall survival for patients with locally
advanced rectal cancer on the basis of European randomized clinical trials. J Clin Oncol. 2011;29:3163-3172.
31. MERCURY Study Group. Diagnostic accuracy of preoperative magnetic resonance imaging in predicting curative resection of rectal cancer: prospective
observational study. BMJ. 2006;333:779.
32. Joye I, Debucquoy A, Deroose CM, et al. Quantitative imaging outperforms molecular markers when predicting response to chemoradiotherapy for rectal cancer.
Radiother Oncol. 2017;124:104-109.
33. Lambregts DMJ, Maas M, Boellaard TN, et al. Long-term imaging characteristics of clinical complete responders during watch-and-wait for rectal cancer-an
evaluation of over 1500 MRIs. Eur Radiol. 2020;30:272-280.
34. Maxime J, van der Valk M, Hilling DE, et al; IWWD Consortium. Long-term outcomes of clinical complete responders after neoadjuvant treatment for rectal cancer
in the International Watch & Wait Database (IWWD): an international multicentre registry study. Lancet. 2018;391:2537-2545.
35. Breugom AJ, Swets M, Bosset JF, et al. Adjuvant chemotherapy after preoperative (chemo)radiotherapy and surgery for patients with rectal cancer: a systematic
review and meta-analysis of individual patient data. Lancet Oncol. 2015;16:200-207.
36. Bosset JF, Calais G, Mineur L, et al; EORTC Radiation Oncology Group. Fluorouracil-based adjuvant chemotherapy after preoperative chemoradiotherapy in
rectal cancer: long-term results of the EORTC 22921 randomised study. Lancet Oncol. 2014;15:184-190.
37. Glynne-Jones R, Grainger J, Harrison M, et al. Neoadjuvant chemotherapy prior to preoperative chemoradiation or radiation in rectal cancer: should we be more
cautious? Br J Cancer. 2006;94:363-371.
38. Garcia-Aguilar J, Chow OS, Smith DD, et al; Timing of Rectal Cancer Response to Chemoradiation Consortium. Effect of adding mFOLFOX6 after neoadjuvant
chemoradiation in locally advanced rectal cancer: a multicentre, phase 2 trial. Lancet Oncol. 2015;16:957-966.
39. Nilsson PJ, van Etten B, Hospers GA, et al. Short-course radiotherapy followed by neo-adjuvant chemotherapy in locally advanced rectal cancer—the RAPIDO
trial. BMC Cancer. 2013;13:279.
40. Bujko K, Wyrwicz L, Rutkowski A, et al; Polish Colorectal Study Group. Long-course oxaliplatin-based preoperative chemoradiation versus 5 5 Gy and
consolidation chemotherapy for cT4 or fixed cT3 rectal cancer: results of a randomized phase III study. Ann Oncol. 2016;27:834-842.
41. Ciseł B, Pietrzak L, Michalski W, et al. Long-course preoperative chemoradiation versus 5 5 Gy and consolidation chemotherapy for clinical T4 and fixed clinical
T3 rectal cancer: long-term results of the randomized Polish II study. Ann Oncol. 2019;30:1298-1303.
42. Hartley A, Ho KF, McConkey C, et al. Pathological complete response following pre-operative chemoradiotherapy in rectal cancer: analysis of phase II/III trials. Br
J Radiol. 2005;78:934-938.
43. Park IJ, You YN, Agarwal A, et al. Neoadjuvant treatment response as an early response indicator for patients with rectal cancer. J Clin Oncol. 2012;
30:1770-1776.
44. Cercek A, Goodman KA, Hajj C, et al. Neoadjuvant chemotherapy first, followed by chemoradiation and then surgery, in the management of locally advanced
rectal cancer. J Natl Compr Canc Netw. 2014;12:513-519.
45. Perez K, Safran H, Sikov W, et al. Complete neoadjuvant treatment for rectal cancer: the Brown University Oncology Group CONTRE Study. Am J Clin Oncol.
2017;40:283-287.

46. Fernandez-Martos C, Garcia-Albeniz X, Pericay C, et al. Chemoradiation, surgery and adjuvant chemotherapy versus induction chemotherapy followed by
chemoradiation and surgery: long-term results of the Spanish GCR-3 phase II randomized trial. Ann Oncol. 2015;26:1722-1728.
47. Dewdney A, Cunningham D, Tabernero J, et al. Multicenter randomized phase II clinical trial comparing neoadjuvant oxaliplatin, capecitabine, and preoperative
radiotherapy with or without cetuximab followed by total mesorectal excision in patients with high-risk rectal cancer (EXPERT-C). J Clin Oncol. 2012;
30:1620-1627.
48. Nogué M, Salud A, Vicente P, et al; AVACROSS Study Group. Addition of bevacizumab to XELOX induction therapy plus concomitant capecitabine-based
chemoradiotherapy in magnetic resonance imaging-defined poor-prognosis locally advanced rectal cancer: the AVACROSS study. Oncologist. 2011;
16:614-620.
49. Fernández-Martos C, Pericay C, Losa F, et al. Effect of aflibercept plus modified FOLFOX6 induction chemotherapy before standard chemoradiotherapy and
surgery in patients with high-risk rectal adenocarcinoma: the GEMCAD 1402 randomized clinical trial. JAMA Oncol. 2019;5:1566.
50. Kosinski L, Habr-Gama A, Ludwig K, et al. Shifting concepts in rectal cancer management: a review of contemporary primary rectal cancer treatment strategies.
CA Cancer J Clin. 2012;62:173-202.
51. Sanghera P, Wong DW, McConkey CC, et al. Chemoradiotherapy for rectal cancer: an updated analysis of factors affecting pathological response. Clin Oncol (R
Coll Radiol). 2008;20:176-183.
52. Habr-Gama A, Lynn PB, Jorge JM, et al. Impact of organ-preserving strategies on anorectal function in patients with distal rectal cancer following neoadjuvant
chemoradiation. Dis Colon Rectum. 2016;59:264-269.
53. Habr-Gama A, Gama-Rodrigues J, São Julião GP, et al. Local recurrence after complete clinical response and watch and wait in rectal cancer after neoadjuvant
chemoradiation: impact of salvage therapy on local disease control. Int J Radiat Oncol Biol Phys. 2014;88:822-828.
54. Smith FM, Rao C, Oliva Perez R, et al. Avoiding radical surgery improves early survival in elderly patients with rectal cancer, demonstrating complete clinical
response after neoadjuvant therapy: results of a decision-analytic model. Dis Colon Rectum. 2015;58:159-171.
55. Habr-Gama A, São Julião GP, Perez RO. Nonoperative management of rectal cancer: identifying the ideal patients. Hematol Oncol Clin North Am. 2015;
29:135-151.
56. Smith FM, Waldron D, Winter DC. Rectum-conserving surgery in the era of chemoradiotherapy. Br J Surg. 2010;97:1752-1764.
57. Habr-Gama A, Perez RO, Wynn G, et al. Complete clinical response after neoadjuvant chemoradiation therapy for distal rectal cancer: characterization of clinical
and endoscopic findings for standardization. Dis Colon Rectum. 2010;53:1692-1698.
58. Habr-Gama A, Perez RO, Nadalin W, et al. Operative versus nonoperative treatment for stage 0 distal rectal cancer following chemoradiation therapy: long-term
results. Ann Surg. 2004;240:711-717, NaN-718.
59. Brown G. Thin section MRI in multidisciplinary pre-operative decision making for patients with rectal cancer. Br J Radiol. 2005;78:S117-S127.
60. Habr-Gama A, São Julião GP, Fernandez LM, et al. Achieving a Complete Clinical Response After Neoadjuvant Chemoradiation That Does Not Require Surgical
Resection: It May Take Longer Than You Think!. Dis. Colon Rectum. 2019;62:802-808.
61. Perez RO, Habr-Gama A, Pereira GV, et al. Role of biopsies in patients with residual rectal cancer following neoadjuvant chemoradiation after downsizing: can
they rule out persisting cancer? Colorectal Dis. 2012;14:714-720.
62. Van den Begin R, Kleijnen J-P, Engels B, et al. Tumor volume regression during preoperative chemoradiotherapy for rectal cancer: a prospective observational
study with weekly MRI. Acta Oncol. 2018;57:723-727.
63. Perez RO, Habr-Gama A, São Julião GP, et al. Predicting complete response to neoadjuvant CRT for distal rectal cancer using sequential PET/CT imaging. Tech
Coloproctol. 2014;18:699-708.
64. Shihab OC, Taylor F, Salerno G, et al. MRI predictive factors for long-term outcomes of low rectal tumours. Ann Surg Oncol. 2011;18:3278-3284.
65. Appelt AL, Pløen J, Harling H, et al. High-dose chemoradiotherapy and watchful waiting for distal rectal cancer: a prospective observational study. Lancet Oncol.
2015;16:919-927.
66. Vaccaro CA, Yazyi FJ, Ojra Quintana G, et al. Locally advanced rectal cancer: preliminary results of rectal preservation after neoadjuvant chemoradiotherapy. Cir
Esp. 2016;94:274-279.
67. Araujo RO, Valadão M, Borges D, et al. Nonoperative management of rectal cancer after chemoradiation opposed to resection after complete clinical response. A
comparative study. Eur J Surg Oncol. 2015;41:1456-1463.
68. Dalton RS, Velineni R, Osborne ME, et al. A single-centre experience of chemoradiotherapy for rectal cancer: is there potential for nonoperative management?
Colorectal Dis. 2012;14:567-571.
69. Smith RK, Fry RD, Mahmoud NN, et al. Surveillance after neoadjuvant therapy in advanced rectal cancer with complete clinical response can have comparable
outcomes to total mesorectal excision. Int J Colorectal Dis. 2015;30:769-774.
70. Renehan AG, Malcomson L, Emsley R, et al. Watch-and-wait approach versus surgical resection after chemoradiotherapy for patients with rectal cancer (the
OnCoRe project): a propensity-score matched cohort analysis. Lancet Oncol. 2016;17:174-183.
71. Martens MH, Maas M, Heijnen LA, et al. Long-term outcome of an organ preservation program after neoadjuvant treatment for rectal cancer. J Natl Cancer Inst.
2016;108:djw171.
72. Dattani M, Heald RJ, Goussous G, et al. Oncological and survival outcomes in watch and wait patients with a clinical complete response after neoadjuvant
chemoradiotherapy for rectal cancer: a systematic review and pooled analysis. Ann Surg. 2018;268:955-967.

Sun et al
73. Chadi SA, Malcomson L, Ensor J, et al. Factors affecting local regrowth after watch and wait for patients with a clinical complete response following che-
moradiotherapy in rectal cancer (InterCoRe consortium): an individual participant data meta-analysis. Lancet Gastroenterol Hepatol. 2018;3:825-836.
74. São Julião GP, Karagkounis G, Fernandez LM, et al. Conditional survival in patients with rectal cancer and complete clinical response managed by watch and wait
after chemoradiation: recurrence risk over time. Ann Surg. 2019;1.
75. van der Valk MJM, Hilling DE, Bastiaannet E, et al; IWWD Consortium. Long-term outcomes of clinical complete responders after neoadjuvant treatment for rectal
cancer in the International Watch & Wait Database (IWWD): an international multicentre registry study. Lancet. 2018;391:2537-2545.
76. Habr-Gama A, Perez RO, Proscurshim I, et al. Interval between surgery and neoadjuvant chemoradiation therapy for distal rectal cancer: does delayed surgery
have an impact on outcome? Int J Radiat Oncol Biol Phys. 2008;71:1181-1188.
77. Smith JJ, Strombom P, Chow OS, et al. Assessment of a watch-and-wait strategy for rectal cancer in patients with a complete response after neoadjuvant therapy.
JAMA Oncol. 2019;5:e185896.
78. Habr-Gama A, São Julião GP, Gama-Rodrigues J, et al. Baseline T classification predicts early tumor regrowth after nonoperative management in distal rectal
cancer after extended neoadjuvant chemoradiation and initial complete clinical response. Dis Colon Rectum. 2017;60:586-594.
79. Habr-Gama A, Sabbaga J, Gama-Rodrigues J, et al. Watch and wait approach following extended neoadjuvant chemoradiation for distal rectal cancer: are we
getting closer to anal cancer management? Dis Colon Rectum. 2013;56:1109-1117.
80. Garcia-Aguilar J, Shi Q, Thomas CR Jr., et al. A phase II trial of neoadjuvant chemoradiation and local excision for T2N0 rectal cancer: preliminary results of the
ACOSOG Z6041 trial. Ann Surg Oncol. 2012;19:384-391.
81. Perez RO. Complete clinical response in rectal cancer: a turning tide. Lancet Oncol. 2016;17:125-126.

Redefining Colorectal Cancer by Tumor Biology

Mohamed E. Salem, MD1; Alberto Puccini, MD2; and Jeanne Tie, MD3,4
overview
Colorectal cancer treatment has undergone a paradigm shift. We no longer see this disease as a singular,
anatomic tumor type but rather a set of disease subgroups. Largely because of a better understanding of
cancer biology and the introduction and integration of molecular biomarkers—the premise of precision
therapy—we are beginning to direct treatments toward the right tumor target(s) in the right patients. The field
of molecular profiling is continually evolving, and new biomarkers are constantly being discovered that have
investigational, therapeutic, and/or prognostic implications—negative or positive. To date, only a few bio-
markers have sufficient actionable, clinical implication to earn international guideline-recommended routine
testing. Hence, it is vital that the treating oncologist should know which biomarkers to assess, when in the
treatment course to test for them, and how the test is to be done. Correct interpretation of profiling results is
imperative. Herein, we focus on international guideline-recommended mutation testing for patients prior to
their colorectal cancer treatment initiation. The clinical applications of circulating tumor DNA (ctDNA) in
patients with metastatic disease, based on our current knowledge and capabilities, are also addressed.
MOLECULAR TESTING IN THE CLINIC FOR METASTATIC combination with anti-EGFR therapy is shown to
COLORECTAL CANCER improve overall survival compared with doublet che-
Consensus guidelines for biomarker/molecular testing motherapy in combination with bevacizumab in the
in metastatic colorectal cancer have evolved over the first-line setting. Additionally, both European Society
past decade in line with the emergence of several for Medical Oncology (ESMO) and NCCN guidelines
effective targeted therapies—anti-EGFR–, BRAFV600E-, have recommended testing for BRAFV600E mutation at
HER2-, and NTRK-targeted treatments, as well as diagnosis, both for prognostication and for identifying
checkpoint inhibitors. However, the guideline- patients suitable for BRAFV600E-targeted treatment, such
recommended molecular testing rate remains low in as encorafenib, binimetinib, and cetuximab, in the
routine practice,1,2 with patients being more likely to second- and third-line settings. Because tumors with the
have guideline-aligned testing if they are treated at an BRAFV600E mutation are associated with poor prognosis
academic center, are diagnosed with de novo meta- compared with BRAFV600E wild-type (WT) tumors, the
static disease, are younger than 65, and are female.2 knowledge of this mutation status at diagnosis may
Because the field of molecular profiling is continually encourage oncologists to administer the more intensive
evolving, it is vital that the treating oncologist knows FOLFOXIRI (leucovorin, fluorouracil, oxaliplatin, and
what, when, and how to test and, importantly, how to irinotecan) regimen in the first-line setting. Routine
correctly interpret the profiling results to avoid over- testing of other components downstream of the EGFR
interpretation or misinterpretation of actionable vari- signaling pathway, such as PIK3CA and PTEN muta-
ants, which could lead to the treatment of patients with tions, is not currently recommended. For disease re-
ineffective but expensive therapies.3,4 fractory to standard-of-care treatment, testing for other
WHAT TO PROFILE? rare therapeutic targets such as HER2 amplification
Author affiliations
(via immunohistochemistry [IHC], fluorescence in situ
and support Table 1 describes the current National Comprehensive
information (if hybridization, or next-generation sequencing [NGS]
Cancer Network (NCCN; Version 1.2020) guideline-
applicable) appear testing) and fusion events in NTRK1–3, ALK, or ROS1
at the end of this
based recommendations for molecular testing in met-
should be considered. Of note, most colorectal cancer
article. astatic colorectal cancer. Testing for activating KRAS
with targetable kinase fusions is KRAS, NRAS, BRAF
Accepted on and NRAS mutations (codons 12 and 13 of exon 2; 59
WT and also mismatch repair (MMR) deficient, pro-
February 28, 2020 and 61 of exon 3; and 117 and 146 of exon 4; so-called
viding a rationale for the screening of oncogenic kinase
and published at expanded or extended RAS panel) is essential to select
ascopubs.org on fusions in this subset of colorectal cancer.5
patients who are likely to benefit from the anti-EGFR
March 24, 2020:
therapies cetuximab or panitumumab. This should be MMR testing is recommended for all patients with
10.1200/EDBK_ done at diagnosis, especially for patients with left-sided colorectal cancer, both for the detection of possible
279867 primary tumors for whom doublet chemotherapy in Lynch syndrome and for the selection of patients for

Salem, Puccini, and Tie
Tumor Tissue–Based Molecular Profiling: Single Gene or

NGS Testing?
• Testing for activating KRAS and NRAS muta- The field of molecular profiling is evolving at a tremendous
tions is essential to select patients who are likely speed, from a limited number of disease-specific “one
to benefit from anti-EGFR therapy of cetuximab biomarker–one drug” polymerase chain reaction (PCR)-
or panitumumab. based assays to broad commercial or academic multigene
• BRAFV600E mutation at diagnosis has both panel NGS assays covering several hundred cancer-related
prognostic and predictive value for identifying genes. Based on massive parallel sequencing of DNA with
patients suitable for BRAFV600E-targeted treat- subsequent sequence alignment, NGS is capable of ana-
ment, such as encorafenib, binimetinib, and lyzing multiple genetic alterations simultaneously, includ-
cetuximab. ing single nucleotide variants, small insertions/deletions
• DNA MMR testing is recommended for all pa- (indels), copy number variants, and complex fusions/rear-
tients with colorectal cancer, both for potential rangements, as well as providing an assessment of tumor
Lynch syndrome and as a predictive biomarker mutation burden. Although there are more than 20,000
for immunotherapy. genes in the human genome, only approximately 500 genes
• Testing for other rare therapeutic targets such are potentially related to cancer, and among those cancer
as HER2 amplification and fusion events in genes, very few mutations (2% across 7,000 tumors of a 28-
NTRK1–3, ALK, or ROS1 should be considered cancer type cohort) are known oncogenic events.10 Given
for patients with refractory disease. these findings, targeted gene panels, which sequence
a discrete number of loci or genes of interest at a greater
depth and lower cost than whole-exome sequencing, are
immunotherapy (pembrolizumab or nivolumab/ipilimu- the method of choice for most laboratories for clinical
mab). The MMR system is critical for repairing base sub- implementation.
stitution errors that arise during DNA replication. Loss of Current commercial and academic oncogenic panel testing
function of the MMR system (MMR deficiency) results in the can vary substantially in size from a small number of ac-
accumulation of base-pair mismatches across the genome, tionable genes (hotspot panels) to hundreds of genes (e.g.,
particularly in highly repetitive regions called microsatellites, FoundationOne, MSK-IMPACT, Tempus, Caris Molecular
resulting in genomic instability. Most MMR deficiency is Intelligence, Oncomine). Targeted panels use either a PCR
related to somatic alterations in the MMR pathway, whereas amplicon–based or a hybrid capture–based approach, both
the presence of a pathogenic germline mutation in one of of which have been shown to reliably detect single nucle-
the MMR genes is diagnostic of Lynch syndrome, ac- otide variants and indels for a variety of clinical applications.
counting for 3% of all colorectal cancer diagnoses. There It is important to note that some NGS assays also involve
are two methods to detect MMR deficiency in current testing of matched normal/germline samples, resulting in
clinical use: IHC for the four MMR proteins (MLH1, MSH2,
the possibility of detecting secondary pathogenic germline
PMS2, and MSH6) or DNA-based microsatellite instability
mutations. 11-13 In this setting, the American College of
(MSI) analysis using the Bethesda Panel.6 In MMR IHC
Medical Genetics and Genomics has recommended that 59
testing, loss or absence of staining of one or more MMR
genes should be actively evaluated for pathogenic muta-
proteins indicates deficient MMR. It is noteworthy that
tions and reported back to patients unless they decline to
missense mutations may lead to a nonfunctional but anti-
receive these germline findings. 14 ASCO has therefore
genically intact protein giving falsely intact results. Loss of
recommended pretest communication to inform patients of
MLH1 IHC should be followed by tumor testing for
the possibility of discovering pathogenic germline variants
BRAFV600E mutation; the presence of this mutation is con-
and the opportunity to opt-out from receiving germline
sistent with sporadic cancer. However, approximately 1% of
information.15
patients with colorectal cancer with absent MLH1 and
BRAFV600E mutations have Lynch syndrome, suggesting that With the increasing affordability of NGS, multigene tumor
germline screening should still be considered in patients somatic profiling is becoming more commonly adopted in
with a strong family history. More recently, it has been oncology. However, the extent to which incorporating NGS
shown that NGS panels can reliably determine the MSI into clinical care improves patient outcomes, such as
status of tumors.7,8 In fact, using the University of Wash- treatment response and disease-free survival, remains
ington OncoPlex targeted NGS sequencing panel, Hampel unclear. Studies have shown that only a subset (, 10%) of
et al found that tumor sequencing alone had better sen- patients who have undergone tumor sequencing are en-
sitivity and equal specificity to IHC plus BRAF and MSI plus rolled in genotype-relevant clinical trials.16-18 Numer-
BRAF testing.9 ous “basket” studies, such as NCI-MATCH, ASCO TAPUR

Redefining Colorectal Cancer by Tumor Biology
TABLE 1. NCCN Guideline–Based Recommendations for Molecular Testing in Metastatic Colorectal Cancer
Biomarker Type of Alteration When to Test Testing Method Recommendation
KRAS/ Point mutation* At diagnosis Individual gene or Patients with any known KRAS or NRAS mutations should
NRAS NGS not be treated with either cetuximab or panitumumab
BRAF V600E point mutation At diagnosis Individual gene or BRAFV600E mutation makes response to cetuximab or
NGS panitumumab highly unlikely unless given with a BRAF
inhibitor
MMR/MSI MMR: protein expression; MSI: At diagnosis IHC (MMR), PCR Screening for Lynch syndrome; patients with deficient
changes in short repetitive DNA (MSI) or NGS MMR or MSI-high tumors should be considered for
sequences (MSI) immunotherapy
HER2 Overexpression or amplification At diagnosis or after IHC, FISH, or NGS HER2 testing is not recommended for those with known
failure of at least one RAS or BRAF mutation; consider anti-HER2 therapy in
line of therapy patients with HER2-amplified tumors
NTRK Fusions After failure of at least IHC, FISH, NGS Limit NTRK testing to patients with wild-type KRAS/NRAS/
one line of therapy (DNA or RNA BRAF and dMMR/MSI-H tumors; TRK inhibitors
based)** (larotrectinib or entrectinib) are treatment options for
patients with NTRK gene-fusion positive tumors
Abbreviations: NGS, next-generation sequencing; MMR, mismatch repair; MSI, microsatellite instability; IHC, immunohistochemistry; PCR, polymerase chain
reaction; FISH, fluorescence in situ hybridization; dMMR, mismatch repair deficiency; MSI-H, microsatellite instability high.
*Codons 12 and 13 of exon 2; 59 and 61 of exon 3; and 117 and 146 of exon 4.
**Tumors that test positive by IHC should be confirmed by RNA-based NGS.
(Targeted Agent and Profiling Utilization Registry), and Health care stakeholders are concerned about the added
EORTC-SPECTA, are enrolling patients with potentially ac- “financial toxicity” of using multigene panel NGS testing
tionable genomic alterations in their tumors, irrespective of (currently costing $4,000 to $5,000) in cancer diagnostics
disease type, and matching them to U.S. Food and Drug rather than more affordable single-gene testing, which can
Administration (FDA)–approved targeted therapies or in- be performed for a few hundred dollars each. Interestingly,
vestigational agents. Results for patients with colorectal a recent study assessing the economic impact of using NGS
cancer from three study cohorts of the TAPUR study were versus single-gene testing strategies among patients with
recently presented at the 2020 Gastrointestinal Cancers metastatic non–small cell lung cancer suggested that
Symposium (abstracts 122,19 132,20and 13321). Objective upfront NGS testing was associated with substantial cost
response and disease control rates were as follows: 29% savings compared with sequential testing and hotspot
and 57% for patients with a BRAFV600E mutation treated with panels.23 Although NCCN guidelines currently do not pro-
cobimetinib and vemurafenib; 14% and 50% for patients vide specific recommendations on the type of testing (single
with HER2 overexpression or amplification treated with gene vs. NGS panels) in metastatic colorectal cancer, cli-
pertuzumab and trastuzumab; and 28% and 4% for pa- nicians should consider the clinical context (treatment-
tients with microsatellite-stable disease with high tumor naı̈ve vs. treatment-refractory), the amount of available
mutation burden (defined as at least nine mutations/mega- tumor tissue (resection vs. limited biopsy), and the in-
base by FoundationOne test) treated with pembrolizumab. formation they are aiming to obtain when choosing the
Intriguingly, the only “precision medicine” randomized trial appropriate testing strategy.
to date (SHIVA trial) has failed to demonstrate an im-
provement in progression-free survival (PFS) when using Which Tissue to Test?
NGS to match patients to off-label targeted treatments
Comparative genomic analysis of primary and metastatic
compared with treatment by physician’s choice.22
colorectal tumor tissue have demonstrated a very high
Although NGS testing increases our ability to identify rare concordance for driver-gene mutations (e.g., KRAS, NRAS,
targetable alterations (fusion events in NTRK1–3, ALK, BRAF mutations) and MSI status for treatment-naı̈ve
ROS1, RET, or POLE mutations) and may facilitate enroll- patients.24-29 Therefore, patient tumor testing for clinically
ment in genomically matched clinical trials, the current relevant genes can be performed on either primary or
number of actionable driver events in metastatic colorectal metastatic tumor tissue, with the exception of patients with
cancer remains modest (KRAS, NRAS, BRAF, HER2, or a history of multiple primary colorectal cancers, in which
MMR genes) and can be adequately and inexpensively case a metastatic site biopsy is recommended. However,
assessed using PCR or immunohistochemical assays. clonal evolution and phenotypic plasticity of cancer cells in

response to treatment, especially under the selection pres- adjuvant treatment monitoring in early-stage disease to
sure of targeted therapy such as anti-EGFR inhibition, can genotyping for therapy selection and tracking resistant
affect the molecular landscape and result in lesion-specific mechanisms in metastatic disease. This section will focus
genomic heterogeneity.30-32 In this context, a single tissue on the clinical applications of ctDNA in metastatic co-
biopsy would be inadequate to capture the molecular lorectal cancer.
landscape of all metastatic lesions.
ctDNA RAS and BRAF Profiling
THE ROLE OF ctDNA
Access to sufficient patient tumor tissue for genomic pro-
ctDNA fragments are released by cancer cells into the filing is usually not an issue in the case of colorectal cancer.
bloodstream and, in theory, should contain genetic and However, plasma genotyping may play a role when a quick
epigenetic alterations identical to the cancer cells they turnaround time is needed, for example, at the time of
originated from. Although the phenomenon of cell-free DNA metastatic colorectal cancer diagnosis, because ctDNA
(cfDNA) was first described in 1948,33 the clinical potential profiling takes a few days, whereas solid tumor sequencing
of cfDNA as a cancer marker was only recognized in 1994 typically takes 2 to 4 weeks or even longer. Many studies
when mutated RAS was detected in circulating DNA frag- have shown a high concordance (approximately 90%)
ments in patients with pancreatic and hematologic cancer.34,35 between RAS status in matched cfDNA and RAS status in
ctDNA represent only a small fraction of the total cfDNA, but tumor tissue samples, with sensitivities and specificities in
this fraction is highly variable, ranging from less than 0.1% the range of 70% to 96% and 83% to 98%, respectively
to greater than 10% depending on tumor stage, disease (Table 2).38,45-50 Several studies have examined the reasons
burden, biologic shedding, and anatomic factors such as for the suboptimal sensitivity of ctDNA testing and found
disease site.36-38 that the failure to detect mutated RAS in the plasma is
Broadly speaking, there are three approaches to ctDNA largely caused by the lack of ctDNA being released by the
analysis for clinical application in the metastatic setting. For tumor, which in turn is associated with low tumor burden;
detection of individual or a small number of point mutations peritoneal, nodal, and lung metastases; and mucinous
(e.g., RAS or BRAF mutations), assays based on PCR histology.38,47,48,50 The high specificity of ctDNA RAS testing
analysis such as BEAMing (beads, emulsion, amplification, means that the detection of RAS mutations in plasma can
and magnetics) or droplet digital PCR (ddPCR) analysis can reliably predict RAS mutation status in the tumor tissue. In
detect mutations with mutant allele fraction (MAF) as low as the study with the lowest cfDNA test specificity (high rate of
0.01%. An advantage of this approach is the fast turnaround tissue WT/plasma mutant cases), analysis of tumor tissue
time (a few hours to a few days) and relatively low cost; with more sensitive assays (droplet digital PCR or deep
hence, it is most suited to the identification of actionable sequencing) identified the same RAS mutations found in
targets for therapy. The second approach is with a targeted plasma by BEAMing in all cases.50 Several of the RAS
NGS panel (e.g., IonTorrent, Guardant360), which provides
a higher genomic coverage with the ability to detect fusions
and copy number alterations. These assays are, however, TABLE 2. Summary of Studies Assessing the Concordance of RAS Mutation
less sensitive (limit of detection approximately 0.1%) and Between Tumor Tissue and Cell-Free DNA in Anti-EGFR–Naı̈ve Disease
more expensive and have a slower turnaround time (1 to 2 ctDNA Concordance Sensitivity Specificity
Author Assay N (%) (%) (%)
weeks). The final approach is with whole-exome sequencing
or whole-genome sequencing without prior knowledge of Bachet et al38 Targeted 412 85 76 98
NGS
the genomic alterations in the tumor tissue, allowing for
identification of novel changes that develop during treat- Vidal et al47 OncoBEAM 115 93 96 90
ment. The breadth of genomic coverage offered by whole- Garcı́a- OncoBEAM 236 92 90 94
exome sequencing or whole-genome sequencing comes at Foncillas
et al49
the expense of the depth of coverage, with sensitivity av-
eraging 1% to 5%. For detection of minimal residual disease Grasseli OncoBEAM 146 90 89 90
et al48
after curative resection in early-stage cancer, tumor-informed
(where tumor-specific mutations are first identified) targeted Thierry et al45 IntPlex 95 96 92 98
sequencing of a limited gene panel offers the highest Schmiegel OncoBEAM 98 92 92 93
sensitivity, especially if modified with molecular barcoding et al46
methods and error-suppression algorithms (SafeSeqS, Normanno OncoBEAM 92 78 70 83
CAPP-Seq).39-44 et al50
ctDNA holds many promises in the clinics: from early Abbreviations: ctDNA, circulating tumor DNA; NGS, next-generation sequencing;
cancer detection, minimal residual disease detection, and BEAM, beads, emulsion, amplification, and magnetics.

concordance studies also examined BRAFV600E mutation RAS mutations was 9.5%, whereas plasma cell-free DNA
status. Overall, the concordance for ctDNA BRAFV600E revealed a much higher mutant RAS emergence rate of
mutation appears very high (97% to 100%).31,45,51 36.7%, suggesting that ctDNA analysis provides a more
Despite the analytical challenges of ctDNA testing, it is comprehensive assessment of tumor genomic landscape
reassuring that several retrospective studies have observed than tissue testing.54
similar outcomes in patients with metastatic colorectal Using serial ctDNA tracking, two landmark studies in-
cancer who were treated with anti-EGFR–based regimens dependently revealed KRAS alterations as secondary
based on RAS status that was determined by tissue or mechanisms of resistance to anti-EGFR therapy. 55,56
plasma testing.47,48,50 As ctDNA assays become increasingly Importantly, the emerging mutations preceded radiologic
sensitive, ctDNA analysis can identify patients with a very progression by several months. Several studies have sub-
low number of RAS-mutant cells who may still benefit from sequently described other molecular mechanisms of re-
anti-EGFR therapy. Indeed, Grasselli et al demonstrated sistance predominantly in the mitogen-activated protein kinase
worsening PFS with increasing MAF for patients treated with (MAPK) pathway including NRAS, BRAF, MEK, and EGFR
second- or third-line anti-EGFR antibody plus irinotecan. ectodomain mutations and MET and HER2 amplification.37,57-60
Intriguingly, there appears to be no difference in PFS be- Emergence and co-occurrence of multiple alterations in the
tween patients with plasma RAS WT and those with low MAF MAPK pathway is also seen in patients with BRAFV600E
of less than 0.1, suggesting that some patients with low-level metastatic colorectal cancer treated with a combination of
mutations may still benefit from RAS-targeted treatment. BRAF/MEK/EGFR inhibitors.53,61-64 Targeting these hetero-
Future studies should prospectively validate a threshold of geneous secondary resistance alterations poses a clinical
detectable mutation rate that confers intrinsic resistance to challenge because most emerging mutations are not thera-
EGFR inhibition. Until then, tumor tissue profiling of RAS peutically actionable. A preliminary attempt to target the
status should remain the gold standard for selecting patients emerging EGFR ectodomain mutations by second-generation
for anti-EGFR therapy, unless access to tumor tissue is not anti-EGFR (Sym004) has been disappointing.65
possible.
Of particular therapeutic relevance is the recent observation
Assessment of Treatment Response that temporal withdrawal of an anti-EGFR antibody resulted
The short half-life of ctDNA allows for early assessment of in a decline in RAS-resistant clones, opening up a window
tumor burden during treatment; hence, changes can be of opportunity for anti-EGFR rechallenge.31,66,67 To highlight
observed more or less in real time and certainly before the use of this ctDNA-guided rechallenge strategy, ctDNA
changes can be appreciated on CT imaging or by serum analysis of the single-arm phase II CRICKET (rechallenge for
protein marker such as carcinoembryonic antigen (CEA). patients with RAS- and BRAF-WT metastatic colorectal
A study of 53 patients with metastatic colorectal cancer cancer with acquired resistance to first-line cetuximab and
receiving first-line chemotherapy found that a 10-fold re- irinotecan) study demonstrated that none of the six patients
duction in ctDNA MAF 2 weeks after starting chemotherapy who responded to anti-EGFR rechallenge had a RAS mu-
correlated with a radiologic response at 8 to 10 weeks.36 The tation in their circulation, translating into a significantly
PLACOL study, which included patients treated with first-, longer PFS in patients with RAS-WT ctDNA than those with
second-, and third-line treatment have also confirmed that RAS-mutated ctDNA at the time of rechallenge (median
reduction in ctDNA MAF predicted for treatment re- PFS, 4.0 vs. 1.9 months; p = .03).68 To further validate this
sponse.52 A similar pattern is seen in patients with strategy, the ongoing ctDNA-guided single-arm phase II
BRAFV600E metastatic colorectal cancer treated with BRAF- CHRONOS trial (Rechallenge With Panitumumab Driven by
targeted combination therapy.53 The clinical use of ctDNA RAS Dynamic of Resistance; NCT03227926) aims to assess
to provide an earlier assessment of treatment response the efficacy of panitumumab rechallenge when circulating
compared with traditional imaging remains to be validated in RAS mutation load decreases by greater than 50% com-
prospective studies. pared with the mutation load at the time of progression on
first-line anti-EGFR therapy.
Tracking Resistance Mutations and Timing of
Anti-EGFR Rechallenge MOVING AWAY FROM A ONE-SIZE-FITS-ALL APPROACH
Serial ctDNA measurements allow longitudinal tracking of As mentioned earlier, molecular testing is crucial to per-
the genomic evolution and identify resistance mechanisms sonalize the optimal treatment of metastatic colorectal
to molecularly targeted therapies. Analysis of archival pre- cancer and has become standard practice for management
and posttreatment tumor biopsy, as well as plasma samples of these patients.69 Because of advances in the availability of
for emergence of mutations during/posttreatment from several targeted agents over the last 2 decades, international
a phase II study of panitumumab in combination with iri- guidelines recommend molecular testing of those biomarkers
notecan, showed that the rate of emergence of tumor tissue that provide clinically actionable information in terms of both

prognosis and treatment decisions (Table 1). Novel targeted However, not all patients with RAS-WT tumors will respond
agents including VEGF, EGFR, and, more recently, immune to treatment. In fact, it is conceivable that only 50% to 65%
checkpoints and BRAF inhibitors have become available for of these patients have EGFR-dependent cancer.
the treatment of metastatic colorectal cancer, adding to Subsequent studies described other mutations in genes of
standard chemotherapy with 5-fluorouracil, oxaliplatin, and the EGFR signaling pathways involving other exons of KRAS
irinotecan.70 and NRAS, BRAF, and PTEN that may affect response of
colorectal cancer to anti-EGFR antibody therapies.78,79
Extended-RAS Panel
Several additional mechanisms of primary resistance to anti-
The EGFR signaling pathway plays a critical role in co- EGFRs have been identified in RAS-WT metastatic co-
lorectal cancer development, and EGFR inhibitors are well lorectal cancer thus far, based on preclinical data and
established therapeutic agents in metastatic colorectal retrospective evaluations. However, the routine use of these
cancer treatment. A subgroup of metastatic colorectal biomarkers in clinical practice is not recommended at
cancers is highly dependent on EGFR signaling, and the use present, and further prospective validation is warranted.
of EGFR inhibitors has been demonstrated effective in early These include HER2 amplification, PIK3CA mutations
lines of treatment, as well as in patients with colorectal cancer (exons 9 and 20), MET amplification, and FGFR1 and
that was heavily pretreated chemorefractory.71 Currently, two PDGFRA mutations.80
anti-EGFR drugs are available in clinical practice with similar
indications: panitumumab and cetuximab. More recently, the PRESSING panel of multiple combined
genomic alterations comprising activating mutations of the
Panitumumab is currently approved by the FDA and Eu- MAPKs or the PI3K/AKT axis, NTRK/ROS1/ALK/RET rear-
ropean Medicines Agency for the treatment of RAS-WT rangements, HER2 amplification or mutations, and MET
metastatic colorectal cancer in combination with FOLFOX amplification was shown to predict primary resistance to
(5-fluorouracil, leucovorin, and oxaliplatin) or FOLFIRI (5- anti-EGFRs in RAS/BRAF-WT metastatic colorectal cancer.81
fluorouracil, leucovorin, and irinotecan) in the first-line
setting; in combination with FOLFIRI in the second-line MSI or DNA MMR Deficiency
setting; and as monotherapy following disease progres- MSI status or DNA MMR deficiency (dMMR) is one of the
sion after prior chemotherapy treatment (fluoropyrimidine-, few biomarkers that have been approved for clinical use in
oxaliplatin-, and irinotecan-containing regimens).72,73 colorectal cancer by regulatory authorities.69
Cetuximab is currently approved by the FDA and European Notably, the likelihood of MSI/dMMR in colorectal cancer
Medicines Agency for the treatment of RAS-WT metastatic varies according to the stage of the disease, with a higher
colorectal cancer in combination with FOLFIRI or FOLFOX incidence in the early stages (approximately 20% in stages I
for first-line treatment; in combination with irinotecan in and II and 12% in stage III) and a lower incidence in the
patients with disease that is refractory to irinotecan-based metastatic setting (4% to 5%).82,83 Most MSI/dMMR cases
chemotherapy; and as a single agent in patients who have are defined as sporadic because of the epigenetic silencing
previously received oxaliplatin- and irinotecan-based che- of the MLH1 gene promoter by hypermethylation,84 whereas
motherapy or whose disease is intolerant to irinotecan.74,75 approximately 20% are affected by Lynch syndrome.85 This
Anti-EGFR monoclonal antibodies require knowledge of the estimate suggests that the prevalence of Lynch syndrome
mutational status of genes in the EGFR pathway as pre- among all patients with colorectal cancer is roughly 3%.
dictive biomarkers. Initial clinical trial data demonstrated Lynch syndrome is the most common cause of hereditary
that patients with colorectal cancer carrying activating colorectal cancer and is an autosomal-dominant hereditary
mutations of KRAS affecting exon 2 codons 12 and 13 did disease caused by germline mutations in the DNA MMR
not benefit from anti-EGFR monoclonal antibody therapy.76,77 genes (MLH1, MSH2, MSH6, PMS2) or EPCAM, leading to
Subsequently, less frequent activating mutations in KRAS epigenetic inactivation of MSH2.86,87 Lynch syndrome is
exons 3 and 4 and NRAS exons 2, 3, and 4, which are associated with an increased risk of extracolonic cancers of
present in approximately 15% to 20% of KRAS exon 2 WT the uterus, ovary, stomach, small intestine, and kidney.85
tumors, have been identified as other biomarkers of intrinsic Patients with Lynch syndrome–associated colorectal cancer
cancer cell resistance to cetuximab or panitumumab, as it have a better prognosis compared with patients with spo-
was for the first time found in the randomized phase III study radic colorectal cancer.88 They require personalized treat-
of FOLFOX plus panitumumab versus FOLFOX alone.73 ment and benefit from enrollment in targeted surveillance
Regulatory agencies have also provided recommendations protocols for metachronous primary colorectal cancers and
on validated laboratory techniques and accreditation criteria other Lynch syndrome–associated cancers.89 Therefore,
for RAS mutation testing, which should be performed only in universal screening of Lynch syndrome through the iden-
highly qualified and certified laboratories. tification of MMR deficiency in the tumor tissue of all new

colorectal cancer cases has been proposed as an alternative oxaliplatin, and irinotecan, based on data from CheckMate
diagnostic approach90 and led to an increase of MSI/dMMR 142 (NCT02060188).
detection. In addition to nivolumab, the CheckMate 142 study eval-
MSI/dMMR status has both prognostic and predictive values uated the combination of nivolumab and ipilimumab in
in patients with colorectal cancer. It is widely recognized that patients with MSI/dMMR metastatic colorectal cancer after
MSI/dMMR is associated with both a good prognosis (i.e., a treatment with fluoropyrimidine, oxaliplatin, and irinotecan,98
substantially lower risk for recurrence) and a lack of benefit showing an investigator-assessed ORR of 55% and PFS at 9
from fluorouracil treatment in the adjuvant setting, although and 12 months of 76% and 71%, respectively. Based on
some controversial results have been shown.91,92 Con- these data, in July 2018, the FDA approved the use of
versely, in the metastatic setting, the prognostic value of combined nivolumab plus ipilimumab in the treatment of
MSI/dMMR status has not yet been clarified. Some evidence patients with dMMR metastatic colorectal cancer who are
suggests that the frequent association of BRAF mutation progressing on standard treatments.
with MSI/dMMR colorectal cancer may act as a confounding The CheckMate 142 study subsequently added another
factor.93 However, a recent study showed that patients with cohort evaluating nivolumab as frontline therapy: ORR was
MSI/dMMR and carrying a BRAFV600E mutation do not have 60%, and PFS and overall survival at 12 months were 77%
worse survival than that of patients with BRAF-WT colorectal and 83%, respectively.99
cancers.94
However, currently in Europe, not a single PD-1–blocking
The predictive value of MSI/dMMR status, not only in pa-
antibody has been approved for the treatment of patients
tients with colorectal cancer but also as a pan-cancer
with MSI-H/dMMR cancers.100
biomarker,95 has recently emerged following the un-
precedented results of immune checkpoint inhibitors in Immunotherapy for MSI/dMMR metastatic colorectal can-
MSI/dMMR metastatic colorectal cancer, which have opened cer has made considerable progress. Several studies are
a new era in the treatment of these tumors. ongoing in different settings (e.g., neoadjuvant), and dif-
ferent combination approaches are being investigated to
In fact, in 2017, pembrolizumab—an anti–PD-1 anti-
overcome resistance to immunotherapy and eventually
body—was FDA approved for the treatment of metastatic
improve the overall survival of these patients.
cancers that are characterized by MSI-H status or dMMR,
regardless of the site of tumor origin,96 based on a pooled BRAF Mutations
analysis of five trials: KEYNOTE-016 (NCT01876511), The RAS/RAF/MEK/ERK signaling cascade (MAPK path-
KEYNOTE-164 (NCT02460198), KEYNOTE-158 (NCT02628067), way) participates in cell proliferation, differentiation, sur-
KEYNOTE-012 (NCT01848834), and KEYNOTE-028 vival, and apoptosis.101 BRAF mutations are detected in
(NCT02054806). These five clinical trials collectively many cancer types, such as melanoma, colorectal cancer,
had 149 patients with dMMR cancers. Ninety patients thyroid cancer, non–small cell lung cancer, and hairy cell
had metastatic colorectal cancer, whereas 59 patients leukemia.102 Mutations in BRAF are divided into three
had non-colorectal malignancies. Objective response classes; the most common mutation is BRAFV600E, whereas
rate (ORR) was 39.6% (95% CI, 31.7–47.9) with 11 other BRAF mutations (grouped together as atypical or non-
and 48 complete responses and partial responses, re- V600E mutations) occur in approximately 2% to 4% of
spectively. These responses did not differ between co- patients with metastatic colorectal cancer. The first and
lorectal cancer (36%) and non-colorectal cancer (46%) second classes of mutations result in a protein that signals
malignancies.97 as RAS-independent active monomers (class I, comprising
Currently, KEYNOTE-177 (NCT02563002) is an ongoing the V600E mutation) or dimers (class II). The third class
international, randomized, open-label, phase III study that comprises RAS-dependent BRAF mutations, which result in
will evaluate the safety and efficacy of pembrolizumab as a kinase-impaired or kinase-dead protein.103 Recently,
first-line therapy for MSI/dMMR metastatic colorectal can- a study showed for the first time that different clinical and
cer versus the standard-of-care chemotherapy (mFOL- pathologic features and outcome data exist according to the
FOX6, FOLFIRI, or in combination with bevacizumab or three BRAF mutation classes in metastatic colorectal
cetuximab). cancer.104 This finding may call for the development of
Other immune checkpoint inhibitors proved to be effica- personalized treatment of these patients.
cious alone or in combination as therapy for patients with BRAFV600E-mutated colorectal cancer has very distinct be-
metastatic colorectal cancer. In fact, in 2017, the FDA havior both biologically and clinically. BRAFV600E mutation is
granted accelerated approval to nivolumab for the treatment associated with older age at diagnosis, female sex, right-
of patients with MSI/dMMR metastatic colorectal cancer sided tumor location, poor differentiation, mucinous his-
that has progressed following treatment with fluoropyrimidine, tology, and MSI.105 In the metastatic setting, mutated

BRAFV600E is found in approximately 7% to 10% of cases, or FOLFIRI plus cetuximab in second- or third-line treatment
and it is associated with a poor prognosis and very little of patients with BRAFV600E-mutated metastatic colorectal
overall benefit from standard therapeutic approaches.106 cancer.116 It showed that the triplet dramatically improved
median overall survival (9.0 vs. 5.4 months; HR, 0.52; 95%
The median overall survival of patients with BRAFV600E-
CI, 0.39–0.70) and ORR (26% vs. 2%) compared with the
mutated colorectal cancer is approximately 12 months,
control group. These results suggest that this targeted
compared with 30 to 35 months in patients with BRAF-WT
therapy regimen should be a new standard of care for this
tumors. Accumulating data suggest that BRAF-mutated
patient population.
tumors are resistant to anti-EGFR treatments (i.e., pan-
itumumab or cetuximab).107 Therefore, guidelines have However, ESMO guidelines do not recognize any of these
been updated to limit anti-EGFR inhibitors alone or in combinations as standards of care to date, creating a huge
combination with cytotoxic chemotherapy to patients with disparity in treatment options between nations (United
RAS- and BRAF-WT tumors. States vs. Europe and Asia).
Regarding therapy for patients who have BRAFV600E-mu-
tated metastatic colorectal cancer, attempts to directly in- HER2 Amplification
hibit the active BRAF protein have failed, which suggests Amplification and/or overexpression of HER2 is detected in
a complex carcinogenic process in this disease108—more approximately 3% of patients with colorectal cancer. The
complex, in fact, than in other BRAFV600E-mutated cancers, prognostic value of HER amplification is still a matter of de-
such as melanoma, for which responses have been ob- bate.117 However, HER2 amplification may also serve as
served using BRAF inhibitor monotherapy. Three phase I/II a negative biomarker for EGFR-targeted treatments in colorectal
trials using vemurafenib,109 dabrafenib,110 or encorafenib111 cancer118: results from both preclinical and clinical studies
as monotherapy had disappointing results, with response showed that HER2 amplification hampers the efficacy of anti-
rates ranging from 5% to 16%. EGFR–targeted therapies. This is crucial in clinical practice
Because of the rapid disease progression, resistance to because it could potentially impact the physician’s ability to
standard treatments, and general poor patient prognosis optimize therapeutic sequencing. However, these data should
related to BRAFV600E mutations, more aggressive cytotoxic be interpreted with caution as they are retrospective and must
combinations may be preferred in the first-line treatment of be validated in prospective clinical studies.119
patients with BRAFV600E colorectal cancer. In fact, FOL- HER2 has been investigated as a therapeutic target in
FOXIRI plus bevacizumab is considered a potential first-line metastatic colorectal cancer. HERACLES-A was a proof-of-
treatment option for patients with BRAFV600E-mutated dis- concept phase II trial of dual HER2-targeted therapy
ease, ever since the Italian TRIBE trial demonstrated an (trastuzumab plus lapatinib) in patients with colorectal
increased survival trend for patients treated with FOLFOXIRI cancer refractory to standard-of-care treatments, including
plus bevacizumab compared with FOLFIRI plus bev- cetuximab or panitumumab: complete responses were
acizumab (HR, 0.55; 95% CI, 0.24–1.23).112 However, the observed in two patients (6%) and partial responses in eight
specificity of the recommendation for FOLFOXIRI specifi- patients (24%), giving an ORR of 30%.120 Similar results
cally for patients with BRAFV600E has been called into were reported in the MyPathway trial,121 which investigated
question with the results of the TRIBE2 study, 113 which pertuzumab-trastuzumab combination therapy in 37 pa-
failed to replicate the finding. tients with HER2-amplified colorectal cancer, showing an
In the era of precision oncology, it is logical to imagine that ORR of 38%.
BRAF-targeted therapy would be the best treatment option More recently, encouraging results from three other clinical
for these patients. Accordingly, subsequent to the suc- trials investigating dual HER2 blockade were presented at
cessful testing of different BRAF inhibitor combinations the 2019 ESMO Congress. In the phase II TRIUMPH study,
with other targeted agents with or without chemotherapy, 19 patients with RAS-WT and HER2-amplified metastatic
NCCN guidelines now specify that vemurafenib, cetux- colorectal cancer were treated with trastuzumab plus per-
imab, and irinotecan 114 ; dabrafenib, trametinib, and tuzumab with an ORR of 35%.122 In the phase II MOUN-
panitumumab 53 ; and encorafenib, binimetinib, and TAINEER trial (NCT03043313), 26 patients with the same
cetuximab115 are standard treatment options for patients characteristics were enrolled and treated with the newly
with BRAFV600E-mutated metastatic colorectal cancer. The developed tucatinib, a potent, highly selective, oral tyrosine
latter combination was investigated in the phase III BEACON kinase inhibitor of the HER2 receptor, in combination with
study (NCT02928224), which compared the efficacy of trastuzumab, showing an ORR of 52%.123 Finally, the open-
encorafenib (anti-BRAF) and cetuximab (anti-EGFR) with or label phase II HERACLES-B study evaluated a targeted
without binimetinib (anti-MEK) versus a control group of therapy approach with a combination of pertuzumab and
patients treated with the investigator’s choice of irinotecan T-DM1 in patients with RAS/BRAF-WT, HER2-amplified

refractory metastatic colorectal cancer: ORR was 10% of 30 in the case that disease is refractory to standard-of-care
patients enrolled.124 treatment. Additionally, MMR testing is recommended,
CONCLUSION which could diagnose Lynch syndrome and aid in the se-
lection of patients for immunotherapy. ctDNA assays are
In conclusion, molecular testing for patients with colorectal
being fine-tuned and may soon be included in standard
cancer has evolved over the last decade or so, and in-
testing of patients before and during treatment to weed out
ternational guidelines recommend testing for activating
KRAS and NRAS mutations (for anti-EGFR therapy guid- nonresponders as soon as possible and provide those in-
ance); a BRAFV600E mutation (for prognostic purpose and dividuals with alternative treatments. Suffice it to say, the
encorafenib and binimetinib therapy guidance); and HER2 field of molecular profiling is continually evolving, and every
amplification and fusion events in NTRK1–3, ALK, or ROS1 treating oncologist must stay up to date with developments.

1
Department of Medical Oncology, Levine Cancer Institute, Charlotte, NC Mohamed E. Salem, MD, Department of Medical Oncology, Levine Cancer
2
University of Genoa, Ospedale Policlinico San Martino IRCCS, Genoa, Institute, Carolinas HealthCare System, 1021 Morehead Medical Drive,
Italy Charlotte, NC 28204; email: [email protected].
3
Department of Medical Oncology, Peter MacCallum Cancer Centre,
Melbourne, Victoria, Australia
4
Division of Personalized Oncology, The Walter and Eliza Hall Institute of
Medical Research, Melbourne, Victoria, Australia
EDBK_279867.
REFERENCES
1. Charlton ME, Kahl AR, Greenbaum AA, et al. KRAS testing, tumor location, and survival in patients with stage IV colorectal cancer: SEER 2010-2013. J Natl
Compr Canc Netw. 2017;15:1484-1493.
2. Gutierrez ME, Price KS, Lanman RB, et al. Genomic profiling for KRAS, NRAS, BRAF, microsatellite instability, and mismatch repair deficiency among patients
with metastatic colon cancer. Precision Oncol. 2019;3:1-9.
3. Schram AM, Reales D, Galle J, et al. Oncologist use and perception of large panel next-generation tumor sequencing. Ann Oncol. 2017;28:2298-2304.
4. Brusco LL, Wathoo C, Mills Shaw KR, et al. Physician interpretation of genomic test results and treatment selection. Cancer. 2018;124:966-972.
5. Cocco E, Benhamida J, Middha S, et al. Colorectal carcinomas containing hypermethylated MLH1 promoter and wild-type BRAF/KRAS are enriched for
targetable kinase fusions. Cancer Res. 2019;79:1047-1053.
6. Rubenstein JH, Enns R, Heidelbaugh J, et al; Clinical Guidelines Committee. American Gastroenterological Association Institute guideline on the diagnosis and
management of Lynch syndrome. Gastroenterology. 2015;149:777-782.
7. Stadler ZK, Battaglin F, Middha S, et al. Reliable detection of mismatch repair deficiency in colorectal cancers using mutational load in next-generation
sequencing panels. J Clin Oncol. 2016;34:2141-2147.
8. Salipante SJ, Scroggins SM, Hampel HL, et al. Microsatellite instability detection by next generation sequencing. Clin Chem. 2014;60:1192-1199.
9. Hampel H, Pearlman R, Beightol M, et al; Ohio Colorectal Cancer Prevention Initiative Study Group. Assessment of tumor sequencing as a replacement for lynch
syndrome screening and current molecular tests for patients with colorectal cancer. JAMA Oncol. 2018;4:806-813.
10. Tamborero D, Rubio-Perez C, Deu-Pons J, et al. Cancer Genome Interpreter annotates the biological and clinical relevance of tumor alterations. Genome Med.
2018;10:25.
Oncol. 2016;27:795-800.
12. Schrader KA, Cheng DT, Joseph V, et al. Germline variants in targeted tumor sequencing using matched normal DNA. JAMA Oncol. 2016;2:104-111.
13. Mandelker D, Zhang L, Kemel Y, et al. Mutation detection in patients with advanced cancer by universal sequencing of cancer-related genes in tumor and
normal DNA vs guideline-based germline testing. JAMA. 2017;318:825-835.
14. Kalia SS, Adelman K, Bale SJ, et al. Recommendations for reporting of secondary findings in clinical exome and genome sequencing, 2016 update (ACMG SF
v2.0): a policy statement of the American College of Medical Genetics and Genomics [published correction appears in Genet Med. 2017]. Genet Med. 2017;
19:249-255.
15. Robson ME, Bradbury AR, Arun B, et al. American Society of Clinical Oncology policy statement update: genetic and genomic testing for cancer susceptibility.
J Clin Oncol. 2015;33:3660-3667.

16. Stockley TL, Oza AM, Berman HK, et al. Molecular profiling of advanced solid tumors and patient outcomes with genotype-matched clinical trials: the Princess
Margaret IMPACT/COMPACT trial. Genome Med. 2016;8:109.
17. Massard C, Michiels S, Ferté C, et al. High-throughput genomics and clinical outcome in hard-to-treat advanced cancers: results of the MOSCATO 01 trial.
Cancer Discov. 2017;7:586-595.
18. Meric-Bernstam F, Brusco L, Shaw K, et al. Feasibility of large-scale genomic testing to facilitate enrollment onto genomically matched clinical trials. J Clin
Oncol. 2015;33:2753-2762.
19. Klute K, Garrett-Mayer E, Halabi S, et al. Cobimetinib plus vemurafenib (C+V) in patients (Pts) with colorectal cancer (CRC) with BRAF V600E mutations: results
from the TAPUR Study. J Clin Oncol. 2020;38 15s (suppl; abstr 122).
20. Gupta R, Garrett-Mayer E, Halabi S, et al. Pertuzumab plus trastuzumab (P+T) in patients (Pts) with colorectal cancer (CRC) with ERBB2 amplification or
overexpression: Results from the TAPUR Study. J Clin Oncol. 2020;38 15s (suppl; abstr 132).
21. Meiri E, Garrett-Mayer E, Halabi S, et al. Pembrolizumab (P) in patients (Pts) with colorectal cancer (CRC) with high tumor mutational burden (HTMB): results
from the Targeted Agent and Profiling Utilization Registry (TAPUR) Study. J Clin Oncol. 2020;38 15s (suppl; abstr 133).
22. Le Tourneau C, Delord JP, Gonçalves A, et al; SHIVA investigators. Molecularly targeted therapy based on tumour molecular profiling versus conventional
therapy for advanced cancer (SHIVA): a multicentre, open-label, proof-of-concept, randomised, controlled phase 2 trial. Lancet Oncol. 2015;16:1324-1334.
23. Pennell NA, Mutebi A, Zhou Z-Y, et al. Economic impact of next-generation sequencing versus single-gene testing to detect genomic alterations in metastatic
non–small-cell lung cancer using a decision analytic model. Precision Oncol. 2019;3:1-9.
24. Fujiyoshi K, Yamamoto G, Takahashi A, et al. High concordance rate of KRAS/BRAF mutations and MSI-H between primary colorectal cancer and corre-
sponding metastases. Oncol Rep. 2017;37:785-792.
25. Brannon AR, Vakiani E, Sylvester BE, et al. Comparative sequencing analysis reveals high genomic concordance between matched primary and metastatic
colorectal cancer lesions. Genome Biol. 2014;15:454.
26. Tie J, Lipton L, Desai J, et al. KRAS mutation is associated with lung metastasis in patients with curatively resected colorectal cancer. Clin Cancer Res. 2011;
17:1122-1130.
27. Vakiani E, Janakiraman M, Shen R, et al. Comparative genomic analysis of primary versus metastatic colorectal carcinomas. J Clin Oncol. 2012;30:2956-2962.
28. Bhullar DS, Barriuso J, Mullamitha S, et al. Biomarker concordance between primary colorectal cancer and its metastases. EBioMedicine. 2019;40:363-374.
29. Reiter JG, Baretti M, Gerold JM, et al. An analysis of genetic heterogeneity in untreated cancers. Nat Rev Cancer. 2019;19:639-650.
6:147-153.
31. Siravegna G, Mussolin B, Buscarino M, et al. Clonal evolution and resistance to EGFR blockade in the blood of colorectal cancer patients [published correction
appears in Nat Med; 2017;21:doi:10.1038/nm0715-827b]. Nat Med. 2015;21:795-801.
32. McGranahan N, Swanton C. Biological and therapeutic impact of intratumor heterogeneity in cancer evolution [published correction appears in Cancer Cell.
2015;28:141]. Cancer Cell. 2015;27:15-26.
33. Mandel P, Metais P. Les acides nucléiques du plasma sanguin chez l’homme. C R Seances Soc Biol Fil. 1948;142:241-243.
34. Sorenson GD, Pribish DM, Valone FH, et al. Soluble normal and mutated DNA sequences from single-copy genes in human blood. Cancer Epidemiol
Biomarkers Prev. 1994;3:67-71.
35. Vasioukhin V, Anker P, Maurice P, et al. Point mutations of the N-ras gene in the blood plasma DNA of patients with myelodysplastic syndrome or acute
myelogenous leukaemia. Br J Haematol. 1994;86:774-779.
36. Tie J, Kinde I, Wang Y, et al. Circulating tumor DNA as an early marker of therapeutic response in patients with metastatic colorectal cancer. Ann Oncol. 2015;
26:1715-1722.
37. Bettegowda C, Sausen M, Leary RJ, et al. Detection of circulating tumor DNA in early- and late-stage human malignancies. Sci Transl Med. 2014;6:224ra24.
38. Bachet JB, Bouché O, Taieb J, et al. RAS mutation analysis in circulating tumor DNA from patients with metastatic colorectal cancer: the AGEO RASANC
prospective multicenter study. Ann Oncol. 2018;29:1211-1219.
39. Kinde I, Wu J, Papadopoulos N, et al. Detection and quantification of rare mutations with massively parallel sequencing. Proc Natl Acad Sci USA. 2011;
108:9530-9535.
40. Newman AM, Bratman SV, To J, et al. An ultrasensitive method for quantitating circulating tumor DNA with broad patient coverage. Nat Med. 2014;20:548-554.
41. Newman AM, Lovejoy AF, Klass DM, et al. Integrated digital error suppression for improved detection of circulating tumor DNA. Nat Biotechnol. 2016;
34:547-555.
42. Tie J, Wang Y, Tomasetti C, et al. Circulating tumor DNA analysis detects minimal residual disease and predicts recurrence in patients with stage II colon cancer.
Sci Transl Med. 2016;8:346ra92.
43. Tie J, Cohen JD, Wang Y, et al. Circulating tumor DNA analyses as markers of recurrence risk and benefit of adjuvant therapy for stage III colon cancer. JAMA
Oncol. 2019;5:1710.
44. Reinert T, Henriksen TV, Christensen E, et al. Analysis of plasma cell-free DNA by ultradeep sequencing in patients with stages I to III colorectal cancer. JAMA
Oncol. 2019;5:1124.

45. Thierry AR, Mouliere F, El Messaoudi S, et al. Clinical validation of the detection of KRAS and BRAF mutations from circulating tumor DNA. Nat Med. 2014;
20:430-435.
46. Schmiegel W, Scott RJ, Dooley S, et al. Blood-based detection of RAS mutations to guide anti-EGFR therapy in colorectal cancer patients: concordance of results
from circulating tumor DNA and tissue-based RAS testing. Mol Oncol. 2017;11:208-219.
47. Vidal J, Muinelo L, Dalmases A, et al. Plasma ctDNA RAS mutation analysis for the diagnosis and treatment monitoring of metastatic colorectal cancer patients.
Ann Oncol. 2017;28:1325-1332.
48. Grasselli J, Elez E, Caratù G, et al. Concordance of blood- and tumor-based detection of RAS mutations to guide anti-EGFR therapy in metastatic colorectal
cancer. Ann Oncol. 2017;28:1294-1301.
49. Garcı́a-Foncillas J, Tabernero J, Élez E, et al. Prospective multicenter real-world RAS mutation comparison between OncoBEAM-based liquid biopsy and tissue
analysis in metastatic colorectal cancer. Br J Cancer. 2018;119:1464-1470.
50. Normanno N, Esposito Abate R, Lambiase M, et al; CAPRI-GOIM Investigators. RAS testing of liquid biopsy correlates with the outcome of metastatic colorectal
cancer patients treated with first-line FOLFIRI plus cetuximab in the CAPRI-GOIM trial. Ann Oncol. 2018;29:112-118.
51. Thierry AR, Pastor B, Jiang ZQ, et al. Circulating DNA demonstrates convergent evolution and common resistance mechanisms during treatment of colorectal
52. Garlan F, Laurent-Puig P, Sefrioui D, et al. Early evaluation of circulating tumor DNA as marker of therapeutic efficacy in metastatic colorectal cancer patients
(PLACOL study). Clin Cancer Res. 2017;23:5416-5425.
53. Corcoran RB, André T, Atreya CE, et al. Combined BRAF, EGFR, and MEK inhibition in patients with BRAFV600E-mutant colorectal cancer. Cancer Discov. 2018;
8:428-443.
54. Siena S, Sartore-Bianchi A, Garcia-Carbonero R, et al. Dynamic molecular analysis and clinical correlates of tumor evolution within a phase II trial of
panitumumab-based therapy in metastatic colorectal cancer. Ann Oncol. 2018;29:119-126.
486:532-536.
56. Diaz LA Jr., Williams RT, Wu J, et al. The molecular evolution of acquired resistance to targeted EGFR blockade in colorectal cancers. Nature. 2012;
486:537-540.
2015;21:2157-2166.
3:658-673.
59. Montagut C, Dalmases A, Bellosillo B, et al. Identification of a mutation in the extracellular domain of the Epidermal Growth Factor Receptor conferring
cetuximab resistance in colorectal cancer [published correction appears in Nat Med. 2012;18:1445]. Nat Med. 2012;18:221-223.
60. Yonesaka K, Zejnullahu K, Okamoto I, et al. Activation of ERBB2 signaling causes resistance to the EGFR-directed therapeutic antibody cetuximab. Sci Transl
Med. 2011;3:99ra86.
61. Ahronian LG, Sennott EM, Van Allen EM, et al. Clinical acquired resistance to RAF inhibitor combinations in BRAF-mutant colorectal cancer through MAPK
pathway alterations. Cancer Discov. 2015;5:358-367.
62. Hazar-Rethinam M, Kleyman M, Han GC, et al. Convergent therapeutic strategies to overcome the heterogeneity of acquired resistance in BRAFV600E colorectal
cancer. Cancer Discov. 2018;8:417-427.
63. Oddo D, Sennott EM, Barault L, et al. Molecular landscape of acquired resistance to targeted therapy combinations in BRAF-mutant colorectal cancer. Cancer
Res. 2016;76:4504-4515.
64. Pietrantonio F, Oddo D, Gloghini A, et al. MET-driven resistance to dual EGFR and BRAF blockade may be overcome by switching from EGFR to MET inhibition
in BRAF-mutated colorectal cancer. Cancer Discov. 2016;6:963-971.
66. Morelli MP, Overman MJ, Dasari A, et al. Characterizing the patterns of clonal selection in circulating tumor DNA from patients with colorectal cancer refractory
to anti-EGFR treatment. Ann Oncol. 2015;26:731-736.
67. Parseghian CM, Loree JM, Morris VK, et al. Anti-EGFR-resistant clones decay exponentially after progression: implications for anti-EGFR re-challenge. Ann
Oncol. 2019;30:243-249.
68. Cremolini C, Rossini D, Dell’Aquila E, et al. Rechallenge for patients with RAS and BRAF wild-type metastatic colorectal cancer with acquired resistance to first-
line cetuximab and irinotecan: a phase 2 single-arm clinical trial. JAMA Oncol. 2019;5:343-350.
69. Sepulveda AR, Hamilton SR, Allegra CJ, et al. Molecular biomarkers for the evaluation of colorectal cancer: guideline from the American Society for Clinical
Pathology, College of American Pathologists, Association for Molecular Pathology, and the American Society of Clinical Oncology. J Clin Oncol. 2017;
35:1453-1486.
70. Battaglin F, Puccini A, Ahcene Djaballah S, et al. The impact of panitumumab treatment on survival and quality of life in patients with RAS wild-type metastatic
colorectal cancer. Cancer Manag Res. 2019;11:5911-5924.
71. Salazar R, Ciardiello F. Optimizing anti-EGFR therapy in colorectal cancer. Clin Cancer Res. 2015;21:5415-5416.

72. U.S. Food and Drug Administration. Amgen Inc. Vectibix (Panitumumab). https://www.accessdata.fda.gov/drugsatfda_docs/label/2009/125147s080lbl.pdf.
Accessed February 12, 2020.
73. European Medicines Agency. Amgen Inc. Vectibix (Panitumumab). www.ema.europa.eu/en/documents/product-information/vectibix-epar-product-
information_en.pdf. Accessed February 12, 2020.
74. U.S. Food and Drug Administration. ERBITUX (cetuximab). www.accessdata.fda.gov/drugsatfda_docs/label/2019/125084s273lbl.pdf. Accessed February 12,
2020.
75. European Medicines Agency. ERBITUX (cetuximab). www.ema.europa.eu/en/documents/product-information/erbitux-epar-product-information_en.pdf.
76. Normanno N, Tejpar S, Morgillo F, et al. Implications for KRAS status and EGFR-targeted therapies in metastatic CRC. Nat Rev Clin Oncol. 2009;6:519-527.
77. Douillard JY, Oliner KS, Siena S, et al. Panitumumab-FOLFOX4 treatment and RAS mutations in colorectal cancer. N Engl J Med. 2013;369:1023-1034.
Pathology, College of American Pathologists, Association for Molecular Pathology, and American Society of Clinical Oncology. Arch Pathol Lab Med. 2017;
141:625-657.
80. Bertotti A, Papp E, Jones S, et al. The genomic landscape of response to EGFR blockade in colorectal cancer. Nature. 2015;526:263-267.
81. Morano F, Corallo S, Lonardi S, et al. Negative hyperselection of patients with RAS and BRAF wild-type metastatic colorectal cancer who received
panitumumab-based maintenance therapy. J Clin Oncol. 2019;37:3099-3110.
82. Battaglin F, Naseem M, Lenz HJ, et al. Microsatellite instability in colorectal cancer: overview of its clinical significance and novel perspectives. Clin Adv Hematol
Oncol. 2018;16:735-745.
83. Vilar E, Gruber SB. Microsatellite instability in colorectal cancer-the stable evidence. Nat Rev Clin Oncol. 2010;7:153-162.
84. Boland CR, Goel A. Microsatellite instability in colorectal cancer. Gastroenterology. 2010;138:2073-2087.
85. Latham A, Srinivasan P, Kemel Y, et al. Microsatellite instability is associated with the presence of Lynch syndrome pan-cancer. J Clin Oncol. 2019;37:286-295.
86. Hampel H, Frankel WL, Martin E, et al. Screening for the Lynch syndrome (hereditary nonpolyposis colorectal cancer). N Engl J Med. 2005;352:1851-1860.
87. Sinicrope FA. Lynch syndrome-associated colorectal cancer. N Engl J Med. 2018;379:764-773.
88. Phipps AI, Limburg PJ, Baron JA, et al. Association between molecular subtypes of colorectal cancer and patient survival. Gastroenterology. 2015;148:77-87.
89. Stoffel EM, Mangu PB, Gruber SB, et al; European Society of Clinical Oncology. Hereditary colorectal cancer syndromes: American Society of Clinical Oncology
Clinical Practice Guideline endorsement of the familial risk-colorectal cancer: European Society for Medical Oncology Clinical Practice Guidelines. J Clin Oncol.
2015;33:209-217.
90. National Institute for Health and Care Excellence. NICE guidelines. www.nice.org.uk/guidance/dg27/chapter/1-Recommendations. Accessed February 12,
2020.
91. Sargent DJ, Marsoni S, Monges G, et al. Defective mismatch repair as a predictive marker for lack of efficacy of fluorouracil-based adjuvant therapy in colon
92. Dienstmann R, Mason MJ, Sinicrope FA, et al. Prediction of overall survival in stage II and III colon cancer beyond TNM system: a retrospective, pooled
biomarker study. Ann Oncol. 2017;28:1023-1031.
93. Venderbosch S, Nagtegaal ID, Maughan TS, et al. Mismatch repair status and BRAF mutation status in metastatic colorectal cancer patients: a pooled analysis
of the CAIRO, CAIRO2, COIN, and FOCUS studies. Clin Cancer Res. 2014;20:5322-5330.
94. Cohen R, Buhard O, Cervera P, et al. Clinical and molecular characterisation of hereditary and sporadic metastatic colorectal cancers harbouring microsatellite
instability/DNA mismatch repair deficiency. Eur J Cancer. 2017;86:266-274.
95. Sidaway P. MSI-H: a truly agnostic biomarker? Nat Rev Clin Oncol. 2020;17:68.
2019;25:3753-3758.
97. Thomas J, Leal A, Overman M. Clinical development of immunotherapy for deficient mismatch repair (dMMR) colorectal cancer. Clin Colorectal Cancer. In
press.
98. Overman MJ, Lonardi S, Wong KYM, et al. Durable clinical benefit with nivolumab plus ipilimumab in DNA mismatch repair-deficient/microsatellite instability-
high metastatic colorectal cancer. J Clin Oncol. 2018;36:773-779.
99. Lenz H-J, Lonardi S, Zagonel V, et al. Nivolumab (NIVO) + low-dose ipilimumab (IPI) as first-line (1L) therapy in microsatellite instability-high/DNA mismatch
repair deficient (MSI-H/dMMR) metastatic colorectal cancer (mCRC): clinical update. J Clin Oncol. 2019;37 15s (suppl; abstr 3521).
100. Kok M, Chalabi M, Haanen J. How I treat MSI cancers with advanced disease. ESMO Open. 2019;4 (suppl 2):e000511.
101. Burotto M, Chiou VL, Lee JM, et al. The MAPK pathway across different malignancies: a new perspective. Cancer. 2014;120:3446-3456.
102. Millington GW. Mutations of the BRAF gene in human cancer, by Davies et al. (Nature 2002; 417: 949-54). Clin Exp Dermatol. 2013;38:222-223.
103. Yao Z, Yaeger R, Rodrik-Outmezguine VS, et al. Tumours with class 3 BRAF mutants are sensitive to the inhibition of activated RAS. Nature. 2017;548:234-238.

104. Schirripa M, Biason P, Lonardi S, et al. Class 1, 2, and 3 BRAF-mutated metastatic colorectal cancer: a detailed clinical, pathologic, and molecular
characterization. Clin Cancer Res. 2019;25:3954-3961.
105. Clancy C, Burke JP, Kalady MF, et al. BRAF mutation is associated with distinct clinicopathological characteristics in colorectal cancer: a systematic review and
meta-analysis. Colorectal Dis. 2013;15:e711-e718.
106. Modest DP, Ricard I, Heinemann V, et al. Outcome according to KRAS-, NRAS- and BRAF-mutation as well as KRAS mutation variants: pooled analysis of five
randomized trials in metastatic colorectal cancer by the AIO colorectal cancer study group. Ann Oncol. 2016;27:1746-1753.
107. Rowland A, Dias MM, Wiese MD, et al. Meta-analysis of BRAF mutation as a predictive biomarker of benefit from anti-EGFR monoclonal antibody therapy for
RAS wild-type metastatic colorectal cancer. Br J Cancer. 2015;112:1888-1894.
108. Ducreux M, Chamseddine A, Laurent-Puig P, et al. Molecular targeted therapy of BRAF-mutant colorectal cancer. Ther Adv Med Oncol. 2019;
11:1758835919856494.
109. Kopetz S, Desai J, Chan E, et al. Phase II pilot study of vemurafenib in patients with metastatic BRAF-mutated colorectal cancer. J Clin Oncol. 2015;
33:4032-4038.
110. Falchook GS, Long GV, Kurzrock R, et al. Dabrafenib in patients with melanoma, untreated brain metastases, and other solid tumours: a phase 1 dose-escalation
trial. Lancet. 2012;379:1893-1901.
111. Gomez-Roca CA, Delord J, Robert C, et al. 535Pencorafenib (LGX818), an oral BRAF inhibitor, in patients (pts) with BRAF v600e metastatic colorectal cancer
(MCRC): results of dose expansion in an open-label, phase 1 study. Ann Oncol. 2014;25:iv182-iv183.
112. Cremolini C, Loupakis F, Antoniotti C, et al. FOLFOXIRI plus bevacizumab versus FOLFIRI plus bevacizumab as first-line treatment of patients with metastatic
colorectal cancer: updated overall survival and molecular subgroup analyses of the open-label, phase 3 TRIBE study. Lancet Oncol. 2015;16:1306-1315.
113. Cremolini C, Antoniotti C, Lonardi S, et al. Updated results of TRIBE2, a phase III, randomized strategy study by GONO in the 1st- and 2nd-line treatment of
unresectable mCRC. Ann Oncol. 2019;30(4):iv154.
114. Kopetz S, McDonough SL, Morris VK, et al. Randomized trial of irinotecan and cetuximab with or without vemurafenib in BRAF-mutant metastatic colorectal
cancer (SWOG 1406). J Clin Oncol. 2017;35 15s (suppl; abstr 520).
116. Kopetz S, Grothey A, Van Cutsem E, et al. BEACON CRC: a randomized, 3-Arm, phase 3 study of encorafenib and cetuximab with or without binimetinib vs.
choice of either irinotecan or FOLFIRI plus cetuximab in BRAF V600E–mutant metastatic colorectal cancer. Ann Oncol. 2019;30(4):iv137-iv151.
117. Siena S, Sartore-Bianchi A, Marsoni S, et al. Targeting the human epidermal growth factor receptor 2 (HER2) oncogene in colorectal cancer. Ann Oncol. 2018;
29:1108-1119.
118. Sartore-Bianchi A, Amatu A, Porcu L, et al. HER2 positivity predicts unresponsiveness to egfr-targeted treatment in metastatic colorectal cancer. Oncologist.
2019;24:1395-1402.
119. Bregni G, Sciallero S, Sobrero A. HER2 amplification and anti-EGFR sensitivity in advanced colorectal cancer. JAMA Oncol. 2019;5:605-606.
121. Hainsworth JD, Meric-Bernstam F, Swanton C, et al. Targeted therapy for advanced solid tumors on the basis of molecular profiles: results from Mypathway, an
open-label, phase IIa multiple basket study. J Clin Oncol. 2018;36:536-542.
122. Nakamura Y, Okamoto W, Kato T, et al. TRIUMPH: Primary Efficacy of a Phase II Trial of Trastuzumab (T) and Pertuzumab (P) in Patients (pts) with Metastatic
Colorectal Cancer (mCRC) with HER2 (ERBB2) Amplification (amp) in Tumor Tissue or Circulating Tumor DNA (ctDNA): A GOZILA Sub-study. Ann Oncol.
2019;30(5):v198-v252.
123. Strickler JH, Zemia T, Ou F, et al. Trastuzumab and tucatinib for the treatment of HER2 amplified metastatic colorectal cancer (mCRC): Initial results from the
MOUNTAINEER trial. Ann Oncol. 2019;30(5):v198-v252.
124. Sartore-Bianchi A, Martino C, Lonardi E, et al. Phase II Study of Pertuzumab and Trastuzumab-emtansine (T-DM1) in Patients with HER2-positive Metastatic
Colorectal Cancer: the HERACLES-B (HER2 Amplification for Colo-rectaL cancer Enhanced Stratification, cohort B) Trial. Ann Oncol. 2019;30(5):v851-v934.

Colorectal Cancer in the Young: Epidemiology,

Prevention, Management
Rebecca L. Siegel, MPH1; Christopher Dennis Jakubowski, MD2; Stacey A. Fedewa, PhD1; Anjee Davis, MPPA3; and
Nilofer S. Azad, MD2
overview
Colorectal cancer (CRC) incidence rates in the United States overall have declined since the mid-1980s
because of changing patterns in risk factors (e.g., decreased smoking) and increases in screening. However,
this progress is increasingly confined to older adults. CRC occurrence has been on the rise in patients younger
than age 50, often referred to as early-onset disease, since the mid-1990s. Young patients are more often
diagnosed at an advanced stage and with rectal disease than their older counterparts, and they have numerous
other unique challenges across the cancer management continuum. For example, young patients are less
likely than older patients to have a usual source of health care; often need a more complex treatment protocol
to preserve fertility and sexual function; are at higher risk of long-term and late effects, including subsequent
primary malignancies; and more often suffer medical financial hardship. Diagnosis is often delayed because of
provider- and patient-related factors, and clinicians must have a high index of suspicion if young patients
present with rectal bleeding or changes in bowel habits. Educating primary care providers and the larger
population on the increasing incidence and characteristic symptoms is paramount. Morbidity can further be
averted by increasing awareness of the criteria for early screening, which include a family history of CRC or
polyps and a genetic predisposition.
CHANGING EPIDEMIOLOGY effect.5,8,11 A birth cohort effect occurs when age-

Rising incidence of early-onset CRC (EO-CRC) was first specific incidence rates vary by generation because
noted at the population level in 20031 but did not begin of changes in exposure that influence disease risk, in
to gain traction as a public health concern for another contrast to period effects, in which incidence varies
decade after publication of a second report.2 In the at the same point in time for all age groups. In the
years since, epidemiologic studies have provided clues United States, people born in the 1950s have the
for what might be causing the trends. For example, lowest CRC incidence, and the risk of disease since
incidence patterns are consistent in men and women, has increased with each subsequent generation
after declining in the first half of the 20th century.5
implicating exposures that are not sex-specific but vary
The phenomenon is clearly visible when temporal
by anatomic subsite, stage at diagnosis, race and eth-
trends are stratified by granular age groups, because
nicity, and geographic area of residence. Inclines are
the elevated risk of disease travels with particular
steepest for advanced-stage disease, for tumors in the
birth cohorts as they advance in age. Among young
distal colon and rectum, and among non-Hispanic
adults, for example, previously declining colon
whites.2-7 A recent study based on cancer registry
cancer incidence began increasing in the mid-1980s
data from 47 states found that incidence is increasing
among people age 20 to 29 but not until the late-
among non-Hispanic whites in most states, with the
Author affiliations 1980s in ages 30 to 39 and the mid-1990s in ages 40
most rapid pace in the West.6 From 1995 to 2015, for
and support to 54.5 During the past decade, an accompanying
information (if example, the CRC incidence rate in people younger than
increase in mortality has emerged, with CRC death
applicable) appear age 50 increased by 57% in Colorado and by 73% in
at the end of this
rates increasing by more than 1% per year since
Washington State. However, contemporary incidence
article. 2004 among adults younger than age 55 after pre-
rates remained generally highest in the South and lowest
Accepted on March viously declining trends.12,13 Rates have recently
in the West. The unique increase among young people
17, 2020 and begun to tick up in people age 50 to 64, for reasons
is also occurring outside the United States in many high-
published at that are unknown.12 A consequence of these op-
ascopubs.org on income countries, including Australia, Canada, Ger-
positional trends is an increasingly younger patient
April 21, 2020: many, and the United Kingdom.8-10
population; the median age at CRC diagnosis has
10.1200/EDBK_ In the United States and elsewhere, the CRC inci- decreased from age 72 in the early 2000s to age 66
279901 dence pattern is characterized by a strong birth cohort today (Fig. 1).14

Siegel et al
• The rate of early-onset CRC (, 50 years at
diagnosis) has been increasing for 2 decades
for unknown reasons.
• Screening for CRC in average-risk patients (no
family history or predisposing conditions) is
recommended by the American Cancer Society
to begin at age 45.
• Patients present with characteristic symptoms
of abdominal pain, rectal bleeding, and
changes in bowel habits, but diagnosis often is
delayed.
• Currently, patients with early-onset CRC are
treated in the adjuvant and metastatic setting,
just as their older counterparts are.
• Fertility, pregnancy, sexual health, financial
toxicity, and long-term survivorship challenges FIGURE 1. Median Age at Colorectal Cancer Diagnosis in the United
are some of the unique issues in the man- States, 1990–2016
agement of early-onset CRC.
The strongest known risk factor for CRC is a family history of

the disease. People with a first-degree relative (FDR) who
RISK FACTORS OF EO-CRC has been diagnosed with CRC have two to four times the risk
A strong birth cohort effect indicates population-level of someone without this family history, with higher risk for
changes in behavioral factors that influence cancer risk. diagnosis before age 50 and multiple affected relatives.32 A
Major established modifiable risk factors for CRC are CRC history among more distant relatives is also associated
excess body weight,15 cigarette smoking,16 heavy alcohol with increased risk,33 as is a family history of adenomas.34
consumption,17 a high intake of red or processed meat,18 More than one-quarter of patients younger than age 50 have
and physical inactivity.19 Importantly, however, these as- an FDR with a history of CRC or adenomas, and an addi-
sociations are based almost entirely on cancer occurrence tional 16% have a hereditary syndrome, half of whom have
in older cohorts. The risk of EO-CRC among U.S. women Lynch syndrome.35,36 People with Lynch syndrome are also
was recently shown to have increased by 20% for every 5- at elevated risk for many other cancers, including endo-
unit increase in body mass index and by 69% for more than metrial, ovarian, small intestine, stomach, urinary bladder,
14 hours per week of television watching.20,21 An analysis of and female breast.37 Although rigorous colonoscopy sur-
data on CRC in people younger than age 45 from two veillance leads to early-stage CRC diagnosis and high
European case-control studies found relative risks for other survival,38 most of the estimated 1.2 million Americans (1 in
modifiable factors similar to those in older patients. 22 279) who have Lynch syndrome are undiagnosed.39 For this
However, a recent retrospective study of patients di- reason, the National Comprehensive Cancer Network and
agnosed at a New York academic center found no ASCO recommend universal testing for Lynch syndrome in
significant associations between obesity, smoking, and di- patients with CRC or endometrial cancer.40,41 However,
abetes and EO-CRC risk.23 Although molecular and clinical implementation of screening has been slow in the com-
characteristics in patients age 30 to 49 are similar to those in munity setting,42 despite coverage by most major public and
patients age 50 and older,24 suggesting common carcino- private insurers.43 Identification of high-risk families in the
genic pathways, there remains an urgent need for research absence of a genetic syndrome also offers substantial op-
on the influence of new, highly prevalent early-life exposures portunity to mitigate the cancer burden through early
on CRC risk.25 For example, associations have recently been screening. However, a major obstacle is incomplete patient
reported between colorectal tumors and antibiotic use as family history in medical records. One study found that less
well as high-fructose corn syrup, probably mediated by than half of primary care physicians document information
alterations in the composition of gut microbiota.26-28 The role about family members other than FDRs, and age at cancer
of dietary components in regulating the microbiome and diagnosis, which is a crucial indicator of disease risk, was
its impact on cancer risk is an area of active scientific rarely collected.44 Another study found that only 22% of the
study.29-31 medical records of patients with CRC had family history

Colorectal Cancer in the Young
information sufficient to identify people who should be re- in life years, discord between the simulation models, and
ferred for genetic counseling or testing.45 lack of empiric evidence to support screening among
Although a larger proportion of early CRCs are hereditary people in their 40s. Table 1 outlines recommendations
compared with late CRCs, the majority of cases are sporadic regarding age to initiate CRC screening according to dif-
(i.e., occur in average-risk patients). CRC is the most ferent patient risk categories.
commonly diagnosed cancer and the leading cause of Findings from recent studies support screening before age
cancer death in men younger than age 50, whereas in 50. Although data on colonoscopy outcomes among people
women it ranks fourth (after breast, thyroid, and melanoma) younger than age 50 are limited, average-risk adults age
and second (after breast) in terms of incidence and mor- 40 to 49 appear to have prevalences of any adenoma
tality, respectively.46,47 Although the absolute risk of a CRC (14%–16%), large polyps (3%–4% in women and in
diagnosis by age 50 remains low (0.4% vs. 3.3% from age 5%–6% men), and distal large polyps (5%) similar to those
50 to 85),48 the burden for young adults is substantial and observed among adults age 50 to 54.56-58 In addition, there
growing (Fig. 2). In 2020, 17,930 (12%) of the estimated is a higher burden of prevalent CRC in people age 45 to
147,950 cases of CRC in the United States will be diagnosed 49 than what is suggested by observed incidence rates, as
in people younger than age 50—the equivalent of ap- indicated by the pronounced spike in incidence between
proximately 50 per day, in addition to 3,640 (of 53,200) CRC age 49 and 50 upon screening initiation.59 Recent studies
deaths in that age group.12 According to cancer registry data have also shown screening beginning at age 40 or 45 to be
covering more than 96% of the U.S. population, the age- cost effective.60,61 Although health insurance coverage for
standardized CRC incidence rate in ages 20 to 49 between average-risk screening before age 50 is variable, changes
2012 and 2016 ranged from 30 (per 100,000) in Alaska to USPSTF guidelines, expected in 2020 or 2021, could
Natives to 14 in blacks, 13 in non-Hispanic whites, and nine eliminate that obstacle, because the Affordable Care Act
in Hispanics and Asian and Pacific Islanders.47 However, requires coverage for USPSTF-recommended preventive
rectal cancer incidence rates are now slightly higher in non- services.
Hispanic whites (5.1 per 100,000) than in blacks (4.5 per
Screening before age 50 is universally recommended for
100,000) because of the rapid increase in the incidence of
people at elevated risk of CRC because of familial syn-
these tumors among whites.
dromes (e.g., familial adenomatous polyposis, Lynch),
SCREENING GUIDELINES: TIME FOR A CHANGE? chronic inflammatory bowel disease, or a family history of
CRC or adenoma. 50,51 In addition, some organizations
The majority of people diagnosed with CRC before age 50
recommend beginning screening at age 45 for African
are at average risk with respect to screening,23,35 and half
Americans and Alaskan Natives because of their elevated
(48%) of these patients are age 45 to 49.47 Most organi-
risk.51,62 It is noteworthy that, in recent years (2015–2016),
zations, including the U.S. Preventive Services Task Force
incidence rates among non-Hispanic whites age 20 to 49
(USPSTF), recommend screening average-risk adults for
are the same as those among blacks (14.1 per 100,000).12
CRC with colonoscopy (every 10 years), sigmoidoscopy
(every 5 years), or stool testing (annually) beginning at age On the basis of national estimates, 13% of people age 40 to
50.49-51 However, in 2018 the American Cancer Society 44 and 21% of adults age 45 to 49 met the definition for up-
prompted substantial controversy by lowering their rec- to-date screening in 2018 compared with 67% of those age
ommended age to begin from age 50 to age 45.52-54 The 50 and older.12 CRC screening test use among people in
revision was based on an extensive review of empiric evi- their 40s is more common for those with a family history,
dence on disease risk and the benefits and harms of although less than half of people with an FDR report
screening, along with results from two simulation models screening, and most (. 80%) of those who are tested in
from the Cancer Intervention and Surveillance Modeling their 40s have no family history.63,64 Blacks age 45 to 49 are
Network that compared numerous combinations of screening approximately 30% more likely than whites to have had
strategies with regard to age to begin and end screening, a recent colonoscopy because of the long-standing rec-
type of test, and test interval. These models were the same ommendation for screening in this population.64 Patterns of
as those used to inform the 2016 USPSTF recommenda- CRC screening test use do not appear to explain the in-
tions but with an adjustment to account for the recent increase in EO-CRC through lead-time bias, as previously
crease in underlying risk of CRC. Two of three unadjusted hypothesized.65 Past-year colonoscopy use in people age 40
Cancer Intervention and Surveillance Modeling Network to 44 remained steady at 3% from 2000 to 2015, despite
models used to inform the 2016 USPSTF recommendations a 28% relative increase in CRC incidence, whereas prev-
found that screening beginning at age 45 rather than age 50 alence doubled among people age 45 to 49, but CRC in-
resulted in a more favorable balance of benefit to harm,55 cidence was stable for localized tumors and increased only
but the USPSTF resisted the change, citing a modest gain for advanced disease.64 Among people age 40 to 49,

Siegel et al
FIGURE 2. Opportunities for Mitigating the Burden of Colorectal Cancer Among Patients Younger Than Age 50 Across the Cancer Continuum
increasing incidence is limited to advanced-stage disease.7 model and screening guidelines released in 2018 that rec-
The 5-year survival rate among patients with EO-CRC de- ommend starting screening at age 45 for the normal-risk
creases from 94% for localized-stage disease to 21% for population.68 It is imperative to continue to reassess epide-
distant-stage diagnoses.12 Young patients are diagnosed at miology data and screening guidelines for at-risk populations;
a later stage than older patients, even when screening- currently, the American College of Gastroenterology and the
detected cancers are excluded.66 American Society for Gastrointestinal Endoscopy support
screening from age 45 for African Americans.69 In addition,
OPPORTUNITIES TO REDUCE CRC IN THE YOUNG WITH the U.S. Multi-Society Task Force of Colorectal Cancer rec-
CHANGES IN SCREENING ommends screening at age 40 for all patients with a family
Most current guidelines recommend the initiation of history of CRC at any age.70 Incidence data must continue to
screening for average-risk patients at age 50 (USPSTF, the be analyzed and mined to ensure appropriate threshold ages
Canadian Task Force on Preventive Health Care, the Eu- for beginning of screening.
ropean Council, the American Academy of Family Physi- OTHER POPULATIONS TO TARGET
cians, and the American College of Physicians). 49,67
However, it is becoming increasingly recognized that spo- First-Degree Relatives With Adenomas
radic EO-CRC represents a large number of CRC cases, with Taking advantage of our understanding of the continuum of
growing morbidity and mortality, because these patients colorectal tissue neoplasia, which has a well-described
present more often with stage III or IV disease. The biggest progression from adenoma to dysplasia to carcinoma, we
increases in CRC are occurring among people younger than have an opportunity to modify screening recommendations
age 40, which suggests that consideration should be given for patients with a family history of polyps to potentially
to starting screening at age 40.2,4,5 Cancer screening models capture patients at an earlier, benign point in this
must be updated with the most current data regarding age sequence.71,72 A large proportion of patients with sporadic
and incidence of CRC. The American Cancer Society has EO-CRC have a family history of advanced polyps.36 Ac-
shown leadership on this issue, with a recent update to their cordingly, clinicians should include not only CRC but also

TABLE 1. Age to Initiate CRC Screening Based on Risk Category

Risk Category Family History Age to Initiate Screening Recommended Test
“Average” Risk (no known Age 50a-g Colonoscopy (every 10 years), sigmoidoscopy (every 5
family history) Age 45h years), multitargeted stool DNA (every 3 years), or
fecal occult blood test or FIT (annually)
African American or Alaskan Age 45f
Native Who Are at
“Average" Risk
Familial Adenomatous Age 10–12g Colonoscopy (annually until colectomy)
Polyposis
Lynch Syndrome Age 20–25 or 2–5 years younger than Colonoscopy (every 1–2 years)
youngest age at diagnosis of CRC in
family if diagnosis before age 25f
Inflammatory Bowel Disease 8 years after disease onsetg Colonoscopy (every 1–3 years)
Relative With CRC Cancer in an FDR Age 40 or 10 years younger than age of Colonoscopy every 5 years
diagnosis of FDRf,g
Cancer in 2 SDRs Age 40f
FDR With Advanced Advanced adenoma in 1 Age 40 or 10 years younger than age of Colonoscopy every 5 years
Colorectal Polyp FDR , 60 years or in diagnosis of FDRf
2 FDRs
Advanced adenoma in 1 Age 40f Colonoscopy every 10 years or FIT annually
FDR 60 years
Confirmed advanced Age 40 or at age of diagnosis of Colonoscopy every 5–10 years
polyp in 1 FDR (any advanced adenoma in FDRg
age)
Abbreviations: CRC, colorectal cancer; FDR, first-degree relative; FIT, fecal immunochemical testing; SDR, second-degree relative.
a
U.S. Preventive Services Task Force.
b
Canadian Task Force on Preventive Health Care.
c
European Council.
d
American Academy of Family Physicians.
e
American College of Physicians.
f
U.S. Multi-Society Task Force of Colorectal Cancer, which represents the American College of Gastroenterology, the American Gastroenterological
Association, and the American Society for Gastrointestinal Endoscopy.
g
National Comprehensive Cancer Network.
h
American Cancer Society.
a family history of advanced colorectal polyps in an as- interventions. The National Colorectal Cancer Roundtable
sessment of a patient’s risk. created the Advanced Colorectal Polyp GI Brief to provide
Currently the U.S. Multi-Society Task Force of Colorectal endoscopists and primary care clinicians with a resource
Cancer, American College of Gastroenterology, American to help treat patients with advanced polyps.74,75 One
Gastroenterological Association, and American Society for suggestion is for endoscopists to draft a personalized
Gastrointestinal Endoscopy recommend advanced screening letter detailing patient colonoscopy results, follow-up,
for patients who have an FDR with an advanced adenoma and risk factors so that patients are more likely to share
before age 60; these recommendations include screening this information with relatives about potential screening
colonoscopy at age 40 or 10 years younger than age of di- implications.76
agnosis of advanced adenoma in the FDR, with follow-up Overall, the guidelines from the American Cancer Society,
colonoscopy every 5 years. The National Comprehensive the U.S. Multi-Society Task Force of Colorectal Cancer, and
Cancer Network recommends starting screening at age 40 the American College of Radiology for FDRs with adenoma
for patients with an FDR with a history of advanced polyps at are identical to the recommendations for an FDR with CRC
any age.73 before age 60.77 However, the USPSTF currently does not
All young adults should be counseled to speak to their support this recommendation, because it could overwhelm
family regarding their history of advanced polyps so that screening capacity because of the high prevalence of ad-
they can be considered for earlier screening or other enomas in the age 50 to 59 population and because there

Siegel et al
was insufficient evidence that increased screening would groups to educate the provider community about the
reduce mortality.49,78 changing incidence of this entity. Primary care physicians
Other Risk Factors and other clinicians can have a dramatic impact on the
morbidity and mortality of EO-CRC by ruling out serious
There is also a real opportunity to improve outcomes in EO- causes of these symptoms in young patients (rectal
CRC by maximizing compliance in populations in which bleeding, abdominal pain, change in bowel habits, and
guidelines already recommend early screening. Studies anemia).88-90 Indeed the American Society for Gastroin-
have suggested that there is often poor adherence to early testinal Endoscopy recommends endoscopic evaluation of
screening among patients with known cancer syndromes or all patients with lower gastrointestinal bleeding.91 Although
familial CRC, especially among those at low socioeconomic diagnosis delays do not completely account for the dis-
levels for many reasons, including distrust of the medical proportionate distant-stage disease in young patients,66
community, poor access to health care, and uncertainty in improving access to care and educating patients and cli-
navigating the health care system.79 Interventions aimed at nicians about the symptoms of CRC would undoubtedly
educating patients and providers about recommendations improve outcomes in this population.92
and resources, including social work and patient navigator
supports, could result in significantly increased compliance TREATMENT AND CARE CONSIDERATIONS IN YOUNG
with screening recommendations. Another group at ele- PATIENTS WITH CRC
vated risk that should be considered for early screening is Several reviews have systematically examined and sum-
those who have received pelvic radiation in adolescence or marized studies related to prognosis and treatment out-
early adulthood. Finally, with the predominance of left-sided comes in EO-CRC.93,94 As previously described, younger
or rectal cancers in EO-CRC, sigmoidoscopy screening patients are more likely than their older counterparts to
initiated at an earlier age could be another screening present with advanced regional or distal disease. Compared
solution. 80 with older patients, patients with EO-CRC have higher
BARRIERS TO CARE cancer-specific survival at every stage, despite usually
high-risk pathology, as indicated by population-based
Beyond increasing appropriate screening and surveillance, studies.12,93 This difference may reflect fewer comorbid-
there is an immediate opportunity to reduce mortality ities or more aggressive treatment regimens. In the meta-
through earlier diagnosis. One single-institution study found static setting, patients with EO-CRC treated with standard
a median time from onset of rectal cancer symptom to regimens or clinical trials can have poorer progression-free
treatment of 217 days for patients younger than age 50 survival but do not have worse overall survival or response
compared with 29.5 days for those older than age 50, largely rates.95,96 Patients with EO-CRC are more likely to receive
because of patient delays in presentation to the initial additional surgical therapy for both early-stage and meta-
physician. 81 Some of these delays are caused by static disease, potentially reflecting both patient and pro-
misdiagnoses.82 Patient-based delays in seeking care may vider age-related biases.
be related to poor knowledge of worrisome symptoms,
embarrassment, denial, low health literacy, and poor social Current national and international guidelines do not have
and family support. Additional factors include lack of access different treatment recommendations for patients with EO-
to health care (e.g., being uninsured or underinsured) and CRC and patients with later-onset CRC, but a more ag-
poor access to transportation. More challenging to quantify gressive treatment paradigm is often pursued. Multiple
in terms of effect, younger patients may have competing studies describing adjuvant approaches with more ag-
demands of time with familial and employment re- gressive systemic cytotoxic regimens, targeted agents, or
sponsibilities that supersede attention to personal health. surgical approaches differing from current guidelines have
led to potential overtreatment with unclear benefits.97-102
Patients with EO-CRC can present with characteristic Younger patients can be overtreated because oncologists
symptoms, including abdominal pain, weight loss, and fa- perceive them to have fewer adverse reactions to chemo-
tigue, but tend to have a higher rates of left-sided related therapy, fewer comorbidities, and a worse prognosis at
symptoms at presentation, including rectal bleeding and diagnosis.
changes in bowel habits.2,83,84 Diagnosis can be delayed
because symptoms are attributed to a low index of suspicion Early-Stage Disease
on the part of treating providers, who focus on more Compared with an older cohort (age 65–75), more patients
common conditions in young adults.85,86 A recent study with early-stage EO-CRC (age 18–49) are given adjuvant
found that, among the 52% of patients with EO-CRC who therapy for stage II and III disease, including low- and high-
experienced rectal bleeding, the average time from onset of risk stage II disease.101,103 Kneuertz et al101 reported that
bleeding to diagnosis was 271.17 days.87 This challenge patients with stage II low-risk EO-CRC receive adjuvant
can be addressed by working with national primary care therapy 50% of the time, compared with 19.1% in the 65- to

75-year-old cohort, with no improved survival in younger some outliers.106,107 A genetic study from France sequenced
patients. Another study using the National Cancer Database 39 patients with sporadic EO-CRC (before age 45) for TP53,
analyzed more than 40,000 patients with early-stage CRC to KRAS, BRAF, and PIK3CA mutations and the presence of
examine differences in characteristics of patients with CRC a methylator phenotype. Gene expression studies were also
younger than age 50 compared with those older than 50. It performed to elucidate activated cellular pathways in these
found that patients with EO-CRC were more likely to receive samples. They found fewer BRAF mutations, fewer meth-
National Comprehensive Cancer Network guideline–driven ylator phenotypes, and upregulation of certain signaling
care but had no survival advantage as a result. Alternatively, pathways (Wnt/beta catenin, MAP kinase, growth factor
in the cohort of patients older than age 50, those who re- signaling, TNFR1 pathway), suggesting that EO-CRC may
ceived guideline-driven care had better survival than those be a distinctive molecular entity.108 Another study looking
who did not.104 at tumors from patients age 30 or younger found that
There have also been studies examining the efficacy of microsatellite instability in EO-CRC was more prevalent, was
specific systemic agents based on age. An analysis of not tightly linked to MLH1/PMS2 loss, and was never as-
patients younger than age 60 from the CAO/ARO/AIO- sociated with BRAFV600E mutations.100 These were small
04 (Working Group of Surgical Oncology/Working Group of studies, and larger trials will be needed to truly assess the
Radiation Oncology/Working Group of Medical Oncology of molecular differences between younger and older patients
the Germany Cancer Society), randomized, phase III trial with CRC. A recent study of more than 36,000 patients with
showed that adding oxaliplatin to 5-fluorouracil–based CRC in four study cohorts found that the continuum of
adjuvant chemotherapy reduces local and distal re- clinical and molecular age-associated differences slows
currence in this younger cohort, similar to the entire cohort after age 30 and that characteristics of patients age 40 to 49
in the original study results.105 Regarding targeted agents in are very similar to those of patients age 50 to 59.24
the adjuvant space, when cetuximab or bevacizumab is ADDITIONAL CARE CONSIDERATIONS IN PATIENTS
added to adjuvant fluorouracil, leucovorin, and oxaliplatin WITH EO-CRC
regimens, there is no increase in survival, although younger
Sexuality
patients seem to tolerate multiagent regimens better than
their older counterparts.97-99,102 With a rise in EO-CRC, there is a clear need to identify and
address survivorship concerns that are unique to younger
Advanced and Metastatic Disease
adults who may be married, unmarried, dating, or still ex-
Recommendations for the treatment of metastatic disease ploring their sexuality. Sexual health is often neglected yet
are the same regardless of age. For patients with EO-CRC, critically important for patients with EO-CRC, who may be
one should consider assessment of performance status, reluctant to talk to their health care teams or unaware that
comorbidity, RAS/BRAF status, and primary tumor sided- interventions for sexual challenges may be available. A
ness. Regarding specific regimens, outcomes in the cohort cross-sectional study in France of patients with CRC ages
younger than age 50 from nine phase III, fluorouracil-based, 20 to 84 found that only 20% of men and 11% of
single-agent and combination studies were analyzed by women—11% with colon cancer, and 33% with rectal
Blanke et al.95 They concluded that patients with EO-CRC cancer—discussed sexuality with their cancer team, al-
had lower progression-free survival but no difference in though younger patients (, age 55) received more in-
relative risk of death or overall survival compared with the formation than their older counterparts.109
cohort of patients older than age 50, a finding that persisted
whether the age cutoff was 40 or 50 years. Of note, nausea Studies have shown that patients with CRC have higher rates
was more likely in younger cohorts, but diarrhea and of sexual dysfunction than the general population and are
neutropenia were less common. Another systematic review less sexually active after surgery. Female patients may face
looked at 24 first-line clinical trials, including trials with vaginal reconstruction, dryness, or pain during intercourse,
double and triple therapy and targeted agents. The trials and male patients may experience erectile dysfunction.
included 3,051 patients in total, 15% of whom were younger Overall, women and patients with rectal cancer appear to
than age 50. They concluded that the youngest (closer report more sexual and body image distress than men or
to age 20) and oldest (older than approximately age 65) those with colon cancer.110 The type of surgery can have an
cohorts had the lowest progression-free and overall impact, as one study reported that women who underwent
survivals. 96 abdominoperineal excision were less sexually active than
women who underwent lower anterior resection. 111,112
MOLECULAR BIOLOGY CONSIDERATIONS Among men, erectile dysfunction was reported as a symp-
Regarding sidedness and mutations in the KRAS–NRAS– tom in 54% of rectal cancer survivors and 25% of colon
BRAF pathway, multiple studies have suggested a similar cancer survivors.111 Presence or history of ostomy is another
proportion of KRAS mutations in EO-CRC, but there are important factor in sexual outcomes and body image. One

Siegel et al
survey study found that patients with current or past ostomy increasing out-of-pocket expenses because of coinsurance,
reported worse sexual function, and those with a current copays for expensive cancer drugs, and—in some cases—
ostomy struggled more with body image issues than those drugs that are not covered by insurance. Young cancer
who never had an ostomy.113 survivors are more likely to experience material (e.g., trouble
As the number of patients with EO-CRC increases, there is paying bills), psychological (e.g., worrying about paying
a need to better understand and preserve a patient’s sexual bills), and behavioral (e.g., skipping medications) financial
health after treatment, surgery, and radiation. It is critical hardships. For example, in a nationally representative
that we empower and encourage patients, caregivers, and survey, more than 43%, 54%, and 31% of cancer survivors
medical teams to proactively discuss sexual health and younger than age 50 reported material, psychological, and
long-term post-treatment side effects for patients with EO- behavioral effects of medical financial hardship, respectively—
CRC. Communication and assessment and tools are rates that were higher than those of older cancer survivors
available.114-116 After discussions, interventions can be ex- and cancer-free counterparts.120 We must expand our vo-
plored, including long-term counseling, vaginal lubricants, cabulary to include financial toxicity as a real problem that
or topical estrogen for women and phosphodiesterase-5 patients face as a result of a cancer diagnosis and consider
inhibitors and testosterone replacement for men.117 cost when designing a treatment plan.
Fertility The pace of increasing financial toxicity is alarming, es-

pecially for young patients with CRC. Among young patients
Fluorouracil is the backbone of chemotherapy in the ad-
and survivors of cancer (age , 40), financial toxicity as
juvant and metastatic settings. It has been shown to reduce
measured through an 11-item distress survey was found to
sperm count temporarily and may also cause amenorrhea,
be associated with lower insurance satisfaction, more de-
although risk is low.118 For other treatments commonly used
pression and anxiety symptoms, and lower coping with
in CRC, such as oxaliplatin, irinotecan, and anti-EGFR and
cancer. Furthermore, financial toxicity leads to skipping or
-VEGF therapy, effects on fertility are largely unknown.
delaying treatment in multivariable modeling.121 Patients at
However, radiation can result in decreased or eradicated
the highest risk for financial toxicity are those in the lowest
fertility and early menopause if the radiation field contains
income quartile and those who undergo emergency surgery,
the ovaries or uterus, and it can result in decreased male
are black or Hispanic, and undergo surgery for esophageal
fertility through prostatic and gonadal radiation.118
or colon cancer.122
Pregnancy
Many young patients are juggling competing financial pri-
CRC can occur during pregnancy and may be a more orities and have the added stress of increasing out-of-pocket
common occurrence with delays in childbearing. An article medical costs during treatment. Among patients with stage
outlining CRC systemic treatments and their potential risk III colon cancer undergoing adjuvant chemotherapy, stud-
during pregnancy listed each as having either a C or D ies find that younger age and lower household income are
pregnancy risk category. Category C is defined as treat- strongly associated with financial hardship, and 40% of the
ments for which animal studies have shown adverse effects patients accessed money from savings accounts during
on the fetus and for which there are no adequate studies in treatment.123,124 In addition to direct monetary costs, the
humans. Category D is defined as treatments for which there burden includes patient time costs, including time receiving
is evidence of human fetal risk according to data from in- care rather than working or engaging in other activities.
vestigational or marketing experience or studies in humans. Depending on the type of cancer and phase of care, patient
For both categories, the benefits of the drug must be bal- time costs range from hundreds to many thousands of
anced against risks. Select cases and outcomes of patients dollars per year. In addition to direct costs, patients may also
and infants were described.119 Important recommendations face indirect costs through lost wages and lower earning
included avoiding systemic agents as much as possible potential and may experience job loss or job lock (i.e., being
during the first trimester and avoiding targeted agents al- unable to change jobs) because of concerns about health
together because of the lack of studies in pregnant patients. insurance coverage.125
Both fluorouracil and oxaliplatin have been given in the
second and third trimester, but the risks and benefits for the We must seek opportunities to openly discuss and address
mother must be weighed carefully, with close involvement of financial hardship for patients and their families. Further-
high-risk maternal-fetal medicine. more, we need policy makers to explore new strategies
to lessen the economic impact of new CRC therapies,
Financial Toxicity such as easing restrictions on the federal government’s
The topic of drug costs and comprehensive cancer care is ability to negotiate drug prices and asking drug developers
not often raised with patients in their doctors’ offices or to reassess pricing policies. Evidence-based interven-
elsewhere. Patients with and without insurance are facing tions and patient assistance programs can provide real and

substantive support for families. It is therefore paramount covering a variety of topics in addition to disease and
that advocacy organizations, such as Fight Colorectal treatment information, such as patient stories related to
Cancer, the American Cancer Society’s Action Network, the dating, family, and sexuality.133 Advocacy organizations also
Prevent Cancer Foundation, and Friends of Cancer Re- provide resource libraries with information about clinical
search, engage in a national dialogue with all stakeholders, trials and financial and insurance assistance. Examples of
including health care providers and systems, payers, and advocacy groups with a strong social media presence
patients, to tackle together the unique challenges faced by include COLONTOWN, the Colon Club, Colon Cancer
patients with EO-CRC. Coalition, Fight Colorectal Cancer, the Colon Cancer
Survivorship Foundation, Colorectal Cancer Alliance, and Michael’s
Mission. COLONTOWN is an online community of patients
Fortunately, 5 years after curative treatment, the most and survivors with different “neighborhoods,” including
common cause of death for EO-CRC survivors is the same as Youngstown (patients younger than age 40), Poker Club
for any other person of that age.126 Over time, secondary (male patients only), Tough Chicks (women only), and PTA
cancers become a more common cause of death and the (patients with young children).134 The Colon Club features
eventual leading cause of death 11 to 15 years after a magazine highlighting survivors younger than age 50 and
treatment. Cardiovascular disease also becomes an in- their caregivers, called “On the Rise.”135 Large advocacy
creasingly important cause of death further from treat- groups, such as the Colon Cancer Coalition, Colorectal
ment.126 Still, there is a large and growing population of Cancer Alliance, and Fight Colorectal Cancer, have website
patients with EO-CRC who are treated for early-stage dis- subsections devoted to EO-CRC awareness and symptoms
ease and survive long term. Survivors may experience along with patient testimonials.136-138 The Colon Cancer
chronic side effects after surgery, radiation, and chemo- Alliance regularly conducts and publishes results from
therapy that can last through the remainder of life. Persistent a survey of patients with EO-CRC, survivors, and their
symptoms may include fatigue, anxiety, sleep dysfunction, caregivers to better understand and support this growing
genitourinary problems, bone problems, psychological and population, including information on the most common
body image problems, long-term pain, bowel problems, and symptoms, delays in diagnosis, and QOL. The 2018 report,
neuropathy along with secondary cancer risk.127,128 based on information from more than 1,600 respondents in
One study comparing long-term symptoms (assessed at 38 countries, found that 80% of patients with EO-CRC had
a mean time of 10.8 years since cancer) by age found that children younger than age 18 at the time of diagnosis.139
younger survivors (age 18–50) had higher scores for There are also an increasing number of large scientific
anxiety, body image problems, abdominal and pelvic pain, meetings dedicated to EO-CRC, including the yearly Early
bloated feeling, hair loss, and embarrassment related to Age Onset Colorectal Cancer Summit, which is hosted by the
bowel movements.129 Quality of life (QOL) assessments of Colon Cancer Foundation in partnership with the Colon
CRC survivors reveal high health-related and global QOL Cancer Coalition.140
scores and are similar to those of older age groups (, 60
vs. . 70). However, young survivors tend to report lower CONCLUSIONS
scores in social functioning. QOL has been found to be EO-CRC is an increasing public health problem, with major
most affected by higher residual symptom burden, rectal ramifications for patients and their families. Approximately
cancer, lower education level, and ostomy presence. In 30% of rectal cancer is diagnosed in patients younger than
one cohort, higher QOL was associated with improved age 55, and recent work showing a surge of CRC diagnosed
survival, and the highest-scoring patients had the lowest at age 50 (when screening begins) compared with age
all-cause mortality.130,131 However, there is a paucity of 49 suggests strongly that patients younger than age 50
research focusing on survivorship issues specifically in already are at increased risk of having precancerous
patients with EO-CRC. Considerations include surveillance polyps and cancer. The American Cancer Society has
and radiation exposure, recommended lifestyle changes, taken leadership on this issue and recommends that
and preventive modalities for patients with EO-CRC as well screening begin at age 45 for the average-risk population,
as the mental and social implications of survivorship and a qualified recommendation only because there is a lack of
the impact on family.132 This is a large unmet need for information on screening for patients younger than age 50
which funding should be prioritized nationally. to confirm efficacy. Other national guideline bodies simi-
Role of Advocacy Groups and Social Media larly should reassess recent data in their calculations and
Advocacy groups provide vital information for young men recommendations.
and women with CRC, along with their family and friends. There are unique clinical challenges for patients with EO-
These efforts often leverage survivor stories to raise CRC. Diagnostic delays result from low indices of suspicion
awareness and encourage education about EO-CRC, from primary providers, financial toxicity for patients who are

Siegel et al
in the prime of their earning potential, sexual and fertility for this particular CRC population. Together, these
considerations, and long-term survivorship concerns. steps will allow us to work together as a medical
Focused support for patients with EO-CRC is needed community to improve outcomes for our patients with
during their therapy and survivorship, as well as edu- EO-CRC.
cation for primary providers and oncologists on diag-
nosing and caring for survivors over the long term. Finally, ACKNOWLEDGMENT
we need more research exploring the etiology of this Rebecca L. Siegel, MPH, and Christopher Dennis Jaku-
entity so that we can improve treatment and management bowski, MD, served as co-first authors.

1
American Cancer Society, Atlanta, GA Nilofer S. Azad, MD, 1450 Orleans St., Room 4M10, Baltimore, MD
2
Johns Hopkins Medicine Sidney Kimmel Comprehensive Cancer Center, 21287; email: [email protected].
Baltimore, MD
3
Fight Colorectal Cancer, Springfield, MO
EDBK_279901.
REFERENCES
1. O’Connell JB, Maggard MA, Liu JH, et al. Rates of colon and rectal cancers are increasing in young adults. Am Surg. 2003;69:866-872.
2. Siegel RL, Jemal A, Ward EM. Increase in incidence of colorectal cancer among young men and women in the United States. Cancer Epidemiol Biomarkers
Prev. 2009;18:1695-1698.
3. Austin H, Henley SJ, King J, et al. Changes in colorectal cancer incidence rates in young and older adults in the United States: what does it tell us about
screening. Cancer Causes Control. 2014;25:191-201.
4. Bailey CE, Hu CY, You YN, et al. Increasing disparities in the age-related incidences of colon and rectal cancers in the United States, 1975-2010. JAMA Surg.
2015;150:17-22.
5. Siegel RL, Fedewa SA, Anderson WF, et al. Colorectal cancer incidence patterns in the United States, 1974–2013. J Natl Cancer Inst. 2017;109:djw322.
6. Siegel RL, Medhanie GA, Fedewa SA, et al. State variation in early-onset colorectal cancer in the United States, 1995–2015. J Natl Cancer Inst. 2019;
111:1104-1106.
7. Meester RGS, Mannalithara A, Lansdorp-Vogelaar I, et al. Trends in incidence and stage at diagnosis of colorectal cancer in adults aged 40 through 49 years,
1975–2015. JAMA. 2019;321:1933-1934.
8. Araghi M, Soerjomataram I, Bardot A, et al. Changes in colorectal cancer incidence in seven high-income countries: a population-based study. Lancet
9. Siegel RL, Torre LA, Soerjomataram I, et al. Global patterns and trends in colorectal cancer incidence in young adults. Gut. 2019;68:2179-2185.
10. Vuik FE, Nieuwenburg SA, Bardou M, et al. Increasing incidence of colorectal cancer in young adults in Europe over the last 25 years. Gut. 2019;68:1820-1826.
11. Brenner DR, Ruan Y, Shaw E, et al. Increasing colorectal cancer incidence trends among younger adults in Canada. Prev Med. 2017;105:345-349.
12. Siegel RL, Miller KD, Goding Sauer A, et al. Colorectal cancer statistics, 2020. CA Cancer J Clin. 2020;70:7-30.
13. Siegel RL, Miller KD, Jemal A. Colorectal cancer mortality rates in adults aged 20 to 54 years in the United States, 1970–2014. JAMA. 2017;318:572-574.
14. SEER*Stat Database. November 2018 Submissions: Rate Sessions—Incidence - SEER 9 Regs Research Data with Delay Adjustment, Malignant Only, Nov
2018 Sub (1975-2016) ,Katrina/Rita Population Adjustment.. https://seer.cancer.gov/data-software/documentation/seerstat/nov2018/.
15. Xue K, Li FF, Chen YW, et al. Body mass index and the risk of cancer in women compared with men: a meta-analysis of prospective cohort studies. Eur J Cancer
Prev. 2017;26:94-105.
16. Carter BD, Abnet CC, Feskanich D, et al. Smoking and mortality: beyond established causes. N Engl J Med. 2015;372:631-640.
17. McNabb S, Harrison TA, Albanes D, et al. Meta-analysis of 16 studies of the association of alcohol with colorectal cancer. Int J Cancer. 2020;146:861-873.
18. Vieira AR, Abar L, Chan DSM, et al. Foods and beverages and colorectal cancer risk: a systematic review and meta-analysis of cohort studies, an update of the
evidence of the WCRF-AICR Continuous Update Project. Ann Oncol. 2017;28:1788-1802.
19. Boyle T, Keegel T, Bull F, et al. Physical activity and risks of proximal and distal colon cancers: a systematic review and meta-analysis. J Natl Cancer Inst. 2012;
104:1548-1561.
20. Liu PH, Wu K, Ng K, et al. Association of obesity with risk of early-onset colorectal cancer among women. JAMA Oncol. 2019;5:37-44.

21. Nguyen LH, Liu PH, Zheng X, et al. Sedentary behaviors, TV viewing time, and risk of young-onset colorectal cancer. JNCI Cancer Spectr. 2018;2:pky073.
22. Rosato V, Bosetti C, Levi F, et al. Risk factors for young-onset colorectal cancer. Cancer Causes Control. 2013;24:335-341.
23. Gausman V, Dornblaser D, Anand S, et al. Risk factors associated with early-onset colorectal cancer. Clin Gastroenterol Hepatol. 2019;S1542-3565-5.
24. Willauer AN, Liu Y, Pereira AAL, et al. Clinical and molecular characterization of early-onset colorectal cancer. Cancer. 2019;125:2002-2010.
25. Nimptsch K, Wu K. Is timing important? The role of diet and lifestyle during early life on colorectal neoplasia. Curr Colorectal Cancer Rep. 2018;14:1-11.
26. Dik VK, van Oijen MGH, Smeets HM, et al. Frequent use of antibiotics is associated with colorectal cancer risk: results of a nested case-control study. Dig Dis Sci.
2016;61:255-264.
27. Cao Y, Wu K, Mehta R, et al. Long-term use of antibiotics and risk of colorectal adenoma. Gut. 2018;67:672-678.
28. Goncalves MD, Lu C, Tutnauer J, et al. High-fructose corn syrup enhances intestinal tumor growth in mice. Science. 2019;363:1345-1349.
29. O’Keefe SJ. Diet, microorganisms and their metabolites, and colon cancer. Nat Rev Gastroenterol Hepatol. 2016;13:691-706.
30. Kwong TNY, Wang X, Nakatsu G, et al. Association between bacteremia from specific microbes and subsequent diagnosis of colorectal cancer. Gastro-
enterology. 2018;155:383-390 e388.
31. Scott KP, Gratz SW, Sheridan PO, et al. The influence of diet on the gut microbiota. Pharmacol Res. 2013;69:52-60.
32. Lowery JT, Ahnen DJ, Schroy PC III, et al. Understanding the contribution of family history to colorectal cancer risk and its clinical implications: a state-of-the-
science review. Cancer. 2016;122:2633-2645.
33. Samadder NJ, Smith KR, Hanson H, et al. Increased risk of colorectal cancer among family members of all ages, regardless of age of index case at diagnosis. Clin
34. Tuohy TM, Rowe KG, Mineau GP, et al. Risk of colorectal cancer and adenomas in the families of patients with adenomas: a population-based study in Utah.
Cancer. 2014;120:35-42.
35. Chen FW, Sundaram V, Chew TA, et al. Low prevalence of criteria for early screening in young-onset colorectal cancer. Am J Prev Med. 2017;53:933-934.
36. Pearlman R, Frankel WL, Swanson B, et al; Ohio Colorectal Cancer Prevention Initiative Study Group. Prevalence and spectrum of germline cancer susceptibility
gene mutations among patients with early-onset colorectal cancer. JAMA Oncol. 2017;3:464-471.
37. Win AK, Lindor NM, Young JP, et al. Risks of primary extracolonic cancers following colorectal cancer in lynch syndrome. J Natl Cancer Inst. 2012;
104:1363-1372.
38. Møller P, Seppälä T, Bernstein I, et al; Mallorca Group. Cancer incidence and survival in Lynch syndrome patients receiving colonoscopic and gynaecological
surveillance: first report from the prospective Lynch syndrome database. Gut. 2017;66:464-472.
39. Win AK, Jenkins MA, Dowty JG, et al. Prevalence and penetrance of major genes and polygenes for colorectal cancer. Cancer Epidemiol Biomarkers Prev. 2017;
26:404-412.
40. Gupta S, Provenzale D, Llor X, et al; CGC. NCCN guidelines insights: genetic/familial high-risk assessment: colorectal, version 2.2019. J Natl Compr Canc Netw.
2019;17:1032-1041.
Pathology, College of American Pathologists, Association for Molecular Pathology, and the American Society of Clinical Oncology. J Clin Oncol. 2017;
35:1453-1486.
42. Cohen SA, Laurino M, Bowen DJ, et al. Initiation of universal tumor screening for Lynch syndrome in colorectal cancer patients as a model for the imple-
mentation of genetic information into clinical oncology practice. Cancer. 2016;122:393-401.
43. Green RF, Ari M, Kolor K, et al. Evaluating the role of public health in implementation of genomics-related recommendations: a case study of hereditary cancers
using the CDC Science Impact Framework. Genet Med. 2019;21:28-37.
44. Flynn BS, Wood ME, Ashikaga T, et al. Primary care physicians’ use of family history for cancer risk assessment. BMC Fam Pract. 2010;11:45.
45. Wood ME, Kadlubek P, Pham TH, et al. Quality of cancer family history and referral for genetic counseling and testing among oncology practices: a pilot test of
quality measures as part of the American Society of Clinical Oncology Quality Oncology Practice Initiative. J Clin Oncol. 2014;32:824-829.
46. SEER*Stat Database. Mortality - All COD, Aggregated With State, Total U.S. (1969–2017), National Cancer Institute, DCCPS, Surveillance Research Program,
Cancer Statistics Branch, released 2019. Underlying mortality data provided by NCHS. https://seer.cancer.gov/mortality/.
47. SEER-Stat Database. NAACCR Incidence Data - CiNA Analytic File, 1995–2016, Public Use (which includes data from CDC’s National Program of Cancer
Registries (NPCR), CCCR’s Provincial and Territorial Registries, and the NCI’s Surveillance, Epidemiology and End Results (SEER) Registries), certified by the
North American Association of Central Cancer Registries (NAACCR) as meeting high-quality incidence data standards for the specified time periods, submitted
December 2018.
48. DevCan. Probability of Developing or Dying of Cancer Software, Version 6.7.7. Surveillance Research Program, Statistical Methodology and Applications,
National Cancer Institute, 2019. Bethesda, MD: National Cancer Institute; 2019.
49. Bibbins-Domingo K, Grossman DC, Curry SJ, et al; U.S. Preventive Services Task Force. Screening for colorectal cancer: U.S. Preventive Services Task Force
Recommendation Statement. JAMA. 2016;315:2564-2575.
50. Provenzale D, Gupta S, Ahnen DJ, et al. NCCN guidelines insights: colorectal cancer screening, version 1.2018. J Natl Compr Canc Netw. 2018;16:939-949.
51. Rex DK, Boland CR, Dominitz JA, et al. Colorectal cancer screening: recommendations for physicians and patients from the U.S. Multi-Society Task Force on
Colorectal Cancer. Am J Gastroenterol. 2017;112:1016-1030.

Siegel et al
52. Wolf AMD, Fontham ETH, Church TR, et al. Colorectal cancer screening for average-risk adults: 2018 guideline update from the American Cancer Society. CA
Cancer J Clin. 2018;68:250-281.
53. Anderson JC, Samadder JN. To screen or not to screen adults 45–49 years of age: that is the question. Am J Gastroenterol. 2018;113:1750-1753.
54. Imperiale TF, Kahi CJ, Rex DK. Lowering the starting age for colorectal cancer screening to 45 years: who will come . . . and should they? Clin Gastroenterol
Hepatol. 2018;16:1541-1544.
55. Knudsen AB, Zauber AG, Rutter CM, et al. Estimation of benefits, burden, and harms of colorectal cancer screening strategies: modeling study for the U.S.
Preventive Services Task Force. JAMA. 2016;315:2595-2609.
56. Lieberman DA, Williams JL, Holub JL, et al. Race, ethnicity, and sex affect risk for polyps .9 mm in average-risk individuals. Gastroenterology. 2014;
147:351-358, NaN-e15.
57. Rundle AG, Lebwohl B, Vogel R, et al. Colonoscopic screening in average-risk individuals ages 40 to 49 vs 50 to 59 years. Gastroenterology. 2008;
134:1311-1315.
58. Thoma MN, Castro F, Golawala M, et al. Detection of colorectal neoplasia by colonoscopy in average-risk patients age 40–49 versus 50-59 years. Dig Dis Sci.
2011;56:1503-1508.
59. Abualkhair WH, Zhou M, Ahnen D, et al. Trends in incidence of early-onset colorectal cancer in the United States among those approaching screening age.
JAMA Netw Open. 2020;3:e1920407.
60. Ladabaum U, Mannalithara A, Meester RGS, et al. Cost-effectiveness and national effects of initiating colorectal cancer screening for average-risk persons at age
45 years instead of 50 years. Gastroenterology. 2019;157:137-148.
61. Azad NS, Leeds IL, Wanjau W, et al. Cost-utility of colorectal cancer screening at 40 years old for average-risk patients. Prev Med. 2020;133:106003.
62. Alaska Native Medical Center. Alaska Native Medical Center Colorectal Cancer Screening Guidelines. Anchorage, AK: Alaska Native Medical Center; 2013.
63. Tsai MH, Xirasagar S, Li YJ, et al. Colonoscopy screening among U.S. adults aged 40 or older with a family history of colorectal cancer. Prev Chronic Dis. 2015;
12:E80.
64. Fedewa SA, Siegel RL, Jemal A. Are temporal trends in colonoscopy among young adults concordant with colorectal cancer incidence? J Med Screen. 2019;
26:179-185.
65. Murphy CC, Lund JL, Sandler RS. Young-onset colorectal cancer: earlier diagnoses or increasing disease burden? Gastroenterology. 2017;152:1809-1812.
66. Chen FW, Sundaram V, Chew TA, et al. Advanced-stage colorectal cancer in persons younger than 50 years not associated with longer duration of symptoms or
time to diagnosis. Clin Gastroenterol Hepatol. 2017;15:728-737 e723.
67. Qaseem A, Crandall CJ, Mustafa RA, et al; Clinical Guidelines Committee of the American College of Physicians. Screening for colorectal cancer in asymptomatic
average-risk adults: a guidance statement from the American College of Physicians. Ann Intern Med. 2019;171:643-654.
68. Peterse EFP, Meester RGS, Siegel RL, et al. The impact of the rising colorectal cancer incidence in young adults on the optimal age to start screening:
microsimulation analysis I to inform the American Cancer Society colorectal cancer screening guideline. Cancer. 2018;124:2964-2973.
69. Mannucci A, Zuppardo RA, Rosati R, et al. Colorectal cancer screening from 45 years of age: thesis, antithesis and synthesis. World J Gastroenterol. 2019;
25:2565-2580.
70. Rex DK, Boland CR, Dominitz JA, et al. Colorectal cancer screening: recommendations for physicians and patients from the U.S. Multi-Society Task Force on
Colorectal Cancer. Gastroenterology. 2017;153:307-323.
71. Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell. 1990;61:759-767.
72. Terzić J, Grivennikov S, Karin E, et al. Inflammation and colon cancer. Gastroenterology. 2010;138:2101-2114.e5.
73. National Comprehensive Cancer Network. Clinical Practice Guidelines in Oncology: Colorectal Cancer Screening, v. 2.2019. https://www.nccn.org/
professionals/physician_gls/pdf/colorectal_screening.pdf. Accessed February 4, 2020.
74. Molmenti CL, Kolb JM, Karlitz JJ. Advanced colorectal polyps on colonoscopy: a trigger for earlier screening of family members. Am J Gastroenterol. 2020;
115:311-314.
75. National Colorectal Cancer Round Table. Advanced Colorectal Polyp Brief. https://nccrt.org/resource/advanced-colorectal-polyp-brief/. Accessed February 8,
2020.
76. Schroy PC III, Glick JT, Wilson S, et al. An effective educational strategy for improving knowledge, risk perception, and risk communication among colorectal
adenoma patients. J Clin Gastroenterol. 2008;42:708-714.
77. Levin B, Lieberman DA, McFarland B, et al; American College of Radiology Colon Cancer Committee. Screening and surveillance for the early detection of
colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the U.S. Multi-Society Task Force on Colorectal Cancer,
and the American College of Radiology. Gastroenterology. 2008;134:1570-1595.
78. Lin JS, Piper MA, Perdue LA, et al. Screening for colorectal cancer: updated evidence report and systematic review for the U.S. Preventive Services Task Force.
JAMA. 2016;315:2576-2594.
79. Hampel H, Frankel WL, Martin E, et al. Screening for the Lynch syndrome (hereditary nonpolyposis colorectal cancer). N Engl J Med. 2005;352:1851-1860.
80. Segev L, Kalady MF, Church JM. Left-sided dominance of early-onset colorectal cancers: a rationale for screening flexible sigmoidoscopy in the young. Dis Colon
Rectum. 2018;61:897-902.
81. Scott RB, Rangel LE, Osler TM, et al. Rectal cancer in patients under the age of 50 years: the delayed diagnosis. Am J Surg. 2016;211:1014-1018.

82. Yarden RI, Newcomer KL; Never Too Young Advisory Board, Colorectal Cancer Alliance. Young onset colorectal cancer patients are diagnosed with advanced
disease after multiple misdiagnoses. https://www.abstractsonline.com/pp8/#!/6812/presentation/7708.
83. Dozois EJ, Boardman LA, Suwanthanma W, et al. Young-onset colorectal cancer in patients with no known genetic predisposition: can we increase early
recognition and improve outcome? Medicine (Baltimore). 2008;87:259-263.
84. O’Connell JB, Maggard MA, Livingston EH, et al. Colorectal cancer in the young. Am J Surg. 2004;187:343-348.
85. Mitchell E, Macdonald S, Campbell NC, et al. Influences on pre-hospital delay in the diagnosis of colorectal cancer: a systematic review. Br J Cancer. 2008;
98:60-70.
86. You YN, Xing Y, Feig BW, et al. Young-onset colorectal cancer: is it time to pay attention? Arch Intern Med. 2012;172:287-289.
87. Sandhu GS, Anders R, Walde A, et al. High incidence of advanced stage cancer and prolonged rectal bleeding history before diagnosis in young-onset patients
with colorectal cancer. J Clin Oncol. 2019;37:3576.
88. Bleyer A. CAUTION! Consider cancer: common symptoms and signs for early detection of cancer in young adults. Semin Oncol. 2009;36:207-212.
89. Wilhelm S, Carter C, Lynch M, et al. Discovery and development of sorafenib: a multikinase inhibitor for treating cancer. Nat Rev Drug Discov. 2006;5:835-844.
90. Liang J, Church J. How to manage the patient with early-age-of-onset (,50 years) colorectal cancer? Surg Oncol Clin N Am. 2010;19:725-731.
91. Pasha SF, Shergill A, Acosta RD, et al; ASGE Standards of Practice Committee. The role of endoscopy in the patient with lower GI bleeding. Gastrointest Endosc.
2014;79:875-885.
92. Siminoff L, Thomson M, Dumenci L. Factors associated with delayed patient appraisal of colorectal cancer symptoms. Psychooncology. 2014;23:981-988.
93. Abdelsattar ZM, Wong SL, Regenbogen SE, et al. Colorectal cancer outcomes and treatment patterns in patients too young for average-risk screening. Cancer.
2016;122:929-934.
94. Mauri G, Sartore-Bianchi A, Russo A-G, et al. Early-onset colorectal cancer in young individuals. Mol Oncol. 2019;13:109-131.
95. Blanke CD, Bot BM, Thomas DM, et al. Impact of young age on treatment efficacy and safety in advanced colorectal cancer: a pooled analysis of patients from
nine first-line phase III chemotherapy trials. J Clin Oncol. 2011;29:2781-2786.
96. Lieu CH, Renfro LA, de Gramont A, et al; Aide et Recherche en Cancérologie Digestive Foundation. Association of age with survival in patients with metastatic
colorectal cancer: analysis from the ARCAD Clinical Trials Program. J Clin Oncol. 2014;32:2975-2984.
97. Alberts SR, Sargent DJ, Nair S, et al. Effect of oxaliplatin, fluorouracil, and leucovorin with or without cetuximab on survival among patients with resected stage III
colon cancer: a randomized trial. JAMA. 2012;307:1383-1393.
98. Allegra CJ, Yothers G, O’Connell MJ, et al. Phase III trial assessing bevacizumab in stages II and III carcinoma of the colon: results of NSABP protocol C-08. J Clin
Oncol. 2011;29:11-16.
99. de Gramont A, Van Cutsem E, Schmoll H-J, et al. Bevacizumab plus oxaliplatin-based chemotherapy as adjuvant treatment for colon cancer (AVANT): a phase 3
randomised controlled trial. Lancet Oncol. 2012;13:1225-1233.
100. Khan SA, Morris M, Idrees K, et al. Colorectal cancer in the very young: a comparative study of tumor markers, pathology and survival in early onset and adult
onset patients. J Pediatr Surg. 2016;51:1812-1817.
101. Kneuertz PJ, Chang GJ, Hu C-Y, et al. Overtreatment of young adults with colon cancer: more intense treatments with unmatched survival gains. JAMA Surg.
2015;150:402-409.
102. Taieb J, Balogoun R, Le Malicot K, et al; PETACC8 Investigators. Adjuvant FOLFOX +/- cetuximab in full RAS and BRAF wildtype stage III colon cancer patients.
Ann Oncol. 2017;28:824-830.
103. Quah HM, Joseph R, Schrag D, et al. Young age influences treatment but not outcome of colon cancer. Ann Surg Oncol. 2007;14:2759-2765.
104. Kolarich A, George TJ Jr., Hughes SJ, et al. Rectal cancer patients younger than 50 years lack a survival benefit from NCCN guideline-directed treatment for
stage II and III disease. Cancer. 2018;124:3510-3519.
105. Hofheinz RD, Arnold D, Fokas E, et al; German Rectal Cancer Study Group. Impact of age on the efficacy of oxaliplatin in the preoperative chemoradiotherapy
and adjuvant chemotherapy of rectal cancer: a post hoc analysis of the CAO/ARO/AIO-04 phase III trial. Ann Oncol. 2018;29:1793-1799.
106. Perea J, Arriba M, Rodrı́guez Y, et al. Frequency and impact of KRAS mutation in early onset colorectal cancer. Hum Pathol. 2017;61:221-222.
107. Watson R, Liu T-C, Ruzinova MB. High frequency of KRAS mutation in early onset colorectal adenocarcinoma: implications for pathogenesis. Hum Pathol.
2016;56:163-170.
108. Kirzin S, Marisa L, Guimbaud R, et al. Sporadic early-onset colorectal cancer is a specific sub-type of cancer: a morphological, molecular and genetics study.
PLoS One. 2014;9:e103159.
109. Almont T, Bouhnik A-D, Ben Charif A, et al. Sexual health problems and discussion in colorectal cancer patients two years after diagnosis: a national cross-
sectional study. J Sex Med. 2019;16:96-110.
110. Reese JB, Handorf E, Haythornthwaite JA. Sexual quality of life, body image distress, and psychosocial outcomes in colorectal cancer: a longitudinal study.
Support Care Cancer. 2018;26:3431-3440.
111. Den Oudsten BL, Traa MJ, Thong MSY, et al. Higher prevalence of sexual dysfunction in colon and rectal cancer survivors compared with the normative
population: a population-based study. Eur J Cancer. 2012;48:3161-3170.
112. Tekkis PP, Cornish JA, Remzi FH, et al. Measuring sexual and urinary outcomes in women after rectal cancer excision. Dis Colon Rectum. 2009;52:46-54.

Siegel et al
113. Reese JB, Finan PH, Haythornthwaite JA, et al. Gastrointestinal ostomies and sexual outcomes: a comparison of colorectal cancer patients by ostomy status.
114. Althof SE, Parish SJ. Clinical interviewing techniques and sexuality questionnaires for male and female cancer patients. J Sex Med. 2013;10:35-42.
115. Dowswell G, Ismail T, Greenfield S, et al. Men’s experience of erectile dysfunction after treatment for colorectal cancer: qualitative interview study. BMJ. 2011;
343:d5824.
116. Traa MJ, De Vries J, Roukema JA, et al. The sexual health care needs after colorectal cancer: the view of patients, partners, and health care professionals.
117. El-Shami K, Oeffinger KC, Erb NL, et al. American Cancer Society colorectal cancer survivorship care guidelines. CA Cancer J Clin. 2015;65:428-455.
118. Lee SJ, Schover LR, Partridge AH, et al; American Society of Clinical Oncology. American Society of Clinical Oncology recommendations on fertility preservation
in cancer patients. J Clin Oncol. 2006;24:2917-2931.
119. Rogers JE, Dasari A, Eng C. The treatment of colorectal cancer during pregnancy: cytotoxic chemotherapy and targeted therapy challenges. Oncologist. 2016;
21:563-570.
120. Zheng Z, Jemal A, Han X, et al. Medical financial hardship among cancer survivors in the United States. Cancer. 2019;125:1737-1747.
121. Thom B, Benedict C. The impact of financial toxicity on psychological well-being, coping self-efficacy, and cost-coping behaviors in young adults with cancer.
J Adolesc Young Adult Oncol. 2019;8:236-242.
122. Farooq A, Merath K, Hyer JM, et al. Financial toxicity risk among adult patients undergoing cancer surgery in the United States: an analysis of the National
Inpatient Sample. J Surg Oncol. 2019;120:397-406.
123. Shankaran V, Jolly S, Blough D, et al. Risk factors for financial hardship in patients receiving adjuvant chemotherapy for colon cancer: a population-based
exploratory analysis. J Clin Oncol. 2012;30:1608-1614.
124. Veenstra CM, Regenbogen SE, Hawley ST, et al. A composite measure of personal financial burden among patients with stage III colorectal cancer. Med Care.
2014;52:957-962.
125. Mehnert A. Employment and work-related issues in cancer survivors. Crit Rev Oncol Hematol. 2011;77:109-130.
126. van Erning FN, van Steenbergen LN, Lemmens VEPP, et al. Conditional survival for long-term colorectal cancer survivors in the Netherlands: who do best? Eur
J Cancer. 2014;50:1731-1739.
127. Denlinger CS, Barsevick AM. The challenges of colorectal cancer survivorship. J Natl Compr Canc Netw. 2009;7:883-893, quiz 894.
128. O’Gorman C, Stack J, O’Ceilleachair A, et al. Colorectal cancer survivors: an investigation of symptom burden and influencing factors. BMC Cancer. 2018;
18:1022.
129. Bailey CE, Tran Cao HS, Hu C-Y, et al. Functional deficits and symptoms of long-term survivors of colorectal cancer treated by multimodality therapy differ by age
at diagnosis. J Gastrointest Surg. 2015;19:180-188.
130. Kunitake H, Russell MM, Zheng P, et al. Quality of life and symptoms in long-term survivors of colorectal cancer: results from NSABP protocol LTS-01. J Cancer
Surviv. 2017;11:111-118.
131. Ratjen I, Schafmayer C, Enderle J, et al. Health-related quality of life in long-term survivors of colorectal cancer and its association with all-cause mortality:
a German cohort study. BMC Cancer. 2018;18:1156.
132. Sharp L, Deady S, Gallagher P, et al. The magnitude and characteristics of the population of cancer survivors: using population-based estimates of cancer
prevalence to inform service planning for survivorship care. BMC Cancer. 2014;14:767.
133. Colon Club. Dating. http://colonclub.com/category/dating/. Accessed February 22, 2020.
134. Seybold N. COLONTOWN Neighborhoods. https://colontown.org/colontown-neighborhoods/. Accessed February 22, 2020.
135. Colon Club. On The Rise. https://colonclub.com/on-the-rise/. Accessed February 22, 2020.
136. Colon Cancer Coalition. Age Is Not a Factor. https://coloncancercoalition.org/get-educated/age-is-not-a-factor/. Accessed February 22, 2020.
137. Fight Colorectal Cancer. Young Adult Colorectal Cancer: On the Rise. https://fightcolorectalcancer.org/colorectal-cancer/young-adult-colorectal-cancer/.
138. Colorectal Cancer Alliance. Young Onset. https://www.ccalliance.org/colorectal-cancer-information/young-onset. Accessed February 22, 2020.
139. Colorectal Cancer Alliance. 2018 Young-Onset Colorectal Cancer Survey Report. ccalliance.org/about/never-too-young/survey/2018-young-onset-colorectal-
cancer-survey-report.
140. Colon Cancer Foundation. Early AOCCS. https://www.coloncancerfoundation.org/about/eao-crc/. Accessed February 22, 2020.

GASTROINTESTINAL CANCER—
GASTROESOPHAGEAL,
PANCREATIC, AND
HEPATOBILIARY

Overcoming Resistance to Targeted Therapies in

Gastrointestinal Cancers: Progress to Date and
Progress to Come
Christopher Chen, MD1; Maria Di Bartolomeo, MD2; Salvatore Corallo, MD2; John H. Strickler, MD3; and Lipika Goyal, MD, DPhil1
overview
Targeted therapies have transformed the treatment paradigm in diseases such as non–small cell lung cancer
and melanoma but have shown relatively modest clinical benefit in gastrointestinal malignancies. Anti-EGFR
therapy in RAS wild-type colorectal cancer, anti-HER2 therapy in HER2- amplified esophagogastric cancer,
and FGFR and isocitrate dehydrogenase 1 (IDH1) inhibitors in FGFR2 fusion-positive and IDH1-mutant biliary
tract cancers offer antitumor efficacy, but the clinical benefit and durability of response in each case are
typically limited. We review targeted therapies in each of these therapeutic areas and discuss their clinical
efficacy, mechanisms of primary and acquired resistance, and strategies to overcome this resistance. We
discuss lessons learned that we hope will lead to an expanded role for molecularly targeted therapy options for
patients with gastrointestinal cancers.
Molecularly targeted therapies in diseases such as stabilization or response. In this article, we review
non–small cell lung cancer and melanoma offer sig- targeted therapy strategies in GI cancers, including
nificantly higher response rates and, at times, dramatic anti-EGFR therapy in colorectal cancer, anti-HER2
antitumor responses compared with conventional therapy in gastric and gastroesophageal junction
chemotherapy. However, the benefits of a personalized cancer, and FGFR and IDH1 inhibitors in chol-
medicine approach in gastrointestinal (GI) malignan- angiocarcinoma. For each of these topics, we dis-
cies have been relatively modest. Of the 58 targeted cuss the clinical efficacy of the drug class, review
oncology drugs approved by the U.S. Food and mechanisms of primary and acquired resistance, and
Drug Administration (FDA) between 2006 and discuss strategies to overcome resistance mecha-
2018, only four were in GI cancers: imatinib for gas- nisms. We conclude by discussing lessons learned
trointestinal stromal tumors, panitumumab and from these disease areas that we hope will expand the
cetuximab for RAS wild-type colorectal cancer, and number of targeted therapy options for patients with
trastuzumab for HER2+ gastroesophageal cancer.1 GI cancer and the subsets of patients who may
Therapeutic strategies that have exhibited striking benefit.
efficacy in non-GI malignancies, such as BRAF in-
hibition in BRAFV600E mutated melanoma and HER2-
directed therapy in HER2+ breast cancer, have had TARGETING EGFR IN COLORECTAL CANCER
more limited benefit in BRAF-mutated colorectal
cancer and HER2+ esophagogastric tumors. Addi- The EGFR (ErbB-1) is a transmembrane receptor
tionally, the optimal treatment approach for patients tyrosine kinase (RTK) involved in cell proliferation,
Author affiliations angiogenesis, migration, survival, and adhesion.2
with acquired resistance to targeted therapy is gen-
and support Upon ligand binding, EGFR forms homo- or hetero-
information (if erally less clear in GI malignancies compared with
dimers with another member of the ErbB RTK su-
applicable) appear other diseases.
at the end of this
perfamily (ErbB2 [HER2], ErbB3 [HER3], or ErbB4
article. On the other hand, increased recognition and avail- [HER4]), leading to tyrosine phosphorylation and
Accepted on ability of tumor molecular profiling has expanded our activation of multiple downstream signaling cascades,
February 27, 2020 understanding of oncogenic mutations and molecu- including the RAS-RAF mitogen–activated protein
and published at lar mechanisms of resistance. Primary resistance is kinase (MAPK) pathway, the phosphoinositide 3-
ascopubs.org on defined as radiographic or clinical disease progres- kinase (PI3K) pathway, and the STAT pathway.3
May 18, 2020:
sion as the best response to an anticancer therapy. Due to its role in the growth and spread of meta-
10.1200/EDBK_ Acquired or secondary resistance is defined as static colorectal cancer (CRC), EGFR is an important
280871 therapeutic resistance after an initial period of disease therapeutic target.

Chen et al
signaling pathway. The most common RAS mutations

occur in codons 12 and 13 of KRAS exon 2 and are
established drivers of anti-EGFR resistance.11,12 Mutations
• The anti-EGFR antibodies cetuximab and in KRAS exons 3 and 4 and NRAS exons 2–4 occur less
panitumumab are standard treatments for RAS
commonly but have shown similar ability to induce primary
wild-type metastatic colorectal cancer, but
resistance to cetuximab and panitumumab.13,14 Given
nearly all patients eventually develop
resistance. Evidence suggests that some the role of RAS mutations in driving primary EGFR re-
patients who develop resistance to anti-EGFR sistance, current National Comprehensive Cancer Network
antibodies may benefit from a rechallenge. guidelines recommend that all patients with metastatic
CRC have tumor tissue genotyping for KRAS and NRAS
• The anti-HER2 antibody trastuzumab prolongs
survival when combined with chemotherapy for exon 2–4 mutations, and patients with a RAS mutation
first-line treatment of advanced should not be treated with an anti-EGFR monoclonal
HER2-amplified esophagogastric cancer, antibody.15
although phase III trials of other anti-HER2 BRAF V600E BRAF is downstream of EGFR and RAS in the
agents have not shown benefit in this MAPK signaling pathway. Approximately 5% to 10% of
population. Sensitivity to anti-HER2 drugs may
patients with metastatic CRC have a BRAFV600E- mutated
be greatest in the presence of high HER2
tumor.16 Mutually exclusive with RAS mutations, BRAFV600E
amplification and the absence of concomitant
oncogenic drivers. mutations cause constitutive activation of the MAPK path-
way, thereby rescuing colorectal tumors from EGFR
• FGFR and IDH1 inhibitors have emerged as key
inhibition.17-19 Patients with BRAFV600E mutated tumors
therapeutic strategies in FGFR2-fusion and
are unlikely to benefit from EGFR inhibitors as mono-
IDH1-mutant cholangiocarcinoma; after
progression on ATP-competitive FGFR therapy or in combination with chemotherapy, but they
inhibitors, FGFR2-fusion cholangiocarcinoma may benefit from anti-EGFR monoclonal antibodies when
may still respond to covalent inhibitors that can combined with BRAF and/or MEK inhibitors. In the phase II
overcome FGFR2 kinase domain mutations. SWOG 1406 study, the BRAF inhibitor vemurafenib in
• Advancing targeted therapies for gastrointestinal combination irinotecan and cetuximab provided superior
malignancies may require scientific advances progression-free survival (PFS) compared with irinotecan
on three fronts: (1) biomarker-driven and cetuximab.20 In another phase I/II trial, triple therapy
enrichment strategies, (2) improved selectivity with the BRAF inhibitor dabrafenib in combination with
and potency of candidate drugs, and (3) deeper panitumumab and the MEK inhibitor trametinib provided
understanding of tumor resistance a 21% response rate.21 More recently in the three-arm
mechanisms. phase III BEACON trial with 1:1:1 randomization, BRAF/
EGFR/MEK inhibition and BRAF/EGFR inhibition were com-
pared with EGFR inhibition plus investigator’s choice of
Anti-EGFR antibodies bind the extracellular domain of irinotecan or FOLFIRI. The triplet of encorafenib/cetuximab/
EGFR, thereby preventing ligand binding. Additionally, binimetinib (BRAF/EGFR/MEK inhibitors, respectively) dem-
anti-EGFR antibodies induce receptor endocytosis and onstrated superior median overall survival (OS) compared
degradation.4 Anti-EGFR antibodies may also induce antibody- with cetuximab and chemotherapy. Although the trial did not
dependent cell-mediated cytotoxicity, which supports an directly compare BRAF/EGFR/MEK inhibition with BRAF/
immunologic mechanism of action.5 The two anti-EGFR EGFR inhibition, the objective response rate (ORR) and
antibodies in clinical use for the treatment of metastatic median OS were numerically similar in these two arms.22
CRC are cetuximab and panitumumab. Although these Based on these results, current National Comprehensive
molecules are structurally distinct, they have equivalent Cancer Network guidelines recommend BRAFV600E tumor
efficacy.6 Both are active as monotherapy in the refractory mutation testing in all patients with metastatic CRC and
setting and in combination with cytotoxic chemotherapy in the combination of an anti-EGFR monoclonal antibody
the first line and later.7-10 with either dabrafenib and trametinib or encorafenib with
or without binimetinib for patients with tumors that test
Primary Resistance positive.
Primary resistance can rise from both genomic and non- HER2 (ERBB2) amplification HER2 amplification occurs in
genomic mechanisms. Genomic alterations are described approximately 3% of all patients with metastatic CRC and is
below. enriched in RAS and BRAF wild-type and left-sided co-
KRAS and NRAS (RAS) mutations Activating mutations in lorectal tumors.23,24 Cell line and patient-derived xenograft
KRAS or NRAS result in constitutive activation of the MAPK models have shown that amplification of HER2, a potent

Overcoming Resistance to Targeted Therapy in GI Cancers
driver of MAPK and PI3K signaling, leads to cetuximab alterations to influence sensitivity and resistance to EGFR-
resistance.25,26 Consistent with these models, several ret- directed therapy; in a pooled analysis from six ran-
rospective clinical studies have identified a strong as- domized trials, Arnold et al49 reported that patients with
sociation between HER2 amplification and anti-EGFR right-sided tumors had a worse prognosis and less
resistance.24,27 Nevertheless, select patients with HER2- benefit from anti-EGFR antibodies compared with pa-
amplified tumors may still benefit from anti-EGFR therapy, tients with left-sided tumors.
such as patients with tumors with low-level HER2 copy
Acquired Resistance
number gain or borderline expression by immunohisto-
chemistry (IHC).28 A randomized phase II trial is underway Nearly all patients with metastatic CRC who experience
comparing an anti-HER2 combination (trastuzumab and benefit with anti-EGFR monoclonal antibodies ultimately
pertuzumab) with cetuximab and irinotecan in patients experience disease progression. Postprogression tumor
with RAS wild-type/HER2+ metastatic CRC (SWOG 1613; biopsies can yield important insights about resistance
NCT03365882). mechanisms to these therapies, but they are often difficult to
obtain and may fail to capture interlesional and intralesional
Other genomic alterations Amplification of KRAS and MET,
genomic heterogeneity. Circulating tumor DNA (ctDNA)
each found in about 1% of CRC tumors, has been asso-
analysis from a peripheral blood draw is a complementary,
ciated with EGFR resistance in preclinical models but
noninvasive technology that can provide insight into
has not been clinically validated.29-32 Mutations in EGFR,
mechanisms of resistance and offer an opportunity for serial
PDGFRA, and MAP2K1; amplification of ALK, FGFR1, and
monitoring to study tumor evolution.50
FLT3; and fusions involving NTRK, ALK, and ROS1 have
also been associated with resistance to anti-EGFR anti- Blood-based molecular profiling of ctDNA reveals KRAS or
bodies, but insufficient evidence exists to withhold anti- NRAS mutations as the most common genomic alterations
EGFR therapy in patients with tumors with these molecular associated with acquired resistance to anti-EGFR therapy.51-54
features.33-35 PIK3CA encodes the catalytic subunit of RAS mutations often exist as rare clones prior to anti-EGFR
PI3K, an essential protein in the PI3K signaling pathway. antibody exposure but expand relative to RAS and BRAF wild-
PIK3CA mutations are found in 15% to 20% of patients type clones under selective pressure. Additionally, EGFR
with metastatic CRC and are often co-altered with other ectodomain mutations emerge under the selective pressure of
mutations, including KRAS mutations.35-37 Results from EGFR-directed therapy.55-57 These mutations disrupt cetux-
retrospective studies are inconclusive regarding the im- imab and/or panitumumab binding to EGFR, thereby con-
pact of PIK3CA mutations on EGFR resistance.37-39 It is ferring resistance.58 Additional genomic alterations associated
possible that only certain alterations (e.g., PIK3CA exon with acquired EGFR resistance include mutations in BRAF,
20) confer EGFR resistance. 37,40 Nonetheless, given the ERBB2, MAP2K1, and MET and copy number gain of ERBB2,
low frequency of PIK3CA exon 20 mutations, the con- KRAS, and MET.56 These alterations are not mutually exclu-
founding effect of concomitant chemotherapy, and co- sive; in most cases, multiple concomitant resistance-conferring
alterations in the PI3K and MAPK pathways, there is cur- genomic alterations are detectable in ctDNA.59 This heteroge-
rently insufficient evidence to exclude patients with these neity complicates efforts to reverse resistance.
PIK3CA mutations from anti-EGFR therapy. Similarly, ret- Nongenomic mechanisms also drive acquired resistance to
rospective studies have shown conflicting results regarding anti-EGFR therapy in metastatic CRC. A recent study has
the efficacy of anti-EGFR antibodies in patients with tumors shown that anti-EGFR antibody–resistant tumor cells have
that harbor PTEN loss, which can activate the PI3K decreased expression of genes critical to DNA repair,60 en-
pathway.41,42 hancing their mutability and generating a diverse range of
A number of nongenomic mechanisms of resistance have clones potentially capable of overcoming therapeutic pres-
also been implicated. One study of patients with KRAS sure. The tumor microenvironment may also play an im-
wild-type metastatic CRC treated with cetuximab found portant role: secretion of the EGFR ligands tumor growth
that low amphiregulin and low epiregulin expression cor- factor alpha (TGF-α) and amphiregulin is associated with
related with inferior PFS and survival.43 Additionally, ac- cetuximab resistance.61 Additionally, preclinical models have
tivation of alternative signaling pathways may rescue shown that increased secretion of the ligand hepatocyte
CRC tumors from EGFR inhibition; preclinical models show growth factor activates its receptor (c-MET) and induces
JAK/STAT signaling can drive EGFR resistance in both resistance to cetuximab.62 Finally, VEGF signaling may
head and neck cancer and lung cancer.44,45 Epithelial-to- drive resistance to anti-EGFR therapies. Elevated expres-
mesenchymal transition has also been implicated as sion of VEGF is associated with resistance to anti-EGFR
a cause of EGFR resistance, although clinical evidence in drugs, although combinations of anti-VEGF and anti-EGFR
patients with metastatic CRC is limited.46-48 A clinical phe- therapies have not consistently demonstrated therapeutic
notype, tumor sidedness, may interact with genomic benefit.63,64

Chen et al
Strategies to Overcome Resistance HER2+ gastric or GEJ cancers. Based on these results,
The complex and heterogeneous mechanisms of EGFR trastuzumab in combination with chemotherapy has been
resistance complicate a rational therapeutic strategy to approved by the FDA and the European Medicines Agency
pre-empt or overcome resistance. A strategy of admin- as first-line treatment of advanced gastric or GEJ cancer with
istering EGFR antibodies with MEK inhibitors has been HER2 overexpression/ERBB2 amplification.73,74 To date, no
proposed based on preclinical models, but clinical ex- additional anti-HER2 agents have been approved for the
perience remains limited.65,66 treatment of HER2+ gastric and GEJ cancer, including none
in the refractory or adjuvant setting. This is in contrast to
Loss of detection of secondary resistance mutations by HER2+ breast cancer, for which trastuzumab has been
ctDNA analysis may predict which patients may benefit approved by the FDA for adjuvant and metastatic disease
from an anti-EGFR antibody rechallenge. Siravegna et al53 and four additional anti-HER2 agents have been approved in
showed that acquired RAS mutations emerge under the the metastatic setting, including pertuzumab, trastuzumab-
selective pressure of anti-EGFR therapy but decline relative emtasine (T-DM1), fam-trastuzumab deruxtecan, and
to RAS wild-type clones after EGFR-directed therapy is lapatinib.
discontinued. Additional analysis from Parseghian et al67
has shown that the decay half-lives for EGFR ectodomain Primary Resistance
mutations and RAS mutations are 6.9 and 3.4 months, In the ToGA study, the benefits in overall response rate
respectively. Thus, ctDNA may identify patients who have (47% vs. 35%), PFS (6.7 vs. 5.5 months), and duration of
lost secondary resistance mutations. Retrospective analyses response (6.9 vs. 4.8 months) favoring the trastuzumab arm
of clinical studies evaluating anti-EGFR rechallenge have were numerically small, highlighting the considerable pro-
shown that clinical benefit is concentrated in patients who portion of patients with primary refractory disease. A post
previously responded to EGFR blockade and who have no hoc exploratory analysis showed that patients with tumors
detectable BRAF, EGFR, or RAS mutations in blood prior to that stained HER2 3+ by IHC or HER2 2+ and HER2
initiating rechallenge.68,69 Based on these findings, pro- amplified by fluorescence in situ hybridization (FISH) de-
spective studies are underway to evaluate whether ctDNA rived the greatest benefit from trastuzumab plus chemo-
can be used to identify patients appropriate for anti-EGFR therapy. Therefore, all major guidelines have recommended
rechallenge (NCT03227926; NCT03992456). HER2 testing through combined IHC and FISH analysis.
HER2 status should be evaluated by IHC first, followed by
TARGETING HER2 IN GASTRIC AND GASTROESOPHAGEAL FISH when the IHC result is 2+ (equivocal). Positive (3+) or
JUNCTION CANCERS negative (0 or 1+) HER2 IHC results do not require further
Identification of predictive biomarkers to guide therapeutic FISH testing.75
decisions in the management of gastric cancer and gas-
Despite this more sophisticated approach to HER2 testing,
troesophageal junction (GEJ) cancers has been an area of
a substantial proportion of HER2+ patients experience
great scientific effort over the last several years. The first
disease progression early in the course of treatment with
major achievement was the identification of HER2 over-
trastuzumab, thus revealing a primary resistance. Intra-
expression and HER2/neu (ERBB2) amplification in ap-
patient and interpatient heterogeneity in HER2 expression
proximately 20% of gastric and GEJ cancers, with variances
may be key drivers of primary resistance to trastuzumab.76
by tumor histologic subtype and location.70 HER2 belongs to
Studies have estimated that 23% to 29% of gastric and GEJ
the EGFR tyrosine kinase family, but, in contrast to EGFR
cancers have heterogeneous HER2 expression, which is
itself, the kinase activity of HER2 is ligand independent.
a higher rate than in breast adenocarcinomas.77 Multi-
HER2 acts as coreceptor after other EGFR family receptors
platform analyses have also revealed interpatient hetero-
are activated via ligand binding, leading to activation of the
geneity among gastric cancer78,79 and have shown that
canonical MAPK/PI3K signaling cascades. Therefore,
HER2 amplification may be coupled with other oncogenic
HER2 does not pivot between active and inactive con-
drivers, such as amplification of EGFR, MET, and KRAS and
formations but rather is constitutively active. The anti-
mutations in EGFR, MET, KRAS, HER3, PIK3CA, and
HER2 antibody trastuzumab acts by blocking the activity
PTEN.80,81 Some of these co-alterations have been shown to
of the HER2 receptor and weakening subsequent down-
confer resistance to anti-HER2 therapy in vitro, and, in
stream signaling.71
a small case series, trastuzumab resistance was success-
In 2010, Bang et al72 reported the results of the phase III fully reversed by combining blockade of HER2 and the
ToGA study, which demonstrated improvement in median secondary driver mutation.82 Moreover, paired analyses of
OS with trastuzumab combined with a fluoropyrimidine and primary tumor and metastasis show recurrent genomic
cisplatin compared with fluoropyrimidine and cisplatin biomarker discrepancies between primary and metastatic
alone in the first-line treatment of patients with advanced lesions.83

Investigation of trastuzumab primary resistance mech- oncogenes may require dual inhibition of HER2 and the
anisms in vivo is challenging given the heterogeneity and co-altered pathway.
multiplicity of the candidate genomic mechanisms.84 Ad-
ditionally, the use of concurrent chemotherapy can con- Acquired Resistance
found understanding of primary resistance mechanisms to Nearly all patients receiving benefit from anti-HER2 therapy
trastuzumab; for example, patients with defects in homol- for advanced disease eventually develop drug resistance
ogous recombination may have exquisite platinum sensi- and experience tumor progression. In this setting, pre- and
tivity with concurrent trastuzumab resistance, and the latter postprogression tumor biopsies can be sequenced to un-
may go unrecognized in the setting of clinical benefit to the cover mechanisms of acquired resistance. Case series
combined regimen. To overcome some of these limits, studies have identified HER2 loss as a potential driver of
Pietrantonio et al85 carried out a multicenter, prospective, acquired resistance to trastuzumab in 16% to 34% of
case-control study (AMNESIA-1) investigating biomarkers patients with gastric cancer based on such analysis.80,87
of primary resistance to trastuzumab in patients with Postprogression biopsies have also identified enrichment of
HER2+ metastatic gastric cancer. Tissue from 37 patients alterations in other tyrosine kinase receptors and in-
with HER2+ gastric cancer treated with trastuzumab, intracellular signaling nodes, including exon 16 deletion of the
cluding 17 patients with sensitive cancers and 20 patients HER2 gene and mutations in KRAS and PI3K pathways.86
with cancers with primary resistance, were tested against These results suggest that both heterogeneity of the HER2
a panel of candidate genomic alterations, including EGFR/ proteome itself as well as co-altered oncogenic drivers may
MET/KRAS/PI3K/PTEN mutations and EGFR/MET/KRAS limit long-term tumor response to trastuzumab.
amplifications. Panel alterations were significantly more
frequent in baseline samples of primary resistant tumors as Strategies to Overcome Resistance
compared with sensitive tumors (55 vs. 0%; p , .001) and in
Three main strategies have been tested to overcome pri-
HER2 IHC 2+ versus 3+ tumors (53.8% vs. 16.7%; p = .028).
mary and secondary resistance: (1) improving upon the
Moreover, the absence of any candidate alteration was as-
results of the ToGA trial through more robust HER2
sociated with a longer PFS (5.2 vs. 2.6 months; HR, 0.34; p =
blockade, (2) dual HER2 and EGFR blockade in the first-line
.001) and OS (16.1 vs. 7.6 months; HR, 0.38; p = .015). The
setting, and (3) restoring anti-HER2 sensitivity after pro-
predictive accuracy regarding treatment outcome of the
gression on trastuzumab-based therapy (Table 1). In terms
targeted panel and HER2 IHC combined was 84%.
of the first strategy, the JACOB study investigated the ad-
Consistent with these results, Janjigian et al86 recently dition of pertuzumab, a monoclonal antibody that blocks
reported the results of targeted genetic sequencing of 50 HER2 homodimerization, to trastuzumab and fluoropyr-
HER2+ tumors collected before treatment with first-line imidine/cisplatin in a randomized, placebo-controlled phase
trastuzumab. Patients with higher levels of ERBB2 am- III trial and found a numerically superior but not statistically
plification had a significantly longer PFS on trastuzumab, significant OS difference (17.5 vs. 14.2 months; HR, 0.84;
whereas those with co-alterations in RTK–RAS–PI3K/AKT p = .057).88 The phase III HELOISE study tested a higher
pathway genes had a significantly shorter PFS (24.3 vs. dose of trastuzumab (10 mg/kg every 3 weeks) versus the
8.4 months, respectively). Although further validation is standard schedule of an 8 mg/kg loading dose followed by
needed, these results suggest that HER2 sensitivity is 6 mg/kg every 3 weeks89 but failed to demonstrate any
greatest in tumors with high-level HER2 amplification clinical benefit to higher dosing. In terms of dual blockade of
without concomitant oncogenic drivers and that effec- HER2 and EGFR, the TRIO-LOGIC trial tested the multi-
tive treatment of HER2+ gastric cancer with co-altered kinase inhibitor lapatinib in combination with CAPOX
TABLE 1. Completed Phase III Trials With HER2-Inhibitor in Gastric and Gastroesophageal Cancer
Agent Trial/Line Selection Biomarker/Trial Design Results: mOS
71
Trastuzumab ToGA/1°-line HER2+: CT with or without trastuzumab Positive HR: 0.74; 13.8 vs. 11.1
91
Trastuzumab Eloise/1°-line HER2+ (bulkydisease): CT with trastuzumab HD vs. Negative HR: 1.24; 12.5 vs. 10.6
CT with trastuzumab SD
Trasuzumab/pertuzumab90 Jacob/1°-line HER2+: CT with trastuzumab-pertuzumab/CT with Negative HR: 0.84; 17.5 vs. 14.2
trastuzumab-placebo
Lapatinib92 TRIO-LOGiC/1°-line HER2+: CT with or without lapatinib Negative HR: 0.91; 12.2 vs. 10.5
Lapatinib93 Tytan/2°-line HER2+: paclitaxel with or without lapatinib Negative HR: 0.84; 11 vs. 8.9
T-DM194 Gasby/2°-line HER2+: T-DM1 vs. taxan Negative HR: 1.15; 7.9 vs. 8.6
Abbreviations: mOS, median overall survival; CT, chemotherapy; HD, high dose; SD, standard dose.

Chen et al
compared with CAPOX alone in the first-line setting of Third, novel therapeutic agents and targeted therapy com-
HER2+ gastroesophageal cancer90 and found no significant binations must be brought forward into the clinic. DS-8201,
improvement in median OS (12.2 vs. 10.5 months; HR, a novel antibody-drug conjugate with a topoisomerase I in-
0.91; p = .3492). In the second-line setting, the phase III hibitor, demonstrated remarkable activity in heavy pretreated
TyTAN trial tested the addition of lapatinib to weekly patients with gastric cancer in early trials96 and is currently
paclitaxel compared with paclitaxel alone as second-line undergoing phase II testing in patients with gastric and GEJ
treatment in HER2 FISH-positive Asian patients with ad- cancers who have progressed on trastuzumab. Fourth, given
vanced gastric cancer (most of whom were naive to anti-
the extensive heterogeneity seen in gastric and gastroesoph-
HER2 treatment).91 The study exhibited no significant ageal cancers, addressing HER2 resistance may require
difference in the mOS and PFS between the two arms, a more global approach whereby we target both the tumor
although both mOS and PFS were improved in HER2 IHC and the tumor microenvironment. Studies of immune
3+ patients treated with lapatinib plus paclitaxel compared checkpoint inhibitors combined with novel HER2-targeted
with paclitaxel. agents have demonstrated evidence of clinical activity;
Within the post-trastuzumab progression space, the GATSBY positive OS, PFS, and ORR results were reported using
trial investigated the role of T-DM1, an antibody-drug pembrolizumab with trastuzumab and chemotherapy as
conjugate, in HER2-positive patients with gastric cancer frontline treatment in HER2-positive gastric cancer.97 A
experiencing progressive disease during or after first- phase III randomized study comparing trastuzumab plus
line trastuzumab-containing therapy92 and found no im- chemotherapy with or without pembrolizumab is ongoing
(NCT03615326). Finally, margetuximab, a chimeric anti-
provement in median OS (7.9 vs. 8.6 months; HR, 1.15;
HER2 IgG1 antibody, is under evaluation in combination
one-sided p = .86). The utility of continuing trastuzumab
with pembrolizumab as second-line therapy.
therapy beyond progression is an active area of research,
based on the efficacy of this strategy in breast cancer and We believe that ongoing trials evaluating new anti-HER2
on some retrospective promising data in gastric cancer.93 approaches in HER2+ gastric and GEJ cancers, combined
The phase II T-ACT trial prospectively examined the effi- with a deeper understanding of HER2 resistance mecha-
cacy of the addition of trastuzumab to weekly paclitaxel nisms, will lead to prolonged and expanded benefit of HER2
compared with paclitaxel alone in patients who have ex- inhibition in patients with these cancers.
perienced disease progression during or after a trastuzumab- TARGETING FGFR AND IDH1 IN CHOLANGIOCARCINOMA
containing first-line therapy.94 The study did not show any
significant difference in PFS, OS, and ORR between the two FGFR
arms, although subgroup analysis showed a therapeutic The fibroblast growth factor (FGF) family consists of four
benefit to continuing trastuzumab in patients who had a PFS transmembrane tyrosine kinase receptors (FGFR1–4) and
of at least 6 months to first-line trastuzumab-based therapy. 22 distinct FGF ligands; FGFR activation by FGFs result in
Despite the studies above being negative, they suggest that a cascade of intracellular events, including downstream
a subset of patients may benefit from more intensive HER2- activation of the RAS/RAF/MEK, JAK/STAT, and PI3K/AKT
directed therapy in the first line and continued HER2- pathways that drive a wide array of biologic processes from
directed therapy in the second line. embryonal development to wound repair.98 In cancer,
deregulation of the FGFR signaling axis via amplification,
We propose four strategies for addressing HER2 resistance fusions, or activating mutations leads to oncogenesis,
in gastric and gastroesophageal cancers. First, we must proliferation, migration, and angiogenesis in a variety of
assess and possibly quantitate heterogeneity of tumor HER2 malignancies. In a study of 4,853 solid tumor samples
expression within individual patients to identify which pa- analyzed by next-generation sequencing, FGFR was the
tients are most likely to benefit from anti-HER2 therapy. A most commonly altered tyrosine kinase, with 7.1% of
biopsy of a metastatic lesion as opposed to only the primary tumors harboring FGFR aberrations.99
lesion may provide insight into homogeneity of HER2 ex- Although four FDA-approved multikinase inhibitors (len-
pression. Functional imaging techniques with HER2-PET19 vatinib, ponatinib, pazopanib, and regorafenib) have some
could also help capture the spatial and temporal tumor biologic activity against FGFR, development of second-
heterogeneity typical of HER2-positive gastric cancer. generation, FGFR-selective, ATP-competitive inhibitors
Second, development of more sophisticated HER2 bio- made inhibition of this target a viable clinical strategy. One
markers, such as targeted sequencing panels and HER2 FGFR-selective inhibitor, erdafitinib, was approved for
IHC and FISH assays with higher thresholds for positivity FGFR2- or FGFR3-altered bladder cancer in 2019.100 Within
(CEP17 ratio . vs. , 4.7),95 are needed to optimize patient GI malignancies, about 10% to 16% of intrahepatic chol-
selection. angiocarcinomas harbor FGFR2 fusions, and multiple

FGFR-selective inhibitors have demonstrated antitumor ef- reported that, in FGFR2-mutant cell lines, acquired re-
ficacy in FGFR2-fusion advanced cholangiocarcinoma.101 sistance may occur primarily through activating mutations in
the FGFR2 kinase domain. Subsequent studies using tissue
Pemigatinib, an oral selective FGFR inhibitor, was studied in
biopsies, ctDNA analysis, and autopsy samples of patients
a phase II single-arm study (FIGHT 202) and showed
with FGFR2 fusion+ cholangiocarcinoma confirmed the
a response rate of 35.5% among 107 pretreated patients
emergence of kinase domain mutations as a common
with FGFR2 fusion+ cholangiocarcinoma.102 The duration of
mechanism of clinical acquired resistance. Goyal et al105
response was 7.5 months, with median PFS and median OS
identified three patients who initially responded to treatment
of 6.9 months and 21.1 months, respectively. The most
with infigratinib but subsequently developed resistance, two
common adverse events were hyperphosphatemia (60%),
of whom had polyclonal recurrent mutations in the FGFR2
alopecia (49%), diarrhea (47%), fatigue (42%), nail tox-
kinase domain. An FGFR2 V564F gatekeeper mutation
icities (42%), and dysgeusia (40%). Discontinuation, dose
emerged in all three patients. Consistent with these find-
reduction, and interruption due to adverse events occurred
ings, other investigators identified the emergence of
in 9%, 14%, and 42% of patients, respectively.
FGFR2 kinase domain mutations in patients with FGFR2-
Infigratinib, another oral selective FGFR inhibitor with sig- fusion cholangiocarcinoma who had initially responded
nificant structural homology to pemigatinib, showed an to pemigatinib. 106,107 Structural modeling revealed that
overall response rate of 26.9% among 71 pretreated pa- FGFR2 kinase mutations can (1) induce steric clash,
tients in a single-arm phase II study, with a duration of preventing drugs from entering the ATP-binding pocket
response of 5.4 months and median PFS and median OS of (gatekeeper mutation), or (2) stabilize the active form of the
6.8 months and 12.5 months, respectively. The adverse kinase by either disrupting a series of bonds that form as
event profile for infigratinib was similar to pemigatinib.103 a molecular “brake” on the active form or pushing the ki-
Both drugs are now being tested against standard of care nase into an active form.108 These studies suggest that
cisplatin/gemcitabine in the first line for locally advanced or a salient mechanism of acquired resistance is via mutation
metastatic cholangiocarcinoma in randomized phase III of the drug target itself.
trials (NCT03773302 and NCT03656536).
Futibatinib, an oral selective third-generation covalent FGFR Strategies to Overcome Resistance
inhibitor, irreversibly binds the kinase P-loop cysteine and Covalent versus noncovalent inhibitors The identification of
has shown potent preclinical activity against multiple FGFR2 the gatekeeper mutation and other FGFR2 kinase domain
kinase domain mutations that emerge upon resistance to mutations in patients treated with the ATP-competitive
second-generation, ATP-competitive FGFR inhibitors. Futi- FGFR selective inhibitors pemigatinib and infigratinib
batinib is being evaluated in a randomized global phase III raised the question as to whether next-generation covalent
frontline study (FOENIX-CCA3, NCT04093362) against inhibitors could overcome or suppress selection of such
gemcitabine/cisplatin in FGFR2-fusion or rearrangement mutations. Novel tool compounds that formed a covalent
positive intrahepatic cholangiocarcinoma and in a single-arm bond with a conserved cysteine residue in the FGFR kinase
phase II study in the same molecular subset of patients but in P-loop demonstrated the ability to overcome FGFR gate-
the refractory setting (FOENIX-CCA2, NCT04189445). Ad- keeper mutations in the laboratory109 and helped lead to the
ditional studies of other FGFR inhibitors, such as dera- development of the irreversible inhibitor futibatinib. Notably,
zantinib and erdafitinib, are ongoing (NCT03230318; Goyal et al108 described four patients with FGFR2-fusion
NCT04083976). cholangiocarcinoma who developed acquired resistance
to infigratinib or another ATP competitive FGFR inhibitor
FGFR Inhibitor Primary and Acquired Resistance
Debio1347 and subsequently responded to futibatinib,
Response rates to FGFR inhibitors to pretreated FGFR2 likely because its covalent binding allowed for effective
fusion+ cholangiocarcinomas have been in the 20% to target engagement regardless of the conformation of the
36% range, which is lower than that of targeted therapies FGFR2 kinase.
in other pretreated oncogene-driven malignancies, such
Isoform-selective inhibitors FGFR1, FGFR2, and FGFR3
as ALK-fusion+ lung cancer. In addition, response rates for
are all highly conserved, with FGFR4 having a relatively
FGFR2-mutated and amplified cholangiocarcinoma have
distinct structure; FGFR1–3 inhibitors thus tend to have
been disappointing relative to FGFR2-fusion–positive can-
limited activity against FGFR4 while showing similar po-
cers; no responses were seen with pemigatinib or infigratinib
tency against FGFR1, FGFR2, and FGFR3. This lack of
in their initial studies in patients with FGFR2-amplified or
isoform selectivity results in dose-limiting toxicity that pre-
FGFR2-mutated patients.
vents greater FGFR2 pathway inhibition. For example, in-
The biologic basis for both primary and acquired resis- hibition of the FGFR23-FGFR1 interaction in the kidney
tance is an area of active investigation. Byron et al104 initially results in decreased phosphorous renal clearance, causing

Chen et al
the hyperphosphatemia frequently seen in clinical studies of IDH1 mutations across multiple tumor types, including
pan-FGFR inhibitors to date.110 Drugs with true FGFR2 cholangiocarcinoma (NCT03684811). The IDH1 inhibitor
isoform selectivity could avoid adverse events such as BAY1436032 has shown efficacy in AML and is currently
hyperphosphatemia inherent to nonselective FGFR1–3 in- undergoing a phase I trial in advanced solid tumors
hibitors and allow higher maximum tolerated doses that (NCT02746081).113 The PARP inhibitor olaparib is being
could result in greater target inhibition. studied in phase II trials of refractory advanced solid tumors
More precise biomarkers Genomic analysis of patients en- with IDH1/2 mutations (NCT03212274) and in combination
rolled in FIGHT-202 found that 0% (0/9) patients with with ATR inhibitors (NCT03878095) based on preclinical
mutated TP53 had an objective response to pemigatinib data that IDH1- or IDH2-mutant cancers have defects in
versus 38.8% (38/98) of those with wild-type TP53.107 Al- homologous recombination that may leave them vulnerable
though this finding needs to be verified, it raises the broader to PARP inhibition.114
question of whether identification of a FGFR2 fusion alone is Primary and Acquired Resistance and Strategies to
sufficient to predict clinical response to FGFR inhibition. Overcome Resistance
The biologic basis for the lack of response to date in pa-
Potential mechanisms of IDH1 resistance include both on-
tients with FGFR2 mutations and amplifications, despite
pathway and off-pathway alterations. In the phase I study of
presumed FGFR2 pathway activation, also needs to be
ivosidenib in IDH1-mutant cholangiocarcinoma, 37 (59%)
understood.
of patients underwent paired pretreatment and post-
IDH1/2 progression tumor sequencing. Six patients, including four
IDH1 and IDH2 are enzymes involved in the tricarboxylic with stable disease and one with a partial response, de-
acid cycle, catalyzing the conversion of isocitrate to veloped new oncogenic mutations. These included muta-
α-ketoglutarate. Mutations in IDH1 and IDH2 can alter their tions in IDH1 and IDH2 (IDH2-R172V, IDH1-R132F) and
enzymatic function to promote reduction of α-ketoglutarate mutations in TP53, ARID1A, POLE, PIK3R1, and TBX3.115
to 2-hydroxyglutarate, an “oncometabolite” that drives The roles of these mutations in resistance to IDH1 inhibitors
aberrant DNA and histone methylation and oncogenic must be validated via functional laboratory studies.
epigenetic changes. IDH mutations were initially char- Harding et al116 described three patients with IDH1-mutated
acterized in hematologic malignancies, and the IDH2 cancers (two with AML and one with intrahepatic chol-
inhibitor enasidenib and the IDH1 inhibitor ivosidenib angiocarcinoma) who had sustained responses to ivoside-
gained FDA approval for the treatment of IDH-mutant nib but subsequently developed IDH2 mutations. They also
acute myeloid leukemia (AML) in 2017 and 2018, reported a fourth case of a patient with AML with IDH2-
respectively. mutant disease who progressed on the IDH2 inhibitor
Within GI malignancies, mutations in IDH1/2 occur in ap- enasidenib after emergence of a new IDH1 mutation. Ad-
proximately 20% to 25% of intrahepatic cholangiocarcinomas ditional prospective studies in larger cohorts of patients are
and are seen in up to 10% of clear cell hepatocellular needed to understand the frequency of isoform switching as
carcinomas and sporadically in pancreatic cancer and a mechanism of IDH inhibitor resistance, to elucidate other
colorectal cancer.111 In the phase III ClarIDHy trial, 185 mechanisms of primary acquired resistance, and to assess
patients with pretreated IDH1-mutated cholangiocarcinoma whether sequential or combination strategies with IDH1 and
were randomly selected 2:1 to receive ivosidenib or placebo, IDH2 inhibitors can provide clinical benefit.
with crossover to ivosidenib allowed at disease progression.
CONCLUSION
The trial met its primary endpoint, with a median PFS of 2.7
vs. 1.4 months favoring the ivosidenib arm (HR, 0.37; p , Targeted therapy strategies in colorectal cancer, esoph-
.001)112; PFS at 6 and 12 months was 32% and 22%, re- agogastric cancer, and cholangiocarcinoma have provided
spectively, for ivosidineb versus 0% at both timepoints for meaningful clinical benefit and additional therapeutic op-
placebo. Median OS showed a trend toward benefit (10.8 vs. tions for subsets of patients. Anti-EGFR therapy in RAS wild-
9.7 months; p = .06), and all subgroups favored ivosidenib. type colorectal cancer, anti-HER2 therapy in HER2+ gastric
Adverse events included nausea (32%), diarrhea (29%), and GEJ tumors, and FGFR and IDH1 inhibitors in FGFR2-
fatigue (24%), cough (19%), abdominal pain (19%), ascites fusion and IDH1-mutant cholangiocarcinoma are key
(19%), decreased appetite (17%), anemia (16%), and therapeutic strategies. However, only a small percentage of
vomiting (16%). patients with GI cancers who undergo tumor genomic
profiling are matched to an FDA-approved agent, and the
Other inhibitors of IDH1/2 currently being evaluated in GI
optimal strategy after resistance is often uncertain.
malignancies include the IDH1 inhibitor FT2012, which
is being studied alone and in combination with other drugs Reflecting on the progress to date, we believe that overcoming
in an early-phase basket trial in refractory patients with primary and acquired resistance to targeted therapies for GI

malignancies may require scientific advances on three fronts: pre- and post-treatment issue biopsies, ctDNA, and au-
(1) biomarker-driven enrichment strategies, (2) improved topsy studies can identify novel mechanisms of primary
drug candidates, and (3) deeper understanding of tumor and acquired resistance that guide development of more
resistance mechanisms. In the biomarker space, consid- potent therapeutic strategies.
eration of co-mutations in addition to the primary molecular
target may better identify patients whose tumors may re- Although many key questions remain to be solved, the
spond to targeted therapeutics. ctDNA may better capture scientific and clinical landscape is primed for innovation,
tumor heterogeneity than single tissue biopsies. Within drug with important advances in the efficiency of genetic and
discovery efforts, novel compounds that are more isoform- transcriptomic analysis, improvements in medicinal chemistry
or mutant specific or that have greater target binding and drug discovery platforms, and the potential for faster drug
affinity may overcome resistance to earlier-generation approvals for biomarker-driven agents. We are optimistic that
drugs and allow for greater pathway inhibition. Within the breakthroughs for targeted therapy thus far represent only
the clinic, study of tumor heterogeneity and evolution via the beginning of more essential discoveries to come.

1
Massachusetts General Hospital Cancer Center, Boston, MA Lipika Goyal, MD, DPhil, Massachusetts General Hospital Cancer Center,
2
Medical Oncology Department, Fondazione IRCCS Istituto Nazionale dei 55 Fruit Street, Boston, MA 02114; email: [email protected].
Tumori, Milan, Italy
3
Duke University Medical Center, Durham, NC
EDBK_280871.
REFERENCES
1. Marquart J, Chen EY, Prasad V. Estimation of the percentage of US patients with cancer who benefit from genome-driven oncology. JAMA Oncol. 2018;
4:1093-1098.
2. Seshacharyulu P, Ponnusamy MP, Haridas D, et al. Targeting the EGFR signaling pathway in cancer therapy. Expert Opin Ther Targets. 2012;16:15-31.
3. Yarden Y. The EGFR family and its ligands in human cancer. signalling mechanisms and therapeutic opportunities. Eur J Cancer. 2001;37 (Suppl 4):S3-S8.
4. Sunada H, Magun BE, Mendelsohn J, et al. Monoclonal antibody against epidermal growth factor receptor is internalized without stimulating receptor
phosphorylation. Proc Natl Acad Sci USA. 1986;83:3825-3829.
5. Kimura H, Sakai K, Arao T, et al. Antibody-dependent cellular cytotoxicity of cetuximab against tumor cells with wild-type or mutant epidermal growth factor
receptor. Cancer Sci. 2007;98:1275-1280.
6. Price TJ, Peeters M, Kim TW, et al. Panitumumab versus cetuximab in patients with chemotherapy-refractory wild-type KRAS exon 2 metastatic colorectal
cancer (ASPECCT): a randomised, multicentre, open-label, non-inferiority phase 3 study. Lancet Oncol. 2014;15:569-579.
7. Van Cutsem E, Peeters M, Siena S, et al. Open-label phase III trial of panitumumab plus best supportive care compared with best supportive care alone in
patients with chemotherapy-refractory metastatic colorectal cancer. J Clin Oncol. 2007;25:1658-1664.
8. Jonker DJ, O’Callaghan CJ, Karapetis CS, et al. Cetuximab for the treatment of colorectal cancer. N Engl J Med. 2007;357:2040-2048.
9. Bokemeyer C, Van Cutsem E, Rougier P, et al. Addition of cetuximab to chemotherapy as first-line treatment for KRAS wild-type metastatic colorectal cancer:
pooled analysis of the CRYSTAL and OPUS randomised clinical trials. Eur J Cancer. 2012;48:1466-1475.
10. Douillard JY, Siena S, Cassidy J, et al. Final results from PRIME: randomized phase III study of panitumumab with FOLFOX4 for first-line treatment of metastatic
colorectal cancer. Ann Oncol. 2014;25:1346-1355.
11. Lièvre A, Bachet JB, Le Corre D, et al. KRAS mutation status is predictive of response to cetuximab therapy in colorectal cancer. Cancer Res. 2006;
66:3992-3995.
12. Benvenuti S, Sartore-Bianchi A, Di Nicolantonio F, et al. Oncogenic activation of the RAS/RAF signaling pathway impairs the response of metastatic colorectal
cancers to anti-epidermal growth factor receptor antibody therapies. Cancer Res. 2007;67:2643-2648.
13. Douillard JY, Oliner KS, Siena S, et al. Panitumumab-FOLFOX4 treatment and RAS mutations in colorectal cancer. N Engl J Med. 2013;369:1023-1034.
14. Van Cutsem E, Lenz HJ, Köhne CH, et al. Fluorouracil, leucovorin, and irinotecan plus cetuximab treatment and RAS mutations in colorectal cancer. J Clin
Oncol. 2015;33:692-700.
15. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology: Colon Cancer. https://www.nccn.org/professionals/physician_gls/
pdf/colon.pdf. Accessed February 1, 2020.

Chen et al
16 Yaeger R, et al. Clinical sequencing defines the genomic landscape of metastatic colorectal cancer. Cancer Cell. 2018;33:125-136 e3.
17. Ikenoue T, Hikiba Y, Kanai F, et al. Functional analysis of mutations within the kinase activation segment of B-Raf in human colorectal tumors. Cancer Res.
2003;63:8132-8137.
18. Davies H, Bignell GR, Cox C, et al. Mutations of the BRAF gene in human cancer. Nature. 2002;417:949-954.
19. Wan PT, Garnett MJ, Roe SM, et al.; Cancer Genome Project. Mechanism of activation of the RAF-ERK signaling pathway by oncogenic mutations of B-RAF.
Cell. 2004;116:855-867.
20. Kopetz S, McDonough S, Morris VK, et al. Randomized trial of irinotecan and cetuximab with or without vemurafenib in BRAF-mutant metastatic colorectal
cancer (SWOG 1406). J Clin Oncol. 2017;35 (suppl; abstr 520).
21. Corcoran RB, André T, Atreya CE, et al. Combined BRAF, EGFR, and MEK inhibition in patients with BRAF V600E-mutant colorectal cancer. Cancer Discov. 2018;
8:428-443.
24. Raghav KP, Overman MJ, Yu R, et al. HER2 amplification as a negative predictive biomarker for anti-epidermal growth factor receptor antibody therapy in
metastatic colorectal cancer. J Clin Oncol. 2016;34 (suppl, abstr 3517).
25. Yonesaka K, Zejnullahu K, Okamoto I, et al. Activation of ERBB2 signaling causes resistance to the EGFR-directed therapeutic antibody cetuximab. Sci Transl
Med. 2011;3:99ra86.
26. Bertotti A, Migliardi G, Galimi F, et al. A molecularly annotated platform of patient-derived xenografts (“xenopatients”) identifies HER2 as an effective therapeutic
target in cetuximab-resistant colorectal cancer. Cancer Discov. 2011;1:508-523.
27. Sartore-Bianchi A, Amatu A, Porcu L, et al. HER2 positivity predicts unresponsiveness to EGFR-Targeted treatment in metastatic colorectal cancer. Oncologist.
2019;24:1395-1402.
28. Yaeger R, Chatila WK, Lipsyc MD, et al. Clinical sequencing defines the genomic landscape of metastatic colorectal cancer. Cancer Cell. 2018;33:125-136 e3.
29. Valtorta E, Misale S, Sartore-Bianchi A, et al. KRAS gene amplification in colorectal cancer and impact on response to EGFR-targeted therapy. Int J Cancer.
2013;133:1259-1265.
30. Mekenkamp LJ, Tol J, Dijkstra JR, et al. Beyond KRAS mutation status: influence of KRAS copy number status and microRNAs on clinical outcome to cetuximab
in metastatic colorectal cancer patients. BMC Cancer. 2012;12:292.
3:658-673.
32. Pietrantonio F, Maggi C, Di Bartolomeo M, et al. Gain of ALK gene copy number may predict lack of benefit from anti-EGFR treatment in patients with advanced
colorectal cancer and RAS-RAF-PI3KCA wild-type status. PLoS One. 2014;9:e92147.
33. Pietrantonio F, Di Nicolantonio F, Schrock AB, et al. ALK, ROS1, and NTRK rearrangements in metastatic colorectal cancer. J Natl Cancer Inst. 2017;109:doi:
10.1093/jnci/djx089.
34. Bertotti A, Papp E, Jones S, et al. The genomic landscape of response to EGFR blockade in colorectal cancer. Nature. 2015;526:263-267.
35. Nosho K, Kawasaki T, Ohnishi M, et al. PIK3CA mutation in colorectal cancer: relationship with genetic and epigenetic alterations. Neoplasia. 2008;10:534-541.
36. Bregni G, Sciallero S, Sobrero A. HER2 amplification and anti-EGFR sensitivity in advanced colorectal cancer. JAMA Oncol. 2019;5:605-606.
38. Sartore-Bianchi A, Martini M, Molinari F, et al. PIK3CA mutations in colorectal cancer are associated with clinical resistance to EGFR-targeted monoclonal
antibodies. Cancer Res. 2009;69:1851-1857.
39. Prenen H, De Schutter J, Jacobs B, et al. PIK3CA mutations are not a major determinant of resistance to the epidermal growth factor receptor inhibitor
cetuximab in metastatic colorectal cancer. Clin Cancer Res. 2009;15:3184-3188.
40. Mao C, Yang ZY, Hu XF, et al. PIK3CA exon 20 mutations as a potential biomarker for resistance to anti-EGFR monoclonal antibodies in KRAS wild-type
metastatic colorectal cancer: a systematic review and meta-analysis. Ann Oncol. 2012;23:1518-1525.
41. Perrone F, Lampis A, Orsenigo M, et al. PI3KCA/PTEN deregulation contributes to impaired responses to cetuximab in metastatic colorectal cancer patients.
Ann Oncol. 2009;20:84-90.
42. Karapetis CS, Jonker D, Daneshmand M, et al; NCIC Clinical Trials Group and the Australasian Gastro-Intestinal Trials Group. PIK3CA, BRAF, and PTEN status
and benefit from cetuximab in the treatment of advanced colorectal cancer: results from NCIC CTG/AGITG CO.17. Clin Cancer Res. 2014;20:744-753.
43. Jacobs B, De Roock W, Piessevaux H, et al. Amphiregulin and epiregulin mRNA expression in primary tumors predicts outcome in metastatic colorectal cancer
treated with cetuximab. J Clin Oncol. 2009;27:5068-5074.
44. Kijima T, Niwa H, Steinman RA, et al. STAT3 activation abrogates growth factor dependence and contributes to head and neck squamous cell carcinoma tumor
growth in vivo. Cell Growth Differ. 2002;13:355-362.
45. Haura EB, Sommers E, Song L, et al. A pilot study of preoperative gefitinib for early-stage lung cancer to assess intratumor drug concentration and pathways
mediating primary resistance. J Thorac Oncol. 2010;5:1806-1814.

46. Thomson S, Petti F, Sujka-Kwok I, et al. Kinase switching in mesenchymal-like non-small cell lung cancer lines contributes to EGFR inhibitor resistance through
pathway redundancy. Clin Exp Metastasis. 2008;25:843-854.
47. Tan TZ, Miow QH, Miki Y, et al. Epithelial-mesenchymal transition spectrum quantification and its efficacy in deciphering survival and drug responses of cancer
patients. EMBO Mol Med. 2014;6:1279-1293.
48. Buck E, Eyzaguirre A, Barr S, et al. Loss of homotypic cell adhesion by epithelial-mesenchymal transition or mutation limits sensitivity to epidermal growth factor
receptor inhibition. Mol Cancer Ther. 2007;6:532-541.
49. Arnold D, Lueza B, Douillard JY, et al. Prognostic and predictive value of primary tumour side in patients with RAS wild-type metastatic colorectal cancer treated
with chemotherapy and EGFR directed antibodies in six randomized trials. Ann Oncol. 2017;28:1713-1729.
50. Misale S, Di Nicolantonio F, Sartore-Bianchi A, et al. Resistance to anti-EGFR therapy in colorectal cancer: from heterogeneity to convergent evolution. Cancer
Discov. 2014;4:1269-1280.
51. Diaz LA, Jr., Williams RT, Wu J, et al. The molecular evolution of acquired resistance to targeted EGFR blockade in colorectal cancers. Nature. 2012;
486:537-540.
486:532-536.
53. Siravegna G, Mussolin B, Buscarino M, et al. Clonal evolution and resistance to EGFR blockade in the blood of colorectal cancer patients. Nat Med. 2015;
21:795-801.
2015;21:2157-2166.
56. Pietrantonio F, Vernieri C, Siravegna G, et al. Heterogeneity of acquired resistance to anti-EGFR monoclonal antibodies in patients with metastatic colorectal
57. Van Emburgh BO, Arena S, Siravegna G, et al. Acquired RAS or EGFR mutations and duration of response to EGFR blockade in colorectal cancer. Nat Commun.
2016;7:13665.
58. Montagut C, Dalmases A, Bellosillo B, et al. Identification of a mutation in the extracellular domain of the Epidermal Growth Factor Receptor conferring
cetuximab resistance in colorectal cancer. Nat Med. 2012;18:221-223.
59. Strickler JH, Loree JM, Ahronian LG, et al. Genomic landscape of cell-Free DNA in patients with colorectal cancer. Cancer Discov. 2018;8:164-173.
60. Russo M, Crisafulli G, Sogari A, et al. Adaptive mutability of colorectal cancers in response to targeted therapies. Science. 2019;366:1473-1480.
61. Hobor S, Van Emburgh BO, Crowley E, et al. TGFα and amphiregulin paracrine network promotes resistance to EGFR blockade in colorectal cancer cells. Clin
Cancer Res. 2014;20:6429-6438.
62. Yonesaka K, Satoh T, Ueda S, et al. Circulating hepatocyte growth factor is correlated with resistance to cetuximab in metastatic colorectal cancer. Anticancer
Res. 2015;35:1683-1689.
63. Tol J, Koopman M, Cats A, et al. Chemotherapy, bevacizumab, and cetuximab in metastatic colorectal cancer. N Engl J Med. 2009;360:563-572.
64. Hecht JR, Mitchell E, Chidiac T, et al. A randomized phase IIIB trial of chemotherapy, bevacizumab, and panitumumab compared with chemotherapy and
bevacizumab alone for metastatic colorectal cancer. J Clin Oncol. 2009;27:672-680.
65. Misale S, Bozic I, Tong J, et al. Vertical suppression of the EGFR pathway prevents onset of resistance in colorectal cancers. Nat Commun. 2015;6:8305.
6:147-153.
67. Parseghian CM, Loree JM, Morris VK, et al. Anti-EGFR-resistant clones decay exponentially after progression: implications for anti-EGFR re-challenge. Ann
Oncol. 2019;30:243-249.
68. Cremolini C, Rossini D, Dell’Aquila E, et al. Rechallenge for patients with RAS and BRAF wild-type metastatic colorectal cancer with acquired resistance to first-
line cetuximab and irinotecan: a phase 2 single-arm clinical trial. JAMA Oncol. 2019;5:343-350.
70. Van Cutsem E, Bang YJ, Feng-Yi F, et al. HER2 screening data from ToGA: targeting HER2 in gastric and gastroesophageal junction cancer. Gastric Cancer.
2015;18:476-484.
71. Okines A, Cunningham D, Chau I. Targeting the human EGFR family in esophagogastric cancer. Nat Rev Clin Oncol. 2011;8:492-503.
72. Bang YJ, Van Cutsem E, Feyereislova A, et al; ToGA Trial Investigators. Trastuzumab in combination with chemotherapy versus chemotherapy alone for
treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet. 2010;
376:687-697.
73. U.S. Food and Drug Administration. Full Prescribing Information for Trastuzumab. www.accessdata.fda.gov/drugsatfda_docs/label/2010/103792s5250lbl.pdf.
74. European Medicines Agency. Committee for Medicinal Products for Human Use PostAuthorisation Summary of Positive Opinion for Herceptin. www.ema.
europa.eu/docs/en_GB/document_library/Summary_of_opinion/human/000278/WC500059913.pdf. Accessed February 1, 2020.

Chen et al
75. Bartley AN, Washington MK, Colasacco C, et al. HER2 testing and clinical decision making in gastroesophageal adenocarcinoma: guideline from the College of
American Pathologists, American Society for Clinical Pathology, and the American Society of Clinical Oncology. J Clin Oncol. 2017;35:446-464.
76. Zhao D, Klempner SJ, Chao J. Progress and challenges in HER2-positive gastroesophageal adenocarcinoma. J Hematol Oncol. 2019;12:50.
77. Albarello L, Pecciani L, Doglioni C.HER2 testing in gastric cancer. Adv Anat Pathol. 2011;18:53-59.
78. Cancer Genome Atlas Research Network. Comprehensive molecular characterization of gastric adenocarcinoma. Nature. 2014;513:202-209.
79. Cristescu R, Lee J, Nebozhyn M, et al. Molecular analysis of gastric cancer identifies subtypes associated with distinct clinical outcomes. Nat Med. 2015;
21:449-456.
80. Pietrantonio F, Caporale M, Morano F, et al. HER2 loss in HER2-positive gastric or gastroesophageal cancer after trastuzumab therapy: Implication for further
clinical research. Int J Cancer. 2016;139:2859-2864.
81. Kim J, Fox C, Peng S, et al. Preexisting oncogenic events impact trastuzumab sensitivity in ERBB2-amplified gastroesophageal adenocarcinoma. J Clin Invest.
2014;124:5145-5158.
82. Kwak EL, Ahronian LG, Siravegna G, et al. Molecular heterogeneity and receptor coamplification drive resistance to targeted therapy in MET-amplified
esophagogastric cancer. Cancer Discov. 2015;5:1271-1281.
83. Pectasides E, Stachler MD, Derks S, et al. Genomic heterogeneity as a barrier to precision medicine in gastroesophageal adenocarcinoma. Cancer Discov. 2018;
8:37-48.
84. Lee JY, Hong M, Kim ST, et al. The impact of concomitant genomic alterations on treatment outcome for trastuzumab therapy in HER2-positive gastric cancer.
Sci Rep. 2015;5:9289.
85. Pietrantonio F, Fucà G, Morano F, et al. Biomarkers of primary resistance to trastuzumab in HER2-positive metastatic gastric cancer patients: the AMNESIA
case-control study. Clin Cancer Res. 2018;24:1082-1089.
86. Janjigian YY, Sanchez-Vega F, Jonsson P, et al. Genetic predictors of response to systemic therapy in esophagogastric cancer. Cancer Discov. 2018;8:49-58.
87. Seo S, Ryu MH, Park YS, et al. Loss of HER2 positivity after anti-HER2 chemotherapy in HER2-positive gastric cancer patients: results of the GASTric cancer
HER2 reassessment study 3 (GASTHER3). Gastric Cancer. 2019;22:527-535.
88. Tabernero J, Hoff PM, Shen L, et al. Pertuzumab plus trastuzumab and chemotherapy for HER2-positive metastatic gastric or gastro-oesophageal junction
cancer (JACOB): final analysis of a double-blind, randomised, placebo-controlled phase 3 study. Lancet Oncol. 2018;19:1372-1384.
89. Shah MA, Xu RH, Bang YJ, et al. HELOISE: Phase IIIb randomized multicenter study comparing standard-of-care and higher-dose trastuzumab regimens
combined with chemotherapy as first-line therapy in patients with human epidermal growth factor receptor 2-positive metastatic gastric or gastroesophageal
junction adenocarcinoma. J Clin Oncol. 2017;35:2558-2567.
90. Hecht JR, Bang YJ, Qin SK, et al. Lapatinib in combination with capecitabine plus oxaliplatin in human epidermal growth factor receptor 2-positive advanced or
metastatic gastric, esophageal, or gastroesophageal adenocarcinoma: TRIO-013/LOGiC–a randomized phase III trial. J Clin Oncol. 2016;34:443-451.
91. Satoh T, Xu RH, Chung HC, et al. Lapatinib plus paclitaxel versus paclitaxel alone in the second-line treatment of HER2-amplified advanced gastric cancer in
Asian populations: TyTAN--a randomized, phase III study. J Clin Oncol. 2014;32:2039-2049.
92. Thuss-Patience PC, Shah MA, Ohtsu A, et al. Trastuzumab emtansine versus taxane use for previously treated HER2-positive locally advanced or metastatic
gastric or gastro-oesophageal junction adenocarcinoma (GATSBY): an international randomised, open-label, adaptive, phase 2/3 study. Lancet Oncol. 2017;
18:640-653.
93. Palle J, Tougeron D, Pozet A, et al. Trastuzumab beyond progression in patients with HER2-positive advanced gastric adenocarcinoma: a multicenter AGEO
study. Oncotarget. 2017;8:101383-101393.
94. Horita Y, Nishino M, Sugimoto S, et al. Phase II clinical trial of second-line weekly paclitaxel plus trastuzumab for patients with HER2-positive metastatic gastric
cancer. Anticancer Drugs. 2019;30:98-104.
95. Gomez-Martin C, Plaza JC, Pazo-Cid R, et al. Level of HER2 gene amplification predicts response and overall survival in HER2-positive advanced gastric cancer
treated with trastuzumab. J Clin Oncol. 2013;31:4445-4452.
96. Shitara K, Iwata H, Takahashi S, et al. Trastuzumab deruxtecan (DS-8201a) in patients with advanced HER2-positive gastric cancer: a dose-expansion, phase 1
study. Lancet Oncol. 2019;20:827-836.
97. Janjigian YY, Chou JF, Simmons M, et al. First-line pembrolizumab (P), trastuzumab (T), capecitabine (C) and oxaliplatin (O) in HER2-positive metastatic
esophagogastric adenocarcinoma (mEGA). J Clin Oncol. 2019;37 (suppl; abstr 62).
98. Babina IS, Turner NC. Advances and challenges in targeting FGFR signalling in cancer. Nat Rev Cancer. 2017;17:318-332.
99. Helsten T, Elkin S, Arthur E, et al. The FGFR landscape in cancer: analysis of 4,853 tumors by next-generation sequencing. Clin Cancer Res. 2016;22:259-267.
100. U.S. Food and Drug Administration. FDA grants accelerated approval to erdafitinib for metastatic urothelial carcinoma. www.fda.gov/drugs/resources-
information-approved-drugs/fda-grants-accelerated-approval-erdafitinib-metastatic-urothelial-carcinoma. Accessed February 15, 2020.
101. Valle JW, Lamarca A, Goyal L, et al. New horizons for precision medicine in biliary tract cancers. Cancer Discov. 2017;7:943-962.
102. Vogel A, Sahai V, Hollebecque A, et al. FIGHT-202: a phase 2 study of pemigatinib in patients with previously treated locally advanced or metastatic
cholangiocarcinoma. https://oncologypro.esmo.org/meeting-resources/esmo-2019-congress/FIGHT-202-a-phase-2-study-of-pemigatinib-in-patients-pts-with-
previously-treated-locally-advanced-or-metastatic-cholangiocarcinomaCCA. Accessed February 1, 2020.
103. Javle M, Kelley R, Roychowdhury S, et al. A phase II study of infigratinib (BGJ398) in previously-treated advanced cholangiocarcinoma containing FGFR2
fusions. Hepatobiliary Surg Nutr. 2019;8 (suppl; abstr AB051).

104. Byron SA, Chen H, Wortmann A, et al. The N550K/H mutations in FGFR2 confer differential resistance to PD173074, dovitinib, and ponatinib ATP-competitive
inhibitors. Neoplasia. 2013;15:975-988.
105. Goyal L, Saha SK, Liu LY, et al. Polyclonal secondary FGFR2 mutations drive acquired resistance to FGFR inhibition in patients with FGFR2 fusion-positive
cholangiocarcinoma. Cancer Discov. 2017;7:252-263.
106. Krook MA, Bonneville R, Chen HZ, et al. Tumor heterogeneity and acquired drug resistance in FGFR2-fusion-positive cholangiocarcinoma through rapid
research autopsy. Cold Spring Harb Mol Case Stud. 2019;5:a004002.
107. Hollebecque A, Silverman I, Owens S, et al. Comprehensive genomic profiling and clinical outcomes in patients with fibroblast growth factor receptor
rearrangement-positive cholangiocarcinoma treated with pemigatinib in the FIGHT-202 trial. Ann Oncol. 2019;30 (suppl_5):v253-v324.
108. Goyal L, Shi L, Liu LY, et al. TAS-120 overcomes resistance to ATP-competitive FGFR inhibitors in patients with FGFR2 fusion-positive iontraheptic chol-
angiocarcinoma. Cancer Discov. 2019;9:1064-1079.
109. Tan L, Wang J, Tanizaki J, et al. Development of covalent inhibitors that can overcome resistance to first-generation FGFR kinase inhibitors. Proc Natl Acad Sci
USA. 2014;111:E4869-E4877.
110. Gattineni J, Alphonse P, Zhang Q, et al. Regulation of renal phosphate transport by FGF23 is mediated by FGFR1 and FGFR4. Am J Physiol Renal Physiol. 2014;
306:F351-F358.
111. Wang P, Dong Q, Zhang C, et al. Mutations in isocitrate dehydrogenase 1 and 2 occur frequently in intrahepatic cholangiocarcinomas and share hyper-
methylation targets with glioblastomas. Oncogene. 2013;32:3091-3100.
112. Abou-Alfa G, Mercade T, Javle M, et al. ClarIDHy: a global phase 3, randomized, goublebind study of ivosidenib vs placebo in patients with advanced
cholangiocarcinoma with an isocitrate dehydrogenase 1 (IDH1) mutation. https://oncologypro.esmo.org/meeting-resources/esmo-2019-congress/ClarIDHy-A-
global-phase-3-randomized-double-blind-study-of-ivosidenib-IVO-vs-placebo-in-patients-with-advanced-cholangiocarcinoma-CC-with-an-isocitrate-dehydrogenase-
1-IDH1-mutation. Accessed February 1, 2020.
113. Chaturvedi A, Herbst L, Pusch S, et al. Pan-mutant-IDH1 inhibitor BAY1436032 is highly effective against human IDH1 mutant acute myeloid leukemia in vivo.
Leukemia. 2017;31:2020-2028.
114. Silkowski P, Corso C, Dobinson N, et al. 2-Hydroxyglutarate produced by neomprohic IDH mutations. Sci Transl Med. 2017;9:eaal2463.
115. Lowery MA, Abou-Alfa GK, Burris HA, et al. Phase I study of AG-120, an IDH1 mutant enzyme inhibitor: Results from the cholangiocarcinoma dose escalation
and expansion cohorts. J Clin Oncol. 2017;35 (suppl; abstr 4015).
116. Harding JJ, Lowery MA, Shih AH, et al. Isoform switching as a mechanism of acquired resistance to mutant isocitrate dehydrogenase inhibition. Cancer Discov.
2018;8:1540-1547.

The Treatment of Hepatocellular Carcinoma With

Portal Vein Tumor Thrombosis
Motaz Qadan, MD, PhD1; Nishita Kothary, MD2; Bruno Sangro, MD, PhD3; and Manisha Palta, MD4
overview
Hepatocellular carcinoma (HCC) is the sixth most common cancer and third leading cause of cancer-related
death worldwide. HCC is also is a tumor with a distinct ability to invade and grow within the hepatic vas-
culature. Approximately 20% of patients with HCC have macrovascular invasion (MVI) at the time of diagnosis.
MVI is associated with dismal prognosis, with median survival ranging from 2 to 5 months. Current staging
systems designate MVI as advanced disease. Recent advances in multimodal approaches, including systemic
therapies, radiation therapy, liver-directed therapies, and surgical approaches, in the treatment of HCC with
MVI have rendered this disease process more treatable with improved outcomes and are discussed here.
INTRODUCTION decompensation, with resultant ascites, jaundice, vari-

Hepatocellular carcinoma (HCC) is the sixth most com- ces, or encephalopathy.
mon cancer and third leading cause of cancer-related Recent advances in multimodal approaches in the
death worldwide.1 HCC is also is a tumor with a distinct treatment of HCC with MVI have rendered this disease
ability to invade and grow within the hepatic vasculature. process more treatable with improved outcomes and
Although macrovascular invasion (MVI) may involve are discussed here.
hepatic veins or portal veins, invasion of portal venous
branches is more common. Approximately 20% of pa- SYSTEMIC THERAPIES
tients have MVI at the time of diagnosis. Among patients HCC is highly resistant to traditional cytotoxic che-
with unresectable tumors, the probability of developing motherapy.6 Oral multi-targeted tyrosine kinase in-
portal vein tumor thrombosis (PVTT) at 1 and 3 years is hibitors (TKIs) are the mainstay systemic agents of
21% and 46%, respectively. choice in the treatment of HCC. There are four drugs
The Barcelona Clinic Liver Cancer (BCLC) staging approved worldwide. Ten years ago, sorafenib proved
system designates PVTT as advanced disease (BCLC superior to placebo in prolonging overall survival (OS)
class C) for which only systemic therapy is currently of patients with advanced HCC, with preserved liver
recommended.2,3 MVI, PVTT, and/or hepatic vein function in the pivotal SHARP trial.7 More recently,
invasion is associated with dismal prognosis, with lenvatinib was found to be noninferior to sorafenib as
median survival ranging from 2 to 5 months with best a first-line agent, with a comparable toxicity profile and
supportive care.4,5 On one hand, MVI impairs liver a similar impact in health-related quality of life.8 Im-
function through reduced liver perfusion as a result portantly, patients with PVTT involving the main trunk
of impaired blood flow, either directly via reduced were excluded from the REFLECT trial; as a result, only
inflow in PVTT or via elevated sinusoidal pressure in 21% had PVTT compared with 38% in SHARP.
hepatic vein invasion. On the other hand, extension Cabozantinib resulted in prolonged survival compared
Author affiliations of tumor within the vasculature promotes tumor with placebo in patients progressing to one or two lines
and support spread beyond the liver via direct seeding and ex- of therapy (including sorafenib) in the CELESTIAL trial.9
information (if tension. The impact on prognosis is thus profound Importantly, patients with hepatic vein invasion were
applicable) appear excluded from this trial. In the RESORCE trial, patients
and predictable. In fact, the relative impact of MVI
at the end of this who stably tolerated sorafenib (at least 400 mg daily for
article. on mortality in untreated patients (odds ratio [OR]
1.9) is stronger than that of extrahepatic spread at least 20 of the 28 days before discontinuation) but
Accepted on
February 20, 2020 (OR 1.6) or performance status (OR 1.2). The lo- experienced radiologic progression had an improved
and published at cation and extent of MVI may further affect prog- survival with regorafenib compared with those who re-
ascopubs.org on nosis through the limitations exerted on the use ceived placebo.10
March 26, 2020:
of treatment modalities such as transplantation, Antiangiogenic monoclonal antibodies, including
10.1200/EDBK_ resection, and transarterial therapies. Finally, MVI bevacizumab11 and ramucirumab,12 failed to prove
280811 can result in portal hypertension and associated liver activity as single agents. However, a post hoc analysis

Treatment of Hepatocellular Carcinoma With Portal Vein Tumor Thrombosis
(13.9 months vs. 10.6 months; p = .023), although the

difference did not meet the prespecified statistical thresh-
old.16 The median duration of response to pembrolizumab
• The treatment of hepatocellular carcinoma with at 13.8 months was comparable to that of nivolumab, in-
macrovascular invasion, including portal vein
dicating a drug class effect. Recently, the CheckMate 459
tumor thrombus, remains challenging to treat
trial compared nivolumab with sorafenib in the first-line
and has poor outcomes.
setting.17 Improved median survival was observed in
• Approximately 20% of patients with hepato- nivolumab-treated patients (16.4 months vs. 14.7 months;
cellular carcinoma have macrovascular in-
p = .07), although the predefined threshold of statistical
vasion at the time of diagnosis.
significance was not met. In the postprogression setting,
• Recent advances in the management of this 31% of patients in the sorafenib arm received an immu-
challenging condition have emerged. notherapeutic or investigational agent that was often a check-
• Multimodal approaches in the treatment of point blockade inhibitor, and a TKI was received by a similar
hepatocellular carcinoma with macrovascular proportion of patients in both arms (36% after nivolumab
invasion have improved outcomes associated and 23% after sorafenib). Nivolumab induced objective
with this advanced entity. responses more frequently than sorafenib (15% vs. 7%),
including 4% of patients exhibiting complete responses and
with responses in patients with PVTT (Fig. 1). Complete
of the negative trial with ramucirumab suggested a potential responses to sorafenib have also been reported among pa-
efficacy among patients with high serum levels of alpha- tients with MVI.18
fetoprotein greater than 400 ng/mL. The signal was later
The presence of MVI and that of extrahepatic disease were
confirmed in the REACH-2 trial and demonstrated a favor-
incorporated as stratification factors in the pivotal SHARP
able safety profile.13
trial and subsequent phase III trials, thereby allowing
Immunotherapy with immune checkpoint inhibitors as an comparison of outcomes in patients with MVI in subgroup
entity is transforming systemic therapy in HCC. The PD-1 post hoc analyses. As shown in Table 1, the treatment
inhibitors nivolumab and pembrolizumab first demonstrated benefit with all of these agents in the overall population is
activity in single-arm trials in the second-line setting.14,15 replicated in the subgroup of patients with MVI, with the
This led to accelerated approval by the U.S. Food and Drug exception of ramucirumab. The median survival was sig-
Administration and other regulatory agencies. The most nificantly higher in the ramucirumab group than in the
important finding in these trials was the 15% to 20% rate of placebo group (8.5 months vs. 7.3 months; p = .0199).
objective remissions that were durable and associated with However, in the post hoc subgroup analysis, the benefit
prolonged survival. In the CheckMate 040 trial, the median persisted among patients without MVI (HR, 0.60; 95% CI,
duration of response to nivolumab was 17 months, and the 0.42–0.87) but not in those with vascular invasion (HR,
2-year survival rate among responders was greater than 0.97; 95% CI, 0.61–1.53). It should be noted that, in the
80%.14 The KEYNOTE 240 trial compared pembrolizumab REFLECT and CheckMate 459 trials, post hoc analyses were
with placebo after progression to sorafenib and showed available for combined MVI and extrahepatic disease but
a statistically significant prolongation of median survival not for MVI alone.
FIGURE 1. Hepatitis C Virus–

Related Cirrhosis and an Advanced
Hepatocellular Carcinoma Tumor
With Portal Venous Invasion
(A) The patient received nivolu-
mab as first-line therapy. A partial
response was observed after 8
weeks, followed by a complete
response, including disappear-
ance of the portal vein tumor
thrombus. Nivolumab was dis-
continued after 1 year of therapy.
(B) The patient remains free of
disease 1 year later.
Abbreviation: AFP, alpha-fetoprotein.

Qadan et al
TABLE 1. Magnitude of Treatment Benefit of Systemic Therapies Among Patients With Hepatocellular Carcinoma and Macrovascular Invasion
CheckMate
Variable SHARP7 REFLECT8 RESORCE10 CELESTIAL9 REACH-213 45917 KEYNOTE 24016
Investigational Sorafenib Lenvatinib Regorafenib Cabozantinib Ramucirumab Nivolumab Pembrolizumab
agent
Control Placebo Sorafenib Placebo Placebo Placebo Sorafenib Placebo
Percentage of 38 21 28 30 35 73 13
patients with
MVI
HR (95% CI)
MVI 0.68 (0.49–0.93) 0.87a (0.73–1.04) 0.67 (0.46–0.98) 0.75 (0.54–1.03) 0.97 (0.61–1.53) 0.74a (0.61–0.90) 0.57 (0.29–1.13)
No MVI 0.74 (0.54–1.00) 1.05a (0.79–1.40) 0.67 (0.52–0.86) 0.80 (0.64–1.01) 0.60 (0.42–0.87) 1.14a (0.81–1.62) 0.82 (0.63–1.06)
Abbreviations: HR, hazard ratio; MVI, macrovascular invasion.

a
HR was calculated for the subgroup of patients with MVI and extrahepatic spread or neither.
Substantial experience has been accumulated over the Technological advances in radiation delivery in the past
years with sorafenib to increase median OS from 10.7 to several decades allow delivery of safe and effective ablative
14.7 months since the last reported trial. When patients are doses to the liver. These advances include respiratory
fit to receive a second agent upon progression, survival motion assessment, respiratory motion management, and
figures greater than 2 years are seen.19 Recently, the treatment planning and delivery. Conformal techniques
IMbrave 150 trial was reported to have met its coprimary allow both dose escalation to target volumes and normal
endpoints, demonstrating superior survival for the com- tissue sparing. These advances have allowed radiotherapy
bination of atezolizumab and bevacizumab (median 16.4 to be an important modality in the management of HCC,
months) compared with sorafenib in the first-line setting.17 demonstrating promising local control and quality of life with
This finding is likely to establish the combination as acceptable rates of toxicity. Given these advances, radio-
a global standard of care in the management of advanced therapy has evolved from a purely palliative treatment to that
HCC, and it illustrates how the activity of current systemic of a bridge to transplantation or even definitive curative
therapies provides clinically significant benefits for patients intent treatment.21
with HCC.
A number of prospective nonrandomized studies have
In summary, available systemic agents, including TKIs, evaluated the use of hypofractionated image-guided radi-
immunotherapy agents with checkpoint blockade inhibitors, ation therapy for treatment of HCC. Select studies are
and vascular endothelial growth factor inhibitors, have all presented in Table 2. These experiences suggest that SBRT
been shown to prolong survival in patients with advanced is well tolerated, with acceptable toxicities and excellent
HCC with strong (level 1) scientific evidence. In addition, local control rates. Follow-up is short, however, because
sequencing the available agents offers prolonged survival many have not reached a median survival endpoint. In
for many patients. Indeed, response to systemic therapies addition, few randomized studies comparing radiation
may open the door for locoregional therapies, including therapy with other locoregional modalities exist.
surgery, to further improve prognosis.
Bujold et al22 reported a sequential phase I and II trial of 102
RADIATION THERAPY patients with Child–Turcotte–Pugh (CTP) class A HCC.
Larger tumors were included (median diameter 7.2 cm),
Clinical Outcomes of Radiotherapy for HCC and more than half of patients had previous liver-directed
Stereotactic body radiotherapy (SBRT) involves treating treatment and tumor vascular thrombosis. The phase I dose
primary tumors or metastases with a few high doses of escalation tested 24 to 54 Gy in six fractions, with a median
ionizing radiation. The benefits of SBRT are that it is dose for the pooled analysis of 36 Gy in six fractions. Despite
a noninvasive outpatient procedure typically delivered in 3 this poorer-prognosis patient population, SBRT was asso-
to 10 fractions. In SBRT, tumor kill is maximized and dose to ciated with reasonable outcomes and modest toxicity.
surrounding tissue is minimized via precise and accurate Median survival was 17 months, with a 1-year OS of 55%.
delivery of multiple radiation beams to the target. This is Thirty-six percent of patients had grade 3+ toxicity, and 29%
particularly challenging with liver malignancies, because had deterioration of CTP scores at 3 months, with most
these lesions move with respiration and are irregular in shape, patients recovering liver function at 12 months. All declines
necessitating careful treatment planning and continual man- in liver function, regardless of etiology from progressive
agement of such motion and patient position during irradiation.20 cirrhosis or sequelae of radiation toxicity, were scored as

TABLE 2. Select Prospective Clinical Trials of More Than 50 Patients Evaluating Role of Radiation in Hepatocellular Carcinoma
VI/ Median
No. of Patients Tumor Size TVT Survival Toxicity (collective ‡ grade
Study Design (No. of lesions) CP Class (%) Dose/Fx (cm) (%) ORR (%) LC (%) OS (%) (months) 3)a
Bujold et al, Phase I, II 102 A (100) 24–54 Gy/6 fx Median 55 54 87, 1 year 55, 1 year 17 36%
201322 (3 week) diameter
7.2
Median 36 Gy Range 29% progression of CP
in 6 fx (1.4–23.1) class at 3 months
6% progression of CP class
at 12 months
7 patients with grade 5
(including 2 with massive
TVT progression)
Lasley et al, Phase I, II 59 (65) A (64) CP-A: Max 20 CP-A: 89 CP-A: 91, 1 CP-A: 94, CP-A: 44.8 11% CP-A
201523 36–48 Gy/3 diameter 6 (includes year, 2 1 year
fx SD) years,
3 years
CP-A: Median,
48 Gy in 3 fx
B (36) CP-B: CP-A: 72, 38% CP-B
40–55 Gy/5 2 years
fx
CP-B: median CP-B: 95 CP-B: 17 14% RILD (CP-B only)
55 Gy in 5 fx (includes
CP-A: 61, CP-A: 42% progression to
SD)
3 years CP-B
CP-B: 82, 1 CP-A: 8% progression to
year, 2 CP-C
years,
CP-B: 14% transient
3 years
progression to CP-C

CP-B: 57, CP-B: 33% progression to
1 year CP-C

CP-B: 33,
2 year
CP-B: 26,
3 year

177
TABLE 2. Select Prospective Clinical Trials of More Than 50 Patients Evaluating Role of Radiation in Hepatocellular Carcinoma (Continued)
178
VI/ Median
No. of Patients Tumor Size TVT Survival Toxicity (collective ‡ grade
Study Design (No. of lesions) CP Class (%) Dose/Fx (cm) (%) ORR (%) LC (%) OS (%) (months) 3)a
Takeda Phase II 90 A (91) 35–40 Gy/5 fx Median NR NR 96, 3 years 67, 3 year 54.7 8%
et al, diameter
B (8) Median 40 Gy 9% progression of CP score
201624 4
in 5 fx by 2 points
Feng et al, Phase II 90 A (77) 23–60 Gy in Median 3 18 99, 1 year 67, 1 year NR NR
201825 3–5 fx
B (23) Range 95, 2 years 36,
(0–13) 2 years
Hong et al, Phase II, 83b A (73) 15.1–67.5 GyE/ Median 30 NR 94, 2 years 77, 1 year 50 6%
201626 proton 15 fx diameter
therapy 5.0
44 HCC (58) B (21); no Median 58 GyE Range 63, 4% progression of CP class
cirrhosis in 15 fx (1.9–12.0) 2 years (A to B)
(7)

Kasuya Phase I and 124 (133) A (77) 15.1–67.5 GyE/ Median 17 NR 95, 1 year 90, 1 year 35.4 20%
et al, II, carbon 12–14 fx diameter
201727 ion 4.0
B (23) Median 52.8 Range 91, 3 years 50, 33% progression of CP
GyE in 4 fx (1.0–12.0) 3 years score at 3 months
90, 5 years 25, 25% progression of CP
5 years score at 6 months
Qadan et al
Abbreviations: CP, Child–Pugh; fx, fractions; GyE, Gy equivalent; LC, local control; NR, not reported; ORR, objective response rate (including complete and partial response); OS, overall survival;
RILD, radiation-induced liver disease; SD, stable disease; TVT, tumor vascular thrombus; VI, vascular invasion.
a
Toxicity was calculated as total grade 3 or higher toxicities as stated or all cumulative grade 3 or higher toxicities as reported, divided by patients in trial; therefore, it may represent multiple grade 3
toxicities within the same patient.
b
Hong et al26 included 39 patients with intrahepatic cholangiocarcinoma. Toxicities are calculated from all patients in the trial (83 patients); other reported values are from patients with HCC only.

radiation-induced liver disease (RILD). Seven patients had SBRT, and selective internal radiation therapy (SIRT).
grade 5 toxicity, with liver failure in five patients (including Thirty-seven studies with a total of 2,513 patients were
“massive” tumor thrombus progression in two patients, included in the analysis. Although no differences in 1-
which probably contributed to the decline in liver function). year OS were seen, the response rates to treatment were
A phase II study by Takeda et al24 of 90 patients treated in best in the cohort that received SBRT, at 70%, compared
Japan with tumors smaller than 4 cm was reported. Ninety- with SIRT at 30%. A single-institution retrospective study
one percent were in CTP class A, and patients received evaluated outcomes of patients receiving transarterial
a median dose of 40 Gy in five fractions. The median survival chemoembolization (TACE) or SBRT between 2006 and
was 54.7 months, with a 3-year OS of 67%. Eight percent 2014. Propensity score analysis was performed to compare
had grade 3+ toxicity (laboratory only), and 9% had pro- outcomes and adjust for imbalances in a cohort of 209
gression of CTP scores by more than two points. patients with one or two tumors (TACE, 84 patients; SBRT,
125 patients). Patients were similarly matched in terms of
Given concern for liver complications and RILD with SBRT, underlying liver disease and baseline liver function. However,
investigators at the University of Michigan evaluated a risk- patients undergoing SBRT were slightly older and had smaller
adapted model such that the radiation dose selected would tumors. The 1-year and 2-year local control favored SBRT
result in predicted risk of RILD of less than 15%. All patients (97% vs. 47% at 1 year and 91% vs. 23% at 2 years). Grade
underwent indocyanine green testing before beginning 3+ toxicity was seen in 13% of patients after TACE and 8% of
radiation and again after completing three treatments. patients after SBRT. Given the retrospective nature of these
Based on the authors’ findings, a determination was made studies, further prospective studies and comparative evalu-
regarding whether the final two fractions would be delivered ation of liver-directed therapies are warranted.
as planned, delivered with a reduced dose, or not delivered.
Ninety patients with 116 tumors were enrolled in this pro- REGIONAL LIVER-DIRECTED THERAPIES
spective phase II study. Local control was 99% and 95% at Recent data, reflected in the Hong Kong Liver Cancer
1 and 2 years, respectively, and OS was 67% and 36% at 1 staging system, underscore the growing role of liver-directed
and 2 years, respectively. The five patients whose disease therapies for carefully selected patients with liver-dominant
recurred all had tumor vascular thrombosis.25 Using bio- disease, limited PVTT, and preserved liver synthetic func-
markers such as indocyanine green to adapt liver radio- tion.29 Liver-directed therapies for PVTT include percuta-
therapy is a promising area for future research and may neous thermal ablation and transarterial therapies. Data on
mitigate some of the toxicities associated with radiotherapy. percutaneous thermal ablation for PVTT are limited,30 and
Another technique for potentially reducing the risk of liver the procedure is technically challenging for central PVTT
toxicity is the use of particle-based radiotherapy. Both protons because of the proximity to the bile ducts and hepatic
and carbon ions have been prospectively evaluated in the vasculature. Transarterial therapies include transarterial
treatment of patients with HCC. A multi-institutional phase II embolization with microspheres, TACE, and transarterial
trial by Hong et al26 evaluated 83 patients with HCC or administration of 90Y-labeled microspheres, often called
intrahepatic cholangiocarcinoma (44 patients with HCC). transarterial radioembolization (TARE) or SIRT. Because
Thirty percent had tumor vascular thrombosis, and 20% were TACE and TARE are the most commonly offered liver-
in CTP class B. All patients were treated with a median dose of directed therapies in clinical practice, the rationale and
58 Gy equivalents (GyE) in 15 fractions. Two-year local control data for both therapies are discussed below.
was 95%, with no nonhematologic toxicities seen in the HCC Transarterial Chemoembolization for PVTT
cohort. Kasuya et al27 reported outcomes from a phase I and II
study demonstrating safety and efficacy of carbon ions in Two randomized controlled trials, several meta-analyses,
a cohort of 127 patients with HCC. The phase I portion of the and smaller studies have demonstrated increased OS after
study established 52.8 GyE in four fractions, which was used TACE for patients with intermediate-stage HCC without
in the phase II portion of the trial. Local control was 95%, 91%, PVTT (BCLC class B).3 TACE is the current liver-directed
and 90% at 1, 3, and 5 years, respectively. therapy of choice to bridge patients to transplantation3 and
can be used for tumor downstaging.31 HCC tumors are fed
Radiotherapy Compared With Other Modalities primarily by hepatic arteries, and the remaining liver pa-
For patients with advanced HCC, no prospective trials have renchyma derives blood supply from the portal circulation.
compared radiotherapy with other liver-directed modalities, The unique dual supply allows transarterial delivery of high-
and much of the available comparative data stems from dose chemotherapeutic agents, usually cisplatin or doxo-
systematic reviews, database analyses, and retrospective rubicin emulsified in ethiodized oil or eluted by drug-loaded
single-institution experiences. A meta-analysis and sys- microspheres, to the tumor, thereby maximizing drug delivery
tematic review by Rim et al28 evaluated the efficacy and while minimizing collateral damage. Because of its embolic
toxicity of three-dimensional conformal radiation therapy, nature, both BCLC and the National Comprehensive Cancer

Qadan et al
Network recommend against TACE for patients with PVTT,32 Although there are no large randomized controlled trials
lest the concurrent interruption of the hepatic arterial flow that compare TACE with TARE for patients with PVTT,
and portal thrombosis incite severe hepatic ischemia. nonrandomized, prospective, and retrospective data have
However, a more nuanced approach that differentiates reported a slightly superior OS and progression-free survival
between intrahepatic, branch vessel, and main PVTT and with TARE, with notably fewer side effects.40,41 In the largest
stratifies patients based on their CTP score is needed. The prospective, nonrandomized study, Salem et al42 reported
safety of superselective TACE for patients with PVTT has a median OS of 10.4 months (95% CI, 7.2–16.6) for patients
been reported in several studies and is associated with in CTP class A. As expected, patients in CTP class A with
increased OS.33-36 In a large prospective, nonrandomized branch vessel PVTT did remarkably better than those with
study, Luo et al36 reported higher 1-year and 2-year survival main PVTT (16.6 month vs. 7.7 months, respectively).42
rates with TACE compared with best supportive care (30.9% Unfortunately, patients in CTP class B fared poorly, irre-
and 9.2% vs. 3.8% and 0%, respectively; p , .001), with spective of the extent of PVTT (6.5 months vs. 4.5 months,
a median OS of 7.1 months in the TACE arm and 4.1 months respectively), underscoring the prognostic role of intact
in the best supportive care arm (p , .001) and a 30-day baseline liver function.42 Other studies performed in Europe
mortality of 1.2%. Of note, although all patients in the TACE and North America have reported comparable median OS of
arm did better than those in the conservative arm, TACE 10 to 13 months for patients with PVTT.43,44 Interventional
patients with only branch vessel PVTT did significantly radiologists continue to refine the technique and dosimetry,
better than those with main PVTT, including greater tumor and recent studies report a median OS of 45 months after
response (77% vs. 51%, respectively) and increased OS selective delivery of radiation doses greater than 200 Gy
(median OS, 10.2 months vs. 5.3 months). Chung et al34 (ablative TARE).45 Reported objective tumor responses
demonstrated the importance of preserved baseline liver ranging from 20% to 77%, with successful downstaging to
function to ensure improved outcomes, with patients in CTP allow resection with curative intent, have been reported in
class B doing poorly irrespective of treatment delivered well-selected patients.42-44,46 Although larger studies are
(2.8 months vs. 1.9 months). Other studies, including needed, it appears that, for patients in CTP class A with
a meta-analysis, reported a median OS of 7.1 to 14.9 branch vessel PVTT, TARE probably confers a superior OS of
months, with increased survival for patients with branch 16.6 months compared with 10.2 months for TACE, with
vessel PVTT.33,37 The reported incidence of grade 3 and fewer side effects and adverse events, thereby making it the
higher complications was uncommon, predominantly re- preferred liver-directed therapy option for patients with HCC
stricted to postembolization syndrome (pain, fever, and and associated PVTT and preserved synthetic function.
nausea).33,36
Rationale for Using Liver-Directed Therapies for Patients
With HCC and PVTT
Transarterial Radioembolization for PVTT
As therapeutic options evolve and new data emerge, the
TARE is a new technique for transarterial delivery of ther-
limitations of the BCLC system for patients with PVTT have
apeutic radiation doses to the tumor. A pure beta-emitting
become evident. A more nuanced approach that takes into
isotope, 90Y decays to 90Zr, with a half-life of 64 hours.
account the extent of the PVTT, underlying liver function as
Although it is colloquially described as radioembolization or
measured by the CTP score, and the presence or absence of
TARE, the radiolabeled microspheres are less than 60 mm in
extrahepatic metastases as critical factors is used to provide
diameter and thus do not alter vascular dynamics for pa-
a more tailored approach for patients within this large
tients with PVTT. In fact, preservation of hepatic arterial flow
spectrum of BCLC class C disease. Emerging data on im-
is critical for radiation-induced free radical cell death. Be-
proved outcomes in the treatment of this heterogenous
cause HCC tumors are hypervascular tumors, 90Y particles
group of patients will probably be included in future staging
preferentially flow to the tumor, delivering 100 to 3,000 Gy in
systems as granularity improves.3
the process. However, because the median depth of pen-
etration for the 90Y particles is only 2.5 mm, the surrounding The synergistic and individual roles of TARE and sorafenib
liver parenchyma is spared despite the lethal dose delivered have been a topic of active research. Two recently com-
to the tumor. Furthermore, the lack of a true embolic effect pleted randomized control trials, SARAH and SIRveNIB,
provides a clear advantage over TACE, and TARE is well compared TARE with sorafenib.47,48 SIRveNIB accrued
tolerated with limited grade 1 or 2 toxicities, including patients from the Asia-Pacific region, whereas SARAH
predominantly fatigue and elevated liver enzymes.38,39 Se- accrued patients across France. Both trials were designed
rious complications, such as radiation pneumonitis, gastric to be superiority trials. However, the reported median OS
ulceration, and radiation-induced liver toxicity, are rare in results for TARE and sorafenib were statistically similar, and
experienced hands (, 2%) and with careful preprocedure neither trial met its primary endpoint. Tumor response rates
planning.39 were significantly higher with TARE, but this too did not

translate to increased progression-free survival. Of note, the 1.5 months, respectively. Interestingly, complication rates
inclusion criteria in these trials were overly broad and in- mirrored technical complexity and were 31.9%, 50.0%, and
cluded patients with BCLC class B and C and patients with 71.4%, respectively, representing the technical challenges
extrahepatic metastases, which may have negated the associated with PVTT resection in healthy and highly se-
beneficial impact of TARE for patients with PVTT specifi- lected patients. Although there were no statistically signif-
cally. In addition, only a minority of patients had PVTT, and icant differences in any of the authors’ reported figures,
subset analysis was not performed. Therefore, the potential these data reflected the low survival figures seen in patients
clinical benefit in patients with PVTT, especially those with with PVTT despite complete surgical extirpation of disease,
segmental PVTT only, was not evaluated. Most importantly, highlighting the aggressive nature of PVTT and need for
both trials reported fewer grade 3 or higher adverse events complementary therapies.
with TARE than with sorafenib (20.8% vs. 35.2% in SIR-
Currently, there are no prospective data comparing surgical
veNIB and 27.7% vs. 50.6% in SARAH) in addition to the
resection against other treatment modalities. However, the
distinct possibility of tumor downstaging to allow potentially
criteria for safe resection and improved survival compared
curative resection. Given that most patients in BCLC class C
with no treatment (best supportive care) continue to ex-
survive less than a year from diagnosis, the importance of
pand.53 Overall, the median survival time among the studies
quality of life in avoidance of side effects and complications
was 25.4 months for highly selected patients.54 In a Japa-
cannot be overstated.
nese nationwide analysis between 2000 and 2007, 6,474
SURGICAL APPROACHES patients with HCC and PVTT were included, of whom 2,093
Surgical approaches for the treatment of HCC with PVTT underwent surgical resection.55 In a propensity-matched
have been described since the early 1980s and 1990s, analysis performed by the authors comparing surgical re-
albeit with dramatically worse survival outcomes compared section with nonsurgical therapy, surgical resection was
with HCC without PVTT.49,50 Numerous retrospective series associated with a median survival of 2.9 years, compared
have evaluated multiple aspects related to surgical resection with 1.1 years among patients in the nonsurgical group.
of HCC with PVTT. Importantly, however, the authors were unable to demon-
strate an association with improved survival in patients with
The tumor biologic determinants for resection of HCC and
PVTT extending into or beyond the main venous bifurcation,
PVTT in highly selected patients arguably are the most
representing the aggressive nature of extensive PVTT at
important determinants of oncologic outcomes. However,
diagnosis.
the technical criteria for surgical treatment of HCC with
PVTT depend largely on the degree of tumor extension These data potentially support surgical resection for highly
within the vasculature. Surgical resection of HCC with PVTT selected patients, although application to Western cohorts
is considered technically challenging51 and is increasingly remains largely unknown. The retrospective nature of all
performed in Asian centers in particular. The objectives of these studies does not preclude confounders that represent
surgery have broadly included complete tumor extirpation exceptional tumor biology, which reflects the broad mes-
and prevention of downstream sequelae of portal hyper- sage in this context. In addition, complication rates in the
tension secondary to obstructive tumor thrombus. Broadly, 50% range merit pause before embarking on management
the options depend on the extent of PVTT (ipsilateral vs. of this technically challenging entity.
extension to, or beyond, the main portal vein bifurcation).
Combination Strategies With Surgical Resection
Varying extents of tumor thrombus, the need for in-
dividualized surgical approaches, and variable manage- Combinations of complementary modalities and surgical
ment modalities have thus limited prospective evaluation of resection have been reported and are potentially associated
outcomes. Surgical options include the following: with increased benefit.56
• Hepatectomy with en bloc resection of ipsilateral tumor For example, TACE after resection has been associated with
thrombus an increase in median OS and decline in overall disease
• Resection of tumor and PVTT extending to or beyond the recurrence, especially for patients with large tumors asso-
main portal vein bifurcation, treated with en bloc vascular ciated with vascular invasion in a study by Peng et al57
resection, repair, and reconstruction (13 months vs. 9 months).
• Tumor thrombectomy in PVTT extending to or beyond the Sorafenib as an adjuvant systemic therapy initially showed
main portal vein bifurcation. promising results in orthotopic animal models. However,
In a report by Chok et al,52 the authors compared these three whether these results will eventually materialize to show
modalities and noted median OS times of 10.1 months, 9.4 benefit after resection of HCC with PVTT remains to be
months, and 8.6 months, respectively. In addition, median determined.58 In a report by Tang et al,59 the authors ret-
disease-free survivals were 4.2 months, 3.8 months, and rospectively compared best supportive care, chemotherapy,

Qadan et al
surgery, and surgery with chemotherapy among 589 pa- Therefore, data are limited to living-donor liver transplantation
tients in a modern case series. The authors noted an as- in conjunction with aggressive downstaging protocols.
sociation with improved survival with any treatment In a report by Han et al,62 the authors evaluated use of
compared with no treatment and a greater effect with the concurrent chemoradiation in conjunction with hepatic
combination of surgery and chemotherapy. Zhou et al60 arterial infusion chemotherapy before living-donor liver
compared hepatectomy with thrombectomy versus hepa- transplantation. Eight patients underwent successful
tectomy with thrombectomy plus chemotherapy in a retro- transplantation after complete responses within their vas-
spective analysis that included other treatment modalities culature. One-year disease-free survival was an astounding
and showed that the addition of chemotherapy was asso- 87.5%, and the median survival was 33 months. The
ciated with substantially increased 1-year survival rates feasibility of liver transplantation for patients with down-
(70% vs. 47%). However, there were minimal differences at staged disease warrants evaluation in the modern era as
the 3-year mark for patients who received chemotherapy response rates to systemic therapies continue to improve.
compared with patients who did not (20% vs. 22%, However, this therapy is probably effective, once again, in
respectively). cases of exceptionally indolent tumor biology that is highly
In a rare prospective, randomized controlled trial in this responsive to neoadjuvant therapy.
setting, Wei et al61 evaluated neoadjuvant radiation for re- Similarly impressive results were established with SIRT
sectable HCC with PVTT in a multicenter controlled study before liver transplantation among four patients reported by
conducted in 2016 and 2017. Patients were randomly Levi Sandri et al.63 Tumor regression was noted among all
assigned to receive neoadjuvant radiation followed by four patients, with disease-free survival of 39.1 months.
hepatectomy versus hepatectomy alone for 164 patients When external beam radiation was examined in a cohort
distributed equally between both study arms. The authors being evaluated for liver transplantation by Chino et al,21 the
reported OS rates at 6, 12, 18, and 24 months of 89.0%, authors noted a complete pathologic response of 27%
75.2%, 43.9%, and 27.4%, respectively, among patients among 10 patients, with 100% survival after 5 years and
who received radiation compared with 81.7%, 43.1%, without recurrences observed.
16.7%, and 9.4%, respectively, in the surgery-alone group.
CONCLUSION
Corresponding disease-free survival rates were 56.9%,
33.0%, 20.3%, and 13.3% in the radiation group and Although diagnosis of HCC with PVTT or hepatic vein in-
42.1%, 14.9%, 5.0%, and 3.3% in the surgery-alone group. vasion portends a poor prognosis, recent advances in the
All findings were statistically significant and established treatment of patients with MVI has rendered this disease
a novel role for radiation in a neoadjuvant setting for pa- entity a treatable one with improved oncologic outcomes.
tients with PVTT followed by complete surgical resection. Importantly, inclusion of highly selected patients with em-
To date, this remains the only prospectively validated phasis on tumor biology is necessary to achieve outstanding
treatment algorithm and provides a potentially effective long-term results. In addition, treatment of HCC with PVTT is
treatment strategy for this patient population. Interestingly, multimodal, with expansive combinations and sequences
the authors were also able to establish that elevated that are becoming increasingly available to optimize out-
preradiation serum interleukin-6 levels were associated comes. However, it is important to note that there are con-
with refractory responses to therapy, thus helping select siderable variations in approach to the treatment of HCC with
patients who may not benefit from aggressive treatment. PVTT between the Eastern and Western hemispheres, which
may also be related to variability in the underlying disease
Liver Transplantation process and management preferences. Finally, careful
Currently, liver transplantation for patients with PVTT re- emphasis should be placed on maintaining patient quality of
mains contraindicated, given the limited availability of or- life, which cannot be underestimated in the aggressive
gans combined with the high recurrence rates noted earlier. treatment of HCC and liver-associated complications.

1
Department of Surgery, Massachusetts General Hospital and Harvard Motaz Qadan, MD, PhD, Division of Surgical Oncology, Department of
Medical School, Boston, MA Surgery, Massachusetts General Hospital, 55 Fruit St., Yawkey 7B,
2
Department of Radiology, Stanford University Medical Center, Palo Alto, Boston, MA 02114; Twitter: @motazqadan; email: mqadan@mgh.
CA harvard.edu.
3
Department of Medicine, Clinica Universidad de Navarra, Pamplona,
Spain
4
Department of Radiation Oncology, Duke University, Durham, NC


EDBK_280811.
REFERENCES
1. Torre LA, Bray F, Siegel RL, et al. Global cancer statistics, 2012. CA Cancer J Clin. 2015;65:87-108.
2. Forner A, Reig ME, de Lope CR, et al. Current strategy for staging and treatment: the BCLC update and future prospects. Semin Liver Dis. 2010;30:61-74.
3. Heimbach JK, Kulik LM, Finn RS, et al. AASLD guidelines for the treatment of hepatocellular carcinoma. Hepatology. 2018;67:358-380.
4. Chan SL, Chong CC, Chan AW, et al. Management of hepatocellular carcinoma with portal vein tumor thrombosis: review and update at 2016. World
J Gastroenterol. 2016;22:7289-7300.
5. Llovet JM, Bustamante J, Castells A, et al. Natural history of untreated nonsurgical hepatocellular carcinoma: rationale for the design and evaluation of therapeutic
trials. Hepatology. 1999;29:62-67.
6. Llovet JM, Zucman-Rossi J, Pikarsky E, et al. Hepatocellular carcinoma. Nat Rev Dis Primers. 2016;2:16018.
7. Llovet JM, Ricci S, Mazzaferro V, et al; SHARP Investigators Study Group. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med. 2008;359:378-390.
8. Kudo M, Finn RS, Qin S, et al. Lenvatinib versus sorafenib in first-line treatment of patients with unresectable hepatocellular carcinoma: a randomised phase
3 non-inferiority trial. Lancet. 2018;391:1163-1173.
9. Abou-Alfa GK, Meyer T, Cheng AL, et al. Cabozantinib in patients with advanced and progressing hepatocellular carcinoma. N Engl J Med. 2018;379:54-63.
10. Bruix J, Qin S, Merle P, et al; RESORCE Investigators. Regorafenib for patients with hepatocellular carcinoma who progressed on sorafenib treatment (RESORCE):
a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet. 2017;389:56-66.
11. Siegel AB, Cohen EI, Ocean A, et al. Phase II trial evaluating the clinical and biologic effects of bevacizumab in unresectable hepatocellular carcinoma. J Clin
Oncol. 2008;26:2992-2998.
12. Zhu AX, Park JO, Ryoo BY, et al; REACH Trial Investigators. Ramucirumab versus placebo as second-line treatment in patients with advanced hepatocellular
carcinoma following first-line therapy with sorafenib (REACH): a randomised, double-blind, multicentre, phase 3 trial. Lancet Oncol. 2015;16:859-870.
13. Zhu AX, Kang YK, Yen CJ, et al; REACH-2 study investigators. Ramucirumab after sorafenib in patients with advanced hepatocellular carcinoma and increased
α-fetoprotein concentrations (REACH-2): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol. 2019;20:282-296.
14. El-Khoueiry AB, Sangro B, Yau T, et al. Nivolumab in patients with advanced hepatocellular carcinoma (CheckMate 040): an open-label, non-comparative, phase
1/2 dose escalation and expansion trial. Lancet. 2017;389:2492-2502.
15. Zhu AX, Finn RS, Edeline J, et al; KEYNOTE-224 investigators. Pembrolizumab in patients with advanced hepatocellular carcinoma previously treated with
sorafenib (KEYNOTE-224): a non-randomised, open-label phase 2 trial. Lancet Oncol. 2018;19:940-952.
16. Finn RS, Ryoo BY, Merle P, et al; KEYNOTE-240 investigators. Pembrolizumab as second-line therapy in patients with advanced hepatocellular carcinoma in
KEYNOTE-240: a randomized, double-blind, phase III trial. J Clin Oncol. 2020;38:193-202.
17. Gershenson DM, Miller A, Brady W, et al. A randomized phase II/III study to assess the efficacy of trametinib in patients with recurrent or progressive low-grade
serous ovarian or peritoneal cancer. Ann Oncol. 2019;30:v851-v934.
18. Rimola J, Dı́az-González Á, Darnell A, et al. Complete response under sorafenib in patients with hepatocellular carcinoma: relationship with dermatologic adverse
events. Hepatology. 2018;67:612-622.
19. Finn RS, Merle P, Granito A, et al. Outcomes of sequential treatment with sorafenib followed by regorafenib for HCC: additional analyses from the phase III
RESORCE trial. J Hepatol. 2018;69:353-358.
20. Kirkpatrick JP, Kelsey CR, Palta M, et al. Stereotactic body radiotherapy: a critical review for nonradiation oncologists. Cancer. 2014;120:942-954.
21. Chino F, Stephens SJ, Choi SS, et al. The role of external beam radiotherapy in the treatment of hepatocellular cancer. Cancer. 2018;124:3476-3489.
22. Bujold A, Massey CA, Kim JJ, et al. Sequential phase I and II trials of stereotactic body radiotherapy for locally advanced hepatocellular carcinoma. J Clin Oncol.
2013;31:1631-1639.
23. Lasley FD, Mannina EM, Johnson CS, et al. Treatment variables related to liver toxicity in patients with hepatocellular carcinoma, Child–Pugh class A and B
enrolled in a phase 1–2 trial of stereotactic body radiation therapy. Pract Radiat Oncol. 2015;5:e443-e449.
24. Takeda A, Sanuki N, Tsurugai Y, et al. Phase 2 study of stereotactic body radiotherapy and optional transarterial chemoembolization for solitary hepatocellular
carcinoma not amenable to resection and radiofrequency ablation. Cancer. 2016;122:2041-2049.
25. Feng M, Suresh K, Schipper MJ, et al. Individualized adaptive stereotactic body radiotherapy for liver tumors in patients at high risk for liver damage: a phase 2
clinical trial. JAMA Oncol. 2018;4:40-47.
26. Hong TS, Wo JY, Yeap BY, et al. Multi-institutional phase II study of high-dose hypofractionated proton beam therapy in patients with localized, unresectable
hepatocellular carcinoma and intrahepatic cholangiocarcinoma. J Clin Oncol. 2016;34:460-468.

Qadan et al
27. Kasuya G, Kato H, Yasuda S, et al; Liver Cancer Working Group. Progressive hypofractionated carbon-ion radiotherapy for hepatocellular carcinoma: combined
analyses of 2 prospective trials. Cancer. 2017;123:3955-3965.
28. Rim CH, Kim CY, Yang DS, et al. Comparison of radiation therapy modalities for hepatocellular carcinoma with portal vein thrombosis: a meta-analysis and
systematic review. Radiother Oncol. 2018;129:112-122.
29. Yau T, Tang VY, Yao TJ, et al. Development of Hong Kong liver cancer staging system with treatment stratification for patients with hepatocellular carcinoma.
Gastroenterology. 2014;146:1691-1700 e3.
30. Long J, Zheng JS, Sun B, et al. Microwave ablation of hepatocellular carcinoma with portal vein tumor thrombosis after transarterial chemoembolization:
a prospective study. Hepatol Int. 2016;10:175-184.
31. Heckman JT, Devera MB, Marsh JW, et al. Bridging locoregional therapy for hepatocellular carcinoma prior to liver transplantation. Ann Surg Oncol. 2008;
15:3169-3177.
32. Bruix J, Sherman M; American Association for the Study of Liver Diseases. Management of hepatocellular carcinoma: an update. Hepatology. 2011;
53:1020-1022.
33. Pinter M, Hucke F, Graziadei I, et al. Advanced-stage hepatocellular carcinoma: transarterial chemoembolization versus sorafenib. Radiology. 2012;
263:590-599.
34. Chung GE, Lee JH, Kim HY, et al. Transarterial chemoembolization can be safely performed in patients with hepatocellular carcinoma invading the main portal
vein and may improve the overall survival. Radiology. 2011;258:627-634.
35. Gorodetski B, Chapiro J, Schernthaner R, et al. Advanced-stage hepatocellular carcinoma with portal vein thrombosis: conventional versus drug-eluting beads
transcatheter arterial chemoembolization. Eur Radiol. 2017;27:526-535.
36. Luo J, Guo RP, Lai EC, et al. Transarterial chemoembolization for unresectable hepatocellular carcinoma with portal vein tumor thrombosis: a prospective
comparative study. Ann Surg Oncol. 2011;18:413-420.
37. Leng JJ, Xu YZ, Dong JH. Efficacy of transarterial chemoembolization for hepatocellular carcinoma with portal vein thrombosis: a meta-analysis. ANZ J Surg.
2016;86:816-820.
38. Gao R, Gabr A, Mouli S, et al. Toxicity and survival of hepatocellular carcinoma patients with hepatitis B infection treated with yttrium-90 radioembolization: an
updated 15-year study. J Vasc Interv Radiol. 2020;31:401-408.e1.
39. Padia SA, Lewandowski RJ, Johnson GE, et al; Society of Interventional Radiology Standards of Practice Committee. Radioembolization of hepatic malignancies:
background, quality improvement guidelines, and future directions. J Vasc Interv Radiol. 2017;28:1-15.
40. Carr BI, Kondragunta V, Buch SC, et al. Therapeutic equivalence in survival for hepatic arterial chemoembolization and yttrium 90 microsphere treatments in
unresectable hepatocellular carcinoma: a two-cohort study. Cancer. 2010;116:1305-1314.
41. Salem R, Lewandowski RJ, Kulik L, et al. Radioembolization results in longer time-to-progression and reduced toxicity compared with chemoembolization in
patients with hepatocellular carcinoma. Gastroenterology. 2011;140:497-507 e2.
42. Salem R, Lewandowski RJ, Mulcahy MF, et al. Radioembolization for hepatocellular carcinoma using yttrium-90 microspheres: a comprehensive report of long-
term outcomes. Gastroenterology. 2010;138:52-64.
43. Sangro B, Carpanese L, Cianni R, et al; European Network on Radioembolization With Yttrium-90 Resin Microspheres (ENRY). Survival after yttrium-90 resin
microsphere radioembolization of hepatocellular carcinoma across Barcelona clinic liver cancer stages: a European evaluation. Hepatology. 2011;54:868-878.
44. Iñarrairaegui M, Thurston KG, Bilbao JI, et al. Radioembolization with use of yttrium-90 resin microspheres in patients with hepatocellular carcinoma and portal
vein thrombosis. J Vasc Interv Radiol. 2010;21:1205-1212.
45. Cardarelli-Leite L, Chung J, Klass D, et al. Ablative transarterial radioembolization improves survival in patients with HCC and portal vein tumor thrombus.
Cardiovasc Intervent Radiol. 2020;43:411-422.
46. Riaz A, Gabr A, Abouchaleh N, et al. Radioembolization for hepatocellular carcinoma: statistical confirmation of improved survival in responders by landmark
analyses. Hepatology. 2018;67:873-883.
47. Chow PKH, Gandhi M, Tan SB, et al; Asia-Pacific Hepatocellular Carcinoma Trials Group. SIRveNIB: Selective internal radiation therapy versus sorafenib in Asia-
Pacific patients with hepatocellular carcinoma. J Clin Oncol. 2018;36:1913-1921.
48. Vilgrain V, Pereira H, Assenat E, et al; SARAH Trial Group. Efficacy and safety of selective internal radiotherapy with yttrium-90 resin microspheres compared with
sorafenib in locally advanced and inoperable hepatocellular carcinoma (SARAH): an open-label randomised controlled phase 3 trial. Lancet Oncol. 2017;
18:1624-1636.
49. Kumada K, Ozawa K, Okamoto R, et al. Hepatic resection for advanced hepatocellular carcinoma with removal of portal vein tumor thrombi. Surgery. 1990;
108:821-827.
50. Yamaoka Y, Kumada K, Ino K, et al. Liver resection for hepatocellular carcinoma (HCC) with direct removal of tumor thrombi in the main portal vein. World J Surg.
1992;16:1172-1176, discussion 1177.
51. Shi J, Lai EC, Li N, et al. Surgical treatment of hepatocellular carcinoma with portal vein tumor thrombus. Ann Surg Oncol. 2010;17:2073-2080.
52. Chok KS, Cheung TT, Chan SC, et al. Surgical outcomes in hepatocellular carcinoma patients with portal vein tumor thrombosis. World J Surg. 2014;38:490-496.
53. Liu PH, Huo TI, Miksad RA. Hepatocellular carcinoma with portal vein tumor involvement: best management strategies. Semin Liver Dis. 2018;38:242-251.
54. Jiang JF, Lao YC, Yuan BH, et al. Treatment of hepatocellular carcinoma with portal vein tumor thrombus: advances and challenges. Oncotarget. 2017;
8:33911-33921.

55. Kokudo T, Hasegawa K, Matsuyama Y, et al; Liver Cancer Study Group of Japan. Survival benefit of liver resection for hepatocellular carcinoma associated with
portal vein invasion. J Hepatol. 2016;65:938-943.
56. Kamiyama T, Kakisaka T, Orimo T, et al. Hepatectomy for hepatocellular carcinoma with portal vein tumor thrombus. World J Hepatol. 2017;9:1296-1304.
57. Peng BG, He Q, Li JP, et al. Adjuvant transcatheter arterial chemoembolization improves efficacy of hepatectomy for patients with hepatocellular carcinoma and
portal vein tumor thrombus. Am J Surg. 2009;198:313-318.
58. Feng YX, Wang T, Deng YZ, et al. Sorafenib suppresses postsurgical recurrence and metastasis of hepatocellular carcinoma in an orthotopic mouse model.
Hepatology. 2011;53:483-492.
59. Tang ZY, Zhou BH, Wang W, et al. Curative analysis of several therapeutic methods for primary hepatocellular carcinoma with portal vein tumor thrombus.
Hepatogastroenterology. 2015;62:703-709.
60. Zhou Q, Wang Y, Zhou X, et al. Prognostic analysis for treatment modalities in hepatocellular carcinomas with portal vein tumor thrombi. Asian Pac J Cancer Prev.
2011;12:2847-2850.
61. Wei X, Jiang Y, Zhang X, et al. Neoadjuvant three-dimensional conformal radiotherapy for resectable hepatocellular carcinoma with portal vein tumor thrombus:
a randomized, open-label, multicenter controlled study. J Clin Oncol. 2019;37:2141-2151.
62. Han DH, Joo DJ, Kim MS, et al. Living donor liver transplantation for advanced hepatocellular carcinoma with portal vein tumor thrombosis after concurrent
chemoradiation therapy. Yonsei Med J. 2016;57:1276-1281.
63. Levi Sandri GB, Ettorre GM, Colasanti M, et al. Hepatocellular carcinoma with macrovascular invasion treated with yttrium-90 radioembolization prior to
transplantation. Hepatobiliary Surg Nutr. 2017;6:44-48.

Moving Beyond Chemotherapy for

Pancreaticobiliary Tumors: Targeted and
Immunotherapy Strategies
Rebecca Allen, BS1; Naama Halpern, MD2; Sandra Algaze3; Talia Golan, MD2; Anthony B. El-Khoueiry, MD3; and
Rachna T. Shroff, MD, MS4
overview
Pancreaticobiliary cancers are a group of malignancies affecting the pancreas and biliary tract and are often
associated with poor prognosis. Existing treatment strategies for these malignancies are limited. However,
with the development of more advanced genomic analysis techniques, several mutations have been identified
that may be targeted for the development of novel treatments. Key targets of interest include DNA damage
repair (DDR) pathways for both pancreatic ductal adenocarcinoma (PDAC) and biliary tract cancer (BTC) as
well as isocitrate dehydrogenase 1 (IDH1) and fibroblast growth factor receptor (FGFR) in BTC and mismatch
repair (MMR) genes and germline mutations in PDAC. Additionally, a better understanding of the immune
microenvironment of pancreatic and biliary cancers has revealed cell types and signaling pathways that may
be leveraged for treatment. This includes PD-L1 and CTLA-4 immune checkpoints, tumor-associated
macrophages (TAMs), myeloid-derived suppressor cells (MDSCs), and dendritic cells (DCs). Together, tar-
geted agents and immunotherapy strategies have the potential to significantly improve the existing treatment
landscape for pancreaticobiliary cancers.
INTRODUCTION pancreaticobiliary tumors and supports the need for

Pancreaticobiliary cancers are a group of malignancies more individualized treatment approaches.
affecting the pancreas and biliary tract, including the In the advanced setting, treatment options are limited
gallbladder and intrahepatic and extrahepatic bile for all pancreaticobiliary cancers. For patients with
ducts. Pancreatic cancer is one of the leading causes metastatic pancreatic cancer and good performance
of cancer deaths worldwide, with a 5-year survival rate status, the current standard-of-care first-line treat-
between 9% and 10%.1,2 BTC, although not as ments are FOLFIRINOX (leucovorin, 5-fluorouracil,
prevalent as pancreatic cancer, has an increasing irinotecan, oxaliplatin), gemcitabine with albumin-
incidence and poor prognosis, with a 5-year survival bound paclitaxel, and gemcitabine with cisplatin for
rate between 5% and 15% depending on stage.3 those with known BRCA1/2 or PALB2 mutations. For
Both pancreatic cancer and BTCs are often not di- patients with pancreatic cancer and a poor performance
agnosed until advanced stages, when patients are no status, the standard of care is limited to gemcitabine,
longer candidates for curative resection. This occurs capecitabine, or continuous 5-fluorouracil infusion.4 For
in part because the presentation of symptoms as- patients with advanced BTCs, standard-of-care first-line
sociated with pancreaticobiliary cancer varies widely, treatment is the combination gemcitabine plus cis-
and many patients are initially asymptomatic. Ad- platin.5 Second-line treatments are limited, with many
ditionally, for BTCs, presentation and risk factors treatments failing to exceed the benefit of active
Author affiliations differ between gallbladder cancer (GBC), extrahe- symptom control.3,6 The poor prognosis and lack of
and support patic cholangiocarcinoma (ECC), and intrahepatic diverse treatment options associated with pan-
information (if
cholangiocarcinoma (ICC). For example, the risk fac- creaticobiliary cancers demonstrate the immense need
applicable) appear to develop an improved treatment paradigm for these
at the end of this tors of GBC include gallstones, gallbladder polyps,
malignancies, especially in the refractory setting.
article. obesity, and diabetes, whereas cirrhosis or infection with
Accepted on May 1, the liver flukes Opisthorchis viverrini or Clonorchis
BILIARY TRACT CANCER
2020 and published sinensis can be risk factors for cholangiocarcinoma.
at ascopubs.org on Current Targets
GBC and ICC are usually incidental findings, whereas
doi.org/10.1200/ ECC typically presents with obstructive jaundice.3 Because of efforts to better understand the patho-
EDBK_280901 This variability contributes to complexities in treating genesis of BTCs and identify targetable mutations,

Allen et al
patients with advanced ICC, whole genome sequencing

revealed translocations of FGFR in three of the six patients.
The investigators treated one of these patients with an FGFR
• Pancreaticobiliary tumors have poor prognosis inhibitor, which achieved stable disease in the patient.12
and limited treatment options.
Egan et al13 conducted exome sequencing on 308 chol-
• Promising targets in biliary tract cancer include angiocarcinoma tumors and identified mutations in FGFR2
IDH1, FGFR, and HER2. and FGFR4. The patient with the mutated FGFR2 tumor had
• The majority of mutations seen in PDAC are a sustained tumor response when treated with an FGFR
somatic, but there are several targetable inhibitor. FGFR2 fusions are involved in up to 15% of ICC,
germline mutations as well. whereas FGFR4 overexpression is seen in approximately
• DDR genes may serve as targets for treatment of 50% of all cholangiocarcinoma cases and is associated with
both biliary tract cancer and pancreatic poor prognosis.16 The prevalence of these mutations and
adenocarcinoma. the benefit of targeted therapy in individual cases make
• Immune checkpoint inhibition and modulation FGFR a favorable therapeutic target.
of cells contributing to the immune microen- HER In an effort to elicit the therapeutic value of targeting
vironment are promising immunotherapy HER2, Galdy et al17 performed a systematic review and
strategies for the treatment of pancreaticobiliary
meta-analysis of HER2 and HER3 overexpression and
cancer.
amplification in BTCs. This analysis found a HER3 over-
expression rate of 27.9% and an overall expression rate of
HER2 of 26.5%, with a higher HER2 overexpression in
numerous genomic analysis studies of these tumors have
extrahepatic BTC. Additionally, HER2 overexpression ap-
been conducted. Although mutations have been reported in
pears in up to 20% of extrahepatic BTC cases, with ap-
all forms of BTC, there is variation across GBC, ECC, and
proximately 60% HER2 amplified. This analysis contributes
ICC. In a study of eight cases of liver fluke–associated
to the foundation for clinical trials investigating HER2/HER3
cholangiocarcinoma and 46 individual cholangiocarcinoma
targeted therapy in BTC.
cases, mutations to known cancer-related genes TP53,
KRAS, and SMAD4 occurred most often. Additionally, several Other targets Several other possible targets have been
genes related to histone modification, G protein signaling, identified through genomic analysis. These include muta-
and genome stability were mutated in two or more cases.7 tions to angiogenesis growth factors and receptors and the
Jiao et al8 performed exomic sequencing of 32 ICC cases and KRAS–BRAF–MEK–ERK pathway. The most common al-
found inactivation of one or more chromatin remodeling terations include point mutations to KRAS and BRAF and
genes in 47% of cases. These studies reveal the prominent overexpression of EGFR and VEGFR. The prevalence of
role of chromatin remodeling in the development of BTCs and these mutations and alterations varies between GBC, ECC,
may serve as the basis for targeted therapy in the future. and ICC. KRAS mutations occur most often in ECC, BRAF
mutations in ICC, and EGFR mutations in GBC.3,9,18 Al-
Isocitrate dehydrogenase 1 Multiple studies of BTC have though KRAS mutations are prevalent in BTC, targeting
found hotspot mutations in either IDH1 or IDH2. In a cohort these mutations remains difficult.3 However, there is evi-
of 260 BTC cases (145 ICC, 86 ECC, 29 GBC), IDH1 and dence of successful BRAF + MEK inhibition for treatment of
IDH2 mutations were identified only in ICC.9 Others have BRAF V600E–mutated tumors.19 Additionally, DDR
reported IDH1/2 mutations in 18% to 24% of ICC cases and mechanisms are expected to be deficient in about 25% of
IDH1/2 substitutions in 15% to 23% of ICCs and 3% to 4% patients with BTC.20 This wide range of targets identified
of ECCs.10 IDH mutations were found to be significantly provides an excellent basis for the development of novel
associated with prognosis in a study that reported a 3-year therapeutic options for patients with BTC.
survival of 33% in patients whose tumors had IDH1 or IDH2
mutations compared with a 3-year survival of 81% in those Current and Published Studies
whose tumors expressed wild-type (WT) IDH genes.8 As the development of BTCs is further understood, the door
However, another study analyzing 85 ICC tumor samples has been opened for more individualized, targeted therapy.
determined that IDH1 mutations correlated with a beneficial Key contributors to biliary tract tumorigenesis in GBC, ECC,
prognosis and smaller tumor size.11 Although the prognostic and ICC that are promising include IDH1/2, FGFR, HER2,
value of IDH mutations is unclear, IDH mutations remain and several components of cell proliferation pathways.
promising therapeutic targets. Several agents targeting these mutations are under
Fibroblast growth factor receptor Several genome-wide investigation.17,21-24
sequencing studies have identified the presence of (FGFR IDH1 An oral, targeted inhibitor of mutant IDH1, ivosidenib,
mutations and fusions in BTCs.9,12-15 In a study of six was first investigated in a phase I study for patients with

Targeted and Immunotherapy Strategies for Pancreaticobiliary Tumors
mutant IDH1 solid tumors. For the 73 patients enrolled on infigratinib or other FGFR-targeting agents who develop
the study with cholangiocarcinoma, no dose-limiting tox- resistance, TAS-120 may provide another line of treatment.
icities were reported. Median progression-free survival Several other targeted agents are under investigation, in-
(PFS) was 3.8 months (95% CI, 3.6–7.3), median overall cluding derazantinib (NCT01752920) and debio 1347
survival (OS) was 13.8 months (95% CI, 11.1–29.3; data (NCT01948297).
censored for 66% of patients), and four patients had
a partial response (PR).21 ClarIDHy, the recently completed HER2 Case studies have demonstrated the positive effect of
phase III randomized study comparing ivosidenib and anti-HER2 monoclonal antibodies in the treatment of HER2-
placebo in the second- or third-line settings, included 185 positive cholangiocarcinoma.29 However, several studies
patients (ivosidenib arm, 124 patients; placebo arm, 61 investigating the role of HER2-targeted therapy in the
patients). The study met its primary endpoint of PFS (HR, treatment of bile duct cancer yielded disappointing results.
0.37; 95% CI, 0.25–0.54; p , .001). Median PFS was Additionally, recent studies with afatinib (NCT01679405)
2.7 months in the ivosidenib arm and 1.4 months for and trastuzumab (NCT00478140) in BTCs were terminated
placebo.23,24 These results suggest a promising role for early because of lack of signal and slow accrual, re-
ivosidenib in the treatment of IDH1-mutated chol- spectively. However, preliminary data from the MyPathway
angiocarcinoma. In addition to ivosidenib, several other tar- basket study suggest that the combination of pertuzumab
geted treatments for IDH1 mutations are under investigation, and trastuzumab has activity in BTC with HER2 mutations,
including BAY1436032 (NCT0274608) and dasatinib amplification, or overexpression. In the preliminary analysis,
(NCT02428855). for the eight patients with HER2-amplified or -overexpressed
tumors, ORR was 37.5% and median PFS was 4.2 months
FGFR Fibroblast growth factor and FGFR have wide-
(95% CI, 1.2–5.4). For those with HER2-mutated disease
reaching biologic effects, making FGFR a recent area of
(3 patients), ORR was 33.3%, with a median PFS of
interest in several clinical trials. A multicenter, phase II study
2.8 months (95% CI, 1.4–2.8).30 Results of the TreeTopp
of infigratinib, a selective pan-FGFR kinase inhibitor, in-
study investigating the role of varlitinib, a small molecule
cluded 61 patients with FGFR2 fusion, mutated, or am-
HER1/2/4 inhibitor, in combination with capecitabine
plified advanced cholangiocarcinoma. Overall response rate
for advanced or metastatic BTC are pending, although
(ORR) was 14.8%, with a median PFS of 5.8 months (95%
this is in an unselected, second-line patient population
CI, 4.3–7.6), and all responders had tumors with FGFR2
(NCT03093870). Likewise, results of ongoing studies of
fusions.22 Pemigatinib, a selective FGFR1, 2, and 3 in-
bispecific antibodies targeting HER2 such as PRS-343
hibitor, was evaluated in a phase II study of patients with
(NCT03330561), ZW25 (NCT02892123), and ZW49
previously treated locally advanced or metastatic chol-
(NCT03821233) are eagerly awaited.
angiocarcinoma. Of the 146 patients enrolled in the study,
107 had disease with FGFR2 gene rearrangements or fu- Other targets Although numerous studies targeting angio-
sions. The ORR for these patients was 35.5% (95% CI, genesis pathways, c-MET, and BRAF as well as MEK, in
26.5–45.5), with three complete responses and a median BTCs have yielded disappointing results,3,6 encouraging
PFS of 6.9 months (95% CI, 6.2–9.6).25 Infigratinib has results were reported from the ROAR basket study of
U.S. Food and Drug Administration (FDA) breakthrough combination treatment with dabrafenib, a BRAF inhibitor,
designation for patients with cholangiocarcinoma, and and trametinib, a MEK inhibitor, in patients with BRAF
pemigatinib recently received FDA approval for the treat- V600E–mutated cholangiocarcinoma.19 Patients included
ment of cholangiocarcinoma with a FGFR2 fusion or other in this study had advanced or metastatic cancer and had
rearrangement as detected by an FDA-approved test.26,27 prior treatment with one or more systemic therapies. For
Both are currently being tested in phase III frontline the 33 patients included in the study, ORR was 41% (95%
studies against gemcitabine plus cisplatin for treatment of CI, 24–59) and median PFS was 7.2 months (95% CI, 4.6–
patients with FGFR2 fusion–positive cholangiocarcinoma 10.1).19 Another recent study of ramucirumab, a direct
(NCT03656536, NCT03773302). VEGFR2 antagonist, or merestinib, a tyrosine kinase in-
TAS-120, an irreversible FGFR inhibitor, is undergoing inhibitor targeting MEK, among others, in combination with
vestigation in a phase I study of patients with advanced solid cisplatin/gemcitabine for first-line therapy, did not meet
tumors (NCT02052778). A preliminary analysis of the 45 its primary endpoint for PFS or secondary endpoints for
patients with cholangiocarcinoma included in the study OS and ORR. Although the addition of ramucirumab or
revealed tumor shrinkage for 20 of the 28 patients with merestinib to the current standard of care did not signif-
FGFR2 fusions (71%), and seven patients had confirmed icantly increase survival or response, combination therapy
PRs (25%). Additionally, of the 13 patients who had pre- was well tolerated, and translational studies are ongoing
viously undergone treatment with FGFR inhibition, four (NCT02711553).31 Additionally, a study of the PARP in-
(31%) had confirmed PRs.28 For patients treated with hibitor olaparib for patients with advanced BTC with

Allen et al
mutations in DNA repair genes will soon begin recruitment proteins, which have a role in neural development. Fusion of
(NCT04042831). a variety of differing partners with one of the TRK-encoding
genes causes oncogenic proteins that activate the RAS–
PANCREATIC CANCER
MEK–ERK and PI3K–AST signaling pathways (e.g., NTRK1,
Progress in genomic analysis in PDAC has been seen over NTRK2, or NTRK3). The prevalence of these mutations in
the past several years. Whole genome sequencing has PDAC is less than 1%. In 2018 the FDA approved laro-
subdivided PDAC tumors into four major classes.32 More trectinib for a tissue-agnostic approval for patients harboring
recently, genomic and transcriptional signatures have been NTRK fusions. Entrectinib was also recently (2019) given
developed to identify clinically relevant subpopulations.33,34 FDA approval for this indication. Specifically, in PDAC, for
However, the clinical applicability of these subtypes is still example, entrectinib was given to three patients with PDAC
limited in daily use or clinical trials. The limitations include harboring NTRK fusion–positive tumors, with all three
paucity of cancer cells for analysis and the high cost of these patients demonstrating PR based on RECIST evaluation
more sophisticated and detailed analyses. The most com- (STRTRK-2 TRIAL).
mercially available and applicable genomic analyses of
PDAC include capture-based targeted genomic profiling of Germline mutations Germline mutations in PDAC are identi-
cancer-associated genes, intron regions of specific genes fied in approximately 6% to 7% of unselected patients.
that are rearranged in cancer, tumor mutation burden, and In an examination of geographic and ethnic heterogeneity
microsatellite instability (MSI). These genomic profiles and in the BRCA1/2 prescreening population for the randomized
their applicability in targeted treatment of PDAC will be phase III POLO study of olaparib maintenance in metastatic
discussed further. pancreatic cancer, the prevalence in certain ethnic groups
The most common genomic alterations in PDAC are KRAS, was about 15%.38 Germline mutations in BRCA1 and
TP53, CDKN2A, and SMAD4. However, PDAC also contains BRCA2 have prognostic impact in PDAC. Retrospective and
marked genomic heterogeneity with additional mutations randomized phase II studies have shown that these patients
present at low-level frequency (less than 5%). Genomic demonstrate exquisite sensitivity to platinum-based therapy
sequencing studies of pancreatic cancer have revealed and have extended PFS and OS when exposed to platinum-
a small set of consistent mutations found in most pancreatic based chemotherapy.39 However, the toxicity profile, spe-
cancers and, beyond that, a low prevalence of targetable cifically the cumulative neuropathy, limits long-term use and
mutations. The majority of alterations in PDAC are somatic; exposure to platinum-based chemotherapy. The POLO
however, germline mutations such as BRCA1/2 are also maintenance randomized phase III trial in PDAC was
implicated in approximately 6% to 7% of patients with PDAC designed to address whether PARP inhibitors may be given
and have clinical impact. Actionable genomic alterations are in a maintenance setting. Maintenance treatments aim to
defined as drugs that have received FDA approval for PDAC. delay disease progression after chemotherapy, without
Potentially actionable genomic alterations are based on compromising quality of life. Identified patients were ran-
hypothesis-derived ongoing clinical studies. domly assigned in a 2:1 ratio to receive olaparib 300 mg
twice daily or placebo. Primary endpoint PFS was
FDA-Approved Targeted Agents in PDAC 7.4 months on olaparib versus 3.8 months in the placebo
MSI high and mismatch repair deficiency The first genomic arm (HR 0.53; 95% CI, 0.35–0.82; p = .0038). Interim
alteration to receive FDA approval for an agnostic tumor OS data (at 46% maturity) showed no difference between
label was MSI high and MMR deficiency. The incidence of arms. Final OS results will be evaluated at 69% data ma-
MMR deficiency in PDAC is less than 1%.35 The anti–PD-1 turity. No statistical differences were noted in quality of life
antibody pembrolizumab was approved by the FDA for measurements between the olaparib and placebo arms.
a tissue-agnostic indication for use in MMR-deficient tu- Patients in the olaparib arm were more likely to achieve
mors. In the pivotal phase II study,36 eight patients with a response to treatment or maintain disease control; re-
MMR-deficient tumors and PDAC were treated with pem- sponses were durable, lasting a median of more than
brolizumab and demonstrated two complete responses, 2 years.40
three PRs, and one stable disease. In a more recent
The combinatorial approach of PARP inhibitor veliparib in
publication of KEYNOTE-158, 22 patients with pancreatic
combination with gemcitabine and cisplatin in a random-
cancer and MSI or MMR deficiencies were enrolled. One
ized phase II trial showed no additional benefit in com-
complete response and three PRs were demonstrated.
parison with the chemotherapy-alone arm.41 To date, with
Median PFS was 2.1 months, and median OS was 4
the published available data, it seems that the staggered
months.37
approach with induction chemotherapy followed by main-
Tropomyosin receptor kinase family The tropomyosin re- tenance PARP inhibitor in the germline BRCA setting is the
ceptor kinase (TRK) family has three transmembrane recommended approach.

Ongoing Studies and Potential New Targets PDAC harboring an in-frame BRAF deletion describes a PR
DNA damage repair pathways The POLO study demon- to treatment with the MEK1/2 inhibitor trametinib, which is
strated efficacy with PARP inhibitors in patients with FDA approved for use in BRAF V600E–mutant melanoma.
germline BRCA PDAC responding or stable to platinum- The durable response lasted for 6 months.45 Patients with
based chemotherapy. Furthermore, a favorable response to PDAC with tumors harboring a BRAF alteration might be
platinum agents in the advanced setting has been described candidates for basket trials such as the MATCH screening
for patients with various DDR pathway alterations beyond trial (NCT02465060), where they can get BRAF and MEK
BRCA. The DDR alterations included mutations in ATM, inhibitors.
ATR, ATRX, or BAP1, BARD1, BRIP1, CHEK1/2, RAD50/ HER2 HER2 amplification is described in up to 3.3% of
51/51B, or FANCA/C/D2/E/F/G/L.42 It has been hypothe- KRAS WT PDACs. A phase II trial evaluating capecitabine
sized that patients harboring DDR alterations and demon- and trastuzumab in HER2-overexpressed (immunohisto-
strating response to platinum-based agents may benefit chemistry +3) metastatic PDAC did not demonstrate
from a more targeted approach with biologic agents tar- a favorable PFS or OS. Ongoing clinical trials are eval-
geting the DDR pathway, such as PARP, ATR, ATM, and uating additional anti-HER2 agents in various malig-
DNA-PK inhibitors. However, it is important to note that not nancies including metastatic PDACs (NCT03602079,
all patients responding to platinum-based chemotherapy NCT02465060).
with a broader mechanism of damage to the tumor will ALK ALK rearrangements are a clinically effective target in
respond to a more selected and targeted approach. Addi- numerous malignancies. These rearrangements are rare in
tionally, the differing DDR alterations will affect the DDR PDACs, with a described incidence of 0.16% in all PDACs
pathways in alternative ways and affect tumor progression and up to 1.3% in KRAS WT PDACs. They are more
and resistance mechanisms differently. These critical in- common in young patients (age , 50).46,47 Clinical benefit
sights should be taken into consideration when conducting (stable disease, radiologic response, or CA19-9 decrease)
further clinical trials in DDR alterations beyond BRCA. was described in three out of four patients with PDAC with
Ongoing phase II studies evaluating the efficacy of olaparib ALK rearrangements treated with ALK inhibitors in a case
for patients with PDAC with DDR deficiencies other than series described by Singhi et al. A phase I trial to further
BRCA are ongoing. Preliminary results from a combined evaluate the use of the ALK inhibitor ceritinib is ongoing
international study (MD Anderson and Sheba Medical (NCT02227940).
Center) have shown a response rate of 20% (11 patients) in RET RET alterations has been described in up to 1.35% of
the U.S. cohort, and no responses have been reported in the KRAS WT PDACs.46 LOXO-292 is FDA approved for RET
trial conducted in Israel (21 patients).43 It is important to fusion–positive non–small cell lung cancer and RET
note that in the Israeli cohort, patients with platinum re- mutation–positive medullary thyroid carcinoma based on
sistance were allowed into the clinical trials initially. How- results from the LIBRETTO-001 trial. Two patients with
ever, in the MD Anderson cohort, only platinum-sensitive PDAC with RET alterations received LOXO-292 treatment in
tumors were allowed. Emerging strategies to increase the the clinical trial and demonstrated stable disease.48 An
potency of PARP inhibitors is ongoing, evaluating combi- open-label expanded access program for patients with
natorial drugs such as PARP inhibitors in combination with RET48 alterations is ongoing (NCT03906331).
immunotherapy, MEK inhibitors, antiangiogenic agents,
and DDR drug development. NRG1 NRG1 fusions, another rare PDAC genomic alter-
ation, have a higher prevalence only in a specific subgroup
KRAS KRAS alterations are the most common type in PDAC of young patients with KRAS WT PDAC (17%).49 NRG1 is
and can be found in up to 90% of patients.44 Unfortunately, a cell adhesion molecule that is encoded by the NRG1 gene
there are no effective established agents to target KRAS in and is one of four proteins in the neuregulin family that act
PDAC. Additional targetable alterations in patients with on the EGFR family of receptors. Early efficacy signals have
KRAS-mutant PDAC are rare. Alternatively, when exploring been noted; two patients with NRG1 fusions received EGFR-
the subgroup of KRAS WT PDACs, the chances of finding targeted therapies (afatinib or erlotinib plus pertuzumab)
potentially targetable alterations are much higher and are and had a PR. 49 Ongoing studies are evaluating the
described in up to 38% of samples. The most common HER2/HER3 bispecific antibody MCLA-128 activity in pa-
potentially actionable genomic alterations in which nu- tients with advanced NRG1 fusion–positive solid tumors
merous clinical trials are ongoing include BRAF, 10.79%; (NCT04100694).
ERBB2, 3.37%; ALK, 1.35%; and RET, 1.35%.
CDK CDKN2A aberrations are described in up to 44% of
BRAF BRAF alterations on the MAPK pathway are found in PDACs.46 CDKN2A encodes P16INK4A, which inhibits CDK4/6.
1% to 2% of patients with PDAC and in up to 10% of those Thus, phase I ongoing clinical trials to evaluate the use
with KRAS WT PDAC.45,46 A case report of a patient with of CDK4/6 inhibitors in combination with other agents

Allen et al
are ongoing (NCT03065062, NCT03454035). The clinical months).61 Ipilimumab, the anti–CTLA-4 antibody, was
utility in PDAC has not yet been defined. evaluated in patients with advanced pancreatic cancer with
no objective responses and short survival.62 Similarly, there
IMMUNE MICROENVIRONMENT
was no meaningful single-agent activity for single-agent
The tumor microenvironment of pancreatic and biliary anti–PD-1 or anti–PD-L1 antibodies.63 There are several
adenocarcinoma is complex and heterogeneous, with var- reasons for the disappointing results seen with single-agent
ious cellular components and a dense desmoplastic stroma. ICIs, including the low tumor mutation burden of pancreatic
Pancreatic and biliary adenocarcinomas belong to the adenocarcinoma compared with immunogenic tumors such
category of nonimmunogenic tumors and deploy multiple as melanoma and non–small cell lung cancer.64 Along the
mechanisms of immune evasion. Despite the presence of same lines, the incidence of MMR deficiency, which has
high levels of CD4+ and CD8+ T cells in a subset of pan- been associated with single-agent anti–PD-1 activity in
creatic adenocarcinomas and its positive impact on prog- solid tumors, is less than 1% in pancreatic cancer.36,65
nosis, most pancreatic cancers are infiltrated by an array Lastly, the unfavorable balance of a microenvironment
of immunosuppressive cells including regulatory T cells rich in MDSCs, Tregs, and TAMs, with minimal or no
(Tregs), TAMs with M2 polarization, and MDSCs.50-52 The infiltration of cytotoxic T cells, provides both an expla-
secretion of suppressive chemokines and cytokines and nation for the poor activity of ICIs and a strong rationale for
the expression of cell surface proteins, such as PD-L1 combinatorial approaches. The emerging immunother-
and CTLA-4, contribute to dampening of anticancer apy targets in pancreatic adenocarcinoma are thought to
immunity.53-55 Cancer-associated fibroblasts (CAFs) are an contribute to the trafficking of cytotoxic T cells into the
important stromal cell population with contradictory roles tumor and the reprogramming of the suppressive immune
reported in the literature, including tumor promoting and microenvironment.
tumor inhibitory effects; these effects appear to be caused
by the presence of different subtypes of CAFs with inter- Targeting Tumor-Associated Macrophages
tumoral and intratumoral heterogeneity, which affects CSF1R TAMs contribute to the immunosuppressive envi-
the immune microenvironment as well.56 Perhaps the ronment of pancreatic cancer; specifically, M2-like tumor-
complexity and heterogeneity of the pancreatic cancer micro- promoting macrophages represent the predominant
environment are best captured by the reported classifica- population in later stages of pancreatic cancer devel-
tions that divide pancreatic ductal carcinoma into five opment and are associated with poor survival.66-68 Preclinical
subtypes: pure basal-like, stroma-activated, desmoplastic, evaluation of the CSF1R tyrosine kinase inhibitor PLX3397 in
pure classic, and immune classic. The stroma-activated and a syngeneic mouse model resulted in downregulation of genes
the pure basal-like tumor subtypes both have low immune involved in inflammatory responses, chemotaxis, myeloid
infiltrates and worse prognosis than the other subtypes.57 leukocyte–mediated immunity, and proteolysis, whereas
Despite the more limited literature related to biliary cancers, genes involved in antigen presentation, allograft rejection,
existing data highlight several similarities with pancreatic interferon responses, and TH1 immunity were upregulated.
cancer in regard to the multitude of immunosuppressive Furthermore, CSF1R blockade depleted TAMs, reprog-
mechanisms and the role of CAFs.58 Here we review how the rammed remaining macrophages to support antitumor
various mechanisms of immune evasion may be leveraged immunity, and increased the activity of PD-1 blockade.69
as potential therapeutic targets and provide several exam- Cabiralizumab, a humanized immunoglobulin G4 mono-
ples of relevant preclinical and clinical research with a focus clonal antibody that blocks CSF1R, showed durable clinical
on pancreatic carcinomas. benefit in 16% of patients and a confirmed objective re-
Immune Checkpoints sponse rate of 10% in a small cohort of 31 patients in
combination with nivolumab.70 Despite the promising early
PD-1 and CTLA-4 Immune checkpoint inhibitors (ICIs) have signal, a randomized phase II trial of cabiralizumab and
not manifested reliable and sufficient antitumor activity in nivolumab with or without chemotherapy failed to show an
pancreatic and biliary cancer. PD-1 and PD-L1 over- increase in PFS over standard-of-care chemotherapy
expression in biliary tumor cells has been associated with according to a press release in February 2020.71
increased invasiveness and poor prognosis.59,60 In a BTC
cohort of the KEYNOTE-158 trial—an ongoing phase II, CCL2/CCR2 A variety of other modalities to target TAMs and
single arm, open-label trial of pembrolizumab in advanced promote the infiltration of effector T cells are at various
cancers—the ORR was 5.8% in 104 patients with BTC, and stages of preclinical and clinical development, including
the duration of response ranged from 6.2 to more than 15 approaches focused on certain chemokines released by
months. The ORR was slightly higher in the cohort of pa- TAMs. The CCL2/CCR2 axis plays a critical role in the
tients with BTC with PD-L1–positive tumors. However, there mobilization of inflammatory monocytes into the tumor,
was no significant difference in the OS (7.2 vs. 9.6 where they acquire an M2 phenotype.72 The ratio of

peripheral blood to bone marrow inflammatory monocytes Cytotoxic Chemotherapy for Immune Modulation
has been associated with postresection survival.73 An in- Cytotoxic chemotherapy is thought to have immunomodu-
triguing response rate of 49% has been reported from latory effects including increasing neoantigen load as a re-
a phase I study of FOLFIRINOX plus PF-04136309, a small sult of tumor cell death85 and upregulation of major
molecule CCR2 inhibitor.74 histocompatibility complex class I expression, which en-
CD40 CD40 agonists promote macrophage and DC matu- hances antigen presentation.86 Several chemotherapeutic
ration, promote macrophage tumoricidal activity, and assist agents including paclitaxel have been associated with de-
antigen-presenting cells to more effectively present antigen pletion of immunosuppressive cell populations such as
to T cells.75,76 In a phase I study, the addition of the fully Tregs.87,88 In a phase I study of gemcitabine, nab-paclitaxel,
human agonist CD40 monocolonal antibody CP-870,893 to and nivolumab, the ORR was 18%, with a median PFS and
gemcitabine resulted in promising clinical activity man- OS comparable to those of gemcitabine and nab-paclitaxel.89
ifested by a fluorodeoxyglucose–PET metabolic response Additional studies to elucidate the effects of cytotoxic che-
rate of 88% and a RECIST response rate of 19%. The motherapy on the pancreatic cancer immune environment
median PFS for the 21 patients was 5.6 months (95% CI, and on the efficacy of immunotherapy are needed.
4.0 months to not estimable), and the median OS was Improving Dendritic Cell Function and Antigen
7.4 months (95% CI, 5.5–12.8 months). This result com- Presentation With Vaccine-Based Combination Strategies
pared favorably with the reported single-agent gemcitabine Vaccine therapy potentiates antigen presentation by DCs
activity, with a median PFS of 2.3 months and median OS of and can activate pathogen-specific effector and memory
5.4 months.77,78 Using a genetically engineered mouse T cells.90 For instance, GVAX, an allogeneic granulocyte–
model, the authors showed that the activity of the CD40 macrophage colony-stimulating factor vaccine therapy, in-
agonist depended on macrophages that upregulated major duced the formation of intratumoral tertiary lymphoid ag-
histocompatibility complex class II and the costimulatory gregates, which are associated with improved survival.91
molecule CD86 in response to treatment; this result was More recently, a randomized phase II study evaluated the
concurrent with a cytokine surge with elevated serum levels combination of cyclophosphamide and GVAX followed by
of interleukin-12, tumor necrosis factor-α, and interferon-γ CRS-207, an attenuated Listeria monocytogenes that ex-
but not interleukin-10. A more recent clinical trial evaluated presses the tumor-associated antigen mesothelin, with or
the combination of gemcitabine, nab-paclitaxel, and the without nivolumab.92 Although there was no difference in OS
CD40 agonist APX005M with or without nivolumab. Early between the two arms, the median OS of 6 months was
results from the dose-escalation phase of this trial, pre- comparable to that seen with chemotherapy in second-line
sented in abstract format, showed an intriguing response pancreatic cancer. Furthermore, treatment-emergent changes
rate of 53% by RECIST criteria.79 in the tumor microenvironment, including expansion of
Targeting Myeloid-Derived Suppressor Cells functional CD8+ T cells and a decrease in TAMs and myeloid
cells, were associated with higher OS. A currently enrolling
MDSCs, both granulocytic and monocytic subtypes, deplete study is evaluating dual checkpoint inhibition with a vacci-
L-arginine (thereby blocking T-cell proliferation), induce PD-
nation approach (NCT03190265). Other combinations in-
L1 expression, and recruit Tregs.80,81 Epigenetic modulating spired by the correlative biomarker findings are warranted.
drugs have been shown to suppress inhibitory myeloid cells
and synergize with ICIs in various preclinical models of Interplay of Genomics and Immunotherapy
different solid tumors.82,83 In the Panc02 metastatic pan- As discussed earlier, patients with pancreatic and biliary
creatic cancer model, addition of the histone deacetylase cancers may harbor DDR alterations including MMR de-
inhibitor entinostat to anti–PD-1 and anti–CTLA-4 anti- ficiencies and BRCA mutations.20,93 DNA damage has been
bodies increased tumor-free survival and shifted the MDSC shown to induce immune priming and proinflammatory
population to be dominated by the less immunosuppressive pathways in multiple tumor types via activation of stimulator
granulocyte-like MDSCs. Furthermore, the combination of of interferon genes and activation of the proinflammatory
entinostat with ICIs resulted in altered expression of im- transcription factor nuclear factor-κB.94 Preclinical and
portant signaling pathways that converged to control and clinical studies have shown promising antitumor activity
regulate myeloid immunosuppressive activities and infor the combination of PARP inhibitors and anti–PD-1 an-
creased infiltration of effector CD8+ T cells independent of tibodies in various tumor types.95-98 Ongoing trials are
Treg infiltration.84 Ongoing early-phase clinical trials are assessing similar combinations for patients with advanced
evaluating the combination of histone deacetylase inhibitors pancreatic cancer (NCT03404960). Patients with pan-
or demethylating agents in combination with anti–PD-1 creatic cancer or cholangiocarcinoma harboring MMR
therapies in pancreatic and biliary cancers (NCT03250273, deficiencies benefit from single-agent ICI targeting PD-1
NCT03257761). with ORRs of 18% and 41% that are durable.99

Allen et al
FUTURE DIRECTIONS AND CONCLUSIONS identification of several promising targets in numerous

Pancreaticobiliary cancers are a group of malignancies cellular pathways. The development of agents targeting
characterized by limited treatment options and poor these elements will enhance existing treatment paradigms
prognosis. Genomic analysis and a deeper understanding and improve the care of patients with pancreaticobiliary
of the immune microenvironment have contributed to the cancers.

1
University of Arizona, College of Medicine, Tucson, AZ Rachna T. Shroff, MD, MS, Division of Hematology/Oncology, University
2
Institute of Oncology, Sheba Medical Center, Tel Hashomer, Israel of Arizona Cancer Center, 1515 N. Campbell Ave., Tucson, AZ 85724;
3
University of Southern California, Keck School of Medicine, USC Norris Twitter: @rachnatshroff; email: [email protected].
Comprehensive Cancer Center, Los Angeles, CA
4
Division of Hematology/Oncology, University of Arizona Cancer Center,
Tucson, AZ
EDBK_280901.
REFERENCES
1. Martens S, Lefesvre P, Nicolle R, et al. Different shades of pancreatic ductal adenocarcinoma, different paths towards precision therapeutic applications. Ann
Oncol. 2019;30:1428-1436.
2. Rawla P, Sunkara T, Gaduputi V. Epidemiology of pancreatic cancer: global trends, etiology and risk factors. World J Oncol. 2019;10:10-27.
3. Valle JW, Lamarca A, Goyal L, et al. New horizons for precision medicine in biliary tract cancers. Cancer Discov. 2017;7:943-962.
4. Tempero MA. NCCN guidelines updates: pancreatic cancer. J Natl Compr Canc Netw. 2019;17:603-605.
5. Valle J, Wasan H, Palmer DH, et al; ABC-02 Trial Investigators. Cisplatin plus gemcitabine versus gemcitabine for biliary tract cancer. N Engl J Med. 2010;
362:1273-1281.
6. Martinez FJ, Shroff RT. Biliary tract cancers: systemic therapy for advanced disease. Chin Clin Oncoli. 2020;9:5.
7. Ong CK, Subimerb C, Pairojkul C, et al. Exome sequencing of liver fluke-associated cholangiocarcinoma. Nat Genet. 2012;44:690-693.
8. Jiao Y, Pawlik TM, Anders RA, et al. Exome sequencing identifies frequent inactivating mutations in BAP1, ARID1A and PBRM1 in intrahepatic chol-
angiocarcinomas. Nat Genet. 2013;45:1470-1473.
9. Nakamura H, Arai Y, Totoki Y, et al. Genomic spectra of biliary tract cancer. Nat Genet. 2015;47:1003-1010.
10. Bridgewater JA, Goodman KA, Kalyan A, et al. Biliary tract cancer: epidemiology, radiotherapy, and molecular profiling. Am Soc Clin Oncol Educ Book. 2016;
35:e194-e203.
11. Wang J, Zhang ZG, Ding ZY, et al. IDH1 mutation correlates with a beneficial prognosis and suppresses tumor growth in IHCC. J Surg Res. 2018;231:116-125.
12. Borad MJ, Champion MD, Egan JB, et al. Integrated genomic characterization reveals novel, therapeutically relevant drug targets in FGFR and EGFR pathways in
sporadic intrahepatic cholangiocarcinoma. PLoS Genet. 2014;10:e1004135.
13. Egan JB, Marks DL, Hogenson TL, et al. Molecular modeling and functional analysis of exome sequencing-derived variants of unknown significance identify
a novel, constitutively active FGFR2 mutant in cholangiocarcinoma. JCO Precis Oncol. 2017;2017:1-13.
14. Wardell CP, Fujita M, Yamada T, et al. Genomic characterization of biliary tract cancers identifies driver genes and predisposing mutations. J Hepatol. 2018;
68:959-969.
15. Jusakul A, Cutcutache I, Yong CH, et al. Whole-genome and epigenomic landscapes of etiologically distinct subtypes of cholangiocarcinoma. Cancer Discov.
2017;7:1116-1135.
16. Mahipal A, Tella SH, Kommalapati A, et al. FGFR2 genomic aberrations: Achilles heel in the management of advanced cholangiocarcinoma. Cancer Treat Rev.
2019;78:1-7.
17. Galdy S, Lamarca A, McNamara MG, et al. HER2/HER3 pathway in biliary tract malignancies; systematic review and meta-analysis: a potential therapeutic target?
Cancer Metastasis Rev. 2017;36:141-157.
18. Marks EI, Yee NS. Molecular genetics and targeted therapeutics in biliary tract carcinoma. World J Gastroenterol. 2016;22:1335-1347.
19. Wainberg Z, Lassen U, Elez E, et al. Efficacy and safety of dabrafenib (D) and trametinib (T) in patients (pts) with BRAF V600E–mutated biliary tract cancer (BTC):
a cohort of the ROAR basket trial. J Clin Oncol. 2019;37:4s(suppl; abstr 187).
20. Lamarca A, Barriuso J, McNamara MG, et al. Biliary tract cancer: state of the art and potential role of DNA damage repair. Cancer Treat Rev. 2018;70:168-177.

21. Lowery MA, Burris HA III, Janku F, et al. Safety and activity of ivosidenib in patients with IDH1-mutant advanced cholangiocarcinoma: a phase 1 study. Lancet
22. Javle M, Lowery M, Shroff RT, et al. Phase II study of BGJ398 in patients with FGFR-altered advanced cholangiocarcinoma. J Clin Oncol. 2018;36:276-282.
23. Abou-Alfa G, Valle J, Kelley R, et al. ClarIDHy: a phase 3 multicenter randomized double-blind study of AG-120 versus placebo in patients with non-resectable or
metastatic cholangiocarcinoma with an IDH1 mutation. J Clin Oncol. 2018;36:4s(suppl; abstr TPS545).
24. Abou-Alfa GK, Macarulla Mercade T, Javle M, et al. LBA10_PRClarIDHy: a global, phase III, randomized, double-blind study of ivosidenib (IVO) vs placebo in
patients with advanced cholangiocarcinoma (CC) with an isocitrate dehydrogenase 1 (IDH1) mutation. Ann Oncol. 2019;30(suppl 5):v851-v934.
25. Vogel A, Sahai V, Hollebecque A, et al. LBA40FIGHT-202: a phase II study of pemigatinib in patients (pts) with previously treated locally advanced or metastatic
cholangiocarcinoma (CCA). Ann Oncol. 2019;30(suppl 5):v851-v934.
26. BridgeBio Pharma Inc. BridgeBio Pharma’s QED Therapeutics receives fast track designation for infigratinib in adults with first-line advanced or metastatic
cholangiocarcinoma and orphan drug designation for infigratinib for treatment of cholangiocarcinoma. https://bridgebio.com/news/bridgebio-pharmas-
qedtherapeutics-receives-fast-track-designation-for-infigratinib-in-adults-with-first-lineadvanced-or-metastatic-cholangiocarcinoma-and-orphan-
drug-designation-for-infigra. Accessed June 1, 2020.
27. U.S. Food and Drug Administration. FDA grants accelerated approval to pemigatinib for cholangiocarcinoma with an FGFR2 rearrangement or fusion. https://
www.fda.gov/drugs/resources-information-approveddrugs/fda-grants-accelerated-approval-pemigatinib-cholangiocarcinoma-fgfr2-rearrangement-or-fusion.
Accessed June 1, 2020.
28. Tran B, Meric-Bernstam F, Arkenau H-T, et al. Efficacy of TAS-120, an irreversible fibroblast growth factor receptor inhibitor (FGFRi), in patients with
cholangiocarcinoma and FGFR pathway alterations previously treated with chemotherapy and other FGFRi’s. Ann Oncol. 2018;29(suppl 9):ix46-ix66.
29. Mou H-B, Li W-D, Shen Y-J, et al. Trastuzumab, not lapatinib, has therapeutic effects on Chinese patients with HER2-positive cholangiocarcinoma. Hepatobiliary
Pancreat Dis Int. 2018;17:477-479.
30. Javle M, Hainsworth J, Swanton C, et al. Pertuzumab + trastuzumab for HER2-positive metastatic biliary cancer: preliminary data from MyPathway. J Clin Oncol.
2017;35:4s(suppl; abstr 402).
31. Valle J, Bai L-Y, Orlova R, et al. Ramucirumab (RAM) or merestinib (MER) or placebo (PL) plus gemcitabine (GEM) and cisplatin (CIS) as first-line treatment for
advanced or metastatic biliary tract cancer (BTC): a randomized, double-blind, phase II study. J Clin Oncol. 2020;38:4s(suppl; abstr 477).
32. Waddell N, Pajic M, Patch AM, et al; Australian Pancreatic Cancer Genome Initiative. Whole genomes redefine the mutational landscape of pancreatic cancer.
Nature. 2015;518:495-501.
33. Bailey P, Chang DK, Nones K, et al; Australian Pancreatic Cancer Genome Initiative. Genomic analyses identify molecular subtypes of pancreatic cancer. Nature.
2016;531:47-52.
34. Collisson EA, Bailey P, Chang DK, et al. Molecular subtypes of pancreatic cancer. Nat Rev Gastroenterol Hepatol. 2019;16:207-220.
35. Hu ZI, Shia J, Stadler ZK, et al. Evaluating mismatch repair deficiency in pancreatic adenocarcinoma: challenges and recommendations. Clin Cancer Res. 2018;
24:1326-1336.
37. Marabelle A, Le DT, Ascierto PA, et al. Efficacy of pembrolizumab in patients with noncolorectal high microsatellite instability/mismatch repair–deficient cancer:
results from the phase II KEYNOTE-158 Study. J Clin Oncol. 2019;38:4-10.
38. Golan T, Kindler HL, Park JO, et al. Geographic and ethnic heterogeneity of germline BRCA1 or BRCA2 mutation prevalence among patients with metastatic
pancreatic cancer screened for entry into the POLO trial. J Clin Oncol. 2020;38:1442-1454.
39. Golan T, Kanji ZS, Epelbaum R, et al. Overall survival and clinical characteristics of pancreatic cancer in BRCA mutation carriers. Br J Cancer. 2014;
111:1132-1138.
41. O’Reilly EM, Lee JW, Zalupski M, et al. Randomized, multicenter, phase II trial of gemcitabine and cisplatin with or without veliparib in patients with pancreas
adenocarcinoma and a germline BRCA/PALB2 mutation. J Clin Oncol. 2020;38:1378-1388.
42. Pishvaian MJ, Blais EM, Brody JR, et al. Outcomes in pancreatic adenocarcinoma (PDA) patients (pts) with genetic alterations in DNA damage repair (DDR)
pathways: results from the Know Your Tumor (KYT) program. J Clin Oncol. 2019;37:191.
43. Golan T, Varadhachary G, Sela T, et al. Phase II study of olaparib for BRCAness phenotype in pancreatic cancer. J Clin Oncol. 2018;36:4s(suppl; abstr 297).
44. Buscail L, Bournet B, Cordelier P. Role of oncogenic KRAS in the diagnosis, prognosis and treatment of pancreatic cancer. Nat Rev Gastroenterol Hepatol. 2020;
17:153-168.
45. Aguirre AJ, Nowak JA, Camarda ND, et al. Real-time genomic characterization of advanced pancreatic cancer to enable precision medicine. Cancer Discov.
2018;8:1096-1111.
46. Singhi AD, George B, Greenbowe JR, et al. Real-time targeted genome profile analysis of pancreatic ductal adenocarcinomas identifies genetic alterations that
might be targeted with existing drugs or used as biomarkers. Gastroenterology. 2019;156:2242-2253 e4.
47. Singhi AD, Ali SM, Lacy J, et al. Identification of targetable ALK rearrangements in pancreatic ductal adenocarcinoma. J Natl Compr Canc Netw. 2017;
15:555-562.
48. Alexander E, Drilon VS, Geoffrey R, et al. A phase 1 study of LOXO-292, a potent and highly selective RET inhibitor, in patients with RET-altered cancers. J Clin
Oncol. 2018;36:101-102.

Allen et al
49. Heining C, Horak P, Uhrig S, et al. NRG1 fusions in KRAS wild-type pancreatic cancer. Cancer Discov. 2018;8:1087-1095.
50. Wartenberg M, Zlobec I, Perren A, et al. Accumulation of FOXP3+T-cells in the tumor microenvironment is associated with an epithelial-mesenchymal-transition-
type tumor budding phenotype and is an independent prognostic factor in surgically resected pancreatic ductal adenocarcinoma. Oncotarget. 2015;
6:4190-4201.
51. Attri KS, Mehla K, Singh PK. Evaluation of macrophage polarization in pancreatic cancer microenvironment under hypoxia. Methods Mol Biol. 2018;
1742:265-276.
52. Carstens JL, Correa de Sampaio P, Yang D, et al. Spatial computation of intratumoral T cells correlates with survival of patients with pancreatic cancer. Nat
Commun. 2017;8:15095.
53. Martinez-Bosch N, Vinaixa J, Navarro P. Immune evasion in pancreatic cancer: from mechanisms to therapy. Cancers (Basel). 2018;10:E6.
54. Chen L, Flies D. Molecular mechanisms of T cell co-stimulation and co-inhibition. Nat Rev Immunol. 2013;13:227-242.
55. Knudsen ES, Vail P, Balaji U, et al. Stratification of pancreatic ductal adenocarcinoma: combinatorial genetic, stromal, and immunologic markers. Clin Cancer
Res. 2017;23:4429-4440.
56. Neuzillet C, Tijeras-Raballand A, Ragulan C, et al. Inter- and intra-tumoural heterogeneity in cancer-associated fibroblasts of human pancreatic ductal ad-
enocarcinoma. J Pathol. 2019;248:51-65.
57. Puleo F, Nicolle R, Blum Y, et al. Stratification of pancreatic ductal adenocarcinomas based on tumor and microenvironment features. Gastroenterology. 2018;
155:1999-2013.e3.
58. Fabris L, Perugorria M, Mertens J, et al. The tumour microenvironment and immune milieu of cholangiocarcinoma. Liver Int. 2019;39(suppl 1):63-78.
59. Zhu Y, Wang XY, Zhang Y, et al. Programmed death ligand 1 expression in human intrahepatic cholangiocarcinoma and its association with prognosis and CD8+
T-cell immune responses. Cancer Manag Res. 2018;10:4113-4123.
60. Fontugne J, Augustin J, Pujals A, et al. PD-L1 expression in perihilar and intrahepatic cholangiocarcinoma. Oncotarget. 2017;8:24644-24651.
61. Ueno M, Chung HC, Nagrial A, et al. Pembrolizumab for advanced biliary adenocarcinoma: results from the multicohort, phase II KEYNOTE-158 study. Ann
Oncol. 2018;29:viii210.
62. Royal RE, Levy C, Turner K, et al. Phase 2 trial of single agent ipilimumab (anti-CTLA-4) for locally advanced or metastatic pancreatic adenocarcinoma.
J Immunother. 2010;33:828-833.
63. Brahmer JR, Tykodi SS, Chow LQ, et al. Safety and activity of anti–PD-L1 antibody in patients with advanced cancer. N Engl J Med. 2012;366:2455-2465.
65. Humphris JL, Patch AM, Nones K, et al; Australian Pancreatic Cancer Genome Initiative. Hypermutation in pancreatic cancer. Gastroenterology. 2017;
152:68-74.e2.
66. Liou GY, Bastea L, Fleming A, et al. The presence of interleukin-13 at pancreatic ADM/PanIN lesions alters macrophage populations and mediates pancreatic
tumorigenesis. Cell Reports. 2017;19:1322-1333.
67. Kurahara H, Shinchi H, Mataki Y, et al. Significance of M2-polarized tumor-associated macrophage in pancreatic cancer. J Surg Res. 2011;167:e211-e219.
68. Chen SJ, Zhang QB, Zeng LJ, et al. Distribution and clinical significance of tumour-associated macrophages in pancreatic ductal adenocarcinoma: a retro-
spective analysis in China. Curr Oncol. 2015;22:e11-e19.
69. Zhu Y, Knolhoff BL, Meyer MA, et al. CSF1/CSF1R blockade reprograms tumor-infiltrating macrophages and improves response to T-cell checkpoint im-
munotherapy in pancreatic cancer models. Cancer Res. 2014;74:5057-5069.
70. Wainberg Z, Piha-Paul S, Luke J, et al. First-inhuman phase 1 dose escalation and expansion of a novel combination, anti–CSF-1 receptor (cabiralizumab) plus
anti–PD-1 (nivolumab), in patients with advanced solid tumors. J Immunother Cancer. 2017;5(suppl 3_O42):89.
71. Five Prime Therapeutics, Inc. Five Prime Therapeutics Provides Update on Phase 2 Trial of Cabiralizumab Combined with Opdivo® in Pancreatic Cancer. http://
investor.fiveprime.com/news-releases/news-release-details/five-primetherapeutics-provides-update-phase-2-trial. Accessed June 1, 2020.
72. Shi C, Pamer EG. Monocyte recruitment during infection and inflammation. Nat Rev Immunol. 2011;11:762-774.
73. Sanford DE, Belt BA, Panni RZ, et al. Inflammatory monocyte mobilization decreases patient survival in pancreatic cancer: a role for targeting the CCL2/CCR2
axis. Clin Cancer Res. 2013;19:3404-3415.
74. Nywening TM, Wang-Gillam A, Sanford DE, et al. Targeting tumour-associated macrophages with CCR2 inhibition in combination with FOLFIRINOX in patients
with borderline resectable and locally advanced pancreatic cancer: a single-centre, open-label, dose-finding, non-randomised, phase 1b trial. Lancet Oncol.
2016;17:651-662.
75. Luheshi NM, Coates-Ulrichsen J, Harper J, et al. Transformation of the tumour microenvironment by a CD40 agonist antibody correlates with improved responses
to PD-L1 blockade in a mouse orthotopic pancreatic tumour model. Oncotarget. 2016;7:18508-18520.
76. Vonderheide RH, Glennie MJ. Agonistic CD40 antibodies and cancer therapy. Clin Cancer Res. 2013;19:1035-1043.
77. Beatty GL, Chiorean EG, Fishman MP, et al. CD40 agonists alter tumor stroma and show efficacy against pancreatic carcinoma in mice and humans. Science.
2011;331:1612-1616.
78. Burris HA III, Moore MJ, Andersen J, et al. Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas
cancer: a randomized trial. J Clin Oncol. 1997;15:2403-2413.

79. O’Hara MH, O’Reilly EM, Rosemarie M, et al. Abstract CT004: a phase Ib study of CD40 agonistic monoclonal antibody APX005M together with gemcitabine
(Gem) and nab-paclitaxel (NP) with or without nivolumab (Nivo) in untreated metastatic ductal pancreatic adenocarcinoma (PDAC) patients. Clin Trials. 2019;
79:CT004.
80. Bronte V, Serafini P, Mazzoni A, et al. L-arginine metabolism in myeloid cells controls T-lymphocyte functions. Trends Immunol. 2003;24:302-306.
81. Kusmartsev S, Nagaraj S, Gabrilovich DI. Tumor-associated CD8+ T cell tolerance induced by bone marrow-derived immature myeloid cells. J Immunol. 2005;
175:4583-4592.
82. Orillion A, Hashimoto A, Damayanti N, et al. Entinostat neutralizes myeloid-derived suppressor cells and enhances the antitumor effect of PD-1 inhibition in
murine models of lung and renal cell carcinoma. Clin Cancer Res. 2017;23:5187-5201.
83. Tomita Y, Lee M-J, Lee S, et al. The interplay of epigenetic therapy and immunity in locally recurrent or metastatic estrogen receptor-positive breast cancer:
correlative analysis of ENCORE 301, a randomized, placebo-controlled phase II trial of exemestane with or without entinostat. OncoImmunology. 2016;
5:e1219008.
84. Christmas BJ, Rafie CI, Hopkins AC, et al. Entinostat converts immune-resistant breast and pancreatic cancers into checkpoint-responsive tumors by
reprogramming tumor-infiltrating MDSCs. Cancer Immunol Res. 2018;6:1561-1577.
85. Bezu L, Gomes-de-Silva LC, Dewitte H, et al. Combinatorial strategies for the induction of immunogenic cell death. Front Immunol. 2015;6:187.
86. Liu WM, Fowler DW, Smith P, et al. Pre-treatment with chemotherapy can enhance the antigenicity and immunogenicity of tumours by promoting adaptive
immune responses. Br J Cancer. 2010;102:115-123.
87. Alizadeh D, Larmonier N. Chemotherapeutic targeting of cancer-induced immunosuppressive cells. Cancer Res. 2014;74:2663-2668.
88. Zhang L, Dermawan K, Jin M, et al. Differential impairment of regulatory T cells rather than effector T cells by paclitaxel-based chemotherapy. Clin Immunol.
2008;129:219-229.
89. Wainberg Z, Hochster H, George B, et al. Phase I study of nivolumab (nivo) + nab-paclitaxel (nab-P) 6 gemcitabine (Gem) in solid tumors: Interim results from the
pancreatic cancer (PC) cohorts. J Clin Oncol. 2017;35:4s(suppl; abstr 412).
90. Soares KC, Zheng L, Edil B, et al. Vaccines for pancreatic cancer. Cancer J. 2012;18:642-652.
91. Lutz ER, Wu AA, Bigelow E, et al. Immunotherapy converts nonimmunogenic pancreatic tumors into immunogenic foci of immune regulation. Cancer Immunol
Res. 2014;2:616-631.
92. Tsujikawa T, Crocenzi T, Durham J, et al. Evaluation of cyclophosphamide/GVAX pancreas followed by Listeria-mesothelin (CRS-207) with or without nivolumab
in patients with pancreatic cancer. Clin Cancer Res. Epub 2020 Apr 9.
93. Pishvaian M, Blais E, Brody J, et al. Outcomes in pancreatic adenocarcinoma (PDA) patients (pts) with genetic alterations in DNA damage repair (DDR) pathways:
results from the Know Your Tumor (KYT) program. J Clin Oncol. 2019;37:4s(suppl; abstr 191).
94. Stewart RA, Pilié PG, Yap TA. Development of PARP and immune-checkpoint inhibitor combinations. Cancer Res. 2018;78:6717-6725.
95. Shen J, Zhao W, Ju Z, et al. PARPi triggers the STING-dependent immune response and enhances the therapeutic efficacy of immune checkpoint blockade
independent of BRCAness. Cancer Res. 2019;79:311-319.
96. Huang J, Wang L, Cong Z, et al. The PARP1 inhibitor BMN 673 exhibits immunoregulatory effects in a Brca1( / ) murine model of ovarian cancer. Biochem
Biophys Res Commun. 2015;463:551-556.
97. Konstantinopoulos PA, Waggoner SE, Vidal GA, et al. TOPACIO/Keynote-162 (NCT02657889): a phase 1/2 study of niraparib + pembrolizumab in patients (pts)
with advanced triple-negative breast cancer or recurrent ovarian cancer (ROC)—results from ROC cohort. J Clin Oncol. 2018;36:15s (suppl; abstr 106).
98. Friedlander M, Meniawy T, Markman B, et al. A phase 1b study of the anti-PD-1 monoclonal antibody BGB-A317 (A317) in combination with the PARP inhibitor
BGB-290 (290) in advanced solid tumors. J Clin Oncol. 2017;35:15s (suppl; abstr 3013).
99. Marabelle A, Le DT, Ascierto PA, et al. Efficacy of pembrolizumab in patients with noncolorectal high microsatellite instability/mismatch repair-deficient cancer:
results from the phase II KEYNOTE-158 study. J Clin Oncol. 2020;38:1-10.

GENITOURINARY CANCER—
KIDNEY AND BLADDER

Current Approaches to the Treatment of Advanced

or Metastatic Renal Cell Carcinoma
Matthew Tenold, MD1; Praful Ravi, MD2; Monika Kumar, MD3; Alex Bowman, MD3; Hans Hammers, MD, PhD3; Toni K. Choueiri, MD2; and
Primo N. Lara Jr., MD1
overview
The optimal management approach to advanced or metastatic renal cell cancer of the clear cell type continues
to rapidly evolve. Risk stratification of patients into favorable-, intermediate-, and poor-risk categories is now
routinely performed. In selected individuals with low-volume indolent disease, active surveillance may be an
appropriate option. Cytoreductive nephrectomy and/or surgical metastasectomy may be also be considered for
selected patients after evaluation by a multidisciplinary tumor board. Systemic frontline therapy options now
include immune checkpoint inhibitor–based combination (IBC) therapies such as pembrolizumab/axitinib,
nivolumab/ipilimumab, and avelumab/axitinib. With unusual exceptions, monotherapy with vascular growth
factor receptor tyrosine kinase inhibitors or mTOR inhibitors are no longer appropriate options in the frontline
setting. Despite the established efficacy of frontline IBC, most patients will ultimately require additional lines
of therapy, and oncologists must think carefully when switching to another therapy, particularly in situations of
drug intolerance or apparent disease progression. Systemic therapy options after IBC are generally tyrosine
kinase inhibitor–based, and ongoing clinical trials will help optimize the treatment algorithm further. Despite
many recent drug approvals for renal cell cancer (RCC), there remains a pressing need to identify new
therapeutic targets. Finally, other systemic therapy or supportive care approaches must be considered for
special patient populations such as those with poor performance status, end-organ dysfunction, brain me-
tastases, or who have undergone metastasectomy.
FIRST-LINE TREATMENT OF ADVANCED OR METASTATIC increase this to intermediate risk (IMDC-Int), and
RENAL CELL CARCINOMA a patient with three or more risk factors has poor-
Overview risk disease (IMDC-Poor): 5
RCC is the eighth most common cancer in the United • Karnofsky performance status less than 80%
States, with an estimated 73,820 new cases and • Time from diagnosis to treatment less than 1 year
14,770 deaths in 2019. Early-stage disease can often • Hemoglobin concentration less than the lower limit
be asymptomatic, and 16% of patients present with of normal
metastatic RCC (mRCC).1 The landscape of available • Serum calcium more than the upper limit of normal
therapies has rapidly evolved, with many new highly • Neutrophil count more than the upper limit of
active therapies entering the therapeutic armamen- normal
tarium in the past decade. The goal of treatment, even • Platelet count more than the upper limit of normal.
in the metastatic setting, is cure. If cure is not achiev- Active Surveillance
able, then long-term survival is a reasonable goal in the For many patients with mRCC, it is appropriate to start
Author affiliations
context of many active agents. Clear cell mRCC, often therapy right away, but in select patients who are
and support characterized by alterations in the Von Hippel-Lindau asymptomatic and have a relatively limited volume of
information (if gene, is the most common subtype, accounting for slowly growing disease, it may be appropriate to pursue
applicable) appear about 75% to 85% of cases.2-4 a strategy of active surveillance. A phase II prospective,
at the end of this
article. multicenter international trial in 2016 enrolled and
Risk Stratification monitored 52 patients with asymptomatic mRCC with
Accepted on March
2, 2020 and The first step in developing a treatment plan for people a median follow-up of 38 months.6 They were radio-
published at with advanced clear cell mRCC is to classify a patient’s graphically assessed at baseline, every 3 months for
ascopubs.org on
risk status by using the International mRCC Database year 1, every 4 months for year 2, and then every
April 2, 2020:
Consortium Prognostic Model (IMDC score). Absence 6 months thereafter. The median time to initiation of
10.1200/EDBK_ of any of the following risk factors confers the most treatment of symptomatic disease was 14.9 months,
279881 favorable risk status (IMDC-Fav), one to two risk factors and IMDC-Poor risk and a higher number of metastatic

Tenold et al
(VEGFR) tyrosine kinase inhibitors (TKIs). The CARMENA

trial showed that sunitinib alone provided noninferior overall
survival compared with CN in patients with IMDC-Int and
• In most patients with advanced or metastatic IMDC-Poor mRCC. This trial reported a survival hazard ratio
renal cell cancer who require systemic therapy,
(HR) of 0.89 (95% CI, 0.71–1.10), with noninferiority
immune checkpoint inhibitor–based combi-
margin at no more than 1.20. Median overall survival
nation therapy is currently considered the
frontline standard of care. Monotherapy with (OS) was 18.4 months in the sunitinib-alone group and
a VEGF-TK or mTOR inhibitor is no longer 13.9 months in the nephrectomy/sunitinib group.9 Neo-
a reasonable option in the frontline setting adjuvant sunitinib before CN was compared with immediate
except in unusual circumstances (e.g., im- CN in the SURTIME trial, which failed to show an improved
munotherapy ineligibility). progression-free rate at 28 weeks, with 42% progression
• In selected patients, active surveillance and/or free in the immediate CN arm and 43% in the sunitinib
cytoreductive surgery (including nephrectomy before CN arm (p = .61).10 Both of these trials were limited
or metastasectomy) remain reasonable by poor patient accrual. Both also occurred during a shifting
considerations. landscape of available frontline therapies. A potential syn-
• Despite highly active frontline immunotherapy, ergistic relationship with modern immunotherapy has not
most patients will require additional lines of been explored sufficiently. Additionally, evaluation outside
therapy, and oncologists must think carefully the realm of clear cell mRCC has also been limited. Based
when switching to another therapy, particularly on the available data, it remains reasonable to offer CN for
in situations of drug intolerance or apparent patients with IMDC-Fav and selected patients with IMDC-Int
progression. mRCC who have a low metastatic volume. The key guidance
• Systemic therapy options after immunotherapy is to avoid reflexive CN without a careful multidisciplinary
are generally TKI-based; ongoing clinical trials tumor board evaluation. For many patients with IMDC-Int
will help optimize the treatment algorithm and nearly all patients with IMDC-Poor risk disease, CN can
further. be deferred or eliminated in favor of adequate systemic
• Special metastatic renal cell carcinoma pop- therapy. For those patients who defer CN, the response to
ulations such as those with poor performance systemic therapy may be a critical consideration to proceed
status, brain metastases, and end-organ dys- (or not) to subsequent CN. Additionally, palliative CN for
function will require individualized treatment tumor-related symptoms should be considered for the ap-
approaches. propriate individual.
Surgical metastasectomy may be another potentially ben-
disease sites were associated with shorter time to treatment eficial frontline intervention for patients. Although data
in multivariate analysis. Although median overall survival showing a survival benefit are limited to retrospective
was an exploratory outcome, it was found to be 38.6 studies and a meta-review, patients with good performance
months. A randomized study powered to confirm this status and isolated or oligometastatic disease who can
finding would be useful to further validate this approach, obtain a complete resection appear to benefit significantly.11
which may prolong survival for those who do not need A disease-free interval after nephrectomy of more than
immediate treatment. 2 years, absence of lymph node involvement, and lung-only
involvement are favorable features, whereas patients with
Frontline Local Therapy: The Role of Cytoreductive a primary T-stage 3, high histologic grade, or sarcomatoid
Nephrectomy and Metastasectomy features are less likely to benefit.
In a subset of patients evaluated by a multidisciplinary
Frontline Systemic Therapy: Immunotherapy-Based
tumor board, cytoreductive nephrectomy (CN) may be in-
Combination Therapy
dicated. The hypothesized advantages of CN include al-
teration of humoral interaction of metastatic implants with The historical immunotherapy-based frontline systemic
the primary tumor, triggering of an immune-mediated re- options such as interferon alpha (IFN-α) and high-dose
sponse following surgery, or reduction of genetic hetero- interleukin-2 were poorly tolerated by most patients and
geneity to reduce resistance and metastasis mechanisms had limited efficacy.12 A new standard of care was forged in
via removal of a genetically diverse primary lesion. Prior 2007 with the comparison of the VEGFR TKI sunitinib versus
randomized trials in the interferon era established the role of IFN-α, showing an overall response rate (ORR) of 47%
CN as part of a combined modality approach.7,8 More re- versus 12% and an 11-month versus 5-month median
cently the role of CN in mRCC was studied in the context of progression-free survival (PFS), favoring sunitinib (HR,
treatment with vascular endothelial growth factor receptor 0.42; 95% CI, 0.32–0.54; p , .001).13 For several years, an

Treatment of Advanced or Metastatic Renal Cell Carcinoma
antiangiogenesis approach with VEGFR-TKI monotherapy .0001) for avelumab/axitinib versus sunitinib in the intention
dominated frontline systemic treatment. Although TKIs to treat arm.15 Data from the previous study lacked maturity
continue to have an important role, immunotherapy with to demonstrate a substantial OS difference. Impressively,
antiprogrammed-death (anti–PD-1)/programmed death– the combination of pembrolizumab/axitinib was tested in the
ligand (PD-L1) immune checkpoint antibodies (immune KEYNOTE-426 trial, in which this combination was com-
checkpoint blockade [ICB]) has more recently become the pared with sunitinib, showing an 82.3% OS versus 72.1%
backbone of frontline systemic treatment in clear cell OS at 18 months (p , .0001). This study also included all
mRCC. PD-1/PD-L1 targeted immune checkpoint inhibitors risk subgroups and showed an ORR of 59.3% versus 35.7%
have essentially supplanted IFN-α and high-dose IL2 as the and median PFS of 15.1 months versus 11.1 months (p ,
preferred immunotherapeutic agents. Table 1 summarizes .001) in favor of the combination versus sunitinib alone.16
selected randomized trials that established standard of care The combination of anti–PD-1/PD-L1 plus anti–CTLA-4 has
options in mRCC. not yet been compared head-to-head with ICB plus VEGFR-
TKI; a randomized clinical trial is under development.
ICBs in combination with either a second immune-
Nevertheless, for most patients across all IMDC risk cate-
modulating antibody or a VEGFR-TKI have emerged to be
gories, the combination of an ICB plus VEGFR-TKI such
the optimal choice for most treatment-naive patients. The
as axitinib/pembrolizumab is the preferred treatment op-
CheckMate-214 trial demonstrated superiority for the
tion. In patients with IMDC-Int or IMDC-Poor risk disease,
combination of nivolumab (anti–PD-1 antibody)/ipilimumab
nivolumab/ipilimumab is also a reasonable option. Indeed,
(anti–CTLA-4 antibody) over sunitinib in patients with IMDC-
some clinicians choose to preserve the introduction of
Int/Poor risk mRCC, with an ORR of 42% versus 27% (p ,
VEGF-TKI therapy for the second-line setting in IMDC-Int
.0001), PFS of 11.6 months versus 8.4 months (HR, 0.82;
and IMDC-Poor disease by offering nivolumab/ipilimumab
99% CI, 0.64–1.05; p = .03), and median OS not reached
upfront. However, the impact of sequencing of VEGFR-TKI
versus 26 months respectively in the nivolumab plus ipili-
in the anti–PD-1/PD-L1 era has not yet been adequately
mumab arm versus sunitinib alone.14 Interestingly, the
investigated.
favorable-risk subgroup in CheckMate-214 seemed to fair
better with sunitinib, with combination therapy providing an Frontline Systemic Therapy: The Role of VEGFR-TKI
ORR of 29% versus 52% (p = .0002) and PFS of 15.3 months Monotherapy, mTOR Inhibition, and ICB Monotherapy
versus 25.1 months (p , .0001) when compared with Given the successes seen with ICB-based combination
sunitinib monotherapy; survival appeared to be similar in therapy, it is increasingly difficult to justify the use of VEGFR-
this subgroup. Updated results of this trial (now with a 49- TKIs as monotherapy for most treatment-naive patients with
month median follow-up) were presented in abstract form at mRCC. Potential exceptions would include patients who
the 2020 Genitourinary Cancers Symposium; those results may be considered high risk or ineligible for ICB therapy,
continued to show a persistent OS benefit in favor of the such as those with active autoimmune disease, a history of
nivolumab/ipilimumab arm in both the intermediate-/poor- solid organ transplantation, or on chronic immunosup-
risk group and in the ITT population. The Javelin-101 pressive therapies, because there is theoretical risk of au-
evaluated patients with mRCC from all IMDC risk sub- toimmune flare in patients with active autoimmune disease
groups and demonstrated the superiority of avelumab receiving ICB treatment or risk of impaired efficacy of ICBs
(anti–PD-L1 AB)/axitinib (VEGFR-TKI) over sunitinib alone with concurrent immunosuppressive therapy. Patients in
for patients regardless of PD-L1 status, with an ORR of these latter two clinical scenarios were excluded from the
55.2% versus 25.7% and a median PFS of 13.8 months previously discussed clinical trials. Other rare exceptions
versus 8.4 months (HR, 0.69; 95% CI, 0.56–0.84; p = would be prior adverse reaction to ICBs or patient refusal.
TABLE 1. Pivotal Trials in Advanced Renal Cell Cancer in the Frontline Setting
Pivotal Trial (Year Reported) No. Response Rate (%) Median PFS (Mo) Median OS (Mo)
Sunitinib vs. IFN-α (2007) 750 47 vs. 12 11 vs. 5 26.4 vs. 21.8
Pazopanib vs. sunitinib (2013) 1,110 31 vs. 25 8.4 vs. 9.5 28.4 vs. 29.3
Cabozantinib vs. sunitinib (poor/intermediate risk, 2017) 157 46 vs. 18 8.2 vs. 5.6 30 vs. 21.8
Temsirolimus vs. IFN-α (poor risk, 2007) 626 8.6 vs. 4.8 5.5 vs. 3.1 10.9 vs. 7.3
Nivolumab/ipilimumab vs. sunitinib (poor/intermediate risk, 2018) 1,070 41.6 vs. 26.5 11.5 vs. 8.4 NR vs. 26
Avelumab/axitinib vs. sunitinib (2019) 886 55 vs. 25.5 13.8 vs. 8.4 NR
Pembrolizumab/axitinib vs. sunitinib (2019) 840 59 vs. 36 15 vs. 11 NR

Tenold et al
Historically, the VEGFR-TKIs sunitinib and pazopanib have all approved as first-line therapy; however, the disease
shown similar efficacy as frontline monotherapy.17 However, management after first-line therapy, and particularly after
in the phase II CABOSUN trial, cabozantinib was shown to receipt of ICB, is nuanced and requires consideration of
be superior to sunitinib in a study of IMDC-Int/Poor disease, many patient- and disease-related factors.
with an ORR of 46% versus 18%, median PFS of 8.2 versus
5.6 months, and median OS of 30 versus 21.8 months; Principles Underlying Decision-Making in the Previously
however, the response rate in the sunitinib arm of this study Treated Setting
was exceptionally poor and did not appear concordant with
Primary treatment failure Overall, 15% to 25% of patients
prior studies. An interesting observation was the better
experience disease progression as their best response to
response seen in patients with osseous metastatic disease
first-line TKI17 or combination ICB (ipilimumab/nivolumab).14
treated with cabozantinib.18 Although the mTOR inhibitor
Although this proportion has fallen to approximately 10%
temsirolimus remains US Food and Drug Administration
with TKI-ICB combinations (axitinib/avelumab, axitinib/
approved in the frontline setting, and at one time was often
pembrolizumab),16 primary treatment refractoriness re-
used for patients with poor-risk disease, it is becoming
mains an important clinical problem.
evident that this choice is no longer viable in the context of
more active agents given its relatively low response rate and When managing disease with apparent treatment fail-
the lack of comparative trials in the VEGFR-TKI and ICB ure, we consider the possibility of mixed response or
era.19 pseudoprogression, which is usually encountered at the
first restaging scans; consideration should be given to
Finally, there may be a subset of patients deemed
continuing therapy and obtaining a short interval scan to
“immunotherapy-eligible” but who may be unable to tolerate
confirm progression. Notably, 13% of patients treated
(or who refuse) combination therapy or TKIs. Single-agent
beyond progression with nivolumab in the CheckMate-
pembrolizumab was assessed in KEYNOTE-427 cohort A,
025 study experienced a 30% or greater decline in tumor
a phase II single-arm trial in treatment-naive patients with
burden. 21
mRCC and clear cell histology.20 The study demonstrated
an encouraging single agent ORR of 33.6% (95% CI, If disease progression is confirmed, switching to an alter-
24.8–43.4) and a median PFS of 6.9 months (95% CI, native class of agent is recommended. If ICB is used in first-
5.1–NR), with 86% of responders having their response line, a TKI is preferred (see the “Systemic Treatment Options
lasting at least 3 months. After ICB” section); the only available prospective data in
this setting are for axitinib from a phase II single-arm trial,22
Conclusions though subgroup analysis of the phase III METEOR trial
Risk stratification using the IMDC prognostic model con- showed activity for cabozantinib in the post-ICB setting.23 If
tinues to be a helpful tool for predicting disease outcomes single-agent TKI is used in first-line therapy, data from
and is a good starting place for new patient assessment. CheckMate-025 support the use of nivolumab as second
Active surveillance may be an appropriate option for selected line,24 but the combination of ipilimumab/nivolumab is also
patients who are asymptomatic and have slow-growing, used in practice. There are limited data on optimal man-
low-volume disease. Cytoreductive nephrectomy may be agement after progression on first-line TKI-ICB combination
helpful for selected patients evaluated by a multidisci- therapy; in our practice, we switch to cabozantinib or the
plinary tumor board. Surgical metastasectomy may also combination of lenvatinib and everolimus in this setting (if
be considered in select patients with oligometastatic no prior exposure to these agents) or enroll patients in
disease, but prospective trials are needed. The com- a clinical trial.
bination of VEGFR-TKI therapy with checkpoint inhibitor There are ongoing phase II and III trials assessing adaptive
immunotherapy is a reasonable first-line option for most approaches to immunotherapy that will provide information
patients, but it may also be reasonable to preserve the on appropriate therapy after first-line ICB. These are testing
introduction of VEGFR-TKIs until the second line for whether patients who experience disease progression on
patients with higher risk disease. With unusual excep- first-line nivolumab can be “salvaged” with the addition of
tions, VEGFR-TKI monotherapy and mTOR inhibitors are ipilimumab (TITAN-RCC [NCT02917772], OMNIVORE
no longer appropriate options in the frontline setting. [NCT03203473], HCRN: GU16-260 [NCT03117309]). The
BEYOND FIRST-LINE TREATMENT OF METASTATIC RENAL PDIGREE study (NCT03793166) is testing the activity of
CELL CARCINOMA cabozantinib in patients who experience disease progres-
sion on ipilimumab/nivolumab, and randomly selecting
Introduction those with stable disease or partial response to receive
There have been many advances in the treatment of mRCC maintenance nivolumab or the combination of nivolumab
in the past decade, with TKIs, ICB, or TKI-ICB combinations and cabozantinib.

Intolerance More than 50% of patients receiving TKI or TKI- disease (oligoprogressive disease) with other areas con-
based combinations experience sufficiently severe side tinuing to respond or remain stable. If possible, local
effects to require holding, dose-reducing, or stopping therapy to sites of oligoprogression (e.g., resection of a lung
therapy. Before we stop or switch therapy, we consider the or brain lesion, stereotactic radiotherapy to bone, brain or
following: soft-tissue lesions, ablation of liver lesions) should be
• Individualized dosing regimens: TKIs lend themselves pursued and the same systemic therapy may be contin-
well to adjustments of dose and frequency. The rec- ued. The rationale is that resistant clones may be ame-
ommended starting dose of cabozantinib is 60 mg but nable to eradication with local therapy while other disease
reduction to 40 mg or 20 mg (or the use of a higher/ sites remain sensitive to ongoing systemic therapy; retro-
lower dose on alternate days in an individualized spective data26,27 support this approach, and a phase II
schedule) can be attempted if a patient experiences trial is ongoing (NCT03696277).
bothersome side effects.25 The same is true of sunitinib In some cases, disease progression may be more wide-
(recommended dose, 50 mg daily for 4 weeks on/ spread and this is not feasible. Before abandoning the line of
2 weeks off), for which the dose can be lowered to 37.5 treatment, one consideration is to see whether dose es-
or 25 mg and the interval adjusted (e.g., to 2 weeks on/ calation is possible. For example, this means increasing the
1 week off). The feasibility and clinical activity of an TKI dose, if not at maximum already, to see whether the
individualized dosing regimen for axitinib was shown in disease can be stabilized with a higher dose (i.e., cabo-
a phase II study.22 zantinib 60 mg, axitinib 10 mg twice daily). If single-agent ICB
• Treatment holiday: Holding therapy, especially at a time is being used, one could consider adding in a second ICB
when disease is stable or responding, for a short period agent (i.e., ipilimumab to single-agent nivolumab), with
(e.g., 1–2 weeks) can be helpful in allowing patients to preliminary data suggesting that this may lead to responses in
recover from side effects and enable them to resume the approximately 10% of patients who initially have either stable
drug. Similarly, if disease has been stable or in ongoing disease or disease progression on nivolumab monotherapy.28
response for a prolonged period, a longer treatment However, this study also showed a much lower rate of
break can be considered, particularly if there is ongoing complete responses with sequential ipilimumab as compared
low-grade but bothersome toxicity. with using both agents concomitantly.
• Maximizing supportive care and compliance: Proactive
Obtaining a biopsy at the time of disease progression and
management of symptoms, such as hand-foot syndrome
performing next-generation sequencing can also be ben-
(moisturizers), stomatitis (mouthwash), and diarrhea
eficial. Although the likelihood of identifying an actionable
(antimotility agents), and aggressive treatment of hy-
mutation is low, off-label use of an approved targeted therapy
pertension can help prevent or alleviate many side effects
or the possibility of a biomarker-based clinical trial may exist
associated with TKIs.
for the small number of patients with a specific alteration,
• Involvement of appropriate specialists: Prompt referral to
especially in variant RCC subtypes.29
specialists in endocrinology, gastroenterology, derma-
tology, or nephrology can be particularly invaluable in Systemic treatment options after ICB The majority of patients
management of ICB-associated toxicities. ultimately experience disease progression on ICB, and there
In the current era where combination therapy (e.g., ICB-TKI) are limited data to guide us on optimal treatment after ICB.
is increasingly used, it is difficult to tease apart which of the Options include:
two agents is responsible for certain adverse effects (e.g., • TKI: Three retrospective studies (combined number of
diarrhea). Careful dose titration or discontinuation of one of patients, 136) have assessed the activity of a TKI
the drugs with clinical monitoring should be attempted before (sunitinib, pazopanib, axitinib, or cabozantinib) after
the combination is stopped for toxicity; in our practice, we disease progression on first-line ICB (single-agent,
have many examples of patients continuing to experience doublet, or VEGF/TKI-ICB combination).30-32 All have
response to monotherapy after one agent of the combination reported response rates on the order of 30% to 40%
has been discontinued. with second-line TKI (comparable to that seen in first-
line) with a median PFS of 6 to 13 months and no
Disease progression after response Disease growth after
apparent increase in toxicity. A prospective trial of
a period of response is common and poses a challenge to
axitinib after ICB involving 40 patients was recently
treating oncologists given that there are limited classes of
published, with the majority receiving axitinib in the
therapy (TKI, ICB, mTOR) approved for mRCC.
second- or third-line setting; 45% of patients respon-
In such cases, the nature of disease progression should first ded, with most continuing to respond beyond 12
be characterized. In many instances, there may be the months.22 Although the data are insufficiently granular
emergence of treatment resistance in one or a few sites of to allow determination of which TKI is best after prior

Tenold et al
ICB, we prefer axitinib or cabozantinib in our practice. This recommendation is supported by the findings from
Finally, a phase III study comparing a novel VEGF-TKI CheckMate-214 and KEYNOTE-426.14,15 In these phase III
(tivozanib) to sorafenib in the third- or fourth-line setting trials, ICB (nivolumab plus ipilimumab or pembrolizumab
has also been published; a PFS benefit was seen with plus axitinib, respectively) was compared with sunitinib in
tivozanib in the overall cohort as well as the approxi- previously untreated advanced clear cell RCC. In both
mately 25% of patients previously treated with a VEGF- studies, a large proportion of the patients were categorized
TKI/ICB combination.33 If approved, tivozanib may as having IMDC-Int or IMDC-Poor risk disease. Combination
therefore represent a reasonable option in this particular immunotherapy was favored in both trials because of higher
subset. ORR, improved quality of life, and longer OS seen in patients
• Lenvatinib/everolimus: This is approved after prior TKI across all risk groups and levels of tumor PD-L1 expression.
based on phase II data, but there are no data on its Patients with poor-risk disease receiving nivolumab plus
efficacy after prior ICB.34 It may represent a reasonable ipilimumab were also noted to have better responses
option as third-line therapy after ICB and TKI. compared with good-risk disease.14
• ICB: There are sparse data on the outcomes of using
Although patients with poor performance status were ex-
a different ICB (alone or in combination) after patients
cluded, it is believed that the clinical benefit, manageable
experience disease progression on ICB, but in our ex-
toxicities of immunotherapy, and improvements in quality of
perience in RCC, this approach has limited efficacy.35
life seen in practice supports its use in patients with low
• New targets: There is an urgent need for additional
performance status. A subset analysis of CheckMate-171
classes of therapy in RCC beyond VEGF-TKI and ICB.
demonstrated no increased toxicity signal for patients with
One promising target in clear cell RCC is HIF-2α, which
Eastern Cooperative Oncology Group PS 2 and age 70 or
is linked to the Von Hippel-Lindau pathway and oxygen
older treated for squamous non–small cell carcinoma of the
sensing. Phase I and II studies with HIF-2α inhibitors are
lung with nivolumab.37 Elderly patients (. age 70) treated
under way, with initial results showing evidence of
with sunitinib had an acceptable safety profile and similar
single-agent activity in heavily pretreated patients.36
clinical outcomes compared with younger patients, despite
There are efforts underway to combine this agent with
more frequent dose modification.38 If a component of de-
other TKIs or ICB in an attempt to boost response and
bilitation is thought to be related to extensive tumor burden,
circumvent resistance.
the TKI plus checkpoint inhibitor combination would be
Conclusions preferred given the higher and more rapid overall response
rates.
ICB, either as a doublet or in combination with VEGF-TKI, is
quickly becoming the standard of care for first-line treat- End-Organ Dysfunction
ment of mRCC. However, most patients ultimately require
RCC treatment often requires a multimodality approach to
additional lines of therapy, and oncologists must think
care including surgical management as well as systemic
carefully when switching to another therapy, particularly
therapy affecting both liver and kidney function critical in
in situations of drug intolerance or apparent disease pro-
drug elimination and toxicity. The extent of organ dys-
gression. Systemic therapy options after ICB are generally
function should be considered when selecting systemic
TKI-based, and ongoing clinical trials will help optimize the
options; however, at present, most therapies can be used
treatment algorithm. Nevertheless, despite many recent
with close monitoring.
drug approvals for RCC, there remains a pressing need to
identify new therapeutic targets in this disease. Although TKIs are metabolized in the liver and are known for
their potential to cause various levels of hepatotoxicity,
MANAGEMENT APPROACH TO SPECIAL impaired liver function is generally not a contraindication for
PATIENT POPULATIONS TKI therapy. Studies evaluating the use of sorafenib in
Poor Performance Status hepatocellular carcinoma (HCC) showed that although
outcomes were different in patients with Child-Pugh A
Some locally advanced or metastatic cases of RCC are di-
versus Child-Pugh B liver function, the safety and tolerability
agnosed in patients who exhibit important laboratory ab-
were similar.39 A further subset analysis of patients in the
normalities, site-related symptomology, and low performance
phase III SHARP trial found that sorafenib was safe and
status that in the past may have been a contraindication for
effective in patients with advanced HCC despite alterations
systemic therapy or suggestive of a more palliative approach.
in baseline liver enzymes.40 Of note, patients with Child-
As the systemic therapy landscape continues to evolve,
Pugh B or C were not included, thus it is difficult to ex-
immunotherapy has now been incorporated into first-line
trapolate any assessment of safety for patients with more
treatment for patients across all IMDC risk groups with
advanced disease. Current recommendations suggest
substantial clinical benefit.

reduction in starting dose based on initial liver function with disease progression on systemic therapy, or low/intermediate
subsequent titration based on individual tolerability. grade with complete resection.47
Immunotherapy with ICB has been found to be safe in At present, there is no role for systemic adjuvant therapy
patients with underlying liver dysfunction as well. CheckMate- after metastasectomy. A randomized, double-blind, placebo-
040 found that nivolumab elicited objective response rates in controlled multicenter phase III trial conducted by the Eastern
the phase I/II trial with a manageable safety profile in patients Cooperative Oncology Group-American College of Radiology
with advanced HCC with or without hepatitis B or C in- Imaging Network group evaluated the role of adjuvant
fection.41 This finding was reassuring, and the presence of pazopanib versus placebo in patients with no evidence of
chronic hepatitis is not considered a contraindication for disease after metastasectomy.48 The patients received
immunotherapy. In KEYNOTE-426, patients in the pem- treatment for 52 weeks and were additionally stratified by
brolizumab/axitinib group were found to have a higher in- number of sites of resected disease as well as disease-free
cidence of grade 3 or 4 elevations in liver enzyme levels that interval. The study was unable to meet a primary endpoint for
were managed successfully with treatment interruptions and disease-free survival with a hazard ratio of 0.85 (0.55–1.31, p
corticosteroids.15 Additionally, clinical experience continues = .47) in favor of placebo. There was also a trend toward
to support that immunotherapy can be used safely and ef- worse OS with pazopanib. Negative results were also ob-
fectively in patients with baseline organ dysfunction with served with postmetastasectomy sorafenib.49 There are on-
appropriate clinical monitoring.42 going trials studying use of adjuvant sunitinib or checkpoint
Although there are limited data evaluating the use of either inhibitors, but currently there is insufficient evidence to
TKIs or immunotherapy in patients with end-stage renal support use of these over observation.
disease or chronic kidney disease, there have been multiple With data emerging on molecular characteristics as pre-
case reports in which patients were treated safely and dictive markers as well as the morphing landscape of
responded well to treatment.43,44 Product information for systemic immunotherapy and targeted therapy, it is unclear
many of these drugs does not recommend any dosing what role localized surgical intervention and/or consolida-
adjustments before initiation based on renal impairment, tion will continue to have. For now, the clinical experience of
and it is believed that these therapies can be effectively used physicians on multidisciplinary teams should guide the
in these situations. recommendations for metastasectomy in patients with ad-
Post-metastasectomy vanced RCC.
Although metastasectomy in advanced RCC has been in Brain Metastasis

clinical practice since the 1930s, to date there have been no Brain metastases occur in approximately 25% of patients
randomized controlled trials that have evaluated the clinical with RCC, with only 6% to 8% presenting with symptomatic
benefit of such a procedure. The evidence in support of lesions. Initially thought to confer a poor prognosis, more
favorable outcomes has largely come from retrospective active systemic therapies and the increasing use of ste-
case series and case reports in both RCC as well as other reotactic radiation has produced survival outcomes similar
solid tumors. A systematic review by Achkar et al identified to patients without brain metastases.50 Corticosteroids are
44 studies that reported outcomes in patients who un- indicated in any symptomatic patient, especially those with
derwent metastasectomy at any time. In general, complete cerebral edema, with initial dosing of dexamethasone at
metastasectomy was independently associated with a re- 4 mg to 8 mg per day in divided doses and tapered as
duction in mortality and improved median OS com- clinically indicated. Additionally, antiepileptics should be
pared with incomplete surgery (range, 36.5–142 months vs. initiated in patients who present with seizures.
8.4–27 months).45 Although historically thought of as a sanctuary site, systemic
There are no clear guidelines for patient selection for whom therapies, in particular immune checkpoint inhibitors, do
will benefit most from surgery. The National Comprehensive have activity in the central nervous system. Based on hy-
Cancer Network guidelines suggest metastasectomy could potheses related to neo-angiogenesis necessary for meta-
be considered in patients with clear cell or non–clear cell static spread to the brain, VEGFR TKIs have been used with
histology who initially present with primary RCC and oligo- very modest responses reported. A phase II study showed
metastatic sites or develop oligometastases after a prolonged no objective responses in patients receiving sunitinib, al-
disease-free interval from nephrectomy.46 The European So- though central nervous system disease remained stable in
ciety for Medical Oncology recommends localized surgery 31%.51 Results with immune checkpoint inhibitors appear
upon recommendation by a multidisciplinary team to identify more favorable. NIVOREN, a phase II study of second-line
patients who in general have good performance status, nivolumab in patients with RCC demonstrated an ORR of
solitary or oligometastases, metachronous disease with 23% in untreated, asymptomatic brain metastases.52 Brain
disease-free interval for longer than 2 years, absence of metastases response rates as high as 46% have been

Tenold et al
reported in patients with melanoma treated with ipilimumab both stereotactic radiation to amenable lesions and sys-
and nivolumab.53 Although the optimal approach for pa- temic therapy with immune checkpoint inhibitors show the
tients with RCC and brain metastases remains to be determined, most promise.
AFFILIATIONS Sacramento, CA 95817; Twitter: @PrimoLara; email: pnlara@

1
UC Davis Comprehensive Cancer Center, Sacramento, CA ucdavis.edu.
2
Dana Farber Cancer Institute, Boston, MA
3
University of Texas Southwestern Harold C. Simmons Comprehensive
Cancer Center, Dallas, TX
Primo N. Lara, Jr., MD, University of California Davis School of Medicine,
UC Davis Comprehensive Cancer Center, 4501 X St., Suite 3003,
REFERENCES
1. Stat Fact Sheets SEER. Kidney and Renal Pelvis. http://seer.cancer.gov/statfacts/html/kidrp.html.
2. Patard JJ, Leray E, Rioux-Leclercq N, et al. Prognostic value of histologic subtypes in renal cell carcinoma: a multicenter experience. J Clin Oncol. 2005;
23:2763-2771.
3. Richie JP, Skinner DG. Renal Neoplasia. In Brenner BM, Rector F, (eds). The Kidney, 2nd ed. Philadelphia: WB Saunders; 1981;2109.
4. Ridge CA, Pua BB, Madoff DC. Epidemiology and staging of renal cell carcinoma. Semin Intervent Radiol. 2014;31:3-8.
5. Heng DY, Xie W, Regan MM, et al. External validation and comparison with other models of the International Metastatic Renal-Cell Carcinoma Database
Consortium prognostic model: a population-based study. Lancet Oncol. 2013;14:141-148.
6. Rini BI, Dorff TB, Elson P, et al. Active surveillance in metastatic renal-cell carcinoma: a prospective, phase 2 trial. Lancet Oncol. 2016;17:1317-1324.
7. Flanigan RC, Salmon SE, Blumenstein BA, et al. Nephrectomy followed by interferon alfa-2b compared with interferon alfa-2b alone for metastatic renal-cell
8. Mickisch GH, Garin A, van Poppel H, et al; European Organisation for Research and Treatment of Cancer (EORTC) Genitourinary Group. Radical nephrectomy
plus interferon-alfa-based immunotherapy compared with interferon alfa alone in metastatic renal-cell carcinoma: a randomised trial. Lancet. 2001;
358:966-970.
9. Méjean A, Ravaud A, Thezenas S, et al. Sunitinib alone or after nephrectomy in metastatic renal-cell carcinoma. N Engl J Med. 2018;379:417-427.
10. Bex A, Mulders P, Jewett M, et al. Comparison of immediate vs deferred cytoreductive nephrectomy in patients with synchronous metastatic renal cell carcinoma
receiving sunitinib: the SURTIME Randomized Clinical Trial. JAMA Oncol. 2019;5:164-170.
11. Ouzaid I, Capitanio U, Staehler M, et al; Young Academic Urologists Kidney Cancer Working Group of the European Association of Urology. Surgical meta-
stasectomy in renal cell carcinoma: a systematic review. Eur Urol Oncol. 2019;2:141-149.
12. Verman R, Lara PN. Cytokines in the Management of Advanced Renal Cell Cancer. In Lara PN, Jonasch E, (eds). Kidney Cancer: Principles and Practice, 2nd ed.
Switzerland: Springer International; 2015;245-258.
13. Motzer RJ, Hutson TE, Tomczak P, et al. Sunitinib versus interferon alfa in metastatic renal-cell carcinoma. N Engl J Med. 2007;356:115-124.
14. Motzer RJ, Tannir NM, McDermott DF, et al; CheckMate 214 Investigators. Nivolumab plus ipilimumab versus sunitinib in advanced renal-cell carcinoma. N Engl
J Med. 2018;378:1277-1290.
15. Motzer RJ, Penkov K, Haanen J, et al. Avelumab plus axitinib versus sunitinib for advanced renal-cell carcinoma. N Engl J Med. 2019;380:1103-1115.
16. Rini BI, Plimack ER, Stus V, et al; KEYNOTE-426 Investigators. Pembrolizumab plus axitinib versus sunitinib for advanced renal-cell carcinoma. N Engl J Med.
2019;380:1116-1127.
17. Motzer RJ, Hutson TE, Cella D, et al. Pazopanib versus sunitinib in metastatic renal-cell carcinoma. N Engl J Med. 2013;369:722-731.
18. Choueiri TK, Halabi S, Sanford BL, et al. Cabozantinib versus sunitinib as initial targeted therapy for patients with metastatic renal cell carcinoma of poor or
intermediate risk: the Alliance A031203 CABOSUN Trial. J Clin Oncol. 2017;35:591-597.
19. Hudes G, Carducci M, Tomczak P, et al; Global ARCC Trial. Temsirolimus, interferon alfa, or both for advanced renal-cell carcinoma. N Engl J Med. 2007;
356:2271-2281.
20. Donskov F, McDermott DF, Lee JL, et al. KEYNOTE-427 COHORT A: pembrolizumab monotherapy as first-line therapy in advanced clear cell renal cell carcinoma
(CCRCC). Ann Oncol. 2018;29 (suppl 8).
21. Escudier B, Motzer RJ, Sharma P, et al. Treatment beyond progression in patients with advanced renal cell carcinoma treated with nivolumab in CheckMate 025.
Eur Urol. 2017;72:368-376.

22. Ornstein MC, Pal SK, Wood LS, et al. Individualised axitinib regimen for patients with metastatic renal cell carcinoma after treatment with checkpoint inhibitors:
a multicentre, single-arm, phase 2 study. Lancet Oncol. 2019;20:1386-1394.
23. Powles T, Motzer RJ, Escudier B, et al. Outcomes based on prior therapy in the phase 3 METEOR trial of cabozantinib versus everolimus in advanced renal cell
carcinoma. Br J Cancer. 2018;119:663-669.
24. Motzer RJ, Escudier B, McDermott DF, et al; CheckMate 025 Investigators. Nivolumab versus everolimus in advanced renal-cell carcinoma. N Engl J Med. 2015;
373:1803-1813.
25. McElwee JH, Gourdin TS, Mikoll J, et al. Cabozantinib use in metastatic renal cell carcinoma patients in clinical practice: Evaluation of dosing patterns, tolerability,
and outcomes compared to clinical trials. J Oncol Pharm Pract. Epub 29 Sept 2019.
26. Santini D, Ratta R, Pantano F, et al. Outcome of oligoprogressing metastatic renal cell carcinoma patients treated with locoregional therapy: a multicenter
retrospective analysis. Oncotarget. 2017;8:100708-100716.
27. Barata PC, Mendiratta P, Kotecha R, et al. Effect of switching systemic treatment after stereotactic radiosurgery for oligoprogressive, metastatic renal cell
carcinoma. Clin Genitourin Cancer. 2018;16:413-419.
28. Grimm MO. Tailored immunotherapy approach with nivolumab in advanced renal cell carcinoma (TITAN-RCC). Ann Oncol. 2019;30 (suppl 5):v851-v934.
29. Pal SK, Ali SM, Ross J, et al. Exceptional response to palbociclib in metastatic collecting duct carcinoma bearing a CDKN2A homozygous deletion. JCO Precision
Oncol. 2017;1:1-5.
30. Barata PC, De Liano AG, Mendiratta P, et al. The efficacy of VEGFR TKI therapy after progression on immune combination therapy in metastatic renal cell
carcinoma. Br J Cancer. 2018;119:160-163.
31. Auvray M, Auclin E, Barthelemy P, et al. Second-line targeted therapies after nivolumab-ipilimumab failure in metastatic renal cell carcinoma [published erratum
appears in Eur J Cancer. 2019;119:200-201]. Eur J Cancer. 2019;108:33-40.
32. Shah AY, Kotecha RR, Lemke EA, et al. Outcomes of patients with metastatic clear-cell renal cell carcinoma treated with second-line VEGFR-TKI after first-line
immune checkpoint inhibitors. Eur J Cancer. 2019;114:67-75.
33. Rini BI, Pal SK, Escudier BJ, et al. Tivozanib versus sorafenib in patients with advanced renal cell carcinoma (TIVO-3): a phase 3, multicentre, randomised,
controlled, open-label study. Lancet Oncol. 2020;21:95-104.
34. Motzer RJ, Hutson TE, Glen H, et al. Lenvatinib, everolimus, and the combination in patients with metastatic renal cell carcinoma: a randomised, phase 2, open-
label, multicentre trial. Lancet Oncol. 2015;16:1473-1482.
35. Martini DJ, Lalani AA, Bossé D, et al. Response to single agent PD-1 inhibitor after progression on previous PD-1/PD-L1 inhibitors: a case series. J Immunother
Cancer. 2017;5:66.
36. Courtney KD, Infante JR, Lam ET, et al. Phase I dose-escalation trial of PT2385, a first-in-class hypoxia-inducible factor-2α antagonist in patients with previously
treated advanced clear cell renal cell carcinoma. J Clin Oncol. 2018;36:867-874.
37. Felip E, Ardizzoni A, Ciuleanu T, et al. CheckMate 171: a phase 2 trial of nivolumab in patients with previously treated advanced squamous non-small cell lung
cancer, including ECOG PS 2 and elderly populations. Eur J Cancer. 2020;127:160-172.
38. Hutson TE, Bukowski RM, Rini BI, et al. Efficacy and safety of sunitinib in elderly patients with metastatic renal cell carcinoma. Br J Cancer. 2014;
110:1125-1132.
39. McNamara MG, Slagter AE, Nuttall C, et al. Sorafenib as first-line therapy in patients with advanced Child-Pugh B hepatocellular carcinoma-a meta-analysis. Eur
J Cancer. 2018;105:1-9.
40. Raoul JL, Bruix J, Greten TF, et al. Relationship between baseline hepatic status and outcome, and effect of sorafenib on liver function: SHARP trial subanalyses.
J Hepatol. 2012;56:1080-1088.
41. El-Khoueiry AB, Sangro B, Yau T, et al. Nivolumab in patients with advanced hepatocellular carcinoma (CheckMate 040): an open-label, non-comparative, phase
1/2 dose escalation and expansion trial. Lancet. 2017;389:2492-2502.
42. Kanz BA, Pollack MH, Johnpulle R, et al. Safety and efficacy of anti-PD-1 in patients with baseline cardiac, renal, or hepatic dysfunction. J Immunother Cancer.
2016;4:60.
43. Ansari J, Ali M, Farrag A, et al. Efficacy of nivolumab in a patient with metastatic renal cell carcinoma and end-stage renal disease on dialysis: case report and
literature review. Case Reports Immunol. 2018;2018:1623957.
44. Czarnecka AM, Kawecki M, Lian F, et al. Feasibility, efficacy and safety of tyrosine kinase inhibitor treatment in hemodialyzed patients with renal cell cancer:
10 years of experience. Future Oncol. 2015;11:2267-2282.
45. Achkar T, Maranchie JK, Appleman LJ. Metastasectomy in advanced renal cell carcinoma: a systematic review. Kidney Cancer. 2019;3:31-40.
46. National Comprehensive Cancer Network. Kidney Cancer (version 2.2020). https://www.nccn.org/professionals/physician_gls/pdf/kidney.pdf.
47. Escudier B, Porta C, Schmidinger M, et al; ESMO Guidelines Committee. Renal cell carcinoma: ESMO Clinical Practice Guidelines for diagnosis, treatment and
follow-up. Ann Oncol. 2019;30:706-720.
48. Appleman L, Puligandla M, Pal SK, et al. Randomized, double-blind phase III study of pazopanib versus placebo in patients with metastatic renal cell carcinoma
who have no evidence of disease following metastasectomy: a trial of the ECOG-ACRIN cancer research group (E2810). J Clin Oncol. 2019;37 15s (suppl; abstr
4502).
49. Procopio G, Apollonio G, Cognetti F, et al. Sorafenib versus observation following radical metastasectomy for clear-cell renal cell carcinoma: results from the phase
2 randomized open-label RESORT Study. Eur Urol Oncol. 2019;2:699-707.

Tenold et al
50. Wardak Z, Christie A, Bowman A, et al. Stereotactic radiosurgery for multiple brain metastases from renal-cell carcinoma. Clin Genitourin Cancer. 2019;
17:e273-e280.
51. Chevreau C, Ravaud A, Escudier B, et al; French Group on Renal Cancer. A phase II trial of sunitinib in patients with renal cell cancer and untreated brain
metastases. Clin Genitourin Cancer. 2014;12:50-54.
52. Flippot R, Dalban C, Laguerre B, et al. Safety and efficacy of nivolumab in brain metastases from renal cell carcinoma: results of the GETUG-AFU 26 NIVOREN
Multicenter phase II study. J Clin Oncol. 2019;37:2008-2016.
53. Long GV, Atkinson V, Lo S, et al. Combination nivolumab and ipilimumab or nivolumab alone in melanoma brain metastases: a multicentre randomised phase 2

Renal Cell and Urothelial Carcinoma: Biomarkers

for New Treatments
Andrew L. Schmidt, MD1; Arlene Siefker-Radtke, MD2; David McConkey, PhD, MD3; and Bradley McGregor, MD1
overview
Therapies for genitourinary malignancies have evolved considerably in the past 5 years. Combination
treatment targeting biologically relevant immune and angiogenic pathways is improving patient survival in
metastatic renal cell carcinoma (RCC), whereas immune checkpoint blockade (ICB), novel targeted therapy,
and antibody drug conjugates have changed the landscape of urothelial cancer (UC) treatment. A daily
challenge for clinicians is identifying patients who derive a preferential benefit from the available therapeutic
options. The completion of large-scale genomics projects has yielded comprehensive descriptions of the
molecular heterogeneity present in RCC and UC, although clinical applications of these data continue to
evolve. Major molecular subtypes of RCC align well with histology subtype, and although some molecular
characteristics appear to carry prognostic information, biomarkers predicting benefit from tyrosine kinase
inhibitor (TKI) or immunotherapy are generally lacking. Unexpectedly, similar work has demonstrated that UC
can be grouped into “molecular subtypes” that share properties with those found in breast cancer and other
solid tumors. Furthermore, this molecular subtype classification is prognostic and potentially predictive of
differential benefit from conventional and targeted therapies. This article provides an update on the current
state of molecular biomarker development and potential clinical utility in RCC and UC.
INTRODUCTION estimates differ among risk groups: 43.2 months, 22.5

Predicting therapeutic benefit and acquired resistance months, and 7.8 months, respectively.1 As frontline
to therapies in any malignancy remains a daily chal- combination trials stratified patients and defined pri-
lenge for clinicians. This dilemma is no better exem- mary endpoints based on IMDC risk group, this
plified than in RCC and UC, for which biomarkers prognostic information also has implications for treat-
with a binary actionable outcome are not available. ment selection or even the decision to pursue obser-
As physicians are inundated with an increase in in- vation. It should be noted that IMDC criteria are based
formation from patient characteristics to next-generation on data collected from TKI therapy, and the applica-
sequencing (NGS), utilizing this information to guide bility of these criteria in the context of combination
treatment decisions can be challenging. We explore treatment with immunotherapy will likely need re-
the current biomarkers available in RCC and UC and evaluation and evolve with time. In particular, the
their predictive and prognostic relationships to avail- paradox of poor-risk patients who have a remarkable
able therapies. response to immunotherapy and long-term survival
BIOMARKERS IN METASTATIC CLEAR CELL remains unresolved under the current criteria.
RENAL CARCINOMA For patients with favorable-risk disease, surveillance
Although tissue biomarkers continue to evolve, the is an important clinical tool, primarily used for those
International Metastatic Renal Cell Cancer Database with low-volume or quiescent disease.2 More accurate
Consortium (IMDC) risk score or Heng criteria remain risk stratification allows clinicians and patients greater
Author affiliations
a robust set of prognostic criteria that also form the clarity when selecting this as a strategy. One proposed
and support tool is the neutrophil/lymphocyte ratio (NLR), a vali-
information (if
basis of risk stratification for clinical trials. The criteria
applicable) appear include low Karnofsky performance status (less than dated prognostic factor across a multitude of clinical
at the end of this 80%), low serum hemoglobin, high corrected serum settings in RCC, including the risk of recurrence
article. calcium, time from initial RCC diagnosis to start of postnephrectomy, progression post-cytoreductive
Accepted on XXX and therapy of less than 1 year, increased neutrophil count, nephrectomy, and outcome with VEGF-TKI or ICB
published at treatment.3-6 As such, it is intuitive to use NLR to
and increased platelets. Patients with no risk factors
ascopubs.org on May
7, 2020: DOI https://
have favorable risk, those with one or two risk factors refine risk and identify patients with a low risk for
doi.org/10.1200/ have intermediate risk, and patients with three or more progression to undergo surveillance. However, the
EDBK_279905 risk factors have an unfavorable risk profile. Survival presence of an unfavorable/favorable NLR overlaps

Schmidt et al
receive U.S. Food and Drug Administration approval or

achieve an advantage in good-risk disease, IMDC risk
classification remains an important tool in discussions about
• IMDC risk category remains an important frontline systemic therapy. However, although nivo-ipi did
prognostic tool that may guide first-line treat-
not improve response rate or progression-free survival
ment selection.
versus sunitinib in good-risk disease, patients with good-risk
• Although patients respond to therapies in- disease still responded to nivo-ipi, and the complete re-
dependent of PD-L1 expression, and it is not sponse (CR) rate approached 10%. Furthermore, there was
routinely checked in RCC, it may be useful to
no detriment in overall survival.6 Although not endorsed by
define the relative risk of progression versus
guidelines or regulatory approvals, given the chance at long-
response with dual-ICB versus ICB/VEGF
blockade. term remission, this remains an attractive option, even in
good-risk disease. Furthermore, in updated analyses, the
• Other predictive biomarkers for management of
benefit of axitinib-pembrolizumab in good-risk disease was
RCC are lacking.
less impressive, suggesting that VEGF monotherapy could
• Activating mutations in FGFR select patients still be considered. Further biomarker research is clearly
with UC for potential treatment with erdafitinib. needed to subset good-risk disease and identify the re-
• The antibody drug conjugates enfortumab sponders who benefit from ICB, with or without VEGF.
vedotin and sacituzumab govitecan target
In patients with intermediate- or poor-risk disease, for whom
proteins with near-universal expression in UC;
thus, treatment selection based on protein all options are approved by regulatory agencies, the likeli-
expression is not required. hood of response and toxicity profile are important variables
when considering treatment. If a patient has pending vis-
• Differentiating UC into basal and luminal mo-
ceral crisis, the improvement in overall response rate (ORR)
lecular subtypes shows promise at predicting
sensitivity to chemotherapy, but it needs to near 60% with axitinib-pembrolizumab versus less than
rigorous prospective evaluation. 40% with nivo-ipi supports its use. However, other than
clinical variables, is there a reliable marker to determine the
best therapeutic option? Unlike in some other solid-organ
with a conventional IMDC risk factor (i.e., elevated neu- malignancies, such as lung cancer and melanoma, tumor
trophil count) that already guides prognostic decisions.1 As mutational burden has represented a poor correlate for
such, the addition of NLR within the context of validated response to ICB in RCC, with responses agnostic to tumor
clinical variables and IMDC risk, which largely captures mutational burden.9 As such, although often reported in
a favorable NLR by categorizing patients as favorable risk, NGS reports, it should not drive treatment decisions in
adds little resolution when considering surveillance. metastatic RCC. A strong association exists between ex-
When treatment is chosen, first-line combination treatments pression of PD-L1 via immunohistochemistry and response
utilizing ICB have been approved by the regulatory agencies rate to nivo-ipi. Demonstrated in CheckMate 214, 26% of
in metastatic RCC. Nivolumab plus ipilimumab (nivo-ipi) patients in the nivo-ipi cohort harbor PD-L1+ disease, which
and axitinib plus pembrolizumab confer an overall survival is associated with a higher objective response rate than
advantage over sunitinib in intermediate-risk and poor-risk PD-L1 disease, both superior to sunitinib although overall
clear cell RCC (ccRCC), with the latter offering a survival survival was maintained with nivo-ipi independent of
advantage independent of risk classification. Axitinib plus PD-L1 status, underscoring the importance of ICB in
avelumab offers an improvement in progression-free sur- RCC (Table 1).6,10
vival over sunitinib across risk groups.6-8 As nivo-ipi did not PD-L1 status should not be used to determine whether ICB
should be offered to a patient with RCC. However, when
TABLE 1. Relative Risk of Progression Versus Response to Nivo-Ipi in expressing the relative risk of response versus progression,
CheckMate 214 IMDC Intermediate/Poor-Risk Cohort Based on PD-L1 patients with PD-L1+ disease are 4.14 times more likely to
Status have a response to nivo-ipi than to experience disease
Response All Patients PD-L1 < 1% PD-L1 ‡ 1% progression; the relative risk is 1.85 in patients with PD-L1
ORR 42% 37% 58% disease. As such, in patients with PD-L1 disease who risk
PD 20% 20% 14%
being unable to receive salvage treatment upon progres-
sion, the combination of a TKI with ICB is an attractive option
PD:ORR 1:2.1 1:1.85 1:4.14
with a more favorable risk-benefit profile, because axitinib-
Abbreviations: IMDC, International Metastatic Renal Cell Cancer ICB combinations demonstrated high response rates and
Database Consortium; ORR, overall response rate; PD, progressive did not have a differential outcome based on PD-L1 status
disease. (despite that being the primary outcome with axitinib-

Biomarkers in Renal Cell and Urothelial Carcinoma
avelumab).7,8 It should be noted that there has not been relative to pRCCs or chromophobe RCCs,16,17 consistent
a significant difference with regard to survival outcome with the unique role of the VHL-HIF pathway in their de-
based on PD-L1 expression levels, but trials exclude pa- velopment and their relative sensitivity to immunotherapy.
tients who have critical disease, and immediate response is News of the 2019 Nobel Prize in Physiology or Medicine
more highly linked to survival in this context because many being awarded to three investigators who discovered how
patients do not receive second-line treatment. the VHL-HIF pathway controls the cellular response to
hypoxia has focused interest on the development of novel
The presence of sarcomatoid or rhabdoid histology is
agents targeting this pathway, with a promising small
associated with a more aggressive disease course and
molecule inhibitor of HIF-2 under clinical evaluation and
a low likelihood of response to TKI monotherapy.11 A
encouraging results to date.23,24 Studies in cohorts of pa-
uniformly strong efficacy signal in frontline clinical trials
tients with ccRCC have identified potentially targetable
that incorporates checkpoint inhibitors has highlighted
genomic alterations in 13% of patients, with the clear
the importance of incorporating ICB therapy for these
majority being alterations in TSC1, PIK3A, or MET.25,26 Al-
patients.12-14 However, a differential response is noted with
terations in key components of the PI3K/mTOR pathway
nivo-ipi; ORR approaches 60% and CR rate approaches
have been candidates to confer a response to rapalog
20%, although it continues to carry twice the risk for primary
therapy, but clinical data have yielded conflicting results
progressive disease compared with axitinib-pembrolizumab
on retrospective analyses.27,28 In a similar fashion, MET
(25% vs. 13.7%). Thus, as is the case with RCC without
alterations are thought to represent a dominant pathologic
sarcomatoid features, in patients who have limited conse-
pathway actionable through MET-directed therapy.29
quences from progression (e.g., presence of low-volume
However, because the therapeutic sequence of treatment
pulmonary metastases), a greater chance for long-term
lines commonly involves exposure to the rapamycin analog
remission utilizing nivo-ipi may be the favored treatment.
everolimus and TKI therapy involving agents that target
For patients in whom progression may compromise sub-
MET, such as cabozantinib (a multikinase inhibitor that
sequent therapy, the lower primary progression rate while
inhibits VEGF, MET, and AXL), the relevance of these
on axitinib-ICB treatment is appealing, but it comes at the
common alterations remains speculative and they do not yet
expense of a reduction in the number of complete
inform treatment choice, although they may play a greater
responders.7,8,13,15
role in the management of patients with variant histol-
With regard to prognosis, inactivation of TP53 or the CDKN2A ogy. For a summary of biomarkers related to RCC, refer to
tumor suppressors or a CpG island hypermethylation phe- Table 2.
notype is associated with poor prognosis across histologic
Variant histology remains a poor prognostic factor.30 This
subtypes, whereas inactivating mutations in the BAP1 gene
is an umbrella category for a heterogeneous group of ma-
are associated with poor prognosis in ccRCC and type 1
lignancies including papillary, chromophobe, medullary,
papillary RCC (pRCC).16,17 The PBRM1 gene encodes the
collecting duct, TFE3 translocation, and unclassified RCC
protein polybromo-1, which acts as a tumor suppressor in
variants.31 The Cancer Genome Atlas project largely con-
RCC.18 Functional loss is present in 40% of cases of RCC,
firmed that there are strong molecular bases for the his-
and it has been associated with increased survival in ret-
tologic heterogeneity observed in variant RCC.17,32,33 Whether
rospective trial analyses; recent data from the NIVOREN
these metabolic features can be used to develop novel
GETUG-AFU26 trial confirmed its prognostic significance
targeted therapies remains to be determined. As an ex-
in the largest ICB cohort examined to date, with a modest
ample, transport genes are enriched in chromophobe RCC,
effect size.19 Results from retrospective analyses have
and differences in metabolic gene expression and RNA
suggested that PBRM1 mutations confer sensitivity to
splicing distinguish it from the major RCC histologic
ICB20,21; in contrast, a study failed to reproduce this ob-
variants.16,17,32 MET activating mutations and copy number
servation and instead linked mutations to benefit from
alterations are a hallmark of pRCCs, which can be segre-
sunitinib.22 Several ongoing analyses examining PBRM1
gated into type 1 and type 2 subsets based on messenger
mutation status will be critical to resolve these contrasting
RNA expression.16,17 The role of activating MET alterations
data sets before it can be recommended as a predictive
in response to MET inhibitors remains unclear.34 In one
biomarker in the context of patients receiving ICB therapy
phase II clinical trial of the MET inhibitor foretinib, a cor-
upfront or in sequence.
relation was observed between the presence of a germline
The role of NGS in RCC is evolving to consider the muta- MET mutation and response.35 In contrast, in a retrospective
tional landscape of metastatic RCC, ideally to identify pre- real-world study of the effects of cabozantinib (a multikinase
dictive markers that inform first-line treatment choice or inhibitor incorporating significant activity against MET),
direct patients to a targeted therapy. ccRCC exhibits en- somatic alterations were not predictive of benefit.36 A large
hanced vascular development and immune gene signatures collaborative prospective trial evaluating the MET inhibitor

Schmidt et al
cabozantinib versus sunitinib in pRCC (PAPMET/SWOG TARGETING METASTATIC UROTHELIAL CANCER

1500) is underway and should help to reconcile these Similar challenges to RCC exist when assessing the treat-
observations.37 ment landscape of metastatic UC; again, the selection of
Salvolitinib is another small molecule inhibitor of MET that systemic treatment is not aided to a great extent by readily
has demonstrated activity in pRCC.38,39 Although closed available biomarkers. The recent approval of five immune
early to recruitment, the phase III trial result examining checkpoint inhibitors targeting PD-1 and PD-L1 has
salvolitinib versus sunitinib is expected and may shed light resulted in much fanfare, primarily as a result of a durability
on the differential activity against sunitinib.40 However, at of response that has not been observed in the era of
this time, treatment continues to be extrapolated from cisplatin-based chemotherapy. However, the proportion of
ccRCC, and a phase II trial examining atezolizumab and patients who respond remains low, with most trials reporting
bevacizumab in combination for variant histology (ORR, an approximately 15% to 20% objective response rate in
26% vs. 37% in the ccRCC trial IMmotion151) supports this the second-line setting47,48 and slightly higher rates in the
approach, albeit with less efficacy.41,42 More conventional frontline setting.49,50 Attempts to use PD-L1 expression for
biomarkers of immune infiltration (with the exception of patient selection has yielded little benefit in patients who
tumor mutational burden or tumor neoantigen burden) have received prior treatment. Recent data from IMvi-
appear to more consistently enrich for clinical benefit,22 gor130, a trial of first-line chemotherapy with or without
such as PD-L1 expression, although they are still not ro- a checkpoint inhibitor versus single-agent immunotherapy,
bust enough for patient selection. suggest that patients who are not eligible for treatment with
cisplatin and have PD-L1high disease benefit from single-
Finally, the potential value of circulating tumor DNA (ctDNA)
agent immunotherapy, and that single-agent immunother-
in monitoring the molecular evolution of metastatic disease
apy should be avoided in favor of systemic chemotherapy in
is being evaluated in RCC research projects. Early work
patients with PD-L1low tumors.51 A variety of other factors,
suggested that ctDNA may be more difficult to detect in RCC
including tumor mutational burden,50 bladder cancer
compared with other solid tumors, particularly in patients
subtyping by gene-expression profiling,47,48 and the in-
with earlier-stage disease.43,44 Commercial panels detect
terferon gamma gene signature,52 have been looked at as
ctDNA in about three-fourths of patients with overt meta-
a means of improving our ability to select patients who may
static disease45; rates of TP53 and mTOR pathway alter-
benefit the most from an immune checkpoint inhibitor.
ations are higher than those observed in localized tumors.
Despite these attempts, it still appears that the best way of
It may also be possible to detect RCC-related mutations
selecting patients who benefit from an immune checkpoint
in urine,46 although large-scale studies of feasibility must
inhibitor is the patient’s actual response to immune
still be performed. Layering the development of conve-
checkpoint inhibition. One landmark analysis of patients
nient noninvasive mutational identification and disease-
receiving second-line nivolumab for more than 1 year
monitoring methods over tissue-based markers will be
suggested that patients with a clinical CR do extraordinarily
important to maximize the utility and validity of new assays,
well, with only two deaths among 17 patients achieving
which will increase our understanding of the importance
a clinical CR.53 This also raises the question as to whether
and temporal variation of peripherally detected mutations
a depth of response matters. If so, one could envision
in RCC.
a future in which clinical trials aim for improvement in CR, in
addition to assessing a durability of response, as a potential
TABLE 2. Summary of Clinical Utility for Biomarkers in Metastatic surrogate for long-term clinical outcomes.
Renal Cell Carcinoma FGFR3-mutated UC recently emerged as a potential marker
Study Use in Level of
of resistance to immune checkpoint inhibition based upon
Biomarker Context Practice When to Test Evidence
its enrichment for an immunologically cold, luminal papillary
IMDC Risk Retrospective Yes Always Strong subtype of UC.54 Erdafitinib is the first-in-class pan–FGFR1-
Score
4 inhibitor approved for the treatment of metastatic UC; it is
PD-L1 Prospective Limited Critical disease; Weak highly effective in the 20% of patients with activating mu-
which ICB
combination
tations in FGFR.55,56 Early trials of FGFR3 inhibitors used an
to use intermittent schedule because of their toxicity.57,58 As a re-
PBRM1 Prospective No Not indicated N/A
sult, the erdafitinib trial initially assigned patients to an in-
termittent dose of 10 mg daily (1 week on/1 week off) versus
NGS Retrospective No Not indicated N/A
continuous dosing of 6 mg daily. Early results from this
Abbreviations: IMDC, International Metastatic Renal Cell Cancer randomization suggested that the continuous dose may
Database Consortium; ICB, immune checkpoint blockade; N/A, not have better effects on toxicity and clinical activity. Addi-
applicable; NGS, next-generation sequencing. tional pharmacokinetics testing performed during this trial

suggested an increase in continuous dosing to 8 mg daily, studies are in progress to confirm this finding, which could
with uptitration to 9 mg daily if the phosphorous level on day change the treatment paradigm for UC.
15 was less than 5.5 mg/dL. The use of targeted phosphorus
Many other targeted therapies show promise in the treat-
levels reflects on-target inhibition of FGFR3, providing ment of UC. Sacituzumab govitecan is an antibody drug
a means of increasing the dose in patients who may benefit conjugate targeting Trop-2, which is widely expressed in
from more optimal dosing to enhance response. With a 40% multiple tumor types, including UC; it brings the active
objective response rate and median overall survival of ap- metabolite of irinotecan directly to tumor cells. Bempe-
proximately 13.8 months,55 the U.S. Food and Drug Ad- galdesleukin is a pegylated formulation of IL-2 with selective
ministration granted erdafitinib accelerated approval for the signaling through the IL-2 beta gamma receptor, which
second-line treatment of FGFR3-altered metastatic UC in results in clonal expansion of lymphocytes associated with
April 2019. an immune response; it has shown evidence of clinical
However, we still do not have an answer to the question of activity, even in PD-L1low tumors, when combined with an
whether FGFR3-altered urothelial tumors derive greater immune checkpoint inhibitor.63 VEGF inhibitors, including
benefit from treatment with an immune checkpoint inhibitor lenvatinib and sitravatinib,64 which target AXL and MER in
or FGFR3-targeted therapy. In the phase II clinical trial of addition to VEGF, and many more targeted agents are being
erdafitinib, 22 patients treated had received prior immune studied for the treatment of UC in single-agent trials, as well
checkpoint inhibition, and one partial response was reas in combination with immune checkpoint inhibitors.
ported (ORR, 5%).55 Similar results were reported with The molecular subtyping of tumors into relevant classes
rogaratinib; one in 10 patients had stable disease, which is based on differential biology may have utility in guiding
the best response reported with immunotherapy.59 How- future clinical decisions. Using whole-transcriptome mes-
ever, Galsky et al reported a similar objective response rate senger RNA expression profiling and unsupervised hierar-
to an immune checkpoint inhibitor, regardless of whether an chical clustering, several groups independently concluded
FGFR3 mutation was present.60 The phase III trial THOR that muscle-invasive UC can be grouped into basal and
has the goal of answering this important question. Patients luminal molecular subtypes that resemble the ones that had
who have received prior chemotherapy, but no prior im- previously been identified in breast cancers.65-68 Recently,
mune checkpoint inhibitor, for their FGFR3-altered meta- an international team created an open-access single-
static UC will be randomly assigned to receive erdafitinib or sample molecular subtyping algorithm that can be used
pembrolizumab. Patients who have received a prior immune with whole-transcriptome data to assign UC to one of six
checkpoint inhibitor will be randomly assigned to receive consensus subtypes.69
erdafitinib or taxanes (or vinflunine in Europe). A second Basal cancers are enriched with regard to squamous his-
clinical trial (NORSE) is also accruing patients to test topathologicg features and stem cell biomarkers, have the
whether the addition of an immune checkpoint inhibitor to most cellular proliferation with high levels of cyclins, and are
erdafitinib will enhance response or durability of response associated with advanced-stage and metastatic disease
to combination therapy. However, because this mutation at presentation. 67,68 Cisplatin-based chemotherapy has
is present in up to 20% of patients, other therapies are remained the mainstay of treatment for UC for more than
needed. 30 years as a result of its ability to promote DNA damage
Enfortumab vedotin is another recent addition as the first resulting in cell death of the most rapidly proliferating cells.
antibody drug conjugate approved for the treatment of Early evidence from patients treated on two clinical trials of
metastatic UC. The antibody targets nectin-4, which is chemotherapy with methotrexate-vinblastine-doxorubicin-
expressed at a high level on 93% of urothelial tumors; cisplatin (MVAC) and dose-dense MVAC suggested that the
therefore, testing for nectin-4 is not required and should not basal subtype had the best clinical outcomes when treated
be pursued prior to starting therapy. This antibody uses with neoadjuvant cisplatin-based chemotherapy, a finding
a protease-cleavable linker to release the antimicrotubule that has been replicated in a separate retrospective data set
agent auristatin-E following internalization of the antibody. In of cisplatin-treated patients.67,70,71
patients who have received prior chemotherapy and im- Luminal cancers appear to be more heterogeneous and can
mune checkpoint inhibition, treatment with enfortumab be subdivided into as many as four subtypes.69 Luminal
vedotin resulted in a 44% objective response rate, with papillary tumors are enriched with regard to activating
a median overall survival of approximately 11 months.61 FGFR3 mutations and fusions, are associated with low rates
Early results from a small phase II clinical trial of enfortu- of pathologic upstaging,72 and are associated with the best
mab vedotin plus pembrolizumab in the frontline treatment prognoses.69 Although luminal unstable and luminal infil-
of UC in patients who could not receive cisplatin suggest trated tumors are more aggressive, they may also be more
objective response rates as high as 70%.62 Additional sensitive to immunotherapy.73,74 Finally, neuroendocrine

Schmidt et al
tumors are characterized by combined inactivation of TP53 Although pathologic downstaging in patients treated with
and RB1 and expression of neuronal differentiation markers.69 neoadjuvant therapies correlates well with survival, it is
Although they are very clinically aggressive, they may be only a surrogate for the direct effects of systemic therapy
highly sensitive to immunotherapy.75 Importantly, these on micrometastatic disease. Most of the companies that
relationships must be prospectively validated before mo- have panel sequencing products for tumor biopsies also
lecular subtyping can inform clinical decision making. One offer products for panel sequencing of ctDNA in plasma,
such prospective study examining the relationship between although the relatively modest sequencing depth gener-
molecular subtype and benefit from neoadjuvant chemo- ated by these platforms makes them unreliable as tools to
therapy has completed enrollment (Southwest Oncology measure minimal residual disease. However, collaborative
Group’s S1314 COXEN trial), and the results are currently efforts involving a university-based academic group and
being analyzed. Natera have produced a more personalized test that is an
attractive candidate solution for this purpose.81 Tumor
Predicting clinical benefit from neoadjuvant chemotherapy
biopsies are first subjected to whole-exome sequencing,
is also the purpose of a second class of biomarkers:
and droplet-digital polymerase chain reaction–sequencing
inactivating mutations in DNA damage and response
assays are designed to measure 16 patient-specific mu-
genes. Using the MSK-IMPACT panel exome-sequencing
tations. The assay is then applied to plasma samples that
platform and focusing on extreme responders, one group
are sequenced to an average of 109,000 coverage.
linked inactivating mutations in ERCC2 to response
Applying the assay to longitudinal plasma samples from
(i.e., pathologic downstaging) 76 and went on to perform
patients treated with neoadjuvant chemotherapy, the in-
functional studies that demonstrated that these mutations
vestigators demonstrated that it exhibited extremely high
actually cause sensitivity to cisplatin.76,77 A second group
sensitivity and specificity, was highly prognostic at sentinel
performed FoundationOne panel sequencing on tumor
points in clinical management (after transurethral re-
specimens from a completed trial of dose-dense MVAC and
section of bladder tumor/before neoadjuvant chemo-
identified mutations in RB1, FANCC, or ATM in all of the
therapy, after neoadjuvant chemotherapy, and before
responding tumors.78 They next performed validation studies
recurrence), and performed better than molecular sub-
with samples from an independent clinical trial of dose-dense
types or DNA damage and response mutations to predict
gemcitabine/cisplatin. Both groups plan to prospectively
response.81
validate their findings by performing panel exome se-
quencing on the tumors from the COXEN trial. The in- There is also an unmet need to develop accurate methods
vestigators have also opened two prospective clinical trials for measuring residual disease in the bladder. Although it
that are designed to test the hypothesis that patients whose is possible that plasma ctDNA can also be used for this
tumors contain DNA damage and response mutations can purpose, a potentially more attractive approach would be
be managed with a thorough transurethral resection of to develop assays to measure tumor DNA in urine.46,80
bladder tumor and neoadjuvant chemotherapy without Early work using mass spectrometric sequencing ap-
going on to cystectomy. For a summary of biomarkers re- proaches established feasibility,82 and more recent ef-
lated to metastatic UC, refer to Table 3. forts have involved applying panel or personalized NGS
approaches to increase sensitivity and specificity.46,79,80
A third area of major progress in UC is the development of
Although these assays are not ready for clinical imple-
“liquid biopsy” approaches using peripheral blood or urine
mentation, the performance characteristics of the best
to monitor minimal residual disease burden and select
ones rival blue light cystoscopy (without the need for an
patients for conventional and targeted therapy. 46,79,80
experienced urologist).46,80 Ongoing work is aimed at
determining whether these urine tumor DNA assays can be
TABLE 3. Summary of Clinical Utility for Biomarkers in Metastatic used to monitor local residual disease burden in patients
Urothelial Carcinoma treated with intravesical therapies and/or neoadjuvant
Study Use in Level of chemotherapy.
Biomarker Context Practice When to Test Evidence
FGFR Prospective Yes Second-line Strong CONCLUSIONS
alterations treatment
PD-L1 Prospective Sometimes Cisplatin- Strong
Biomarker development in RCC and UC has lagged behind
ineligible first- the recent advances in systemic treatment. Although
line setting VEGF-targeted therapy has been a mainstay in RCC for
Gene- Retrospective No Not indicated N/A more than a decade, it has been used without guidance
expression
from any biomarker, and emerging antibody drug conju-
profiling
gates in UC look to do the same. Prognostic markers
Abbreviation: N/A, not applicable. continue to evolve; however, other than FGFR alterations,

there is no predictive biomarker to guide treatment de- detection methods may define the future landscape for
cisions in UC or RCC. Molecular subtyping of disease biomarker-directed therapy. To achieve this, successful
has generated a greater understanding of the landscape prospective validation of treatment assignment based on
of alterations and their potential impact on treatment re- mutation or disease subtyping will be required before
sponse; further improvements in these sequencing and routine clinical utilization.

1
Dana-Farber Cancer Institute, Boston, MA AND DATA AVAILABILITY STATEMENT
2
University of Texas MD Anderson Cancer Center, Houston, TX Disclosures provided by the authors and data availability statement (if
3
Johns Hopkins Greenberg Bladder Cancer Institute, Baltimore, MD applicable) are available with this article at DOI https://doi.org/10.1200/
EDBK_279905.
Bradley McGregor, Dana-Farber Cancer Institute, 450 Brookline Ave.,
Boston, MA 02215; Twitter: @Bradmcg04; email: bradley_mcgregor@
dfci.harvard.edu.
REFERENCES
1. Heng DY, Xie W, Regan MM, et al. External validation and comparison with other models of the International Metastatic Renal-Cell Carcinoma Database
Consortium prognostic model: a population-based study. Lancet Oncol. 2013;14:141-148.
3. Viers BR, Houston Thompson R, Boorjian SA, et al. Preoperative neutrophil-lymphocyte ratio predicts death among patients with localized clear cell renal
carcinoma undergoing nephrectomy. Urol Oncol. 2014;32:1277-1284.
4. Peyton CC, Abel EJ, Chipollini J, et al. The value of neutrophil to lymphocyte ratio in patients undergoing cytoreductive nephrectomy with thrombectomy. Eur Urol
Focus. 2020;6:104-111.
5. Templeton AJ, Knox JJ, Lin X, et al. Change in neutrophil-to-lymphocyte ratio in response to targeted therapy for metastatic renal cell carcinoma as a prog-
nosticator and biomarker of efficacy. Eur Urol. 2016;70:358-364.
6. Tannir NM, McDermott DF, Escudier B, et al. Overall survival and independent review of response in CheckMate 214 with 42-month follow-up: first-line
nivolumab + ipilimumab (N+I) versus sunitinib (S) in patients (pts) with advanced renal cell carcinoma (aRCC). J Clin Oncol. 2020;38:15s (suppl; abstr 609).
2019;380:1116-1127.
9. Labriola M, Zhu J, Gupta R, et al. Characterization of tumor mutational burden (TMB), PD-L1, and DNA repair genes to assess correlation with immune
checkpoint inhibitors (ICIs) response in metastatic renal cell carcinoma (mRCC). J Clin Oncol. 2019;37:15s (suppl; abstr e16079).
10. Motzer RJ, Tannir NM, McDermott DF, et al. Nivolumab plus ipilimumab versus sunitinib in advanced renal-cell carcinoma. N Engl J Med. 2018;378:1277-1290.
11. Bakouny Z, Vokes N, Gao X, et al. Efficacy of immune checkpoint inhibitors (ICI) and genomic characterization of sarcomatoid and/or rhabdoid (S/R) metastatic
renal cell carcinoma (mRCC). J Clin Oncol. 2019;37:15s (suppl; abstr 4514).
12. Rini BI, Plinick ER, Stus V, et al. Pembrolizumab (pembro) plus axitinib (axi) versus sunitinib as first-line therapy for metastatic renal cell carcinoma (mRCC):
outcomes in the combined IMDC intermediate/poor risk and sarcomatoid subgroups of the phase 3 KEYNOTE-426 study. J Clin Oncol. 2019;37:15s (suppl; abstr
4500).
13. McDermott DF, Choueiri TK, Motzer RJ, et al. CheckMate 214 post-hoc analyses of nivolumab plus ipilimumab or sunitinib in IMDC intermediate/poor-risk
patients with previously untreated advanced renal cell carcinoma with sarcomatoid features. J Clin Oncol. 2019;37:15s (suppl; abstr 4513).
14. Rini BI, Motzer RJ, Powles T, et al. Atezolizumab (atezo) + bevacizumab (bev) versus sunitinib (sun) in pts with untreated metastatic renal cell carcinoma (mRCC)
and sarcomatoid (sarc) histology: IMmotion151 subgroup analysis. J Clin Oncol. 2019;37:15s (suppl; abstr 4512).
15. Voss MH. Clinically Useful Biomarkers: Ready for Prime Time in 2020? Presented at: Genitourinary Cancers Symposium; February 15, 2020; San Francisco, CA.
16. Linehan WM, Ricketts CJ. The Cancer Genome Atlas of renal cell carcinoma: findings and clinical implications. Nat Rev Urol. 2019;16:539-552.
17. Ricketts CJ, De Cubas AA, Fan H, et al. The Cancer Genome Atlas comprehensive molecular characterization of renal cell carcinoma. Cell Rep. 2018;
23:313-326.e5.
18. Cancer Genome Atlas Research Network. Comprehensive molecular characterization of clear cell renal cell carcinoma. Nature. 2013;499:43-49.
19. Vano Y-A, Rioux-Leclerq N, Dalban C, et al. NIVOREN GETUG-AFU 26 translational study: association of PD-1, AXL, and PBRM-1 with outcomes in patients (pts)
with metastatic clear cell renal cell carcinoma (mccRCC) treated with nivolumab (N). J Clin Oncol. 2020;38:15s (suppl; abstr 618).

Schmidt et al
20. Miao D, Margolis CA, Gao W, et al. Genomic correlates of response to immune checkpoint therapies in clear cell renal cell carcinoma. Science. 2018;
359:801-806.
21. Abou Alaiwi S, Nassar A, El Bakouny Z, et al. Association of polybromo-associated BAF (PBAF) complex mutations with overall survival (OS) in cancer patients
(pts) treated with checkpoint inhibitors (ICIs). J Clin Oncol. 2019;37:15s (suppl; abstr 103).
22. McDermott DF, Huseni MA, Atkins MB, et al. Clinical activity and molecular correlates of response to atezolizumab alone or in combination with bevacizumab
versus sunitinib in renal cell carcinoma. Nat Med. 2018;24:749-757.
23. Courtney KD, Infante JR, Lam ET, et al. Phase I dose-escalation trial of PT2385, a first-in-class hypoxia-inducible factor-2alpha antagonist in patients with
previously treated advanced clear cell renal cell carcinoma. J Clin Oncol. 2018;867-874.
24. Chen W, Hill H, Christie A, et al. Targeting renal cell carcinoma with a HIF-2 antagonist. Nature. 2016;539:112-117.
25. Attalla K, DiNatale RG, Reznik E, et al. Prevalence and landscape of actionable genomic alterations in renal cell carcinoma. J Clin Oncol. 2020;38:15s (suppl;
abstr 616).
26. Zehir A, Benayed R, Shah RH, et al. Mutational landscape of metastatic cancer revealed from prospective clinical sequencing of 10,000 patients. Nat Med. 2017;
23:703-713.
27. Kwiatkowski DJ, Choueiri TK, Fay AP, et al. Mutations in TSC1, TSC2, and MTOR are associated with response to rapalogs in patients with metastatic renal cell
carcinoma. Clin Cancer Res. 2016;22:2445-2452.
28. Nassar AH, Hamieh L, Gray KP, et al. Mutations and response to rapalogs in patients with metastatic renal cell carcinoma. Mol Cancer Ther. 2020;19:690-696.
29. Gibney GT, Aziz SA, Camp RL, et al. c-Met is a prognostic marker and potential therapeutic target in clear cell renal cell carcinoma. Ann Oncol. 2013;24:343-349.
30. Kroeger N, Xie W, Lee JL, et al. Metastatic non-clear cell renal cell carcinoma treated with targeted therapy agents: characterization of survival outcome and
application of the International mRCC Database Consortium criteria. Cancer. 2013;119:2999-3006.
31. Linehan WM, Srinivasan R, Garcia JA. Non-clear cell renal cancer: disease-based management and opportunities for targeted therapeutic approaches. Semin
Oncol. 2013;40:511-520.
32. Davis CF, Ricketts CJ, Wang M, et al. The somatic genomic landscape of chromophobe renal cell carcinoma. Cancer Cell. 2014;26:319-330.
33. Linehan WM, Spellman PT, Ricketts CJ, et al; Cancer Genome Atlas Research Network. Comprehensive molecular characterization of papillary renal-cell
carcinoma. N Engl J Med. 2016;374:135-145.
34. Albiges L, Guegan J, Le Formal A, et al. MET is a potential target across all papillary renal cell carcinomas: result from a large molecular study of pRCC with CGH
array and matching gene expression array. Clin Cancer Res. 2014;20:3411-3421.
35. Choueiri TK, Vaishampayan U, Rosenberg JE, et al. Phase II and biomarker study of the dual MET/VEGFR2 inhibitor foretinib in patients with papillary renal cell
carcinoma. J Clin Oncol. 2013;31:181-186.
36. Martinez Chanzá N, Xie W, Asim Bilen M, et al. Cabozantinib in advanced non-clear-cell renal cell carcinoma: a multicentre, retrospective, cohort study. Lancet
Oncol. 2019;20:581-590.
37. NCT02761057. Cabozantinib S-Malate, Crizotinib, Savolitinib, or Sunitinib Malate in Treating Patients With Locally Advanced or Metastatic Kidney Cancer.
https://clinicaltrials.gov/ct2/show/NCT02761057. Accessed March 3, 2020.
38. Choueiri TK, Plimack E, Arkenau HT, et al. Biomarker-based phase II trial of savolitinib in patients with advanced papillary renal cell cancer. J Clin Oncol. 2017;
35:2993-3001.
39. Rodriguez CS, Larkin JMG, Patel P, et al. Overall survival results for durvalumab and savolitinib in metastatic papillary renal cancer. J Clin Oncol. 2020;38:15s
(suppl; abstr 619).
40. NCT03091192. Savolitinib vs. Sunitinib in MET-Driven PRCC. https://clinicaltrials.gov/ct2/show/NCT03091192. Accessed March 3, 2020.
41. Flippot RMB, Flaifel A, Gray K, et al. Atezolizumab plus bevacizumab in non-clear cell renal cell carcinoma (NccRCC) and clear cell renal cell carcinoma with
sarcomatoid differentiation (ccRCCsd): Updated results of activity and predictive biomarkers from a phase II study. J Clin Oncol. 2019;37:15s (suppl; abstr 4583).
42. Rini BI, Huseni M, Atkins MB, et al. Molecular correlates differentiate response to atezolizumab (atezo) + bevacizumab (bev) vs sunitinib (sun): results from
a phase III study (IMmotion151) in untreated metastatic renal cell carcinoma (mRCC). Ann Oncol. 2018;29 (suppl 8):viii724-viii725.
44. Phallen J, Sausen M, Adleff V, et al. Direct detection of early-stage cancers using circulating tumor DNA. Sci Transl Med. 2017;9:eaan2415.
45. Pal SK, Sonpavde G, Agarwal N, et al. Evolution of circulating tumor DNA profile from first-line to subsequent therapy in metastatic renal cell carcinoma. Eur Urol.
2017;72:557-564.
46. Springer SU, Chen CH, Rodriguez Pena MDC, et al. Non-invasive detection of urothelial cancer through the analysis of driver gene mutations and aneuploidy.
Elife. 2018;7:e32143.
47. Sharma P, Retz M, Siefker-Radtke A, et al. Nivolumab in metastatic urothelial carcinoma after platinum therapy (CheckMate 275): a multicentre, single-arm,
phase 2 trial. Lancet Oncol. 2017;18:312-322.
49. Balar AV, Castellano D, O’Donnell PH, et al. First-line pembrolizumab in cisplatin-ineligible patients with locally advanced and unresectable or metastatic
urothelial cancer (KEYNOTE-052): a multicentre, single-arm, phase 2 study. Lancet Oncol. 2017;18:1483-1492.

50. Balar AV, Galsky MD, Rosenberg JE, et al; IMvigor210 Study Group. Atezolizumab as first-line treatment in cisplatin-ineligible patients with locally advanced and
metastatic urothelial carcinoma: a single-arm, multicentre, phase 2 trial. Lancet. 2017;389:67-76.
51. Grande E, Galsky M, Arija J, et al. IMvigor130: Efficacy and safety from a phase 3 study of atezolizumab (atezo) as monotherapy or combined with platinum-based
chemotherapy (PBC) vs placebo + PBC in previously untreated locally advanced or metastatic urothelial carcinoma (MUC). Ann Oncol. 2019;30(suppl_5): v851-
v934.
52. Gao J, Shi LZ, Zhao H, et al. Loss of IFN-gamma pathway genes in tumor cells as a mechanism of resistance to anti-CTLA-4 therapy. Cell. 2016;167:397-404.e9.
53. Siefker-Radtke A, Baron A, Necchi A, et al. Nivolumab monotherapy in patients with advanced platinum-resistant urothelial carcinoma: efficacy and safety
update from CheckMate 275. J Clin Oncol. 2019;37:15s (suppl; abstr 4524).
54. McConkey DJ, Choi W, Ochoa A, et al. Therapeutic opportunities in the intrinsic subtypes of muscle-invasive bladder cancer. Hematol Oncol Clin North Am. 2015;
29:377-394, x-xi.
55. Loriot Y, Necchi A, Park SH, et al; BLC2001 Study Group. Erdafitinib in locally advanced or metastatic urothelial carcinoma. N Engl J Med. 2019;381:338-348.
56. Robertson AG, Kim J, Al-Ahmadie H, et al. Comprehensive molecular characterization of muscle-invasive bladder cancer. Cell. 2017;171:540-556.e25.
57. Milowsky MI, Dittrich C, Durán I, et al. Phase 2 trial of dovitinib in patients with progressive FGFR3-mutated or FGFR3 wild-type advanced urothelial carcinoma.
Eur J Cancer. 2014;50:3145-3152.
58. Pal SK, Rosenberg JE, Hoffman-Censits JH, et al. Efficacy of BGJ398, a fibroblast growth factor receptor 1-3 inhibitor, in patients with previously treated advanced
urothelial carcinoma with FGFR3 alterations. Cancer Discov. 2018;8:812-821.
59. Joerger M, Cassier, P, Penel N, et al. Rogaratinib treatment of patients with advanced urothelial carcinomas prescreened for tumor FGFR mRNA expression. J
Clin Oncol. 2018;36:15s (suppl; abstr 494).
60. Wang L, Gong Y, Saci A, et al. Fibroblast growth factor receptor 3 alterations and response to PD-1/PD-L1 blockade in patients with metastatic urothelial cancer.
Eur Urol. 2019;76:599-603.
61. Rosenberg JE, O’Donnell PH, Balar AV, et al. Pivotal trial of enfortumab vedotin in urothelial carcinoma after platinum and anti-programmed death 1/programmed
death ligand 1 therapy. J Clin Oncol. 2019;37:2592-2600.
62. Hoimes CJ, Rosenberg JE, Srinivas S, et al. EV-103: initial results of enfortumab vedotin plus pembrolizumab for locally advanced or metastatic urothelial
carcinoma. Ann Oncol. 2019;30(suppl 5):v356-v402.
63. Siefker-Radtke AO, Fishman MN, Balar AK, et al. NKTR-214 + nivolumab in first-line advanced/metastatic urothelial carcinoma (mUC): Updated results from
PIBOT-02. J Clin Oncol. 2019;37:15s (suppl; abstr 388).
64. Msaouel P, Siefker-Radtke A, Sweis R, et al. Sitravatinib in combination with nivolumab demonstrates clinical activity in platinum-experienced patients with
urothelial carcinoma (UC) who progressed on prior immune checkpoint inhibitor (CPI). Presented at: SITC; National Harbor, MD. Abstract O23.
65. Sjödahl G, Lauss M, Lövgren K, et al. A molecular taxonomy for urothelial carcinoma. Clin Cancer Res. 2012;18:3377-3386.
66. Damrauer JS, Hoadley KA, Chism DD, et al. Intrinsic subtypes of high-grade bladder cancer reflect the hallmarks of breast cancer biology. Proc Natl Acad Sci
USA. 2014;111:3110-3115.
67. Choi W, Porten S, Kim S, et al. Identification of distinct basal and luminal subtypes of muscle-invasive bladder cancer with different sensitivities to frontline
chemotherapy. Cancer Cell. 2014;25:152-165.
68. Cancer Genome Atlas Research Network. Comprehensive molecular characterization of urothelial bladder carcinoma. Nature. 2014;507:315-322.
69. Kamoun A, de Reyniès A, Allory Y, et al; Bladder Cancer Molecular Taxonomy Group. A consensus molecular classification of muscle-invasive bladder cancer.
Eur Urol. 2020;77:420-433.
70. Seiler R, Ashab HAD, Erho N, et al. Impact of molecular subtypes in muscle-invasive bladder cancer on predicting response and survival after neoadjuvant
chemotherapy. Eur Urol. 2017;72:544-554.
71. McConkey DJ, Choi W, Shen Y, et al. A prognostic gene expression signature in the molecular classification of chemotherapy-naive urothelial cancer is predictive
of clinical outcomes from neoadjuvant chemotherapy: a phase 2 trial of dose-dense methotrexate, vinblastine, doxorubicin, and cisplatin with bevacizumab in
urothelial cancer. Eur Urol. 2016;69:855-862.
72. Lotan Y, Boorjian SA, Zhang J, et al. Molecular subtyping of clinically localized urothelial carcinoma reveals lower rates of pathological upstaging at radical
cystectomy among luminal tumors. Eur Urol. 2019;76:200-206.
73. Mariathasan S, Turley SJ, Nickles D, et al. TGFbeta attenuates tumour response to PD-L1 blockade by contributing to exclusion of T cells. Nature. 2018;
554:544-548.
75. Kim J, Kwiatkowski D, McConkey DJ, et al. The Cancer Genome Atlas expression subtypes stratify response to checkpoint inhibition in advanced urothelial cancer
and identify a subset of patients with high survival probability. Eur Urol. 2019;75:961-964.
76. Van Allen EM, Mouw KW, Kim P, et al. Somatic ERCC2 mutations correlate with cisplatin sensitivity in muscle-invasive urothelial carcinoma. Cancer Discov. 2014;
4:1140-1153.
77. Li Q, Damish AW, Frazier Z, et al. ERCC2 helicase domain mutations confer nucleotide excision repair deficiency and drive cisplatin sensitivity in muscle-invasive
bladder cancer. Clin Cancer Res. 2019;25:977-988.

Schmidt et al
78. Plimack ER, Dunbrack RL, Brennan TA, et al. Defects in DNA repair genes predict response to neoadjuvant cisplatin-based chemotherapy in muscle-invasive
bladder cancer. Eur Urol. 2015;68:959-967.
79. Birkenkamp-Demtröder K, Nordentoft I, Christensen E, et al. Genomic alterations in liquid biopsies from patients with bladder cancer. Eur Urol. 2016;70:75-82.
80. Dudley JC, Schroers-Martin J, Lazzareschi DV, et al. Detection and surveillance of bladder cancer using urine tumor DNA. Cancer Discov. 2019;9:500-509.
81. Christensen E, Birkenkamp-Demtröder K, Sethi H, et al. Early detection of metastatic relapse and monitoring of therapeutic efficacy by ultra-deep sequencing of
plasma cell-free DNA in patients with urothelial bladder carcinoma. J Clin Oncol. 2019;37:1547-1557.
82. Critelli R, Fasanelli F, Oderda M, et al. Detection of multiple mutations in urinary exfoliated cells from male bladder cancer patients at diagnosis and during follow-
up. Oncotarget. 2016;7:67435-67448.

GENITOURINARY
CANCER—PROSTATE,
TESTICULAR, AND PENILE

Sorting Through the Maze of Treatment Options

for Metastatic Castration-Sensitive
Prostate Cancer
Brian Schulte, MD1; Alicia K. Morgans, MD, MPH1; Neal D. Shore, MD2; and Carmel Pezaro, BHB, MBChB, FRACP, DMedSc, MHPE3
overview
Since 1944, when Huggins and Hodges demonstrated the effectiveness of bilateral orchiectomy for meta-
static prostate cancer (PCa), androgen deprivation therapy (ADT) has been the first-line treatment for men with
advanced PCa. The proportion of PCa cases that are metastatic at diagnosis ranges globally, from 5%–20% in
countries with widespread screening practices to upward of 30%–60% where screening is minimal. In the
United States alone, there will be an estimated 191,000 new cases of PCa diagnosed in the year 2020, of
which approximately 20% will be metastatic.1 Ongoing controversy around prostate-specific antigen (PSA)
screening practices, increased access to novel imaging modalities, and a globally aging population will drive
increased rates of metastatic castration-sensitive prostate cancer (mCSPC).2,3 At the same time, advances in
upfront hormonal or chemohormonal therapy have driven a dramatic shift in treatment paradigms. In this
article, we review recent advances in treatment choices for men with newly diagnosed mCSPC and the impact
of upfront treatment on subsequent disease biology. Options include treatment with chemohormonal therapy,
androgen receptor (AR)–directed therapy in addition to ADT, or, less commonly, ADT alone. Treatment choice
must include consideration of clinical and disease characteristics, as well as patient preferences and lim-
itations of geography and financial concerns.
WHICH PATIENTS WITH mCSPC SHOULD RECEIVE was a nonsignificant trend toward docetaxel benefit
INITIAL CHEMOHORMONAL THERAPY? (median OS, 46.5 vs. 60.9 months; HR, 0.9; 95% CI,
Docetaxel has been used in advanced PCa for more 0.7–1.2). When the CHAARTED definition of tumor
than 15 years, where it has been offered as a palliative volume was retrospectively applied, docetaxel benefit
treatment for men with disease progression despite seemed greater in the high-volume subset (48% of the
castrate levels of testosterone (castration-resistant PCa overall trial population) but again failed to achieve
[CRPC]). The early use of six cycles every 3 weeks of statistical significance.
docetaxel for men starting ADT for mCSPC is also now In the CHAARTED trial, 790 men with mCSPC were
a standard of care. However, the best use of, and randomly selected to receive treatment with ADT
optimal patient selection for, chemohormonal therapy alone or with six cycles of docetaxel, delivered without
remains unclear, because of subtle differences in the continuous corticosteroids.6 Docetaxel was started at
key clinical trials and gaps in our understanding of the a median of 5 weeks after ADT, although an interval of
biology underpinning the benefits observed. up to 4 months was allowed. The primary endpoint was
The GETUG-AFU 15 trial was the first randomized OS, with prospective stratification of disease volume,
Author affiliations phase III trial to report the use of early docetaxel in aiming to identify patients with poor prognosis.7 High-
and support addition to ADT. Three hundred eighty-five men with volume disease was defined by the presence of visceral
information (if mCSPC were randomly selected to receive treatment metastases and/or at least four bone metastases with at
applicable) appear least one outside the spine/pelvis, representing two-
with standard ADT versus ADT plus a maximum of nine
at the end of this
article. cycles of docetaxel, without concomitant prednisone/ thirds of the trial population. Updated survival for the
Accepted on prednisolone.4 After a median follow-up of 50 months, complete trial population after a median follow-up of
February 23, 2020 the trial was negative for the primary endpoint of overall 53.7 months showed a median OS of 47.2 months with
and published at survival (OS), with median OS being 54.2 months in the ADT and 57.6 months with ADT/docetaxel (HR, 0.72;
ascopubs.org on ADT cohort and 58.9 months in the ADT/docetaxel 95% CI, 0.59–0.89; p , .001).8 As in the initial report,
March 17, 2020:
cohort (hazard ratio [HR], 1.01; 95% CI, 0.75–1.36). chemotherapy benefit appeared confined to men with
10.1200/EDBK_ Updated survival was presented in abstract form in high-volume mCSPC, although it should be noted that
278845 2015, with a median follow-up of 82.9 months.5 There the trial was not powered for the preplanned volume-

Treatment Choice for Metastatic Castration-Sensitive Prostate Cancer
enrollment, as did 83% of patients in the high-volume

cohort of CHAARTED. However, the low-volume cohort of
CHAARTED and GETUG 15 had a larger population of men
• Treatment decisions for first-line mCSPC who had prior local therapy to the prostate and also had
require consideration of all available options,
longer OS with ADT alone.9 Although these variations offer
including AR-directed therapies and chemo-
a compelling narrative, this hypothesis has not been for-
therapy, clinical considerations, and patient
preferences to match the right treatment with mally tested. Although prior treatment of localized PCa may
an individual patient. be considered when recommending escalated treatment
with chemohormonal therapy, it is not known what weight it
• Chemohormonal therapy with docetaxel in
should carry in decision-making.
addition to ADT appears effective in men with
high- or low-volume mCSPC but may be most The addition of docetaxel to ADT is not without risk. In
effective in patients with de novo mCSPC rather mCRPC, men treated outside of the controlled environment
than recurrent disease after local treatment. of clinical trials experienced greater toxicity and shorter
• Androgen receptor–directed therapy is effective survival compared with the TAX 327 study that initially
in men with high- or low-risk mCSPC, and showed survival benefit.12 In GETUG 15, four treatment-
options include abiraterone acetate, apaluta- related deaths led the data monitoring committee to rec-
mide, and enzalutamide. ommend the use of granulocyte colony-stimulating factor to
• When choosing a first-line treatment for prevent neutropenia.4 In STAMPEDE, 14% of men treated
mCRPC, clinicians should consider treatment with docetaxel experienced febrile neutropenia, and five
used in mCSPC and should choose treatment men died of neutropenic sepsis, with higher rates of severe
that uses a different mechanism of action if toxicity in the ADT/docetaxel/zoledronic acid arm.13 In
initial treatment of mCSPC included an AR- CHAARTED, approximately 6% of the docetaxel cohort
directed therapy. experienced febrile neutropenia, and there was one
treatment-related death.6 The use of granulocyte colony-
stimulating factor as primary prophylaxis varies between
based analysis. A combined analysis using CHAARTED and clinicians but is recommended in men at high risk for
GETUG 15 data again suggested heterogeneity of response neutropenia.14 Although most men are able to complete
based on tumor volume.9 treatment, patients who experience nonlethal toxicity may
STAMPEDE is an ongoing multiarm, multistage United still have morbidity. In STAMPEDE, contemporaneous data
Kingdom–led trial, assessing treatments for men with newly of upfront docetaxel and abiraterone cohorts showed poorer
diagnosed locally advanced cancer or mCSPC. The initial quality of life in men receiving docetaxel, an effect that
docetaxel comparison included men treated with ADT, with persisted for almost 12 months from randomization.15 In
or without six cycles of docetaxel and continuous pred- situations where there are multiple choices for escalated
nisolone, and also incorporated randomized allocation to upfront treatment, it is understandable that men may
zolendronic acid.10 Outcomes were compared for each choose to defer chemotherapy toxicity until they have
treatment group against the control arm of ADT alone. Of the mCRPC.
2,962 men included, 61% had metastatic cancer, and Although the oncology community has been consumed by
docetaxel was started at a median of 9 weeks after ADT. discussions about the predictive impact of tumor volume for
Benefit was observed with docetaxel use, with a median OS chemohormonal treatment, the marker that we actually
of 71 months in the ADT arm and 81 months in the ADT/ desire is tumor biology. Patients with rapidly progressive
docetaxel arm (HR for docetaxel, 0.78; 95% CI, 0.66–0.93; PCa are most likely to benefit from effective escalated
p = .005). Updated survival was published after a median upfront treatment (and enhanced monitoring). Some pa-
follow-up of 78.2 months. For the 1,086 men with mCSPC, tients appear to progress rapidly to mCRPC even despite
the median OS was 43.1 months with ADT versus 59.1 docetaxel; in both CHAARTED and STAMPEDE, approxi-
months with ADT/docetaxel (HR, 0.81; 95% CI, 0.69– mately 10% of men had progression within 9 months of
0.95; p = .009).11 Retrospective assignment of tumor vol- starting treatment. It seems unlikely that a measurement of
ume using the CHAARTED definition failed to identify volume at a static time point, using conventional imaging
heterogeneity of effect by disease volume or location. It has techniques and any of the existing criteria, will remain the
been hypothesized that the discordance in volume-based optimal means of judging cancer behavior. It is hoped that
effect may relate to differences in the proportion of men in translational studies may uncover the population most likely
each study with de novo mCSPC versus recurrent mCSPC to derive treatment benefit, and indeed, reveal the mech-
after prior local therapy. In STAMPEDE, 95% of patients with anism by which docetaxel is contributing, perhaps by tar-
mCSPC had de novo metastatic disease at the time of geting non-AR clones and preventing resistance or by

Schulte et al
targeting AR-driven cells made vulnerable by the loss of ADT, there is a paucity of direct comparator trials, and there
ligand. are no established predictive biomarkers to guide treatment
In men with mCRPC receiving docetaxel, a number of re- selection. Clinicians should consider tumor burden spe-
sistance mechanisms have been described.16 Translational cifics, patient comorbidities and concomitant medications,
studies have also identified circulating cytokine and epi- drug accessibility, and patient preferences in selecting
genetic changes associated with poorer response and a combination regimen.
shorter survival during docetaxel treatment.17,18 It is yet to Several phase III trials have shown the benefit of upfront
be proven whether these factors will also contribute to re- ARB treatments. Abiraterone was studied in men with high-
sistance in the mCSPC setting, but it is plausible that dif- risk mCSPC in the LATITUDE trial, as well as the broader
ferent mechanisms may be responsible for early innate STAMPEDE population.13,27 The LATITUDE study included
resistance versus later progression. only men who met two of three predefined criteria, in-
We understand very little about patient and tumor hetero- dicating high-risk disease: Gleason score 8; presence
geneity as it impacts cancer in men with mCSPC. PSA ki- of at least three lesions on bone scan; and presence of
netics may be useful prognostic markers for men on ADT, measurable visceral lesions. Both trials used a lower dose of
with longer survival associated with more sustained (time to prednisone/prednisolone than had been used in the CRPC
nadir . 6 months) and complete PSA response (PSA na- trials, aiming to lessen the impact of steroid-related toxicity.
dir , 0.2).19 Baseline metabolic dysregulation and can- In LATITUDE, OS was significantly longer in the abiraterone
cer subtyping may also impact the speed of emergence of group than the placebo group (median OS not reached vs.
mCRPC.20-22 Circulating biomarkers are set to be a domi- 34.7 months; HR, 0.62; 95% CI, 0.51–0.76; p , .001).27
nant research focus over the coming years, including STAMPEDE also demonstrated reduced risk of death in the
a correlative study from the CHAARTED trial presented at abiraterone arm compared with ADT alone (HR, 0.63; 95%
the ASCO Genitourinary Cancers Symposium in February CI, 0.52–0.76).13 Benefit, however, was not confined only to
2020, showing an association between the luminal B those with metastatic disease, or high-risk features, because
subtype and both poorer OS with ADT alone and benefit STAMPEDE also incorporated patients with nonmetastatic
from addition of docetaxel.23 It is not yet known whether PCa. Furthermore, secondary outcomes such as skeletal
baseline risk features or PSA kinetics can be modified by the events (measured in STAMPEDE and LATITUDE) and pain
addition of upfront treatment such as docetaxel. At present, progression (LATITUDE) were significantly improved over
there is no set of agreed-on predictive factors to encourage controls.13,28
choice of upfront treatment; clinicians must embrace the The ENZAMET and ARCHES trials, presented in 2019,
uncertainty alongside their patients. showed the utility of enzalutamide in first-line treatment of
Based on two large phase III trials with a positive primary mCSPC.29,32 As these studies were conducted after pre-
endpoint of improved OS, the expert panel at the Advanced sentation of CHAARTED and STAMPEDE, docetaxel was
Prostate Cancer Consensus Conference concluded that received by a minority of patients in both. ENZAMET was
upfront docetaxel is a standard of care for men with mCSPC, unique among the ARB trials, in that the control arm re-
started within 4 months of ADT, with or without continuous ceived nonsteroidal antiandrogens in conjunction with ADT
steroid treatment.24 With more understanding of the un- rather than single-agent ADT.29 ENZAMET achieved its
derlying biology, we may be able to make rationale selection primary endpoint, with an HR for all-cause mortality of 0.67
of patients, revisit the optimal timing of docetaxel treatment, (95% CI, 0.52–0.86).29 The ARCHES trial used a combined
and perhaps even look to de-escalate treatment of some primary endpoint of radiographic progression-free sur-
patients, and further intensify treatment of others, using vival (PFS) or death and achieved an HR of 0.39 (95% CI,
additional systemic or tumor-focused treatments. 0.30–0.50; p , .001), favoring enzalutamide.30 The primary
WHICH PATIENTS SHOULD RECEIVE INITIAL AR-DIRECTED outcomes were preserved across all prespecified groups,
INTENSIFICATION FOR mCSPC? including those individuals who had previously received
docetaxel therapy.
There is widespread agreement that testosterone sup-
pression with ADT alone is no longer the standard of care for Apalutamide was tested in the TITAN trial, using dual-
most men with mCSPC.25,26 ADT has been combined with primary endpoints of radiographic PFS and OS.31 Nota-
a number of upfront therapies, including docetaxel, abir- bly, the TITAN trial was halted at first interim analysis, after
aterone acetate (with concomitant corticosteroid therapy), meeting the preset efficacy criteria. The HR for radiographic
enzalutamide, and apalutamide. Table 1 details the treat- PFS was 0.48 (95% CI, 0.39–0.60; p , .001) and for OS
ment effects that led to U.S. Food and Drug Administration was 0.67 (95% CI, 0.51–0.89; p = .005). Further data
approval of these agents for mCSPC. When selecting presented at the ASCO Genitourinary Cancers Symposium
a specific AR blockade (ARB) agent to be combined with in 2020 confirmed that upfront apalutamide treatment

TABLE 1. U.S. Food and Drug Administration–Approved Agents for the Treatment of mCSPC
Trial, Phase; Selected Grade 3-4
Publication Study Adverse Events
Treatment Year Population Comparator Size Outcome Treatment vs. Control (Treatment vs. Control)
Abiraterone LATITUDE, mCSPC ADT + placebo III; 1,199 OS 53.3 vs. 36.5 mos, AEs from LATITUDE:
acetate with 201727,28 (HR, 0.66; 95% CI, hypertension (20%
prednisone 0.56–0.78; p , vs. 10%);
.0001) hypokalemia (10% vs.
1%);
STAMPEDE, mCSPC and locally ADT alone III; 1,917 OS Estimated 83% vs.
elevated ALT (5% vs.
201713 advanced PCa 73% alive at 3 yrs
1%);
(HR, 0.63; 95% CI,
elevated AST (4% vs.
0.52–0.76; p ,
1%);
.001)
cardiac disorder (4%
vs. 1%)
Enzalutamide ENZAMET, mCSPC ADT+ III; 1,125 OS Estimated 80% vs. AEs from ENZAMET:
201929 nonsteroidal 72% alive at 3 yrs hypertension (8% vs.
ART (HR, 0.67; 95% CI, 4%);
0.52–0.86; p = .002) fatigue (6% vs. 1%);
falls (1% vs. , 1%);
ARCHES, mCSPC—stratified by ADT + placebo III; 1,150 rPFS or NR vs. 19 mos (HR,
seizures (, 1% vs. 0%);
201930 CHAARTED criteria death 0.39; 95% CI, 0.3–0.5;
rash (6.3% vs. 0.6%);
p , .001)
fractures (1.3% vs.
Apalutamide TITAN, 201931 mCSPC ADT + placebo III; 1,052 rPFS or 68.2% vs. 47.5% at 24 0.8%);
death mos (HR, 0.48; 95% hypothyroidism, all
CI, 0.39–0.60; p , grades (6.5% vs.
.001) 1.1%);
OS 82.4% vs. 73.5% alive seizures, all grades
at 24 mos (HR, 0.67; (0.6% vs. 0.4%)
95% CI, 0.51–0.89;
p = .005)
Docetaxel CHAARTED, mCSPC ADT alone III; 790 OS 57.6 vs. 44 mos (HR, AEs from GETUG-AFU
20156 0.61; 95% CI, 15:
0.47–0.80; p , .001) febrile neutropenia (7%
vs. 0%);
GETUG-AFU mCSPC ADT alone III; 192 OS 58.9 vs. 54.2 mos (NS)
elevated AST (2% vs.
15, 20134
, 1%);
STAMPEDE, mCSPC and locally ADT alone III; 1,086 OS 5-yr survival of 49% vs. elevated ALT (2% vs.
201711 advanced PCa 37%, (HR, 0.81; , 1%);
95% CI, 0.69–0.95; nervous system change
p = .009) (2% vs. 0%);
fatigue (7% vs. 1%)
Abbreviations: mCSPC, metastatic castration-sensitive prostate cancer; ADT, androgen deprivation therapy; OS, overall survival; AE, adverse effect; PCa,
prostate cancer; ART, androgen receptor inhibitor; NS, not significant; AST, aspartate aminotransferase; ALT, alanine aminotransferase.
resulted in survival improvement from trial baseline to agents is the preferred approach. Clinicians and patients
second progression, regardless of the choice of treatment must then consider other potential toxicities associated with
with subsequent ARB or docetaxel.33 AR-targeted treatments, including whether the financial
toxicity of daily oral AR-directed therapy is acceptable.
Disease volume, tumor burden (bone, soft tissue, or vis-
ceral), and de novo/recurrent metastatic status are key Although some adverse events are general to all ARB
considerations for decision-making and regimen selection. agents used to treat mCSPC, there are some unique effects
Despite the lack of supporting data, many clinicians fa- that may influence treatment choice. For example, abir-
vor chemohormonal therapy for patients with visceral in- aterone must be given in combination with prednisone to
volvement, in part supported by the evidence of benefit in avoid toxicities related to excess mineralocorticoid activity,
patients with high-volume disease in the CHAARTED trial.34 whereas the other AR-targeted agents are given without
For the many patients who decline chemotherapy or are steroids. Both enzalutamide and apalutamide are associ-
considered unfit because of age, comorbidities, or risk of ated with an increased risk of falls (3.7%, and 7.4%, re-
toxicity, an intensification strategy that relies on AR-targeted spectively) and fractures (6.5% and 6.3%, respectively)

Schulte et al
compared with ADT alone.31,32 Apalutamide was associated agents for mCSPC. Subgroup analyses, some preplanned
with rash in 27% of patients and hypothyroidism in 6.5% of and prespecified in the various trial protocols, may be hy-
patients, whereas enzalutamide was associated with seizure pothesis generating but are not considered definitive.40
in 0.3% of patients.31,32 Specific comorbidities or patient Ongoing biomarker studies, prospective comparator trials,
characteristics may be negatively affected by these toxic- and various meta-analyses may help address some of the
ities. These include diabetes (may wish to avoid treatments unanswered clinical gaps for mCSPC and support treat-
that require concomitant corticosteroids), history of falls or ment decisions between chemohormonal and AR-directed
balance issues (may wish to avoid apalutamide and therapies.
enzalutamide), seizure predisposition (should avoid enza-
TREATMENT CONSIDERATIONS FOR PCa PROGRESSING
lutamide), hepatic disease (should avoid abiraterone), and
AFTER COMBINATION THERAPY FOR mCSPC
cardiovascular disease (may wish to avoid abiraterone or
enzalutamide). In a population-based study of men with The pace of change in frontline treatment of mCSPC has
mCRPC receiving abiraterone or enzalutamide, those with rapidly incorporated data from the CHAARTED, STAMPEDE,
multiple cardiovascular comorbidities had increased short- LATITUDE, TITAN, ARCHES, and ENZAMET studies, creating
term mortality compared with patients without cardiovas- a new paradigm for upfront treatment of mCSPC.5,6,27,29-31
cular disease.35 Within a randomized, phase II trial in the Although the dramatic increase in the number of treatment
mCRPC setting, men age 75 or older reported poorer phys- combinations is exciting, questions remain regarding appro-
ical and functional well-being during treatment with enza- priate subsequent lines of therapy in patients who develop
lutamide but not with abiraterone.36 Other nonefficacy CRPC after progression on these agents in the mCSPC setting.
treatment factors that may affect quality of life and con- Treatment selection should include consideration of treat-
tribute to decision-making are still being defined. ments used for mCSPC, the clinical characteristics of the
patient, performance status, location of metastatic sites, de-
Understanding patient and family preferences, as well as the gree of symptom burden, and patient preferences. Future
challenges faced by patients receiving particular therapies, is decision-making is likely to include consideration of germline
a critical part of selecting treatments for mCSPC. AR-targeted and somatic genetic information and will eventually be in-
agents are given in the mCSPC setting for prolonged periods formed by several pivotal trials that are currently underway in
that often extend beyond 2 years, and patients must be will- the mCRPC setting.
ing to continue daily oral treatment. This can be a challenge
for men with cognitive impairment, with hectic personal or Considerations After AR-Directed Agents for mCSPC
professional schedules, or who face financial toxicity or other For men who underwent treatment of mCSPC with ADT
socioeconomic burdens. Financial toxicity is a complex and and an AR-directed agent (apalutamide, enzalutamide, or
understudied issue that some argue should be considered as abiraterone), use of a treatment option with a distinct
an adverse effect and incorporated within shared decision mechanism of action is generally preferred for patients with
discussions with patients.37 disease progression to mCRPC. Multiple studies in men with
mCRPC have shown limited response to a second AR-
When choosing an upfront systemic therapy, clinicians may
targeted agent following progression on an ARB therapy,
also consider complementary localized treatments for some
although these data do not include men with initial treat-
patients. One benefit of treatment with AR-directed therapy
ment of mCSPC.41-46 PLATO was a phase II study that
is that intensified treatment can proceed simultaneously,
evaluated patients with mCRPC who experienced PSA
such as with radiation of the primary tumor for men with
progression after treatment with enzalutamide. Patients
low-volume mCSPC, based on data from STAMPEDE.38
were treated with abiraterone and prednisone, with or
Although sequential therapy is possible (18% of those
without continued enzalutamide treatment. Disease control
enrolled in STAMPEDE arm H received docetaxel before
was poor in both arms, with a median PFS of 5.7 months in
radiotherapy), this does extend the duration of clinic visits
the combination group versus 5.6 months in the sequential
and may prolong treatment-related toxicity. Regardless of
group (HR, 0.83; 95% CI, 0.61–1.12; p = .22), and re-
whether patients proceed with chemohormonal or AR-
sponse rates of 0% versus 5% in single agent and com-
directed therapy, men with mCSPC should also be en-
bination arms, respectively.41 A separate phase II study of
couraged to consider participation in clinical trials of lo-
202 men with mCRPC randomly assigned men to sequential
calized therapy, including consideration of SWOG 1802, in
treatment with abiraterone followed by enzalutamide at PSA
which men received systemic therapy for 6 months before
progression (arm A) or to enzalutamide followed by abir-
undergoing randomization to prostatectomy or continued
aterone (arm B) and compared PSA decline and time to PSA
monitoring.39
progression on the second therapy.42 The trial showed
Despite the pace of change in mCSPC, there are not yet a difference in the PSA response to the second AR-directed
results from direct comparator trials evaluating the approved agent with a 50% or greater PSA decline of 34% versus 4%

(p , .001) for arms A and B, respectively, but a nonsig- improved OS compared with ADT plus placebo in the
nificant difference in median time to PSA progression on prechemotherapy CRPC setting (HR, 0.71; 95% CI,
second therapy of 2.7 versus 1.3 months (HR, 0.38; 95% 0.6–0.81; p , .001).48 In men with progression of disease
CI, 0.26–0.56). after treatment with docetaxel, enzalutamide was associated
In the third-line advanced mCRPC setting, the CARD study with improved survival compared with ADT plus placebo
was a multicenter open-label trial that evaluated patients (HR, 0.63; 95% CI, 0.53–0.75; p , .001).49 Abiraterone
who had previously been treated with docetaxel and an AR- was associated with improved survival compared with ADT
directed agent (abiraterone or enzalutamide); men were plus placebo for men with mCRPC before treatment with
randomly selected to receive treatment with the other AR- docetaxel (HR, 0.81; 95% CI, 0.7–0.93; p = .0033). Im-
directed agent versus cabazitaxel. Cabazitaxel was asso- proved survival was likewise shown with abiraterone in the
ciated with a 36% reduction in mortality compared with the postchemotherapy setting (HR, 0.65; 95% CI, 0.54–0.77;
alternate AR-directed agent (HR, 0.64; p = .0078).46 No- p , .001).50 No direct comparator trials have shown su-
tably, adverse events were not substantially higher using periority for individual treatments or treatment sequences in
chemotherapy than with an AR-directed agent, with 56.3 the CRPC setting. Thus, choice of agent involves a careful,
versus 52.4 patients experiencing at least grade 3 adverse shared decision process with patients and caregivers.
events in the cabazitaxel and AR arms, respectively.
Considerations for Specific Populations
Given these data suggesting a poor response to a second
AR-directed agent in the mCRPC setting, options for men Sipuleucel-T is a treatment option in the United States for
developing mCRPC on an ARB treatment should aim to men with mCRPC who have a low symptom burden and lack
change the mechanism of action rather than sequencing liver metastases. Sipuleucel-T was assessed in a random-
a second AR-directed therapy. Options therefore include ized phase III clinical trial in which asymptomatic men with
docetaxel, radium-223 for men with symptomatic pro- mCRPC without visceral metastases were randomly se-
gression of bone metastases, sipuleucel-T for asymptomatic lected in a 2:1 ratio to receive either sipuleucel-T or pla-
or minimally symptomatic patients, or treatment on a clinical cebo.52 With sipuleucel-T, there was a significant reduction
trial. Treatment on a clinical trial is always encouraged and in the risk of death (HR, 0.78; 95% CI, 0.61–0.98; p = .03),
preferred if patients are interested and they meet eligibility with a 4.1-month improvement in median survival
criteria. (25.8 months in the treatment group vs. 21.7 months with
placebo).52 Sipuleucel-T is generally well tolerated, and
Considerations After Docetaxel for mCSPC treatment is completed within approximately 6 weeks. Al-
Options differ for men who have been treated in the mCSPC though some men may have disease control without ra-
setting with chemohormonal therapy with ADT and doce- diographic progression for several months, the treatment is
taxel. Although a change in mechanism of action can still be not associated with prolonged PFS, and most men require
considered for men treated with chemohormonal therapy, subsequent therapy within 3–6 months of completing
cabazitaxel chemotherapy is effective for the treatment of treatment. The benefit of sipuleucel-T has not been formally
mCRPC in men with disease progression after docetaxel.47 assessed in men treated with combination CSPC treatment,
Treatment of mCRPC after chemohormonal therapy but the distinct mechanism of action makes shared re-
with ADT and docetaxel therefore include cabazitaxel, sistance mechanisms unlikely.
AR-targeted treatment (enzalutamide or abiraterone), Radium-223 is a radiopharmaceutical that is U.S. Food and
radium-223, sipuleucel-T, or treatment in a clinical trial. Drug Administration–approved for treatment of men with
Additionally, for patients who have had an extended period symptomatic bone metastatic CRPC. Therapy is adminis-
of disease control after initial docetaxel for mCSPC ( 12–18 tered in the postchemotherapy setting, for men not fit for
months), retreatment with docetaxel may be considered, chemotherapy, or for those who decline chemotherapy.53 It
although there are no prospective, randomized data de- was assessed in the ALSYMPCA trial, which is a phase III
fining response rates or other specific outcomes for these randomized, double-blind, placebo-controlled study in
patients. Treatment teams should monitor patients un- which patients were assigned in a 2:1 ratio to receive
dergoing retreatment closely with regular scans and PSA radium-223 or matching placebo every 4 weeks. The study
assessments to ensure that disease progression is identified showed improvement in OS with a median of 14.9 versus
promptly. 11.3 months (HR, 0.7; 95% CI, 0.58–0.83; p , .001).
Enzalutamide and abiraterone are currently the only AR- Treatment with radium requires coordination between
targeted agents that are approved in the pre- and post- oncology teams and nuclear medicine but is generally
chemotherapy settings in men with mCRPC.48-51 They can well tolerated. Radium-223 is not suitable for men with
be used for men treated with ADT alone or chemohormonal visceral metastases or soft tissue metastases (lymph
therapy for mCSPC. Enzalutamide was associated with nodes . 3.0 cm).54

Schulte et al
Although becoming increasingly uncommon, men receiving 0.64; 95% CI, 0.43–0.97; p = .0173).57 Exploitation of this
ADT alone in the mCSPC setting will need treatment in- vulnerability may move earlier in the treatment algorithm as
tensification when they experience progression to mCRPC. additional studies are completed.
Fortunately, there are ample phase III data to support Multiple clinical trials are ongoing to define an optimal
treatment choices for these men. In this setting, physicians method for treating patients in the frontline mCRPC set-
must consider patient performance status, comorbid con- ting after treatment of mCSPC. The CHAARTED2 trial
ditions, personal preferences, and potential limitations to (NCT03419234) is randomly assigning patients treated with
obtaining or safely receiving combination therapy with ADT. docetaxel in the mCSPC setting to receive treatment with
Importance of Shared Decision-Making abiraterone with or without cabazitaxel in the frontline
In addition to recognizing the treatment options available, mCRPC setting to determine how to best use chemotherapy
physicians must consider both clinical factors and patient or AR-targeted therapy.58 BRACAWAY (NCT03012321) is
preferences when making treatment choices with men facing a phase II trial in which patients with germline or somatic
treatment of first-line mCRPC. Clinical factors that may be DNA repair defect (DRD) mutations are randomly assigned
important include the pace and type of progression of dis- to receive treatment with abiraterone, olaparib, or abir-
ease. For example, progression with a PSA that is dispro- aterone and olaparib in a 1:1:1 ratio.59 The phase II
portionately low for the volume of disease or progression ImmunoProst trial (NCT03040791) likewise seeks to cap-
with predominantly liver metastases may prompt biopsy of italize on the effect of DRD in patients with mCRPC.60 This
a metastatic site to ensure that the patient is not developing study incorporates a single arm of nivolumab in patients who
neuroendocrine differentiation or an AR-independent phe- progressed on taxane-based therapy. Patients should al-
notype. Progression of disease that is concerning for small ways be encouraged to consider clinical trials if they ex-
cell differentiation may be best treated with a platinum and perience disease progression throughout the course of
etoposide chemotherapy combination rather than standard disease, including in the frontline mCRPC setting.
treatment of mCRPC, although there is little formal evidence In the absence of durable disease control, we aim to move
to support this approach. Additionally, patients progressing sequentially through treatments that extend patient survival
with highly symptomatic bone metastases may benefit from and delay, or altogether prevent, cancer-related morbidity.
consideration of palliative radiation, supportive treatment with As one may expect, second-, third-, and fourth-line CRPC
bone-targeted agents, and consultation with palliative care treatment options will all be impacted by upfront or prior
services, in addition to changes in systemic therapy for first- treatment selection. As of yet, there are few data to guide
line mCRPC. Furthermore, physicians should always dis- treatment choices, but anchor points exist. Notably, FIRSTANA
cuss which options exist with patients to support a shared showed that moving cabazitaxel forward, before docetaxel,
decision-making process, particularly when there is not was not superior.61 Further prospective data on sequential
a clinical reason to pursue one treatment over another. treatment activity will be gathered through worldwide col-
Future Considerations laborations such as the IRONMAN registry.
Although treatment of first-line mCRPC does not yet in- CONCLUSIONS

corporate genetic information, germline or somatic genetic With an increasing number of treatment options for mCSPC,
data can provide important information for later-line treat- it is paramount to understand the benefits and risks of each
ment decisions and is likely to influence initial treatment potential therapy so that we might optimize treatment of our
choice in the future. Although the prevalence of mismatch patients. Regardless of which treatment is chosen, a careful
repair deficiency, or microsatellite instability-high status, discussion with patients and families is critical. It is im-
within PCa is low (between 2% and 5%), early-phase portant to consider not only the potential toxicities but also
studies of immunotherapy in this population have shown logistical and financial barriers associated with treatment.
partial and even complete responses.55,56 Long-term data Furthermore, there must be realistic expectations for out-
supporting immunotherapy in PCa, however, are lacking. comes and responses to treatment, which can be limited to
Targeted therapy represents another potentially exciting a few months in duration for some men with highly ag-
alternative. The PROfound phase III study showed the ef- gressive disease. Last, whenever possible, treatment within
ficacy of olaparib in the treatment of men with metastatic the context of a clinical trial is always preferred, so that we
CRPC and DNA repair deficiencies, with disease progres- may gain further data to support treatment decisions and
sion after treatment with AR-directed therapy. Data pre- understand how best to serve our patients. Although the
sented at the 2019 European Society for Medical Oncology treatment of metastatic PCa has shifted dramatically since
Congress demonstrated an improved OS compared with the time of Huggins and Hodges, we must continue to
a second AR-targeted therapy (18.5 vs. 15.1 months; HR, advance the field to thwart this fatal disease.

AFFILIATIONS 60611; Twitter: @capsurvivorship; email: alicia.morgans@

1
Robert H. Lurie Comprehensive Cancer Center, Northwestern University, northwestern.edu.
Chicago, IL
2
Carolina Urologic Research Center, Myrtle Beach, SC
3 AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Yorkshire Cancer Research Weston Park Hospital, Sheffield, United
Kingdom
EDBK_278845.
Alicia Morgans, MD, MPH, Robert H. Lurie Comprehensive Cancer Center,
Northwestern University, 676 North St. Clair, Suite 850, Chicago, IL
REFERENCES
2. Herlemann A, Washington SL III, Cooperberg MR. Health care delivery for metastatic hormone-sensitive prostate cancer across the globe. Eur Urol Focus. 2019;
5:155-158.
3. Lee DJ, Mallin K, Graves AJ, et al. Recent changes in prostate cancer screening practices and epidemiology. J Urol. 2017;198:1230-1240.
4. Gravis G, Fizazi K, Joly F, et al. Androgen-deprivation therapy alone or with docetaxel in non-castrate metastatic prostate cancer (GETUG-AFU 15): a randomised,
5. Gravis G, Boher JM, Joly F, et al; GETUG. Androgen deprivation therapy (ADT) plus docetaxel versus ADT alone in metastatic non castrate prostate cancer: impact
of metastatic burden and long-term survival analysis of the randomized phase 3 GETUG-AFU15 trial. Eur Urol. 2016;70:256-262.
6. Sweeney CJ, Chen YH, Carducci M, et al. Chemohormonal therapy in metastatic hormone-sensitive prostate cancer. N Engl J Med. 2015;373:737-746.
7. Eisenberger MA, Crawford ED, Wolf M, et al. Prognostic factors in stage D2 prostate cancer; important implications for future trials: results of a cooperative
intergroup study (INT.0036). The National Cancer Institute Intergroup Study #0036. Semin Oncol. 1994;21:613-619.
8. Kyriakopoulos CE, Chen YH, Carducci MA, et al. Chemohormonal therapy in metastatic hormone-sensitive prostate cancer: long-term survival analysis of the
randomized phase III E3805 chaarted trial. J Clin Oncol. 2018;36:1080-1087.
9. Gravis G, Boher JM, Chen YH, et al. Burden of metastatic castrate naive prostate cancer patients, to identify men more likely to benefit from early docetaxel:
further analyses of CHAARTED and GETUG-AFU15 studies. Eur Urol. 2018;73:847-855.
10. James ND, Sydes MR, Clarke NW, et al; STAMPEDE investigators. Addition of docetaxel, zoledronic acid, or both to first-line long-term hormone therapy in
prostate cancer (STAMPEDE): survival results from an adaptive, multiarm, multistage, platform randomised controlled trial. Lancet. 2016;387:1163-1177.
11. Clarke NW, Ali A, Ingleby FC, et al. Addition of docetaxel to hormonal therapy in low- and high-burden metastatic hormone sensitive prostate cancer: long-term
survival results from the STAMPEDE trial. Ann Oncol. 2019;30:1992-2003.
12. Templeton AJ, Vera-Badillo FE, Wang L, et al. Translating clinical trials to clinical practice: outcomes of men with metastatic castration resistant prostate cancer
treated with docetaxel and prednisone in and out of clinical trials. Ann Oncol. 2013;24:2972-2977.
13. James ND, de Bono JS, Spears MR, et al; STAMPEDE Investigators. Abiraterone for prostate cancer not previously treated with hormone therapy. N Engl J Med.
2017;377:338-351.
14. Becker PS, Griffiths EA, Alwan LM, et al. NCCN guidelines insights: hematopoietic growth factors, version 1.2020. J Natl Compr Canc Netw. 2020;18:12-22.
15. Rush HL, Cook AD, Brawley CD, et al. Comparative quality of life in patients randomized contemporaneously to docetaxel or abiraterone in the STAMPEDE trial. J
Clin Oncol. 2020;38 (suppl; abstr 14).
16. Bumbaca B, Li W. Taxane resistance in castration-resistant prostate cancer: mechanisms and therapeutic strategies. Acta Pharm Sin B. 2018;8:518-529.
17. Mahon KL, Qu W, Lin HM, et al. Serum free methylated glutathione S-transferase 1 DNA levels, survival, and response to docetaxel in metastatic, castration-
resistant prostate cancer: post hoc analyses of data from a phase 3 trial. Eur Urol. 2019;76:306-312.
18. Mahon KL, Lin H-M, Lee-Ng M, et al. Clinical validation of circulating cytokines as markers of prognosis and response to docetaxel in men with metastatic
castration-resistant prostate cancer. J Clin Oncol. 2019;37 (suppl; abstr 230).
19. Choueiri TK, Xie W, D’Amico AV, et al. Time to prostate-specific antigen nadir independently predicts overall survival in patients who have metastatic hormone-
sensitive prostate cancer treated with androgen-deprivation therapy. Cancer. 2009;115:981-987.
20. Flanagan J, Gray PK, Hahn N, et al. Presence of the metabolic syndrome is associated with shorter time to castration-resistant prostate cancer. Ann Oncol. 2011;
22:801-807.
21. Sharma J, Gray KP, Evan C, et al. Elevated insulin-like growth factor binding protein-1 (IGFBP-1) in men with metastatic prostate cancer starting androgen
deprivation therapy (ADT) is associated with shorter time to castration resistance and overall survival. Prostate. 2014;74:225-234.
22. Zhao SG, Chang SL, Erho N, et al. Associations of luminal and basal subtyping of prostate cancer with prognosis and response to androgen deprivation therapy.
JAMA Oncol. 2017;3:1663-1672.

Schulte et al
23. Hamid A, Xin VW, Chen Y-H, et al. Luminal B subtype as a predictive biomarker of docetaxel benefit for newly diagnosed metastatic hormone sensitive prostate
cancer (mHSPC): a correlative study of E3805 CHAARTED. J Clin Oncol. 2020;38 (suppl, abstr 162).
24. Gillessen S, Attard G, Beer TM, et al. Management of patients with advanced prostate cancer: the Report of the Advanced Prostate Cancer Consensus Conference
APCCC 2017. Eur Urol. 2018;73:178-211.
25. Mohler JL, Antonarakis ES, Armstrong AJ, et al. Prostate cancer, version 2.2019, NCCN clinical practice guidelines in oncology. J Natl Compr Canc Netw. 2019;
17:479-505.
26. Heidenreich A, Bastian PJ, Bellmunt J, et al; European Association of Urology. EAU guidelines on prostate cancer. Part II: treatment of advanced, relapsing, and
castration-resistant prostate cancer. Eur Urol. 2014;65:467-479.
27. Fizazi K, Tran N, Fein L, et al; LATITUDE Investigators. Abiraterone plus prednisone in metastatic, castration-sensitive prostate cancer. N Engl J Med. 2017;
377:352-360.
28. Fizazi K, Tran N, Fein L, et al. Abiraterone acetate plus prednisone in patients with newly diagnosed high-risk metastatic castration-sensitive prostate cancer
(LATITUDE): final overall survival analysis of a randomised, double-blind, phase 3 trial. Lancet Oncol. 2019;20:686-700.
29. Davis ID, Martin AJ, Stockler MR, et al; ENZAMET Trial Investigators and the Australian and New Zealand Urogenital and Prostate Cancer Trials Group.
Enzalutamide with standard first-line therapy in metastatic prostate cancer. N Engl J Med. 2019;381:121-131.
30. Armstrong AJ, Szmulewitz RZ, Petrylak DP, et al. Phase III study of androgen deprivation therapy (ADT) with enzalutamide (ENZA) or placebo (PBO) in metastatic
hormone-sensitive prostate cancer (mHSPC): the ARCHES trial. J Clin Oncol. 2019;37 (suppl, abstr 687).
31. Chi KN, Agarwal N, Bjartell A, et al; TITAN Investigators. Apalutamide for metastatic, castration-sensitive prostate cancer. N Engl J Med. 2019;381:13-24.
32. Armstrong AJ, Szmulewitz RZ, Petrylak DP, et al. ARCHES: a randomized, phase III study of androgen deprivation therapy with enzalutamide or placebo in men
with metastatic hormone-sensitive prostate cancer. J Clin Oncol. 2019;37:2974-2986.
33. Agarwal N, Chowdhury S, Bjartell A, et al. Time to second progression (PFS2) in patients (pts) from TITAN with metastatic castration-sensitive prostate cancer
(mCSPC) by first subsequent therapy (hormonal vs. taxane). J Clin Oncol. 2020;38 (suppl, abstr 82).
34. Gillessen S, Attard G, Beer TM, et al. Management of patients with advanced prostate cancer: report of the Advanced Prostate Cancer Consensus Conference
2019. Eur Urol. In press. Available at: 10.1016/j.eururo.2020.01.012.
35. Lu-Yao G, Nikita N, Keith SW, et al. Mortality and hospitalization risk following oral androgen signaling inhibitors among men with advanced prostate cancer by
pre-existing cardiovascular comorbidities. Eur Urol. 2020;77:158-166.
36. Khalaf DJ, Sunderland K, Eigl BJ, et al. Health-related quality of life for abiraterone plus prednisone versus enzalutamide in patients with metastatic castration-
resistant prostate cancer: results from a phase II randomized trial. Eur Urol. 2019;75:940-947.
37. Penson DF. Keynote: Financial toxicity and quality of life: understanding and improving patient-centered outcomes in genitourinary malignancies. Presented at:
2020 Genitourinary Cancers Symposium; February 13, 2020; San Francisco, CA.
38. Parker CC, James ND, Brawley CD, et al; Systemic Therapy for Advanced or Metastatic Prostate cancer: Evaluation of Drug Efficacy (STAMPEDE) investigators.
Radiotherapy to the primary tumour for newly diagnosed, metastatic prostate cancer (STAMPEDE): a randomised controlled phase 3 trial. Lancet. 2018;
392:2353-2366.
39. NCT03678025. Standard Systemic Therapy With or Without Definitive Treatment in Treating Participants With Metastatic Prostate Cancer. https://clinicaltrials.
40. Halabi S. Subgroup analysis in clinical trials. Paper presented at: Advanced Prostate Cancer Consensus Conference; August 2019; Basel, Switzerland.
41. Attard G, Borre M, Gurney H, et al; PLATO collaborators. Abiraterone alone or in combination with enzalutamide in metastatic castration-resistant prostate cancer
with rising prostate-specific antigen during enzalutamide treatment. J Clin Oncol. 2018;36:2639-2646.
42. Khalaf D, Annala M, Finch DL, et al. Phase 2 randomized cross-over trial of abiraterone + prednisone (ABI+P) vs enzalutamide (ENZ) for patients (pts) with
metastatic castration resistant prostate cancer (mCPRC): results for 2nd-line therapy. J Clin Oncol. 2018;36 (suppl, abstr 5015).
43. Smith MR, Saad F, Rathkopf DE, et al. clinical outcomes from androgen signaling-directed therapy after treatment with abiraterone acetate and prednisone in
patients with metastatic castration-resistant prostate cancer: post hoc analysis of COU-AA-302. Eur Urol. 2017;72:10-13.
44. Zhang T, Dhawan MS, Healy P, et al. Exploring the clinical benefit of docetaxel or enzalutamide after disease progression during abiraterone acetate and
prednisone treatment in men with metastatic castration-resistant prostate cancer. Clin Genitourin Cancer. 2015;13:392-399.
45. Azad AA, Eigl BJ, Murray RN, et al. Efficacy of enzalutamide following abiraterone acetate in chemotherapy-naive metastatic castration-resistant prostate cancer
patients. Eur Urol. 2015;67:23-29.
46. de Wit R, de Bono J, Sternberg CN, et al; CARD Investigators. Cabazitaxel versus abiraterone or enzalutamide in metastatic prostate cancer. N Engl J Med. 2019;
381:2506-2518.
47. de Bono JS, Oudard S, Ozguroglu M, et al; TROPIC Investigators. Prednisone plus cabazitaxel or mitoxantrone for metastatic castration-resistant prostate cancer
progressing after docetaxel treatment: a randomised open-label trial. Lancet. 2010;376:1147-1154.
48. Beer TM, Armstrong AJ, Rathkopf DE, et al; PREVAIL Investigators. Enzalutamide in metastatic prostate cancer before chemotherapy. N Engl J Med. 2014;
371:424-433.
49. Scher HI, Fizazi K, Saad F, et al; AFFIRM Investigators. Increased survival with enzalutamide in prostate cancer after chemotherapy. N Engl J Med. 2012;
367:1187-1197.

50. de Bono JS, Logothetis CJ, Molina A, et al; COU-AA-301 Investigators. Abiraterone and increased survival in metastatic prostate cancer. N Engl J Med. 2011;
364:1995-2005.
51. Ryan CJ, Smith MR, Fizazi K, et al; COU-AA-302 Investigators. Abiraterone acetate plus prednisone versus placebo plus prednisone in chemotherapy-naive men
with metastatic castration-resistant prostate cancer (COU-AA-302): final overall survival analysis of a randomised, double-blind, placebo-controlled phase 3
52. Kantoff PW, Higano CS, Shore ND, et al; IMPACT Study Investigators. Sipuleucel-T immunotherapy for castration-resistant prostate cancer. N Engl J Med. 2010;
363:411-422.
53. Parker C, Nilsson S, Heinrich D, et al; ALSYMPCA Investigators. Alpha emitter radium-223 and survival in metastatic prostate cancer. N Engl J Med. 2013;
369:213-223.
54. Smith M, Parker C, Saad F, et al. Addition of radium-223 to abiraterone acetate and prednisone or prednisolone in patients with castration-resistant prostate
cancer and bone metastases (ERA 223): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol. 2019;20:408-419.
56. Graff JN, Alumkal JJ, Drake CG, et al. Early evidence of anti-PD-1 activity in enzalutamide-resistant prostate cancer. Oncotarget. 2016;7:52810-52817.
57. Hussain M, Mateo J, Fizazi K, et al. PROfound: phase III study of olaparib versus enzalutamide or abiraterone for metastatic castration-resistant prostate cancer
(mCRPC) with homologous recombination repair (HRR) gene alterations. Paper presented at: ESMO 2019; September 2019; Barcelona, Spain.
58. NCT03419234. Abiraterone acetate and antiandrogen therapy with or without cabazitaxel and prednisone in treating patients with metastatic, castration-resistant
prostate cancer previously treated with docetaxel. https://clinicaltrials.gov/ct2/show/NCT03419234. Accessed March 4, 2020.
59. NCT03012321. Abiraterone/prednisone, olaparib, or abiraterone/prednisone + olaparib in patients with metastatic castration-resistant prostate cancer with DNA
repair defects. https://clinicaltrials.gov/ct2/show/NCT03012321. Accessed March 4, 2020.
60. NCT03040791. Nivolumab in prostate cancer with DNA repair defects (ImmunoProst Trial) (ImmunoProst). https://clinicaltrials.gov/ct2/show/NCT03040791.
Accessed March 4, 2020.
61. Oudard S, Fizazi K, Sengeløv L, et al. Cabazitaxel versus docetaxel as first-line therapy for patients with metastatic castration-resistant prostate cancer:
a randomized phase III trial-FIRSTANA. J Clin Oncol. 2017;35:3189-3197.

Immune Checkpoint Blockade for Prostate

Cancer: Niche Role or Next Breakthrough?
Daniel Vargas P. de Almeida, MD1,2; Lawrence Fong, MD3; Matthew B. Rettig, MD4,5; and Karen A. Autio, MD, MSc1
overview
A number of trials have evaluated the use of single-agent immune checkpoint inhibitors for the treatment of
metastatic castration-resistant prostate cancer (mCRPC). The benefit appears to be limited to a small subset
of patients, such as those with tumors with microsatellite instability, highlighting the importance of bio-
markers to identify which patients may be more likely to respond. Given the lack of efficacy for most patients
with mCRPC, our understanding of the mechanisms of primary resistance to checkpoint inhibitors and of the
tumor immune microenvironment in prostate cancer is critical. Knowledge gained in these key areas will allow
for the identification of novel combination therapies that will circumvent resistance mechanisms and should
be tested in clinical trials. Improving our understanding of the effects of androgen deprivation therapy on
immune cells and of the most favorable disease setting (e.g., biochemically recurrent vs. castration-resistant
prostate cancer) may aid in the optimal use of checkpoint inhibitors in combination with other agents. If
successful, this may move immune checkpoint inhibitors into the treatment armamentarium of prostate
cancer management.
BACKGROUND background, diversity of cytokines, and the vascular

In the past 5 years, the earlier use of chemotherapy 1,2 network are relevant factors in the T-cell–mediated
or the addition of more potent therapies targeting antitumor response. A complex interaction exists
the androgen receptor (AR) or androgen biosynthesis among the fibroblast-infiltrated stromal cellular
inhibitors3-6 have been shown to improve the survival background, the metabolic state promoted by the
of patients with advanced prostate cancer. Although disturbed vasculature, and subsequent hypoxia
checkpoint inhibitors (CPIs) are standard of care in that contributes to an immunosuppressive TME. 10
patients with melanoma, lung cancer, and other geni- Cytokines, such as VEGF, transforming growth
tourinary tumors, such as urothelial and renal cell factor β (TGF-β), interleukin-10 (IL-10), and pros-
carcinoma, clinical trials assessing the efficacy of taglandin E2, in the TME are responsible for the
CPIs in advanced prostate cancer have been associ- recruitment of regulatory T cells (Tregs) and the
ated with low response rates, with benefits in only inhibition of proliferation, activation, and infiltration
a minority of patients.7-9 Efforts are ongoing to understand of cytotoxic lymphocytes.11 The importance of the
mechanisms of resistance and strategies to overcome TME in prostate cancer can be demonstrated early
the primary resistance to CPIs (Fig. 1); these studies in the natural history, with chronic persistent in-
are likely to guide future clinical trials with immuno- flammation being associated with prostate cancer
therapy. Biomarkers associated with a higher chance development. 12
of response are critical to identify the small percentage
The presence of tumor-infiltrating lymphocytes (TILs)
of patients with prostate cancer who do experience
Author affiliations is associated with response to CPIs in melanoma13
a response to monotherapy checkpoint inhibition. This
and support and with a better prognosis in colorectal carcinoma,
information (if
article reviews the characteristics of the immune mi-
melanoma, non–small cell lung cancer, and ovarian
applicable) appear croenvironment in prostate cancer, existing evidence
cancer14-16; however, the prostate cancer TIL pop-
at the end of this on the use of immune checkpoint blockade in this
article.
ulation largely consists of CD4+FOXP3+CD25+ T cells,
disease, biomarkers of response to CPIs, and future
Accepted on March a subpopulation of Tregs that dampens the immune
clinical trials that incorporate CPIs as part of combi-
17, 2020 and response by producing inhibitory cytokines and main-
nation strategies.
published at taining self-tolerance.17,18 A multi-institutional study
ascopubs.org on IMMUNE MICROENVIRONMENT OF PROSTATE CANCER demonstrated that higher levels of CD4+ T cells in the
April 28, 2020:
The tumor microenvironment (TME) plays a crucial surgical prostate specimen were associated with sig-
10.1200/EDBK_ role in the antitumor immune response. T-cell com- nificantly worse distant metastasis–free survival (HR,
278853 position, the spatial relationship of the stromal cellular 1.57; 95% CI, 1.25–1.97; p , .001), suggesting that

de Almeida et al
androgen suppression.23 Neoadjuvant ADT prior to radical

prostatectomy is associated with tumoral infiltration with
Tregs and CD4+ and CD8+ lymphocytes.24-26 The total
• The tumor immune microenvironment of macrophage population (CD68 + ) also increases after
prostate cancer harbors a predominance of
ADT.24,25 Analysis of tumor samples collected after ADT
immunosuppressive cells (e.g., regulatory
demonstrated that the T-cell repertoire is modified within
T cells and M2 macrophages).
the first week after treatment administration.24 Of interest,
• Low levels of circulating testosterone resulting ADT leads to increased migration of cells to the tumor site
from androgen deprivation therapy modulate
as well as increased proliferation of the cells within the
the prostate cancer immune
organ, with the exception of Tregs.24,26 Androgen sup-
microenvironment.
pression also has been associated with increased secretion
• Data from clinical trials demonstrate that only of the myeloid chemoattractant chemokine IL-8 by re-
a small subset of patients with prostate cancer
leasing AR-mediated transcriptional repression, resulting
have tumors that respond to single-agent
in the infiltration of MDSCs into the prostate tumor.27
immune checkpoint inhibitors.
Enhancing our understanding of the immunomodulatory
• Although many biomarkers associated with effects of more potent AR inhibitors (e.g., apalutamide)
response in other tumor types have been
and androgen biosynthesis inhibitors (e.g., abiraterone
explored in prostate cancer, the only
acetate with prednisone) in the hormone-sensitive and
actionable biomarker that predicts
response to checkpoint inhibition for men with castration-resistant settings will be important for rational
metastatic castration-resistant prostate cancer trial designs.
is microsatellite instability/mismatch repair
deficiency. COMPLETED CLINICAL TRIALS EVALUATING CHECKPOINT
BLOCKADE IN PROSTATE CANCER
• Multiple trials evaluating different
combination therapy strategies using Nearly 2 decades ago, studies of transgenic prostate
immune checkpoint inhibitors and their use cancer mouse models treated with anti–CTLA-4 therapy
across the disease continuum of prostate resulted in eradication of residual disease after surgery,28
cancer are ongoing. suggesting the potential of checkpoint inhibition as
a therapeutic strategy in prostate cancer. A decade later,
the results of the first phase I/II trial of patients with
metastatic castration-resistant prostate cancer (mCRPC)
these early immune cell populations may be associated with to receive ipilimumab, an antibody directed against CTLA-
recurrence or lack of immune surveillance to prevent re- 4, as monotherapy or in combination with radiation therapy
currence.19 In addition to Tregs, the prostate TME is popu- to bone metastases were published. Treatment efficacy
lated by tumor-associated macrophages (TAMs), including was assessed in 50 patients treated with ipilimumab
protumorigenic M2-subtype macrophages, which are re- (10 mg/kg); eight patients (16%) had a prostate-specific
sponsible for the secretion of high levels of TGF-β. The antigen (PSA) decline of at least 50%, and there was one
presence of M2 macrophages also has been associated complete radiographic response among the 28 evaluable
with a higher risk for metastatic disease at diagnosis (HR, patients.7 Given these promising results and the enthusi-
1.98; 95% CI, 1.17–3.33; p = .011).20 The cytokine milieu asm regarding CPIs in melanoma, ipilimumab was evalu-
is equally relevant, and the presence of chemokines, such ated in two phase III clinical trials enrolling men with mCRPC,
as CCL22, in the TME of prostate cancer contributes to with or without prior exposure to docetaxel chemotherapy.8,9
immune suppression. Several chemokine receptors (e.g., Neither trial was able to demonstrate an overall survival (OS)
CXCR4 and CXCR5) are expressed by Tregs and myeloid- advantage over placebo.
derived suppressor cells (MDSCs), leading to the integration
of innate and adoptive immune responses against prostate The first study enrolled patients with mCRPC and at least
cancer cells.21,22 one patient with bone metastasis who experienced pro-
gression after treatment with docetaxel. On the basis of data
TME Repertoire Is Modified by Androgens suggesting a synergistic effect of radiotherapy in combi-
Because androgen deprivation therapy (ADT) is the pre- nation with CPIs to activate an immune response,29,30 the
dominant therapy for many patients with prostate cancer, treatment schema consisted of a single dose of radiation for
the effects of low androgen levels on the immune milieu are at least one, and up to five, bone fields followed by ipili-
highly relevant. Increased levels of peripheral naive T cells mumab or placebo. The median OS was 11.2 months for
have been observed in patients initially undergoing ADT, ipilimumab and was 10.0 months for placebo (HR, 0.85;
demonstrating a reactivation of thymic function following 95% CI, 0.72–1.00; p = .053). Results of a post hoc analysis

A Critical Review of Immune Checkpoint Blockade in Prostate Cancer
FIGURE 1. Factors Affecting the Results of Immunotherapy Trials in Prostate Cancer

Abbreviations: AR, androgen receptor; IL-6, interleukin-6; PSMA, prostate-specific membrane antigen, TGF-β, tumor growth factor β; TIL, tumor-
infiltrating lymphocyte; TMB, tumor mutational burden; TME, tumor microenvironment.
suggested a potential treatment benefit in patients with with heavily pretreated mCRPC were included in a phase I
favorable prognostic features: an alkaline phosphatase trial assessing the safety and efficacy of the anti–PD-1
concentration of less than 1.5 times the upper limit of antibody nivolumab; all 17 patients had RECIST measur-
normal, a hemoglobin concentration of 11.0 g/dL or higher, able disease, and none had a radiographic response.31 In
and no visceral metastases (HR for OS, 0.62; 95% CI, vitro data revealed the presence of a high number of cir-
0.45–0.86; p = .0038).8 Those results led to speculation as culating PD-L1/L2+ dendritic cells in men who were expe-
to how these prognostic features, particularly visceral me- riencing progression while taking enzalutamide, suggesting
tastases, may be associated with the underlying TME. Pa- a potential role for anti–PD-L1 antibodies in this setting.32 A
tients with visceral metastasis were excluded from the clinical trial of the anti–PD-1 antibody pembrolizumab in
subsequent trial carried out in the mCRPC chemotherapy- patients experiencing progression while taking enzaluta-
naive setting, in which more patients had the previously mide reported that three (30%) of the first 10 patients had
defined favorable prognostic factors. However, there was no a 50% or greater reduction in serum PSA—two of whom had
a radiographic response.33 Later, it was confirmed that one
significant improvement in OS in this patient population,
of the patients experiencing a response had microsatellite
with a median OS of 28.7 months in the ipilimumab arm and
instability (MSI), which may explain why these positive re-
29.7 months in the placebo arm (HR, 1.1; 95% CI, 0.88–
sults were not substantiated in larger trials. Updated data
1.39; p = .3667).9
with a larger sample of patients from this trial demonstrated
As new CPIs were developed, such as antibodies targeting PSA reductions of 50% or greater in four patients (13%),
anti–PD-1 and its ligand (PD-L1), they were tested in pa- and six (25%) of 24 patients with measurable disease had
tients with metastatic prostate cancer. Seventeen patients radiographic responses.34 This scientific rationale also was

de Almeida et al
explored in the design of the phase III trial IMbassador250, setting or after progression on a prior standard treatment;
in which patients with mCRPC were randomly assigned to however, even in this setting, responses are generally only of
receive treatment with enzalutamide alone or enzalutamide durable benefit in a small percentage of patients. As such,
and atezolizumab; however, the trial was stopped because multiple efforts to identify predictive biomarkers are un-
of futility in a preliminary analysis.35 derway in malignancies in which there is a U.S. Food and
The antitumor activity of pembrolizumab also was eval- Drug Administration (FDA)–approved CPI. Many of these
uated in a phase Ib trial (KEYNOTE-028) that enrolled potential biomarkers of response have been evaluated in
patients with advanced solid tumors, who had RECIST 1.1 patients with prostate cancer.
measurable disease, and who were PD-L1+ (at least 1% PD-L1 Expression
expression in the tumor or stromal cells). Among the 245
PD-L1 expression in advanced melanoma, non–small cell
patients with mCRPC who were screened, 35 met criteria
lung cancer, and cervical cancer is associated with a higher
for PD-L1+ expression (14.3%) using a prototype assay
likelihood of response to CPIs. However, there is no asso-
(QualTek, Goleta, CA), and 23 were enrolled and received
ciation between PD-L1 expression and response in renal cell
treatment. Seven patients had a confirmed radiographic
carcinoma.44 Whether PD-L1 expression is associated with
response, resulting in an objective response rate (ORR) of
better response to CPIs in patients with mCRPC has been
17.4% (95% CI, 5.0%–38.8%).36 These results were not
assessed by several studies. The KEYNOTE-028 trial in-
replicated in a larger cohort studied in the phase II trial
cluded 23 patients with mCRPC and initially suggested that
(KEYNOTE-199) evaluating pembrolizumab in patients
PD-L1 expression ( 1% modified proportion score or interface
with mCRPC and prior exposure to docetaxel (258 pa-
pattern)45 could predict response to CPIs (ORR, 17.4%;
tients); this trial stratified patients according to the level
95% CI, 5%–38.8%).36 However, the larger KEYNOTE-199
of PD-L1 expression and the presence of measurable
trial demonstrated similar response rates between the
disease. The ORRs were 5% (95% CI, 2%–11%) and 3%
PD-L1 and PD-L1+ cohorts (ORRs, 3% and 5%, respec-
(95% CI, , 1%–11%) in those presenting with mea-
tively), which defined PD-L1 positivity as a combined pos-
surable disease in the PD-L1 + and PD-L1 cohorts,
itive score of 1 or greater (22C3 pharmaDx assay, Agilent
respectively.37
Technologies, Carpiteria, CA).37 Such results illustrate the
Given the limited activity with single-agent checkpoint low accuracy of PD-L1 as a biomarker of response to CPIs in
inhibition, the promising preclinical data using combina- patients with mCRPC. Even in malignancies for which PD-
tion therapy,38 and the clinical data of improved efficacy L1 expression has been associated with clinical outcomes, it
with combination therapy in melanoma, lung cancer, and remains an imperfect predictive biomarker, and there is
renal cell carcinoma,39-42 the combination of ipilimumab debate as to what constitutes positivity, with different criteria
(3 mg/kg) plus nivolumab (1 mg/kg) was evaluated in two and assays utilized across trials.46-49
cohorts of patients with mCRPC with or without prior
Microsatellite Instability–High Status or Mismatch
treatment with chemotherapy (CheckMate 650). Baseline
Repair Deficiency
measurable disease was present in 30 patients with prior
chemotherapy treatment and in 23 patients without prior In 2017, the FDA approved pembrolizumab for the treat-
chemotherapy treatment. Of patients with measurable ment of unresectable and metastatic tumors with MSI-high
disease, the ORRs were 25% (95% CI, 10%–48%) and (MSI-H) status or dMMR after progression on standard lines
10% (95% CI, 2%–27%) in those without and with prior of treatment.50 This approval was a hallmark in the field,
exposure to docetaxel, respectively. When analyzing only because it was tumor agnostic; interestingly, the registry that
the patients with PD-L1 positivity (defined as 1% ex- led to this approval included only two patients with mCRPC.
pression), the ORRs increased to 33.3% and 40%, respectively. Using Memorial Sloan Kettering Integrated Mutation Pro-
An exploratory analysis of all patients demonstrated that the filing of Actionable Cancer Targets (i.e., MSK-IMPACT) next-
ORR of those with tumors with alterations in DNA damage generation sequencing, the prevalence of MSI-H/dMMR
repair (DDR) genes was 40%; of those with tumors with tumors was 3.1% in a large series of patients with prostate
a tumor mutational burden (TMB) above the median, the cancer,51 which is consistent with other reports.52-54 How-
ORR was 56.3%.43 The prevalence of patients with tumors ever, the frequency at which MSI-H status can be detected
with mismatch repair deficiency (dMMR) has not been reis assay dependent, with a prevalence as high as 12%
ported for this trial. having been reported.55 In a retrospective series of 11
patients with mCRPC characterized by MSI-H/dMMR and
BIOMARKERS OF RESPONSE TO IMMUNE CHECKPOINT treated with anti–PD-1/PD-L1 inhibitors alone or in com-
BLOCKADE IN PROSTATE CANCER bination with another CPI, the ORR was 50%; 54.5% of the
CPIs are systemic treatment options for multiple types of patients had a reduction in PSA greater than 50%, including
advanced and metastatic tumors either in the first-line four with a decline greater than 99%.51

Ultramutated Genomic Status (57%) of seven patients treated with an anti–PD-1 antibody
There are emerging data regarding the relationship be- had a PSA decline greater than 50%.62 Treatment with
tween ultramutant status and alterations in the genes ipilimumab and nivolumab resulted in an ORR of 56.3% in
that encode DNA polymerase epsilon and delta1 (POLE patients with a TMB above the median (74.5 mutations/
and POLD1, respectively), which are responsible for patient) in the CheckMate 650 trial as well as longer ra-
proofreading and DNA replication, irrespective of the diographic progression-free survival when compared with
microsatellite status.56 In a pooled analysis of 47,721 those who had a TMB below the median (7.4 months [95%
patients with different primary tumors registered in CI, 6.5 months to not estimated] vs. 2.4 months [95% CI,
a genomic database, the prevalence of POLE and/or POL- 1.8–3.9 months], p , .0001).43 It is plausible that high
D1 alterations in patients with prostate cancer was 1.8%.56 TMB may be a genomic manifestation of dMMR.63 How-
The POLE/POLD1-mutated tumors carry very high rates of ever, given the limited data demonstrating the association
single nucleotide substitutions, often exceeding 100 mu- between TMB and response to CPIs in randomized clinical
tations/megabyte.57 These ultramutated tumors may have trials, TMB is not considered a biomarker of response
a greater number of tumor-associated antigens, which has validated for clinical use.64-66
been described in at least one case report of a man with DNA Damage Repair
prostate cancer and a POLE-mutated microsatellite-stable Alterations in DDR genes can lead to genomic instability,
tumor that responded to CPI therapy.58 If confirmed in which may yield increased neoantigen formation and
additional analyses, these genomic alterations could be greater immunogenicity. DDR alterations can be found in
further analyzed as potential biomarkers of response 22.7% of patients with prostate cancer, with BRCA2 and
to CPIs. ATM being the most frequently affected genes.52 The
CDK12 Loss of Function different DNA sequencing methods and definitions used
The possibility of biallelic somatic loss-of-function muta- to define an alteration (e.g., monoallelic versus biallelic)
tions in CDK12 as a biomarker of response to CPIs was pose challenges in our understanding of DDR alterations
proposed by Wu et al,59 because CDK12 alterations are as biomarkers of sensitivity to CPIs.67 With this in mind, an
associated with focal tandem duplications that lead to gene exploratory analysis of the phase II clinical trial KEYNOTE-
fusions that can generate neoantigens. Corroborating their 199 suggested a potential association between alter-
hypothesis, the investigators reported clinical activity, with ations in DDR genes captured by whole-exome DNA
PSA decreases in two of four patients with CDK12 muta- sequencing (Genome Analysis Toolkit [GATK] and MuTec;
tions treated with CPIs.59 Because alterations in CDK12 Broad Institute, Cambridge, MA) and response to an
can be found in approximately 6.9% of patients with anti–PD-1 antibody, but the ORR was still low (11%).37
mCRPC,59 this result could have important implications, Results of a phase II trial exploring combined treatment
although the clinical data are still largely anecdotal. A using the anti–PD-L1 antibody durvalumab and the PARP
multi-institutional retrospective series of cases presented inhibitor olaparib included 17 patients with mCRPC, with or
at the 2019 Congress of the European Society of Medical without a DDR gene alteration; nine patients (53%) had
Oncology included eight patients with monoallelic or reductions in PSA of at least 50%, and four (33%) of 12
biallelic CDK12 alterations who were treated with an patients with measurable disease had a radiographic
anti–PD-1/PD-L1 inhibitor; a 50% reduction in PSA levels response per RECIST. Genomic analysis detected a DDR
was observed in three (37.5%) patients. The median gene alteration in six (66%) of the nine patients with
progression-free survival among the eight patients was a PSA response (three with germline BRCA2, one with
6.6 months (95% CI, 2.3–10.8 months).60 Clinical trials germline NBN, and two with homologous somatic
to prospectively assess the efficacy of CPIs in patients BRCA2).68 Further studies are exploring the combination of
with CDK12 alterations are underway (NCT03570619 CPIs and PARP inhibitors. In the phase II clinical trial
and NCT03810105). CheckMate 650, treatment with ipilimumab plus nivolumab
was associated with an ORR of 40% among the 10 patients
Tumor Mutational Burden with DDR gene alterations presenting with measurable
TMB is associated with response to CPIs in some cancers disease.43
(e.g., non–small cell lung cancer) but not in others (e.g., RATIONAL COMBINATIONS WITH CHECKPOINT INHIBITORS
renal cell carcinoma).61 However, the mutational burden in
metastatic prostate cancer is generally low (median, 2.9 Combination With Existing Standard Therapies That
mutations/megabyte),62 and only 3.0% to 8.3% of ad- Induce Cell Death
vanced prostate cancer tumors have high TMB.51,62 In Radiation therapy Treatment with radiation therapy is as-
a small retrospective series of 12 patients with high TMB sociated with changes in the tumor immune microenvi-
(defined as more than 10 mutations/megabyte), four ronment and can cause immunogenic cell death. In addition

de Almeida et al
to cell death from direct DNA damage, radiation therapy proposed mechanisms for the efficacy of the combination;
induces inflammatory cytokines and the recruitment of in a breast cancer model, PARP inhibition leads to
dendritic cells that can convert the tumor into an in situ upregulation of PD-L1 and superior efficacy of the com-
vaccine, thus activating the tumor-specific T cells.69 Dose, bined treatment over each agent alone.84 It has also been
fractionation, and, possibly, metastatic site are components proposed that PARP inhibition causes DNA fragments to
that can contribute to the immune impact of radiation.70-73 be released into the cytosol, leading to activation of the
The rationale for combining treatment with CPIs and radiation cyclic GMP-AMP synthase-stimulator of the interferon
therapy is to induce a more potent immune response at the genes pathway.85,86 Clinically, the combination of olaparib
sites of irradiated disease, as well as at distant locations—a and durvalumab has been evaluated in 17 patients with
phenomenon referred to as the “abscopal effect.” 29,30 mCRPC in a phase II trial. Nine patients (53%) experi-
Compiled data from 35 studies in a meta-analysis demon- enced a radiographic and/or PSA response, resulting in
strated that radiotherapy and CPI appeared to be safe, but the a 12-month progression-free survival of 51.5% (95% CI,
combination was not associated with an OS benefit for the 25.7%–72.3%). Genomic analysis identified alterations
majority of patients.73 Given the rationale and safety profile, in DDR genes in nearly two-thirds of the patients with
the combination of other immune checkpoint agents and responses. The 12-month progression-free survival was
radiation therapy is under investigation in high-risk localized higher in the DDR-mutated group than in the group with-
prostate cancer (NCT03543189), oligometastatic prostate out DDR mutations (83.3% [95% CI, 27.3%–94.5%] vs.
cancer (NCT03795207), and mCRPC (NCT03217747 and 36.4% [95% CI, 11.2%–62.7%]; p = .031).68 How this
NCT01303705). compares to PARP inhibition alone has not been stud-
Chemotherapy Often misconceived as an immunosuppres- ied in a randomized fashion in prostate cancer. Addi-
sive treatment, chemotherapy can have immunomodula- tional trials evaluating CPIs and PARP inhibitors in
tory effects. Chemotherapy in combination with CPIs is patients with mCRPC (NCT03572478, NCT03330405,
now standard of care in small cell lung cancer, non–small and NCT02484404) are underway. A trial assessing treat-
cell lung cancer, and triple-negative breast cancer.74-77 ment of molecularly selected patients with biochemically
Cytotoxic chemotherapy has been used in combination recurrent prostate cancer initially without ADT is also in progress
with immunotherapy for several purposes, including re- (NCT03810105).
duction of tumor burden, activation of an antigen cas- Antiangiogenics Regulation of the tumor vasculature has
cade, and reduction of MDSCs.78,79 The type, dose (low, been shown to promote a proinflammatory state in the
standard, or submyeloablative), and mechanism of action TME.11 The improved vascular network induces the po-
of the chemotherapy can produce different effects.80 larization of TAMs toward the M1-like phenotype and
Taxanes are of particular interest, because prostate can- facilitates tumor infiltration with CD4+ and CD8+ T cells in
cer is taxane sensitive. In mice, treatment with docetaxel addition to decreasing the MDSC population in preclinical
was shown to increase cytotoxic T-lymphocyte response models.87 The combination of CPIs with antiangiogenic
and decrease MDSCs.81 Studies in which tumor-bearing drugs has been shown to be effective in the treatment
mice were administered docetaxel after a booster vac- of some solid tumors, such as renal and endometrial
cination demonstrated that antigen-specific T-cell re- cancers.88-90 The combination of the multiple-receptor
sponses to tumor-derived antigens distinct from the tyrosine kinase inhibitor cabozantinib with atezolizumab
antigen used in the vaccine were induced, suggestive of is being studied in patients with mCRPC who have been
antigen cascade. 82 These effects were not demon- previously treated with abiraterone acetate with predni-
strated when docetaxel was given prior to or during sone and/or enzalutamide. Initial results of the first pa-
vaccination, highlighting the importance of sequencing tients treated demonstrated an ORR of 32% (80% CI,
with combination treatment. Preclinical data demon- 23%–42%) among 44 patients with measurable disease
strate that paclitaxel decreases Tregs and skews mac- and a disease control rate of 80%. Ten (29%) of 34
rophages to an M1 proinflammatory phenotype.83 Of patients had a reduction in serum PSA level of at least
course, murine models may not faithfully recapitulate 50%.91
human disease, and whether these results hold true in
humans needs to be determined. However, multiple Combination Strategies With Other Immune Therapies
trials combining chemotherapy and CPIs are underway Targeting cytokines: IL-8 and TGF-β IL-8 is a proinflammatory
in patients with metastatic castration-sensitive prostate chemokine that is associated with cell survival and pro-
cancer (NCT03951831 and NCT03879122) and in patients liferation. Elevated serum levels have been associated with
with mCRPC (NCT03248570 and NCT03834506). poor prognosis in metastatic castration-sensitive disease.92,93
PARP inhibitors The combination of PARP inhibitors with IL-8 is also modulated by androgens, making this a rele-
CPIs has been explored preclinically, and there are several vant target in prostate cancer. Castration in murine

models leads to secretion of IL-8 and infiltration of MDSCs, from its receptors, such as the adenosine A2A receptor
although the effects could be blocked using an antibody and adenosine A2B receptor, or by inhibiting key enzy-
against CXCR2, the receptor for IL-8.27 A clinical trial using matic pathways involved in the catabolism of adenosine
an IL-8 antibody in combination with nivolumab and ADT is diphosphate and nicotinamide adenine dinucleotide,
underway in recurrent hormone-sensitive prostate cancer such as monoclonal antibodies against the nucleotidases
(NCT03689699). CD73 or CD39.101 Initial results of a phase I trial evalu-
ating the A2A receptor inhibitor AZD4653, alone or in
TGF-β is present in the TME and appears to be an im-
combination with the anti–PD-L1 antibody durvalumab,
portant mediator of immune resistance in prostate can-
demonstrated a confirmed response rate of 37.5% in
cer, potentially as it relates to bone metastases.94 Results
patients with mCRPC and RECIST-evaluable disease.
from a preclinical study using castration-resistant animal
Additionally, a durable PSA decline greater than 99% was
models suggest that bone metastases induce osteoclast-
found in one (25%) of the four patients evaluated.103 The
mediated bone resorption, releasing TGF-β in the bone
A2A receptor antagonist ciforadenant is being evaluated
microenvironment; in combination with IL-6, TGF-β con-
when used alone or combined with the anti–PD-L1 an-
tributes to the polarization of CD4+ T cells toward the im-
tibody atezolizumab in an ongoing phase I trial in patients
munosuppressive Treg and T helper 17 lineages and away
with mCRPC and measurable disease. Initial results of the
from the effector T helper 1 lineage.95 In that study, com-
first 35 patients demonstrate a disease control rate of
bined treatment with a CPI and an anti–TGF-β antibody
20%, including one patient treated with the combination
induced CD4+ lymphocyte polarization to the T helper 1 cell
who had an objective response. A PSA reduction of at
subset, promoting the expansion of CD8+ effector memory
least 50% occurred in three patients. 104 Further studies
cells and control of tumor growth.95 Inhibition of TGF-β is
are being conducted to explore this pathway in patients
under evaluation in combination with hormonal therapy in
with mCRPC, including in combination with CPIs and
the treatment of patients with mCRPC (NCT03685591 and
chemotherapy (NCT03454451, NCT03629756, and
NCT02452008).
NCT04089553).
Targeting TAMs via colony-stimulating factor 1 receptor Other immune checkpoints: agonists and inhibitors Targeting
Other mechanisms to target the immune microenvironment immune agonists, such as glucocorticoid-induced tumor
have focused on the elimination of TAMs. In a phase I trial necrosis factor receptor, CD134 (OX-40), and inducible
of patients with breast cancer and advanced mCRPC, in- costimulator, has resulted in few objective responses in
hibition of colony-stimulating factor 1 (CSF1) receptor to phase I trials of patients with solid tumors,105-107 although
reduce M2 macrophages and TAMs has been evaluated. several have reported evidence of immune activation. One
Although target engagement was demonstrated through notable exception was with a CD40 agonist: when patients
elevated levels in the CSF1 receptor ligands CSF1 and IL-34, with metastatic pancreatic cancer were treated with
there was no efficacy signal in this small cohort of patients a CD40 agonist, gemcitabine, and nab-paclitaxel che-
with prostate cancer to support further monotherapy.96 In an motherapy, with or without nivolumab, 14 patients (64%)
all–solid tumor trial, little clinical activity was reported (ORR, who were evaluable for dose-limiting toxicity had a partial
0.7%) when the anti–CSF1 receptor monoclonal antibody response.108 Because pancreatic cancer shares some
emactuzumab was used in combination with paclitaxel. immunologic similarities with prostate cancer, such as
Although analysis of pre- and post-treatment metastatic the role of fibroblasts and an immunosuppressive TME,
biopsies suggested a decrease in M2 macrophages, the treatment using a CD40 agonist in combination with
treatments could not reprogram the remaining macro- chemotherapy may be a strategy to consider for patients
phages into proinflammatory TAMs.97 Future combination with prostate cancer as well.109-112
therapies may build upon these findings to decrease TAMs,
but it is important to emphasize the necessity for combi- CPIs other than anti–CTLA-4 and anti–PD-1/PD-L1 also may
nation therapies to modulate other aspects of the immune have a role in prostate cancer; there was a notable increase
system to drive T effector cells and generate an antitumor in the immune infiltrate in the tumor tissue of patients
response. with localized prostate cancer treated with a neoadjuvant
course of ipilimumab and ADT. Interestingly, the majority
Adenosine pathway Another area of great interest in prostate of the T-cell population had upregulation of PD-L1 and
cancer is targeting the adenosine signaling pathway.98 El- V-domain immunoglobulin suppressor of T-cell activa-
evated levels of adenosine in the TME can impair T-cell tion.113 Coexpression with other inhibitory receptors, such
function, and adenosine appears to enhance Tregs and as lymphocyte-activating gene 3 and T-cell immunor-
MDSCs, limit the functionality of dendritic cells and T cells, eceptor with immunoglobulin and tyrosine-based in-
and support protumorigenic and angiogenic fibroblasts.99-102 hibition motif domains, also may play a role in the lack of
The pathway can be targeted by blocking adenosine immune response to PD-1/PD-L1 and CTLA-4 inhibitors.

de Almeida et al
Each of these checkpoints has at least one known in- immune-based therapies. Designed to engage a tumor
vestigational therapy targeting this pathway. Although cell surface–specific antigen (e.g., PSMA) and the CD3 co-
dual-checkpoint inhibition may offer advantages, it is receptor on T cells, bispecific T-cell engagers (also known
unlikely that this will be the optimal immunologic ap- as "BiTEs") drive T cells toward tumor sites, and the efficacy
proach for most patients with prostate cancer, whose in prostate cancer is under investigation. Preclinical data
tumors harbor a more inert microenvironment lacking demonstrated regression of tumors and improved survival
T-cell infiltration. in xenograft models.124 Pasotuxizumab, a BiTE-targeting
PSMA, was shown to be active in patients with mCRPC in
Cancer vaccines Cancer vaccines use a cellular antigen a dose-escalation study of 15 patients reported at the
to induce an antigen-specific adaptive immune re- American Society for Clinical Oncology Annual Meeting in
sponse. In prostate cancer, vaccines have been studied 2019. Two patients had durable responses of greater than
for decades because of the common tumor antigens 1 year, with others experiencing decreases in PSA.125
PSA, prostate-specific membrane antigen (PSMA), pros- Currently, at least three phase I clinical trials are ex-
tatic stem cell antigen, and prostatic acid phosphatase ploring BiTEs in patients with mCRPC, some in combi-
(PAP), which are expressed on the surface of prostate nation with CPIs (NCT00635596, NCT01723475, and
cancer cells.114,115 Sipuleucel-T is a cellular immuno- NCT03792841).
therapy product derived from a patient’s peripheral blood
mononuclear cells that are exposed to a fusion protein Chimeric antigen receptor T-cell therapy Given the success
of PAP and granulocyte-macrophage colony-stimulating of chimeric antigen receptor (CAR) T-cell therapy in
factor. It is the only FDA-approved immunotherapy in the treatment of hematologic malignances,126-132 cellular
prostate cancer, with a modest survival benefit in therapies are also under investigation in solid tumors.
castration-resistant prostate cancer, although a biomarker Different tumor-associated antigens can potentially be
to assess efficacy on an individual level is lacking. Of note, targeted in the development of CAR T-cell agents for
black patients with low serum PSA levels treated with prostate cancer: PSA, PAP, T-cell receptor gamma alter-
sipuleucel-T had longer median OS compared with white nate reading frame protein, transient receptor potential-
patients.116 p8, prostatic stem cell antigen, and PSMA, among
Several vaccine-based approaches for prostate cancer others.133 Most studies have focused on prostatic stem cell
have failed in the late stages of development.117-121 Im- antigen and PSMA because of their specific expression in
mune tolerance to self-antigen is one mechanism medi- prostate cancer and high expression in advanced disease.
ating the intrinsic resistance to vaccine therapies in Two phase I trials have reported results on these agents in
prostate cancer, and successful vaccination may indeed the treatment of mCRPC. A first-generation PSMA CAR
be enhanced with combination approaches, including T-cell was evaluated in combination with the continuous
checkpoint inhibition. Many vaccine trials incorporate administration of low-dose IL-2 in five patients with mCRPC.
intravenous administration of CPIs. Others have used Two patients had clinical responses, with PSA decreases of
subcutaneous administration at the same site as the 50% and 70%.134 In another phase I study, Slovin et al135
vaccination, with the goal of maximizing delivery to the reported on seven patients who were treated with a sec-
draining lymph nodes, optimizing antigen presentation, ond-generation PSMA CAR T cell. The treatment was
and potentially reducing toxicity by limiting systemic ex- safe, and two patients had prolonged stable disease
posure (NCT02616185). Initial results using the combi- (more than 6 and 16 months). Multiple clinical trials
nation of pembrolizumab and the PAP-encoding DNA evaluating this class of agents are ongoing, as are studies
vaccine pTVG-HP in patients with mCRPC showed with CAR T cells targeting other tumor-associated antigens,
promising efficacy and safety (NCT02499835).122 In the such as epithelial cell adhesion molecules (NCT03873805,
phase I/II KEYNOTE-046 trial, the combination of pem- NCT03089203, NCT04053062, NCT03013712, and
brolizumab and ADXS31-142 (an attenuated Listeria NCT02744287). Clinical trials using CAR natural killer cells
monocytogenes–listeriolysin O immunotherapy) is being are also underway (NCT04107142 and NCT03692663).
explored in the treatment of patients with mCRPC SELECTION OF PATIENTS IN CLINICAL TRIALS
(NCT02325557).123 There are no trials of personalized
cancer vaccines specifically for patients with prostate As the next wave of studies explores new immunotherapy
cancer, perhaps over concerns that low TMB will decrease targets and different combinations (Table 1), un-
the likelihood of finding predictive neoepitopes to generate derstanding the optimal timing of these therapies and the
a personalized vaccine. patient populations in which to test them is key. The
majority of reported trials using CPIs have focused on
Bispecific T-cell engagers The targeting of common tu- mCRPC and have yielded disappointing results; however,
mor antigens is also the foundation for other nonvaccine evaluation of CPIs in earlier disease states of prostate

TABLE 1. The Current Landscape: Ongoing Clinical Trials Using Checkpoint Inhibition in Prostate Cancer
Treatment Regimens Phase Status NCT Number
Localized Prostate Cancer
CPI
Nivolumab + ADT + brachytherapy + EBRT I/II Recruiting NCT03543189
Pembrolizumab + ADT + enzalutamide + surgery II Recruiting NCT03753243
Atezolizumab + enzalutamide + surgery; atezolizumab + surgery II Recruiting NCT03821246
CPI + vaccine
Atezolizumab + PROSTVAC-V/F + MVA-BN-Brachyury + surgery II Recruiting NCT04020094
Ipilimumab + surgery; PROSTVAC-V/F + surgery; ipilimumab + II Active, not recruiting NCT02506114
PROSTVAC-V/F + surgery
Nivolumab + PROSTVAC-V/F + surgery I/II Recruiting NCT02933255
Nivolumab + ChAdOx1-MVA 5T4 vaccine + surgery I/II Recruiting NCT03815942
Biochemical Recurrent Prostate Cancer
CPI
Nivolumab II Recruiting NCT03637543
Nivolumab II Recruiting NCT04019964a
Ipilimumab + degarelix II Active, not recruiting NCT02020070
CPI + cytokine-targeted agent
Nivolumab + degarelix; nivolumab + degarelix + BMS-986253 I/II Recruiting NCT03689699
(anti–IL-8 antibody)
CPI + PARP inhibitor
Durvalumab + olaparib (with or without degarelix) II Recruiting NCT03810105b
CPI + vaccine
Nivolumab + pTVG-HP (plasmid DNA vaccine) + GM-CSF II Recruiting NCT03600350
PF-06801591 (anti–PD-1 antibody) + tremelimumab + PF- I Recruiting NCT02616185
06755992 (adenovirus) + PF-06755990 (plasmid) via TDS-IM
electroporation
Metastatic Castration-Sensitive Prostate Cancer
CPI
Durvalumab + SBRT; SBRT II Recruiting NCT03795207
Ipilimumab + degarelix + radical prostatectomy II Active, not recruiting NCT02020070
CPI + chemotherapy
Cemiplimab + ADT + docetaxel II Recruiting NCT03951831
Nivolumab + ADT + docetaxel; nivolumab + ADT + ipilimumab II/III Recruiting NCT03879122
alternating with docetaxel; ADT + docetaxel
CPI + vaccine
Nivolumab + ipilimumab + PROSTVAC-V/F + ADT; nivolumab + I Recruiting NCT03532217
ipilimumab + PROSTVAC-V/F + ADT + neoantigen DNA vaccine
via TriGrid electroporation
CPI + other
Pembrolizumab + SD-101 (TLR9 agonist) + ADT + abiraterone + II Recruiting NCT03007732
SBRT; pembrolizumab + ADT + abiraterone + SBRT
mCRPCc
CPI
Nivolumab + ipilimumab IId Recruiting NCT03570619

de Almeida et al
TABLE 1. The Current Landscape: Ongoing Clinical Trials Using Checkpoint Inhibition in Prostate Cancer (Continued)
Pembrolizumab II Recruiting NCT03506997e
Pembrolizumab II Recruiting NCT04104893f
Durvalumab + tremelimumab; durvalumab II Active, not recruiting NCT02788773
Durvalumab + tremelimumab II Active, not recruiting NCT03204812
Ipilimumab II Active, not recruiting NCT01498978
Tremelimumab II Active, not recruiting NCT00378482
Avelumab I Active, not recruiting NCT01772004
Nivolumab + ipilimumab II Recruiting NCT02601014g
Nivolumab + ipilimumab II Recruiting NCT03061539g
Nivolumab II Recruiting NCT03040791b
Nivolumab + ipilimumab; ipilimumab; cabazitaxel II Recruiting NCT02985957
Pembrolizumab + XmAb22841 (anti–CTLA-4/LAG-3 bispecific I Recruiting NCT03849469
antibody); XmAb22841
CPI + adenosine pathway–targeted agent
Pembrolizumab + CPI-006 (anti-CD73 antibody); CPI-006; CPI-006 + I Recruiting NCT03454451
ciforadenant (A2AR antagonist)
Atezolizumab + ciforadenant I Recruiting NCT02655822
Durvalumab + AZD4635 (A2AR antagonist); oleclumab (anti-CD73 II Recruiting NCT04089553
antibody) + AZD4635
Durvalumab + AZD4635; durvalumab + oleclumab + AZD4635; I Recruiting NCT02740985
abiraterone + AZD4635; enzalutamide + AZD4635; docetaxel +
AZD4635
CPI + androgen receptor axis agent
Pembrolizumab + enzalutamide; pembrolizumab II Active, not recruiting NCT02787005i
Pembrolizumab + enzalutamide; enzalutamide III Recruiting NCT03834493
Pembrolizumab + enzalutamide II Active, not recruiting NCT02312557
Atezolizumab + enzalutamide; ADT + enzalutamide III Active, not recruiting NCT03016312
Ipilimumab + abiraterone I/II Active, not recruiting NCT01688492
Ipilimumab + abiraterone + apalutamide; abiraterone + II Active, not recruiting NCT02703623
apalutamide; abiraterone + apalutamide + cabazitaxel +
carboplatin
Avelumab + abiraterone; avelumab + enzalutamide II Recruiting NCT03770455
Nivolumab + enzalutamide II Recruiting NCT03338790
Pembrolizumab + enzalutamide or abiraterone I Recruiting NCT02861573
CPI + BiTE
Pembrolizumab + AMG 160 (anti-PSMA/CD3); AMG 160 I Recruiting NCT03792841
CPI + chemotherapy
Nivolumab + docetaxel II Recruiting NCT03338790
Pembrolizumab + docetaxel or cabazitaxel II Recruiting NCT03248570j
Pembrolizumab + docetaxel; docetaxel III Recruiting NCT03834506
Durvalumab + tremelimumab + vinorelbine I/II Recruiting NCT03518606
Avelumab + PT-112 (platinum-pyrophosphate agent) I/II Recruiting NCT03409458
Nivolumab + ipilimumab + cyclophosphamide + cryosurgery II Recruiting NCT04090775
Pembrolizumab + docetaxel I Recruiting NCT02861573

Pembrolizumab + guadecitabine I Recruiting NCT02998567
CPI + cytokine-targeted agent
Pembrolizumab + navarixin (CXCR1/2 antagonist) II Recruiting NCT03473925
Nivolumab + NKTR-214 (IL-2 agonist); nivolumab + SBRT + CDX- I Recruiting NCT03835533
301 (recombinant Flt3L) + poly-ICLC (TLR3 agonist)
CPI + PARP inhibitor
Nivolumab + rucaparib; nivolumab; rucaparib I/II Recruiting NCT03572478
Pembrolizumab + olaparib; abiraterone or enzalutamide III Recruiting NCT03834519
Durvalumab + olaparib + cediranib; durvalumab + olaparib; I/II Recruiting NCT02484404
durvalumab + cediranib
Avelumab + talazoparib II Recruiting NCT03330405j
Nivolumab + rucaparib II Recruiting NCT03338790
Pembrolizumab + olaparib I Recruiting NCT02861573
CPI + radioligand
177
Pembrolizumab + Lu-PSMA I/II Recruiting NCT03658447
223 223
Pembrolizumab + Ra; Ra II Active, not recruiting NCT03093428
CPI + vaccine
Nivolumab + ChAdOx1-MVA 5T4 vaccine I/II Recruiting NCT03815942
PF-06801591 (anti–PD-1 antibody) + tremelimumab + PF- I Recruiting NCT02616185
06755992 (adenovirus) + PF-06755990 (plasmid) via TDS-IM
electroporation
Nivolumab + PROSTVAC-V/F I/II Recruiting NCT02933255
Nivolumab + CDX-301 + INO-5151 (combined antigens for PSA and I Recruiting NCT03835533
PSMA and vector expressing IL-12)
Pembrolizumab + ADXS31-142 (Lm-LLO-PSA vaccine); ADXS31- I/II Active, not recruiting NCT02325557
142
Pembrolizumab + pTVG-HP (plasmid) + rhGM-CSF; pembrolizumab + II Recruiting NCT04090528
pTVG-HP + pTVG-AR (plasmid) + rhGM-CSF
Pembrolizumab + pTVG-HP I/II Recruiting NCT02499835
CPI + other
Nivolumab + TPST-1120 (PPARα antagonist); docetaxel + TPST- I Recruiting NCT03829436
1120; cetuximab + TPST-1120; TPST-1120
Nivolumab + NeoTCR-P1 (adoptive cell therapy); NeoTCR-P1 I Recruiting NCT03970382
Nivolumab + testosterone cypionate II Recruiting NCT03554317
Pembrolizumab + enzalutamide + fecal microbiota transplant II Recruiting NCT04116775
Pembrolizumab + HER2Bi-armed ATCs II Recruiting NCT03406858
Ipilimumab + sargramostim I Active, not recruiting NCT00064129
Pembrolizumab + GB1275 (CD11b agonist); GB1275 I/II Recruiting NCT04060342
Atezolizumab + sipuleucel-T I Active, not recruiting NCT03024216
Atezolizumab + ipatasertib I/II Recruiting NCT03673787
Atezolizumab + cabozantinib; atezolizumab; cabozantinib I/II Recruiting NCT03170960
Durvalumab + naptumomab (5T4-targeted immunotoxin) I Recruiting NCT03983954
Cemiplimab + REGN5678 (anti-PSMA/CD28 bispecific antibody) I/II Recruiting NCT03972657
Cemiplimab + isatuximab (anti-CD38 antibody) I/II Active, not recruiting NCT03367819

de Almeida et al
Ipilimumab + sipuleucel-T II Active, not recruiting NCT01804465
Avelumab + cyclophosphamide + TRX518 (anti-GITR antibody) I/II Active, not recruiting NCT03861403
Avelumab + utomilumab (4-1BB antibody); avelumab + PF- I/II Recruiting NCT03217747
04518600 (anti-OX40 antibody); avelumab + utomilumab + PF-
04518600; avelumab + utomilumab + RT; avelumab + PF-
04518600 + RT; avelumab + utomilumab + PF-04518600 + RT
Durvalumab + tremelimumab + poly-ICLC I/II Recruiting NCT02643303
Abbreviations: CPI, checkpoint inhibitor; ADT, androgen deprivation therapy; EBRT, external beam radiation therapy; PROSTVAC-V/F, rilimogene
galvacirepvec and rilimogene glafolivec; MVA-BN, modified vaccine Ankara–Bavarian Nordic; ChAdOx1-MVA, chimpanzee adenovirus Oxford
University #1 and modified vaccine Ankara; 5T4, trophoblast glycoprotein; pTVG-HP, cancer vaccine containing plasmid DNA encoding human
prostatic acid phosphatase; GM-CSF, granulocyte-macrophage colony-stimulating factor; TDS-IM, TriGrid Delivery System for intramuscular
delivery; SBRT, stereotactic body radiation therapy; TLR, Toll-like receptor; mCRPC, metastatic castration-resistant prostate cancer; LAG-3,
lymphocyte-activation gene 3; CD73, ecto-5 0 -nucleotidase; A2AR, adenosine A2A receptor; CD3, cluster of differentiation 3; IL, interleukin; BiTE,
bispecific T cell engager; PSMA, prostate-specific membrane antigen; CXCR, CXC chemokine receptor; Flt3L, FMS-like tyrosine kinase 3 ligand;
CDX-301, recombinant human FMS-like tyrosine kinase-3 ligand (Flt3L); poly-ICLC, polyinosinic-polycytidylic acid, and poly-L-lysine; INO-5151,
combined formulation antigens encoding for PSA and PSMA and DNA vector expressing IL-12; Lm, L. monocytogenes, LLO, Listeriolysin O; rhGM-
CSF, recombinant human GM-CSF; pTVG-AR, cancer vaccine containing pTVG4 plasmid DNA encoding the human AR ligand-binding domain;
PPARα, peroxisome proliferator-activated receptor α; HER2Bi-armed ATCs, anti-CD3/Her2 bispecific antibody-armed activated T cells; CD11b,
integrin αM; CD28, cluster of differentiation 28; CD38, cluster of differentiation 38; GITR, glucocorticoid-induced TNFR-related protein; 4-1BB,
tumor necrosis factor receptor superfamily member 9; OX40, tumor necrosis factor receptor superfamily member 4; RT, radiation therapy; PSA,
prostate-specific antigen; ATC, activated T cell; MSI-H, microsatellite instability high; dMMR, mismatch repair deficiency; DDR, DNA damage
repair.
a
Restricted to patients harboring MSI-H status, dMMR, TMB of 20 or more mutations/megabyte, or a CDK12 alteration.
b
Restricted to patients with a DDR alteration.
c
All study regimens include ADT.
Restricted to patients with a CDK12 alteration.
d
e
Includes one cohort is restricted to patients harboring one of the following: dMMR, high TMB without MMR defects, DDR, deleterious nucleotide excision
repair aberrations, deleterious base excision repair aberrations, or deleterious aberrations in nonhomologous end-joining.
Restricted to patients with dMMR or a CDK12 alteration.
f
g
Restricted to patients with androgen receptor variant 7.
h
Restricted to patients with dMMR, DDR, or inflammatory infiltrate by immunohistochemistry.
i
Includes one cohort restricted to patients who were PD-L1+.
j
Includes one cohort restricted to patients with dMMR.
cancer is underway (Fig. 2). It has been hypothesized immune system are highly relevant and dynamic and re-
that the immune system may be more susceptible to quire further study. In addition to having a better un-
modulation earlier in the disease course, such as in derstanding of which clinical state may be optimal for
the localized, biochemically recurrent or metastatic a given therapy, there is a lack of understanding of
castration-sensitive setting. Tumor burden also may the challenges that may be inherent to a bone-tropic
have a complex role; greater tumor heterogeneity in the disease and the bone-metastatic niche that require
later stages of disease may yield increased potential for investigation.
the creation of neoantigens. Patients with biochemically CONCLUSIONS
recurrent prostate cancer may have a lower burden of
Although CPIs have transformed the therapeutic land-
disease and possibly fewer MDSCs and immune sup-
scape for many tumors, their role in prostate cancer as
pressive cells, as well as less tumor heterogeneity. Many a monotherapy is very limited, with FDA approval for pa-
trials are underway in the neoadjuvant setting as window- tients with MSI-H status or dMMR tumors only. Insights
of-opportunity studies that will allow for interrogation of from other malignancies that have a “cold” microenvi-
tissue samples before and after treatment. It is possible ronment and that are intrinsically resistant to CPIs may be
that immunotherapy combinations with different mecha- applicable to prostate cancer. However, understanding
nisms of action may be better suited to one clinical state what is unique about prostate cancer from an immune
over another. As previously reviewed, the effects of an- perspective—the bone-metastatic niche; hormone sensi-
drogen deprivation and castration resistance on the tivity and the use of ADT; the presence of common tumor

FIGURE 2. Breakdown of Current Clinical Trials Using CPIs by Clinical State and Phase in Development
Abbreviations: BCR, biochemically recurrent prostate cancer; mCRPC, metastatic castration-resistant prostate cancer; mCSPC, metastatic castration-
sensitive prostate cancer.
antigens, such as PSMA; and the well-defined clinical states— ACKNOWLEDGMENT

will aid in the design of combination immunotherapies that We thank Sara DiNapoli (Memorial Sloan Kettering Cancer
just may take CPIs from niche to breakthrough. Center) for editorial support.

1
Department of Medicine, Genitourinary Oncology Service, Memorial Karen A. Autio, MD, MSc, Memorial Sloan Kettering Cancer Center, 353
Sloan Kettering Cancer Center, New York, NY East 68th St., New York, NY 10065; email: [email protected].
2
Medical Oncology Department, Beneficencia Portuguesa de Sao Paulo,
Sao Paulo, SP, Brazil
3
Division of Hematology and Oncology, Department of Medicine,
University of California, San Francisco, San Francisco, CA AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
4
Departments of Medicine and Urology, University of California, Los AND DATA AVAILABILITY STATEMENT
Angeles, CA Disclosures provided by the authors and data availability statement (if
5
VA Greater Los Angeles Healthcare System, Los Angeles, CA applicable) are available with this article at DOI https://doi.org/10.1200/
EDBK_278853.
REFERENCES
1. Sweeney CJ, Chen Y-H, Carducci M, et al. Chemohormonal therapy in metastatic hormone-sensitive prostate cancer. N Engl J Med. 2015;373:737-746.
2. James ND, Sydes MR, Clarke NW, et al; STAMPEDE Investigators. Addition of docetaxel, zoledronic acid, or both to first-line long-term hormone therapy in
2017;377:338-351.
377:352-360.
6. Davis ID, Martin AJ, Stockler MR, et al; ENZAMET Trial Investigators and the Australian and New Zealand Urogenital and Prostate Cancer Trials Group.
Enzalutamide with standard first-line therapy in metastatic prostate cancer. N Engl J Med. 2019;381:121-131.
7. Slovin SF, Higano CS, Hamid O, et al. Ipilimumab alone or in combination with radiotherapy in metastatic castration-resistant prostate cancer: results from an
open-label, multicenter phase I/II study. Ann Oncol. 2013;24:1813-1821.

de Almeida et al
8. Kwon ED, Drake CG, Scher HI, et al; CA184-043 Investigators. Ipilimumab versus placebo after radiotherapy in patients with metastatic castration-resistant
prostate cancer that had progressed after docetaxel chemotherapy (CA184-043): a multicentre, randomised, double-blind, phase 3 trial. Lancet Oncol. 2014;
15:700-712.
9. Beer TM, Kwon ED, Drake CG, et al. Randomized, double-blind, phase III trial of ipilimumab versus placebo in asymptomatic or minimally symptomatic patients
with metastatic chemotherapy-naive castration-resistant prostate cancer. J Clin Oncol. 2017;35:40-47.
10. Jayaprakash P, Ai M, Liu A, et al. Targeted hypoxia reduction restores T cell infiltration and sensitizes prostate cancer to immunotherapy. J Clin Invest. 2018;
128:5137-5149.
11. Fukumura D, Kloepper J, Amoozgar Z, et al. Enhancing cancer immunotherapy using antiangiogenics: opportunities and challenges. Nat Rev Clin Oncol. 2018;
15:325-340.
12. Crusz SM, Balkwill FR. Inflammation and cancer: advances and new agents. Nat Rev Clin Oncol. 2015;12:584-596.
14. Gooden MJ, de Bock GH, Leffers N, et al. The prognostic influence of tumour-infiltrating lymphocytes in cancer: a systematic review with meta-analysis.
Br J Cancer. 2011;105:93-103.
15. Fridman WH, Zitvogel L, Sautès-Fridman C, et al. The immune contexture in cancer prognosis and treatment. Nat Rev Clin Oncol. 2017;14:717-734.
16. Gibney GT, Weiner LM, Atkins MB. Predictive biomarkers for checkpoint inhibitor-based immunotherapy. Lancet Oncol. 2016;17:e542-e551.
17. Miller AM, Lundberg K, Ozenci V, et al. CD4+CD25high T cells are enriched in the tumor and peripheral blood of prostate cancer patients. J Immunol. 2006;
177:7398-7405.
18. Kiniwa Y, Miyahara Y, Wang HY, et al. CD8+ Foxp3+ regulatory T cells mediate immunosuppression in prostate cancer. Clin Cancer Res. 2007;13:6947-6958.
19. Zhao SG, Lehrer J, Chang SL, et al. The immune landscape of prostate cancer and nomination of PD-l2 as a potential therapeutic target. J Natl Cancer Inst.
2019;111:301-310.
20. Lundholm M, Hagglof C, Wikberg ML, et al. Secreted factors from colorectal and prostate cancer cells skew the immune response in opposite directions. Sci
Rep. 2015;5:15651.
21. Nagarsheth N, Wicha MS, Zou W. Chemokines in the cancer microenvironment and their relevance in cancer immunotherapy. Nat Rev Immunol. 2017;
17:559-572.
22. Rani A, Dasgupta P, Murphy JJ. Prostate cancer: the role of inflammation and chemokines. Am J Pathol. 2019;189:2119-2137.
23. Sutherland JS, Goldberg GL, Hammett MV, et al. Activation of thymic regeneration in mice and humans following androgen blockade. J Immunol. 2005;
175:2741-2753.
24. Mercader M, Bodner BK, Moser MT, et al. T cell infiltration of the prostate induced by androgen withdrawal in patients with prostate cancer. Proc Natl Acad Sci
USA. 2001;98:14565-14570.
25. Gannon PO, Poisson AO, Delvoye N, et al. Characterization of the intra-prostatic immune cell infiltration in androgen-deprived prostate cancer patients.
J Immunol Methods. 2009;348:9-17.
26. Sorrentino C, Musiani P, Pompa P, et al. Androgen deprivation boosts prostatic infiltration of cytotoxic and regulatory T lymphocytes and has no effect on
disease-free survival in prostate cancer patients. Clin Cancer Res. 2011;17:1571-1581.
27. Lopez-Bujanda ZA, Haffner MC, Chaimowitz MG, et al. Castration-mediated IL-8 promotes myeloid infiltration and prostate cancer progression. https://www.
biorxiv.org/content/10.1101/651083v2. Accessed March 11, 2020.
28. Hurwitz AA, Foster BA, Kwon ED, et al. Combination immunotherapy of primary prostate cancer in a transgenic mouse model using CTLA-4 blockade. Cancer
Res. 2000;60:2444-2448.
29. Kaur P, Asea A. Radiation-induced effects and the immune system in cancer. Front Oncol. 2012;2:191.
30. Postow MA, Callahan MK, Barker CA, et al. Immunologic correlates of the abscopal effect in a patient with melanoma. N Engl J Med. 2012;366:925-931.
31. Topalian SL, Hodi FS, Brahmer JR, et al. Safety, activity, and immune correlates of anti–PD-1 antibody in cancer. N Engl J Med. 2012;366:2443-2454.
32. Bishop JL, Sio A, Angeles A, et al. PD-L1 is highly expressed in enzalutamide resistant prostate cancer. Oncotarget. 2015;6:234-242.
33. Graff JN, Alumkal JJ, Drake CG, et al. Early evidence of anti–PD-1 activity in enzalutamide-resistant prostate cancer. Oncotarget. 2016;7:52810-52817.
34. Graff JN, Moran AE, Slottke RE, et al. 848PD - Phase II study of pembrolizumab with enzalutamide (Enz) in metastatic, castration-resistant prostate cancer
(mCRPC): 30 patient expansion with examination of tumour-infiltrating immune cells and fecal microbiota. Ann Oncol. 2019;30 (suppl 5):v329.
35. Roche. Pipeline Summary. https://www.roche.com/dam/jcr:9fb4878c-e3a2-4fdc-80f6-62d435df7367/en/pharmahy19.pdf. Accessed March 10, 2020.
36. Hansen AR, Massard C, Ott PA, et al. Pembrolizumab for advanced prostate adenocarcinoma: findings of the KEYNOTE-028 study. Ann Oncol. 2018;
29:1807-1813.
37. Antonarakis ES, Piulats JM, Gross-Goupil M, et al. Pembrolizumab for treatment-refractory metastatic castration-resistant prostate cancer: multicohort, open-
label phase II KEYNOTE-199 study. J Clin Oncol. 2020;38:395-405.
38. Selby M, Engelhardt J, Lu L-S, et al. Antitumor activity of concurrent blockade of immune checkpoint molecules CTLA-4 and PD-1 in preclinical models. J Clin
Oncol. 2013;31 (suppl; abstr 3061).
39. Wolchok JD, Kluger H, Callahan MK, et al. Nivolumab plus ipilimumab in advanced melanoma. N Engl J Med. 2013;369:122-133.

40. Antonia SJ, López-Martin JA, Bendell J, et al. Nivolumab alone and nivolumab plus ipilimumab in recurrent small-cell lung cancer (CheckMate 032):
a multicentre, open-label, phase 1/2 trial. Lancet Oncol. 2016;17:883-895.
41. Hellmann MD, Gettinger SN, Goldman JW, et al. CheckMate 012: safety and efficacy of first-line (1L) nivolumab (nivo; N) and ipilimumab (ipi; I) in advanced
(adv) NSCLC. J Clin Oncol. 2016;34 (suppl; abstr 3001).
42. Hammers HJ, Plimack ER, Infante JR, et al. Safety and efficacy of nivolumab in combination with ipilimumab in metastatic renal cell carcinoma: the CheckMate
016 study. J Clin Oncol. 2017;35:3851-3858.
43. Sharma P, Pachynski RK, Narayan V, et al. Initial results from a phase II study of nivolumab (NIVO) plus ipilimumab (IPI) for the treatment of metastatic
castration-resistant prostate cancer (mCRPC; CheckMate 650). J Clin Oncol. 2019;37 (suppl; abstr 142).
16:275-287.
45. Ott PA, Bang Y-J, Piha-Paul SA, et al. T-cell–inflamed gene-expression profile, programmed death ligand 1 expression, and tumor mutational burden predict
efficacy in patients treated with pembrolizumab across 20 cancers: KEYNOTE-028. J. Clin. Oncol. 2019;37:318-327.
46. Nishino M, Ramaiya NH, Hatabu H, et al. Monitoring immune-checkpoint blockade: response evaluation and biomarker development. Nat Rev Clin Oncol.
2017;14:655-668.
16:275-287.
48. Lu S, Stein JE, Rimm DL, et al. Comparison of biomarker modalities for predicting response to PD-1/PD-L1 checkpoint blockade. JAMA Oncol. 2019;5:1195.
49. Davis AA, Patel VG. The role of PD-L1 expression as a predictive biomarker: an analysis of all U.S. Food and Drug Administration (FDA) approvals of immune
2019;25:3753-3758.
51. Abida W, Cheng ML, Armenia J, et al. Analysis of the prevalence of microsatellite instability in prostate cancer and response to immune checkpoint blockade.
JAMA Oncol. 2019;5:471-478.
52. Robinson D, Van Allen EM, Wu Y-M, et al. Integrative clinical genomics of advanced prostate cancer [published correction appears in Cell. 2015;162:454]. Cell.
2015;161:1215-1228.
53. Nava Rodrigues D, Rescigno P, Liu D, et al. Immunogenomic analyses associate immunological alterations with mismatch repair defects in prostate cancer.
J Clin Invest. 2018;128:4441-4453.
54. Antonarakis ES, Shaukat F, Isaacsson Velho P, et al. Clinical features and therapeutic outcomes in men with advanced prostate cancer and DNA mismatch
repair gene mutations. Eur Urol. 2019;75:378-382.
55. Pritchard CC, Morrissey C, Kumar A, et al. Complex MSH2 and MSH6 mutations in hypermutated microsatellite unstable advanced prostate cancer. Nat
Commun. 2014;5:4988.
56. Wang F, Zhao Q, Wang Y-N, et al. Evaluation of POLE and POLD1 mutations as biomarkers for immunotherapy outcomes across multiple cancer types. JAMA
Oncol. 2019;5:1504.
57. Campbell BB, Light N, Fabrizio D, et al. Comprehensive analysis of hypermutation in human cancer. Cell. 2017;171:1042-1056.e10.
58. Lee L, Ali S, Genega E, et al. Aggressive-variant microsatellite-stable POLE mutant prostate cancer with high mutation burden and durable response to immune
checkpoint inhibitor therapy. JCO Precis Oncol. 2018;2:1-8.
59. Wu Y-M, Cieślik M, Lonigro RJ, et al; PCF/SU2C International Prostate Cancer Dream Team. Inactivation of CDK12 delineates a distinct immunogenic class of
advanced prostate cancer. Cell. 2018;173:1770-1782.e14.
60. Antonarakis ES, Velho PI, Agarwal N, et al. 845PD - CDK12-altered prostate cancer: clinical features and therapeutic outcomes to standard systemic therapies,
PARP inhibitors, and PD1 inhibitors. Ann Oncol. 2019;30 (suppl 5):v327.
61. Samstein RM, Lee C-H, Shoushtari AN, et al. Tumor mutational load predicts survival after immunotherapy across multiple cancer types. Nat Genet. 2019;
51:202-206.
62. Mehra N, van Riet J, Smits M, et al. 798PD - In-depth assessment of metastatic prostate cancer with high tumour mutational burden. Ann Oncol. 2018;29 (suppl
8):viii274.
63. Chalmers ZR, Connelly CF, Fabrizio D, et al. Analysis of 100,000 human cancer genomes reveals the landscape of tumor mutational burden. Genome Med.
2017;9:34.
64. Fancello L, Gandini S, Pelicci PG, et al. Tumor mutational burden quantification from targeted gene panels: major advancements and challenges. J Immunother
Cancer. 2019;7.
65. Chan TA, Yarchoan M, Jaffee E, et al. Development of tumor mutation burden as an immunotherapy biomarker: utility for the oncology clinic. Ann Oncol. 2019;
30:44-56.
66. Büttner R, Longshore JW, López-Rı́os F, et al. Implementing TMB measurement in clinical practice: considerations on assay requirements. ESMO Open. 2019;
4:e000442.
67. Lang S, Swift S, White H, et al. A systematic review of the prevalence of DNA damage response gene mutations in prostate cancer. Int J Oncol. 2019;
55:597-616.

de Almeida et al
68. Karzai F, VanderWeele D, Madan RA, et al. Activity of durvalumab plus olaparib in metastatic castration-resistant prostate cancer in men with and without DNA
damage repair mutations. J Immunother Cancer. 2018;6:141.
69. Demaria S, Golden EB, Formenti SC. Role of local radiation therapy in cancer immunotherapy. JAMA Oncol. 2015;1:1325-1332.
70. Dewan MZ, Galloway AE, Kawashima N, et al. Fractionated but not single-dose radiotherapy induces an immune-mediated abscopal effect when combined with
anti-CTLA-4 antibody. Clin Cancer Res. 2009;15:5379-5388.
71. Schaue D, Ratikan JA, Iwamoto KS, et al. Maximizing tumor immunity with fractionated radiation. Int J Radiat Oncol Biol Phys. 2012;83:1306-1310.
72. Bang A, Wilhite TJ, Pike LRG, et al. Multicenter evaluation of the tolerability of combined treatment with PD-1 and CTLA-4 immune checkpoint inhibitors and
palliative radiation therapy. Int J Radiat Oncol Biol Phys. 2017;98:344-351.
73. Welsh JW, Tang C, de Groot P, et al. Phase II trial of ipilimumab with stereotactic radiation therapy for metastatic disease: outcomes, toxicities, and low-dose
radiation-related abscopal responses. Cancer Immunol Res. 2019;7:1903-1909.
J Med. 2018;379:2220-2229.
J Med. 2018;379:2108-2121.
76. West H, McCleod M, Hussein M, et al. Atezolizumab in combination with carboplatin plus nab-paclitaxel chemotherapy compared with chemotherapy alone as
first-line treatment for metastatic non-squamous non-small-cell lung cancer (IMpower130): a multicentre, randomised, open-label, phase 3 trial. Lancet Oncol.
2019;20:924-937.
77. Gandhi L, Rodrı́guez-Abreu D, Gadgeel S, et al; KEYNOTE-189 Investigators. Pembrolizumab plus chemotherapy in metastatic non–small-cell lung cancer.
N Engl J Med. 2018;378:2078-2092.
78. Dosset M, Vargas TR, Lagrange A, et al. PD-1/PD-L1 pathway: an adaptive immune resistance mechanism to immunogenic chemotherapy in colorectal cancer.
Oncoimmunology. 2018;7:e1433981.
79. Alizadeh D, Trad M, Hanke NT, et al. Doxorubicin eliminates myeloid-derived suppressor cells and enhances the efficacy of adoptive T-cell transfer in breast
cancer. Cancer Res. 2014;74:104-118.
80. Wang Z, Till B, Gao Q. Chemotherapeutic agent-mediated elimination of myeloid-derived suppressor cells. Oncoimmunology. 2017;6:e1331807.
81. Kodumudi KN, Woan K, Gilvary DL, et al. A novel chemoimmunomodulating property of docetaxel: suppression of myeloid-derived suppressor cells in tumor
bearers. Clin Cancer Res. 2010;16:4583-4594.
82. Garnett CT, Schlom J, Hodge JW. Combination of docetaxel and recombinant vaccine enhances T-cell responses and antitumor activity: effects of docetaxel on
immune enhancement. Clin Cancer Res. 2008;14:3536-3544.
83. Vicari AP, Luu R, Zhang N, et al. Paclitaxel reduces regulatory T cell numbers and inhibitory function and enhances the anti-tumor effects of the TLR9 agonist
PF-3512676 in the mouse. Cancer Immunol Immunother. 2009;58:615-628.
84. Jiao S, Xia W, Yamaguchi H, et al. PARP inhibitor upregulates PD-L1 expression and enhances cancer-associated immunosuppression. Clin Cancer Res. 2017;
23:3711-3720.
85. Kruczek K, Ratterman M, Tolzien K, et al. A phase II study evaluating the toxicity and efficacy of single-agent temsirolimus in chemotherapy-naı̈ve castration-
resistant prostate cancer. Br J Cancer. 2013;109:1711-1716.
86. Pantelidou C, Sonzogni O, De Oliveria Taveira M, et al. PARP inhibitor efficacy depends on CD8+ T-cell recruitment via intratumoral STING pathway activation in
BRCA-deficient models of triple-negative breast cancer. Cancer Discov. 2019;9:722-737.
87. Huang Y, Yuan J, Righi E, et al. Vascular normalizing doses of antiangiogenic treatment reprogram the immunosuppressive tumor microenvironment and
enhance immunotherapy. Proc Natl Acad Sci U S A. 2012;109:17561-17566.
88. Makker V, Rasco D, Vogelzang NJ, et al. Lenvatinib plus pembrolizumab in patients with advanced endometrial cancer: an interim analysis of a multicentre,
open-label, single-arm, phase 2 trial. Lancet Oncol. 2019;20:711-718.
89. Rini BI, Plimack ER, Stus V, et al. KEYNOTE-426 Investigators. Pembrolizumab plus axitinib versus sunitinib for advanced renal-cell carcinoma. N Engl J Med.
2019;380:1116-1127.
91. Agarwal N, Loriot Y, McGregor BA, et al. Cabozantinib (C) in combination with atezolizumab (A) in patients (pts) with metastatic castration-resistant prostate
cancer (mCRPC): results of cohort 6 of the COSMIC-021 study. J Clin Oncol. 2020;38 (suppl; abstr 139).
92. Waugh DJJ, Wilson C. The interleukin-8 pathway in cancer. Clin Cancer Res. 2008;14:6735-6741.
93. Sharma J, Gray KP, Harshman LC, et al. Elevated IL-8, TNF-α, and MCP-1 in men with metastatic prostate cancer starting androgen-deprivation therapy (ADT)
are associated with shorter time to castration-resistance and overall survival. Prostate. 2014;74:820-828.
94. Batlle E, Massagué J. Transforming growth factor-β signaling in immunity and cancer. Immunity. 2019;50:924-940.
95. Jiao S, Subudhi SK, Aparicio A, et al. Differences in tumor microenvironment dictate T helper lineage polarization and response to immune checkpoint therapy.
Cell. 2019;179:1177-1190.e13.
96. Autio KA, Klebanoff CA, Schaer D, et al. Phase 1 study of LY3022855, a colony-stimulating factor-1 receptor (CSF-1R) inhibitor, in patients with metastatic
breast cancer (MBC) or metastatic castration-resistant prostate cancer (MCRPC). J Clin Oncol. 2019;37 (suppl; abstr 2548).

97. Gomez-Roca CA, Italiano A, Le Tourneau C, et al. Phase I study of emactuzumab single agent or in combination with paclitaxel in patients with advanced/
metastatic solid tumors reveals depletion of immunosuppressive M2-like macrophages. Ann Oncol. 2019;30:1381-1392.
98. Leclerc BG, Charlebois R, Chouinard G, et al. CD73 Expression is an independent prognostic factor in prostate cancer. Clin Cancer Res. 2016;22:158-166.
99. Sitkovsky MV, Hatfield S, Abbott R, et al. Hostile, hypoxia-A2-adenosinergic tumor biology as the next barrier to overcome for tumor immunologists. Cancer
Immunol Res. 2014;2:598-605.
100. Li J, Wang L, Chen X, et al. CD39/CD73 upregulation on myeloid-derived suppressor cells via TGF-β-mTOR-HIF-1 signaling in patients with non–small cell lung
cancer. OncoImmunology. 2017;6:e1320011.
101. Leone RD, Emens LA. Targeting adenosine for cancer immunotherapy. J Immunother Cancer. 2018;6:57.
102. Yu M, Guo G, Huang L, et al. CD73 on cancer-associated fibroblasts enhanced by the A2B-mediated feedforward circuit enforces an immune checkpoint. Nat
Commun. 2020;11:515.
103. Bendell J, Bauer T, Patel M, et al. Evidence of immune activation in the first-in-human phase Ia dose escalation study of the adenosine 2a receptor antagonist,
AZD4635, in patients with advanced solid tumors. In: Proceedings: AACR Annual Meeting 2019. Atlanta, GA: American Association for Cancer Research; 2019.
Abstract CT026.
104. Harshman LC, Chu M, George S, et al. Adenosine receptor blockade with ciforadenant +/ atezolizumab in advanced metastatic castration-resistant prostate
cancer (mCRPC). J Clin Oncol. 2020;38 (suppl; abstr 129).
105. Yap TA, Burris HA, Kummar S, et al. ICONIC: biologic and clinical activity of first in class ICOS agonist antibody JTX-2011 +/CONIC: biologic and clinical activity
of first in class IC. J Clin Oncol. 2018;36 (suppl; abstr 3000).
106. Papadopoulos KP, Autio KA, Golan T, et al. Phase 1 study of MK-4166, an anti-human glucocorticoid-induced tumor necrosis factor receptor (GITR) antibody,
as monotherapy or with pembrolizumab (pembro) in patients (pts) with advanced solid tumors. J Clin Oncol. 2019;37 (suppl; abstr 9509).
107. Hansen AR, Infante JR, McArthur G, et al. Abstract CT097: A first-in-human phase I dose escalation study of the OX40 agonist MOXR0916 in patients with
refractory solid tumors. Cancer Res. 2016:76 (suppl; abstr CT097).
108. Vonderheide RH. Abstract I12: CD40 immunotherapy for pancreatic cancer. Cancer Res. 2019;79 (suppl; abstr I12).
109. Helm O, Mennrich R, Petrick D, et al. Comparative characterization of stroma cells and ductal epithelium in chronic pancreatitis and pancreatic ductal
adenocarcinoma. PLoS One. 2014;9:e94357.
110. Comito G, Giannoni E, Segura CP, et al. Cancer-associated fibroblasts and M2-polarized macrophages synergize during prostate carcinoma progression.
Oncogene. 2014;33:2423-2431.
111. Cioni B, Nevedomskaya E, Melis MHM, et al. Loss of androgen receptor signaling in prostate cancer-associated fibroblasts (CAFs) promotes CCL2- and CXCL8-
mediated cancer cell migration. Mol Oncol. 2018;12:1308-1323.
112. Jachetti E, Cancila V, Rigoni A, et al. Cross-talk between myeloid-derived suppressor cells and mast cells mediates tumor-specific immunosuppression in
prostate cancer. Cancer Immunol Res. 2018;6:552-565.
113. Gao J, Ward JF, Pettaway CA, et al. VISTA is an inhibitory immune checkpoint that is increased after ipilimumab therapy in patients with prostate cancer. Nat
Med. 2017;23:551-555.
114. Zahm CD, Colluru VT, McNeel DG. DNA vaccines for prostate cancer. Pharmacol Ther. 2017;174:27-42.
115. Colluru VT, Johnson LE, Olson BM, et al. Preclinical and clinical development of DNA vaccines for prostate cancer. Urol Oncol. 2016;34:193-204.
116. Sartor AO, Armstrong AJ, Ahaghotu C, et al. Overall survival (OS) of African-American (AA) and Caucasian (CAU) men who received sipuleucel-T for metastatic
castration-resistant prostate cancer (mCRPC): Final PROCEED analysis. J Clin Oncol. 2019;37:15s (suppl; abstr 5035).
117. Small E. A phase III trial of GVAX immunotherapy for prostate cancer in combination with docetaxel versus docetaxel plus prednisone in symptomatic,
castration-resistant prostate cancer (CRPC). In: Proceedings of the 2009 Genitourinary Cancer Symposium. Orlando, FL: American Society of Clinical Oncology;
2009.
118. Higano C. A phase III trial of GVAX immunotherapy for prostate cancer versus docetaxel plus prednisone in asymptomatic, castration-resistant prostate cancer
(CRPC). In: Proceedings of the 2009 Genitourinary Cancer Symposium. Orlando, FL: American Society of Clinical Oncology; 2009.
119. Gulley JL, Borre M, Vogelzang NJ, et al. Phase III trial of PROSTVAC in asymptomatic or minimally symptomatic metastatic castration-resistant prostate cancer.
J Clin Oncol. 2019;37:1051-1061.
120. Comiskey MC, Dallos MC, Drake CG. Immunotherapy in prostate cancer: teaching an old dog new tricks. Curr Oncol Rep. 2018;20:75.
121. Bilusic M, Madan RA, Gulley JL. Immunotherapy of prostate cancer: facts and hopes. Clin Cancer Res. 2017;23:6764-6770.
122. McNeel DG, Eickhoff JC, Wargowski E, et al. Concurrent, but not sequential, PD-1 blockade with a DNA vaccine elicits anti-tumor responses in patients with
metastatic, castration-resistant prostate cancer. Oncotarget. 2018;9:25586-25596.
123. Haas NB, Stein MN, Tutrone R, et al. Phase I-II study of ADXS31-142 alone and in combination with pembrolizumab in patients with previously treated
metastatic castration-resistant prostate cancer (mCRPC): the KEYNOTE-046 trial. J Immunother Cancer. 2015;3 (suppl 2):153.
124. Friedrich M, Raum T, Lutterbuese R, et al. Regression of human prostate cancer xenografts in mice by AMG 212/BAY2010112, a novel PSMA/CD3-Bispecific
BiTE antibody cross-reactive with non-human primate antigens. Mol Cancer Ther. 2012;11:2664-2673.
125. Hummel H-D, Kufer P, Grüllich C, et al. Phase 1 study of pasotuxizumab (BAY 2010112), a PSMA-targeting Bispecific T cell Engager (BiTE) immunotherapy for
metastatic castration-resistant prostate cancer (mCRPC). J Clin Oncol. 2019;37 (suppl; abstr 5034).

de Almeida et al
126. Porter DL, Levine BL, Kalos M, et al. Chimeric antigen receptor-modified T cells in chronic lymphoid leukemia. N Engl J Med. 2011;365:725-733.
127. Porter DL, Hwang W-T, Frey NV, et al. Chimeric antigen receptor T cells persist and induce sustained remissions in relapsed refractory chronic lymphocytic
leukemia. Sci Transl Med. 2015;7:303ra139.
128. Maude SL, Frey N, Shaw PA, et al. Chimeric antigen receptor T cells for sustained remissions in leukemia. N Engl J Med. 2014;371:1507-1517.
130. Park JH, Rivière I, Gonen M, et al. Long-term follow-up of CD19 CAR therapy in acute lymphoblastic leukemia. N Engl J Med. 2018;378:449-459.
131. Neelapu SS, Locke FL, Bartlett NL, et al. Axicabtagene ciloleucel CAR T-cell therapy in refractory large B-cell lymphoma. N Engl J Med. 2017;377:2531-2544.
132. Schuster SJ, Svoboda J, Chong EA, et al. Chimeric antigen receptor T cells in refractory B-cell lymphomas. N Engl J Med. 2017;377:2545-2554.
133. Kiessling A, Wehner R, Füssel S, et al. Tumor-associated antigens for specific immunotherapy of prostate cancer. Cancers (Basel). 2012;4:193-217.
134. Junghans RP, Ma Q, Rathore R, et al. Phase I trial of anti-PSMA designer CAR-T cells in prostate cancer: possible role for interacting interleukin 2-T cell
pharmacodynamics as a determinant of clinical response. Prostate. 2016;76:1257-1270.
135. Slovin SF, Wang X, Hullings M, et al. Chimeric antigen receptor (CAR+) modified T cells targeting prostate-specific membrane antigen (PSMA) in patients (pts)
with castrate metastatic prostate cancer (CMPC). J Clin Oncol. 2013;31:6s (suppl; abstr 72).

Recent Advances in the Management of High-Risk

Localized Prostate Cancer: Local Therapy,
Systemic Therapy, and Biomarkers to Guide
Treatment Decisions
Rana R. McKay, MD1; Felix Y. Feng, MD2; Alice Y. Wang, MD3; Christopher J. D. Wallis, MD3; and Kelvin A. Moses, MD, PhD3
overview
High-risk prostate cancer accounts for approximately 15% of all prostate cancer diagnoses. Patients with
high-risk disease have an increased risk of developing biochemical recurrence, metastases, and death from
prostate cancer. As the optimal management of high-risk disease in patients with prostate cancer continues to
evolve, the contemporary treatment paradigm is moving toward a multidisciplinary integrated approach of
systemic and local therapy for patients with high-risk disease. The strategies for definitive, adjuvant, and
salvage local treatment, including radical prostatectomy or radiation therapy, serve as the backbone of therapy
for patients with localized disease. Systemic therapy decisions regarding use in combination with surgery,
choice of therapy (hormone therapy, chemotherapy), and treatment duration continue to be refined. As more
effective hormonal agents populate the treatment landscape for advanced prostate cancer, including abir-
aterone and next-generation antiandrogens, an opportunity is provided to explore these treatments in patients
with localized disease in the hope of improving the long-term outcome for patients. Integration of innovative
blood and tissue-based biomarkers to guide therapy selection for patients with high-risk disease is an area of
active research. Contemporary studies are using such biomarkers to stratify patients and select therapies. In
this review, we summarize contemporary evidence for local treatment strategies, systemic therapy options,
and biomarkers in development for the management of high-risk prostate cancer in patients.
INTRODUCTION high-risk localized prostate cancer has not been estab-

Prostate cancer is the second most common malig- lished. Additionally, biomarkers, beyond clinicopatho-
nancy among men worldwide, and most patients in logic features, have not been routinely used in practice to
Western countries (greater than 80%) present with guide clinical decision-making and are still under de-
velopment. In this article, we aim to summarize the current
localized disease (T1-4N0M0).1 Approximately 15% of
evidence for (1) local treatment strategies of patients with
individuals with localized disease are identified as at
high-risk localized disease; (2) use of neoadjuvant and
high risk for disease recurrence.2,3 Treatment options
adjuvant systemic therapy, including a discussion on
have historically included surgery, radiation therapy,
novel therapeutics for disease control; and (3) integration
and/or androgen deprivation therapy (ADT). However,
of innovative blood- and tissue-based biomarkers to guide
despite treatment, a subset of men with localized
therapy selection for patients with high-risk disease.
disease develop recurrent lethal prostate cancer. Given
the potential for cure of patients with localized disease, DEFINING HIGH-RISK DISEASE
Author affiliations effective treatment strategies are crucial to improving There are multiple definitions used to categorize in-
and support long-term outcomes for patients at this critical stage. dividuals with high-risk prostate cancer. Pretreatment
information (if Multimodal treatment strategies that integrate local
applicable) appear
parameters, including clinical stage, prostate-specific
and systemic treatments have the potential to de- antigen (PSA), and Gleason score, are established
at the end of this
article. crease recurrence rates and prolong survival. Such ap- predictors of disease recurrence and have historically
Accepted on April proaches that integrate neoadjuvant and/or adjuvant been used in high-risk disease classifications.4 In
24, 2020 and systemic therapy, including chemotherapy and/or 1998, using an endpoint of PSA recurrence, D’Amico
published at hormonal therapy, with localized treatment, including et al2 defined high-risk disease as a clinical T stage of at
ascopubs.org on May
surgery and/or radiation therapy, are standard across least cT2c, a Gleason score of at least 8, or a PSA
14, 2020: DOI https://
doi.org/10.1200/ several solid tumor malignancies. However, consen- greater than 20 ng/mL. This definition is widely used,
EDBK_279459 sus regarding the optimal treatment of patients with given its simplicity and ease of use, and has been

McKay et al
heterogeneity in the outcomes observed within the high-risk

groups defined by each stratification tool. Additionally, the
stratification tools to date define T stage clinically based on
• Patients with high-risk prostate cancer have an digital rectal examination. Interobserver variability and under-
increased risk of disease recurrence and death
and overstaging are challenges with digital rectal exami-
from prostate cancer.
nation staging.12 The American Joint Committee on Cancer
• Local treatment strategies include definitive staging recognizes the use of MRI in clinical T stage cat-
radiotherapy or radical prostatectomy with or egorization. Incorporation of multiparametric prostate MRI
without adjuvant or salvage radiation therapy.
can be useful to evaluate for the presence of extraprostatic
• The backbone of systemic therapy for patients disease, seminal vesicle invasion, and pelvic node in-
with high-risk disease includes androgen dep- volvement. However, challenges also arise with results in-
rivation therapy, although many questions re- terpretation, given differences in MRI quality (magnet
main regarding the use with surgery, intensity,
strength, use of coils), reader variability, prostate factors
and duration of androgen deprivation therapy.
(size of gland, size and grade of tumor), and patient factors
• Blood- and tissue-based biomarkers to guide (motion artifact, metal artifact, biopsy-related hemorrhage).13
therapy selection continue to be an area of active Molecular biomarker tests (e.g., Decipher, Oncotype DX
research, and contemporary clinical trials are
Prostate, Prolaris, or ProMark) have further improved risk
integrating such predictive biomarkers to better
stratification algorithms. Although previous guidelines panels
guide therapy selection for patients at high risk.
have recommended consideration of the use of any of these
four assays for initial risk stratification in patients (with life
adopted by the American Urologic Association,5 European expectancy greater than 10 years) with low or favorable-
Association of Urology,6 and United Kingdom National In- intermediate risk of prostate cancer, the most recent 2020
stitute for Health and Clinical Excellence.7 The Radiation NCCN guidelines expand the use of biomarkers into more
Therapy Oncology Group (RTOG) developed the first clas- aggressive patient populations, stating that “Men with un-
sification system using factors associated with overall sur- favorable intermediate- and high-risk disease and life ex-
vival and prostate cancer–specific survival as opposed to pectancy 10 years may consider use of Decipher and
PSA recurrence. The RTOG definition of high-risk disease is Prolaris tumor-based molecular assays.”
(1) PSA of 20 to 100 ng/mL, biopsy Gleason score of at least
7, and any clinical T stage or (2) PSA less than 100 ng/mL, LOCAL TREATMENT OF PATIENTS WITH HIGH-RISK DISEASE
Gleason score 8 to 10, and clinical T stage cT2c.8 The Radical Prostatectomy
National Comprehensive Cancer Network (NCCN) defines
high risk as clinical T stage cT3a, Gleason score of at least Historically, high-risk prostate cancer in men, especially if
8, or PSA of at least 20 ng/mL, and it defines very high risk clinically advanced, was managed with systemic mono-
as T3b or T4 disease.9 Although these definitions do not therapy using ADT, radiation therapy, or a combination of
integrate disease extent within the gland, the Cancer of the both. Radical prostatectomy was discouraged because of
Prostate Risk Assessment score considers the percent of concerns regarding positive surgical margins, inadequate
biopsy cores positive for cancer in risk stratification, along with disease control, and morbidity.14 However, studies have
age, clinical T stage, and Gleason score, to predict prostate investigated men with locally advanced prostate cancer and
cancer–specific mortality independent of treatment.10 Despite found that excellent long-term survival outcomes are
the introduction of this additional classification system, given achieved. Carver and colleagues15 showed cancer-specific
variability in reliably reproducing estimates of percentage of survival was 85% at 10 years and 76% at 15 years. Although
cores involved with tumor, traditional classification systems there is concern for micrometastatic disease in patients at
are more commonly used for risk stratification. high risk, patients undergoing radical prostatectomy benefit
from local control, with rates of local recurrence approxi-
In addition to tiered classifications systems, nomograms that
mating 10%.16 The important local control and debulking
incorporate several interacting continuous and categoric clinical
not only improve the efficacy of sequential therapy with
variables have been established to stratify patients by risk. The
either radiation therapy or ADT aimed at micrometastatic
Kattan preoperative nomogram uses a multivariable model that
and locoregional disease control but also prevent clinical
combines clinical T stage, Gleason score, and PSA to generate
complications, such as hematuria and obstruction.17 Fur-
an estimate of the risk of treatment failure after radical pros-
thermore, radical prostatectomy allows for accurate staging
tatectomy on a continuous scale.11 Application of the nomogram
that can guide selection of patients who may benefit from
requires a multistep paper tool or computer program for use.
adjuvant therapies. There are also data showing that down-
Although the development of instruments to aid in grading to a lower Gleason score is not infrequent after radi-
risk stratification has been helpful, there is considerable cal prostatectomy for high-risk prostate cancer.18

Optimizing Outcomes for Patients With High-Risk Localized Prostate Cancer
Role of Pelvic Lymph Node Dissection patients in the adjuvant arm were more likely to experience
The current NCCN prostate cancer guidelines recommend clinically relevant genitourinary toxicity. In addition, the
performing extended pelvic lymph node dissection (ePLND) RADICALS-RT trial, which used a trigger for early salvage
in patients with 2% or greater predicted probability of nodal radiotherapy of PSA greater than 0.1 ng/mL, was recently
metastases by nomograms. The ePLND includes lymph presented at the 2019 European Society of Medical On-
node removal of all node-bearing tissue from a predefined cology meeting, showing that between patients who re-
anatomic region; however, the decision to perform PLND ceived adjuvant versus early salvage radiation, there was no
and the extent have historically been left to the surgeon’s difference in biochemical progression-free survival. A meta-
discretion. As a result, most studies looking at the risks and analysis of these trials (as well as the GETUG-AFU 17 trial,
benefits of PLND are retrospective, with associated con- which has not been independently presented or pub-
founding variables.17 A systemic review analyzing com- lished), termed the ARTISTIC collaboration, found that,
parative studies found the data to be mixed regarding despite some difference in patient population and study
oncologic outcomes, and, overall, there was no solid quality design, the results were remarkably similar; there was no
evidence indicating that any type of PLND improves bio- significant improvement in biochemical event-free survi-
chemical recurrence, distant metastasis, or survival com- val for patients receiving adjuvant radiotherapy compared
pared with no PLND for high-risk prostate cancer. 19 with early salvage radiotherapy. Thus, early salvage ra-
Furthermore, PLND was associated with a high rate of diotherapy is an accepted approach for patients with high-
lymphocele but no statistical significance in urinary conti- risk disease undergoing initial surgical treatment, although
nence and erectile function recovery.20,21 Although the additional follow-up data are warranted to assess long-term
benefits of PLND are not clear, this may be because of a outcomes.
lack of adequately powered randomized controlled trials.
Radiation Therapy Paradigms
The rationale for PLND still holds true in that it provides the
most accurate staging to determine node-positive disease, Conformal techniques, particularly intensity-modulated ra-
which may impact therapeutic treatment decisions. In ad- diation therapy and image-guided radiation therapy, are the
dition, PLND may potentially be curative for patients with contemporary standard of care for treatment of high-risk
limited nodal involvement at time of PLND. localized prostate cancer for patients electing radiation
therapy. These conformal techniques allow higher doses
Adjuvant Versus Early Salvage Radiotherapy to the target while minimizing toxicity to normal tissues,
The optimal timing of postoperative radiation for patients compared with older approaches. The role of moderate
with adverse pathologic findings, such as seminal vesicle hypofractionation continues to evolve, and the optimal
invasion, extraprostatic extension, and/or positive surgical regimen for hypofractionation for high-risk prostate cancer
margins, has been widely studied.22 Historic trials, including has not yet been established. In the Dutch HYPRO trial, 820
SWOG-8794,22,23 ARO-96-02,24 EORTC-22911,25,26 and patients, including more than 73% with high-risk prostate
FinnProstate,27 investigated the role of adjuvant radiotherapy cancer, were randomly assigned to radiation therapy with
versus observation in patients with high-risk features. Overall, conventional fractionation (39 fractions of 2 Gy over 8
these studies demonstrated improvements in biochemical weeks) or hypofractionation (19 fractions of 3.4 Gy in 6.5
progression-free survival, but only SWOG 8794 showed im- weeks).28 At 60 months, there was no difference in 5-year
provements in metastasis-free survival and overall survival. De- relapse-free survival; however, gastrointestinal toxicity was
spite these studies, the use of adjuvant radiotherapy in patients more common in the hypofractionation group.
at high risk is low, given the concern for overtreatment of patients For men with localized high-risk prostate cancer without
in whom the cancer is never destined to recur after surgery. clinical pelvic adenopathy, the role of whole-pelvis radiation
Contemporary prospective studies have compared adjuvant therapy remains a controversial area. Whole-pelvis radiation
versus early salvage radiation therapy. The RAVES trial, therapy can be considered in men with an estimated risk of
recently presented at the 2019 American Society for Ra- nodal involvement exceeding 15% on the basis of Partin
diation Oncology (ASTRO) meeting, was the first ran- tables or other tools. Two randomized trials (RTOG 9413
domized study comparing adjuvant and early salvage and GETUG-01) investigated the role of whole-pelvis radi-
radiotherapy. Patients were randomly selected to either ation therapy and did not demonstrate a clear benefit of
adjuvant radiotherapy or an early salvage strategy in which whole-pelvis radiation therapy compared with prostate-only
radiotherapy was started at PSA of 0.2 ng/mL or greater. radiation therapy.29,30 RTOG 0924 (NCT01368588), which has
ADT was not used concurrently with radiotherapy. At a patients with high-risk or locally advanced prostate cancer
follow-up median of 6.1 years, there was no evidence of receiving ADT in conjunction with either prostate-only or
improved biochemical progression-free survival among whole-pelvis radiation therapy, will provide additional informa-
patients who received adjuvant radiotherapy. Furthermore, tion on the extent of radiation therapy for localized disease.

McKay et al
The use of brachytherapy with external beam radiother- androgen receptors and binding of androgen receptors
apy has been studied extensively in patients with both to androgen response elements on DNA.46
intermediate- and high-risk disease. The ASCENDE-RT trial Several contemporary studies have investigated neoadjuvant
demonstrated that the addition of brachytherapy boost abiraterone, enzalutamide, and apalutamide in patients with
to external beam radiotherapy and ADT in men with localized prostate cancer undergoing radical prostatectomy
intermediate- and high-risk disease was associated with (Table 1).47-50 Although the systemic therapy regimens
improved biochemical control and comparable overall differed among these studies (leuprolide 6 months +
survival.31 However, brachytherapy boost was associated abiraterone 3 vs. 6 months [NCT00924469]50; enzalu-
with increased genitourinary toxicity but no difference in tamide vs. enzalutamide + dutasteride + leuprolide
gastrointestinal toxicity. Furthermore, among patient-reported 6 months [NCT01547299]48; enzalutamide + leuprolide
outcomes, brachytherapy boost was associated with worse with or without abiraterone [NCT02268175]47), eligible
overall health, sexual function, and urinary function. patients were required to have unfavorable intermediate-
SYSTEMIC THERAPY FOR PATIENTS WITH or high-risk prostate cancer. Additionally, the studies in-
HIGH-RISK DISEASE tegrated central pathology review with predefined criteria
for measurement and reporting of pathologic outcomes.
Systemic Therapy in Combination With
Overall, combination therapy resulted in pathologic com-
Radical Prostatectomy
plete responses in 4% to 10% of patients and minimum
Neoadjuvant therapy Several clinical trials have evaluated residual disease in 17% to 30% of patients. A critical
neoadjuvant therapy before radical prostatectomy. Initial component of neoadjuvant therapy is that pathologic re-
studies were conducted in the early 1990s with the primary sponse correlates with long-term outcomes. A meta-analysis
intent of improving pathologic surgical outcomes, mainly the of pooled contemporary clinical trials of more intense
rate of positive surgical margins.32,33 Most studies included ADT demonstrated that with a median follow-up of 3.4
luteinizing hormone-releasing hormone agonists or antag- years, 3-year biochemical-free survival was 70% and
onists with or without first-generation antiandrogens. In 3-year metastasis-free survival was 98%.51 All patients
aggregate, these studies included a small number of pa- who experienced a pathologic response (defined as a
tients, most of whom had lower-risk prostate cancer, and pathologic complete response or minimum residual dis-
did not robustly evaluate pathology responses and long- ease no more than 5 mm of residual tumor) were still
term outcomes. Gleave and colleagues34 evaluated 8 ver- disease free at the time of last follow-up. In light of the
sus 3 months of leuprolide plus flutamide in patients with results of these phase II studies, an international, phase
T1b-T2 tumors. This study enrolled 547 patients and dem- III, randomized, placebo-controlled study is underway
onstrated a lower positive surgical margin rate (12% vs. 23% evaluating leuprolide with apalutamide versus placebo
with 8 vs. 3 months) and improved pathologic response rate for 6 months before and 6 months after radical prosta-
(9.3% vs. 5.1% with 8 vs. 3 months) with longer therapy; tectomy (NCT03767244). The coprimary endpoints are
however, disease-free survival and overall survival were not the pathologic complete response rate and metastasis-
reported. free survival. The results will inform evaluation of patho-
The introduction of more effective hormonal agents for logic complete response as a surrogate for metastasis-free
patients with advanced disease has revived interest in this survival in prostate cancer.52
approach for patients with localized prostate cancer at high Studies have evaluated the impact of neoadjuvant che-
risk for recurrence. Abiraterone—a CYP17 inhibitor that motherapy or chemohormonal therapy before radical pros-
decreases testosterone production from the adrenal gland, tatectomy with heterogeneous results. Docetaxel, a taxane
testicles, and prostate cancer cells—has demonstrated chemotherapy that prevents microtubule depolymerization,
improved survival for patients with metastatic castration- has substantial activity in patients with advanced prostate
resistant prostate cancer (CRPC)35,36 and also advanced cancer, including those with CRPC53 and metastatic hormone-
hormone-sensitive disease.37,38 Additionally, three next- sensitive disease.54,55 To date, the most robust data for use
generation antiandrogens, including enzalutamide (for of neoadjuvant docetaxel with leuprolide come from the
nonmetastatic39 and metastatic40,41 CRPC and metastatic PUNCH trial (NCT00430183).56 The trial enrolled 750 pa-
hormone-sensitive disease42), apalutamide (for non- tients with clinically localized high-risk prostate cancer (de-
metastatic CRPC 43 and metastatic hormone-sensitive fined as a Kattan preoperative nomogram probability of less
disease44), and darolutamide (for nonmetastatic CRPC45), than 60% biochemical progression-free survival at 5 years
are currently used for patients with advanced prostate after radical prostatectomy or biopsy Gleason score of 8–10).
cancer. Unlike first-generation antiandrogens, in addition Patients were randomly assigned 1:1 to docetaxel plus leu-
to blocking androgens from binding to the androgen prolide versus surgery alone. Although there was no differ-
receptor, these agents prevent nuclear translocation of ence in the trial primary endpoint of 3-year biochemical

TABLE 1. Summary of PSA and Pathologic Outcomes of Contemporary Clinical Trials of Intense Neoadjuvant ADT Before RP
NeoAbi50 NeoEnza48 NeoAbiEnza47
12wAA 24wAA Enza EDL ELAP EL

Clinical Outcomes (n = 27) (n = 29) (n = 25) (n = 23) (n = 50) (n = 25)
Median PSA pre-RP visit, ng/mL 0.06 0.04 0.51 0.04 0.03 0.02
‡ ypT3 59% 48% 72% 61% 50% 56%
n = 16 n = 14 n = 18 n = 14 n = 25 n = 14
Positive nodes 11% 24% 4% 26% 10% 12%
n=3 n=7 n=1 n=6 n=5 n=3
Positive margins 19% 10% 16% 22% 18% 12%
n=5 n=3 n = 17 n = 13 n=9 n=3
pCR (%) 4% 10% 0% 4% 10% 8%
n=1 n=3 n=0 n=1 n=5 n=2
MRD (largest cross-sectional dimension £ 5 mm) 0% 14% — — 20% 8%
n=0 n=4 n = 10 n=2
MRD (largest cross-sectional dimension £ 3 mm) — — 0% 13% — —
n=0 n=3
pCR or MRD (largest cross-sectional dimension 4% 24% 0% 17% 30% 16%
£ 3 or £ 5 mm)
n=1 n=7 n=0 n=4 n = 15 n=4
RCB £ 0.25 cm 3
44% 52% 36% 74% — —
n = 12 n = 15 n=9 n = 17
Abbreviations: PSA, prostate-specific antigen; ADT, androgen deprivation therapy; RP, radical prostatectomy; 12wAA, 12-week abiraterone acetate; 24wAA,
24-week abiraterone acetate; AA, abiraterone acetate; Enza, enzalutamide; leuprolide; EDL, enzalutamide, dutasteride, leuprolide; ELAP, enzalutamide,
leuprolide, abiraterone, prednisone; EL, enzalutamide; pCR, pathologic complete response; MRD, minimum residual disease; RCB, residual cancer burden.
progression-free survival, with 8 years of follow-up, rates of and prostate cancer–specific survival and reduced the risk of
biochemical progression-free survival were greatly improved recurrence in patients with node-positive prostate cancer, this
in patients receiving docetaxel and leuprolide. Additionally, single-institution, relatively small-scale study did not test the
although the study was not powered to detect differences in use of ADT for PSA recurrence and did not investigate the
overall survival, there was a strong signal of improved overall long-term quality-of-life impact of long-term ADT.59 At the
survival in patients receiving systemic therapy (HR, 0.66; present time, use of adjuvant hormonal therapy after radical
95% CI, 0.42–1.03).56 prostatectomy remains controversial. The phase III AFU-
GETUG-20 trial (NCT01442246) will evaluate adjuvant leu-
Adjuvant therapy The role of adjuvant hormonal therapy
prolide for 2 years after radical prostatectomy. The primary
after radical prostatectomy is uncertain, given the lack of
endpoint is 10-year metastasis-free survival. The trial has
prospective clinical trials of this approach without radiation
completed accrual of 700 patients, and results, which are
therapy after radical prostatectomy. The addition of high-
expected in 2027, will inform the role of adjuvant hor-
dose bicalutamide to watchful waiting, radical prostatectomy,
monal therapy after radical prostatectomy. Additionally, the
and radiation therapy was investigated in three complemen-
ERADICATE trial will investigate the role of adjuvant ADT
tary, double-blind, placebo-controlled trials enrolling 8,113
with darolutamide after prostatectomy in patients with high-risk
patients.57 Treatment was for 2 years in Trial 23 and until
prostate cancer. Several studies have investigated the role
disease progression in Trials 24 and 25. At a median follow-
of adjuvant chemohormonal therapy after radical prostatec-
up of 10 years, bicalutamide resulted in an improvement in
tomy. Collectively, these trials have largely been unsuccessful,
progression-free survival in men with extraprostatic disease
and this approach is not used in clinical practice after surgery.
(T3-4), regardless of choice of local therapy, although no
overall survival benefit was observed. For patients with
lymph node–positive disease at the time of radical prosta- Systemic Therapy in Combination With Radiation Therapy
tectomy, immediate ADT until disease progression versus Hormone therapy Hormonal therapy with radiation therapy
delayed treatment was investigated in a study of 98 pa- is a standard of care for patients with high-risk localized
tients.58 Although immediate ADT improved overall survival prostate cancer (Table 2). Several clinical trials have

McKay et al
established the role of ADT with radiation therapy, given superior to 18 months of ADT, the trial was not designed
improvements in cancer-specific and overall survival. The as a noninferiority trial, and therefore equivalency be-
landmark EORTC 22863 trial evaluated radiation therapy tween the two arms cannot be established from the results of
with or without 3 years of ADT.60 At a median follow-up of this study. Guidelines from several panels, including ASCO,
9 years, the 10-year disease-free survival (48% vs. 23%; American Urologic Association, ASTRO, and the Society of
HR, 0.42; 95% CI, 0.33–0.53) and overall survival (58% vs. Urologic Oncology, generally recommend prolonged ADT
40%; HR, 0.60; 95% CI, 0.45–0.80) were improved with for men with high-risk localized prostate cancer (24–36
combination therapy compared with radiation alone. Ad- months) rather than a shorter duration of therapy.5,6,9 The
ditionally, prostate cancer mortality was decreased (10% vs. use of a luteinizing hormone-releasing hormone agonist
30%; HR, 0.38; 95% CI, 0.24–0.60). with or without an antiandrogen has not been formally
addressed in a randomized trial. Although data from ob-
Efforts to decrease the toxicity associated with prolonged
servational studies support combination therapy, these
ADT have led to several clinical trials comparing short- and
studies did not systematically evaluate longer durations
long-course ADT. The optimal duration of therapy has not
(greater than 4 months) of combination therapy.69
been firmly established. RTOG 92-0263 and DART 01/05
GICOR,66 which investigated 4 versus 28 months of ADT, The addition of abiraterone to ADT for patients with newly
and EORTC 22961,64 which investigated 6 versus 36 months diagnosed hormone-sensitive prostate cancer was investigated
of ADT, demonstrated improvements in overall survival with in the STAMPEDE trial, a multiarm, multistage, randomized
prolonged ADT. Nabid and colleagues67 investigated 18 controlled trial.38 This study included a heterogeneous
versus 36 months of ADT. Although 36 months was not group of patients, including 27% with newly diagnosed
TABLE 2. Phase III Randomized Trials of ADT Use in Combination With Radiation Therapy for Localized Prostate Cancer
No. of Primary
Trial Year Stage Patients Arms RT Endpoint OS
RTOG 2005 T3N0-1 (15% RP) 977 RT 6 LHRHa/orchiectomy until 65–70 Gy Improved survival (10-
85-3161 progression year survival 49% vs.
39%)
RTOG 2008 T2-4N0-1 456 RT 6 LHRHa + flutamide 65–70 Gy 10–year OS No difference (10-year
86-1062 2 months before and OS 43% vs. 34%)
concurrent vs. no ADT
RTOG 2008 T2c-4N0-1M0 1,554 LHRHa + flutamide 65–70 Gy 10–year DFS Improved OS in Gleason
92-0263 4 months + RT 6 24-month 8–10 subgroup (45%
adjuvant LHRHa vs. 32%)
EORTC 2009 T1c-2abN1M0 or 970 LHRHa + antiandrogen 70 Gy OS Improved OS (5-year
2296164 T2c-4N0-1M0 6 months + RT 6 30-month (noninferior) mortality 15% vs.
adjuvant LHRHa 19%)
EORTC 2010 T1-2 WHO grade 3 or 415 RT 6 LHRHa 3 years 70 Gy DFS Improved OS (10-year
2283660 T3-4 OS 58% vs. 40%)
TROG 2011 T2b-4N0M0 802 Neoadjuvant LHRHa + flutamide 66 Gy Time to local Improved PCSM (10-
96-0165 (0 vs. 3 vs. 6 months) + RT failure; year PCSM 11% vs.
PCSM 22% for 6 mos vs.
no ADT)
DART 2015 T1c-T3bN0M0 255 LHRHa + antiandrogen 76 Gy BDFS Improved OS (5-year OS
01/05 4 months + RT 6 24-month 95% vs. 86%)
GICOR66 adjuvant LHRHa
Nabid67 2018 T3-4 or PSA . 20 630 LHRHa + antiandrogen 70 Gy OS No difference (5-year
ng/mL or Gleason 4 months + RT + 36- vs. OS 91% vs. 86%)
.7 18-month adjuvant LHRHa
TROG 2019 T2b-N0M0 or T2a + 1,071 Neoadjuvant LHRHa 6 months 66–74 Gy (or PCSM Improved PCSM (10%
03-0468 PSA 10 ng/mL + RT 6 12-month adjuvant 46 Gy + HDB vs. 13% for 18 vs.
and Gleason 7 LHRHa 6 18-month boost 19.5 Gy) 6 months of ADT)
zoledronic acid
Abbreviations: ADT, androgen deprivation therapy; RT, radiation therapy; OS, overall survival; RP, radical prostatectomy; LHRHa, lutenizing hormone
releasing hormone agonist; DFS, disease-free survival; PCSM, prostate cancer–specific mortality; BDFS, biochemical disease-free survival; PSA, prostate-
specific antigen; HDB, high-dose brachytherapy boost.

node-negative, nonmetastatic prostate cancer. These patients changes in NCCN guidelines. In the 2019 guidelines,
received abiraterone plus ADT versus ADT alone for 2 years biomarkers were not recommended for initial risk stratifi-
and were mandated to receive definitive radiation therapy. cation within high-risk populations, but the most recent
In subgroup analyses by metastasis status, nodal status, and 2020 guidelines advocate for consideration of the Deci-
planned radiation therapy, overall survival favored abiraterone pher and Prolaris tumor-based molecular assays in men
plus ADT compared with ADT alone (M0: HR, 0.75; 95% CI, with high-risk disease and life expectancy of at least
0.48–1.18; N0: HR, 0.69; 95% CI, 0.49–0.96; planned radio- 10 years. To date, Decipher is the most extensively stud-
therapy: HR, 0.64; 95% CI, 0.38–1.08). The use of abiraterone ied biomarker assay within the high-risk population. Large
in conjunction with ADT and radiation therapy is controversial studies have demonstrated that the addition of Decipher
and requires a thoughtful discussion between patients and to clinicopathologic variables improves the estimation of
clinicians on the risks and benefits of this approach. Two risk of distant metastases compared with risk stratifica-
phase III studies are investigating the addition of apaluta- tion by NCCN groups alone.72 In addition, in studies fo-
mide (ATLAS, NCT02531516) and enzalutamide (ENZARAD, cused specifically within patients with high-risk prostate
NCT02446444) combined with ADT for patients with high-risk cancer, Decipher is an independent predictor of meta-
prostate cancer undergoing primary radiation therapy. Both static progression even when accounting for clinicopath-
studies have completed accrual. The primary endpoint for the ologic features in patients treated with surgery or with
ATLAS study is metastasis-free survival, and the primary radiotherapy.73,74 Recent data also support the prognostic
endpoint for the ENZARAD study is overall survival. value of Prolaris in predicting metastatic progression in
patients with NCCN intermediate- or high-risk prostate
Chemotherapy The addition of chemotherapy to radiation cancer.75
therapy for patients with localized prostate cancer has not
As a consequence of these retrospective studies demon-
been established as a standard practice. The phase III RTOG
strating the benefit of biomarkers, a large, prospective,
05-21 trial investigated the benefit of adjuvant docetaxel for
biomarker-driven phase III clinical trial (the PREDICT-RT
six cycles to 24 months of ADT and radiotherapy in patients
trial, NRG GU009) is being initiated for patients with high-
with high-risk prostate cancer.70 At a median follow-up of 5.7
risk prostate cancer. In this study, patients with high-risk
years, disease-free survival and overall survival were im-
disease, by NCCN criteria, will undergo biomarker testing of
proved with adjuvant chemotherapy (4-year overall sur-
their tumor sample with the Decipher assay. Those with high
vival, 93% vs. 89%; HR, 0.69; 95% CI, 0.49–0.97). In
Decipher scores will be enrolled in a phase III randomized
addition, evidence supporting the potential utility of che-
trial of treatment intensification, assessing the addition of
motherapy for patients with localized prostate cancer stem
intensified androgen ablation (with abiraterone and apa-
from the STAMPEDE trial.54 In the stage of the study testing
lutamide) to the standard of care (radiation and 24 months
docetaxel for hormone-sensitive disease, 22% of patients
of ADT). Those with low or intermediate Decipher scores will
has localized disease. Radiotherapy was initially encour-
be enrolled in a phase III randomized trial of treatment de-
aged and then later mandated for patients with N0M0
intensification, in which a de-intensified regimen (radiation
disease. Overall survival in patients with disease staged as
and 12 months of ADT) is compared with the standard of
M0 and N0 favored docetaxel (M0: HR, 0.95; 95% CI,
care (radiation and 24 months of ADT).
0.62–1.47; N0: HR, 0.58; 95% CI, 0.41–0.81); however, in
patients with planned radiation therapy, overall survival In addition to improving risk prediction in treatment-naı̈ve
favored ADT alone (HR, 1.11; 95% CI, 0.67–1.85). In patients who are at high risk, biomarkers also have a role in
a meta-analysis of three trials evaluating docetaxel of M0 improving the assessment of risk of disease progression
prostate cancer (GETUG-12, RTOG 05-21, and STAM- after radical prostatectomy, particularly for patients at high
PEDE), no overall survival benefit was observed with the risk. Current NCCN guidelines state that the “Decipher
addition of docetaxel chemotherapy.71 Additional data are molecular assay is recommended to inform adjuvant treat-
needed to evaluate the effects of docetaxel for patients with ment if adverse features are found post-radical prostatec-
M0 disease. The PEACE2 trial will investigate the role of tomy”; these guidelines are based on a number of studies
cabazitaxel for patients with localized high-risk prostate validating the prognostic value of Decipher in the postprosta-
cancer (NCT01952223). The trial, which has a primary tectomy space, including a large meta-analyses of 975
endpoint for progression-free survival, is actively enrolling patients demonstrating that Decipher independently pre-
patients and is expected to report results in December 2025. dicts metastatic progression within nearly all clinicopathologic,
demographic, and treatment subgroups.76 More recently,
BIOMARKERS TO GUIDE THERAPY SELECTION FOR a study validated the prognostic value of Decipher in patient
LOCALIZED PROSTATE CANCER samples from the RTOG 9601 trial, in which patients with PSA
The use of biomarkers to guide treatment approaches in recurrences after surgery were randomly assigned to radia-
high-risk prostate cancer is evolving, as evidenced by recent tion alone or in combination with 2 years of high-dose

McKay et al
bicalutamide.77 Intriguingly, this study also suggested that only assessed via germline testing include BRCA1, BRCA2,
patients with intermediate to high Decipher scores derived ATM, PALB2, MLH1, MSH2, MSH6, and PMS2.81 Several
benefit from the antiandrogen therapy, and patients with studies have investigated the impact of germline mutations
low Decipher scores did not.78 Based on these findings, in BRCA2, a key DNA repair gene involved in homologous
a number of trials have used or are incorporating Decipher recombination, on treatment outcomes in patients with
to select patients at high risk for adjuvant therapies after prostate cancer. Germline mutations in DNA repair genes
prostatectomy. These include the NRG-GU002 RADD occur in approximately 4.6% of patients with localized
randomized trial (NCT03070886), which is investigating the prostate cancer and 11.8% to 16.2% of patients with
addition of adjuvant docetaxel to radiation and ADT for metastatic disease; among these, BRCA2 alterations are by
patients with persistently elevated PSAs after prostatectomy, far the most common.82 For men with localized prostate
as well as the ECOG ERADICATE trial, in which patients who cancer treated with local therapy (i.e., either radical pros-
had a prostatectomy with high Decipher scores were ran- tatectomy or radiation), the presence of a pathogenic
domly assigned to 12 months of ADT with or without germline BRCA2 mutation is associated with worse
12 months of darolutamide. metastases-free survival and prostate cancer cause-specific
It should be noted that the major assays (Decipher, Prolaris, survival compared with patients who do not harbor such
Oncotype DX, or Promark) were designed as prognostic, but a mutation.81 A more recent study suggested that, in a
not specifically predictive, biomarker panels. A prognostic cohort of 67 carriers of BRCA2 or BRCA1 germline muta-
biomarker provides information on outcomes independent tions (compared with 1,235 noncarriers), prostate cancer–
of the treatment received. Thus, aggressive disease, as specific survival was decreased in patients with the BRCA
identified by a prognostic test, may be aggressive regard- mutation treated with radiotherapy compared with patients
less of the treatment received. In contrast, a predictive who did not have germline BRCA alterations; no difference
biomarker specifically identifies response or resistance to was seen in prostate cancer–specific survival between
a particular therapy, but not to all treatments. Although BRCA carriers and noncarriers treated with radical pros-
prognostic biomarkers are useful (e.g., in identifying pa- tatectomy.83 Although this study has led some researchers
tients with aggressive disease who should be enrolled in to conclude that patients with BRCA2-mutant prostate
studies of treatment intensification), predictive biomarkers cancer might be better treated with surgery, it is difficult to
are ultimately needed to select the right treatment of each draw practice-changing conclusions from a small retro-
patient. A predictive biomarker panel has been developed spective study in which patients treated with radiation had
and validated to predict the benefit of postoperative radi- far more aggressive disease than patients treated with
ation; this panel is called the postoperative radiation ther- surgery (radiation therapy vs. radical prostatectomy cohort:
apy outcome score and is available through the Decipher median PSA, 14 vs. 7.5 ng/mL; Gleason score of at least 8
platform79; however, it should be noted that postoperative in 20% vs. 10%; patients at high risk, 24% vs. 56%) and
radiation therapy outcome score has not been validated in which there was no significant statistical interaction
prospectively. In addition, another predictive biomarker between BRCA status and treatment modality for either
panel, the PAM50 classifier, has been optimized and val- metastasis-free survival or prostate cancer–specific survival.
idated to predict the benefit of postoperative ADT; this Thus, the issue of whether BRCA status should influence
panel, which was originally derived in breast cancer sam- treatment selection remains an area of active research.
ples and predicts benefit from endocrine therapy in that CONCLUSIONS
disease, has also been validated, in retrospective cohorts, Multimodal treatment strategies of surgery, radiation ther-
to predicting benefit from androgen-directed therapies in apy, and systemic therapy offer the greatest potential for
prostate cancer.80 PAM50 is also being prospectively tested improved long-term outcomes for patients with high-risk
in the NRG GU006 BALANCE trial (NCT03371719), in prostate cancer who may harbor occult metastatic disease.
which patients with PSA recurrences after prostatectomy Integrated multidisciplinary teams of urologists, medical
are randomly assigned to salvage radiation alone or in oncologists, radiation oncologists, radiologists, and pathol-
combination with 6 months of apalutamide. ogists will be instrumental in shifting the treatment tide for
The biomarker assays discussed thus far have involved patients. Novel systemic therapies, including hormonal,
testing of samples of prostate cancer. However, it should be cytotoxic, targeted, and immunologic agents, tested in the
noted that NCCN guidelines now includes a recommenda- context of rationally designed clinical trials will help better
tion of germline genetic testing (i.e., testing of blood or saliva refine therapies for this heterogeneous group of patients.
samples for alterations present in a patient) for all patients Such biomarker-driven trials will be of increasing impor-
with high-risk prostate cancer. Specific genes that should be tance in the era of precision medicine.

AFFILIATIONS #0987, La Jolla, CA 92093; Twitter: @DrRanaMcKay; email: rmckay@

1
University of California San Diego, San Diego, CA ucsd.edu.
2
University of California San Francisco, San Francisco, CA
3
Vanderbilt University Medical Center, Nashville, TN
CORRESPONDING AUTHOR applicable) are available with this article at DOI https://doi.org/10.1200/
Rana R. McKay, Department of Medicine, Division of Hematology/ EDBK_279459.
Oncology, University of California San Diego, 3855 Health Sciences Dr.,
REFERENCES
1. Winter A, Sirri E, Jansen L, et al; Association of Population-based Cancer Registries in Germany (GEKID) Cancer Survival Working Group. Comparison of prostate
cancer survival in Germany and the USA: can differences be attributed to differences in stage distributions? BJU Int. 2017;119:550-559.
2. D’Amico AV, Whittington R, Malkowicz SB, et al. Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation
therapy for clinically localized prostate cancer. JAMA. 1998;280:969-974.
3. Cooperberg MR, Broering JM, Carroll PR. Time trends and local variation in primary treatment of localized prostate cancer. J Clin Oncol. 2010;28:1117-1123.
4. Partin AW, Yoo J, Carter HB, et al. The use of prostate specific antigen, clinical stage and Gleason score to predict pathological stage in men with localized prostate
cancer. J Urol. 1993;150:110-114.
5. Sanda MG, Cadeddu JA, Kirkby E, et al. Clinically localized prostate cancer: AUA/ASTRO/SUO guideline. Part I: risk stratification, shared decision making, and
care options. J Urol. 2018;199:683-690.
6. Mottet N, Bellmunt J, Bolla M, et al. EAU-ESTRO-SIOG guidelines on prostate cancer. Part 1: screening, diagnosis, and local treatment with curative intent. Eur
Urol. 2017;71:618-629.
7. Dasgupta P, Davis J, Hughes S. NICE guidelines on prostate cancer 2019. BJU Int. 2019;124:1.
8. Roach M, Lu J, Pilepich MV, et al. Four prognostic groups predict long-term survival from prostate cancer following radiotherapy alone on Radiation Therapy
Oncology Group clinical trials [published correction appears in Int J Radiat Oncol Biol Phys. 2000;48:313]. Int J Radiat Oncol Biol Phys. 2000;47:609-615.
9. Mohler JL, Antonarakis ES. NCCN guidelines updates: management of prostate cancer. J Natl Compr Canc Netw. 2019;17:583-586.
10. Cooperberg MR, Pasta DJ, Elkin EP, et al. The University of California, San Francisco, Cancer of the Prostate Risk Assessment score: a straightforward and reliable
preoperative predictor of disease recurrence after radical prostatectomy. J Urol. 2005;173:1938-1942.
11. Kattan MW, Eastham JA, Stapleton AM, et al. A preoperative nomogram for disease recurrence following radical prostatectomy for prostate cancer. J Natl Cancer
Inst. 1998;90:766-771.
12. Gosselaar C, Kranse R, Roobol MJ, et al. The interobserver variability of digital rectal examination in a large randomized trial for the screening of prostate cancer.
Prostate. 2008;68:985-993.
13. Barrett T, Haider MA. The emerging role of MRI in prostate cancer active surveillance and ongoing challenges. AJR Am J Roentgenol. 2017;208:131-139.
14. Morlacco A, Karnes RJ. High-risk prostate cancer: the role of surgical management. Crit Rev Oncol Hematol. 2016;102:135-143.
15. Carver BS, Bianco FJ Jr., Scardino PT, et al. Long-term outcome following radical prostatectomy in men with clinical stage T3 prostate cancer. J Urol. 2006;
176:564-568.
16. Inman BA, Davies JD, Rangel LJ, et al. Long-term outcomes of radical prostatectomy with multimodal adjuvant therapy in men with a preoperative serum prostate-
specific antigen level . or =50 ng/mL. Cancer. 2008;113:1544-1551.
17. Karnes RJ, Hatano T, Blute ML, et al. Radical prostatectomy for high-risk prostate cancer. Jpn J Clin Oncol. 2010;40:3-9.
18. Boorjian SA, Karnes RJ, Crispen PL, et al. The impact of discordance between biopsy and pathological Gleason scores on survival after radical prostatectomy.
J Urol. 2009;181:95-104, discussion 104.
19. Fossati N, Willemse PM, Van den Broeck T, et al. The benefits and harms of different extents of lymph node dissection during radical prostatectomy for prostate
cancer: a systematic review. Eur Urol. 2017;72:84-109.
20. Liss MA, Palazzi K, Stroup SP, et al. Outcomes and complications of pelvic lymph node dissection during robotic-assisted radical prostatectomy. World J Urol.
2013;31:481-488.
21. Østby-Deglum M, Brennhovd B, Axcrona K, et al. A comparative study of erectile function and use of erectile aids in high-risk prostate cancer patients after robot-
assisted laparoscopic prostatectomy. Scand J Urol. 2015;49:433-439.
22. Thompson IM, Valicenti RK, Albertsen P, et al. Adjuvant and salvage radiotherapy after prostatectomy: AUA/ASTRO Guideline. J Urol. 2013;190:441-449.
23. Thompson IM Jr., Tangen CM, Paradelo J, et al. Adjuvant radiotherapy for pathologically advanced prostate cancer: a randomized clinical trial. JAMA. 2006;
296:2329-2335.
24. Wiegel T, Bottke D, Steiner U, et al. Phase III postoperative adjuvant radiotherapy after radical prostatectomy compared with radical prostatectomy alone in pT3
prostate cancer with postoperative undetectable prostate-specific antigen: ARO 96-02/AUO AP 09/95. J Clin Oncol. 2009;27:2924-2930.

McKay et al
25. Bolla M, van Poppel H, Collette L, et al; European Organization for Research and Treatment of Cancer. Postoperative radiotherapy after radical prostatectomy:
a randomised controlled trial (EORTC trial 22911). Lancet. 2005;366:572-578.
26. Bolla M, van Poppel H, Tombal B, et al; European Organisation for Research and Treatment of Cancer, Radiation Oncology and Genito-Urinary Groups.
Postoperative radiotherapy after radical prostatectomy for high-risk prostate cancer: long-term results of a randomised controlled trial (EORTC trial 22911).
Lancet. 2012;380:2018-2027.
27. Hackman G, Taari K, Tammela TL, et al; FinnProstate Group. Randomised trial of adjuvant radiotherapy following radical prostatectomy versus radical
prostatectomy alone in prostate cancer patients with positive margins or extracapsular extension. Eur Urol. 2019;76:586-595.
28. Incrocci L, Wortel RC, Alemayehu WG, et al. Hypofractionated versus conventionally fractionated radiotherapy for patients with localised prostate cancer
(HYPRO): final efficacy results from a randomised, multicentre, open-label, phase 3 trial. Lancet Oncol. 2016;17:1061-1069.
29. Pommier P, Chabaud S, Lagrange JL, et al. Is there a role for pelvic irradiation in localized prostate adenocarcinoma? Preliminary results of GETUG-01. J Clin
Oncol. 2007;25:5366-5373.
30. Roach M, Moughan J, Lawton CAF, et al. Sequence of hormonal therapy and radiotherapy field size in unfavourable, localised prostate cancer (NRG/RTOG
9413): long-term results of a randomised, phase 3 trial. Lancet Oncol. 2018;19:1504-1515.
31. Morris WJ, Tyldesley S, Rodda S, et al. Androgen Suppression Combined with Elective Nodal and Dose Escalated Radiation Therapy (the ASCENDE-RT Trial): an
analysis of survival endpoints for a randomized trial comparing a low-dose-rate brachytherapy boost to a dose-escalated external beam boost for high- and
intermediate-risk prostate cancer. Int J Radiat Oncol Biol Phys. 2017;98:275-285.
32. McKay RR, Choueiri TK, Taplin ME. Rationale for and review of neoadjuvant therapy prior to radical prostatectomy for patients with high-risk prostate cancer.
Drugs. 2013;73:1417-1430.
33. Ryan ST, Patel DN, Parsons JK, et al. Neoadjuvant approaches prior to radical prostatectomy. Cancer J. 2020;26:2-12.
34. Gleave ME, Goldenberg SL, Chin JL, et al; Canadian Uro-Oncology Group. Randomized comparative study of 3 versus 8-month neoadjuvant hormonal therapy
before radical prostatectomy: biochemical and pathological effects. J Urol. 2001;166:500-506, NaN-507.
35. de Bono JS, Logothetis CJ, Molina A, et al; COU-AA-301 Investigators. Abiraterone and increased survival in metastatic prostate cancer. N Engl J Med. 2011;
364:1995-2005.
36. Ryan CJ, Smith MR, de Bono JS, et al; COU-AA-302 Investigators. Abiraterone in metastatic prostate cancer without previous chemotherapy. N Engl J Med. 2013;
368:138-148.
377:352-360.
2017;377:338-351.
39. Hussain M, Fizazi K, Saad F, et al. Enzalutamide in men with nonmetastatic, castration-resistant prostate cancer. N Engl J Med. 2018;378:2465-2474.
40. Beer TM, Armstrong AJ, Rathkopf DE, et al; PREVAIL Investigators. Enzalutamide in metastatic prostate cancer before chemotherapy. N Engl J Med. 2014;
371:424-433.
41. Scher HI, Fizazi K, Saad F, et al; AFFIRM Investigators. Increased survival with enzalutamide in prostate cancer after chemotherapy. N Engl J Med. 2012;
367:1187-1197.
42. Armstrong AJ, Szmulewitz RZ, Petrylak DP, et al. ARCHES: a randomized, phase III study of androgen deprivation therapy with enzalutamide or placebo in men
with metastatic hormone-sensitive prostate cancer. J Clin Oncol. 2019;37:2974-2986.
43. Smith MR, Saad F, Chowdhury S, et al; SPARTAN Investigators. Apalutamide treatment and metastasis-free survival in prostate cancer. N Engl J Med. 2018;
378:1408-1418.
45. Fizazi K, Shore N, Tammela TL, et al; ARAMIS Investigators. Darolutamide in nonmetastatic, castration-resistant prostate cancer. N Engl J Med. 2019;
380:1235-1246.
46. Tran C, Ouk S, Clegg NJ, et al. Development of a second-generation antiandrogen for treatment of advanced prostate cancer. Science. 2009;324:787-790.
47. McKay RR, Ye H, Xie W, et al. Evaluation of intense androgen deprivation before prostatectomy: a randomized phase II trial of enzalutamide and leuprolide with or
without abiraterone. J Clin Oncol. 2019;37:923-931.
48. Montgomery B, Tretiakova MS, Joshua AM, et al. Neoadjuvant enzalutamide prior to prostatectomy. Clin Cancer Res. 2017;23:2169-2176.
49. Mostaghel EA, Nelson PS, Lange P, et al. Targeted androgen pathway suppression in localized prostate cancer: a pilot study. J Clin Oncol. 2014;32:229-237.
50. Taplin ME, Montgomery B, Logothetis CJ, et al. Intense androgen-deprivation therapy with abiraterone acetate plus leuprolide acetate in patients with localized
high-risk prostate cancer: results of a randomized phase II neoadjuvant study. J Clin Oncol. 2014;32:3705-3715.
51. McKay RR, Montgomery B, Xie W, et al. Post prostatectomy outcomes of patients with high-risk prostate cancer treated with neoadjuvant androgen blockade.
Prostate Cancer Prostatic Dis. 2018;21:364-372.
52. Xie W, Regan MM, Buyse M, et al; ICECaP Working Group. Metastasis-free survival is a strong surrogate of overall survival in localized prostate cancer. J Clin
Oncol. 2017;35:3097-3104.
53. Tannock IF, de Wit R, Berry WR, et al; TAX 327 Investigators. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl
J Med. 2004;351:1502-1512.

54. James ND, Sydes MR, Clarke NW, et al; STAMPEDE investigators. Addition of docetaxel, zoledronic acid, or both to first-line long-term hormone therapy in
55. Sweeney CJ, Chen YH, Carducci M, et al. Chemohormonal therapy in metastatic hormone-sensitive prostate cancer. N Engl J Med. 2015;373:737-746.
56. Eastham JA, Heller G, Halabi S, et al. CALGB 90203 (Alliance): Radical prostatectomy (RP) with or without neoadjuvant chemohormonal therapy (CHT) in men
with clinically localized, high-risk prostate cancer (CLHRPC). J Clin Oncol. 2019;37 (suppl; abstr 5079).
57. Iversen P, McLeod DG, See WA, et al; Casodex Early Prostate Cancer Trialists’ Group. Antiandrogen monotherapy in patients with localized or locally advanced
prostate cancer: final results from the bicalutamide Early Prostate Cancer programme at a median follow-up of 9.7 years. BJU Int. 2010;105:1074-1081.
58. Messing EM, Manola J, Sarosdy M, et al. Immediate hormonal therapy compared with observation after radical prostatectomy and pelvic lymphadenectomy in
men with node-positive prostate cancer. N Engl J Med. 1999;341:1781-1788.
59. Lepor H. Adjuvant hormonal therapy in men with node-positive prostate cancer. Rev Urol. 2000;2:90-91.
60. Briganti A, Wiegel T, Joniau S, et al. Early salvage radiation therapy does not compromise cancer control in patients with pT3N0 prostate cancer after radical
prostatectomy: results of a match-controlled multi-institutional analysis. Eur Urol. 2012;62:472-487.
61. Bolla M, Van Tienhoven G, Warde P, et al. External irradiation with or without long-term androgen suppression for prostate cancer with high metastatic risk: 10-
year results of an EORTC randomised study. Lancet Oncol. 2010;11:1066-1073.
62. Mottet N, Peneau M, Mazeron JJ, et al. Addition of radiotherapy to long-term androgen deprivation in locally advanced prostate cancer: an open randomised
phase 3 trial. Eur Urol. 2012;62:213-219.
63. Pilepich MV, Winter K, Lawton CA, et al. Androgen suppression adjuvant to definitive radiotherapy in prostate carcinoma--long-term results of phase III RTOG 85-
31. Int J Radiat Oncol Biol Phys. 2005;61:1285-1290.
64. Horwitz EM, Bae K, Hanks GE, et al. Ten-year follow-up of radiation therapy oncology group protocol 92-02: a phase III trial of the duration of elective androgen
deprivation in locally advanced prostate cancer. J Clin Oncol. 2008;26:2497-2504.
65. Roach M III, Bae K, Speight J, et al. Short-term neoadjuvant androgen deprivation therapy and external-beam radiotherapy for locally advanced prostate cancer:
long-term results of RTOG 8610. J Clin Oncol. 2008;26:585-591.
66. Zapatero A, Guerrero A, Maldonado X, et al. High-dose radiotherapy with short-term or long-term androgen deprivation in localised prostate cancer (DART01/05
GICOR): a randomised, controlled, phase 3 trial. Lancet Oncol. 2015;16:320-327.
67. Nabid A, Carrier N, Martin AG, et al. Duration of androgen deprivation therapy in high-risk prostate cancer: a randomized phase III trial. Eur Urol. 2018;
74:432-441.
68. Denham JW, Joseph D, Lamb DS, et al. Short-term androgen suppression and radiotherapy versus intermediate-term androgen suppression and radiotherapy,
with or without zoledronic acid, in men with locally advanced prostate cancer (TROG 03.04 RADAR): 10-year results from a randomised, phase 3, factorial trial.
Lancet Oncol. 2019;20:267-281.
69. Nanda A, Chen MH, Moran BJ, et al. Total androgen blockade versus a luteinizing hormone-releasing hormone agonist alone in men with high-risk prostate
cancer treated with radiotherapy. Int J Radiat Oncol Biol Phys. 2010;76:1439-1444.
70. Rosenthal SA, Hu C, Sartor O, et al. Effect of chemotherapy with docetaxel with androgen suppression and radiotherapy for localized high-risk prostate cancer: the
randomized phase III NRG Oncology RTOG 0521 trial. J Clin Oncol. 2019;37:1159-1168.
71. Vale CL, Burdett S, Rydzewska LHM, et al; STOpCaP Steering Group. Addition of docetaxel or bisphosphonates to standard of care in men with localised or
metastatic, hormone-sensitive prostate cancer: a systematic review and meta-analyses of aggregate data. Lancet Oncol. 2016;17:243-256.
72. Spratt DE, Zhang J, Santiago-Jiménez M, et al. Development and validation of a novel integrated clinical-genomic risk group classification for localized prostate
73. Nguyen PL, Martin NE, Choeurng V, et al. Utilization of biopsy-based genomic classifier to predict distant metastasis after definitive radiation and short-course
ADT for intermediate and high-risk prostate cancer. Prostate Cancer Prostatic Dis. 2017;20:186-192.
74. Tosoian JJ, Birer SR, Jeffrey Karnes R, et al. Performance of clinicopathologic models in men with high risk localized prostate cancer: impact of a 22-gene
genomic classifier. Prostate Cancer Prostatic Dis. Epub 2020 Mar 30.
75. Tward JD, Schlomm T, Bardot S, et al. Ability of the combined clinical cell-cycle risk score to identify patients that benefit from multi versus single modality therapy
in NCCN intermediate and high-risk prostate cancer. J Clin Oncol. 2020;38 (suppl; abstr 346).
76. Spratt DE, Yousefi K, Deheshi S, et al. Individual patient-level meta-analysis of the performance of the decipher genomic classifier in high-risk men after
prostatectomy to predict development of metastatic disease. J Clin Oncol. 2017;35:1991-1998.
77. Shipley WU, Seiferheld W, Lukka HR, et al; NRG Oncology RTOG. Radiation with or without antiandrogen therapy in recurrent prostate cancer. N Engl J Med.
2017;376:417-428.
78. Feng FY, Sandler HM, Huang H-C, et al. Transcriptome profiling of NRG Oncology/RTOG 9601: validation of a prognostic genomic classifier in salvage ra-
diotherapy prostate cancer patients from a prospective randomized trial. J Clin Oncol. 2020;38:6s (suppl; abstr 276).
79. Zhao SG, Chang SL, Spratt DE, et al. Development and validation of a 24-gene predictor of response to postoperative radiotherapy in prostate cancer: a matched,
retrospective analysis. Lancet Oncol. 2016;17:1612-1620.
80. Zhao SG, Chang SL, Erho N, et al. Associations of luminal and basal subtyping of prostate cancer with prognosis and response to androgen deprivation therapy.
JAMA Oncol. 2017;3:1663-1672.

McKay et al
81. Castro E, Goh C, Olmos D, et al. Germline BRCA mutations are associated with higher risk of nodal involvement, distant metastasis, and poor survival outcomes in
prostate cancer. J Clin Oncol. 2013;31:1748-1757.
82. Pritchard CC, Mateo J, Walsh MF, et al. Inherited DNA-repair gene mutations in men with metastatic prostate cancer. N Engl J Med. 2016;375:443-453.
83. Castro E, Goh C, Leongamornlert D, et al. Effect of BRCA mutations on metastatic relapse and cause-specific survival after radical treatment for localised prostate
cancer. Eur Urol. 2015;68:186-193.

GERIATRIC ONCOLOGY

GERIATRIC ONCOLOGY
Toward Modernization of Geriatric Oncology by

Digital Health Technologies
Armin Shahrokni, MD, MPH1; Kah Poh Loh, MBBCh BAO2; and William A. Wood, MD3
overview
The number of older adults with cancer is increasing. Over the past 3 decades, geriatric oncology research has
focused on improving the assessment of frailty and fitness of older adults with cancer as well as methods of
improving their outcomes. At the same time, advances in digital health technologies have opened new
frontiers for reaching this goal. Digital health technologies encompass a variety of solutions, from electronic
patient-reported outcomes (ePROs) to big data and wireless sensors. These solutions have the potential to
further advance our understanding of patients’ experiences during cancer treatment. Whereas the data on the
feasibility and utility of such solutions in the care of older adults with cancer are limited, interest from digital
health oncology researchers to further explore the benefits of these products is increasing. In this article, we
describe the focus of geriatric oncology, the rationale behind the need to explore digital health technologies in
this setting, and emerging data and ongoing studies, as well as provide guidelines for proper selection,
implementation, and testing of digital health solutions in the context of geriatric oncology.
INTRODUCTION feasibility, acceptability, and utility of such platforms in

1
It is no longer news that global aging is real. It is also all disease-related contexts remain under investigation,
abundantly clear that cancer is still primarily a disease the promise of such tools in geriatric oncology is worthy
of aging populations.2 In the past 3 decades, evidence of more attention.
has emerged that age alone should not be a factor in As discussed above, the main concept of geriatric
decision making, and tolerating treatment is not a oncology is to assess frailty or a lack of fitness of older
function of age but rather a patient’s frailty or lack of adults with cancer. Until now, such an assessment
fitness.3 The gold standard by which to assess the mainly occurred in the outpatient clinic setting; how-
fitness of older adults with cancer is geriatric assess- ever, older adults with cancer spend little of their time
ment, a multidimensional assessment of older adults.3 in clinics, even during chemotherapy. Instead, they
On the basis of geriatric assessment and other simple spend the majority of their time at home. It is during this
clinical and laboratory results, statistical models with time that symptoms, adverse effects from medical
which to predict chemotherapy toxicities have been cancer treatment, or complications from surgery are
developed that predict outcomes much better than more likely to emerge. Without timely diagnosis and
routine functional assessment methods, such as the management of severe symptoms, a fit patient can
Karnofsky performance scale.4 Components of geri- become frail between clinic visits, leading to hospi-
atric assessment are also associated with other out- talization and a further decrease in fitness level. A
comes, such as early mortality,5 surgical complications, decline in fitness of older adults with cancer may be
and readmission.6 On the basis of this level of evidence, more problematic for those with poor social support,
many organizations and societies, including ASCO, have especially those patients who do not have someone to
Author affiliations
and support
published guidelines and recommendations for the take them to doctor appointments, because these
information (if implementation of geriatric assessment at each phase of patients are at higher risk for under-reporting their
applicable) appear cancer care.7,8 chemotherapy toxicity.10 Traditionally, other methods,
at the end of this
article. Although geriatric oncology has advanced with the such as visiting nursing services or relying on care-
Accepted on March promotion of more comprehensive assessment and givers/patients to call via phone, have been used to fill
4, 2020 and care of older adults with cancer, technology—mainly such gaps in care; however, emerging digital health
published at web-based and wireless technologies—has also ad- technologies may also assist in filling the gap. Another
ascopubs.org on vanced. Digital health technology 9 encompasses use for such tools is the ability to collect a vast and
April 2, 2020:
multiple platforms, such as ePROs, either web or useful amount of data from older adults with cancer
10.1200/EDBK_ application based, wearables and sensors, and remote before, during, and after treatment that can provide
279505 monitoring tools, such as smart homes. Whereas the real-world and precise information about the challenges

Shahrokni, Loh, and Wood
many older adults are not confident in their ability to learn

and use technology devices, in part because of their per-
ception that the technology is too complicated14; therefore,
• To properly apply digital health technology to proactive approaches to increase support for this population
research or clinical practice in older adults with
(e.g., instruction manuals, videos, and staff) are needed.
cancer, carefully consider appropriate mea-
Simple technology devices are also preferred.14 Second,
surement concepts, fit for purpose technolo-
gies, analytic plan, and study designs. older adults are more likely to have impairments that serve
as barriers to using technology devices.15 For example,
• Big data plays a vital role in defining the real-
touchscreen devices may be challenging for those with
world experience of older adults with cancer
during treatment. visual impairment, but this barrier may be mitigated by
large-print keyboard and screen-reading features.15 Such
• Both opportunities and challenges for digital features as larger font sizes, bigger icons, and magnification
health in geriatric oncology are significant.
are also helpful in this population. Despite this known
challenge with touchscreen interfaces, there is less of a
and outcomes of older adults with cancer as they receive performance gap between older and younger adults with
cancer-related treatments. touchscreens compared with keyboards and mice.16 Such
In this article, we will describe the opportunities and findings indicate that touchscreens are an attractive option
challenges of using these solutions for modernizing geriatric for older adults. Older adults are also less anxious when
oncology. using touchscreens versus a standard keyboard terminal.17
Third, certain older adults are skeptical about the impor-
OLDER ADULTS AND ACCEPTANCE OF DIGITAL tance of technology devices.14 In this group, education
HEALTH TECHNOLOGIES about how the technology device gathers information and
Digital health technologies have become increasingly prev- how the health care team uses the information they provide
alent, but their utility, feasibility, and roles may differ in older to guide their management and care may help reinforce
and younger adults. According to the Pew Research Center, their adoption.
in 2017, approximately 42% of adults age 65 and older
BIG DATA TO SHED LIGHT ON REAL-WORLD
reported owning smartphones, which increased from 18% in
OUTCOMES OF OLDER ADULTS WITH CANCER
2013.11 In addition, 67% of adults age 65 and older reported
having internet access, 42% reported having home broad- Older adults with cancer are under-represented in clinical
band, 32% reported having tablets, and 34% reported using trials.18 Published clinical trials also under-report the out-
social media.11 It is important to note that use decreased with comes of older adults with cancer who participate in those
age (i.e., 59% in those age 65 to 69 reported owning clinical trials.19 Moreover, the fitness of those who partici-
a smartphone, and 17% of those age 80 and older reported pate in clinical trials may not fully represent the fitness of
owning a smartphone).11 However, these numbers have average older adults with cancer in the community because
steadily increased and will continue to with the increasing of strict inclusion and exclusion criteria.20 As a result, there
familiarity of older adults with technology as well as the growth is a need for alternative methods by which to assess the
of the population. outcomes of older adults with cancer who resemble average
older adults with cancer in the community.
Not all older adults are afraid of technology. In a survey
conducted by the Pew Research Center, two distinct groups Over the past decade, big data has emerged as a method of
of older adults in the United States were identified. The assessing the outcomes of patients with cancer in general
younger, more educated, and wealthier group was more and in older adults with cancer specifically. In general, we
likely to use technology devices and have a favorable view define big data as large and complex data sets that grow so
of them. In contrast, the older, less wealthy, and more rapidly that existing analytical tools fail to take advantage of
comorbid group was more disconnected from the digital the full potential of big data.21 The CancerLinQ project,
world.12 Nevertheless, most older adults are willing to learn which started in 2012 and is led by ASCO, is probably the
and adopt new technology13,14 and should therefore not be most prominent initiative to develop oncologic big data.22
left out. With the goal of rapid learning from patients and their
outcomes, CancerLinQ goals are to collect and aggregate
CHALLENGES TO THE USE OF DIGITAL HEALTH data from patient records, analyze the data, and develop
TECHNOLOGIES FOR OLDER ADULTS real-time decision support systems for clinicians, as well as
There are several challenges that older adults face when to explore the data to generate relevant hypotheses. In
adopting technology devices. It is essential to consider these 2015, CancerLinQ was officially implemented and adopted
challenges in both the clinical and research settings. First, by 85 community and academic oncology practices.23

Digital Health and Geriatric Oncology
Additional progress was made by linking CancerLinQ to cancer, many oncologists are uncertain about how to
other data sets, such as the National Cancer Institute Sur- evaluate and implement these technologies in practice.
veillance, Epidemiology, and End Results program.24 Recently, Here, we propose a practical guide for the oncology re-
data are emerging on the utility of CancerLinQ to better un- searcher or clinician. We focus the discussion on sensors
derstand the outcomes of patients with cancer and their and specifically on those that are part of connected bio-
contributing factors. For example, using the CancerLinQ data metric monitoring technologies (BioMeTs).33 We focus on
set, the prevalence of autoimmune disease among patients BioMeTs as these types of devices and systems, which can
receiving immune checkpoint inhibitors (ICIs) was assessed. capture continuous physiologic data from patients for
The prevalence of autoimmune disease was between 23% prolonged periods, are probably the most relevant to po-
and 27% for patients with lung cancer and other patients who tential use cases in older adults with cancer. A BioMeT can
were receiving ICIs—much higher than expected, especially detect and monitor one or more metrics, such as activity,
because a majority of clinical trials for ICIs exclude patients with sleep, heart rate, heart rhythm, respiratory rate, and others.
autoimmune diseases.25 As a result, CancerLinQ provides When considering the deployment of a BioMeT in research
a great opportunity to better understand the outcomes of older or practice, it is essential to first consider the potential
adults who receive ICIs despite having autoimmune diseases. context of use and, from there, the health concept of in-
As older adults with cancer become more tech savvy, many terest. Then, potential metrics that are needed to measure
health care institutions are transforming their paper-format this health concept of interest at baseline and over time can
questionnaires to electronic and web-based formats for be identified. In the setting of geriatric oncology, as noted
ease of data capturing. At Memorial Sloan Kettering Cancer earlier in this work, potential contexts of use include baseline
Center, geriatric assessment has been converted into a web- frailty assessments (e.g., comprehensive geriatric assessments34
based questionnaire and named electronic rapid fitness or eRFAs)35 and monitoring for functional deterioration or ad-
assessment (eRFA).26 A pilot study on more than 600 older verse events between clinic visits. The best current example of
adults with cancer demonstrated that eRFA can be com- the latter is likely home-based ePRO monitoring,31 although the
pleted in approximately 10 minutes, and an overwhelming robustness of this approach in older individuals with cancer is still
majority of older adults with cancer expressed their satis- being evaluated.32
faction with completing eRFA in the geriatrics clinics.26
Measuring Physical Function With BioMeTs
Subsequently, Memorial Sloan Kettering Cancer Center tho-
racic surgery and bone marrow transplantation services suc- To augment the use of baseline frailty assessments and
cessfully implemented eRFA at a point of care.27,28 between-visit monitoring, a BioMeT that specifically eval-
uates physical function may be of interest. An abundance of
ePROs are another method with which to capture a large
literature highlights the capacity of physical function or
amount of data inside and outside of clinics, which can be
performance status to predict morbidity, resource utiliza-
useful for both research and clinical practice.29 Investigators
tion, and mortality in patients with cancer.36-39 A frailty
have shown that the use of ePROs for assessing and
assessment, at its root, has the exploration of participant
subsequently managing symptoms improves the quality of
physical function, and between-visit deterioration or adverse
life and overall survival of patients.30,31 However, with par-
events can reasonably be expected to correlate with con-
ticipants’ median age of 62, the feasibility and efficacy of
current physical function changes. Thus, in characterizing
ePROs for older adults with cancer was unknown. To address
baseline physical function and changes in function over
this question, Nipp and colleagues32 performed a secondary
time, much like an early warning system,40,41 a BioMeT
analysis and found that overall survival and the rate of
could augment the current roles of frailty assessments and
emergency room visits were similar between older adults with
PRO monitoring.
cancer who were reporting their symptoms via ePRO and
those who were observed by routine care. In contrast, the use Once a health concept of interest has been identified, one or
of ePROs has led to a reduction in hospitalization and an more metrics that help to measure this concept should be
improvement in quality of life. This study highlights the need ascertained. In some cases, the best metrics may not be
for additional exploration and expansion of ePROs for the obvious, and additional development may be needed to
assessment and care of older adults with cancer, especially construct composite measures derived from BioMeTs.
outside of clinical trials in which participants are more similar Physical function is likely to be an example of this. On the
to average older adults with cancer in the community. basis of previous research, we know that physical activity is
related to physical function, and thus a measure of activity
SELECTION AND USE OF WEARABLE SENSORS IN (e.g., steps per day) at baseline and over time can tell us
ONCOLOGY PRACTICE something about a participant’s function.42 However, we
Although digital health technologies have the potential to also know that activity is behaviorally influenced and that the
augment clinical care and research in older adults with relationship between function and activity is not entirely

overlapping. Modern BioMeTs contain additional information agreements, terms of service, and privacy policies, which
that can help to correct for this. For example, resting heart specify the rights of users and the ability of device manu-
rate, heart rate response to activity increases or decreases, facturers to monitor, aggregate, and share users’ digital
respiratory variation as a function of heart rate and activity, biospecimens. Utility and usability help to determine
and other physiologic data are related to physical whether a BioMeT has the features that users need and
function,43,44 are obtainable from BioMeTs,45 and could be evaluate the user experience with these features. These
included within a BioMeT-derived physical function mea- include human factor considerations, such as battery life,
sure. Furthermore, other types of clinical outcomes as- water resistance, form, user interface, and other issues.
sessments may also provide nonoverlapping data to give Finally, economic feasibility refers to the business model
additional precision around physical function assessment, and fees of the sensor, which may include a subscription or
such as self-reported physical function (e.g., PROMIS46,47), or long-term fees around data storage and analysis. Many
the results of periodic standardized physical assessments, frameworks, such as that described here, include ways to
such as a 6-minute walk test48 or short physical performance evaluate how a candidate-connected sensor technology
battery.49,50 In other words, it may be possible to construct performs against the five dimensions and may include vi-
a model around physical function prediction that includes, sualizations to help understand this, like a nutrition label for
but is not limited to, data from BioMeTs while incorporating a particular technology or a side-by-side comparison.
information from other clinical outcomes assessments.
Considerations for Analytic Plans and Study Designs Using
Selecting Connected Health Technologies That Are Fit Connected Sensor Technologies
for Purpose After a fit-for-purpose technology has been identified, the
When a health concept of interest (e.g., physical function) user must determine an analytic plan for measuring the
and potential metrics of interest (e.g., activity, heart rate, or health concept of interest using the chosen metrics with
respiratory rate) have been identified, a user can start to look the fit-for-purpose technology. This is not an easy task
for a device that is fit for purpose to collect the identified considering that large volumes of data are continuously
metric(s). Here, it is essential to note that the marketplace is generated from these devices. Missing data are often
filled with a variety of connected sensor technologies that a concern, as is the context within which the data are
exist throughout consumer-oriented and health-directed generated, as well as their comparisons to gold standard
spaces. The quality of data that these sensors generate is metrics.52 Furthermore, as in our example of physical
variable, and some connected sensor technologies have function, how to combine metrics to appropriately measure
other vulnerabilities or concerns that limit their appropri- the concept of interest is not always obvious. Pilot studies
ateness for use with research or clinical participants. Several may be appropriate for model development and follow-up
organizations have attempted to develop frameworks to validation studies before deploying the composite metric in
evaluate whether a device is fit for purpose. One such a larger population as part of a larger study. Machine
framework includes five key dimensions for evaluation: learning techniques may facilitate the aggregation and
validation, security practices, data rights and governance, modeling of high dimensionality, multivariable data, such as
utility and usability, and economic feasibility.33 that generated by connected sensor technologies.53
In this framework, validation is used as a shorthand for After these steps, it is critical to develop an appropriate study
a three-stage process of verification, analytical validation, design and enrollment and monitoring strategy. This is
and clinical validation, collectively known as V3.51 Verifi- especially true with connected sensor technologies, as
cation refers to sample-level data generated by a sensor a variety of factors influence adherence to wearing the
within a BioMeT against a prespecified set of criteria. An- BioMeT and the willingness of sites and participants to
alytical validation looks at how well the BioMeT component participate in a project involving these technologies. The
algorithms perform in measuring, detecting, or predicting Clinical Trials Transformation Initiative, a public/private
physiologic metrics. Finally, clinical validation determines partnership with a mission to “develop and drive adoption of
whether the BioMeT identifies, measures, or predicts a practices that will increase the quality and efficiency of
meaningful clinical, biologic, physical, functional state, or clinical trials,”54 recently developed recommendations around
experience in a specified population and context of use. optimizing mobile clinical trials by engaging patients and
Without a robust V3 process, it is difficult to ascertain the sites.55 When considering the deployment of BioMeTs in ge-
quality and reliability of data generated by a sensor and riatric oncology, a researcher or clinician might not have
associated BioMeT. Security practices refers to the degree a multicenter clinical trial in mind. With that said, the Clinical
to which a BioMeT is protected from unauthorized access Trials Transformation Initiative recommendations, which cover
and attacks, including human error. Data rights and gov- topics that include engaging patients and sites in planning
ernance concern such processes as end-user license clinical trials using mobile technology (e.g., protocol design,

technology selection, and pilot testing), maximizing value and based system was used to deliver interventions on the basis
minimizing burden for study participants (e.g., setting ex- of geriatric assessment impairments.15 For example, if
pectations, protecting privacy, returning individual data, en- a patient has impairment in physical function, he or she was
hancing participant-site interactions, and providing technical educated about energy conservation and exercise and were
support), and addressing challenges for investigative sites asked to track daily steps. If patients were on five or more
(e.g., contracting and budgeting recommendations, evaluating medications or screened positive for cognitive impairment,
site readiness, and implementing practical and streamlined medication monitoring was instituted. The intervention was
training), may still be instructive in planning for any study that demonstrated to be feasible and usable. Loh and colleagues15
involves connected sensor technologies. also found challenges specific to older adults that were used to
At one of our institutions (University of North Carolina), we adapt the intervention, including reliable internet access;
have followed this basic framework in developing a health providing a stylus; and incorporating nonmedical functions,
coaching program for patients with cancer or survivors of such as daily jokes or words, to increase motivation. Hong and
cancer of all ages. Participants undergo baseline evaluation, colleagues57 evaluated an interactive mobile-enabled web app
which includes several components that are similar to ge- to promote physical activity among older survivors of cancer in
riatric assessment remote PRO assessment, including symp- a single-arm pilot. Of the 26 patients, two patients were on
toms, physical function, and unmet needs during treatment and active treatment. Participants had overall positive experiences
survivorship; and BioMeT deployment, with regular monitoring with the mobile app, with improvement found in self-rated
of activity, heart rate, and other metrics. With this program, we health, quality of life, and sleep quality. Specific suggestions to
can concurrently develop composite metrics of physical function improve the system include automated functions (e.g., an
from the BioMeTs using data iteratively acquired from partici- ability to enter activity automatically). Finally, a third study
pants throughout our 6-month intervention. evaluated the differential effects of an electronic symptom
monitoring intervention for patients with advanced cancer.
EMERGING DATA ON THE USE OF DIGITAL HEALTH Nipp and colleagues found that older adults may not derive the
TECHNOLOGIES FOR OLDER ADULTS WITH CANCER same benefits from an electronic symptom monitoring in-
Several studies have investigated or are currently investigating tervention as younger adults.32 Specifically, both the older and
the use of digital health technologies for older adults with younger populations in the intervention arm had lower risks of
cancer. Their functions may include one or more of the fol- hospitalization and better quality of life, but emergency room
lowing: monitor cancer- and treatment-related adverse effects; visits and survival were similar between the intervention and
monitor vital signs, such as blood pressure, weight, and sleep; control arms in the older age group. This study highlights the
manage daily activities (e.g., medications and appointments); need to tailor technology-based interventions to the unique
provide education; and promote healthy lifestyles (e.g., needs of older adults.
physical activity). Studies in the oncology setting have been
CONCLUSION
previously summarized, and we highlight studies that are
specific to older adults with cancer.9 Advances in technology and digital health solutions provide
a great opportunity to further advance the field of geriatric
In terms of wearable sensors, in a pilot study of 40 adults age
oncology. However, additional studies are needed to
65 and older starting first-line chemotherapy, Soto and
explore proper solutions, context, patient population, and
colleagues56 found that using a smartphone with a pe-
outcomes.
dometer application to monitor daily steps was feasible. If
patients had a decline in daily steps, they were contacted for ACKNOWLEDGMENT
assessment of symptoms and chemotherapy-related tox- Dr. Loh receives support from the National Cancer Institute
icity. A randomized controlled trial is currently ongoing (K99CA237744) and Wilmot Research Fellowship Award.
(ClinicalTrials.gov identifier: NCT04040881). This work was supported by the Memorial Sloan Kettering
Three studies evaluated mobile health applications in older Summer Research Fellowship Grant 5R25CA020449 and
adults with cancer. In the first study of 18 older adults National Institutes of Health/National Cancer Institute Cancer
receiving cancer treatment and their caregivers, a tablet- Center Support Grant P30CA008748.

1
Memorial Sloan Kettering Cancer Center, New York, NY Armin Shahrokni, MD, MPH, Geriatrics Service, Department of Medicine,
2
James P. Wilmot Cancer Institute, University of Rochester School of Memorial Sloan Kettering Cancer Center, Box 205, 1275 York Ave., New
Medicine and Dentistry, Rochester, NY York, NY 10065; Twitter: @MSK_GeriOnc; email: [email protected].
3
Lineberger Comprehensive Cancer Center, University of North Carolina at
Chapel Hill, Chapel Hill, NC


EDBK_279505.
REFERENCES
1. Daniels N. Global Aging and the Allocation of Health Care Across the Life Span. Abingdon, United Kingdom: Taylor & Francis; 2013.
2. White MC, Holman DM, Boehm JE, et al. Age and cancer risk: a potentially modifiable relationship. Am J Prev Med. 2014;46:S7-S15.
3. Korc-Grodzicki B, Sun SW, Zhou Q, et al. Geriatric assessment as a predictor of delirium and other outcomes in elderly patients with cancer. Ann Surg. 2015;
261:1085-1090.
4. Hurria A, Mohile S, Gajra A, et al. Validation of a prediction tool for chemotherapy toxicity in older adults with cancer. J Clin Oncol. 2016;34:2366-2371.
5. Soubeyran P, Fonck M, Blanc-Bisson C, et al. Predictors of early death risk in older patients treated with first-line chemotherapy for cancer. J Clin Oncol. 2012;
30:1829-1834.
6. Kristjansson SR, Nesbakken A, Jordhøy MS, et al. Comprehensive geriatric assessment can predict complications in elderly patients after elective surgery for
colorectal cancer: a prospective observational cohort study. Crit Rev Oncol Hematol. 2010;76:208-217.
7. Mohile SG, Dale W, Somerfield MR, et al. Practical assessment and management of vulnerabilities in older patients receiving chemotherapy: ASCO guideline for
geriatric oncology. J Oncol Pract. 2018;14:442-446.
8. Extermann M, Aapro M, Bernabei R, et al. Use of comprehensive geriatric assessment in older cancer patients: recommendations from the task force on CGA of
the International Society of Geriatric Oncology (SIOG). Crit Rev Oncol Hematol. 2005;55:241-252.
9. Fallahzadeh R, Rokni SA, Ghasemzadeh H, et al. Digital health for geriatric oncology. JCO Clin Cancer Inform. 2018;2:1-12.
10. Shahrokni A, Sun CL, Tew WP, et al. The association between social support and chemotherapy-related toxicity in older patients with cancer. J Geriatr Oncol.
2020;11:274-279.
11. Badgwell B, Stanley J, Chang GJ, et al. Comprehensive geriatric assessment of risk factors associated with adverse outcomes and resource utilization in cancer
patients undergoing abdominal surgery. J Surg Oncol. 2013;108:182-186.
12. Carneiro D, Novais P, Pêgo JM, et al. Using mouse dynamics to assess stress during online exams. In: International Conference on Hybrid Artificial Intelligence
Systems. Bilbao, Spain: Hybrid Artificial Intelligent Systems; 2015; pp 345-356.
13. Hoogland AI, Mansfield J, Lafranchise EA, et al. eHealth literacy in older adults with cancer. J Geriatr Oncol. Epub 2020 Jan 6.
14. Vaportzis E, Clausen MG, Gow AJ. Older adults perceptions of technology and barriers to interacting with tablet computers: a focus group study. Front Psychol.
2017;8:1687.
15. Loh KP, Ramsdale E, Culakova E, et al. Novel mHealth app to deliver geriatric assessment-driven interventions for older adults with cancer: pilot feasibility and
usability study. JMIR Cancer. 2018;4:e10296.
16. Findlater L, Froehlich J, Fattal K, et al. Age-Related Differences in Performance with Touchscreens Compared to Traditional Mouse Input. In CHI 2013. Paris,
France: Association for Computing Machinery; 2013;343-346.
17. Umemuro H. Lowering elderly Japanese users’ resistance towards computers by using touchscreen technology. Univers Access Inf Soc. 2004;3:276-288.
18. Hurria A, Levit LA, Dale W, et al. Improving the evidence base for treating older adults with cancer: American Society of Clinical Oncology statement. J Clin Oncol.
2015;33:3826-3833.
19. BrintzenhofeSzoc K, Krok-Schoen JL, Canin B, et al. The underreporting of phase III chemo-therapeutic clinical trial data of older patients with cancer. J Geriatr
Oncol. Epub 2020 Jan 10.
20. Kim ES, Bruinooge SS, Roberts S, et al. Broadening eligibility criteria to make clinical trials more representative: American Society of Clinical Oncology and Friends
of Cancer Research joint research statement. J Clin Oncol. 2017;35:3737-3744.
21. Andreu-Perez J, Poon CC, Merrifield RD, et al. Big data for health. IEEE. 2015;19:1193-1208.
22. Sledge GW Jr., Miller RS, Hauser RJ. CancerLinQ and the future of cancer care. Am Soc Clin Oncol Educ Book. 2013;33:430-434.
23. Rubinstein SM, Warner JL. CancerLinQ: origins, implementation, and future directions. JCO Clin Cancer Infor. Epub 2018 Feb 16.
24. Rivera D, Rubinstein WS, Schussler NC, et al. NCI and ASCO CancerLinQ collaboration to advance quality of cancer care and surveillance. J Clin Oncol. 2019;37:
15s (suppl; abstr e18317).
25. Chen L, Walker MS, Zhi J, et al. Real-world prevalence of autoimmune disease (AD) among patients (pts) receiving immune checkpoint inhibitors (ICI) in ASCO’s
CancerLinQ database. J Clin Oncol. 2019;37:15s (suppl; abstr 6583).
26. Shahrokni A, Tin A, Downey RJ, et al. Electronic rapid fitness assessment: a novel tool for preoperative evaluation of the geriatric oncology patient. J Natl Compr
Canc Netw. 2017;15:172-179.
27. Lin RJ, Dahi PB, Shahrokni A, et al. Feasibility of a patient-reported, electronic geriatric assessment tool in hematopoietic cell transplantation–a single institution
pilot study. Leuk Lymphoma. 2019;60:3308-3311.

28. Downey RJ, Korc-Grodzicki B, Weber R, et al. Assessing the clinical feasibility of implementing a novel assessment of frailty: the electronic rapid fitness
assessment in diverse thoracic surgery clinics. J Clin Oncol. 2017;35:15s (suppl; abstr e21687).
29. Jensen RE, Snyder CF, Abernethy AP, et al. Review of electronic patient-reported outcomes systems used in cancer clinical care. J Oncol Pract. 2014;10:e215-e222.
JAMA. 2017;318:197-198.
Oncol. 2016;34:557-565.
32. Nipp RD, Horick NK, Deal AM, et al. Differential effects of an electronic symptom monitoring intervention based on the age of patients with advanced cancer. Ann
Oncol. 2020;31:123-130.
33. Coravos ADM, Doerr M, Goldsack J, et al. Modernizing and designing evaluation frameworks for connected sensor technologies in medicine. NPJ Digit Med. 2020;3:37.
34. Hurria A, Cirrincione CT, Muss HB, et al. Implementing a geriatric assessment in cooperative group clinical cancer trials: CALGB 360401. J Clin Oncol. 2011;
29:1290-1296.
35. Shahrokni A, Tin A, Downey RJ, et al. Electronic rapid fitness assessment: a novel tool for preoperative evaluation of the geriatric oncology patient. J Natl Compr
Canc Netw. 2017;15:172-179.
36. Atkinson TM, Andreotti CF, Roberts KE, et al. The level of association between functional performance status measures and patient-reported outcomes in cancer
patients: a systematic review. Support Care Cancer. 2015;23:3645-3652.
37. Evers PD, Logan JE, Sills V, et al. Karnofsky performance status predicts overall survival, cancer-specific survival, and progression-free survival following radical
cystectomy for urothelial carcinoma. World J Urol. 2014;32:385-391.
38. Sorror M, Storer B, Sandmaier BM, et al. Hematopoietic cell transplantation-comorbidity index and Karnofsky performance status are independent predictors of
morbidity and mortality after allogeneic nonmyeloablative hematopoietic cell transplantation. Cancer. 2008;112:1992-2001.
39. Su J, Barbera L, Sutradhar R. Do repeated assessments of performance status improve predictions for risk of death among patients with cancer? A population-
based cohort study. Palliat Med. 2015;29:547-553.
40. Fringer A, Arrer E, Maier E, et al. Development of an early warning system to prevent crises in the palliative home care setting of patients and their informal
caregivers: protocol for a mixed method study. JMIR Res Protoc. 2019;8:e13933.
41. Hu SB, Wong DJ, Correa A, et al. Prediction of clinical deterioration in hospitalized adult patients with hematologic malignancies using a neural network model.
PLoS One. 2016;11:e0161401.
42. Bennett AV, Reeve BB, Basch EM, et al. Evaluation of pedometry as a patient-centered outcome in patients undergoing hematopoietic cell transplant (HCT):
a comparison of pedometry and patient reports of symptoms, health, and quality of life. Qual Life Res. 2016;25:535-546.
43. Groarke JD, Mahmood SS, Payne D, et al. Case-control study of heart rate abnormalities across the breast cancer survivorship continuum. Cancer Med. 2019;8:447-454.
44. Ha D, Malhotra A, Ries AL, et al. Heart rate variability and heart rate recovery in lung cancer survivors eligible for long-term cure. Respir Physiol Neurobiol. 2019;269:103264.
45. Perez MV, Mahaffey KW, Hedlin H, et al; Apple Heart Study Investigators. Large-scale assessment of a smartwatch to identify atrial fibrillation. N Engl J Med.
2019;381:1909-1917.
46. Jensen RE, Potosky AL, Reeve BB, et al. Validation of the PROMIS physical function measures in a diverse US population-based cohort of cancer patients. Qual
Life Res. 2015;24:2333-2344.
47. Jensen RE, Potosky AL, Moinpour CM, et al. United States population-based estimates of patient-reported outcomes measurement information system symptom
and functional status reference values for individuals with cancer. J Clin Oncol. 2017;35:1913-1920.
48. Enright PL. The six-minute walk test. Respir Care. 2003;48:783-785.
49. Klepin HD, Geiger AM, Tooze JA, et al. Geriatric assessment predicts survival for older adults receiving induction chemotherapy for acute myelogenous leukemia.
Blood. 2013;121:4287-4294.
50. Guralnik JM, Simonsick EM, Ferrucci L, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and
prediction of mortality and nursing home admission. J Gerontol. 1994;49:M85-M94.
51. Goldsack J. Verification, analytical validation, and clinical validation (V3): the foundation of determining fit-for-purpose for biometric monitoring technologies
(BioMeTs). NPJ Digit Med. In press.
52. Izmailova ES, Wagner JA, Perakslis ED. Wearable devices in clinical trials: hype and hypothesis. Clin Pharmacol Ther. 2018;104:42-52.
53. Meng Y, Speier W, Shufelt C, et al. A machine learning approach to classifying self-reported health status in a cohort of patients with heart disease using activity
tracker data. IEEE J Biomed Health Inform. 2020;24:878-884.
54. Tenaerts P, Madre L, Archdeacon P, et al. The Clinical Trials Transformation Initiative: innovation through collaboration. Nat Rev Drug Discov. 2014;13:797-8.
55. Clinical Trials Transformation Initiative. CTTI recommendations: optimizing mobile clinical trials by engaging patients and sites. https://www.ctti-clinicaltrials.org/
sites/www.ctti-clinicaltrials.org/files/ctti_recommendations_-_mct_engaging_patients_and_sites_final.pdf. Accessed March 18, 2020.
56. Soto-Perez-De-Celis E, Kim H, Rojo-Castillo MP, et al. A pilot study of an accelerometer-equipped smartphone to monitor older adults with cancer receiving
chemotherapy in Mexico. J Geriatr Oncol. 2018;9:145-151.
57. Hong YA, Goldberg D, Ory MG, et al. Efficacy of a mobile-enabled Web app (iCanFit) in promoting physical activity among older cancer survivors: a pilot study.
JMIR Cancer. 2015;1:e7.

GERIATRIC ONCOLOGY
The Globalization of Geriatric Oncology: From

Data to Practice
Ravindran Kanesvaran, MD1; Supriya Mohile, MD2; Enrique Soto-Perez-de-Celis, MD3; and Harpreet Singh, MD4
Older adults with cancer are a unique group of patients who require specialized care and treatment. The field
overview
of geriatric oncology (GO) was born more than 3 decades ago to address the needs of this growing group of
patients. Some challenges in the GO field include establishing a GO clinical service, educating and training
personnel, and conducting research in GO. These issues are addressed to varying extents with global initiatives
in GO, which are largely dependent on the socioeconomic status of the countries involved. To overcome
disparities seen globally, scientific journals that reach an international cancer audience should publish
content related to improving care of older adults with cancer around the world, develop an organizational
structure that encourages global dissemination of GO knowledge, and advance reporting policies that en-
courage higher-quality reporting of data relevant to older adults with cancer worldwide. A number of in-
ternational scientific journals have risen to the occasion to address these disparities. A key battle in enabling
access of this vulnerable group of patients to clinical trials is now being fought and won on the global front with
numerous regulatory initiatives. The U.S. Food and Drug Administration (FDA) Oncology Center of Excellence
(OCE) recently issued draft guidance on the inclusion of older adults in cancer clinical trials. This and other
global initiatives led by the FDA have the potential to further improve the evidence base for older adults with
cancer.
GERIATRIC ONCOLOGY INITIATIVES ACROSS THE GLOBE is forthcoming. The latest document, coauthored by
Cancer Demographics Worldwide: The Implications of a list of global experts representing SIOG, aims to
Epidemiologic Transitions address four key domains (education, clinical practice,
research, and collaborations and partnerships) with
The global population is not only growing rapidly but is a list of 12 priorities.
also aging at an accelerated pace. The number of
people aged 65 or older is projected to grow from 703 Treating an older adult with cancer is complex and
million in 2019 to 1.5 billion in 2050.1 This impending requires multidisciplinary teams.5 Since the inception
explosion in population aging comes with a rising of GO, many studies have demonstrated the benefit of
number of older adults who have multiple comorbid- having geriatric assessments and interventions in
ities and chronic illnesses such as cancer. More than terms of influencing treatment selection, prognosis,
60% of all cancers are diagnosed for adults age 60 or and prediction of chemotherapy toxicity.6,7 However,
older.2 In view of this confluence of cancer and aging there are very few cancer centers around the world that
and the complexities that arise from treating an older have a dedicated GO clinic or program, and most are
adult with cancer, the field of GO was born in the 1980s located in high-income countries (HICs).5,8 Most
with pioneering work done by Dr. B.J. Kennedy and leading GO centers have a shared care model that
colleagues.3 incorporates a geriatrician or an oncologist who is
Author affiliations
and support
dually trained in geriatrics (geriatric oncologist) pro-
GO went global with the formation of the International
information (if viding recommendations to the primary oncologist as
Society of Geriatric Oncology (SIOG) in 2000. SIOG
applicable) appear part of the patient’s cancer care. However, this model is
at the end of this
now consists of more than 1,700 members in more
not possible in most countries globally (including some
article. than 80 countries. The SIOG annual scientific meeting
HICs), owing to the lack of geriatricians as well as
Accepted on April is a platform at which many of the educational and
oncologists trained in geriatrics to assist in tailoring the
13, 2020 and research initiatives pertaining to GO are discussed.
published at treatment of these older adults with cancer.5
Apart from this, SIOG also spearheaded the SIOG 10
ascopubs.org on
Priorities Initiative in 2011 to define the top priorities for Although it is very challenging to engage countries to
April 28, 2020:
the development of GO globally.4 This initiative is increase their training of geriatricians and geriatric
10.1200/EDBK_ currently being revised for the first time to reflect the oncologists in their respective populations, key
279513 changes in the field in the past decade and publication stakeholders like SIOG have created short, advanced

Kanesvaran et al
has contributed significantly to the development of the field

in areas ranging from clinical trial design10 to prognostic and
predictive tools (Cancer and Aging Research Group score7)
• GO initiatives across the globe are revolutionizing that are now used globally. ASCO has played a key role in
the way older adults with cancer are being treated.
accelerating the development of GO by initiating the B.J.
• Leading oncology societies have now in- Kennedy Award in 2007 and also offering a robust GO
corporated GO as part of their global oncology oncology program as part of its annual meeting.
curriculum and have provided recommenda-
tions on how to improve GO clinical practice, Europe is another part of the world where GO has a strong
training, and research. foothold. Because the European continent has the highest
• There is a global regulatory effort led by the FDA population share of older adults in the world, it is not sur-
to increase the evidence base for older adults prising that many European countries have mature GO
with cancer. programs.9 These programs range from specialized GO
• In view of the substantial proportion of patients clinics whose management is driven by national guidelines
with cancer being older adults, it is vital for (Italy, France) to GO departments within tertiary centers
oncologists globally to embrace GO: “every (Greece, Switzerland). Older adults with cancer are also
oncologist is a geriatric oncologist.” encouraged to enroll in clinical trials in France with the
formation of dedicated research groups like DIALOG.11
From a research angle, there are a number of large col-
courses to bridge this gap and expand the number of on- laborative efforts in Europe looking at under-representation
cologists and geriatricians who are familiar with GO prin- of older adults with cancer and its impact on cancer care for
ciples in clinical practice.9 These courses, which were this vulnerable group of patients.9
initially available only in Europe (Treviso, Italy), will now be
available globally. In Oceania, Australia is currently the only HIC offering
comprehensive GO services in a limited number of cen-
GO Initiatives Around the World ters.12 Starting with the GO service in Adelaide, there are
The field of GO, like any other specialty in medicine, is very now GO consultation services (mainly hospital based) in
resource dependent. A key resource that is lacking globally most major cities in Australia.13 Although there is no formal
is skilled personnel (presence of multidisciplinary teams); GO training available in Australia, a number of GO spe-
hence, it is not surprising that most of the GO initiatives are cialists have undertaken training in both geriatrics and
in HICs.9 Another reason for this is that older adults rep- oncology. The Clinical Oncology Society of Australia leads
resent a higher share of the population of HICs compared the way in terms of GO research in Australia, and GO re-
with low- and middle-income countries (LMICs). In the next search is actively undertaken by many of its members in
few paragraphs, we will broadly discuss GO initiatives based academic institutions all over the country.9
on the income level of the country according to the World With more than half of the global cancer incidence and mortality
Bank (Fig. 1). in Asia, the stage was set for rapid growth in establishing GO
The United States and Canada have some of the most services in many countries across the region.14 HICs like South
developed GO services in the world with a few different Korea, Japan, Hong Kong, and Singapore have GO clinical
models of care. Some cancer centers provide compre- services that follow a consultative model with multidisciplinary
hensive cancer care for older adults with cancer with the GO input. These countries also have robust GO research platforms
team providing care from diagnosis to treatment and follow- that are constantly exploring ways to improve the care of
up (Moffit Cancer Center), whereas the majority provide GO older adults in their respective countries.6,15 Asian LMICs
consultative services with the primary oncologist driving such as China and India have also shown keen interest in
cancer-specific care (Memorial Sloan Kettering, University building GO services within their respective countries de-
of Rochester, McGill University).9 The model selected by the spite challenges they face related to limited resources.16,17
GO centers depends not only on the resources available but To help LMICs recognize the benefits of having GO services
also on the preferences of the geriatric oncologist driving the within their health care systems, SIOG organized its first
development of the service. In terms of training, there are Asian public policy meeting in the Philippines in May 2019.
a number of pathways as well. There are a limited number of This meeting brought together key Ministry of Health offi-
formal GO training fellowships in the United States and cials from nine Southeast Asian countries to highlight the
Canada (e.g., McGill University, University of North Caro- key benefits GO services can provide to the growing number
lina), whereas others undergo separate oncology and ge- of older adults with cancer in their countries. This meeting
riatric fellowships before getting into the GO field. Research was coupled with a GO advanced course for oncologists and
from a number of collaborative North American GO groups geriatricians in the Philippines.

The Globalization of Geriatric Oncology
FIGURE 1. Global Geriatric Oncology Initiatives by Continent

Abbreviation: LMICs, low- and middle-income countries.
Latin America is also seeing a rapidly growing population of of research to improve outcomes of older adults with cancer
older adults with cancer. Recognizing the importance of GO and to improve care delivery in GO have been developed
in the clinical care of these patients, Brazil started the first and implemented in HICs, and there remain inequalities in
two GO clinics in 2011 and 2012, respectively. In Mexico, GO initiatives between LMICs and HICs.9 Understanding the
the first multidisciplinary GO clinic was started in 2015. successes and challenges of GO in all countries around the
Whereas the GO clinics in Brazil were based on a consul- world and fostering international collaboration in research is
tative model with a geriatrician performing the assessment essential for improving the care of all older adults with
and providing the input to the treating oncologist, a com- cancer. Because GO is a “disparity within a disparity,”18
prehensive multidisciplinary model was used in Mexico scientific journals that reach an international cancer audi-
instead.9 ence should (1) publish content related to improving care of
older adults with cancer around the world, (2) develop an
In summary, older adults with cancer globally form a bur-
organizational structure that encourages global dissemi-
geoning group within our patient population. This chal-
nation of GO knowledge, and (3) advance reporting policies
lenging situation must be addressed urgently before the
that encourage higher-quality reporting of data relevant to
health outcome disparities between older adults and their
older adults with cancer worldwide.
younger counterparts become untenable.18 It is heartening
to note that there is substantial effort from interested groups Content Scientific journals with an international reach and
all over the world who have recognized this problem and focus should prioritize publication of initiatives in care de-
taken positive steps to address it. It is important for these livery research and education that are relevant to older
groups to work together to help shed light on the importance adults with cancer living in both HICs and LMICs. In ad-
of GO in the care of older adults with cancer worldwide. dition, scientific journals should solicit and engage our in-
ternational audience to provide perspectives from LMICs on
Global dissemination of GO knowledge through scientific how to best care and promote research for older adults with
journals Aging is a global issue, and the majority of older cancer in these areas of the world.
adults will live in LMICs. Because of the growing numbers of
older adults with cancer in LMICs, scientific journals should Care Delivery Research
prioritize efforts to ensure that scientific progress addresses Over the past 5 years, there has been an increase in GO
the needs of older adults with cancer in LMICs. The majority clinics and programs in HICs. Although a few centers

Kanesvaran et al
(mostly in the United States) use geriatric oncologists (dually chemotherapy in the United States,23 did not identify older
trained geriatricians and oncologists) to deliver aging- Australians at risk for toxicity.24 Prioritizing research to
sensitive care to vulnerable and frail older adults with understand how epidemiology, access to care, and local
cancer, the demand for care greatly outstrips the availability clinical practice influence the accuracy of predictive models
of dually trained clinicians, even in HICs.8,9 Other models and availability of geriatric-specific interventions will help to
incorporate a geriatrician or geriatrician oncologist who disseminate and implement GO models of care worldwide.21
delivers consultations at initial decision-making8 (in support
Education
of ASCO GO guidelines19) or along the disease trajectory.
Even in these models, the need is greater than the avail- Despite the worldwide need, the number of clinicians
ability of clinicians who have specialized training in aging. A trained in geriatric medicine remains low. In the United
large cluster-randomized study in the United States that States, a high number of physician fellowship positions in
trained staff in oncology clinics to conduct geriatric as- geriatrics are not filled.25
sessment showed that providing geriatric assessment in- Geriatric medicine is not recognized as a distinct specialty in
formation to oncologists improved communication about many areas of the world, especially LMICs where resources
aging among older adults with cancer and their caregivers.20 are scarce.9 Although awareness has increased, there also
Patients and caregivers in the intervention arm were also remains a need to train advanced practice providers,
more satisfied with care. This study demonstrates the nurses, and allied health team members who care for older
feasibility of integrating GO principles and care into com- patients with cancer in the principles of geriatrics.26 The lack
munity oncology clinics without requiring the presence of of emphasis in health care systems for growing a workforce
a geriatrician or dually trained geriatric oncologist. The trained in how to care for older adults with cancer adversely
majority of countries continue to lack resources to integrate affects the ability to develop clinical care initiatives that
GO principles and care into treatment decision-making to improve outcomes for this vulnerable population. Despite
best address the needs of older patients with cancer, and efforts to integrate geriatrics content in curricula for clini-
most oncology teams practice in settings where geriatric cians in training and sponsor training opportunities for
expertise and training is not available.18 clinicians in practice, the majority still do not include ad-
There have been efforts to increase an understanding of the equate geriatrics content to properly prepare clinicians to
epidemiology of cancer among older adults and how GO evaluate and treat older adults with cancer. In LMICs, in-
models of care may or may not work in LMICs.21 The age tegration of geriatric principles into the care of older adults
cutoff for who is considered “older” may vary based on with cancer is quite limited.9 As a result, several societies
access to health care and life expectancy estimates for the have developed GO initiatives. ASCO facilitates education
region. Although age 65 to 70 is considered older in the about global oncology and GO through efforts like this very
United States, age 60 is defined as older by the United review and education symposium. In collaboration with the
Nations,17 and younger ages such as 50 to 55 may be used European Society for Medical Oncology, ASCO has sup-
for LMICs in sub-Saharan Africa.21 In addition to differences ported a GO track and a global curriculum in medical on-
in how older age is defined, there are differences in the cology, which includes attention to global oncology.27
ability for geriatric domains to be evaluated in clinical care. Other efforts to develop curriculum and training have
Literacy rates may affect the ability of an older patient to been spearheaded by SIOG.28,29 The training is also now
inclusive of nonphysician specialties, including nursing
complete the patient-reported outcomes parts of the geri-
and allied health, because practitioners in these disciplines
atric assessment in LMICs, and these countries may not
are often on the frontline of care in LMICs. There is still
have resources to have the clinicians or staff administer the
much to learn about how to train practitioners to best care
geriatric assessment. Research that disseminates novel
for older adults with cancer across the world.
ways to administer geriatric assessment tools with older
adults with cancer in LMICs could add value. For example, Organizational structures Very few journals in oncology are
one study evaluated a short geriatric assessment screening dedicated to understanding global oncology. JCO Global
tool for older adults with cancer in India who had an ele- Oncology is an ASCO journal dedicated to advancing cancer
mentary school literacy level or lower; patients found the care, research, and care delivery issues unique to countries
questions to be easy to answer in a reasonable amount of and settings with limited health care resources. The journal
time.22 Furthermore, tools that identify patients at high risk is open access and all articles are immediately and freely
for adverse outcomes may not be as predictive in countries available to the public upon online publication, which
with health care systems that have different care patterns truly facilitates dissemination to LMICs. Editor-in-Chief
and/or resources. For example, the Cancer and Aging Dr. Gilberto Lopes and the editorial board, which includes
Research Group Chemotherapy Toxicity Risk Score, which researchers and clinicians from all over the world, foster
was found to predict toxicity in older adults receiving a home for high-quality research that addresses the needs

of patients with cancer in settings where there are lower have been identified as a key issue (e.g., not offering trials to
resources. In addition to having a diverse editorial board, older adults), trial design and eligibility restrictions are also
other organizational practices that facilitate submission of key barriers. When older patients do participate, they are
research are in place, including limited formatting re- usually a disproportionately small representation of the
quirements. Other journals have dedicated attention to patients, they represent the fittest subset of this group, and
global oncology through special series or issues. For ex- their data are under-reported. The FDA, ASCO, and the
ample, Lancet Oncology published a global oncology series Cancer and Aging Research Group have worked closely
in January 2018 that discussed critical issues in LMICs, together to try to increase evidence of safety and efficacy of
such as implementation of a cancer control plan and de- new cancer therapeutics for older adults so that the data can
velopment of registries. In a viewpoint published in JAMA in inform clinical practice when older adults represent a high
June 2019, Gopal and Loehrer30 discussed the emergence proportion of patients with the disease.31,32 The FDA re-
of global oncology as an “academic discipline [that] seeks to cently put forth draft guidance for industry that recommends
promote scientific and clinical advances for cancer world- more attention to enrollment of older adults in trials that
wide.” This viewpoint highlighted representative examples hopefully will be finalized soon to guide future clinical tri-
of global oncology initiatives by the National Cancer In- als.33 Thus far, clinical trials still do not routinely report age-
stitute, cancer centers, professional societies, and industry. related issues, and specific toxicities are rarely reported by
Increasing research that demonstrates how to use data patient characteristics to include a breakdown by age.34
science and mobile health to improve access to care, foster
A recent systematic review identified the reporting of treatment
collaborations to conduct clinical trials in LMICs, and grow
efficacy and adverse events specific to older adults with cancer
workforce capacity will certainly lead to improvements
in phase III clinical trials.35 In 159 articles in 36 different
in care delivery and outcomes. However, there remains
journals published between July 1, 2016, and June 30, 2017,
a dearth of attention to how best to improve care delivery for
117 articles (73.6%) reported age-specific medians and 75
the growing population of older adults with cancer in LMICs.
(47.2%) included stratifications of data by age. Only 39.9%
The Journal of Geriatric Oncology is an international mul-
tidisciplinary journal dedicated to fostering improvements in reported effectiveness of treatment by age, and only 8.9%
care delivery and outcomes for older adults with cancer and reported adverse events stratified by age. The authors
their caregivers worldwide. The editorial board includes concluded that there continued to be inadequate reporting
expertise from LMICs and one of the associate editors, Dr. of treatment efficacy and adverse events as well as basic
Enrique Soto-Perez-de-Celis (a current ASCO board descriptive statistics about the age distribution of study
member), and encourages submissions of care delivery subjects.
research, education, and perspectives from researchers Experts in GO have advocated for full publication of data
and clinicians all over the world. As the official SIOG journal, from clinical trials, either in updated or secondary ana-
the Journal of Geriatric Oncology seeks out reviewers in- lyses. Large single-arm or randomized trials should report
terested in GO worldwide and provides mentoring to these the age distribution of patients in detail, not just the ranges
reviewers by pairing them up (if needed) with an editor or of age. Relationships between age and efficacy and toxicity
a reviewer with more expertise. In addition, the journal should be undertaken according to patient characteristics
provides support to authors who require assistance with (e.g., function and comorbidity) if the population with the
English language editing and offers a “Research Letter” specific disease includes a large proportion of older adults.
option for those manuscripts that demonstrate important Grade 2 toxicities should be reported for this population. If
perspectives from LMICs. there were limited enrollment of older adults despite a high
As demonstrated above, journals can facilitate dissemina- prevalence with the disease in the population, an explicit
tion of knowledge through implementation of organizational plan should be included that discusses next steps to
structures such as (1) a representative and diverse editorial gain data.
board; (2) a reviewer pool that includes reviewers from In a 2015 ASCO white paper on improving the evidence
LMICs; (3) mentorship of new reviewers from LMICs; (4) base for treating older adults with cancer, Hurria et al36
submission policies that facilitate submissions from LMICs, recommended that journals should incentivize research to
such as limited formatting requirements; (5) submission report age distribution and health of participants. Specific
options that allow for LMICs to be published more readily recommendations from this white paper are listed in Sidebar
(e.g., “Research Letters”); (6) partnerships with organiza- 1. These recommendations represent opportunities to
tions that facilitate research and clinical care in global identify differences in safety, efficacy, and dosing associ-
oncology and GO; and (7) offering open access if possible. ated with age (Fig. 2). Utilizing these journal policies could
Reporting of data Older patients continue to be under- improve researchers’ reporting of data relevant to the
represented in clinical trials. Although physician barriers treatment of older adults worldwide.

Kanesvaran et al
FIGURE 2. Recommendations for Improving Global Geriatric Oncology Research, Training, and Clinical Practice
Abbreviation: LMICs, low- and middle-income countries.
THE FDA ONCOLOGY CENTER OF EXCELLENCE UNVEILS Because most of what we know about cancer therapeutics is
PROJECT SILVER: A GLOBAL REGULATORY EFFORT TO based on clinical trials conducted with younger patients,
INCREASE THE EVIDENCE BASE FOR OLDER ADULTS oncology lacks robust knowledge of the risks and benefits of
WITH CANCER cancer treatments for older adults, especially those older
Cancer drug development is an increasingly global en- than age 75. The first guidance to industry on the topic was
deavor. Pivotal clinical trials are commonly conducted in- published in 1989, encouraging sponsors to study new
ternationally, providing critical data for evaluation of new drugs with the full range of patients who will receive them,
drugs. At the same time, older adults remain a growing including older adults.38
segment of our global oncology population.37 The conflu- As mentioned previously, the FDA OCE’s continued interest in
ence of the globalization of cancer clinical trials and this topic is reflected in a recently issued draft guidance on the
a growing number of older adults with cancer calls for inclusion of older adults in cancer clinical trials.33 The guidance
regulatory action. provides recommendations for including an adequate repre-
Enrollment of Older Adults in Oncology Trials: FDA and sentation of older adults in cancer clinical trials to better enable
European Medicines Agency Efforts evaluation of the benefit-risk profile of cancer drugs for this
The FDA has made a concerted effort to encourage en- population. The FDA makes recommendations to consider
rollment of older adults in registration clinical trials.31 evaluating drug efficacy and safety for older adults with cancer in

these findings on the interpretation of clinical trial results

require additional exploration.
SIDEBAR 1. RECOMMENDATIONS FOR IMPROVING THE
EVIDENCE BASE FOR TREATING OLDER ADULTS WITH
CANCER OCE’s Global Regulatory Initiatives
The following recommendations are adapted from The FDA has long recognized the value of collaboration
Hurria et al36: among international regulators. Since 2004, the FDA On-
Recommendation 1: Authors should report the de- cology group has held regular teleconferences with other
tailed age distribution of the population included in regulatory agencies to allow for exchange of information and
the study, not just the age ranges. Authors should collaboration on specific topics related to applications under
include data analyses that could potentially yield review. Currently, the Office of Oncologic Diseases holds
valuable age-related information, such as age- a monthly teleconference with Australia’s Therapeutic Goods
based analyses of response, benefit, and toxicity. Administration, Health Canada, the European Medicines
Recommendation 2: Include geriatric oncology experts Agency, Japan’s Pharmaceuticals and Medical Devices
in the pool of editorial board members who serve as Agency, and Switzerland’s Swissmedic.
peer reviewers of manuscripts. For global oncol-
ogy, include peer reviewers who also have ex-
pertise or perspectives relevant to review research Project Orbis
from the region. Project Orbis, conceived by Dr. Richard Pazdur and led by
Recommendation 3: Instruct peer reviewers to con- Dianne Spillman (OCE associate director for global regu-
sider whether the authors have adequately re- latory outreach), reached a major milestone in 2019 with the
ported the age distribution of the population first simultaneous approval with collaboration among in-
included in the study, the generalizability of the
ternational regulators.41 In September 2019, the FDA, the
results to the population with the disease, and data
Australian Therapeutic Goods Administration, and Health
analyses that could potentially yield valuable age-
related information. Canada granted simultaneous approval for the combina-
tion of pembrolizumab and lenvatinib for patients with
advanced-stage endometrial carcinoma that is not micro-
satellite instability-high or mismatch-repair deficient and
who had disease progression following prior systemic
early clinical development, in pivotal clinical trials, and in the therapy. This landmark approval was the first under Project
postmarket setting. Strategies for trial design, recruitment, col- Orbis, an FDA initiative that provides a framework for
lection of additional information, and reporting data in discrete concurrent submission and review of oncology drugs among
age groups (age 65–74 and 75 or older) are described. international partners. Future collaborations between in-
Similarly, the European Medicines Agency established its ternational regulators could provide patients with earlier
geriatric medicines strategy in 2011 to better address the access to products in other countries in which substantial
needs of older adults in the development and evaluation of delays in regulatory submissions might occur. The FDA took
medicines. The European Medicines Agency also has its second action under Project Orbis in November 2019,
a Geriatric Expert Group, which advises the agency and its again in collaboration with the Australian Therapeutic Goods
Committee for Medicinal Products for Human Use on issues Administration and Health Canada, with the approval of
related to older adults and has published several scientific acalabrutinib for patients with chronic lymphocytic leuke-
guidelines.39 mia or small lymphocytic lymphoma. Continued efforts
under Project Orbis will build on the initial success to in-
The FDA reviewed global enrollment of older adults in
corporate additional global partners, including Swissmedic
oncology trials over a 10-year period.40 More than 170,000
and the Singapore Health Science Authority.
patients with information regarding age and country were
included in the analysis. Forty-five percent of patients were
enrolled from Europe, 36% from North America (including Project Harmony
the United States and Canada), and 8.4% from Asia. Project Harmony was initiated in early 2020 to increase
Countries in Latin America, the Middle East/Africa, and the collaboration with international regulators on issues of on-
Baltic States/Russia enrolled the remaining 10.5% of pa- cology regulatory policy and scientific advice to industry.
tients. Europe enrolled the highest number of patients aged Under Project Harmony, international regulators will share
65 or older with cancer (51.1%) compared with North information on products under review and approved
America (32.4%) and Asia (6.8%). The reasons for differ- products. The project will also entail holding online-based
ential enrollment of older adults between Europe, North workshops two or three times a year on scientific and
America, and other parts of the world and the impact of regulatory topics.

Kanesvaran et al
Project Silver stakeholders on cancer in older adults. Project Silver

In early 2020, the OCE announced Project Silver, a con- provides an exciting new framework with which to build
certed global effort to improve the evidence base for older progress through collaboration. The project builds on
adults with cancer. Under Project Silver, global regulatory existing efforts across global regulatory agencies to
agencies will discuss key applications and development improve the evidence base for older adults with cancer,
programs with indications affecting older adults with cancer, while recruiting additional international partners. This
consider more detailed labeling information that reflects the global regulatory strategy is one of several OCE initia-
clinical experience with older adults, collectively gather tives, including Project Orbis, Project Harmony, and now
data in the postmarket setting on older adults with can- Project Silver, which benefit future drug development for
cer, and conduct educational workshops with global all patients with cancer.

1
National Cancer Centre Singapore, Singapore Ravindran Kanesvaran, MD, National Cancer Centre Singapore, 11
2
Wilmot Cancer Institute, University of Rochester, Rochester, NY Hospital Crescent, Singapore 169610; Twitter: @ravikanesvaran; email:
3
National Institute of Medical Science and Nutrition, Mexico City, Mexico [email protected].
4
Division of Oncology 2, Center for Drug Evaluation and Research, U.S.
Food and Drug Administration, Silver Spring, MD
EDBK_279513.
REFERENCES
1. United Nations Department of Economic and Social Affairs. World Population Ageing 2019: Highlights. www.un.org/en/development/desa/population/
publications/pdf/ageing/WorldPopulationAgeing2019-Report.pdf. Accessed March 1, 2020.
2. White MC, Holman DM, Goodman RA, et al. Cancer risk among older adults: time for cancer prevention to go silver. Gerontologist. 2019;59 (suppl 1):S1-S6.
3. Kennedy BJ. Aging and cancer. J Clin Oncol. 1988;6:1903-1911.
4. Extermann M, Aapro M, Audisio R, et al. Main priorities for the development of geriatric oncology: a worldwide expert perspective. J Geriatr Oncol. 2011;
2:270-273.
5. Wildiers H, Heeren P, Puts M, et al. International Society of Geriatric Oncology consensus on geriatric assessment in older patients with cancer. J Clin Oncol.
2014;32:2595-2603.
6. Kanesvaran R, Li H, Koo KN, et al. Analysis of prognostic factors of comprehensive geriatric assessment and development of a clinical scoring system in elderly
Asian patients with cancer. J Clin Oncol. 2011;29:3620-3627.
7. Hurria A, Togawa K, Mohile SG, et al. Predicting chemotherapy toxicity in older adults with cancer: a prospective multicenter study. J Clin Oncol. 2011;
29:3457-3465.
8. Magnuson A, Dale W, Mohile S. Models of care in geriatric oncology. Curr Geriatr Rep. 2014;3:182-189.
9. Soto-Perez-de-Celis E, de Glas NA, Hsu T, et al. Global geriatric oncology: achievements and challenges. J Geriatr Oncol. 2017;8:374-386.
10. Hurria A, Dale W, Mooney M, et al; Cancer and Aging Research Group. Designing therapeutic clinical trials for older and frail adults with cancer: U13 conference
recommendations. J Clin Oncol. 2014;32:2587-2594.
11. Paillaud E, Cudennec T, Caillet P, et al. [Clinical trials and elderly patients with cancer, the Geriatric Core Dataset (G-Code) tool]. Soins Gerontol. 2019;24:35-37.
12. Steer CB, Marx GM, Singhal N, et al. Cancer in older people: a tale of two disciplines. Intern Med J. 2009;39:771-775.
13. Lakhanpal R, Yoong J, Joshi S, et al. Geriatric assessment of older patients with cancer in Australia—a multicentre audit. J Geriatr Oncol. 2015;6:185-193.
14. Bray F, Ferlay J, Soerjomataram I, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185
countries. CA Cancer J Clin. 2018;68:394-424.
15. Kim JW, Kim SH, Lee YG, et al. Prospective validation of the Korean Cancer Study Group Geriatric Score (KG)-7, a novel geriatric screening tool, in older patients
with advanced cancer undergoing first-line palliative chemotherapy. Cancer Res Treat. 2019;51:1249-1256.
16. Kanesvaran R, Wang W, Yang Y, et al. Characteristics and treatment options of elderly Chinese patients with cancer as determined by comprehensive geriatric
assessment (CGA). J Geriatr Oncol. 2014;5:171-178.
17. Kansal S, Rao S. Demographic transition - cancer trends in geriatric population of North India. J Geriatr Oncol. 2019;10:362-364.
18. Ashing KT, Soto-Perez-de-Celis E. Disparities within a disparity: global health and health equity in geriatric oncology. J Geriatr Oncol. 2020;11:200-202.

19. Mohile SG, Dale W, Somerfield MR, et al. Practical assessment and management of vulnerabilities in older patients receiving chemotherapy: ASCO guideline for
geriatric oncology. J Clin Oncol. 2018;36:2326-2347.
20. Mohile SG, Epstein RM, Hurria A, et al. Communication with older patients with cancer using geriatric assessment: a cluster-randomized clinical trial from the
National Cancer Institute Community Oncology Research Program. JAMA Oncol. 2020;6:196-204.
21. Soto-Perez-de-Celis E. Global geriatric oncology: one size does not fit all. J Geriatr Oncol. 2019;10:199-201.
22. Banerjee J, Satapathy S, Upadhyay AD, et al. A short geriatric assessment tool for the older person with cancer in India—development and psychometric
validation. J Geriatr Oncol. 2019;10:222-228.
23. Hurria A, Mohile S, Gajra A, et al. Validation of a prediction tool for chemotherapy toxicity in older adults with cancer. J Clin Oncol. 2016;34:2366-2371.
24. Moth EB, Kiely BE, Stefanic N, et al. Predicting chemotherapy toxicity in older adults: comparing the predictive value of the CARG Toxicity Score with oncologists’
estimates of toxicity based on clinical judgement. J Geriatr Oncol. 2019;10:202-209.
25. Hurria A, High KP, Mody L, et al. Aging, the medical subspecialties, and career development: where we were, where we are going. J Am Geriatr Soc. 2017;
65:680-687.
26. Nightingale G, Burhenn PS, Puts M, et al. Integrating nurses and allied health professionals in the care of older adults with cancer: a report from the International
Society of Geriatric Oncology Nursing and Allied Health Interest Group. J Geriatr Oncol. 2020;11:187-190.
27. Dittrich C, Kosty M, Jezdic S, et al; ESMO/ASCO Global Curriculum Working Group. ESMO/ASCO recommendations for a global curriculum (GC) in medical
oncology-edition 2016. Ann Oncol. 2016;27:1378-1381.
28. Lund CM, O’Hanlon S. The SIOG Treviso course: students’ perspective. J Geriatr Oncol. 2017;8:389-390.
29. Colloca G, Monfardini S. A contribution to the future of geriatric oncology training: the SIOG Treviso advanced course. J Geriatr Oncol. 2017;8:387-388.
30. Gopal S, Loehrer PJ Sr. Global oncology. JAMA. 2019;322:397-398.
31. Singh H, Beaver JA, Kim G, et al. Enrollment of older adults on oncology trials: an FDA perspective. J Geriatr Oncol. 2017;8:149-150.
32. Levit LA, Singh H, Klepin HD, et al. Expanding the evidence base in geriatric oncology: action items from an FDA-ASCO workshop. J Natl Cancer Inst. 2018;
110:1163-1170.
33. U.S. Food and Drug Administration. Inclusion of Older Adults in Cancer Clinical Trials: Guidance for Industry. Draft Guidance. www.fda.gov/media/135804/
download. Accessed March 1, 2020.
34. Lichtman SM. Call for changes in clinical trial reporting of older patients with cancer. J Clin Oncol. 2012;30:893-894.
35. BrintzenhofeSzoc K, Krok-Schoen JL, Canin B, et al. The underreporting of phase III chemo-therapeutic clinical trial data of older patients with cancer:
a systematic review. J Geriatr Oncol. 2020;11:369-379.
36. Hurria A, Levit LA, Dale W, et al; American Society of Clinical Oncology. Improving the evidence base for treating older adults with cancer: American Society of
Clinical Oncology statement. J Clin Oncol. 2015;33:3826-3833.
37. Tang M, Joensuu H, Simes RJ, et al. Challenges of international oncology trial collaboration—a call to action. Br J Cancer. 2019;121:515-521.
38. U.S. Food and Drug Administration. Guideline for the Study of Drugs Likely to Be Used in the Elderly. www.fda.gov/regulatory-information/search-fda-guidance-
documents/study-drugs-likely-be-used-elderly. Accessed March 1, 2020.
39. European Medicines Agency. Medicines for Older People. www.ema.europa.eu/en/human-regulatory/research-development/medicines-older-people. Accessed
March 1, 2020.
40. Kanapuru B, Singh H, Fashoyin-Aje LA, et al. FDA analysis of patient enrollment by region in clinical trials for approved oncological indications. J Clin Oncol. 2017;
35:15s (suppl; abstr 2539).
41. Singh H, Blumenthal G, Pazdur R. Approvals in 2019: international review and a new agnostic molecular entity. Nat Rev Clin Oncol. 2020;17:130-131.

GLOBAL HEALTH

GLOBAL HEALTH
Resource-Stratified Guideline-Based Cancer Care

Should Be a Priority: Historical Context and
Examples of Success
Natasha Hunter, MD1; Naomi Dempsey, MD2; Fayez Tbaishat, MD3; Mohammad Jahanzeb, MD, FACP, FASCO4;
Sana Al-Sukhun, MD, MSc5; and Julie R. Gralow, MD, FACP, FASCO1
overview
Low- and middle-income countries (LMICs) are shouldering most of the burden of the rapidly increasing cancer
incidence and mortality worldwide, and this situation is projected to worsen in coming decades. Studies
estimate that more than one million deaths could be prevented annually if all patients received high-quality
care, but most LMICs lack the resources and infrastructure to adopt U.S. or European clinical oncology
practice guidelines. Several organizations have developed resource-stratified guidelines (RSGs) to provide
graduated and/or region-specific strategies for cancer diagnosis and treatment. The birth of these efforts
traces to 2002, when the World Health Organization (WHO) called for tailoring cancer treatments to the level
of available resources by country; the Breast Health Global Initiative (BHGI) formalized the first stratified
guidelines for breast cancer shortly thereafter. Since then, multiple organizations including ASCO and the
National Comprehensive Cancer Network (NCCN) have created guidelines customized for various cancer
subtypes and regions. These RSGs offer roadmaps for policy makers, clinicians, and health care adminis-
trators in LMICs to design projects in implementation science that can gradually and strategically raise the
quality of cancer care in their nation or region. Although the same resource limitations that complicate cancer
care in these areas also pose barriers to data gathering and research, some countries have met the challenge
and are improving cancer care using RSGs as a metric for success.
INTRODUCTION could be avoided with effective prevention and treat-

Cancer incidence and mortality are rapidly increasing ment, and that LMICs account for most avoidable
worldwide. In 2018, the International Agency for Re- cancer mortality.6 Recent studies project that if the
search on Cancer estimated more than 18 million new high standards of cancer care provided in high-resource
cancer diagnoses and 9.6 million cancer deaths world- settings were delivered globally across all economic
wide.1 Cancer is an important cause of morbidity and levels, approximately 1.5 million cancer deaths could be
mortality throughout the world, irrespective of a coun- avoided annually.5
try’s level of economic development, yet increasingly, The unanimous adoption of WHO 2017 Cancer Reso-
LMICs are shouldering much of the cancer burden. lution by WHO Member States from across all regions
Predictions suggest that by 2030, 13 million people will during the 70th World Health Assembly in Geneva has
die of cancer each year, and three-quarters of these focused global attention on the pressing necessity to
deaths will be in LMICs.2,3 The cancer burden is in- address cancer control worldwide, particularly in LMICs.7
creasingly concentrated in the regions and environ- The cancer resolution emphasized the growing impact
Author affiliations
and support ments where health care is most resource limited. of cancer as a leading cause of morbidity and mortality
information (if in LMICs.7
Scientific research has generated unprecedented
applicable) appear
knowledge about cancer, leading to new vaccines, Disparities in available resources for health care are
at the end of this
article. drugs, and technologies contributing to tremendous enormous. WHO estimates that the United States and
Accepted on March progress in cancer prevention, diagnosis, treatment, Canada have 10% of the global burden of disease,
2, 2020 and and palliation. The benefits of these advances, how- 37% of the world’s health workers, and more than 50%
published at ever, are not reaching most patients in LMICs.4 What is of the world’s financial resources for health; in contrast,
ascopubs.org on the African region has 24% of the global burden of
needed to achieve optimal clinical outcomes may be
March 31, 2020:
known, but how to reach these goals with limited health disease, 3% of health workers, and less than 1% of the
10.1200/EDBK_ resources often is not.5 Globally, it is estimated that world’s financial resources for health. This disparity is
279693 between 2.6 and 4.3 million cancer deaths annually even more extreme with cancer.8

Hunter et al
represented by LMICs, RSGs have been developed. RSGs

account for the costs and complexity of health care delivery
across a span of resource levels. Prioritizing and imple-
• RSGs provide evidence-based recommenda- menting interventions to maximize impact is complex,
tions for cancer care that can be tailored to the
however. Governments, institutions, and individuals around
level of resource availability and health systems
the world are working to implement evidence-based RSGs
infrastructure within a specific country or
region. as part of national and regional cancer control plans.4,12 The
safety, effectiveness, replicability, scalability, and short-
• Policy makers, clinicians, and health care ad-
comings of oncology interventions in resource-poor settings
ministrators can use RSGs based on evidence
can only be known through rigorous evaluation in the form of
and expert consensus to inform prioritization
and allocation of resources in the design and implementation science research.4
implementation of cancer control strategies. Here we will review the development of cancer-specific
• Effective implementation of RSGs requires RSGs over the last 2 decades, describe how implementa-
quality metrics, monitoring, and continual re- tion science can promote the adoption of these RSGs and
evaluation to guide improvements in the level of thereby improve standards of cancer care in LMICs, and cite
cancer care in resource-limited settings. examples of successful implementation.
WHO RESOURCE STRATIFICATION
Effective control of cancer requires integration of cancer WHO articulated the concept of tailoring cancer treat-
prevention, early detection, and comprehensive therapeutic ments to the level of available resources by country in
and palliative approaches. Human resources, poverty, 2002, as part of its executive summary of its National
human rights, education, health care infrastructure, com- Cancer Control Programs Policies and Managerial Guide-
peting priorities, and cultural and political issues all influ- lines. In this monograph, WHO describes a stratification of
ence local medical and public health practices. LMICs face national health care resources into low-, medium-, and
a multitude of constraints that pose daunting challenges in high-resource scenarios to facilitate establishment of Na-
the provision of optimal cancer care and control, including tional Cancer Control Plans that ensure the most efficient
lack of well-developed health care infrastructure, human use of resources.13
capacity issues, and affordability of technologies. Policy
In 2013, WHO issued guidelines for screening and treatment
makers and health care providers in LMICs and other
of precancerous lesions for cervical cancer prevention. The
resource-constrained settings face painful decisions re-
guideline included a screen-and-treat approach (visual
garding budget priorities, and the difficulty of offering optimal
detection of abnormalities using acetic acid and treatment
services with available resources is escalating. International
with cryotherapy) tailored to lower-resource settings. HPV
health organizations increasingly acknowledge that evidence-
testing was recommended when feasible. The guideline
based tools are desperately needed to delineate essential
acknowledges that cytology-based programs (including use
packages of cost-effective measures for countries to consider
of Pap smears) are difficult to implement with high screening
and adapt if they are to make successful cancer control
coverage in LMICs; thus, new screening and treatment
investments.9
strategies had to be considered to increase coverage and
Evidence-based cancer clinical practice guidelines (CPGs) decrease cervical cancer incidence and mortality.14 In 2014,
provide recommendations for the multidisciplinary man- WHO released an updated guideline for treatment of cer-
agement of most cancer types, with a goal of improving the vical intraepithelial neoplasia focusing on three treatments
quality of cancer care and optimizing outcomes.10 The most for cervical intraepithelial neoplasia available in LMICs
prominent of these guidelines include those by ASCO, (cryotherapy, large loop excision, and cold knife conization),
NCCN, and the European Society for Medical Oncology. as well as a guide on comprehensive cervical cancer
Most of the published practice guidelines developed by control.15,16
professional societies assume the availability of costly
In 2014, WHO also issued a position paper on mammog-
preventive, diagnostic, and treatment resources applied
raphy screening that distinguishes between two strategies
within a mature and organized health care infrastructure.11
for the early detection of breast cancer: early diagnosis and
They make no recommendations about how resource ex-
screening. Early diagnosis relies on heightened public and
penditures should be prioritized to achieve the greatest
professional awareness of cancer signs and symptoms,
clinical benefit and outcome. This limits the applicability of
improved health care–seeking behavior, prompt clinical
many existing practice guidelines in LMICs.
assessment, and referral of suspected cases for diagnostic
In an attempt to translate CPGs in a rational and clinically evaluation. Screening involves the systematic application of
useful way to the diverse spectrum of health care settings a screening test in an asymptomatic population to detect

Resource Stratified Guidelines for Cancer Risk Reduction, Diagnosis, and Treatment
cancer or precancers before they threaten the well-being of Although BHGI’s work focuses on women with breast
the individual or the community. The WHO paper asserts cancer, the model is applicable to other cancers. This model
that screening is a complex public health undertaking and is for RSG development has subsequently been adopted in
usually cost-effective and justified in settings with high some form by several other oncology programs and soci-
disease burden and adequate health system capacity. In eties, including ASCO, NCCN, WHO, and the Asian On-
limited resource settings, where most women with breast cology Summit.26
cancer are diagnosed at late stages, and mammography NCCN RESOURCE STRATIFICATION
screening is neither cost-effective nor feasible, WHO advises
focusing on early diagnosis (i.e., ensuring that women with NCCN Guidelines are a comprehensive set of evidence-
symptomatic lesions get prompt and effective diagnosis and based, consensus-driven guidelines for delivering multi-
treatment).17 disciplinary cancer care, from risk assessment through
prevention, screening, diagnosis, treatment, survivorship,
BHGI RESOURCE STRATIFICATION and end-of-life care. NCCN guidelines were intentionally
The BHGI pioneered a resource-stratified approach to developed for use at the resource level available in the
guidelines for the early detection, diagnosis, and treatment United States.
of breast cancer in LMICs.11,18 BHGI has focused on de- To address the limited utility of standard NCCN Guidelines
velopment and implementation of evidence-based, eco- in lower-resource settings, NCCN initiated a program to
nomically feasible, and culturally appropriate guidelines to develop a framework for resource-stratifying NCCN Guide-
improve breast health outcomes across a spectrum of re- lines (NCCN Framework).27 The Framework is modified
source settings. BHGI builds its recommendations on from a method developed by the BHGI and identifies four
a four-tiered, resource-stratified system in which cancer resource environments: basic resources, core resources,
management strategies can be prioritized within the context enhanced resources, and standard NCCN Guidelines. It
of available health care resources: basic, limited, enhanced, presents a rational approach for building cancer manage-
and maximum levels (Table 1).19 ment systems by applying available and affordable ser-
Between 2002 and 2018, and through a series of six col- vices in a logical sequence, each level building on the
laborative global summits, BHGI developed a resource- previous one to improve cancer care incrementally. In se-
stratified framework to guide all aspects of breast cancer lected circumstances, treatment options are added for
management. BHGI guidelines use a consensus approach consideration in lower resource settings that are not typi-
with analysis of evidence-based breast cancer research. cally used in the United States. Currently a specific NCCN
Panels of breast cancer experts representing diverse Framework has been created for 17 cancer types, as well
countries and world regions created BHGI guidelines to as for palliative care and breast cancer screening and
address early detection, diagnosis, and treatment of breast diagnosis.28
cancer in countries with limited health care resources. Because of interest among international oncologists in
Summits have also focused on optimizing health care de- adapting NCCN Guidelines to improve the standard of
livery, supportive care, and guideline implementation.18,20-25 cancer care in their own countries, NCCN has also launched
initiatives to enhance use of the guidelines internationally,
TABLE 1. Breast Health Global Initiative Resource Stratification including translation of NCCN content into multiple lan-
Resource guages with free access around the world and collaboration
Level Availability of Services with organizations across the globe to create adaptations
Basic Fundamental or core services that are necessary for of NCCN Guidelines for specific countries and regions,
any cancer system to function (e.g., mastectomy) which take into account metabolic differences in pop-
Limited Second-tier services that intend to produce major ulations, technology access, and the regulatory setting in
improvements in outcomes and are achievable the particular country.29 Each set of region-specific NCCN
with little financial means and modest
Guidelines offer adaptations in green text that apply to that
infrastructure (e.g., tamoxifen as adjuvant therapy)
particular region.
Enhanced Third-tier services that are optional in a resource-
constrained setting but are important and should NCCN has also recently partnered with regional repre-
produce further improvements in outcome (e.g., sentatives to create harmonized guidelines (i.e., region
aromatase inhibitors) specific but also acknowledging varying levels of resource
Maximum Services that might be used in settings where there availability) as part of a collaborative effort to combat
are no resource constraints. To be useful, skyrocketing cancer rates and unique care circumstances
resources at the maximum level always depend on
in specific regions. NCCN harmonized guidelines offer two
the existence and functionality of all lower-level
resources tiers of treatment recommendations, representing both the
optimal care that these countries aspire to provide and

Hunter et al
pragmatic approaches to treatment options for resource- understanding that had been uncovered during the regional
constrained settings. NCCN harmonized guidelines use customization of NCCN Guidelines.31
color coding to represent generally available standard of
The lung cancer committee provides an illustrative example.
care (black text); highly advanced/optimal care that may
Composed of a multidisciplinary team of medical, radiation,
be costly, technologically challenging, and/or have a lesser
and surgical oncologists, as well as experts in pulmonology,
impact on cancer outcome (gray text); regional options that
pathology, and imaging, the group identified five categories
may be considered when standard of care cannot be
of cancer care that would require modification of NCCN
achieved (blue text); and care options that are not feasible
Guidelines for the MENA region: diagnosis, staging, curative
or available at this time (gray strikethrough text). NCCN
treatment, palliative treatment, and supportive care. The
harmonized guidelines currently exist for sub-Saharan
diagnosis of lung cancer requires bronchoscopy or CT-
Africa and the Caribbean.30
guided lung biopsy, available only in tertiary centers in
major cities. Pathology laboratories can perform immuno-
NCCN Regional Adaptation: Middle East and North histochemistry stains, but more advanced testing such as
Africa Example next-generation sequencing may not be available. For
One of the first efforts to adapt NCCN Guidelines to lower staging, PET is recommended, but some countries have no
resource settings occurred in 2009 in the Middle East PET scanners, and many more have only one for the entire
and North Africa (MENA) region. Extending from Mo- nation. Mediastinoscopy is used infrequently because of
rocco to India and from Turkey to Yemen, the MENA a lack of local expertise. Curative therapy requires a multi-
region is geographically vast and culturally varied. The disciplinary approach, but the MENA region suffers from
differences in local and regional resources, health care a shortage of well-trained thoracic surgeons, and tumor
delivery, customs, and cultural expectations high- boards are not widely used. Adjuvant chemotherapy was not
light the challenges of adapting NCCN Guidelines standard at the time, and radiation oncology expertise and
internationally. technology were scarce. Regarding palliative and sup-
portive care, centers often allocated resources to curable
Guideline adaptation in the MENA region began with the
patients, refusing patients with metastatic disease. Access
formation of committees focused on seven areas: breast
to biologic agents for patients with metastatic disease was
cancer, lung cancer, colon cancer, prostate cancer, hepa-
limited, and insufficient access to pain medications com-
tocellular carcinoma, lymphoma, and palliative care. Each
plicated symptom management and end-of-life care.32
committee was composed of a chairperson and subject
matter experts from the MENA region, as well as a U.S. Many of the modifications the lung cancer committee made
expert from the corresponding NCCN panel to serve as to the 2009 NCCN Guidelines have been superceded by
a disease-specific advisor in guideline development and further research, but some representative recommenda-
consensus building. Committees met at an initial launch tions follow. In the initial laboratory evaluation, the group
meeting; subsequently, much of the work occurred elec- advised specifying the tests required (calcium, lactate de-
tronically. A regional chair coordinated the efforts of all hydrogenase, renal function, liver function tests, and al-
committees through a central office in Riyadh, Saudi bumin) rather than ordering a costly comprehensive
Arabia.31 metabolic profile. In the pretreatment evaluation of stage I
and II disease, endobronchial ultrasound was added as an
Using the 2009 version of the NCCN Guidelines as a tem-
alternative to mediastinoscopy or other modalities of lymph
plate, committee members discussed potential modifica-
node evaluation that might be unavailable. For positive
tions and drafted recommendations for evidence-based
resection margins, the committee recommended chemo-
revisions derived from experience and publications relevant
radiotherapy consolidation if reresection proved impossible
to their region. Justifications and references were man-
because of resource constraints or lack of local expertise. If
dated for each change, and the committee met with the
combined chemoradiation was not available, sequential
NCCN expert to discuss any ambiguities or controversies. A
therapy was advised. For superior sulcus tumors, the
regional symposium was then held among committee
committee added surgery alone as an initial option for re-
members, U.S. experts, and NCCN representatives to fi-
sectable disease in centers without access to combined
nalize the recommendations. Local oncologists were invited
chemoradiation induction.31
to each disease-specific symposium to learn the new
standards of care for their region. Committees then pub- The committee also identified a number of areas for future
lished a final manuscript of agreed-on modifications, with research in the MENA region to further evaluate how the
scientific translations provided in each regional language. guideline modifications may affect outcomes and how pa-
Committee members also created lists of potential re- tient and cancer biology in the MENA region may differ from
search projects from the observed gaps in knowledge or data gathered on patients with other backgrounds. The

committee recommended studies to determine how best regions in the world with the highest cancer prevalence.
to use limited PET resources in the MENA region, clinical Cancer is the second leading cause of death in the Ca-
trials to validate phase III data from Western countries in ribbean, and the rate of breast cancer deaths is currently
the MENA population, and molecular studies on muta- twice that of the United States. NCCN Guidelines Panel
tions such as KRAS and EGFR to evaluate their preva- Members participated in a working session with local on-
lence and utility as a target in patients from the MENA cologists to adapt existing NCCN Guidelines and NCCN
region.31 Framework to better reflect the diverse needs and resources
throughout this region. Currently seven cancer types are
NCCN Harmonized Guidelines: Sub-Saharan Africa and
represented in this harmonized guideline, with more under
the Caribbean
development.30
The NCCN Guidelines and NCCN Framework were re-
cently adapted to create the NCCN harmonized guidelines ASCO INTERNATIONAL RESOURCE STRATIFICATION
for sub-Saharan Africa through a joint project of the African
On World Cancer Day 2013, ASCO, recognizing its role as
Cancer Coalition, American Cancer Society, the Clinton
a leading oncology society with one-third of its members
Health Access Initiative, IBM, and NCCN. The goal of this
hailing from outside the United States, launched ASCO
collaboration is to train an expanded group of cancer care
International. This initiative includes an expanding portfolio
providers and to improve the availability of effective thera-
of international programs for professional development,
pies for people with cancer in Africa.30
education, training, and volunteering in LMICs, with the
There are an estimated 811,000 new cases of cancer and aim of connecting the global community of cancer care
527,000 deaths from cancer each year in sub-Saharan providers.36
Africa; this incidence is expected to double by 2040
because of population growth and aging.33,34 Cancer care ASCO has longstanding experience with developing CPGs
in Africa is characterized by late presentation, limited beginning in 1995.37 These guidelines provide detailed,
access to treatment, and poor outcomes relative to other evidence-based recommendations in the areas of greatest
geographic regions. Although access to the latest bio- clinical need and uncertainty to ensure that patients get the
medical advancements may be idealistic, the disparity best-quality medical care. However, they assume maximal
in access to basic cancer care is unacceptable and availability of resources, infrastructure, and technology.
reversible. Like the other standard-setting institutions mentioned
The African Cancer Coalition comprises oncologists from above, ASCO has recognized that lack of resources in-
sub-Saharan Africa working together to improve access to cluding infrastructure problems, personnel shortages, and
high-quality cancer treatment. The Coalition has more than training deficits can make it difficult, if not impossible, for
100 oncologists and other multidisciplinary cancer care some institutions and clinicians outside the United States
physicians from more than 30 hospitals, universities, and to implement ASCO guidelines. With this in mind, the
Ministries of Health representing 13 countries that are home organization has seized an opportunity to marry the ex-
to 55% of the cancer cases in sub-Saharan Africa.35 pertise of its international membership with its own ex-
tensive background in CPG development to create RSGs
The first NCCN harmonized guidelines for sub-Saharan that provide guidance across a spectrum of available
Africa debuted in November 2017, during the biennial resources.
African Organization for Research and Training in Cancer
conference in Kigali, Rwanda. In addition to covering var- Adapting BHGI’s model, ASCO began to undertake RSG
ious cancer types, they also provided treatment recom- development in 2013 by initiating a pilot effort to develop
mendations for pain management, survivorship, smoking RSGs for prevention and treatment of cervical cancer,
cessation, and other aspects of supportive care.30 a globally high-incidence cancer. ASCO’s set of three RSGs
for cervical cancer, the first of which was issued in 2016,
Currently 45 cancer types are represented in the harmo- encompass the full spectrum of disease, from primary and
nized guidelines for sub-Saharan Africa. The medical rec- secondary prevention/screening to work-up and treat-
ommendations have been officially endorsed in a number of ment.38-40 After the release of the cervical cancer RSGs,
countries, including Ethiopia, Malawi, Nigeria, Tanzania, ASCO’s RSG Advisory Group recommended that ASCO
Uganda, and Zambia.30 expand into a planned series of RSGs in other areas, in-
In the Caribbean, NCCN harmonized guidelines have been cluding palliative care and colorectal cancer, with more
created as part of a joint project with the Caribbean As- topics to be developed over time. A palliative care RSG
sociation for Oncology & Hematology and NCCN. The Ca- was published in 2018,41 and three colon cancer guide-
ribbean archipelago, extending from the Bahamas in the lines on early detection and treatment were published in
north to Trinidad and Tobago in the south, ranks among the 2019-2020.42-44

Hunter et al
The primary building block of ASCO’s approach involves work with local oncology societies on disseminating the
ASCO Expert Panels or disease-specific multidisciplinary guidelines as they are released.
committees. RSG panelists hail from various resource set-
RESOURCE-STRATIFIED GUIDELINES: CHALLENGES OF
tings and represent multiple specialties, as well as diverse
ADOPTION, IMPLEMENTATION, AND MONITORING
geographic, gender, and ethnic backgrounds. Panelists
need not be ASCO members; they include patients, patient Given the growing focus on the burgeoning problem of
advocates, public health experts, and health economists cancer in the developing world, the coming decade will see
with expertise in the represented settings.37 further development of guidelines for cancer management
tailored to a variety of resource settings. The next and ar-
The roadmap that RSG Expert Panels follow involves sys-
guably more important question, however, is what impact
tematic reviews and formal consensus building using
these guidelines are actually having on the health care
a modified Delphi strategy (i.e., a method of collecting and
systems and providers they are meant to benefit.
circulating judgments from a group of experts in sequential
iterations to encourage participants to reassess and develop The uptake of CPGs is inconsistent, despite their potential to
their opinions). Because of the paucity of evidence from improve the quality of health care and patient outcomes.48 A
some of these settings, ASCO also used other validated recent literature review to assess BHGI guideline dissemi-
methods for some of their RSGs, including elements of nation, adoption, and implementation demonstrated that
ASCO’s endorsement processes, as well as a modified BHGI guidelines are frequently referenced in the global
ADAPTE method (ADAPTE was an international effort be- health literature, most commonly to support statements
tween 2006 and 2010 to develop and validate a process for related to disease management. However, there was low
adapting guidelines while fostering the users’ sense of documentation of specific implementation of the guidelines
ownership of the guideline).45,46 in comprehensive cancer control plans, indicating that
implementation remains a challenge for most LMICs.49
ASCO’s RSGs are freely available through the ASCO website
and guidelines mobile app. ASCO publishes each RSG in What accounts for the low rate of RSG adoption (or at least
the open access Journal of Clinical Oncology (JCO) Global documentation of adoption) in LMICs? Part of the answer
Oncology and a summary of the guideline in JCO Oncology may lie in the fact that most guidelines themselves remain
Practice. ASCO University has produced a new, case-based mute on the issue of implementation. Surveys and in-
course series designed to provide further essential guideline terviews with clinicians have revealed that they are aware of
information to both U.S.- and international-based ASCO and agree with RSGs but desire guidance and support to
members and has also translated the summaries of its first help implementing them.5 RSG developers and users have
three RSGs into Spanish. also advocated for guidance on providing tools and strat-
egies for implementation of RSGs alongside the guidelines
The guidelines have also been disseminated through ASCO themselves.5
International programs in LMICs. ASCO’s Cancer Control for
Primary Care shares the ASCO RSGs as part of its course to A recent Cochrane Database review of the effectiveness of
train primary care providers on cancer prevention, screening, implementation tools developed by guideline producers to
and referral. In a collaboration with Health Volunteers improve uptake of CPGs concluded that a tool developed by
Overseas, ASCO assists hospitals in LMICs to improve the the guideline author may lead to increased adherence and
quality of care provided to patients with cancer and has that providing health care professionals with tools to improve
incorporated RSGs into that assistance. The ASCO/Conquer implementation of a guideline may lead to little difference in
Cancer International Development and Education Award health care cost.48
supports early career oncologists in LMICs (today there are . Put together, this suggests that involving stakeholders in
300 recipients of this award in . 60 countries)47; RSGs have RSG development, which has been a standard approach, as
been disseminated to the International Development and described earlier, and addressing the issue of imple-
Education Award network for promotion and adoption in their mentation concurrently with RSG development may lead to
practice settings. broader and more consistent adoption of RSGs among
Other groups have endorsed or otherwise recognized ASCO individual providers and policy makers in development of
RSGs, including the Society for Gynecologic Oncology, In- cancer control plans.48
ternational Gynecologic Cancer Society, American College of Improving national or institutional adoption of new guide-
Obstetrics and Gynecology, American Society for Colposcopy lines falls into a field called implementation science, which
and Cervical Pathology, and Gynecologic Cancer Intergroup. requires baseline and ongoing data gathering and analysis
Clinicians in several countries/regions are implementing and/ for success.50,51 A program guided by implementation
or adapting the guidelines, including Uganda, Turkey, Leb- science includes multiple phases and components, which
anon, Central America, and Myanmar. ASCO will continue to were summarized in a recently published checklist by Ilbawi

The Community-Based Program for Breast Health in Peru,

a collaboration between the Peruvian Ministry of Health, the
SIDEBAR 1. CHECKLIST FOR IMPLEMENTATION OF
National Cancer Institute in Lima, and the Regional Cancer
CANCER CONTROL PROGRAMS
Institute in Trujillo, is using a four-step, resource-stratified
1. Assess burden amenable to planned program phased implementation model developed by Breast Cancer
2. Identify prevalence of risk factors and disease
Initiative 2.5, a breast cancer analytics and advocacy group
burden
based at the Fred Hutchinson Cancer Research Center/
3. Define target population for program
4. Identify gaps and barriers to implementation University of Washington. Through this stepwise approach,
5. Estimate implementation costs of programs policy makers in this middle-income Latin American country
6. Determine existing capacity for program aim to improve breast health care capacity and delivery.53
development To date, the program has successfully implemented the first
7. Generate investments from stakeholders in- three steps of a proposed four-step model, with reports of
cluding advocates increased awareness of breast cancer among women,
8. Implement program or programs improved capacity for early diagnosis among health care
9. Survey programs including quality assurance and workers, and creation of stronger and more functional links
patient safety between primary and higher levels of health care.53
10. Reassess and modify programs based on mea-
sured outcomes Similar strides are being made in Paraguay, where the
government launched health care reform in 2008, along
with a cancer control program. Cervical cancer is the
number one cause of cancer death among Paraguayan
et al for designing and implementing cancer control women, with a mortality at 15.7%, among the highest rates
programs5 (Sidebar 1). in Latin America. Between 2012 and 2017, National Cancer
Institute Paraguay partnered with Health Volunteers Over-
This kind of cyclic research involving identifying and
seas, an international organization of volunteer consultants
characterizing a problem, devising a plan based on a metric
and educators, to improve care processes and, hopefully,
(e.g., RSGs), executing the plan, and then reassessing is
outcomes for patients with cervical cancer. Health Volun-
critical in evaluating and improving the expansion of quality
teers Overseas worked with Paraguayan oncologists to
cancer care in resource-limited settings.
describe the baseline health care setting for patients with
The paucity of evidence from the settings where the cervical cancer and identify barriers to timely and effective
guidelines are intended to be applied poses a serious treatment. These barriers included delays in diagnosis, lack
challenge to the successful implementation of RSGs. Un- of coordination between specialists, and a paucity of ap-
fortunately, the same limited resources that lead to the need propriate radiation equipment. The group used ASCO’s RSG
for RSGs also contribute to limited funds for conducting to classify the practice as a basic setting. Over the course of
research, which in turn perpetuates the lack of data around the next 5 years, and aided by Health Volunteers Overseas’
guideline-driven efforts in LMICs. recommendations, motivated cancer center staff worked to
Systematic reviews of the current state of cancer care in make dramatic improvements in care, including better
various regions can help guideline developers identify gaps coordination and communication between specialists,
in research, creating opportunities for future study. Regional faster turnaround times for pathology, and increased re-
conferences such as the ones hosted by NCCN in which source allocation leading to increased staffing and improved
senior clinicians are invited to modify, adapt, and customize radiation technology. Overall, the institution estimated that it
the parent guideline can help identify research objectives.52 progressed from a basic to an enhanced setting per ASCO’s
Overall, however, research efforts in RSG adoption must RSGs.54
originate at policy-making and institutional levels, as the DISCUSSION
implementation research itself is integral to any cancer
control program. The dramatic increase in the cancer toll in LMICs mandates
effective interventions at all levels of cancer care. Consid-
Resource-Stratified Guidelines in Action: Examples ering the rising costs of cancer care globally, optimal use of
of Success resources to optimize outcome becomes essential.
Although data around RSG implementation globally are Over the last 2 decades, multiple organizations have rec-
sparse, some health care systems and institutions have ognized the growing problem of cancer worldwide and the
published their experiences with initiating system-wide need for cancer management strategies that take into ac-
change in cancer care delivery using guidelines as a map count vast gaps in resources across nations and regions.
and a metric for success. When resources are limited, RSGs give individual clinicians

Hunter et al
a practical framework for managing patients based on the optimal treatment and dosing, demand for region-specific
level of health care available in that practice area. They also clinical practice guidelines will only increase.
give policy makers tools for planning and prioritizing Prioritizing and implementing interventions to maximize
resource-appropriate cancer control, establishing consis- the impact of RSGs is complex and expensive and may
tent minimum standards for cancer care while realistically benefit from partnerships across and beyond the health
accounting for differences in resource levels and health sector.55 The planning, implementation, and evaluation
care systems around the world. Finally, they provide phases of RSG-based cancer control policies and pro-
a metric for implementation science efforts through which grams should involve policy makers, health care providers,
research questions can be posed and progress may be scientists, and community-based organizations repre-
measured. senting patients and their families. It is only through these
We can expect that such efforts will gain importance as the collaborations, and the commitment of resources to these
rapid expansion of personalized medicine generates ever efforts, that LMICs will gain a foothold in quality cancer
more data tailored to diverse genetic backgrounds. With control and begin to erase the cancer outcome disparities
potential information identifying regional differences for that we currently witness.

1
Division of Oncology, University of Washington, Seattle, WA Julie R. Gralow, MD, 825 Eastlake Ave. East G3-630, Seattle, WA 98109;
2
University of Miami, Miami, FL Twitter: @jrgralow; email: [email protected].
3
Department of Oncology, Al Bashir Hospital, Amman, Jordan
4
21st Century Oncology, Boca Raton, FL
5
Al Hyatt Oncology Practice, Amman, Jordan
EDBK_279693.
REFERENCES
1. Ferlay J, Colombet M, Soerjomataram I, et al. Estimating the global cancer incidence and mortality in 2018: GLOBOCAN sources and methods. Int J Cancer.
2019;144:1941-1953.
2. Bray F, Jemal A, Grey N, et al. Global cancer transitions according to the Human Development Index (2008-2030): a population-based study. Lancet Oncol.
2012;13:790-801.
3. Goss PE, Lee BL, Badovinac-Crnjevic T, et al. Planning cancer control in Latin America and the Caribbean. Lancet Oncol. 2013;14:391-436.
4. Tapela NM, Mpunga T, Karema N, et al. Implementation science for global oncology: the imperative to evaluate the safety and efficacy of cancer care delivery.
J Clin Oncol. 2016;34:43-52.
5. Ilbawi AM, Anderson BO. Global cancer consortiums: moving from consensus to practice. Ann Surg Oncol. 2015;22:719-727.
6. Knaul FM, Arreola-Ornelas H, Rodriguez NM, et al. Avoidable mortality: the core of the global cancer divide. J Glob Oncol. 2018;4:1-12.
7. World Health Organization. Cancer prevention and control in the context of an integrated approach. 70th World Health Assembly. http://apps.who.int/gb/ebwha/
pdf_files/WHA70/A70_ACONF9-en.pdf?ua=1. Accessed March 18, 2020.
8. World Health Organization. The World Health Report 2006: Working Together for Health. Geneva: World Health Organization; 2006.
9. Gelband H, Sankaranarayanan R, Gauvreau CL, et al. Costs, affordability, and feasibility of an essential package of cancer control interventions in low-income and
middle-income countries: key messages from Disease Control Priorities, 3rd edition. Lancet. 2016;387:2133-2144.
10. Cook DJ, Greengold NL, Ellrodt AG, et al. The relation between systematic reviews and practice guidelines. Ann Intern Med. 1997;127:210-216.
11. Anderson BO, Carlson RW. Guidelines for improving breast health care in limited resource countries: the Breast Health Global Initiative. J Natl Compr Canc Netw.
2007;5:349-356.
12. Harford JB, Otero IV, Anderson BO, et al. Problem solving for breast health care delivery in low and middle resource countries (LMCs): consensus statement from
the Breast Health Global Initiative. Breast. 2011;20 (suppl 2):S20-S29.
13. World Health Organization. National Cancer Control Programmes: Policies and Managerial Guidelines. Geneva: World Health Organization; 2002.
14. World Health Organization. Guidelines for Screening and Treatment of Precancerous Lesions for Cervical Cancer Prevention. Geneva: World Health Organization;
2013.
15. World Health Organization. Treatment of Cervical Intraepithelial Neoplasia 2–3 and Adenocarcinoma In Situ: Cryotherapy, Large Loop Excision of the
Transformation Zone, and Cold Knife Conization: WHO Guidelines. Geneva: World Health Organization; 2014.

16. World Health Organization. Comprehensive Cervical Cancer Control: A Guide to Essential Practice, 2nd ed. Geneva: World Health Organization; 2014.
17. World Health Organization. WHO Position Paper on Mammography Screening. Geneva: World Health Organization; 2014.
18. Anderson BO, Yip CH, Smith RA, et al. Guideline implementation for breast healthcare in low-income and middle-income countries: overview of the Breast Health
Global Initiative Global Summit 2007. Cancer. 2008;113 (8, suppl):2221-2243.
19. Anderson BO, Shyyan R, Eniu A, et al. Breast cancer in limited-resource countries: an overview of the Breast Health Global Initiative 2005 guidelines. Breast J.
2006;12 (s1, suppl 1):S3-S15.
20. Eniu A, Carlson RW, El Saghir NS, et al; Breast Health Global Initiative Treatment Panel. Guideline implementation for breast healthcare in low- and middle-
income countries: treatment resource allocation. Cancer. 2008;113 (8, suppl):2269-2281.
21. Harford J, Azavedo E, Fischietto M; Breast Health Global Initiative Healthcare Systems Panel. Guideline implementation for breast healthcare in low- and middle-
income countries: breast healthcare program resource allocation. Cancer. 2008;113 (8, suppl):2282-2296.
22. Anderson BO, Ilbawi AM, El Saghir NS. Breast cancer in low and middle income countries (LMICs): a shifting tide in global health. Breast J. 2015;21:111-118.
23. El Saghir NS, Adebamowo CA, Anderson BO, et al. Breast cancer management in low resource countries (LRCs): consensus statement from the Breast Health
Global Initiative. Breast. 2011;20 (suppl 2):S3-S11.
24. El Saghir NS, Eniu A, Carlson RW, et al; Breast Health Global Initiative Systemic Therapy Focus Group. Locally advanced breast cancer: treatment guideline
implementation with particular attention to low- and middle-income countries. Cancer. 2008;113 (8, suppl):2315-2324.
25. Distelhorst SR, Cleary JF, Ganz PA, et al; Breast Health Global Initiative Global Summit on Supportive Care and Quality of Life Consensus Panel Members.
Optimisation of the continuum of supportive and palliative care for patients with breast cancer in low-income and middle-income countries: executive summary of
the Breast Health Global Initiative, 2014. Lancet Oncol. 2015;16:e137-e147.
26. Al-Sukhun S, de Lima Lopes G Jr., Gospodarowicz M, et al. Global health initiatives of the International Oncology Community. Am Soc Clin Oncol Educ Book.
2017;37:395-402.
27. Carlson RW, Scavone JL, Koh WJ, et al. NCCN framework for resource stratification: a framework for providing and improving global quality oncology care. J Natl
Compr Canc Netw. 2016;14:961-969.
28. National Comprehensive Cancer Network. NCCN Framework for Resource Stratification of NCCN Guidelines (NCCN Framework). https://www.nccn.org/
framework/default.aspx. Accessed January 30, 2020.
29. Carlson RW, Larsen JK, McClure J, et al. International adaptations of NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw. 2014;12:643-648.
30. National Comprehensive Cancer Network. NCCN Harmonized Guidelines. https://www.nccn.org/harmonized/. Accessed January 30, 2020.
31. Jazieh AR, Azim HA, McClure J, et al. The process of NCCN guidelines adaptation to the Middle East and North Africa region. J Natl Compr Canc Netw. 2010;8
(Suppl 3):S5-S7.
32. Jazieh AR, Bamefleh H, Demirkazik A, et al; MENA Lung Cancer Regional Guidelines Committee. Modification and implementation of NCCN guidelines on non-
small cell lung cancer in the Middle East and North Africa region. J Natl Compr Canc Netw. 2010;8 (Suppl 3):S16-S21.
33. International Agency for Research on Cancer. Cancer today. https://gco.iarc.fr/today/home. Accessed February 5, 2020.
34. International Agency for Research on Cancer. Cancer tomorrow. https://gco.iarc.fr/tomorrow/home. Accessed February 5, 2020.
35. National Comprehensive Cancer Network. NCCN Global: Africa. https://www.nccn.org/global/africa.aspx. Accessed January 30, 2020.
36. American Society of Clinical Oncology. International programs. www.asco.org/international-programs. Accessed February 5, 2020.
37. Al-Sukhun S, Temin S, Chavez-MacGregor M, et al. ASCO resource-stratified guidelines: methods and opportunities. J Glob Oncol. 2018;4:1-8.
38. Jeronimo J, Castle PE, Temin S, et al. Secondary prevention of cervical cancer: ASCO Resource-Stratified Clinical Practice Guideline. J Glob Oncol. 2016;
3:635-657.
39. Arrossi S, Temin S, Garland S, et al. Primary prevention of cervical cancer: American Society of Clinical Oncology Resource-Stratified Guideline. J Glob Oncol.
2017;3:611-634.
40. Chuang LT, Temin S, Camacho R, et al. Management and care of women with invasive cervical cancer: American Society of Clinical Oncology Resource-Stratified
Clinical Practice Guideline. J Glob Oncol. 2016;2:311-340.
41. Osman H, Shrestha S, Temin S, et al. Palliative care in the global setting: ASCO Resource-Stratified Practice Guideline Summary. J Oncol Pract. 2018;
14:431-436.
42. Lopes G, Stern MC, Temin S, et al. Early detection for colorectal cancer: ASCO Resource-Stratified Guideline. J Glob Oncol. 2019;5:1-22.
43. Costas-Chavarri A, Nandakumar G, Temin S, et al. Treatment of patients with early-stage colorectal cancer: ASCO Resource-Stratified Guideline. J Glob Oncol.
2019;5:1-19.
44. Chiorean EG, Nandakumar G, Fadelu T, et al. Treatment of Patients With Late-Stage Colorectal Cancer: ASCO Resource-Stratified Guideline. JCO Glob Oncol.
2020;6:414-438.
45. The ADAPTE Collaboration. The ADAPTE Resource Toolkit for Guideline Adaptation. Version 2.0. www.g-i-n.net. Accessed February 5, 2020.
46. Loblaw DA, Prestrud AA, Somerfield MR, et al; America Society of Clinical Oncology Clinical Practice Guidelines. American Society of Clinical Oncology Clinical
Practice Guidelines: formal systematic review-based consensus methodology. J Clin Oncol. 2012;30:3136-3140.
47. ASCO. International development & education award. www.asco.org/research-guidelines/grants-awards/funding-opportunities/international-development-and-
education-award. Accessed February 5, 2020.

Hunter et al
48. Flodgren G, Hall AM, Goulding L, et al. Tools developed and disseminated by guideline producers to promote the uptake of their guidelines. Cochrane Database
Syst Rev. 2016;CD010669.
49. Echavarria MI, Anderson BO, Duggan C, et al. Global uptake of BHGI guidelines for breast cancer. Lancet Oncol. 2014;15:1421-1423.
50. Sivaram S, Sanchez MA, Rimer BK, et al. Implementation science in cancer prevention and control: a framework for research and programs in low- and middle-
income countries. Cancer Epidemiol. Biomarkers Prev. 2014;23:2273-84.
51. Mitchell SA, Chambers DA. Leveraging Implementation Science to Improve Cancer Care Delivery and Patient Outcomes. J Oncol Pract. 2017;13:523-529.
52. Abulkhair O, Saghir N, Sedky L, et al; MENA Breast Cancer Regional Guidelines Committee. Modification and implementation of NCCN guidelines on breast
cancer in the Middle East and North Africa region. J Natl Compr Canc Netw. 2010;8 (suppl 3):S8-S15.
53. Duggan C, Dvaladze AL, Tsu V, et al. Resource-stratified implementation of a community-based breast cancer management programme in Peru. Lancet Oncol.
2017;18:e607-e617.
54. Jacobson G, Chuang L, Pankow M. Improving quality of care and timely access to radiation therapy for patients with invasive cervical cancer at the National
Cancer Institute Paraguay. Gynecol Oncol Rep. 2018;25:82-86.
55. Perlman PC, Vinson C, Tulika S, et al. Multi-stakeholder partnerships: breaking down barriers to effective cancer-control planning and implementation in low- and
middle-income countries. AAAS Sci Diplomacy. 2016;5.

GLOBAL HEALTH
The Use of Health-Related Technology to Reduce

the Gap Between Developed and Undeveloped
Regions Around the Globe
Wilfred Ngwa, MS, PhD1; Ian Olver, AM, MD, PhD2; and Kathleen M. Schmeler, MD3
overview
Cancer is the second leading cause of death worldwide, with approximately 70% of the 9.6 million deaths per
year occurring in low- and middle-income countries (LMICs), where there is critical shortage of human and
material resources or infrastructure to deal with cancer. If the current trend continues, the burden of cancer is
expected to increase to 22 million new cases annually by 2030, with 81% of new cases and almost 88% of
mortality occurring in LMICs. Global health places a priority on improving health and reducing these dis-
parities to achieve equity in health for all people worldwide. In today’s hyper-connected world, information and
communication technologies (ICTs) will increasingly play an integral role in global health. Here, we focus on
how the use of health-related technology, specifically ICTs and artificial intelligence (AI), can help in closing
the gap between high-income countries (HICs) and LMICs in cancer care, research, and education. Key
examples are highlighted on the use of telemedicine and tumor boards, as well as other online resources that
can be leveraged to advance global health.
INTRODUCTION most common cancer in many regions of sub-Saharan

One of the famous quotes of Martin Luther King Jr. is Africa. This is primarily caused by a lack of access to
that “Of all the forms of inequality, injustice in health screening coupled with a shortage of specialists to treat
care is the most shocking and inhumane.” In “Closing the disease, particularly in the precancerous and early
the Cancer Divide: An Equity Imperative,”1 Knaul et al stages. High rates of cervical cancer are also seen in
discuss the shocking depth of the health inequality that medically underserved areas of the United States, also
exists in the availability and quality of cancer care in because of a lack of screening and skilled providers.
LMICs. The recent World Health Organization World One such area is the Rio Grande Valley of south Texas
Cancer Report2,3 describes the growing global burden located along the Texas-Mexico border, where the
cervical cancer incidence and mortality rates are 30%
of cancer and disparities as alarming, with more than
higher than the rest of Texas.6 Overall, for many LMIC
60% of the global burden being in LMICs in Asia,
and resource-poor settings, advanced ICTs have major
Africa, and Central and South America, where 70% of
potential to help bridge the cancer divide in cancer
cancer deaths occur. If the current trend continues,
care research and education (e.g., in telemedicine,
the burden of cancer is expected to increase to 22
tumor boards, and the use of other online resources).
million new cases annually by 2030, with 81% of new
cases and almost 88% of mortality occurring in LMICs. HOW CAN WE TAKE ADVANTAGE OF TELEMEDICINE?
These disparities in cancer incidence and mortality are The health disparities between HICs and LMICs most
Author affiliations underpinned by a poignant dearth in human and often relate to a lack of access to health care and
and support material resources or infrastructure to deal with the clinical training. There can be a lack of specialist ex-
information (if rising burden of cancer.4 Lack of access to diagnosis pertise and an insufficient medical workforce to pro-
applicable) appear vide diagnoses, treatment, and follow-up. In LMICs,
and treatment is common in LMICs, resulting in most
at the end of this
patients presenting with late-stage disease. In 2017, there may be insufficient funding to provide a full range
article.
less than 30% of LMICs reported available treatment of prevention, screening, and treatment options.7
Accepted on
February 27, 2020 services compared with more than 90% of HICs. Telemedicine or cancer-focused teleoncology is de-
and published at Cervical cancer is a notable example of the disparities fined as delivering health care at a distance and
ascopubs.org on in cancer care between high- and low-resource set- therefore provides a potential solution to some of these
March 31, 2020:
tings.5 In the United States, cervical cancer is now problems.8 In the broadest sense, telemedicine can
10.1200/EDBK_ relatively uncommon, yet it remains the second most involve real-time videoconferencing between clinicians
288613 common cancer among women in LMICs and is the or delayed storage of patient data to be forwarded

Ngwa, Olver, and Schmeler
used to remotely train specialist clinicians.11 Consultations

between remote individual patients and their primary doctors
can occur.
• Health-related technologies provide tremen-
dous opportunities for high-impact global on- In HICs, key elements to the successful use of videocon-
cology collaborations to bridge the gap between ferencing have been identified and may be applied to its use
LMICs and HICs. in developing countries.12 A bottom-up approach where
• ICTs can be particularly leveraged to overcome clinicians see a need that can be addressed by telemedicine
distance-time barriers and facilitate participa- rather than finding a use for available new technology works
tion in global oncology care, education, and best.13 It is helpful if there are “champions” at either end of
research as seen in teleoncology, tumor boards, a link, who encourage its use by their colleagues.
and collaborative research.
Videoconferencing is easier when the participants know
• Telemedicine opens up the opportunity for each other and can meet face to face occasionally. It is vital
improved patient care and upskilling clinicians that videoconferencing does not require changing clini-
and can help address the disparities in patient
cians’ practices. For example, diagnostic specialists such as
care and clinical expertise in developing na-
pathologists and radiologists should not be expected to
tions. New initiatives should be formally eval-
uated to ensure that they match the needs and provide instant opinions on slides or images if their usual
resources of the LMICs. practice is to select optimal fields and spend time examining
images before providing an opinion.14
• Virtual tumor boards such as Project ECHO
provide a way to ensure regular communica- The telemedicine equipment, when purchased, must be
tion, ongoing education and training, and skill accessible as required and easy to use. It is important that
building in patient care and management while teleoncology initiatives do not cut across established referral
also mentoring and supporting clinicians in pathways. The major downside of not being able to examine
medically underserved areas to provide optimal a remote patient means relying on the remote doctor’s
care for patients with cancer.
examination findings; therefore, the remote practitioner
• ASCO, NIH, and many other professional so- becomes part of a team approach.7 Administrative matters,
cieties, institutions, and organizations now such as who takes medicolegal responsibility for a remote
provide a vast amount of online resources to opinion, issues of licensing, and payment for telemedicine
advance cancer care, research, and education.
consultations should be resolved in advance, but the so-
Training in optimal use of ICTs and continuing
lution will differ by jurisdiction.7 Bioethical issues such as
development of AI tools would greatly facilitate
access to meaningful data and help bridge patient confidentiality apply to telemedicine just like other
global cancer health disparities. consultations.
In developed countries, the outcomes of teleoncology in-
terventions have been shown to be at least as effective as
remotely for expert opinions. Some telemedicine involves face-to-face consultations with a high level of patient and
patient to health professional communication or monitoring clinician satisfaction, as access to oncology services
of patients through wearable devices and mobile phones improves.15,16 Teleoncology has been shown to be cost
linking to the patients’ medical team or electronic records.9 effective, largely by savings from reducing travel costs.17
There is a paucity of international research into the out- Telemedicine and Developing Nations
comes of teleoncology. Most observational studies come
from HICs where teleoncology is primarily used to address Where once a limitation to applying telemedicine to LMICs
disparities in cancer outcomes between urban and rural and was the cost of the equipment and networks, now with
remote communities.10 Although lessons from studies in widespread internet access, affordable personal computers
HICs may be applicable to LMICs, the programs in those and one-third of the population from developing nations
countries must be aligned to their needs, budgets, and owning a smartphone, the barrier of unaffordable tech-
priorities.7 nology is less problematic.18 However, the human factors
previously reported with earlier adoption of teleoncology
Teleoncology may still be limiting its use.7,19 For example, some LMICs
Teleoncology allows interaction between health professionals have a proliferation of underused videoconferencing cen-
in real time using videoconferencing. This encompasses ters, whereas others have been very successful.20,21 Failure
multidisciplinary meetings with multiple specialties and allied to use telemedicine has been ascribed to fear of change,
health teams to discuss cases or one-to-one mentoring. loss of political control or professional control over patients,
Videoconferencing is also used for education and can be and a reluctance to seek second opinions.

Bridging Global Cancer Health Disparities With Health-related Technology
The beliefs of Senegal’s physicians toward telemedicine across a country. Teleconsultation could also be used for
were recently published, and the negatives included the patient follow-up.
potential for medical errors, issues with data security, pos- Telesurgery Although feasible, it is unlikely that expensive
sible additional costs to patients through overinvestigating robotic surgeries from a distance can be widely adopted in
them, and consultations with other physicians being inter- LMIC.23 However, with augmented reality telesurgery over
preted as a lack of their doctor’s skill by patients.22 Further the internet, an expert surgeon from a developed nation
barriers included whether the use of telemedicine was seen could be virtually in the operating room to guide local
to be more time consuming, the equipment was too difficult surgeons in procedures in real time. Augmented reality
to use, and inadequate training on its use was offered. The superimposes a computer-generated image over a live video
positive beliefs centered around enhanced care through stream, which could enable communication using hands or
access to expert opinions and the opportunity for continuing diagrams over the real-world image. It only requires cam-
education and research. Being able to treat patients remotely eras, screens, computers, and sufficient internet band-
and the advantage of reduced travel time and expense for width; therefore, it is a feasible application of telemedicine. It
both patients and physicians contradicted the concept of has already been described to assist surgery in Gaza.27
increased costs to patients.23
Diagnosis: Teleradiology and pathology Telemedicine can
Use of teleoncology to link centers within Pakistan has been be used to deliver expert pathology and radiology opinions.
reported as improving coordinated care with more stan- One method is storing the images and forwarding from the
dardization between hospitals and avoiding duplication of LMIC to be read and returned by the expert reviewer in the
investigations.24 Doctors have realized the benefits of HIC.15 In real time, a high-definition video microscopy
continuing medical education by linking with other spe- technician can select a section of a cytopathology slide and
cialists. In a country where oncology services are frag- stream an image to a centralized location for reading.28
mented, teleoncology is reported as reducing the burden on Virtual telepathology slides can be read remotely and can
the health system and the family. be bundled with teleradiology, giving patients rapid access
Patient Care to high-quality diagnostic expertise.29 This could easily be
A common use of teleoncology for developing nations is used between HICs and LMICs.
videoconferencing from leading centers in HICs to add Clinical trials Most clinical trials are conducted in major
specialist expertise to the clinical decision-making in mul- centers; however, a teletrial model in Australia may be
tidisciplinary meetings. This works best if it supplements applicable to LMICs. The remote center is a satellite of the
existing collaborations between medical centers and clini- primary center from where the trial is administered, whereas
cians, and it is ideal if it involves multiple centers in the patients are recruited and treated by the satellite site cli-
LMIC.23 The technology and the opinions given must match nician under the direction of clinicians at the primary site via
the available resources within the LMIC, and clearly it is telemedicine.30
easier if the time zones match. The multidisciplinary Training and Education
meetings can relieve the feeling of isolation from main-
stream oncology, but whether that will result in better re- Telemedicine is well established for distance education,
tention of clinicians is unclear. which can be applied to training programs from HICs to
LMICs. This can be specific group education or that which
Chemotherapy It is possible that, in addition to clinical occurs at multidisciplinary case conferences. It can involve
advice, teleoncology to developing nations could be used individual mentoring of LMIC clinicians, which works best if
to supervise chemotherapy. Of greatest importance would they already know each other. Some programs have used
be safety. An encouraging example from Australia is that telemedicine to supervise junior specialist trainees who are
chemotherapy in remote small towns, supervised by on- on clinical rotations in remote areas.11 To achieve the
cologists from larger centers by telemedicine, has been greatest likelihood of success, the education provider will
found to achieve the same dose intensity and safety as in the require training in how to educate, mentor, or supervise
larger center.25 using telemedicine.31
Radiotherapy Many LMICs do not have radiotherapy WEB-BASED TUMOR BOARDS
equipment because of its expense. Many that do lack the
personnel and expertise to optimally use it. Telemedicine Virtual Tumor Boards Using Project ECHO Telementoring
could be used to consult international experts over patient Virtual tumor boards using Project ECHO provide a poten-
treatments. It is also possible to remotely plan a patient’s tial solution for the disparities in oncology specialists be-
external beam and brachytherapy and deliver the sophis- tween low-resource and high-resource settings in the
ticated planning to the treatment machine in the LMIC.26 United States and globally. Project ECHO was developed in
Centers could be networked and share higher-level expertise 2003 by Sanjeev Arora, a hepatologist at the University of

New Mexico (UNM), to improve both provider capacity and Anderson team partnered with physicians, nurse practi-
access to specialty care for rural and underserved pop- tioners, nurse midwives, and physician assistants in the
ulations.32 Project ECHO links multidisciplinary specialist region and provided hands-on training courses for cervical
teams with community clinicians through regularly sched- cancer prevention procedures such as colposcopy, cervical
uled telementoring sessions, in which the participants use biopsy, and loop electrosurgical excision procedure. In
videoconferencing to comanage patient cases. Specialists parallel, they implemented Project ECHO sessions for
share their expertise via mentoring, guidance, feedback, 1 hour every 2 weeks. Continuing Medical and Nursing
and didactic education. This approach has enabled clini- Education credits are awarded, free of charge, after each
cians in medically underserved areas to develop the skills, session. The first 45 minutes involve case presentations
confidence, and knowledge to treat patients with common, (without patient-identifying information) by the community
complex diseases in their own communities, thereby re- providers, and the specialists provide feedback. This is
ducing travel costs, wait times, and avoidable complica- followed by a 15-minute didactic presentation by a partici-
tions. Project ECHO is different from telemedicine, where pating faculty member or a guest lecturer. This initial
the specialist assumes the care of the patient, but instead Cervical Cancer Prevention ECHO is now in its fifth year and
involves telementoring, where the community clinician re- has grown to include additional partners along the Texas-
tains responsibility for managing the patient, operating Mexico border and other medically underserved commu-
with increasing independence as their skills and self- nities around Texas.
efficacy grow.
MD Anderson has expanded the initiative and currently
The first Project ECHO program was developed for the hosts 14 different Project ECHO programs directed at
management of hepatitis C virus (HCV) in patients in rural cancer prevention (tobacco control and melanoma), cancer
New Mexico. Providers from 16 rural community clinics and treatment (breast, head and neck, gynecologic, and he-
five prisons throughout New Mexico participated in weekly matologic malignancies in Mozambique), palliative care
Project ECHO HCV telementoring sessions with specialists (Africa), and cancer survivorship (Texas). Over the last year,
from UNM, presenting their cases, including patients’ the institution held 163 different ECHO sessions with 794
medical histories, laboratory results, treatment plans, and providers from 27 participating countries. In 2017, UNM
individual challenges; asking questions; and receiving designated MD Anderson the ECHO Superhub for Oncol-
guidance about best practices. Specialists at UNM provided ogy, and they have trained more than 150 individuals from
advice and clinical mentoring. Working together, the 36 organizations and institutions to adopt and implement
community providers and specialists managed the patients’ the Project ECHO model for cancer.
care according to evidence-based guidelines. The effec-
tiveness of the program was evaluated in a prospective Project ECHO Virtual Tumor Boards for Gynecologic
cohort study of 407 patients with chronic HCV published in Oncology Training in LMICs
The New England Journal of Medicine in 2011.33 This study To address the lack of specialists providing cervical cancer
compared outcomes between patients treated by specialists prevention and treatment in LMICs, the International Gy-
at UNM with patients treated by primary care providers at necologic Cancer Society developed a Global Gynecologic
the 21 rural ECHO clinics. There were no significant dif- Oncology Fellowship Program.36 This comprehensive pro-
ferences in cure and/or complication rates between the two gram provides structured support and training in gyneco-
cohorts. Improvements were noted in patient satisfaction, logic oncology to physicians in countries in which no formal
physician self-efficacy, and patient outcomes while con- training program exists by pairing them with gynecologic
comitantly reducing regional disparities in evidence-based oncology mentors from high-income settings. The program
HCV management across New Mexico. Project ECHO has includes a structured web-based curriculum, self-study,
since expanded to cover more than 50 other specialty areas regular assessments, hands-on surgical training, regular
across the United States and globally including diabetes, visits by the international mentors to the training site, and
addictions, rheumatology, HIV/AIDS, dementia, palliative a 3- to 6-month visit by the trainee to the international
care, autism, and cardiovascular disease.34 mentor’s home institution. Project ECHO tumor boards are
an important component of the program and allow for
Adopting Project ECHO for Cancer Prevention a virtual connection on a monthly basis to complement in-
and Treatment person visits. The fellows are responsible for case pre-
The University of Texas MD Anderson Cancer Center (MD sentations. The local mentors, international mentors, and
Anderson) adopted the Project ECHO model for cancer other specialists (radiation oncology, pathology, radiology,
prevention and treatment in 2014.35 The initial project was medical oncology) join remotely for a multidisciplinary
focused on cervical cancer prevention and management of discussion. There are currently 12 International Gyneco-
preinvasive disease along the Texas-Mexico border. The MD logic Cancer Society fellowship training sites around the

globe, with 30 fellows and 26 international and local 180 zettabytes by 2025. Many LMICs have leapfrogged the
mentors. As of December 2019, the program has held 136 analog era directly into the digital era with access to data via
Project ECHO sessions with more than 300 participating the internet, with deep penetration of mobile phones and
providers. other ICTs.
Metrics and Evaluation Examples of Online Resources for Global Oncology
Evaluation is ongoing and includes three components: Online resources for cancer care The internet has a vast
process metrics, provider satisfaction, and levels of col- amount of cancer-related content. Many health care pro-
laborations. Process metrics include number of partici- viders are increasingly making their practice guidelines
pants, number of ECHO sessions held, and the number of available for online viewing, downloading, or both. A growing
individual cases discussed. The provider satisfaction and number of resources are available to also help cancer health
self-efficacy are measured at baseline, with annual follow- professionals answer specific clinical questions at the point
up surveys. Furthermore, collaborative efforts are measured of care by providing extensive cancer-related basic science,
through number of workshops successfully delivered; num- diagnostic, and therapeutic content. Meanwhile, other
ber of joint research programs; number of providers partici- online resources are continuously developed to help im-
pating in colposcopy, surgery, and other workshops; and prove patient adherence to recommended treatment regi-
number of observerships and trainee exchanges completed. mens and healthier lifestyles. Many resources are available
Some programs are also measuring changes in the number to further help health care professionals and organizations
of patients undergoing cancer screening and treatment implement improved systems of cancer care, focusing on
procedures. evidence-based decision-making, effective information
systems, and meeting the needs of populations from diverse
Challenges
backgrounds. Resources available online can also help
ECHO participants, particularly those engaged with in- those wishing to purchase and implement electronic cancer
ternational partners in LMICs, must consider local context registries or medical records. AI is being trained to rapidly
and local resources in care delivery. The standard of care in access big data, including open-access, high-impact re-
the United States is not feasible in many regions, and search publications to make recommendations for physi-
creative solutions for providing basic services, given limited cians. Table 1 outlines some examples of online resources
resources, provide a basis for many discussions. In addition for cancer care. One example is Cure4Kids, developed for
to resource limitations, cultural differences and difficulty health care professionals dedicated to enhancing the care
initiating change can create challenges and often require of children who have cancer and other life-threatening
unique, region-specific strategies for care delivery. Regular diseases in countries around the globe. Closing the sur-
videoconferences help to build trust and encourage de- vival gap using a collective effort with such resources is
velopment of partnerships by exchanging information and already having significant impact in pediatric oncology.39,40
knowledge. Project ECHO videoconferences require inter-
net connections, and in some regions, this is a major Online resources for education For both LMICs and HICs,
challenge. Some alternative strategies have been de- online education resources have revolutionized education,
veloped, including additional phone connections if the in- partly because of their convenience and accessibility, but
ternet connection is unstable; however, this remains an also because they make the entire process of teaching and
issue for some partners. learning more interesting and memorable in many ways.
Health care professionals and trainees may prefer different
ONLINE RESOURCES CAN IMPROVE ACCESS TO resources according to their subjects of interest and
MEANINGFUL DATA learning styles, but there are many to choose from. Today,
Telemedicine and tumor boards are made possible by the internet is a major resource for many growing up in LMIC
advanced ICTs. In general, ICTs refer to devices or systems compared with before the early 2000s, when there was
that allow the storage, retrieval, manipulation, transmission, minimal penetration of ICTs. ASCO has dedicated efforts to
and receipt of digital/electronic data.4,37 Examples in- the education of the next generation of oncologists, pro-
clude smart phones, personal computers, digital television, viding extensive resources and programs designed for all
email, social media platforms, radio, Skype, Zoom, WebEx, levels of oncologists, including medical students, residents,
GoToMeeting, etc. These ICTs can also be leveraged to fellows, and program directors, which can be accessed for
access the exponentially growing amounts of data available global health. Table 1 includes examples of some of these
online (big data) with the help of AI.4,38 Although three- resources that can help for those who are exploring different
quarters of data were analogous in 2000, today more than fields, as well as those who are deciding on a subspecialty.
99% of all data are digital data, with an estimated amount of These resources are also helping practicing oncologists to
44 zettabytes of digital data in 2020, which should grow to continue building on their skills and broaden opportunities

TABLE 1. Examples of Online Resources/Platforms That Can Be Accessed for Global Oncology
Online Resources or Platforms Description
Care
Cure4kids, https://www.cure4kids.org/ Cure4Kids is an online resource for health care professionals dedicated to enhancing the
care of children who have cancer and other life-threatening diseases in countries
around the globe
ProKnow, https://proknowsystems.com/ Vendor-neutral remote treatment planning and quality assurance information and
communication technologies platform, software; can be used in tele-oncology to
increase access to quality treatment
Watson for Genomics Scale precision medicine, help oncologists create personalized treatment strategies for
their patients based on results from genetic testing of tumor tissue, hence more quality
treatment
Video-conferencing software: GoToMeeting, Zoom, WebEx Tumor boards can increase access to quality treatment especially for complex cases
Education
ASCO Resources for Medical Students & Residents Resources to support career development, access to networking and mentoring
opportunities, access to ASCO’s Medical Student & Resident myConnection online
community, etc.
Chartrounds, https://www.chartrounds.com Chartrounds.com, build capacity for quality cancer treatment especially for complex cases
i.treatsafely.org, https://i.treatsafely.org/ High-quality learning videos that deliver practical clinical and quality assurance skills
Online oncology degree programs (e.g., University of Build capacity for quality cancer treatment
Heidelberg)
Video-conferencing information and communication Training residents and oncology professionals during tumor boards to increase access to
technologies quality treatment
ASCO Resources for Program Directors This online platform provides members with the opportunity to connect, collaborate, and
participate in in-depth discussions; receive feedback or resources from other
professional peers; and build meaningful connections
ASCO Social Media Resources Social media serves as one of the major public health engines of the 21st century, granting
doctors and the general public alike unprecedented access to information on cancer
prevention, treatments, research, and survival
ASCO’s patient information website, Cancer.Net Cancer.Net offers trusted and compassionate information for people with cancer and their
families and caregivers; the comprehensive content includes sections devoted to cancer
types, navigating cancer care, coping with cancer, survivorship, caregiving, and
research and advocacy (there is also a blog, as well as podcasts and videos)
Research
ArrayExpress: functional genomics data, ArrayExpress Archive of Functional Genomics Data stores data from high-throughput
https://www.ebi.ac.uk/arrayexpress/ functional genomics experiments and provides these data for reuse to the research
community
Watson for clinical trials Aid patient recruitment into clinical trials, advancing research on new therapies that could
increase access to treatment
Quality Assurance Review Center (QARC), Ensure quality assurance in clinical trials including multicenter clinical trials
https://www.qarc.org/
Cancer Genomics Cloud, Provides database for research to generate evidence base for personalized treatments
http://www.cancergenomicscloud.org/
Electronic health records database Allow for epidemiology research to inform policy toward increased access
Global Oncology Map With the goal of improving access to cancer care for people living in the world’s lowest-
resourced areas, this online map hosts a panapoly of information about who, what,
where, and how cancer care and research are transpiring around the globe; facilitates
research collaborations that will lead to increase access to quality cancer care
DNA Data Bank of Japan (DDBJ) Provides sharing and analysis services for data from life science researches and advances
science
The Database of Genotypes and Phenotypes (dbGaP), Archives and distributes data and results from studies that have investigated the
https://www.ncbi.nlm.nih.gov/gap/ interaction of genotype and phenotype in humans

TABLE 1. Examples of Online Resources/Platforms That Can Be Accessed for Global Oncology (Continued)
Online Resources or Platforms Description
European Nucleotide Archive (ENA), Provides a comprehensive record of the world’s nucleotide sequencing information,
https://www.ebi.ac.uk/ena covering raw sequencing data, sequence assembly information, and functional
annotation
GenBank The National Institutes of Health genetic sequence database provides an annotated
collection of all publicly available DNA sequences
to provide the highest-quality care across the continuum, Using AI in Global Oncology
from prevention to treatment to survivorship. ICTs present enormous opportunities for facilitating activi-
Online resources for research Table 1 also highlights ex- ties that can have a significant impact right away, enhancing
amples of online resources for research and research greater effectiveness of ongoing global oncology initiatives
collaborations. The National Institutes of Health (NIH) in the and converting the major recent upsurge in global health
United States has many online resources available and interest into greater concerted action toward improved
major policies in place for data sharing. One example is the health outcomes, saved lives, and closing the cancer di-
cancer genomics cloud, which provides a database for vide.4 In the meantime, AI is being viewed as the next step in
research to generate an evidence base for personalized ICT evolution, with disruptive technologies being developed
treatments. This could also be very helpful in global health and deeply permeating various applications and markets all
research involving immigrant populations, as the primary around the world including in health care. AI is the ability of
drivers of cancer incidence and deaths in LMIC underlie the a computer to perform tasks such as reasoning and learning
higher cancer-related morbidity and mortality rates also what human intelligence can do and entails the use of al-
seen among immigrant populations. gorithms and software to approximate human cognition in
the analysis of complex data without direct human input.41
With ICTs, cloud computing platforms can also facilitate
Progress in AI along with the accessibility of cloud scaling for
global research collaborations to close the gap between
big data storage and integration of health records now
LMICs and HICs. In 2015, the American Society of Radiation
present unprecedented opportunities in health care.42 AI
Oncology, the National Cancer Institute, and the American
can empower health care professionals and improve the
Association of Physicists in Medicine cosponsored a 2-day
efficiency and quality of care globally.43-45 Consequently,
workshop for oncology health professionals focused on
there is growing optimism that the deployment of AI could
opportunities in oncology in the era of big data. The big data
revolutionize global cancer control.4,37,43-45
Workshop highlighted opportunities in big data research
such as current big data cancer registries, safety and in- Currently available AI includes mainly industry-based so-
cident reporting systems, and other strategies that will have lutions. For example, Watson for Oncology, a Memorial
the greatest impact on research, quality assurance, safety, Sloan Kettering–trained cognitive computing system, has
and outcomes analysis. been developed to extract structured and unstructured data
Another example where online data can help is in radiomics. from medical records; ingest a large body of published
Here, radiomics refers to the extraction and analysis of large evidence and guidelines; and provide evidence-based
amounts of advanced quantitative imaging features with high treatment recommendations tailored to an individual pa-
throughput from medical images obtained with CT, PET, or tient. Some preliminary implementation research has been
MRI. Importantly, these data are usually extracted from conducted on this AI in Manipal, India, where it was found
clinical standard-of-care images, which could result in a very that Watson for Oncology recommendations were 90% in
large potential subject pool. Besides data, online technolo- concordance with expert tumor board recommendations.46
gies are also increasingly being used to advance global health This AI can be used to ensure doctors offer the best-
research collaborations. For example, technologies such as available quality treatment to patients with cancer in
Zoom, WebEx (video-conferencing software), and Goto- LMICs. It is perhaps also well positioned to support tele-
Meeting are used for tumor boards and are increasingly being medicine, allowing for patients to access treatment from
used to enable meetings among collaborators or even remote areas in LMIC where there are no oncologists with the
participation at conferences, providing opportunities for needed specialist knowledge. Such a system would work
greater interaction to begin or nurture scientific collabora- really well where there are already electronic health records
tions. These examples of online resources are by no means for the patients and can allow for task-shifted care where
exhaustive. They represent examples of the vast trove of oncology nurses or other health staff can input the in-
online resources that can be leveraged to improve access to formation. In general, advantages of using electronic re-
meaningful data with the advent of ICTs. cords include enabled access to medical records from

remote locations, improved speed and ease of retrieval of opportunity to engage many different stakeholders and
records, avenues to flag abnormal results, and the elimi- resources.
nation of handwritten prescriptions, which reduces the oc- Examples in cancer care include telemedicine, which al-
currence of prescription errors. Other benefits are the lows the opportunity for improved patient care and ups-
simultaneous access to patient records by multiple users and killing clinicians in developing nations. New initiatives
the ability to perform data queries to inform decision-making. should be formally evaluated to ensure that they match the
Implementation research to test this AI in different LMICs needs and resources of the LMICs. Linked with advances in
could facilitate wide adoption in what is clearly a next major the use of mobile technology in health care, wearables to
frontier in oncology health care to close the cancer divide. monitor patients remotely, and smartphones for enhanced
However, the cost of integrating AI into global oncology has communication, telemedicine can help address the dis-
been raised as a concern,46 and therefore, the cost- parities in patient care and clinical expertise in developing
effectiveness of any AI must be considered in resource- nations.
limited settings like LMICs. Perhaps one way to address the Tumor boards such as Project ECHO virtual tumor boards
costs is to work with partners as done recently in the col- provide a way to ensure regular communication, ongoing
laborative initiative between the American Cancer Society, education and training, and skill building in patient care and
Pfizer, and Cipla to steeply discount the prices of cancer management. Because the tumor board is held virtually,
medicines in Africa. In the deal, top American oncologists there is opportunity for participation of multidisciplinary
will simplify complex cancer treatment guidelines for specialists from around the globe. In some cases, this is the
underequipped African hospitals, and a corps of IBM first exposure providers get to the practice of multidisci-
programmers will build those guidelines into an AI tool plinary cancer care. Project ECHO virtual tumor boards
available to any oncologist with an internet connection.47 provide training, mentoring, and support for clinicians in
Such collaborative efforts provide a model for addressing medically underserved areas to provide optimal care for
costs. Including local computer experts and oncologists in patients with cancer.
cross-disciplinary implementation would be recommended
as a way to reduce costs and ensure tools are optimized for In research, ICTs now allow for the possibility of collabo-
those settings. The use of ICTs and AI is anticipated to be rations in the cloud including the use of platforms like the
cost-effective and increase access to quality cancer care NIH-funded quality assurance review platform (Table 1)48
(e.g., in telemedicine). The NIH is not providing funding for multicenter clinical trials across different countries.
mechanisms to support the development of AI and ICT tools Other online resources include data-sharing platforms like
suitable for the LMIC setting. those provided by NIH for research and online education
resources like those provided by ASCO. The examples
SUMMARY AND CONCLUSIONS presented here are by no means exhaustive but suggest the
The recent World Health Organization World Cancer Report need for greater concerted effort in facilitating the use of
describes the growing global burden of cancer and dis- telemedicine, tumor boards, and other online resources and
parities as alarming. In efforts to reduce these disparities tools in closing the cancer divide. With ICTs and AI in the era
and collaborate in closing the cancer divide, ICT-based of big data, one can now imagine or dream of a future with
approaches and online resources present major opportu- a comprehensive cancer center in the cloud, with tele-
nities in cancer care, research, and education. The advantages medicine, tumor boards, and collaborative research ca-
presented by these technologies and online resources pability providing access from anywhere in the world
include overcoming distance barriers to global health as including from LMICs to the same world-class cancer care,
these can be accessed from any country in the world, research, and education available at the best cancer centers
convenience to access at any time, lower costs, and the in HICs.

1
Department of Radiation Oncology, Brigham and Women’s Hospital, Wilfred Ngwa, MS, PhD, Harvard Institutes of Medicine 347, 4 Blackfan
Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA Circle, Boston, MA 02115; Twitter: @WilNgwa; email: wngwa@lroc.
2
School of Psychology, Faculty of Health and Medical Sciences, harvard.edu.
University of Adelaide, Adelaide, Australia
3
Department of Gynecologic Oncology and Reproductive Medicine, The
University of Texas MD Anderson Cancer Center, Houston, TX
EDBK_288613.

REFERENCES
1. Gerhard D. Closing the cancer divide: an equity imperative. Yale J Biol Med. 2013;86:282-283.
2. Wild CP, Weiderpass E, Stewart BW (eds). World Cancer Report: Cancer Research for Cancer Prevention. World Cancer Reports. Lyon: International Agency for
Research on Cancer; 2020.
3. Stewart BW, Wild CP (eds). World Cancer Report 2014. Lyon: International Agency for Research on Cancer; 2014.
4. Ngwa W, Ngoma T, Zietman A, et al. Closing the cancer divide through Ubuntu: information and communication technology-powered models for global radiation
oncology. Int J Radiat Oncol Biol Phys. 2016;94:440-449.
5. Arbyn M, Weiderpass E, Bruni L, et al. Estimates of incidence and mortality of cervical cancer in 2018: a worldwide analysis. Lancet Glob Health. 2020;8:e191-
e203.
6. Boom K, Lopez M, Daheri M, et al. Perspectives on cervical cancer screening and prevention: challenges faced by providers and patients along the Texas-Mexico
border. Perspect Public Health. 2019;139:199-205.
7. Hazin R, Qaddoumi I. Teleoncology: current and future applications for improving cancer care globally. Lancet Oncol. 2010;11:204-210.
8. Wootton R. Telemedicine: a cautious welcome. BMJ. 1996;313:1375-1377.
9. Tuckson RV, Edmunds M, Hodgkins ML. Telehealth. N Engl J Med. 2017;377:1585-1592.
10. Onega T, Duell EJ, Shi X, et al. Geographic access to cancer care in the U.S. Cancer. 2008;112:909-918.
11. Cameron M, Ray R, Sabesan S. Remote supervision of medical training via videoconference in northern Australia: a qualitative study of the perspectives of
supervisors and trainees. BMJ Open. 2015;5:e006444.
12. Olver IN, Selva-Nayagam S. Evaluation of a telemedicine link between Darwin and Adelaide to facilitate cancer management. Telemed J. 2000;6:213-218.
13. Yellowlees P. Successful development of telemedicine systems—seven core principles. J Telemed Telecare. 1997;3:215-222.
14. Weinstein RS, Graham AR, Richter LC, et al. Overview of telepathology, virtual microscopy, and whole slide imaging: prospects for the future. Hum Pathol. 2009;
40:1057-1069.
15. Sirintrapun SJ, Lopez AM. Telemedicine in cancer care. Am Soc Clin Oncol Educ Book, Vol. 38. 2018;540-545.
16. Doolittle GC, Allen A. Practising oncology via telemedicine. J Telemed Telecare. 1997;3:63-70.
17. Thaker DA, Monypenny R, Olver I, et al. Cost savings from a telemedicine model of care in northern Queensland, Australia. Med J Aust. 2013;199:414-417.
18. Tiene D. Bridging the digital divide in the schools of developing countries. Int J Instr Media. 2004;31:89-99.
19. Wootton R. Telemedicine support for the developing world. J Telemed Telecare. 2008;14:109-114.
20. Hsieh RKC, Hjelm NM, Lee JCK, et al. Telemedicine in China. Int J Med Inform. 2001;61:139-146.
21. Quintana Y, Safran C. Global challenges in people-centered e-health. Stud Health Technol Inform. 2015;216:977.
22. Ly BA, Labonté R, Bourgeault IL. The beliefs of Senegal’s physicians towards the use of telemedicine. Pan Afr Med J. 2019;34:97.
23. Murad MF, Ali Q, Nawaz T, et al. Teleoncology: improving patient outcome through coordinated care. Telemed J E Health. 2014;20:381-384.
24. Cameron MPL, Ray R, Sabesan S. Physicians’ perceptions of clinical supervision and educational support via videoconference: a systematic review. J Telemed
Telecare. 2014;20:272-281.
25. Chan BA, Larkins SL, Evans R, et al. Do teleoncology models of care enable safe delivery of chemotherapy in rural towns? Med J Aust. 2015;203:406-6.e6.
26. Datta NR, Heuser M, Samiei M, et al. Teleradiotherapy network: applications and feasibility for providing cost-effective comprehensive radiotherapy care in low-
and middle-income group countries for cancer patients. Telemed J e-Health. 2015;21:523-532.
27. Greenfield MJ, Luck J, Billingsley ML, et al. Demonstration of the effectiveness of augmented reality telesurgery in complex hand reconstruction in Gaza. Plast
Reconstr Surg Glob Open. 2018;6:e1708.
28. Sirintrapun SJ, Rudomina D, Mazzella A, et al. Successful secure high-definition streaming telecytology for remote cytologic evaluation. J Pathol Inform. 2017;
8:33.
29. López AM, Graham AR, Barker GP, et al. Virtual slide telepathology enables an innovative telehealth rapid breast care clinic. Semin Diagn Pathol. 2009;
26:177-186.
30. Sabesan S, Zalcberg J. Telehealth models could be extended to conducting clinical trials-a teletrial approach. Eur J Cancer Care (Engl). 2018;27:e12587.
31. Pandit T, Ray RA, Sabesan S. Managing Emergencies in Rural North Queensland: The Feasibility of Teletraining. Int J Telemed Appl. 2018;2018:8421346.
32. Arora S, Kalishman S, Thornton K, et al. Expanding access to hepatitis C virus treatment—extension for Community Healthcare Outcomes (ECHO) project:
disruptive innovation in specialty care. Hepatology. 2010;52:1124-1133.
33. Arora S, Thornton K, Murata G, et al. Outcomes of treatment for hepatitis C virus infection by primary care providers. N Engl J Med. 2011;364:2199-2207.
34. ECHO Institute. Project ECHO. https://echo.unm.edu/. Accessed February 2, 2020.
35. Lopez MS, Baker ES, Milbourne AM, et al. Project ECHO: a telementoring program for cervical cancer prevention and treatment in low-resource settings. J Glob
Oncol. 2017;3:658-665.

36. International Gynecologic Cancer Society. Global curriculum & mentorship program. https://igcs.org/mentorship-and-training/global-curriculum/. Accessed
February 29, 2020.
37. Ngwa W, Sajo E, Ngoma T, et al. Potential for information and communication technologies to catalyze global collaborations in radiation oncology. Int J Radiat
Oncol Biol Phys. 2015;91:444-447.
38. Davies SE. Artificial intelligence in global health. Ethics Int Aff. 2019;33:181-192.
39. Richardson S, Kettinger WJ, Banks MS, et al. IT and agility in the social enterprise: a case study of St Jude Children’s Research Hospital’s ‘Cure4Kids’ IT-platform
for international outreach. J Assoc Inf Syst. 2014;15:Article 2.
40. Quintana Y, Van Kirk Villalobos A, Li Z, et al. Cure4Kids for Kids: preliminary results on evaluating knowledge acquisition and knowledge retention. Stud Health
Technol Inform. 2012;172:120-128.
41. Jiang F, Jiang Y, Zhi H, et al. Artificial intelligence in healthcare: past, present and future. Stroke Vasc Neurol. 2017;2:230-243.
42. Syrjala KL. Opportunities for improving oncology care. Lancet Oncol. 2018;19:449.
43. Meskó B, Hetényi G, Gy}
orffy Z. Will artificial intelligence solve the human resource crisis in healthcare? BMC Health Serv Res. 2018;18:545.
44. The Lancet. Artificial intelligence in health care: within touching distance. Lancet. 2018;390:2739.
45. Tarassenko L, Watkinson P. Artificial intelligence in health care: enabling informed care. Lancet. 2018;391:1260.
46. Gyawali B. Does global oncology need artificial intelligence? Lancet Oncol. 2018;19:599-600.
47. McNeil D. As cancer tears through Africa, drug makers draw up a battle plan. The New York Times. https://www.nytimes.com/2017/10/07/health/africa-cancer-
drugs.html. Accessed February 2, 2020.
48. Ngwa W, Ikhile E, Ntekim A, et al. Potential role of the quality assurance review center platform in global radiation oncology. Int J Radiat Oncol Biol Phys. 2017;
99:956-962.

GYNECOLOGIC CANCER

GYNECOLOGIC CANCER
A Review of Immune Checkpoint Blockade

Therapy in Endometrial Cancer
Angela K. Green, MD, MPH1; Jacqueline Feinberg, MD2; and Vicky Makker, MD1
overview
Approximately 30% of primary endometrial cancers are microsatellite instability high/hypermutated (MSI-H),
and 13% to 30% of recurrent endometrial cancers are MSI-H or mismatch repair deficient (dMMR). Given the
presence of immune dysregulation in endometrial cancer as described, immune checkpoint blockade (ICB)
has been explored as a therapeutic mechanism, both as monotherapy and in combination with cytotoxic
chemotherapy, other immunotherapy, or targeted agents. In MSI-H or dMMR advanced endometrial cancers,
PD-1 inhibitors dostarlimab and pembrolizumab have shown response rates of 49% and 57%, respectively,
whereas PD-L1 inhibitors avelumab and durvalumab have shown response rates of 27% and 43%, re-
spectively. In microsatellite stable (MSS) or PD-L1–positive advanced endometrial cancers, modest activity of
PD-1 inhibitors nivolumab and dostarlimab and PD-L1 inhibitors atezolizumab, avelumab, and durvalumab
has been seen, with response rates ranging from 3% to 23%. Based on substantial activity in a phase Ib/II
study, the U.S. Food and Drug Administration (FDA) granted lenvatinib and pembrolizumab combination
therapy accelerated approval in 2019 for the treatment of advanced endometrial cancer that is not MSI-H or
dMMR and has progressed following prior therapy. Although these developments have been highly impactful,
a more robust understanding of the molecular and immunologic drivers of response and resistance will be
critical to optimally design next-generation studies in endometrial cancer.
BACKGROUND mechanisms have been proposed to contribute to

Endometrial cancer is composed of a collection of distinct this obesity-cancer association, including high levels
histologic subtypes that, in aggregate, constitute the most of circulating and local estrogens, altered adipokine
common gynecologic malignancy in the United States, (leptin and adiponectin) composition and physiology,
with an estimated 65,620 new cases and 12,590 deaths disrupted insulin/insulin-like growth factor signaling,
in 2020; prognosis within this diverse group of cancers modifications within the microbiome, cellular and vas-
is based largely on histologic grade and clinical stage.1,2 cular perturbations, and local and systemic effects of
Although 67% of patients present with early-stage inflammation.9
disease that is associated with an 81% 5-year over- ENDOMETRIAL CANCER MANAGEMENT
all survival rate, the 5-year overall survival rate for stage
IVA and IVB endometrial cancer is only 17% and 15%, Initial endometrial cancer management consists of
respectively.3 Endometrial cancer–associated mortality surgery followed by radiation and/or cytotoxic therapy
has increased by approximately 1.4% per year from for patients with adverse risk factors. Advanced or
2005 to 2014, and this alarming trend is expected to recurrent disease is treated with cytotoxic, targeted, or
continue.2,3 hormonal therapy and/or palliative radiation depending
on the histology, disease location, and extent or bulk of
Author affiliations
RISK FACTORS disease.6 Because overall survival remains poor for
and support Risk factors associated with endometrial cancer in- these advanced and recurrent cases, the advent of
information (if
clude early menarche, late menopause, obesity, ex- novel targeted and ICB therapies for advanced en-
applicable) appear
at the end of this
posure to tamoxifen, and nulliparity. 4-6 Obesity is dometrial cancer brings hope to improve outcomes.
article. a strong modifiable risk factor for the development of
endometrial cancer, accounting for approximately ENDOMETRIAL CANCER CLASSIFICATION
Accepted on March
4, 2020 and 50% of cases in Europe and the United States; obesity Historically, endometrial cancer has been classified
published at is also associated with a 6.25 times relative increased into two Bokhman histopathologic categories, which
ascopubs.org on
risk of death.7 The metabolic syndrome, a condition describe the pathogenesis, behavior, and risk factors.10
March 26, 2020:
characterized by obesity, insulin resistance, and dys- Type 1 endometrial cancer describes FIGO (International
10.1200/EDBK_ lipidemia, is rising in prevalence among U.S. adults and Federation of Gynecology and Obstetrics) grade 1 and
280503 is also associated with cancer development.8 Several 2 endometrioid tumors, representing approximately

Immunotherapy in Endometrial Cancer
classifications for endometrial cancer that have informed

potential targeted and immuno-oncology treatment strate-
gies for the various subgroups.14 As described by The
• In the phase II KEYNOTE-158 trial, pem- Cancer Genome Atlas, the four distinct endometrial cancer
brolizumab was associated with an ORR of 57%
molecular subtypes, each with its own prognostic signifi-
for the MSI-H/dMMR endometrial cancer
cance, are as follows: polymerase " (POLE)-mutant/ultra-
cohort, consistent with previously reported
efficacy. mutated, MSI-H, copy number low, and copy number high.
MSI-H is the phenotype of deficient proteins in the mis-
• In an ongoing phase I/II study, dostarlimab
match repair (MMR) pathway, which may be either sporadic
showed an ORR of 49% in MSI-H advanced
or inherited mutations as in Lynch syndrome,15 leading to
endometrial cancer versus 20% in MSS tumors.
the accumulation of high mutational loads. Tumors that are
• In the final efficacy analysis of KEYNOTE-146 (a not dMMR/MSI-H are considered MSS.
phase Ib/II study of lenvatinib and pem-
brolizumab in advanced solid tumors), the ORR Efforts to examine genetic drivers of response and de-
among 108 patients with previously treated termine prognostic genomic predictors in advanced or re-
endometrial cancer was 38% at week 24 per current endometrial cancer have shown that 67% to 91% of
investigator review per immune-related patients have at least one therapeutically actionable ge-
RECIST. “The ORRs for participants with MMR nomic alteration for which therapy approved by the FDA or
proficient (94 patients) and dMMR (11 patients) therapy in clinical investigation exists; of patients who were
endometrial cancers were 36% and 64%, retreated, 47% to 62% derived clinical benefit.16,17
spectively,” ... and responses were seen regard-
less of MSI status, PD-L1 status, or histology.
ENDOMETRIAL CANCER IMMUNE LANDSCAPE
• In 2019, the FDA granted accelerated approval
for lenvatinib in combination with pem- Endometrial cancer cells and the tumor microenvironment
brolizumab for the treatment of advanced en- have been shown to modulate the immune response. First,
dometrial cancer that is not MSI-H or dMMR endometrial cancer cells possess the ability to activate PD-1
and has progressed following prior therapy. signaling, an immune checkpoint and potent mechanism
for downregulation of the immune response, by over-
expressing PD-L1 and PD-L2. PD-L1 and PD-L2 then bind
65% of endometrial cancers. These are generally hor- PD-1 receptors, expressed on tumor-infiltrating CD4 and
monally driven with estrogen and progesterone receptors CD8 T cells, inactivating them in the tumor microenviron-
and are often preceded by precursor intraepithelial lesions. ment. Immunohistochemical studies have detailed PD-1
These tumors are more likely to present in early stages, and PD-L1 expression levels (40%–80% in endometrioid,
owing to the onset of symptoms (e.g., abnormal uterine 10%–68% in serous, and 23%–69% in clear cell subtypes,
bleeding or postmenopausal bleeding). They are associated respectively) in endometrial carcinomas, representing the
with PTEN mutations and are often dMMR.11 highest expression among gynecologic cancers.18-20 Sec-
ond, endometrial cancer subtypes with high tumor muta-
Type 2 endometrial cancer includes FIGO grade 3 endo-
tional burden (e.g., POLE mutant/hypermutated and MSI-H)
metrioid tumors and serous and clear cell histologies. These
are highly immunogenic and exhibit more tumor-specific
tumors are typically hormonally independent without ex-
neoantigens, resulting in increased CD3+ and CD8+ tumor-
pression of estrogen and progesterone receptors; they have
infiltrating lymphocytes and a compensatory upregulation
a more aggressive histologic pattern, typically presenting at
of immune checkpoints and subsequent cytotoxic response.21-26
a later stage, and are associated with overexpression of
The presence of increased tumor-infiltrating lympho-
HER2/neu and p53 mutations.12,13 Although these cate-
cytes, an indicator of an active immune response against
gorizations can aid in the identification of patients at risk for
cancer cells, has been associated with improved out-
developing endometrial cancer and provide prognostic in-
comes in endometrial cancer.27 This combination of in-
formation regarding the risk of recurrence or advanced
creased mutational load, tumor-infiltrating lymphocytes,
disease, they fail to aid in the understanding of novel treat-
and PD-1/PD-L1 expression makes endometrial cancer
ment options.
an ideal target for immunotherapeutic interventions.21
The Cancer Genome Atlas is a proteonomic and tran- When considering therapeutic targets, it is important to
scriptomic evaluation of the genomic and epigenomic note that endometrial cancer was recently shown to have
landscape of primary surgical endometrial cancer that the highest prevalence of MSI across 30 human cancer
delineates four subgroups, which more accurately reflects types,28 and approximately 30% of primary endometrial
the underlying tumor biology and clinical outcomes in en- cancers are MSI-H and 13% to 30% of recurrent en-
dometrial cancer. This endeavor offers new molecular dometrial cancers are MSI-H or dMMR.14,16,17,28,29

Green, Feinberg, and Makker
For these reasons, immunotherapy strategies utilizing both the hypothesis that dMMR tumors are sensitive to ICB,
monotherapy and ICB combination approaches, as well as independent of the primary tumor site. The results of the
ICB therapy in combination with other rationale targeted phase II KEYNOTE-158 trial across 27 advanced MSI-H or
therapies, have emerged as promising approaches to en- dMMR solid tumors, published in January 2020, were also
hance the antitumor immune response in advanced en- consistent with this finding.34 This study included 49 pa-
dometrial cancer. tients with endometrial cancer and had an overall response
IMMUNE CHECKPOINT BLOCKADE MONOTHERAPY rate of 57%, including 16% (eight patients) with a complete
response and 41% (20 patients) with a partial response.
Given the presence of immune dysregulation in endometrial
Fatigue, pruritus, and diarrhea were the most common
cancer as described, ICB has been explored as a thera-
treatment-related adverse events in the cohort of 233 pa-
peutic mechanism, both as monotherapy and in combi-
tients. A total of 22 patients (9%) required cessation of
nation with cytotoxic chemotherapy, other immunotherapy,
therapy as a result of adverse events. Three patients ex-
or targeted agents. Generally, tumors with a high rate of
perienced grade 4 toxicity (Guillain-Barré syndrome, ele-
mutation or mutation burden are less subject to immune
vated transaminases, and enterocolitis with neutropenia).
tolerance and are thus more likely to be recognized and
targeted by T cells induced by ICB. Cancer mutations Preliminary results with dostarlimab, a PD-1 inhibitor in-
generate neoantigens, or protein coding sequences, that are vestigated in a phase I/II study, showed an ORR of 49% in
presented as peptide antigens by the major histocom- MSI-H advanced endometrial cancers versus an ORR of
patibility complex and recognized as foreign by T cells for 20% in MSS tumors; in addition, the median duration of
subsequent immunoediting and, ideally, elimination.30 response was not reached, with approximately 50% of all
Tumors, including endometrial cancer, often evade this responders receiving treatment for less than 1 year. Dos-
immune attack by expressing inhibitory checkpoint mol- tarlimab was well tolerated, with less than 6% of patients
ecules, such as PD-1 and its ligand PD-L1, to garner self- experiencing grade 3 or higher immune-related adverse
tolerance.31 events.35 Dostarlimab is now under investigation in a phase
The molecular classifications described by The Cancer III trial in the first-line setting in combination with carboplatin
Genome Atlas are increasingly being used in clinical re- and paclitaxel chemotherapy (RUBY; NCT03981796).
search to identify patients with endometrial cancer who may Nivolumab monotherapy was associated with an ORR of
have a more robust response to immunotherapy. Given that 23% in a phase II trial in advanced endometrial cancer (23
MSI status is a marker for response to immune checkpoint patients) regardless of MSI status.36
inhibition, likely attributable to its high mutational burden The PD-L1 inhibitors avelumab and durvalumab have shown
and generation of neoantigens, pembrolizumab was ap- ORRs of 26.7% and 43% in advanced endometrial cancer
proved in 2017 for the treatment of MSI-H or dMMR solid dMMR tumors and 6.25% and 3% in MMR retained tumors,
tumors. This was the first site-agnostic cancer drug ap- respectively, as monotherapy.37,38
proved by the FDA.
Clinical efficacy observed among patients with endometrial
Prior to this approval, the first evidence for clinical activity of cancer with PD-L1–positive tumors, regardless of MSI
immunotherapy in advanced endometrial cancer derived status, treated with ICB is generally modest. In a phase Ib
from a phase II study of 41 patients with heavily pretreated (KEYNOTE-028) study of pembrolizumab in advanced PD-
metastatic carcinoma with or without dMMR and also in- L1–positive advanced solid tumors, the 24 patients with
cluded two patients with endometrial carcinomas.32 All recurrent metastatic endometrial cancer had an ORR of
patients were treated with anti–PD-1 monoclonal antibody 13% (three patients with partial response) and an addi-
pembrolizumab. Both patients with endometrial cancer tional 13% (three patients) achieved stable disease.39 Of
achieved a partial response. This pivotal study reported for the three patients achieving a partial response, one had
the first time a connection between the tumor microenvi- non–MSI-H endometrial cancer with a POLE mutation, one
ronment, genotype, and response to checkpoint inhibitors had non–MSI-H endometrial cancer, and the other had
and represented a notable step forward in the identification an unknown status. Thirteen patients (54%) experienced
of predictors of response to ICB therapy. treatment-related adverse events, the most common
The clinical efficacy of pembrolizumab monotherapy was being fatigue, pruritus, and pyrexia. Four patients had
further investigated in a phase II trial published in 2017 grade 3 toxicity, although no patients had grade 4 toxicity
across 12 dMMR tumor types.33 An objective response rate or immune-related adverse events of any grade. Simi-
(ORR) of 53% (46 patients), including 21% with a complete larly, atezolizumab (anti–PD-L1) and nivolumab (anti–
response (18 patients), was observed among the 86 pa- PD-1) were examined as monotherapy in PD-L1–positive
tients enrolled. The cohort with endometrial cancer dem- endometrial cancer, with ORRs of 13% and 23%,
onstrated a similar ORR of 53%. This study further supported respectively.36,40

IMMUNE CHECKPOINT BLOCKADE COMBINATION THERAPY an enzyme that causes breakdown of tryptophan in the
Immune Checkpoint Blockade and Chemotherapy tumor microenvironment and induces immunosuppressive
activity; early clinical studies suggest that its blockade can
Preclinical studies suggest that cytotoxic chemotherapy improve the efficacy of immunotherapy agents.50
may result in robust immune stimulation,41 enhance pre-
sentation of tumor cell–specific antigens,42 and induce PD- Immune Checkpoint Blockade and Targeted Agents
L1 expression on cancer cells.43,44 By disrupting the tumor The FDA recently granted accelerated approval for the
stroma, chemotherapy also facilitates the penetration of combination of the oral multikinase inhibitor lenvatinib
cytotoxic T lymphocytes into tumor tissue and increases (VEGFR1-3, FGFR 1-4, KIT, RET, PDGFRa) with pem-
susceptibility to their cytotoxic effect via granzyme B– brolizumab for the treatment of advanced endometrial
mediated killing.45 The result of these mechanisms is a fa- cancer that is not MSI-H or dMMR and has progressed
vorable milieu for immune activation within the tumor that following prior therapy.51 In preclinical models, lenvatinib
has formed the biologic basis for the design of later-phase decreases the tumor-associated macrophage population
clinical studies examining the combination of cytotoxic and increases proportions of CD8+ T cells, inducing immune
chemotherapy agents and immunotherapy. Clinical trials in activation. Thus, coinhibition of both VEGF by lenvatinib and
endometrial cancer emulate prior work studying combi- PD-1 by pembrolizumab is postulated as an efficacious
nation therapy in other primary cancer sites. To date, antitumor strategy.52 In multiple mouse xenograft models,
a phase III trial of first-line dostarlimab versus placebo in the combination of anti–PD-1/PD-L1 monoclonal antibodies
combination with carboplatin and paclitaxel chemotherapy with lenvatinib had superior antitumor activity compared
(RUBY; NCT03981796) is ongoing among patients with with monotherapy with either agent alone.53,54
primary advanced or recurrent endometrial cancer. Simi-
larly, atezolizumab (AtTEnd; NCT03603184) and pem- KEYNOTE-146 (NCT02501096) was designed to examine
brolizumab (GY018’ NCT02549209) are also being studied the safety and preliminary efficacy of lenvatinib plus
in phase III trials in combination with carboplatin and pembrolizumab in multiple advanced solid tumors (non–
paclitaxel for patients with advanced or recurrent endo- small cell lung cancer, renal cell carcinoma, melanoma,
metrial cancer, with the hope to improve on the previous urothelial cancer, endometrial cancer, and squamous cell
results observed with monotherapy alone. Although these carcinoma of the head and neck). The phase Ib portion of
studies are enrolling patients regardless of MMR status, the this study defined a maximum tolerated dose and recom-
differential outcomes will be evaluated in subgroup analyses mended phase II dose of 20 mg of lenvatinib orally once
of patients with MSI-H and MSS tumors. daily in combination with 200 mg of pembrolizumab given
intravenously once every 3 weeks.55 The promising anti-
Combination Immunotherapy tumor activity observed and an acceptable safety profile led
There are ongoing efforts to study the efficacy of combi- to expansion cohorts for studying this combination in renal
nation immunotherapy with other immunotherapy drugs cell carcinoma and endometrial cancer.
with different mechanisms to overcome key resistance In the phase II study published in 2019, the efficacy of
mechanisms observed with monotherapy. Tumor-specific lenvatinib plus pembrolizumab for patients with primary
and intratumoral regulatory T cells may express multiple advanced or recurrent endometrial cancer, independent of
inhibitory receptors, and efforts have been aimed at tar- MMR status, was further investigated.56 In the final efficacy
geting multiple inhibitory receptors to reverse the sub- analysis, the ORR among 108 previously treated patients
sequent exhaustion and unresponsiveness of the immune was 38% (41 patients) at week 24 per investigator review
system.46-48 The interim results of an ongoing phase II study per immune-related RECIST, with a median progression-
of durvalumab (anti–PD-L1) and tremelimumab (anti– free survival of 7.5 months. 57 “ORRs for participants with
CTLA-4) versus durvalumab alone in recurrent endometrial MMR proficient (94 patients) and dMMR (11 patients)
cancer previously treated with platinum-based therapy were endometrial cancer were 36% and 64%, respectively,”... and
reported for 56 enrolled patients.49 Observed ORRs were responses were seen regardless of MSI status, PD-L1 status,
modest for both arms (14.8% of patients treated with or histology. Treatment-emergent adverse events were
durvalumab monotherapy and 11.1% of patients treated common, resulting in dose reductions or dose interruptions
with combination durvalumab with tremelimumab). Addi- for 70 patients (65%) and 78 patients (72%), respectively.
tional early-phase trials are evaluating the combination of The most common any-grade adverse events were hyper-
nivolumab with ipilimumab (anti–CTLA-4; NCT03508570 tension, diarrhea, decreased appetite, fatigue, and hypo-
and NCT02982486) therapy in advanced endometrial thyroidism. Grade 3/4 toxicities occurred for 75 patients
cancer. Nivolumab is also under study in combination with (69%), and two deaths were deemed treatment related per
or without an indoleamine 2,3-dioxygenase inhibitor (BMS- the investigators (intracranial hemorrhage and Escherichia
986205; NCT04106414). Indoleamine 2,3-dioxygenase is coli sepsis). The ORR observed in this study was superior to

TABLE 1. Select Ongoing Immune Checkpoint Inhibitor Combination Studies

ClinicalTrials.gov
Identifier Phase Trial Description
NCT03517449 III KEYNOTE-775: pembrolizumab and lenvatinib vs. liposomal doxorubicin or paclitaxel
NCT03884101 III LEAP-001: pembrolizumab and lenvatinib vs. carboplatin and paclitaxel
NCT02549209 III NRG-GY018: carboplatin and paclitaxel plus pembrolizumab vs. placebo
NCT03981796 III RUBY: carboplatin and paclitaxel plus dostarlimab vs. placebo
NCT03603184 III AtTEND: carboplatin and paclitaxel plus atezolizumab vs. placebo
NCT03526432 II Atezolizumab plus bevacizumab in advanced endometrial cancers and carcinosarcomas
NCT03951415 II Durvalumab plus olaparib in advanced endometrial cancers and carcinosarcomas
NCT03694262 II EndoBARR: atezolizumab, bevacizumab, and rucaparib in advanced endometrial cancers and carcinosarcomas
NCT03572478 I/II Rucaparib and nivolumab in advanced endometrial cancers
NCT02912572 II Avelumab in MSS, MSI-H, and POLE-mutated advanced endometrial cancer and of avelumab plus talazoparib in MSS
advanced endometrial cancer
NCT03835819 II Mirvetuximab soravtansine plus pembrolizumab in FRα-positive advanced MSS endometrial cancer
Abbreviations: MSS, microsatellite stable; MSI-H, microsatellite instability high/hypermutated; POLE, polymerase "; FR, folate receptor.
a prior phase II study of lenvatinib monotherapy alone for other immunotherapeutic agents, chemotherapy, and radiation
patients with advanced endometrial cancer, which showed therapy, which have the potential to alter the treatment land-
an ORR of only 14% (for 19 of 133 patients enrolled).58 scape, are currently ongoing. In addition, vaccine strategies and
The combination of lenvatinib and pembrolizumab is now adoptive T-cell approaches are generating exciting preliminary
under study in two ongoing phase III trials: lenvatinib with data and are areas of investigation that should be closely fol-
pembrolizumab versus doxorubicin or weekly paclitaxel in lowed. Key to advancing this field forward will be optimal patient
advanced endometrial cancer previously treated with selection based on tumor molecular and immunophenotypic
platinum-based therapy (NCT03517449) and first-line properties (tumor mutational burden, MMR status, PD-L1 status
lenvatinib with pembrolizumab versus carboplatin and [tumor proportion score, combined positive score], and tumor-
paclitaxel chemotherapy in advanced endometrial cancer infiltrating lymphocyte classification). In addition, efforts to better
(NCT03884101). Additional phase II studies are actively delineate mechanisms of resistance including specific somatic
enrolling to evaluate the synergistic potential of ICB with mutations (e.g., JAK1, B2M) as well as innate immune function
other targeted agents, PARP inhibitors (NCT03951415 and with variation in HLA alleles is warranted.32,59,60
NCT03572478), and the folate receptor antibody mirve-
Concerted efforts to better understand host factors, such as
tuximab soravtansine (NCT03835819; Table 1).
microbiome composition and the interplay of body mass
FUTURE DIRECTIONS index, are warranted, as is a more robust understanding of
In aggregate, ICB monotherapy has resulted in modest the molecular and immunologic drivers of response and
activity in advanced endometrial cancers. However, nu- resistance that are critical to the optimal design of next-
merous combination strategies with targeted therapies, generation studies in endometrial cancer.

1
Gynecologic Medical Oncology Service, Memorial Sloan Kettering Cancer Vicky Makker, MD, Gynecologic Medical Oncology Service, Memorial
Center and Weill Cornell Medical College, New York, NY Sloan Kettering Cancer Center and Weill Cornell Medical College, 1275
2
Gynecology Service, Department of Surgery, Memorial Sloan Kettering York Ave., New York, NY 10065; email: [email protected].
Cancer Center, New York, NY
EDBK_280503.

REFERENCES
2. Torre LA, Bray F, Siegel RL, et al. Global cancer statistics, 2012. CA Cancer J Clin. 2015;65:87-108.
4. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines: Uterine Neoplasms, Version 2.2017. Plymouth Meeting, PA: National Com-
prehensive Cancer Network; 2017.
5. Colombo N, Creutzberg C, Amant F, et al; ESMO-ESGO-ESTRO Endometrial Consensus Conference Working Group. ESMO-ESGO-ESTRO Consensus Conference
on Endometrial Cancer: diagnosis, treatment and follow-up. Ann Oncol. 2016;27:16-41.
6. Morice P, Leary A, Creutzberg C, et al. Endometrial cancer. Lancet. 2016;387:1094-1108.
7. Calle EE, Rodriguez C, Walker-Thurmond K, et al. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med.
2003;348:1625-1638.
8. Gunter MJ, Hoover DR, Yu H, et al. Insulin, insulin-like growth factor-I, and risk of breast cancer in postmenopausal women. J Natl Cancer Inst. 2009;101:48-60.
9. Argolo DF, Hudis CA, Iyengar NM. The impact of obesity on breast cancer. Curr Oncol Rep. 2018;20:47.
10. Bokhman JV. Two pathogenetic types of endometrial carcinoma. Gynecol Oncol. 1983;15:10-17.
11. Lax SF, Pizer ES, Ronnett BM, et al. Comparison of estrogen and progesterone receptor, Ki-67, and p53 immunoreactivity in uterine endometrioid carcinoma and
endometrioid carcinoma with squamous, mucinous, secretory, and ciliated cell differentiation. Hum Pathol. 1998;29:924-931.
12. Voss MA, Ganesan R, Ludeman L, et al. Should grade 3 endometrioid endometrial carcinoma be considered a type 2 cancer-a clinical and pathological
evaluation. Gynecol Oncol. 2012;124:15-20.
13. Llobet D, Pallares J, Yeramian A, et al. Molecular pathology of endometrial carcinoma: practical aspects from the diagnostic and therapeutic viewpoints. J Clin
Pathol. 2009;62:777-785.
14. Kandoth C, Schultz N, Cherniack AD, et al; Cancer Genome Atlas Research Network. Integrated genomic characterization of endometrial carcinoma [published
correction appears in Nature. 2013;500:242]. Nature. 2013;497:67-73.
15. Meyer LA, Broaddus RR, Lu KH. Endometrial cancer and Lynch syndrome: clinical and pathologic considerations. Cancer Contr. 2009;16:14-22.
16. Prendergast EN, Holman LL, Liu AY, et al. Comprehensive genomic profiling of recurrent endometrial cancer: implications for selection of systemic therapy.
Gynecol Oncol. 2019;154:461-466.
17. Soumerai TE, Donoghue MTA, Bandlamudi C, et al. Clinical utility of prospective molecular characterization in advanced endometrial cancer. Clin Cancer Res.
2018;24:5939-5947.
18. Vanderstraeten A, Tuyaerts S, Amant F. The immune system in the normal endometrium and implications for endometrial cancer development. J Reprod
Immunol. 2015;109:7-16.
19. Herzog TJ, Arguello D, Reddy SK, et al. PD-1, PD-L1 expression in 1599 gynecological cancers: implications for immunotherapy. Gynecol Oncol. 2015;137
(suppl 1):204-205.
20. Brahmer JR, Tykodi SS, Chow LQ, et al. Safety and activity of anti-PD-L1 antibody in patients with advanced cancer. N Engl J Med. 2012;366:2455-2465.
21. Howitt BE, Shukla SA, Sholl LM, et al. Association of polymerase e-mutated and microsatellite-instable endometrial cancers with neoantigen load, number of
tumor-infiltrating lymphocytes, and expression of PD-1 and PD-L1. JAMA Oncol. 2015;1:1319-1323.
22. Gargiulo P, Della Pepa C, Berardi S, et al. Tumor genotype and immune microenvironment in POLE-ultramutated and MSI-hypermutated endometrial cancers:
new candidates for checkpoint blockade immunotherapy? Cancer Treat Rev. 2016;48:61-68.
23. Nelson BH, McAlpine JN. The more tumors change, the more they stay tame: do T cells keep POLE ultramutated endometrial carcinomas in check? Gynecol
Oncol. 2015;138:1-2.
24. Bellone S, Centritto F, Black J, et al. Polymerase " (POLE) ultra-mutated tumors induce robust tumor-specific CD4+ T cell responses in endometrial cancer
patients. Gynecol Oncol. 2015;138:11-17.
25. van Gool IC, Eggink FA, Freeman-Mills L, et al. POLE proofreading mutations elicit an antitumor immune response in endometrial cancer. Clin Cancer Res. 2015;
21:3347-3355.
26. Eggink FA, Van Gool IC, Leary A, et al. Immunological profiling of molecularly classified high-risk endometrial cancers identifies POLE-mutant and microsatellite
unstable carcinomas as candidates for checkpoint inhibition. OncoImmunology. 2017;6:e1264565.
27. de Jong RA, Leffers N, Boezen HM, et al. Presence of tumor-infiltrating lymphocytes is an independent prognostic factor in type I and II endometrial cancer.
Gynecol Oncol. 2009;114:105-110.
28. Bonneville R, Krook MA, Kautto EA, et al. Landscape of microsatellite instability across 39 cancer types. JCO Precis Oncol. 2017;1:1-15.
29. Kloor M, von Knebel Doeberitz M. The immune biology of microsatellite-unstable cancer. Trends Cancer. 2016;2:121-133.
30. Schreiber RD, Old LJ, Smyth MJ. Cancer immunoediting: integrating immunity’s roles in cancer suppression and promotion. Science. 2011;331:1565-1570.

34. Marabelle A, Le DT, Ascierto PA, et al. Efficacy of pembrolizumab in patients with noncolorectal high microsatellite instability/mismatch repair-deficient cancer:
results from the phase II KEYNOTE-158 study. J Clin Oncol. 2020;38:1-10.
35. Oaknin A, Ellard SL, Leath C, et al. Preliminary safety, efficacy, and PK/PD characterization from GARNET, a phase 1 clinical trial of the anti-PD-1 monoclonal
antibody, TSR-042, in patients with recurrent or advanced MSI-H endometrial cancer. Ann Oncol. 2018;29 (suppl 8):viii334.
36. Hasegawa K, Tamura K, Katsumata N, et al. Efficacy and safety of nivolumab (Nivo) in patients (pts) with advanced or recurrent uterine cervical or corpus
cancers. J Clin Oncol. 2018;36:15s (suppl; abstr 5594).
37. Konstantinopoulos PA, Luo W, Liu JF, et al. Phase II study of avelumab in patients with mismatch repair deficient and mismatch repair proficient recurrent/
persistent endometrial cancer. J Clin Oncol. 2019;37:2786-2794.
38. Antill YC, Kok PS, Robledo K, et al. Activity of durvalumab in advanced endometrial cancer (AEC) according to mismatch repair (MMR) status: the phase II
PHAEDRA trial (ANZGOG1601). J Clin Oncol. 2019;37:15s (suppl; abstr 5501).
39. Ott PA, Bang YJ, Berton-Rigaud D, et al. Safety and antitumor activity of pembrolizumab in advanced programmed death ligand 1-positive endometrial cancer:
results from the KEYNOTE-028 study. J Clin Oncol. 2017;35:2535-2541.
40. Fleming GF, Emens LA, Eder JP, et al. Clinical activity, safety and biomarker results from a phase Ia study of atezolizumab (atezo) in advanced/recurrent
endometrial cancer (rEC). J Clin Oncol. 2017;35:15s (suppl; abstr 5585).
41. Zitvogel L, Kepp O, Kroemer G. Immune parameters affecting the efficacy of chemotherapeutic regimens. Nat Rev Clin Oncol. 2011;8:151-160.
42. Liu WM, Fowler DW, Smith P, et al. Pre-treatment with chemotherapy can enhance the antigenicity and immunogenicity of tumours by promoting adaptive
immune responses. Br J Cancer. 2010;102:115-123.
43. Samanta D, Park Y, Ni X, et al. Chemotherapy induces enrichment of CD47+/CD73+/PDL1+ immune evasive triple-negative breast cancer cells. Proc Natl Acad Sci
USA. 2018;115:E1239-E1248.
44. Peng J, Hamanishi J, Matsumura N, et al. Chemotherapy induces programmed cell death-ligand 1 overexpression via the nuclear factor-κB to foster an
immunosuppressive tumor microenvironment in ovarian cancer. Cancer Res. 2015;75:5034-5045.
45. Ramakrishnan R, Assudani D, Nagaraj S, et al. Chemotherapy enhances tumor cell susceptibility to CTL-mediated killing during cancer immunotherapy in mice.
J Clin Invest. 2010;120:1111-1124.
46. Baitsch L, Legat A, Barba L, et al. Extended co-expression of inhibitory receptors by human CD8 T-cells depending on differentiation, antigen-specificity and
anatomical localization. PLoS One. 2012;7:e30852.
47. Matsuzaki J, Gnjatic S, Mhawech-Fauceglia P, et al. Tumor-infiltrating NY-ESO-1-specific CD8+ T cells are negatively regulated by LAG-3 and PD-1 in human
ovarian cancer. Proc Natl Acad Sci USA. 2010;107:7875-7880.
48. Turnis ME, Andrews LP, Vignali DA. Inhibitory receptors as targets for cancer immunotherapy. Eur J Immunol. 2015;45:1892-1905.
49. Rubinstein MM, Caird I, Zhou Q, et al. A phase II trial of durvalumab with or without tremelimumab in patients with persistent or recurrent endometrial carcinoma
and endometrial carcinosarcoma. J Clin Oncol. 2019;37:15s (suppl; abstr 5582).
50. Soliman H, Mediavilla-Varela M, Antonia S. Indoleamine 2,3-dioxygenase: is it an immune suppressor? Cancer J. 2010;16:354-359.
51. U.S. Food and Drug Administration. Simultaneous review decisions for pembrolizumab plus lenvatinib in Australia, Canada and US. https://www.fda.gov/drugs/
resources-information-approved-drugs/simultaneous-review-decisions-pembrolizumab-plus-lenvatinib-australia-canada-and-us. Accessed March 18, 2020.
52. Mittica G, Ghisoni E, Giannone G, et al. Checkpoint inhibitors in endometrial cancer: preclinical rationale and clinical activity. Oncotarget. 2017;8:90532-90544.
53. Kato Y, Tabata K, Hori Y, et al. Effects of lenvatinib on tumor-associated macrophages enhance antitumor activity of PD-1 signal inhibitors. Mol Cancer Ther.
2015;14 (suppl 2; abstr A92).
54. Kato Y, Bao X, Macgrath S, et al. Lenvatinib mesilate (LEN) enhanced antitumor activity of a PD-1 blockade agent by potentiating Th1 immune response. Ann
Oncol. 2016;27 (suppl 6; abstr 2PD).
55. Taylor M, Dutcus CE, Schmidt E, et al. A phase 1b trial of lenvatinib (LEN) plus pembrolizumab (PEM) in patients with selected solid tumors. Ann Oncol. 2016;
25:776PD.
56. Makker V, Rasco D, Vogelzang NJ, et al. Lenvatinib plus pembrolizumab in patients with advanced endometrial cancer: an interim analysis of a multicentre, open-
label, single-arm, phase 2 trial. Lancet Oncol. 2019;20:711-718.
57. Makker V, Taylor MH, Aghajanian C, et al. Lenvatinib and pembrolizumab in patients with advanced endometrial cancer. Ann Oncol. 2019;30 (suppl 5; abstr
992O).
58. Vergote I, Teneriello M, Powell MA, et al. A phase II trial of lenvatinib in patients with advanced or recurrent endometrial cancer: angiopoietin-2 as a predictive
marker for clinical outcomes. J Clin Oncol. 2013;31:15s (suppl; abstr 5520).
59. Albacker LA, Wu J, Smith P, et al. Loss of function JAK1 mutations occur at high frequency in cancers with microsatellite instability and are suggestive of immune
evasion. PLoS One. 2017;12:e0176181.
60. Chowell D, Morris LGT, Grigg CM, et al. Patient HLA class I genotype influences cancer response to checkpoint blockade immunotherapy. Science. 2018;
359:582-587.

GYNECOLOGIC CANCER
Therapeutic Targets and Opportunities in

Endometrial Cancer: Update on Endocrine
Therapy and Nonimmunotherapy
Targeted Options
Helen J. MacKay, MBChB, BSc, MRCP, MD1; Victor Rodriguez Freixinos, MD, PhD1; and Gini F. Fleming, MD, FASCO2
overview
Worldwide, the incidence of endometrial cancer is increasing. Although the prognosis remains good for
patients diagnosed with early-stage disease, for those diagnosed with recurrent or metastatic disease, options
have been limited, and prognosis is short. Optimizing and identifying new well-tolerated treatments for women
living with endometrial cancer is a top priority. A new era is dawning where we are starting to see the integration
of clinically relevant genomic and pathologic data to inform and refine treatment strategies for women with
endometrial cancer. Here, we focus on reviewing nonimmunotherapy-based targeted treatment options and
emerging directions for women with endometrial cancer.
INTRODUCTION toxicity profile and modes of administration, it remains

Until recently, women with recurrent or metastatic endo- an attractive therapeutic strategy. However, despite 6
metrial cancer had few therapeutic options.1 Choices were decades of clinical use, we do not know who truly
limited to chemotherapy, which has limited efficacy after benefits from this approach. We lack large randomized
first-line treatment, and single-agent endocrine therapy. studies, patient-reported outcome data, and validated
Our understanding of this disease is improving with mo- predictors of response.4
lecular classifications as initially identified through the Endocrine therapy was formerly the accepted frontline
Cancer Genome Analysis and later defined by pared-down therapy for advanced endometrial cancer.5 The stan-
classifiers (Fig. 1), and these are starting to inform our dard agents were progestins, and megestrol acetate is
treatment decisions. Molecular data from The Cancer FDA-approved for treatment of patients with recurrent
Genome Atlas (TCGA) and others provide new directions for endometrial cancer. A recent summary analysis of
study and, given prognostic implications, a new lens to progestins used as first-line therapy for metastatic/
consider how we interpret unselected population studies of recurrent endometrial cancer, largely unselected for
patients with endometrial cancer.2 Finally, and most im- predictive markers of response, found an overall re-
portantly, we are seeing increased recognition that “one size sponse rate of 23.3%, a median progression-free
does not fit all” for women with endometrial cancer. Beyond survival (PFS) of 2.9 months, and a median overall
molecular classifications, many women have specific survival (OS) of 9.2 months. These numbers compare
health needs related to age, obesity, diabetes, or other unfavorably with results seen in frontline trials of
comorbidities, which can affect tolerability of treatment.3 chemotherapy, and multiagent chemotherapy has
This needs to be borne in mind not only for clinical decision- generally become the first-line approach. Nonetheless,
Author affiliations
making but also when clinical trials are in development. a subset of endometrial cancers has long-term responses
and support
information (if Immunotherapy alone or in combination is an exciting to endocrine therapy, and in these cases endocrine
applicable) appear therapeutic direction in the treatment of endometrial therapy may be preferred over chemotherapy as initial
at the end of this
cancer and is covered elsewhere in this volume. This treatment.
article.
review focuses on nonimmunotherapy treatment op- Predictive factors for response to standard endocrine
Accepted on March
2, 2020 and tions and emerging directions. therapy Factors reported to be predictive of response
published at to endocrine therapy include low grade, endometrioid
ascopubs.org on
ENDOCRINE THERAPY
histology, and positive estrogen receptor (ER)/pro-
April 2, 2020: Single-Agent Endocrine Therapy
gesterone receptor (PR) status; representative exam-
10.1200/EDBK_ Endocrine therapy is the “original” targeted treatment ples of each are cited below.6 Most trials of endocrine
280495 for women with endometrial cancer. Given its favorable therapy in endometrial cancer have not selected

MacKay, Freixinos, and Fleming
receptors and thereby increases the efficacy of progestin

therapy. Response rates were 38%, 24%, and 22% for
grade 1, 2, and 3 disease, respectively.
• Endocrine therapy should be considered either
alone or in combination with an mTOR inhibitor ER/PR status The prognostic value of ER/PR is well estab-
for the treatment of estrogen receptor/pro- lished, with higher levels of ER and PR expression asso-
gesterone receptor–positive endometrioid en- ciated with longer OS, longer cancer-specific survival, and
dometrial cancer. longer PFS.8 The predictive value of ER/PR in endometrial
• Future clinical trials with endocrine therapy cancer is less clear and is confounded by a lack of stan-
should establish standardized measurement of dardization in tissue processing or established cutoffs
and cutoffs for endocrine receptors. and mixed reporting of ER/PR status in published studies.
• Future trials with endocrine therapy should However, it is possible to infer an association between ER/
establish whether receptor status of primary PR status and response rates to endocrine therapy. In
disease is sufficient for decision-making or a meta-analysis of studies reporting ER/PR status, endo-
whether biopsy of metastatic disease is needed. metrial cancer tumors negative for both ER and PR do not
• A subset of p53 mutant endometrial cancer respond to endocrine therapy.5 In a small phase II trial
(serous and some high grade endometrioid) are testing continuous tamoxifen with intermittent medrox-
HER2-amplified and may benefit from addition yprogesterone acetate, the overall response rate was 33%.9
of trastuzumab to frontline carboplatin/ In ER+ tumors the response rate was 47%, compared with
paclitaxel. 26% for ER tumors.10 A phase II trial of fulvestrant re-
• Although antiangiogenic therapies generally ported no responses in PR negative tumors, compared
tend to have some activity in endometrial with 20% for those staining positive.11
cancer, the addition of bevacizumab to frontline Histology and molecular subgroups Only a few trials of
therapy has not been shown to have major
endocrine therapy in endometrial cancer have broken down
benefit.
results by histology. In a small phase II study of the com-
• Emerging areas of therapeutic interest in the bination of everolimus plus letrozole, the strongest predictor
treatment of endometrial cancer lie in immu- of nonresponse was serous histology.12 Of the 23 patients
notherapy combinations, in DNA repair, and in
with tumors of endometrioid histology, 13 of 19 patients with
incorporating molecular subgroups into clinical
practice. positive PR staining responded, and 1 of 4 patients with
negative PR staining responded. Interestingly, trials of
single-agent mTOR inhibitors have reported roughly equal
activity in serous and nonserous endometrial cancer.13 ER
patients or have separately reported results on the basis of
and PR expression occurs across the molecular subgroups
these criteria. This has led to considerable uncertainty re-
garding selection of patients for endocrine therapy. of the TCGA. The lowest levels of expression are seen in the
serous and serous-like p53 subgroup: ER+ (67.4% of
Grade GOG 1537 evaluated the strategy of tamoxifen al- cases) and PR+ (44.7% of cases).2 However, it is not clear
ternating with megestrol acetate based on the hypothesis that ER+ serous tumors will respond to endocrine therapy.
that tamoxifen increases the expression of progesterone
Even in grade 1/2 ER/PR+ endometrioid endometrial
cancer, response rates for metastatic disease seem lower
than in metastatic low-grade ER/PR+ breast cancers. Hy-
potheses to explain this have included discordance between
ER/PR status of primary endometrial cancer and recurrent
disease and the importance of PR and ER receptor variants.
Receptor status of primary vs. recurrent disease Soliman
et al14 noted a high correlation between ER and PR status in
the primary tumor and in the recurrent tumor. However,
others have found higher discordance rates.15 For example,
Tangen et al16 noted that metastases from PR+ primary
endometrial cancer lost PR expression in 40% to 50% of
cases. Similarly, metastases from ER+ primary endometrial
cancer lost ER staining in all metastases over 50% of the
time. Metastases from ER/PR primary tumors rarely ac-
FIGURE 1. TCGA Genomic Subgroups quired receptor expression.This raises the question of the

Endocrine and Nonimmunotherapy Targets in Endometrial Cancer
optimal timing and selection of tissue for tumor testing. This majority was in the range of 10% or less.25,26 As with other
question remains unanswered. older endocrine trials, these did not select for histology,
PR isoforms It has been proposed that specific isoforms of grade, or ER/PR status.
PR, PR-A, and PR-B may be relevant, with PR-B contrib- In summary, grade 1/2 ER/PR+ endometrioid endometrial
uting to estrogen and progesterone-mediated endometrial cancers appear to have response rates in the range of at
proliferation and PR-A inhibiting it.17 PR-B may therefore least 30% for endocrine therapies (excluding single-agent
be the predominant isoform responsible for the tumor- aromatase inhibitors) used as frontline therapy, and this
suppressive action of progestins.4 Measurement of “acti- represents a reasonable treatment option for women whose
vated PR” rather than overall PR status may also be of tumors have these characteristics. Furthermore, endocrine
greater predictive value. therapy may be considered as a subsequent line of treat-
ESR1 mutations These activating mutations have been ment especially in those with low volume or indolent
identified in only 2.0% of primary endometrial cancers,18 disease.
with the highest incidence in endometrioid tumors (4.4%),19 Endocrine Therapy: New Directions
so they seem unlikely to play a major role in primary re-
sistance of metastatic endometrial cancer to endocrine PI3K/Akt pathway inhibitors ER alpha signaling is mediated
therapy. However, it is possible that the incidence is in- through the mitogen-activated protein kinase (MAPK)
creased in recurrent or endocrine-resistant tumors. pathway, which further activates downstream molecules
ERK and akt murine thymoma viral oncogene (AKT; Fig.
A greater understanding of the ER/PR receptors and their 2).27 PR signaling also partially acts through the MAPK and
associated pathways in endometrial cancer may lead to PI3K/AKT/mTOR pathways (Fig. 2).28 Combining endocrine
predictive biomarker or biomarker signature development. therapy with downstream blockade has the potential to be
Endocrine therapy for fertility sparing in early-stage endo- synergistic. Based on improved PFS seen in breast cancer
metrial cancer Progestin therapy is used in premenopausal studies for the combination of an mTOR inhibitor and en-
women with atypical endometrial hyperplasia or early-stage, docrine therapy and preclinical studies in endometrial
low-grade endometrial cancer who desire to bear children, cancer,29,30 suggesting inhibition of the PI3K/Akt pathway
and meta-analyses suggest complete remission rates higher reversed progestin resistance in preclinical endometrial
than 70%.20,21 However, relapse rates are also high (30% to cancer models, and this approach has been pursued in
40%), and subsequent hysterectomy is often needed. Trials clinical trials (Fig. 2).31
of progestin intrauterine devices in this setting are ongoing. Temsirolimus with alternating megestrol acetate and ta-
These could mitigate some of the weight gain associated moxifen was investigated in advanced recurrent endometrial
with systemic progestins.4,22 cancer. However, the study was halted after 21 patients were
Alternate endocrine therapies The National Comprehensive treated because of excess venous thrombotic events
Cancer Network (NCCN) guidelines list a number of dif- (33%).13 The combination of everolimus plus letrozole was
ferent classes of agents as options for the treatment of explored in a small group of women with endometrial cancer
patients with endometrial cancer. with a response rate of 32%. A subsequent single-arm,
phase II trial (54 patients), limited enrollment to women with
A recent review including 16 heterogeneous studies on endometrioid endometrial cancer and no more than two
advanced and recurrent endometrial cancer found the prior chemotherapy regimens, added metformin to the
response rates for tamoxifen to range from 10% to 53%, for combination. Activity of the everolimus/letrozole combina-
other selective ER modulators and selective estrogen re- tion was confirmed, although metformin did not appear to
ceptor downregulators from 9% to 31%, for aromatase impprove outcomes; the response rate was 28%. The re-
inhibitors from 8% to 9%, and for combined tamoxifen/ sponse rate for PR– tumors was 9% vs. 45% for PR+ tumors
progestins from 19% to 58%. Only six of the trials in- (p = .06). Median OS was 19.6 months.14 The same in-
vestigated benefit separately for ER+ and ER tumors. vestigators conducted a randomized phase II trial comparing
Response rates in ER+ tumors ranged from 10% to 47%; everolimus plus letrozole to tamoxifen with intermittent
response rates in ER tumors was 0 in all but one study. medroxyprogesterone (GOG 3007). There was no se-
Tumors of endometrioid histology had higher response lection based on ER/PR status or histotype. Preliminary
rates.23 results showed response rates of 24% for everolimus/
Luteinizing hormone-releasing hormone receptors mediate letrozole and 22% for tamoxifen/medroxyprogesterone
antiproliferative activity in endometrial cell lines, and ap- acetate; for patients with no prior chemotherapy, these
proximately 80% of endometrial cancers express luteinizing were values 53% and 43%, respectively.32 Everolimus/
hormone-releasing hormone receptors. Response rates letrozole is suggested as an option for women with meta-
have varied from 0% to 45% in small older studies,24 but the static endometrioid within the NCCN guidelines.

Other steroid receptors Steroid receptors do not work in

isolation. Androgen receptors (ARs) and glucocorticoid
receptors (GR) are members of the same nuclear receptor
superfamily as ER/PR and may interact with ER/PR to affect
outcomes.37 Although agents targeting steroid receptors
other than ER/PR are not currently used in endometrial
cancer, they may play a role in future treatments. AR ex-
pression is generally a good prognostic factor in endometrial
cancer, but endometrial cancer–expressing AR with low ER
expression appears to have a worse prognosis.4,38 Pre-
clinical models showed growth inhibition by the anti-
androgen enzalutamide.16 It has been proposed that ER+/
AR+ tumors might benefit from the use of androgens,
whereas ER /AR+ tumors might benefit from antiandrogen
therapy, but such a strategy would require knowledge of ER/
AR status of the metastatic lesions. There is an ongoing
phase II study (NCT02684227) of the AR enzalutamide in
combination with chemotherapy in advanced endometrioid
endometrial cancer.39
GR expression is substantially associated with poorer
prognosis and higher grade. GR has been reported to be
expressed in 64% of nonendometrioid endometrial cancer
and was more common in metastatic lesions, with 56%
FIGURE 2. The Mechanistic Relationship Between Endocrine Signaling,
PI3K Pathway Activation, and Cell-Cycle in Tumor Cells patients with metastases from a GR negative primary tumors
40,41
Abbreviations: AKT, akt murine thymoma viral oncogene; CDK4/6, expressing GR. GR-specific antagonists are under de-
cyclin-dependent kinase 4/6; ER, estrogen receptor; mTOR, mammalian velopment and may be of use in the treatment of endo-
target of rapamycin; RTK, receptor tyrosine kinase. metrial cancer.
HER2–targeting agents In pretreated metastatic endome-
CDK4/6 Inhibitors Elevated cyclin-depending kinase 4 trial cancer, the HER2 amplification rate is reported to be
(CDK4) expression is seen in 35% to 77% of endometrioid 11.5%, with the highest rates in serous (25%) and clear cell
endometrial cancers.33 The combination of CDK4/6 in- carcinomas (38%).42 Subsequently, TCGA data showed
hibitors with endocrine therapy has become standard HER2 to be focally amplified, with protein overexpression in
frontline therapy in metastatic breast cancer.34 A random- 25% of the serous or serous-like tumors (p53).2 HER2
ized phase II trial of palbociclib versus placebo in combi- overexpression has been associated with a worse overall
nation with letrozole for patients with ER+ advanced or prognosis among patients with serous endometrial cancer.43
recurrent EC (NCT02730429) has completed accrual, and A trial of single-agent trastuzumab in advanced/recurrent
results are awaited. A single-arm trial of ribociclib (400 mg HER2+ endometrial cancer yielded no responses, with
daily) plus letrozole included 20 women with ER+ endo- a median PFS of only 1.8 months.42 Similarly, a phase II
metrial cancer with a median PFS of 5.3 months.35 basket study enrolled seven patients with HER2 amplified/
overexpressing endometrial cancer to the combination of
Epigenetic modulators Silencing of ER and/or PR in en- pertuzumab plus trastuzumab, and no responses were
dometrial cancer via epigenetic modulation has been ex- seen.44 However, a randomized phase II trial evaluating the
tensively reported. Histone deacetylase inhibitors have been addition of trastuzumab to carboplatin/paclitaxel chemo-
shown to restore functional PR in endometrial cancer cell therapy in advanced-stage HER2+ serous endometrial cancer
lines.36 The window of opportunity (presurgical) study (61 patients; the study was stopped early because of poor
GY011 randomized women who had a planned hyster- accrual) reported an improvement in median PFS from 8.0 to
ectomy for early-stage endometrial cancer to medrox- 12.6 months (p = .005). The PFS benefit was particularly
yprogesterone alone or with the addition of the histone evident among the 41 patients with stage III or IV disease
deacetylase inhibitor entinostat. The primary objective was undergoing primary treatment who had an improvement in
to determine whether the addition of entinostat results in median PFS from 9.3 months to 17.9 months (p = .013).45
upregulation of activated PRs when compared with ad- The combination of carboplatin, paclitaxel, and trastuzumab
ministration of medroxyprogesterone acetate alone. The is therefore the preferred chemotherapy regimen for women
trial has completed accrual, and results are awaited.4 with HER2-positive serous endometrial cancer.46

It has been suggested that phosphatase and tensin ho- pathways. The effects of inhibiting one part of the pathway
mologue (PTEN) loss and activating PIK3CA mutations are may be bypassed by other signaling branches downstream
associated with lower rates of response to anti-HER2 from PI3K (Fig. 2).59,60 Feedback loops have been noted to
therapy in breast cancer.47 If the high frequency of ab- be induced by mTOR inhibition, paradoxically resulting in
normalities in the PI3K/AKT/mTOR signaling pathway (Fig. upstream activation of AKT and therefore decreasing the
2) is responsible for impairing the activity of trastuzumab potential efficacy of mTOR inhibitors. It was hypothesized
against HER2-amplified endometrial cancer, then an that agents targeting AKT or PI3K earlier in the pathway
antibody-drug conjugate might be hypothesized to would yield increased activity. Studies are ongoing, al-
overcome the resistance. A trial of trastuzumab emtan- though toxicity has proven to be a problem for these
sine (TDM-1) for patients with HER2-amplified cancer agents, and their efficacy is limited.61,62 Combinations of
(NCT02675829) preliminarily reported responses in 4 agents targeting multiple related pathways are may be an
of 18 patients with endometrial cancer. effective therapeutic strategy. This has proven most ef-
Another explanation put forth for trastuzumab resistance fective to date, as discussed above, when combining
among HER2-amplified endometrial cancers is high expres- a rapalog with endocrine therapy. Where other combina-
sion of the constitutively active p95HER2 variant that lacks the tions have been considered, these have met with less
extracellular domain but preserves the intracellular tyrosine success, with toxicity being a problem. For example, the
kinase domain. One series reported 53% of serous/high-grade further development of the combination of an MEK in-
endometrioid cancers to have high p95HER2 expression.48 hibitor (trametinib) with an AKT inhibitor (uprosertib) was
This might support the use of anti-HER2 tyrosine kinase in- discontinued because of toxicity.63 Further combination
hibitors. A case report documenting a PR to the oral tyrosine studies are underway, including the combination of DNA
kinase inhibitors, afatinib, in HER2-amplified serous endo- repair agents, such as PARP inhibitors and AKT inhibitors,
metrial cancer has been published,49 and a phase II trial of in a phase I study, suggesting further exploration is
afatinib in this population is ongoing (NCT02491099). warranted.64
TARGETING THE PI3K/AKT PATHWAY TARGETING ANGIOGENESIS

Changes within the PI3K/AKT pathway (Fig. 2) are common Targeting angiogenesis in endometrial cancer, as in many
events, with somatic mutations occurring in approximately other tumor types, has been an active area of research for a
50% of endometrioid and 40% of serous endometrial number of years. Translational and preclinical data dem-
cancers,2,50,51 and are seen across TCGA subgroups.2 The first onstrated a strong rationale based on the observations that
class of agents to be investigated in patients with endometrial high microvessel density and proangiogenic gene expres-
cancer were the rapalog mTOR inhibitors. Studies using sion predicts for poor outcome across the histologic and
temsirolimus, everolimus, and ridaforolimus demonstrated molecular subgroups of endometrial cancer.65-67 Modest
modest response rates, with some patients experiencing efficacy in endometrial cancer has been reported across
prolonged stable disease.52-54 Response rates were highest in different classes of antiangiogenic agents, with response
patients with no or limited prior chemotherapy. The clinical rates around 10% to 15% and median PFS between 2 and
benefit rate ranged from 24% to 50% and median PFS of 2.8 5 months (Table 1). Single-agent bevacizumab is currently
to 7.3 months. Biomarkers of response to the rapalogs have included in NCCN recommendations for consideration
proven elusive, likely attributable to the complexity of the when disease has progressed after prior chemotherapy.
pathway, its many interactions, and incomplete inhibition of Bevacizumab has been combined with chemotherapy with
the mTOR complex by these agents.55,56 Temsirolimus was one study (86P), which reported no improvement in PFS
also explored in combination with carboplatin and paclitaxel as compared with a historical control of carboplatin and paclitaxel
part of a randomized clinical trial, with no improvement in alone. This study did report an improvement in OS. However,
outcome compared with historical control. In this study, TSC2 the data must be interpreted with caution given the lack of
was identified as a potential predictor of response, although improvement in PFS, the low response rates, an imbalance in
numbers were low.57 A Cochrane review concluded that, for histotypes, and the lack of a randomized control. Interestingly,
second- or third-line therapy, rapalogs may result in improved somatic CTNNB1 mutation, which occurs in 26% of patients
PFS (HR, 0.53; 95% CI, 0.31–0.91), although there was low with endometrial cancer (most commonly in the endometrioid
certainty of benefit because of the low number of studies.58 histotype), was associated with a better outcome for women
Temsirolimus is considered an option for the treatment of receiving bevacizumab (HR, 0.73; 95% CI, 0.60–0.91)
patients with endometrial cancer in the NCCN guidelines either compared with those whose tumors did not have a mutation.57
alone or in combination with endocrine therapy. A trend toward benefit for women whose tumor harbored
The PI3K/AKT/mTOR pathway is characterized by exten- a CTNNB1 mutation was as also observed in the recent
sive feedback loops and crosstalk with a number of single-agent cabozantanib trial.75 The results of a second,

TABLE 1. Single-Agent Antiangiogenic Trials in Endometrial Cancer cases.2 Preclinical endometrial cancer models have dem-
Agent Median Progression-Free Survival (months) onstrated activity for PARP inhibitors.85,86
Bevacizumab68 4.3 PTEN alterations play a role in the double-strand break
69
Aflibercept 3.3 repair system by regulating the expression of RAD51, a key
Brivanib70 3.5 protein in HR repair. Increased in vitro sensitivity to PARP
Cediranib71 3.6 inhibitors has been suggested in PTEN-null cell lines;
72
however, the relevance of PTEN loss is controversial, with
Nintedanib 3.3
other investigators not finding it to be relevant.86-89
Trebanib73 2.0
74
The MSI high-molecular subgroup of endometrial cancer
Sunitinib 3.0
demonstrates defects in the mismatch repair proteins,
Cabozantinib75 4.8 resulting in high mutational burden.90 Preclinical evidence
suggests that MMR deficiency, particularly loss of MSH2 and
MLH1, may be synthetically lethal, with inhibition of select
randomized, phase II study from demonstrated an improve- DNA polymerases (DNA polymerase b and DNA polymerase
ment in PFS (8.7 vs. 13 months).76 g respectively).90 In addition, MSI tumors may harbor sec-
Combining antiangiogenic agents with other targeted agents is ondary mutations in other DNA repair genes involved in HR,
an active area of research. Targeting angiogenesis with im- such as MRE11A and RAD50, that can lead to PARP in-
munotherapy has a strong rationale and has achieved early hibition sensitivity in vitro.91
success. The combination of lenvatanib and pembrolizumab ARID1A, a suppressor gene component of the chromatin-
recently achieved early approval after demonstrating activity remodeling complex, is recruited to DNA breakage sites
across endometrial cancer subgroups.77 A second area of through interaction with ATR and is required for normal G2/M
interest lies in combination with agents that target DNA repair. checkpoint inhibition. ARID1A defects caused replication
Downregulation of homologous recombination (HR) genes has stress and dependency on ATR activity.2,92 ATR inhibition in
been observed in hypoxic conditions (contextual synthetic endometrial cancer with ARID1A loss of function (LOF) has
lethality) with enhancement of PARP inhibitor sensitivity.78,79 the potential to induce synthetic lethality.93 Preclinical studies
NRG GY012, a randomized trial investigating the combina- of tumors bearing ARID1A mutations confirm sensitivity to
tion of cedarinib and olaparib, recently completed accrual several different types of ATR inhibitors.60 In addition, en-
(NCT03660826). However, as with all therapeutic strategies in dometrial cancer cell line data suggest that ARID1A LOF may
this patient population, vigilance regarding potential toxicities is increase reliance on PARP-dependent DNA repair. Several
essential. One example of this was a trial investigating tem- investigators have demonstrated that ARID1A LOF confers
sirolimus with bevacizumab, which demonstrated a response sensitivity to PARP inhibition.94 ATR inhibition is also po-
rate of 24.5% and a median PFS of 5.6 months. However, 38. tentially lethal, with other cancer-associated changes such as
8% of patients discontinued treatment because of toxicity, with oncogenic stress (MYC and CCNE1 overexpression) and
three treatment-related deaths halting further development.80 deficiencies in other DNA repair proteins.95-98 ATR inhibition
TARGETING DNA REPAIR has demonstrated activity in endometrial cancer in cell
lines.99 Targeting ATR in endometrial cancer alone or in
The TCGA and other researchers have recently identified combination with PARP inhibition is of therapeutic interest.
genomic events that suggest targeting DNA repair may be
a promising strategy in treating patients with endometrial OTHER POTENTIAL TARGETS
cancer.2 Parallels between the serous-like TP53-mutated WNT signaling functions predominantly through both
molecular subgroup and high-grade serous ovarian cancer CTNNB1-dependent and CTNNB1-independent pathways,
or basal-like breast cancers have been described, including often termed canonical and noncanonical WNT signaling,
HR deficiency.2,81,82 Germline and somatic alterations in respectively.100-102 Under normal conditions, CTNNB1 is
BRCA and other non-BRCA HR-related genes have been phosphorylated by WNT pathway members and targeted for
described in patients with endometrial cancer (e.g., ATM, proteasomal degradation through ubiquitination, leading to
BARD1, BRIP1, CHEK2, NBN, and RAD51C).83,84 Con- active repression of CTNNB1 target genes. Activating mu-
sideration of DNA repair mechanisms has also identified tations in CTNNB1 lead to accumulation of CTNNB1 in the
new potential targets. ARID1A mutations occurred in ap- cytoplasm, followed by translocation into the nucleus and
proximately 40%, microsatellite instable (MSI) in 11%, and CTNNB1-mediated transcription. Dickkopf-1 (DKK1) and
PTEN loss in 55% of patients with recurrent metastatic other members of the DKK family modulate the WNT
endometrioid endometrial cancer. ARID1A mutations are pathway. CTNNB1 mutations are common in endometrial
also observed in serous-like endometrial cancer, as are high cancer, and initial data suggest the potential to target DKK1
frequencies of nonsilent TP53 mutations in up to 91% of in endometrial cancer.2 DKN-01 (DKK1 antibody) is currently

being tested in a clinical trial (NCT03395080). Folate receptor a new era with treatment defined by molecular characteristics
alpha (FRα) has been explored as a therapeutic target in of the patient’s tumor. The optimal sequence and agents
ovarian cancer. Studies indicate that 41% of serous cancers have not been defined. In addition to new discoveries,
overexpress FRα, with further expression seen in endometrioid revisiting endocrine therapy, defining its role, identifying
endometrial cancer.2 Xenograft model studies supported the predictive biomarkers, and considering new combinations are
development of a clinical trial for patients with endometrial warranted. As we move forward, it is essential to include all
cancer103 with the antibody-drug conjugate mirvetuximab women living with endometrial cancer and to consider trial
soravtansine (NCT03836157).
designs (and recruitment strategies) that consider the older
CONCLUSION patient population and are inclusive of all groups. Patient-
For decades there were no new treatment options available reported outcomes and quality-of-life measures, which are
for women with endometrial cancer. We are moving into lacking in earlier endometrial cancer trials, are essential.

1
Division of Medical Oncology & Hematology, Sunnybrook Odette Cancer Helen J. MacKay, MBChB, BSc, MRCP, MD,Division of Medical Oncology
Centre, Toronto, Ontario, Canada & Hematology, Sunnybrook Odette Cancer Centre, 2075 Bayview Avenue,
2
Section of Hematology-Oncology, Department of Medicine, The Toronto, Ontario, M4N 3M5; email: [email protected].
University of Chicago, Chicago, IL
EDBK_280495.
REFERENCES
1. Lheureux S, McCourt C, Rimel BJ, et al. Moving forward with actionable therapeutic targets and opportunities in endometrial cancer: a NCI clinical trials
planning meeting report. Gynecol Oncol. 2018;8:84579-84594.
2. Kandoth C, Schultz N, Cherniack AD, et al; Cancer Genome Atlas Research Network. Integrated genomic characterization of endometrial carcinoma. Nature.
2013;497:67-73.
3. McAlpine JN, Temkin SM, Mackay HJ. Endometrial cancer: not your grandmother’s cancer. Cancer. 2016;122:2787-2798.
4. Jerzak KJ, Duska L, MacKay HJ. Endocrine therapy in endometrial cancer: an old dog with new tricks. Gynecol Oncol. 2019;153:175-183.
5. Ethier JL, Desautels DN, Amir E, et al. Is hormonal therapy effective in advanced endometrial cancer? A systematic review and meta-analysis. Gynecol Oncol.
2017;147:158-166.
6. Tangen IL, Werner HM, Berg A, et al. Loss of progesterone receptor links to high proliferation and increases from primary to metastatic endometrial cancer
lesions. Eur J Cancer. 2014;50:3003-3010.
7. Fiorica JV, Brunetto VL, Hanjani P, et al; Gynecologic Oncology Group study. Phase II trial of alternating courses of megestrol acetate and tamoxifen in advanced
endometrial carcinoma: a Gynecologic Oncology Group study. Gynecol Oncol. 2004;92:10-14.
8. Zhang Y, Zhao D, Gong C, et al. Prognostic role of hormone receptors in endometrial cancer: a systematic review and meta-analysis. World J Surg Oncol. 2015;
13:208.
9. Whitney CW, Brunetto VL, Zaino RJ, et al; Gynecologic Oncology Group study. Phase II study of medroxyprogesterone acetate plus tamoxifen in advanced
endometrial carcinoma: a Gynecologic Oncology Group study. Gynecol Oncol. 2004;92:4-9.
10. Singh M, Zaino RJ, Filiaci VJ, et al. Relationship of estrogen and progesterone receptors to clinical outcome in metastatic endometrial carcinoma: a Gynecologic
Oncology Group Study. Gynecol Oncol. 2007;106:325-333.
11. Covens AL, Filiaci V, Gersell D, et al. Phase II study of fulvestrant in recurrent/metastatic endometrial carcinoma: a Gynecologic Oncology Group study. Gynecol
Oncol. 2011;120:185-188.
12. Slomovitz BM, Jiang Y, Yates MS, et al. Phase II study of everolimus and letrozole in patients with recurrent endometrial carcinoma. J Clin Oncol. 2015;
33:930-936.
13. Fleming GF, Filiaci VL, Marzullo B, et al. Temsirolimus with or without megestrol acetate and tamoxifen for endometrial cancer: a gynecologic oncology group
study. Gynecol Oncol. 2014;132:585-592.
14. Soliman PT, Westin SN, Iglesias DA, et al. Everolimus, letrozole, and metformin in women with advanced or recurrent endometrioid endometrial cancer: a multi-
center, single arm, phase II study. Clin Cancer Res. 2020;26:581-587.

15. Runowicz CD, Nuchtern LM, Braunstein JD, et al. Heterogeneity in hormone receptor status in primary and metastatic endometrial cancer. Gynecol Oncol.
1990;38:437-441.
16. Tangen IL, Onyango TB, Kopperud R, et al. Androgen receptor as potential therapeutic target in metastatic endometrial cancer. Oncotarget. 2016;
7:49289-49298.
17. Sletten ET, Arnes M, Lyså LM, et al. Significance of progesterone receptors (PR-A and PR-B) expression as predictors for relapse after successful therapy of
endometrial hyperplasia: a retrospective cohort study. BJOG. 2019;126:936-943.
18. Tan DS, Lambros MB, Marchiò C, et al. ESR1 amplification in endometrial carcinomas: hope or hyperbole? J Pathol. 2008;216:271-274.
19. Gaillard SL, Andreano KJ, Gay LM, et al. Constitutively active ESR1 mutations in gynecologic malignancies and clinical response to estrogen-receptor directed
therapies. Gynecol Oncol. 2019;154:199-206.
20. Erkanli S, Ayhan A. Fertility-sparing therapy in young women with endometrial cancer: 2010 update. Int J Gynecol Cancer. 2010;20:1170-1187.
21. Ramirez PT, Frumovitz M, Bodurka DC, et al. Hormonal therapy for the management of grade 1 endometrial adenocarcinoma: a literature review. Gynecol
Oncol. 2004;95:133-138.
22. Mitsuhashi A, Habu Y, Kobayashi T, et al. Long-term outcomes of progestin plus metformin as a fertility-sparing treatment for atypical endometrial hyperplasia
and endometrial cancer patients. J Gynecol Oncol. 2019;30:e90.
23. van Weelden WJ, Massuger LFAG, et al; ENITEC. Anti-estrogen treatment in endometrial cancer: a systematic review. Front Oncol. 2019;9:359.
24. Lhommé C, Vennin P, Callet N, et al. A multicenter phase II study with triptorelin (sustained-release LHRH agonist) in advanced or recurrent endometrial
carcinoma: a French anticancer federation study. Gynecol Oncol. 1999;75:187-193.
25. Asbury RF, Brunetto VL, Lee RB, et al; Gynecologic Oncology Group. Goserelin acetate as treatment for recurrent endometrial carcinoma: a Gynecologic
Oncology Group study. Am J Clin Oncol. 2002;25:557-560.
26. Covens A, Thomas G, Shaw P, et al. A phase II study of leuprolide in advanced/recurrent endometrial cancer. Gynecol Oncol. 1997;64:126-129.
27. Zhang G, Cheng Y, Zhang Q, et al. ATX-LPA axis facilitates estrogen-induced endometrial cancer cell proliferation via MAPK/ERK signaling pathway. Mol Med
Rep. 2018;17:4245-4252.
28. Zhou L, Cai B, Bao W, et al. Crosstalk between estrogen receptor and mitogen-activated protein kinase signaling in the development and progression of
endometrial cancer. Int J Gynecol Cancer. 2011;21:1357-1365.
29. Bachelot T, Bourgier C, Cropet C, et al. Randomized phase II trial of everolimus in combination with tamoxifen in patients with hormone receptor-positive, human
epidermal growth factor receptor 2-negative metastatic breast cancer with prior exposure to aromatase inhibitors: a GINECO study. J Clin Oncol. 2012;
30:2718-2724.
30. Baselga J, Campone M, Piccart M, et al. Everolimus in postmenopausal hormone-receptor-positive advanced breast cancer. N Engl J Med. 2012;366:520-529.
31. Gu C, Zhang Z, Yu Y, et al. Inhibiting the PI3K/Akt pathway reversed progestin resistance in endometrial cancer. Cancer Sci. 2011;102:557-564.
32. Slomovitz BM, Filiaci VL, Coleman RL, et al. GOG 3007, a randomized phase II (RP2) trial of everolimus and letrozole (EL) or hormonal therapy (medrox-
yprogesterone acetate/tamoxifen, PT) in women with advanced, persistent or recurrent endometrial carcinoma (EC): a GOG Foundation study. Gynecol Oncol.
2018;149:2.
33. Tsuda H, Yamamoto K, Inoue T, et al. The role of p16-cyclin d/CDK-pRb pathway in the tumorigenesis of endometrioid-type endometrial carcinoma. Br
J Cancer. 2000;82:675-682.
34. Finn RS, Crown JP, Lang I, et al. The cyclin-dependent kinase 4/6 inhibitor palbociclib in combination with letrozole versus letrozole alone as first-line treatment
of oestrogen receptor-positive, HER2-negative, advanced breast cancer (PALOMA-1/TRIO-18): a randomised phase 2 study. Lancet Oncol. 2015;16:25-35.
35. Colon-Otero G, Weroha SJ, Zanfagnin V, et al. Results of a phase 2 trial of ribociclib and letrozole in patients with either relapsed estrogen receptor (ER)-positive
ovarian cancers or relapsed ER-positive endometrial cancers. J Clin Oncol. 2019;37 (suppl; abstr 5510).
36. Yang S, Xiao X, Jia Y, et al. Epigenetic modification restores functional PR expression in endometrial cancer cells. Curr Pharm Des. 2014;20:1874-1880.
37. Robinson JL, Macarthur S, Ross-Innes CS, et al. Androgen receptor driven transcription in molecular apocrine breast cancer is mediated by FoxA1. EMBO J.
2011;30:3019-3027.
38. Kamal AM, Bulmer JN, DeCruze SB, et al. Androgen receptors are acquired by healthy postmenopausal endometrial epithelium and their subsequent loss in
endometrial cancer is associated with poor survival. Br J Cancer. 2016;114:688-696.
39. Makker V, Green AK, Wenham RM, et al. New therapies for advanced, recurrent, and metastatic endometrial cancers. Gynecol Oncol Res Pract. 2017;4:19.
40. Tangen IL, Veneris JT, Halle MK, et al. Expression of glucocorticoid receptor is associated with aggressive primary endometrial cancer and increases from
primary to metastatic lesions. Gynecol Oncol. 2017;147:672-677.
41. Vahrenkamp JM, Yang CH, Rodriguez AC, et al. Clinical and genomic crosstalk between glucocorticoid receptor and estrogen receptor α in endometrial cancer.
Cell Rep. 2018;22:2995-3005.
42. Fleming GF, Sill MW, Darcy KM, et al. Phase II trial of trastuzumab in women with advanced or recurrent, HER2-positive endometrial carcinoma: a Gynecologic
Oncology Group study. Gynecol Oncol. 2010;116:15-20.
43. Slomovitz BM, Broaddus RR, Burke TW, et al. Her-2/neu overexpression and amplification in uterine papillary serous carcinoma. J Clin Oncol. 2004;
22:3126-3132.

44. Hainsworth JD, Meric-Bernstam F, Swanton C, et al. Targeted therapy for advanced solid tumors on the basis of molecular profiles: results from MyPathway, an
open-label, phase IIa multiple basket study. J Clin Oncol. 2018;36:536-542.
45. Fader AN, Roque DM, Siegel E, et al. Randomized phase II trial of carboplatin-paclitaxel versus carboplatin-paclitaxel-trastuzumab in uterine serous carcinomas
that overexpress human epidermal growth factor receptor 2/neu. J Clin Oncol. 2018;36:2044-2051.
46. Li BT, Makker V, Buonocore DJ, et al. A multi-histology basket trial of ado-trastuzumab emtansine in patients with HER2 amplified cancers. J Clin Oncol. 2018;
36 (suppl; abstr 2502).
47. Loibl S, von Minckwitz G, Schneeweiss A, et al. PIK3CA mutations are associated with lower rates of pathologic complete response to anti-human epidermal
growth factor receptor 2 (her2) therapy in primary HER2-overexpressing breast cancer. J Clin Oncol. 2014;32:3212-3220.
48. Growdon WB, Groeneweg J, Byron V, et al. HER2 over-expressing high grade endometrial cancer expresses high levels of p95HER2 variant. Gynecol Oncol.
2015;137:160-166.
49. Zhou L, Ren Y, Wang X, et al. Efficacy of afatinib in a HER2 amplification-positive endometrioid adenocarcinoma patient- a case report. OncoTargets Ther. 2019;
12:5305-5309.
50. Rudd ML, Price JC, Fogoros S, et al. A unique spectrum of somatic PIK3CA (p110alpha) mutations within primary endometrial carcinomas. Clin Cancer Res.
2011;17:1331-1340.
51. Urick ME, Rudd ML, Godwin AK, et al. PIK3R1 (p85α) is somatically mutated at high frequency in primary endometrial cancer. Cancer Res. 2011;
71:4061-4067.
52. Colombo N, McMeekin DS, Schwartz PE, et al. Ridaforolimus as a single agent in advanced endometrial cancer: results of a single-arm, phase 2 trial. Br
J Cancer. 2013;108:1021-1026.
53. Oza AM, Pignata S, Poveda A, et al. Randomized phase II trial of ridaforolimus in advanced endometrial carcinoma. J Clin Oncol. 2015;33:3576-3582.
54. Ray-Coquard I, Favier L, Weber B, et al. Everolimus as second- or third-line treatment of advanced endometrial cancer: ENDORAD, a phase II trial of GINECO. Br
J Cancer. 2013;108:1771-1777.
55. Mackay HJ, Eisenhauer EA, Kamel-Reid S, et al. Molecular determinants of outcome with mammalian target of rapamycin inhibition in endometrial cancer.
Cancer. 2014;120:603-610.
56. Myers AP. New strategies in endometrial cancer: targeting the PI3K/mTOR pathway--the devil is in the details. Clin Cancer Res. 2013;19:5264-5274.
57. Aghajanian C, Filiaci V, Dizon DS, et al. A phase II study of frontline paclitaxel/carboplatin/bevacizumab, paclitaxel/carboplatin/temsirolimus, or ixabepilone/
carboplatin/bevacizumab in advanced/recurrent endometrial cancer. Gynecol Oncol. 2018;150:274-281.
58. Roncolato F, Lindemann K, Willson ML, et al. PI3K/AKT/mTOR inhibitors for advanced or recurrent endometrial cancer. Cochrane Database Syst Rev. 2019;
10:CD012160.
59. Carracedo A, Pandolfi PP. The PTEN-PI3K pathway: of feedbacks and cross-talks. Oncogene. 2008;27:5527-5541.
60. Massarelli E, Varella-Garcia M, Tang X, et al. KRAS mutation is an important predictor of resistance to therapy with epidermal growth factor receptor tyrosine
kinase inhibitors in non-small-cell lung cancer. Clin Cancer Res. 2007;13:2890-2896.
61. Del Campo JM, Birrer M, Davis C, et al. A randomized phase II non-comparative study of PF-04691502 and gedatolisib (PF-05212384) in patients with
recurrent endometrial cancer. Gynecol Oncol. 2016;142:62-69.
62. Heudel PE, Fabbro M, Roemer-Becuwe C, et al. Phase II study of the PI3K inhibitor BKM120 in patients with advanced or recurrent endometrial carcinoma:
a stratified type I-type II study from the GINECO group. Br J Cancer. 2017;116:303-309.
63. Westin SN, Sill MW, Coleman RL, et al. Safety lead-in of the MEK inhibitor trametinib in combination with GSK2141795, an AKT inhibitor, in patients with
recurrent endometrial cancer: an NRG Oncology/GOG study. Gynecol Oncol. 2019;155:420-428.
64. Westin S, Litton J, Williams R, et al. Phase I expansion of olaparib (PARP inhibitor) and AZD5363 (AKT inhibitor) in recurrent ovarian, endometrial and triple
negative breast cancer. Ann Oncol. 2017;28:v122-v141.
65. Kamat AA, Merritt WM, Coffey D, et al. Clinical and biological significance of vascular endothelial growth factor in endometrial cancer. Clin Cancer Res. 2007;
13:7487-7495.
66. Ozuysal S, Bilgin T, Ozan H, et al. Angiogenesis in endometrial carcinoma: correlation with survival and clinicopathologic risk factors. Gynecol Obstet Invest.
2003;55:173-177.
67. Stefansson IM, Raeder M, Wik E, et al. Increased angiogenesis is associated with a 32-gene expression signature and 6p21 amplification in aggressive
endometrial cancer. Oncotarget. 2015;6:10634-10645.
68. Aghajanian C, Sill MW, Darcy KM, et al. Phase II trial of bevacizumab in recurrent or persistent endometrial cancer: a Gynecologic Oncology Group study. J Clin
Oncol. 2011;29:2259-2265.
69. Coleman RL, Sill MW, Lankes HA, et al. A phase II evaluation of aflibercept in the treatment of recurrent or persistent endometrial cancer: a Gynecologic
Oncology Group study. Gynecol Oncol. 2012;127:538-543.
70. Powell MA, Sill MW, Goodfellow PJ, et al. A phase II trial of brivanib in recurrent or persistent endometrial cancer: an NRG Oncology/Gynecologic Oncology
Group Study. Gynecol Oncol. 2014;135:38-43.

71. Bender D, Sill MW, Lankes HA, et al. A phase II evaluation of cediranib in the treatment of recurrent or persistent endometrial cancer: an NRG Oncology/
Gynecologic Oncology Group study. Gynecol Oncol. 2015;138:507-512.
72. Dizon DS, Sill MW, Schilder JM, et al. A phase II evaluation of nintedanib (BIBF-1120) in the treatment of recurrent or persistent endometrial cancer: an NRG
Oncology/Gynecologic Oncology Group Study. Gynecol Oncol. 2014;135:441-445.
73. Moore KN, Sill MW, Tenney ME, et al. A phase II trial of trebananib (AMG 386; IND#111071), a selective angiopoietin 1/2 neutralizing peptibody, in patients with
persistent/recurrent carcinoma of the endometrium: an NRG/Gynecologic Oncology Group trial. Gynecol Oncol. 2015;138:513-518.
74. Castonguay V, Lheureux S, Welch S, et al. A phase II trial of sunitinib in women with metastatic or recurrent endometrial carcinoma: a study of the Princess
Margaret, Chicago and California Consortia. Gynecol Oncol. 2014;134:274-280.
75. Dhani NC, Hirte HW, Wang L, et al. Phase II trial of cabozantinib in recurrent/metastatic endometrial cancer: a study of the Princess Margaret, Chicago and
California Consortia (NCI9322/PHL86). Clin Cancer Res. Epub 2020 Jan 28.
76. Lorusso D, Ferrandina G, Colombo N, et al. Randomized phase II trial of carboplatin-paclitaxel (CP) compared to carboplatin-paclitaxel-bevacizumab (CP-B) in
advanced (stage III-IV) or recurrent endometrial cancer: The MITO END-2 trial. J Clin Oncol. 2015;33 (suppl; abstr 5502).
77. Makker V, Rasco D, Vogelzang NJ, et al. Lenvatinib plus pembrolizumab in patients with advanced endometrial cancer: an interim analysis of a multicentre,
open-label, single-arm, phase 2 trial. Lancet Oncol. 2019;20:711-718.
78. Bindra RS, Gibson SL, Meng A, et al. Hypoxia-induced down-regulation of BRCA1 expression by E2Fs. Cancer Res. 2005;65:11597-11604.
79. Tentori L, Lacal PM, Muzi A, et al. Poly(ADP-ribose) polymerase (PARP) inhibition or PARP-1 gene deletion reduces angiogenesis. Eur J Cancer. 2007;
43:2124-2133.
80. Alvarez EA, Brady WE, Walker JL, et al. Phase II trial of combination bevacizumab and temsirolimus in the treatment of recurrent or persistent endometrial
carcinoma: a Gynecologic Oncology Group study. Gynecol Oncol. 2013;129:22-27.
81. de Jonge MM, Auguste A, van Wijk LM, et al. Frequent homologous recombination deficiency in high-grade endometrial carcinomas. Clin Cancer Res. 2019;
25:1087-1097.
82. de Jonge MM, Mooyaart AL, Vreeswijk MP, et al. Linking uterine serous carcinoma to BRCA1/2-associated cancer syndrome: a meta-analysis and case report.
Eur J Cancer. 2017;72:215-225.
83. Ring KL, Bruegl AS, Allen BA, et al. Germline multi-gene hereditary cancer panel testing in an unselected endometrial cancer cohort. Mod Pathol. 2016;
29:1381-1389.
84. Shu CA, Pike MC, Jotwani AR, et al. Uterine cancer after risk-reducing salpingo-oophorectomy without hysterectomy in women with BRCA mutations. JAMA
Oncol. 2016;2:1434-1440.
85. Hansen JM, Ring KL, Baggerly KA, et al. Clinical significance of homologous recombination deficiency (HRD) score testing in endometrial cancer patients. J Clin
Oncol. 2016;34 (suppl; abstr 5584).
86. Miyasaka A, Oda K, Ikeda Y, et al. Anti-tumor activity of olaparib, a poly (ADP-ribose) polymerase (PARP) inhibitor, in cultured endometrial carcinoma cells. BMC
Cancer. 2014;14:179.
87. Bian X, Gao J, Luo F, et al. PTEN deficiency sensitizes endometrioid endometrial cancer to compound PARP-PI3K inhibition but not PARP inhibition as
monotherapy. Oncogene. 2018;37:341-351.
88. Dedes KJ, Wetterskog D, Mendes-Pereira AM, et al. PTEN deficiency in endometrioid endometrial adenocarcinomas predicts sensitivity to PARP inhibitors. Sci
Transl Med. 2010;2:53ra75.
89. Mendes-Pereira AM, Martin SA, Brough R, et al. Synthetic lethal targeting of PTEN mutant cells with PARP inhibitors. EMBO Mol Med. 2009;1:315-322.
90. Martin SA, Lord CJ, Ashworth A. Therapeutic targeting of the DNA mismatch repair pathway. Clin Cancer Res. 2010;16:5107-5113.
91. Vilar E, Bartnik CM, Stenzel SL, et al. MRE11 deficiency increases sensitivity to poly(ADP-ribose) polymerase inhibition in microsatellite unstable colorectal
cancers. Cancer Res. 2011;71:2632-2642.
92. Sundar R, Brown J, Ingles Russo A, et al. Targeting ATR in cancer medicine. Curr Probl Cancer. 2017;41:302-315.
93. Williamson CT, Miller R, Pemberton HN, et al. ATR inhibitors as a synthetic lethal therapy for tumours deficient in ARID1A. Nat Commun. 2016;7:13837.
94. Shen J, Peng Y, Wei L, et al. ARID1A deficiency impairs the dna damage checkpoint and sensitizes cells to PARP inhibitors. Cancer Discov. 2015;5:752-767.
95. Gilad O, Nabet BY, Ragland RL, et al. Combining ATR suppression with oncogenic Ras synergistically increases genomic instability, causing synthetic lethality or
tumorigenesis in a dosage-dependent manner. Cancer Res. 2010;70:9693-9702.
96. Krajewska M, Fehrmann RS, Schoonen PM, et al. ATR inhibition preferentially targets homologous recombination-deficient tumor cells. Oncogene. 2015;
34:3474-3481.
98. Toledo LI, Murga M, Zur R, et al. A cell-based screen identifies ATR inhibitors with synthetic lethal properties for cancer-associated mutations. Nat Struct Mol
Biol. 2011;18:721-727.
99. Teng PN, Bateman NW, Darcy KM, et al. Pharmacologic inhibition of ATR and ATM offers clinically important distinctions to enhancing platinum or radiation
response in ovarian, endometrial, and cervical cancer cells. Gynecol Oncol. 2015;136:554-561.
100. Klaus A, Birchmeier W. Wnt signalling and its impact on development and cancer. Nat Rev Cancer. 2008;8:387-398.

101. Korinek V, Barker N, Morin PJ, et al. Constitutive transcriptional activation by a beta-catenin-Tcf complex in APC-/- colon carcinoma. Science. 1997;
275:1784-1787.
102. Tsukamoto AS, Grosschedl R, Guzman RC, et al. Expression of the int-1 gene in transgenic mice is associated with mammary gland hyperplasia and ad-
enocarcinomas in male and female mice. Cell. 1988;55:619-625.
103. Altwerger G, Bonazzoli E, Bellone S, et al. In vitro and in vivo activity of IMGN853, an antibody-drug conjugate targeting folate receptor alpha linked to DM4, in
biologically aggressive endometrial cancers. Mol Cancer Ther. 2018;17:1003-1011.

GYNECOLOGIC CANCER
PARP Inhibitors for Ovarian Cancer: Current

Indications, Future Combinations, and Novel
Assets in Development to Target DNA
Damage Repair
Panagiotis A. Konstantinopoulos, MD1; Stephanie Lheureux, MD2; and Kathleen N. Moore, MD3
PARP inhibitors (PARPIs) have revolutionized the treatment of epithelial ovarian cancer, first for BRCA-
overview
associated cancer, and, recently, for all epithelial cancers of serous or high-grade endometrioid subtypes in
the front line. Although there is hope that PARPIs will help prevent recurrences when used following frontline
maintenance, cancer will still recur in most women, and the need for active combination strategies as well as
continued development of novel assets, either as monotherapy or in combination, will be urgently needed. This
review article discusses the current indications for PARPIs in both frontline and recurrent settings, current
research in combination approaches, and finally, ongoing research on novel methods to target DNA damage
response in an effort to exploit the common susceptibility to DNA damage repair in epithelial ovarian cancer
and improve outcomes for patients.
PARP INHIBITORS: FOR WHOM, WHEN, AND HOW? study provided the benchmark for median PFS in
The treatment paradigm for epithelial ovarian cancer a study of advanced EOC without selection by his-
(EOC), and, in particular, high-grade serous ovarian tology or molecular subtype of 17.9 months.3 The
cancer (HGSOC), has been changing rapidly over the first big paradigm shift in frontline EOC (again, un-
past 5 years. The typical course of EOC has been well selected by histology or molecular subtype) was the
established for decades. Women present, usually with addition of bevacizumab with and to follow frontline
advanced stage disease; they undergo a combination therapy, which resulted in an improvement in PFS
of paclitaxel and carboplatin chemotherapy and sur- by approximately 4 months but no improvement in
gical cytoreduction as part of frontline management; overall survival (OS).4,5
and they commonly achieve a state of clinical re- Perhaps the most important discovery for EOC and
mission, but unfortunately 80% will have recurrence of progress in outcomes is that EOC is not one disease but
cancer within 3 years of diagnosis. Once recurred, the at least five: HGSOC, endometriod, clear cell, mucin-
disease is no longer curable but fortunately treatable ous, and low-grade serous.6 Each of these have unique
for many years. Treatment decisions are based on molecular signatures and targets. Pulling out specific
many factors inclusive of the time interval from last subtypes has led to the transformative and rapid in-
platinum therapy, histology, molecular profile, residual troduction of PARP inhibitors (PARPIs) into the
toxicities, and resectability. Despite many options for treatment paradigm. In particular, the recognition of
treatment in the recurrent setting, the disease becomes homologous recombination deficiency (HRD) as an
Author affiliations
and support
eventually resistant to all interventions, and women important and potential predictive biomarker for EOC
information (if succumb to their disease.1,2 and HGSOC has led to new indications for PARPIs
applicable) appear
The most contemporary clinical trial to evaluate dif- across the treatment spectrum. As a reminder, ho-
at the end of this
article. ferent delivery schedules for paclitaxel and carboplatin mologous recombination (HR) is the high-fidelity
Accepted on March (and likely the last to do so) was ICON8, which com- process by which double strand DNA damage is
31, 2020 and pared standard every 21-day dosing with weekly repaired. It uses sister chromatids as the template,
published at paclitaxel and every 21-day carboplatin with weekly occurs in the G2/M phase of the cell cycle, and is
ascopubs.org on dosing of both paclitaxel and carboplatin. No mainte- dependent on the presence of proteins encoded by
April 30, 2020:
nance was used. There was no difference in progression- BRCA1, BRCA2, and others in the Fanconi anemia
10.1200/EDBK_ free survival (PFS), making every 21-day paclitaxel pathway for its function. The PARP 1 protein is also
288015 and carboplatin the preferred option. In addition, this important because it recruits MRE11 and NBS1 to the

PARP Inhibitors for Ovarian Cancer
single-agent treatment and maintenance), and among more

heavily pretreated patients with recurrent disease. This
article summarizes the data published to date in these three
• Women with BRCA-associated ovarian cancer treatment settings, with the caveat that data presented at
should be offered maintenance therapy with
the 2020 ASCO Annual Meeting will affect how PARPIs
olaparib or niraparib for 2 3 years.
use continues to evolve.
• Women with epithelial ovarian cancer (high-
grade serous, high-grade endometrioid) may be Frontline Maintenance
offered olaparib or niraparib following complete The potential for PARPI use to be transformative in frontline
or partial response to platinum-based ovarian EOC was first demonstrated in the SOLO-1 study. Women
cancer.
with BRCA-associated, advanced HGSOC, or endometrioid
• Women who are treated with bevacizumab cancer who achieved a complete response (CR) or partial
concurrent with platinum-based chemotherapy response (PR) to frontline, platinum-based therapy were
and who have high grade serous or endo- randomly assigned 2:1 to receive olaparib versus placebo
metrioid cancer may have olaparib added
until progression or 2 years. Use of maintenance olaparib in
as maintenance for up to 2 years following
complete or partial response to therapy. this setting led to a hazard ratio of 0.30 (95% CI, 0.23–0.41)
and median PFS (not including time of chemotherapy) of
• Women with recurrent disease who respond to approximately 49 months versus 13.8 months. OS was
repeated use of platinum-based chemotherapy
immature.11 Exploratory analysis in women who entered the
and have not previously received a PARPI, may
be offered PARPI maintenance with olaparib or study with primary versus interval surgery demonstrated
niraparib to continue until toxicity or similar hazard ratio benefits (0.31 and 0.37, respectively).
progression. Similarly, women who entered the study with no residual
versus residual disease benefitted from olaparib mainte-
• Women with recurrent disease that is associ-
ated with BRCA may use olaparib ( 3 previous nance with hazard ratios of 0.33 and 0.44, respectively.
lines), rucaparib ( 2 previous lines), or nir- Among the “best prognostic” group of women who entered
aparib (homologous recombination deficient, a study with primary cytoreduction to no gross residual, the
platinum sensitive, . 3 previous lines). hazard ratio was 0.32 with 71% PFS versus 35% PFS at
3 years.12 Based on this data, olaparib gained U.S. Food and
Drug Administration approval for use as maintenance fol-
site of DNA damage as well as blocks entry into non-
lowing frontline chemotherapy in BRCA-associated ovarian
homologous end joining (NHEJ), which is the low-fidelity
cancer in January 2019.
manner by which double strand DNA may be repaired. In
cells with loss of proteins key to HR, inhibition of PARP leads As stated previously, there are alterations beyond BRCA that
to complete loss of HR and entry into NHEJ, which repairs indicate HRD and PARPIs may work for these patients as
DNA in an error-prone manner and leads to accumulation well. Because of the lack of a proven test to definitely identify
of DNA damage and cell death.7-9 those patients with HRD compared with homologous re-
combination proficient (HRp), three studies were completed
HR proficiency appears to be lost in 50% of HGSOC, with that allowed all comers to enroll with appropriate stratifi-
approximately 30% loss due to germline (14%), somatic cation. These were the PRIMA, PAOLA1, and Velia studies.
(6%), or epigenetic (10%) loss of BRCA1 or 2 function
PRIMA/ENGOT-OV26/GOG3012 enrolled women with
(epigenetic changes are reported for BRCA1 only). An
HGSOC or endometrioid, advanced-stage, high-risk cancer
additional 11% of HGSOC tumors harbor HRD due to other
(stage IV, neoadjuvant chemotherapy or residual disease
mutations in CDK1/2, Fanconi anemia genes, core RAD
following primary cytoreduction) who were in CR or PR
gene mutations, HR DNA damage gene mutations, or following frontline paclitaxel and carboplatin therapy. This
promoter methylation of RAD51C. Approximately 30% of study randomly assigned participants 2:1 to receive nir-
HGSOC may have other alterations that do not cause HRD aparib or placebo until progression or 3 years. Stratification
but still render tumors sensitive to PARPIs (e.g., nucleotide factors included HR status using an assay that tested loss of
excision repair [NER] mutations and MMR mutations); heterozygosity, telomeric allelic imbalance, and large-scale
these are under study. Fifteen percent of HGSOCs have state transitions. Women were classified as having HRD if
cyclin E1 (CCNE1) amplification and are not sensitive to the score was 42 or higher. The primary endpoint for PRIMA
PARPIs.10 Because of the prevalence of HRD, exploration of was PFS in the intention-to-treat population (ITT) and the
PARPIs in HGSOC (and high-grade endometrioid) was HRD population as determined by blinded radiographic
warranted, and, as of this report, completed as a part of review. This endpoint had a hazard ratio of 0.62 (95% CI,
frontline therapy, platinum-sensitive recurrent therapy (both 0.5–0.76; p , .001) and median PFS of 13.8 months versus

Konstantinopoulos, Lheureux, and Moore
8.2 months. In the HRD population, the hazard ratio was 0.60–1.09) and median PFS of 15 months versus 11.5
0.43 (95% CI, 0.31–0.59; p = .001) and median PFS was months (Table 1).15
21.9 months versus 10.4 months. This population in-
Ultimately, how these assets are used will depend on the
cluded those patients with BRCA-associated cancers. indications allowed by regulatory agencies, which are all still
In exploratory nonhypothesis-tested analysis, the hazard pending at the time of writing, but anticipated in May 2020.
ratio among only BRCA-associated cancers was 0.40 (95% Assuming that PAOLA-1 and PRIMA trials gain regulatory
CI, 0.27–0.62), among HRD/wild-type (BRCAwt) was 0.50 approval, the key questions will be: (1) do women with
(95% CI, 0.31–0.83), and among HRp was 0.68 (95% CI, BRCA-associated cancers need bevacizumab and/or ola-
0.49–0.94).13 parib enough; and (2) although PRIMA was positive for ITT
PAOLA-1/ENGOT-OV25 enrolled women with HGSOC or (for women with HRp tumors), is niraparib equivalent,
endometrioid, advanced-stage cancers who were in CR or better, or less effective than bevacizumab? Neither question
PR following frontline, platinum-based chemotherapy, has level 1 evidence to guide response; however, recent
including bevacizumab. They were then randomly selected data presented at the Society for Gynecologic Oncology
2:1 to continue bevacizumab with olaparib, added for 2 years (SGO) 2020 meeting attempted to compare SOLO-1 and
or until progression versus placebo and bevacizumab. The PAOLA-1 using a population-adjusted indirect treatment
primary endpoint was investigator-assessed PFS in the ITT comparison to compare olaparib plus bevacizumab versus
population, and stratification was by BRCA-associated cancers. olaparib monotherapy, olaparib monotherapy versus bev-
The hazard ratio was 0.59 (95% CI, 0.49–0.72; p , .0001) in acizumab monotherapy, and bevacizumab monotherapy
favor of the addition of olaparib, with a median PFS of versus placebo among women with BRCA-associated
22.1 months versus 16.6 months (this study had an cancers. For the first comparison of olaparib plus bev-
active control compared with only placebo).14 Exploratory acizumab versus olaparib, the hazard ratio was 0.71 (95%
nonhypothesis-tested endpoints included outcomes among CI, 0.454–1.09), for olaparib monotherapy versus bev-
only BRCA-associated cancers, with a hazard ratio of 0.31 acizumab monotherapy, the hazard ratio was 0.48 (95% CI,
(95% CI, 0.2–0.47), a BRCAwt hazard ratio of 0.71 (95% CI, 0.30–0.75), and for bevacizumab versus placebo, the
0.58–0.88), an HRD hazard ratio of 0.33 (95% CI, 0.25–0.45), hazard ratio was 0.65 (95% CI, 0.43–0.95). This exploratory
an HRD/BRCAwt hazard ratio of 0.43 (95% CI, 0.28–0.66), analysis suggested a potential additive benefit of bev-
and an HRp/ukn hazard ratio of 0.92 (95% CI, 0.72–1.17).14 acizumab to olaparib among women with BRCA-associated
cancers and reinforced the understanding that, although an
Velia/GOG 3005 differed from the other studies in that it active asset, bevacizumab was not equivalent to olaparib as
attempted to incorporate the PARPI veliparib with and to monotherapy maintenance in this population.16 As to the
follow chemotherapy. Eligible patients had HGSOC and second question concerning efficacy among patients with
advanced-stage disease and were enrolled at the time of HRp, both accurate identification of patients whose tumor
chemotherapy initiation. Randomization was 1:1:1 to car- truly is proficient in HR, and evaluating effective therapies
boplatin/paclitaxel/veliparib followed by veliparib mainte- are high, as of yet unmet, needs. In all studies, use of HRD
nance (arm 1), carboplatin/paclitaxel/veliparib followed by assays failed to identify patients who would not benefit from
placebo maintenance (arm 2), and carboplatin/paclitaxel/ PARPI maintenance. In another exploratory analysis pre-
placebo followed by placebo maintenance (arm 3). The sented at SGO 2020, Swisher et al evaluated the association
primary endpoint was PFS for arm 1 versus arm 3 in the ITT- of the Myriad myChoice Genomic Instability Score with PFS
and BRCA-associated cancers. Stratification factors in- among patients with BRCAwt in the VELIA trial. Regardless
cluded BRCA status. For the ITT group, the hazard ratio was of what Genomic Instability Score cutpoint was selected,
0.68 (95% CI, 0.56–0.83; p , .001) and median PFS was there was PFS benefit in patients categorized as HRp and
23.5 months versus 17.3 months. Among the BRCA pop- HRD and therefore did not have a predictive roll for PFS
ulation, the hazard ratio was 0.44 (95% CI, 0.28–0.68; p , benefit in this trial.17 Therefore, we can speculate that there
.001) and median PFS of 34.7 months versus 22 months. is a benefit of PARPIs in patients currently classified as HRp,
These PFS values were inclusive of time on chemotherapy but we cannot speculate on the magnitude of benefit in this
and that all patients who started the trial contributed to the population for PARPIs versus bevavcizumab. That will have
PFS, including those with stable disease or progressive to await another trial.
disease during chemotherapy. Exploratory nonhypothesis-
tested cohorts included patients with BRCAwt with a hazard Platinum-Sensitive Disease: Maintenance and Treatment
ratio of 0.80 (95% CI, 0.64–0.997) and a median PFS of The first full approvals for PARPIs came from three phase III
18.2 months versus 15.1 months; an HRD/BRCAwt hazard studies and one phase II study that evaluated PARPIs as
ratio of 0.74 (95% CI, 0.52–1.06) and median PFS of maintenance following response to platinum-based therapy
22.9 versus 19.8; an HRp hazard ratio of 0.81 (95% CI, in the recurrent setting. Although all four studies had some

TABLE 1. Key Phase III Studies of PARP Inhibitors in Frontline Ovarian Cancer
BRCA Study
Study Status Starts With Study Arm Control Arm ITT BRCAm HRD HRD/BRCAwt HRp
11
SOLO1 BRCA+ Maint Olaparib Placebo NA HR, 0.30* NA NA NA
(0.23–0.41);
mPFS: NR vs.
13.8
PRIMA13 All comers Maint Niraparib Placebo HR, 0.62* HR, 0.40 HR, 0.43* HR, 0.50 HR, 0.68
(0.50–0.76); (0.27–0.62) (0.31– (0.31–0.83) (0.49–0.94)
0.59);
mPFS: 13.8 vs. mPFS: 21.9 vs.
8.2 10.4
PAOLA114 All comers Maint Olaparib + Placebo + HR, 0.59* HR, 0.31 HR, 0.33 HR, 0.43 HR, 0.92
Bev Bev (0.49–0.72); (0.20–0.47); (0.25– (0.28–0.66); (0.72–1.17);
0.45);
mPFS: 22.1 vs. mPFS: 37 vs. mPFS: 37 vs. mPFS: 28.1 vs. mPFS: 16.9 vs.
16.6 22 17.7 16.6 16
VELIA15 All comers With chemo Veliparib Placebo HR, 0.68* HR, 0.44* NR HR, 0.74 HR, 0.81
(0.56–0.83); (0.28–0.68); (0.52–1.06); (0.60–1.09);
mPFS: 23.5 vs. mPFS: 34.7 vs. mPFS: 22.9 vs. mPFS: 15 vs.
17.3 22 19.8 11.5
*Primary endpoints.
Abbreviations: Bev, bevacizumab; BRCAwt, wild-type BRCA; HR, hazard ratio; HRp, homologous recombination proficient; HRD, homologous recombination
deficiency; ITT, intention to treat; Maint, maintenance; mPFS, median progression-free survival; NR, not reported.
differences in design, they were largely similar because they the study was declared positive based on an odds rate ratio
enrolled women with CR or PR to platinum-based therapy in of 72% versus 51% (odds ratio [OR], 2.53; 95% CI,
the recurrent setting who were randomized to either PARPIs 1.40 4.58; p = .002).24 Niraparib was studied and approved
or placebo. The ARIEL3, NOVA, and Study 19 trials enrolled in women with recurrent EOC whose tumors had evidence
all comers, and SOLO 2 enrolled only those with BRCA- of HRD based on the QUADRA study (Table 3).25
associated cancers. Table 2 summarizes the key findings
Currently, PARPIs are indicated in frontline OC as main-
of these trials. Interestingly, PARPI maintenance was ef-
tenance for patients with BRCA-associated cancer who are
fective in all treatment subgroups, BRCA-associated can-
in CR or PR following chemotherapy. PARPIs may be in-
cers, HRD, and HRp. This finding, compared with the
dicated beyond BRCA following potential results announced
seemingly poorer efficacy in HRp in the front line, could be
during the 2020 ASCO Annual Meeting. PARPIs are in-
explained by the fact that in the recurrent setting, we had
dicated as maintenance therapy following response to
a clinical biomarker of platinum sensitive to identify, one
platinum-based chemotherapy without reference to BRCA
perhaps better than current assays, which benefits from
status because of the power of the clinical biomarker:
PARPI maintenance.
platinum sensitivity. Finally, for women with recurrent,
Finally, there were single-arm phase II data and randomized PARPI-naı̈ve disease, PARPIs are available as treatment
phase III data that provide efficacy data for use of PARPI for instead of chemotherapy for women with BRCA-associated
treatment of ovarian cancer, instead of chemotherapy not and HRD+ recurrent disease. New challenges include
following chemotherapy. Olaparib was the first PARPI ap- evaluation efficacy with repeat PARPIs use and combina-
proved in EOC for women with BRCA-associated cancers tions that improve outcomes among women who have HRp
who received three or more lines of chemotherapy.22 This or who otherwise do not benefit from a monotherapy PARPI.
was followed by rucaparib, which was approved in women
with both germline and somatic BRCA-associated cancer PARP INHIBITOR THERAPY: CURRENT CHALLENGES AND
who received two or more lines of chemotherapy.23 Re- FUTURE OPPORTUNITIES
cently, SOLO-3 evaluated olaparib compared with investi- Women with HRD, in particular BRCA1/2-mutated HGSOC,
gator choice chemotherapy in a randomized phase III study greatly benefit from PARPI therapy.26 However, because of
among women with recurrent BRCA-associated EOC who the lack of a measurable surrogate to evaluate HRD in the
were considered sensitive to platinum and received more clinic, PARPI approval has been granted as maintenance
than three lines of chemotherapy. Notably, there was no treatment in HGSOC post-response to platinum, regardless
platinum among the investigator choice options; however, of HRD status. In addition, besides HRD, high replication

TABLE 2. Key Phase II and III Studies of PARP Inhibitors in Platinum-Sensitive Recurrent Disease
Study Study 1918 SOLO-219 gBRCAm NOVA20 gBRCAm NOVA20 nongBRCAm ARIEL-321 BRCAm ARIEL-321 ITT
Agent Olaparib Olaparib Niraparib Niraparib Rucaparib Rucaparib
Difference in PFS (months) 8.4 vs. 4.8 19.1 vs. 5.5 21.0 vs. 5.5 9.3 vs. 3.9 16.6 vs. 5.4 10.8 vs. 5.4
PFS HR (investigator assessed) 0.35 (95% CI, 0.30 (95% CI, 0.27 (95% CI, 0.53 (95% CI, 0.23 (95% CI, 0.36 (95% CI,
0.25 –0.49; 0.22–0.41; 0.18–0.40) 0.41–0.68) 0.16–0.34, 0.30–0.45;
p , .001) p , .0001) p , .0001) p , .0001)
PFS HR (BICR) 0.39 (95% CI, 0.25 (95% CI, 0.27 (95% CI, 0.45 (95% CI, 0.20 (95% CI, 0.35 (95% CI,
0.27– 0.55; 0.18-0.35; 0.17–0.41; 0.34– 0.61; 0.13–0.32; 0.28–0.45;
p , .001) p , .0001) p , .0001) p , .0001) p , .0001) p , .0001)
Abbreviations: BICR, blinded radiographic review; HR, hazard ratio; PFS, progression-free survival.
stress induced by either loss of tumor suppressor gene or PARylation33; (4) stabilization of stalled forks34; and (5)
oncogene amplification can contribute toward PARPI restoration of HR function.
sensitivity.27 Although PARPIs have led to considerable
Several of these mechanisms can lead tumors to switch
benefit in women with HGSOC, patient selection remains
from a HR-deficient to a HR-proficient state, and thereby,
a challenge. Not all patients benefit from this treatment, and
develop PARPI resistance. The most common mechanism
response is not definitive, resulting in long-term use of the
is through the restoration of the expression or function of key
drug and development of resistance. Efforts are ongoing to
proteins in the HR pathway by the acquisition of secondary
identify potential ways to augment the overall benefit of
mutations, also known as “reversion mutations.” Reversion
PARPIs in women’s cancer. This involves (1) deeper un-
mutations in multiple HR pathway genes, including BRCA1,
derstanding of the landscape of PARPI response mecha-
BRCA2, RAD51C, RAD51D, and PALB2 have been re-
nisms and ways to overcome resistance; and (2) identifying
ported in both preclinical and clinical studies35-40 and have
new strategies to enhance the scope of PARPIs through
enabled a functional restoration of the HR defect. Wild-type
combining PARPIs with other therapies (Fig. 1).
BRCA1 levels or functionality may also be restored through
Understanding PARPI Resistance Mechanisms demethylation of its hypermethylated promoter or intronic
Alu-mediated gene rearrangements.41,42 In addition, HR
As PARPIs move earlier into the treatment paradigm,
can be re-activated by suppressing the HR-counteracting
identifying mechanisms of primary and acquired resistance
pathway, NHEJ, with the loss of proteins involved in NHEJ
is key to guide treatment and prevent recurrence. Following
(e.g., TP53BP1, RIF1, and REV7), which result in PARPI
approval of various PARPIs, drug resistance has emerged in
resistance.43,44
practice; yet, no standard treatment has been established
for post-PARPI progression. Several PARPI resistance Another important mechanism of PARPI resistance is the
mechanisms have been described in the literature,28,29 stabilization of the stalled replication fork.30 At the time of
including: (1) increased drug efflux, such as overexpression DNA damage, the cell cycle is halted, which allows cells to
of the multidrug resistance gene, ABCB1, which involves repair the damaged DNA with the aid of BRCA1/2, which
the promoter fusion of this gene30; (2) loss of PARP 1 acts by stabilization of the stalled replication forks. In
function due to mutation or deletion31,32; (3) loss of PARG BRCA1/2 deficient tumors, these stalled forks are subjected
(poly [ADP-ribose] glycohydrolase) and restoration of to MRE11- and MUS81-mediated degradation. PARPIs
TABLE 3. Single-Agent PARP Inhibitor in Epithelial Ovarian Cancer

ARIEL2/Study 102 ARIEL2/Study 102, QUADRA QUADRA
Study Study 122 BRCAm23 BRCAwt23 gBRCAm25 HRD+25 SOLO324 SOLO324
Agent Olaparib Rucaparib Rucaparib Niraparib Niraparib Olaparib IC
ORR 34% (95% CI, 53.8% (95% CI, 44–64) 29% (LOH high) 29% 27% 72% 51%
26–42)
10% (LOH low)
DOR 7.9 months (95% CI, 9.2 months (95% CI, 10.8 (5.7-NR) 8.3 (6.6–NR) 9.2 (5.9–15.2) 9.4 (5.6– 10.2 (5.5–
5.6–9.6) 6.6–11.6) 25.7) 15.3)
LOH-H
5.6 (4.6–8.5) LOH-L
LOT 3 2 2 4th–5th line 4th–5th line 2 2
Abbreviations: DOR, duration of response; LOH, loss of heterozygosity; LOT, line of therapy; ORR, overall response rate.

FIGURE 1. Challenges and Opportunities Using Combinatorial Approaches Following Long-Term PARP Exposure
(A) Varied genetic, epigenetic, and functional PARP inhibitor (PARPi) resistance mechanisms are described. (B) Current PARPi therapy can be enhanced
using various combinatorial approaches targeting the tumor immune–microenvironment axis.
Abbreviations: DDR, DNA damage repair; DNMTi, DNA methyltransferase inhibitor; HR, homologous recombination; HRD, homologous recombination
deficiency; NHEJ, nonhomologous end joining; ROS, reactive oxygen species.
further act upon this destabilized replication fork, which cells exhibited increased levels of histone H3 lysine 9
leads to their catastrophic degradation. There are three (H3K9me2) and histone-lysine-N-methyl transferases 1 and
mechanisms described in the literature that are involved 2 (EHMT1/2).45 These cells were re-sensitized to PARPI
in fork degradation.29 Loss of factors involved in these through disruption of EHMT1/2 and involved DNA damage
mechanisms may lead to fork protection or stabilization, and and cell cycle dysregulation.45
thereby, PARPI resistance. The first mechanism involves Although several mechanisms of PARPI resistance were
fork reversal by chromatin remodelers (e.g., SMARCAL1, identified in the preclinical settings and their repertoire
ZRANB3, and HTLF), which, in turn, mediate MRE11- seemed to increase over time, evidence toward their exis-
dependent fork degradation.29 The second mechanism tence in the clinical setting are still limited. In practice,
involves EZH2-mediated methylation of histone H3 at lysine platinum sensitivity is more commonly used as a valuable
27 (H3K27), which then relays MUS8-mediated fork deg- indicator of response to PARPIs.46 Because platinum
radation.29 Lastly, methylation of H3K4 by the histone chemotherapy induces DNA lesions as well as aberrant cell
methyltransferase complex of PTIP, MLL3, and MLL4 also cycle and DNA repair signaling,47 some of the mechanisms
leads to MRE11-mediated fork degradation.29 Besides of platinum-based response and/or resistance may overlap
these mechanisms, the loss of the replication stress effector, with PARPI resistance. For example, CCNE1 amplification,
SLFN11, may also contribute to PARPI resistance.29 a well-described marker of platinum resistance in HGSOC,
Recently, new evidence hinted toward involvement of epi- has been observed with poor response to PARPIs.48,49
genetic factors in PARPI resistance. A preclinical study The other commonly observed resistance mechanisms in
utilizing HGSOC cell lines and in vivo patient-derived xe- the clinic are reversion mutations in HR genes (BRCA1/2,
nograft models showed that PARPI-resistant HGSOC tumor RAD51C/D, and so on).35,36,38,50 Recently, a pilot prospective

study showed that multiple mechanisms of resistance could and Rad3-related (ATR) signaling pathway, and ATR in-
be observed in patients post-PARPI progression, which hibition was able to overcome PARPI resistance.55,57,58
highlighted that PARPI resistance was multifactorial.51 Several early-phase clinical trials investigating ATR and
These included reversion mutations in BRCA1/2 and PARP inhibitors combinations are ongoing (NCT03682289,
other HR genes, ABCB1 upregulation, CCNE1 amplifi- NCT02576444, NCT02723864, NCT03462342). Down-
cation, and SLFN11 downregulation.51 Because multiple stream of ATR/ATM are checkpoint kinases: CHK1/2, fol-
mechanisms can lead to the restoration of DNA repair lowed by Wee1 tyrosine kinase proteins, and those that
function and PARPI resistance, patient-by-patient analysis mediate the G2-M checkpoint. Inhibition of CHK1/2 using
is needed to identify the specific mechanism(s) of re- prexasertib along with olaparib has shown early benefit in
sistance involved for each patient.52 Detecting the mech- BRCA-mutated tumors in a phase I study of HGSOC and
anism involved at the time of recurrence is important for other advanced solids tumors.59 Similarly, combining Wee1
the decision-making process and identifying potential new kinase inhibitor (AZD1775) along with PARPIs in OC is
vulnerabilities to target. This is an area of active investiga- currently underway in two phase II studies (NCT02576444
tion, and current research focuses on: (1) measuring the net and NCT03579316). Besides ATR, CHK1/2, and WEE1,
functional impact of the acquired mechanism(s) of re- other DDR inhibitors currently under development can be
sistance on the HR defect in a patient’s clinical journey; and potentially combined with PARPIs (e.g., ATM, DNA-PK, and
(2) identifying the pathways or cellular mechanisms that POLθ).60 Bone marrow toxicity is a major concern when
have been considerably altered due to acquisition of the combining two DDR inhibitors; therefore, careful dosing and
diverse mechanisms of resistance. sequential administration of the drug is an important con-
sideration while designing trials for these combinations.27,60
Identifying New PARPI Combination Strategies
Combining With Anti-Angiogenics
The rationale of PARPI combination is to enhance PARPI
activity by targeting alternative DNA damage repair (DDR) PARPIs and anti-angiogenics have shown synergy in pre-
dependencies and reprogramming the microenvironment to clinical studies. For example, cediranib-mediated induction
overcome the challenge of resistance.52 Several combina- of hypoxia leads to altered gene expression of DNA repair
torial approaches in both HR deficient and proficient OC genes and thus impaired HR.61 The combination of
tumors are under investigation both in preclinical and cediranib and olaparib is under investigation with several
clinical studies. large ongoing clinical trials for recurrent OC after benefit
was observed in phase II clinical trials (NCT02446600,
Combining With Chemotherapeutic Agents NCT02502266, NCT02889900).53 The synergy of this
Studies targeting different chemotherapeutic agents such strategy was further confirmed in the clinic with the com-
as platinum, topoisomerase inhibitors, or DNA alkylating bination of the PARPI niraparib and anti-VEGF-A antibody,
agents have been tested in combination with PARPIs in bevacizumab, investigated in a phase II trial for platinum-
early-phase trials in OC and other solid tumors.53 The sensitive recurrent OC (NCT02354131).62 Recently, ola-
combination may act synergistically by enhancing PARP parib and bevacizumab were approved in the frontline
trapping or its catalytic activity.53 Alternatively, PARPIs may maintenance setting following a positive result from
act through exploiting the cytotoxic activity of these agents.53 the randomized phase III PAOLA-1 trial (ENGOT-OV25,
The major limitation of this combination is overlapping NCT02477644). 14
toxicities and dosing issues, and thus, PARPIs support the At the time of PARPI progression, the addition of the anti-
sequential use of these therapies in patients who respond angiogenic cediranib, showed activity in some patients
well to chemotherapy. (NCT02681237),51 highlighting the benefit of targeting
Combining With Other DDR or Cell Cycle simultaneous angiogenesis and the HRD pathway. This
Pathways Inhibitors suggested the importance of the tumor microenvironment
for enhancing PARPI activity. Combination with anti-
Another attractive approach is to combine PARPIs with angiogenics can be an attractive strategy, particularly in the
other DDR inhibitors based on the rationale that PARPI- HR proficient tumor or the post-PARPI setting.51
resistant cells exhibit enhanced dependency on other
DNA repair or cell cycle mechanisms, which can then be Combining With Immunotherapy
exploited toward an effective combination.54-57 Preclinical Recently, immune checkpoint inhibitors showed phenom-
and early-phase clinical trials have shown promising results enal response in patients with increased tumor mutation
upon combining DDR and cell cycle inhibitors with PARPIs, burden and led to therapeutic changes in pratice.63,64 DDR
targeting proteins such as ATR, Chk1/2, and Wee1. In deficiency was associated with higher tumor mutation
preclinical studies, PARPI-resistant, BRCA-mutated cancer burden,27 and BRCA1/2 or HR-deficient OC increased the
cells showed increased reliance on the ataxia-telangiectasia neo-antigen load compared with HR-proficient tumors.65

PARPI treatment led to overexpression of PD-L1, which their role as master regulators of DDR, ATM and ATR have
resulted in cancer-associated immunosuppression. There- long been the focus of drug discovery efforts, and inhibitors
fore, combining PD-L1 blockade therapy along with PARPIs of these kinases are currently undergoing clinical evaluation
might circumvent the problem of PARPI-led immune sup- as anticancer agents.60,73
pression and augment the efficacy of PARPI therapy.66,67 In Targeting the ATM-CHK2 DNA Damage Checkpoint
addition, PARPI administration led to acute inflammation,
reprogramming of the tumor immune microenvironment, The ATM–CHK2 pathway primarily responds to DNA double
and a systemic immune response.12,68 Because of the strand breaks (DSBs) by promoting G1 arrest (via phos-
pharmacodynamics changes induced by PARPI, target- phorylation and activation of CHK2 and TP53) and DNA
ing the immune environment at the time of PARPI repair via activation of HRR or NHEJ.74 ATM promotes HRR
treatment seems attractive and is being tested in several by recruiting BRCA1 to DSBs but can also promote NHEJ by
clinical trials in the recurrent OC and frontline mainte- antagonizing BRCA1 and recruiting TP53 binding protein 1
nance settings.53 (53BP1).74,75 These antagonistic functions are cell cycle
regulated, whereby NHEJ is the predominant DSB repair
Because of the intertwined regulation of angiogenesis and mechanism used in the G1 phase, and whereas HRR is
immune modulation, reprogramming the tumor microen- predominant in the S phase.75 Germline mutations of ATM
vironment by combining anti-angiogenic agents and im- result in the well-characterized human autosomal recessive
munotherapy could enhance antitumor responses. Recent disorder ataxia-telangiectasia, which is associated with
studies investigated triplet combinations of PARPIs, anti- hypersensitivity to ionizing radiation, failure of cells to arrest
angiogenics, and immune checkpoint inhibitors. The early the cell cycle after induction of DSBs, and an increased
results showed that the combination seemed tolerable with cancer predisposition (20%–30% lifetime cancer risk).76
manageable side effects.53 Larger clinical trials are ongoing Somatic ATM alterations (mutations or deletions) are ob-
to assess benefit. served in a variety of cancers (including approximately 3%
Combining With Epigenetic Modulators of ovarian cancers, 6% of cervical cancers, and 5% of
uterine carcinosarcomas) and can be exploited by existing
Another emerging combination that is actively under in- therapeutic modalities (e.g., PARPIs, radiation therapy,
vestigation in early-phase clinical trials are PARPIs with DSB-inducing chemotherapy agents) or emerging targeted
epigenetic or chromatin modifiers. There is preclinical ev- therapies (e.g., ATR/WEE1/CHK1 inhibitors) that exhibit
idence that combined inhibition of DNA methyltransferase synthetic lethality to ATM-deficient cells.77,78 Preclinical
inhibitors guadecitabine and PARPI talazoparib lead to studies have demonstrated that inhibition of ATM induces
decreased tumor growth in mouse models and increased radio- and chemo-sensitivity via diverse mechanisms, in-
OS.69 The DNA methyltransferase inhibitor and PARPI cluding abrogation of HRR and NHEJ.79,80 Furthermore,
combination led to increased reactive oxygen species ac- ATM inhibitors exhibit synergistic activity with PARPs and
cumulation and enhanced sensitivity of breast and OC cells ATR inhibitors in a variety of tumor models.81-83 Although
to PARPI in a cAMP/PKA-dependent manner and through development of ATM inhibitors was reported as early as
enhanced PARP trapping,69 regardless of BRCA status. 2004, their clinical development is still lacking compared
Recent evidence also showed the synergistic effect of the with other DDR-targeting agents. Two oral ATM inhibitors
EZH2 inhibitor with PARPI in HR-proficient OC.70 (AZD0156 and the brain-penetrant AZD1390) are currently
In conclusion, several different approaches are currently under investigation in phase I clinical trials (NCT02588105
underway, aiming to expand the repertoire of PARPI benefit and NCT03423628), either alone, or in combination with
by inducing HR deficiency in tumors and limiting emer- chemotherapy, PARPIs, or radiation therapy.84,85 In the
gence of resistance. However, the magnitude of success is phase I study of AZD0156 with olaparib (AToM Study,
dependent on careful rationalization of the choice of the NCT02588105) in patients with advanced malignancies,
combination based on the molecular and clinical attributes doses of AZD0156 at 60 mg orally twice daily (bid) 3 days
of an individual patient during the course of treatment. on/4 days off with olaparib 200 mg orally bid continuously
were well tolerated, and pharmacokinetic evaluation in-
BEYOND PARP INHIBITORS: TARGETING DNA DAMAGE dicated that AZD0156 exposure was consistent with efficacy
CHECKPOINTS IN OVARIAN CANCER in vitro.85 Two confirmed PRs (one in a patient with germline
The DDR signaling network preserves genomic integrity via BRCA2 mutation and one in a patient with unknown tumor
the coordinated regulation of cell cycle progression and genetics) were reported, whereas one patient with a tumor
DNA repair.9,60 This is orchestrated by the DNA damage with somatic BRCA2 deletion exhibited stable disease for 18
checkpoint kinases ATM and ATR, which recognize DNA months. Hematologic toxicity, consistent with the mecha-
damage and phosphorylate key effector proteins to induce nism of action of both agents, was observed in higher doses
cell cycle arrest and facilitate DNA repair.71,72 Because of of AZD0156 and represents the treatment-limited toxicity of

AZD0156/olaparib. Although there are currently no clinical associated with subsequent (daughter) G1 arrest and cell
trials of ATM inhibitors specifically for patients with OC or death (i.e., cells with high replication stress treated with ATR
other gynecologic malignancies, exploration of ATM in- inhibitors enter mitosis and form DNA lesions detected as
hibition in this setting is warranted. 53BP1 nuclear bodies and arrest in G1 phase in the
subsequent cell cycle), WEE1 inhibition is more potent,
Targeting the ATR-CHK1-WEE1 DNA Damage Checkpoint inducing cell cycle arrest in the same cell cycle, with cells
Although ATM/CHK2 signaling is triggered in response to undergoing arrest and death in the S phase before entering
DSB formation, ATR is activated by single-stranded DNA mitosis, presumably due to replication catastrophe.96
(ssDNA) that forms during replication stress or as an in- All of the aforementioned mechanisms of replication stress
termediate during DNA repair via HRR and NER.71 Repli- are highly prevalent in HGSOC (Fig. 2).77,98,99 Specifically,
cation stress is defined as the slowing or stalling of replication large-scale genomic studies demonstrated that HGSOC
fork progression during DNA synthesis.71,74,86 Slowing or exhibit: (1) near universal loss of the G1/S checkpoint (via
stalling of the replication forks leads to formation of ssDNA, deleterious TP53 mutations); (2) premature S-phase entry
which activates ATR kinase signaling that initiates a highly due to CCNE1 amplification, RB1 loss, or CDKN2A mRNA
sophisticated replication stress response that involves cell downregulation; (3) oncogenic driver activation via ampli-
cycle arrest, stabilization of replication forks, suppression fication of the MYC oncogene or NF1 loss; and (4) DNA
of dormant origin firing (which helps prevent unscheduled repair deficiencies, mainly due to HRR alterations and less
DNA synthesis), and activation of DNA repair.74,87 Activa- commonly due to NER alterations.77,98,99 Because of the
tion of ATR triggers the intra-S phase and G2 checkpoints prevalence of increased replication stress in this histol-
via phosphorylation of CHK1,88 which, in turn, inactivates ogy, inhibition of ATR/CHK1/WEE1 signaling might be an
(by phosphorylation) CDC25 phosphatases CDC25A and effective strategy against these tumors. Accordingly, in
CDC25C to inhibit cell cycle progression through the co- an open-label, single-center, phase II study in BRCAwt
ordinate suppression of CDK2 and CDK1 (CDK2 and CDK1 HGSOC,99 the CHK1 inhibitor prexasertib, administered
facilitate entry into the S and M phases of the cell cycle, intravenously at 105 mg/m2 as monotherapy every 14 days,
respectively).71,74,89,90 Similarly, WEE1 kinase also inhibits exhibited an objective response rate of 33% (8 of 24 pa-
both CDK1 and CDK2 (both via inhibitory phosphorylation) tients, all PRs; Table 4). Four of the eight PRs involved
to prevent mitotic entry and replication initiation in the S tumors that exhibited both CCNE1 mRNA upregulation and
phase.88,91 Importantly, CHK1 has been reported to phos- CCNE1 amplification or copy gain. Prexasertib is currently
phorylate and activate the kinase activity of WEE1, suggesting being evaluated in a larger multicenter study in ovarian
that WEE1 kinase is downstream of ATR-CHK1, although this cancer (NCT03414047). In addition, the WEE1 inhibitor
may be context-specific.92,93 Finally, ATR facilitates DNA AZD1775 as monotherapy was recently shown to have
repair via phosphorylation of several proteins involved in HRR promising activity in uterine high-grade serous cancer
and Fanconi anemia/interstrand crosslink repair, including (objective response rate, 29%; Table 4), another tumor type
BRCA2, RAD51, FANCA, FANCI, FANCD2, and FANCE.74,87 that, like HGSOC, is enriched with genomic alterations that
Sources of replication stress that may lead to activation of lead to increased replication stress. ATR inhibitor mono-
ATR and initiation of the replication stress response include therapy has not been evaluated specifically in OC or other
loss of the G1/S checkpoint, premature entry into the S gynecologic malignancies but has produced objective re-
phase, oncogenic drive, and DNA repair deficiencies (Fig. sponses in ATM-mutated and/or ATM-deficient tumors, which
2).10,71,89 Cancer cells with high replication stress depend on is consistent with the preclinical data that indicate a synthetic
ATR and the replication stress response for their survival. lethal interaction between ATM and ATR inhibition.83,90
This cancer-specific dependency can be exploited thera- ATR inhibitors have been shown to sensitize OC cells to
peutically by pharmacologic inhibition of ATR, CHK1, or multiple genotoxic chemotherapy agents used routinely in
WEE1. Accordingly, ATR/CHK1/WEE1 inhibitors exhibit this disease, including platinum agents, topoisomerase-I
strong antitumor activity in tumor models with high repli- inhibitors, gemcitabine, and PARPIs.100-103 Sensitization
cation stress, including models with MYC overexpression, to these genotoxic agents has also been reported with WEE1
CDKN2A deletion, TP53 inactivation, CCNE1 amplification, inhibitors.95,104-107 All these agents induce replication stress
and DNA repair deficiencies (e.g., BRCA1/2- and ATM- and dependency on ATR via different mechanisms that
mutated models).90,94-96 It is important to underscore that culminate in the slowing and/or stalling of the replication of
although it has been proposed that WEE1 inhibitors and fork and the inhibition of DNA replication. Synergism with
ATR inhibitors behave similarly, emerging data in lymphoma platinum is particularly potent, which can be explained by
models using the ATR inhibitor AZD6738 and WEE1 in- the additional role of ATR in the activation of the Faconi
hibitor AZD1775 indicate that their mechanisms of action anemia and/or interstrand crosslink DNA repair pathway
may be different.96 Specifically, although ATR inhibition is that is responsible for repair of the DNA interstrand

FIGURE 2. HGSOCs Exhibit High Replication Stress and Depend on ATR Kinase Activation and the Replication Stress Response
(A) Various mechanisms of replication stress lead to slowing/stalling of the replication fork, formation of single strand DNA that induces activation of ATR
and the replication stress response. (B) Prevalence of various mechanisms of high replication stress in high-grade serous ovarian cancers (HGSOCs) of
The Cancer Genome Atlas77,97 highlighting that most HGSOCs exhibit at least two mechanisms of increased replication stress. Each column represents one
HGSOC (n = 316).
Abbreviation: FA/ICL, Fanconi anemia/interstrand crosslink
crosslinks induced by platinum.74,87 It is important to un- inhibitor AZD1775 and one on the ATR inhibitor M6620,
derscore that ATR inhibitors also sensitize BRCA1/2- have suggested that addition of these agents to standard
mutated cells to platinum, topoisomerase-I inhibitors, and chemotherapy improves PFS in OC. Furthermore, one
PARPIs, which is beyond the potent sensitization already phase II study of AZD1775 and carboplatin in TP53-mutated,
caused by their defective HRR. 100 Preclinical studies platinum-resistant and/or refractory OC showed a promising
have also demonstrated that ATR inhibitors may reverse objective response rate of 43%. These results, combined with
resistance PARPIs in BRCA1/2-mutated models by addressing the prexasertib and AZD1775 monotherapy data in high-
the two major mechanisms of resistance in this setting: (1) grade serous ovarian and uterine cancers as previously
restoration of HRR; and (2) replication fork stabilization.108,109 discussed, support further development of these agents in
Other studies have also indicated synergism of ATR these tumors with high replication stress. Although preclinical
inhibitors with WEE1 inhibitors 96,110,111 and immune data suggest that WEE1 inhibition as monotherapy may be
checkpoint inhibitors.112 However, unlike ATR and WEE1 more potent than ATR inhibition as monotherapy (as pre-
inhibitors that sensitize to a broad spectrum of genotoxic viously discussed), combinations of chemotherapy with ATR
agents, available data suggest that CHK1 inhibitors might inhibitors appear to better tolerated than combinations with
not uniformly sensitize to genotoxic drugs, which suggests WEE1 inhibitors. Taken together, ATR inhibitor development
different roles of ATR and CHK1 in the response to geno- may be more likely to succeed using combinatorial ap-
toxic chemotherapy.95,99,113 proaches, whereas WEE1 inhibition may be more likely to
Table 4 summarizes the results of key phase II studies of succeed as monotherapy. A major challenge for further
ATR/WEE1/CHK1 inhibitors, alone or in combination in development of these agents is the absence of predictive
patients with OC or other gynecologic malignancies.114-118 biomarkers of response. Although preclinical data indicate
Three randomized phase II studies, two on the WEE1 that CCNE1 amplification, MYC amplification, CDKN2A loss,

e126
TABLE 4. Key Phase II Studies of ATR/CHK1/WEE1 Inhibitors in Ovarian Cancer and Other Gynecologic Malignancies
Study Primary
Identifier Agents Target Design Patients/Accrual Endpoint Results Comments
NCT02151292 Gemcitabine/ WEE1 Multicenter, randomized, Platinum-resistant/refractory PFS Median PFS: In Gem/AZD1775:
AZD1775 phase II, double-blind (2:1 HGSOC, unlimited Gem/AZD1775: Thrombocytopenia G 3:
(Adavosertib) randomization) previous lines 4.6 months; 31%;
N=124 Gem/Placebo: 3.0 months; Neutropenia G 3: 62%
vs. Gemcitabine/
HR, 0.56
placebo
NCT01164995 Carboplatin/ WEE1 Multicenter, randomized, TP53-mutated ovarian PFS (volumetric Median PFS: Carboplatin/paclitaxel/
paclitaxel/ phase II, double-blind (1:1 cancer RECIST 1.1.) Carboplatin/paclitaxel/ AZD1775 G 3 AEs: 78%;
AZD1775 randomization) N = 121 AZD1775: 34.14 weeks; Carbo/paclitaxel/placebo
Carboplatin/paclitaxel/ G 3 AEs: 65%
vs. Carboplatin/
placebo: 31.86 weeks;
paclitaxel/

HR, 0.63
placebo
NCT02595892 Gemcitabine/ ATR Multicenter, randomized, Platinum-resistant HGSOC, PFS Median PFS: In Gem/M6620:
M6620 phase II, open label (1:1 unlimited previous lines Gemcitabine/M6620: 22.9 Thrombocytopenia G 3:
(Berzosertib) randomization) but no more than 1 wks; 24%
previous regimen in Gemcitabine: 14.7 wks;
vs. Gemcitabine In Gem alone:
platinum-resistant setting HR, 0.61
Thrombocytopenia G 3:
N = 70
6%
NCT01164995 Carboplatin/ WEE1 Phase II, single arm TP53-mutated ovarian ORR ORR: 43%, 9 patients (1 CR + Thrombocytopenia G 3:
AZD1775 cancer, refractory or 8 PR) 48%; Neutropenia G 3:

resistant (, 3 months) to 37%
first-line platinum
N = 21 (evaluable)
NCT02203513 Prexasertib CHK1 Phase II, single arm, single HGS or HG endometrioid ORR ORR: 33%, 8 patients, all Thrombocytopenia G 3:

center ovarian cancer, germline PRs, 4 of 8 PRs in tumors 25%; Neutropenia G 3:
BRCAwt with CCNE1 overexpression 93%;

N = 24 (evaluable) Febrile neutropenia G 3:
7%
NCT03668340 AZD1775 WEE1 Phase II, single arm Recurrent high-grade serous ORR and PFS ORR: 29% Neutropenia G 3: 32%
uterine cancer
PFS: 59%
N = 34 (evaluable)
Abbreviations: AE, adverse event; BRCAwt, wild-type BRCA; CR, complete response; HG, high grade; HGSOC, high-grade serous ovarian cancer; HGS, high-grade serous; HR, hazard ratio; ORR,
overall response rate; PFS, progression-free survival, PR, partial response.
TP53 mutations, ATM deficiency, and HRR deficiency of patients that would be prime candidates for these
may predict response to ATR/CHK1/WEE1 inhibitors, therapies.
clinical studies have not yet confirmed these (except for
ATM deficiency being a predictor for response to ATR ACKNOWLEDGMENT
inhibition). Correlative work from these and future studies Panagiotis Konstantinopoulos, MD, and Stephanie Lheur-
will hopefully identify biomarkers that can guide selection eux, MD, are co-first authors.

1
Dana Farber Cancer Institute, Boston, MA Kathleen Moore, MD, 800 NE 10th Street, Oklahoma City, OK 73104;
2
Princess Margaret Cancer Center, Toronto, Ontario, Canada email: [email protected].
3
Stephenson Cancer Center at the University of Oklahoma Health
Sciences Center, Oklahoma City, OK
EDBK_288015.
REFERENCES
1. Ledermann JA, Raja FA, Fotopoulou C, et al; ESMO Guidelines Working Group. Newly diagnosed and relapsed epithelial ovarian carcinoma: ESMO Clinical
Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2013;24(Suppl 6):vi24-vi32.
2. Wilson MK, Pujade-Lauraine E, Aoki D, et al; Participants of the Fifth Ovarian Cancer Consensus Conference. Fifth Ovarian Cancer Consensus Conference of the
Gynecologic Cancer InterGroup: recurrent disease. Ann Oncol. 2017;28:727-732.
3. Clamp AR, James EC, McNeish IA, et al. Weekly dose-dense chemotherapy in first-line epithelial ovarian, fallopian tube, or primary peritoneal carcinoma
treatment (ICON8): primary progression free survival analysis results from a GCIG phase 3 randomised controlled trial. Lancet. 2019;394:2084-2095.
4. Burger RA, Brady MF, Bookman MA, et al; Gynecologic Oncology Group. Incorporation of bevacizumab in the primary treatment of ovarian cancer. N Engl
J Med. 2011;365:2473-2483.
5. Perren TJ, Swart AM, Pfisterer J, et al; ICON7 Investigators. A phase 3 trial of bevacizumab in ovarian cancer. N Engl J Med. 2011;365:2484-2496.
6. Vaughan S, Coward JI, Bast RC Jr., et al. Rethinking ovarian cancer: recommendations for improving outcomes. Nat Rev Cancer. 2011;11:719-725.
7. Walsh CS. Two decades beyond BRCA1/2: homologous recombination, hereditary cancer risk and a target for ovarian cancer therapy. Gynecol Oncol. 2015;
137:343-350.
8. Konecny GE, Kristeleit RS. PARP inhibitors for BRCA1/2-mutated and sporadic ovarian cancer: current practice and future directions. Br J Cancer. 2016;
115:1157-1173.
9. O’Connor MJ. Targeting the DNA damage response in cancer. Mol Cell. 2015;60:547-560.
10. Konstantinopoulos PA, Ceccaldi R, Shapiro GI, et al. Homologous recombination deficiency: exploiting the fundamental vulnerability of ovarian cancer. Cancer
Discov. 2015;5:1137-1154.
11. Moore K, Colombo N, Scambia G, et al. Maintenance olaparib in patients with newly diagnosed advanced ovarian cancer. N Engl J Med. 2018;379:2495-2505.
12. Mathews CMK, Colombo N, Scambia G, et al. Maintenance olaparib after platinum-based chemotherapy in patients (pts) with newly diagnosed advanced
ovarian cancer (OC) and a BRCA mutation (BRCAm): Efficacy by surgical and tumor status in the Phase III SOLO1 trial. J Clin Oncol. 2019;37 (suppl; abstr
5541).
13. González-Martı́n A, Pothuri B, Vergote I, et al; PRIMA/ENGOT-OV26/GOG-3012 Investigators. Niraparib in patients with newly diagnosed advanced ovarian
14. Ray-Coquard I, Pautier P, Pignata S, et al; PAOLA-1 Investigators. Olaparib plus bevacizumab as first-line maintenance in ovarian cancer. N Engl J Med. 2019;
381:2416-2428.
15. Coleman RL, Fleming GF, Brady MF, et al. Veliparib with first-line chemotherapy and as maintenance therapy in ovarian cancer. N Engl J Med. 2019;
381:2403-2415.
16. Vergote IMK, Hettle R, Rohodes K, et al. Population adjusted indirect comparison of the SOLO-1 and PAOLA-1/ENNGOT-ov25 studies of olaparib with or without
bevacizumab, bevacizumab alone and placebo in the maintenance treatment of women with newly diagnosed stage III/IV ovarian cancer with BRCA mutation.
Gynecol Oncol. 2020, In press.
17. Swisher EKS, Birrer MJ, Levine DA, et al. Exploring the relationship between homologous recombination score and progression free survival in BRCA wildtype
ovarian carcinoma: analysis of velparib plus carboplatin/paclitaxel in the phase 3 VELIA/GOG 3005 study. Gynecol Oncol. 2020, In press.
18. Ledermann J, Harter P, Gourley C, et al. Olaparib maintenance therapy in patients with platinum-sensitive relapsed serous ovarian cancer: a preplanned
retrospective analysis of outcomes by BRCA status in a randomised phase 2 trial. Lancet Oncol. 2014;15:852-861.

19. Pujade-Lauraine E, Ledermann JA, Selle F, et al; SOLO2/ENGOT-Ov21 Investigators. Olaparib tablets as maintenance therapy in patients with platinum-
sensitive, relapsed ovarian cancer and a BRCA1/2 mutation (SOLO2/ENGOT-Ov21): a double-blind, randomised, placebo-controlled, phase 3 trial. Lancet
Oncol. 2017;18:1274-1284.
N Engl J Med. 2016;375:2154-2164.
21. Coleman RL, Oza AM, Lorusso D, et al; ARIEL3 Investigators. Rucaparib maintenance treatment for recurrent ovarian carcinoma after response to platinum
therapy (ARIEL3): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet. 2017;390:1949-1961.
22. Kaufman B, Shapira-Frommer R, Schmutzler RK, et al. Olaparib monotherapy in patients with advanced cancer and a germline BRCA1/2 mutation. J Clin
Oncol. 2015;33:244-250.
23. Swisher EM, Lin KK, Oza AM, et al. Rucaparib in relapsed, platinum-sensitive high-grade ovarian carcinoma (ARIEL2 Part 1): an international, multicentre,
24. Penson RT, Valencia RV, Cibula D, et al. Olaparib versus nonplatinum chemotherapy in patients with platinum-sensitive relapsed ovarian cancer and a germline
BRCA1/2 mutation (SOLO3): a randomized phase III trial. J Clin Oncol. Epub 2020 February 19.
25. Moore KN, Secord AA, Geller MA, et al. Niraparib monotherapy for late-line treatment of ovarian cancer (QUADRA): a multicentre, open-label, single-arm, phase
2 trial. Lancet Oncol. 2019;20:636-648.
26. Lord CJ, Ashworth A. PARP inhibitors: the first synthetic lethal targeted therapy. Science. 2017;355:1152-1158.
27. Pilié PG, Gay CM, Byers LA, et al. PARP inhibitors: extending benefit beyond BRCA-mutant cancers. Clin Cancer Res. 2019;25:3759-3771.
28. Wakefield MJ, Nesic K, Kondrashova O, et al. Diverse mechanisms of PARP inhibitor resistance in ovarian cancer. Biochim Biophys Acta Rev Cancer. 2019;
1872:188307.
29. Noordermeer SM, van Attikum H. PARP inhibitor resistance: a tug-of-war in BRCA-mutated cells. Trends Cell Biol. 2019;29:820-834.
30. Christie EL, Pattnaik S, Beach J, et al. Multiple ABCB1 transcriptional fusions in drug resistant high-grade serous ovarian and breast cancer. Nat Commun.
2019;10:1295.
31. Pettitt SJ, Krastev DB, Brandsma I, et al. Genome-wide and high-density CRISPR-Cas9 screens identify point mutations in PARP1 causing PARP inhibitor
resistance. Nat Commun. 2018;9:1849.
32. Ding X, Ray Chaudhuri A, Callen E, et al. Synthetic viability by BRCA2 and PARP1/ARTD1 deficiencies. Nat Commun. 2016;7:12425.
33. Gogola E, Duarte AA, de Ruiter JR, et al. Selective loss of PARG restores PARylation and counteracts PARP inhibitor-mediated synthetic lethality. Cancer Cell.
2018;33:1078-1093.e12.
34. Liao H, Ji F, Helleday T, et al. Mechanisms for stalled replication fork stabilization: new targets for synthetic lethality strategies in cancer treatments. EMBO Rep.
2018;19:e46263.
35. Lin KK, Harrell MI, Oza AM, et al. BRCA reversion mutations in circulating tumor DNA predict primary and acquired resistance to the PARP inhibitor rucaparib in
high-grade ovarian carcinoma. Cancer Discov. 2019;9:210-219.
36. Christie EL, Fereday S, Doig K, et al. Reversion of BRCA1/2 germline mutations detected in circulating tumor DNA from patients with high-grade serous ovarian
37. Edwards SL, Brough R, Lord CJ, et al. Resistance to therapy caused by intragenic deletion in BRCA2. Nature. 2008;451:1111-1115.
38. Kondrashova O, Nguyen M, Shield-Artin K, et al; AOCS Study Group. Secondary somatic mutations restoring RAD51C and RAD51D associated with acquired
resistance to the PARP inhibitor rucaparib in high-grade ovarian carcinoma. Cancer Discov. 2017;7:984-998.
39. Barber LJ, Sandhu S, Chen L, et al. Secondary mutations in BRCA2 associated with clinical resistance to a PARP inhibitor. J Pathol. 2013;229:422-429.
40. Goodall J, Mateo J, Yuan W, et al; TOPARP-A Investigators. Circulating cell-free DNA to guide prostate cancer treatment with PARP inhibition. Cancer Discov.
2017;7:1006-1017.
41. Wang Y, Bernhardy AJ, Nacson J, et al. BRCA1 intronic Alu elements drive gene rearrangements and PARP inhibitor resistance. Nat Commun. 2019;10:5661.
42. ter Brugge P, Kristel P, van der Burg E, et al. Mechanisms of therapy resistance in patient-derived xenograft models of BRCA1-deficient breast cancer. J Natl
Cancer Inst. 2016;108:djw148.
43. Escribano-Dı́az C, Orthwein A, Fradet-Turcotte A, et al. A cell cycle-dependent regulatory circuit composed of 53BP1-RIF1 and BRCA1-CtIP controls DNA repair
pathway choice. Mol Cell. 2013;49:872-883.
44. Xu G, Chapman JR, Brandsma I, et al. REV7 counteracts DNA double-strand break resection and affects PARP inhibition. Nature. 2015;521:541-544.
45. Watson ZL, Yamamoto TM, McMellen A, et al. Histone methyltransferases EHMT1 and EHMT2 (GLP/G9A) maintain PARP inhibitor resistance in high-grade
serous ovarian carcinoma. Clin Epigenetics. 2019;11:165.
46. Fong PC, Yap TA, Boss DS, et al. Poly(ADP)-ribose polymerase inhibition: frequent durable responses in BRCA carrier ovarian cancer correlating with platinum-
free interval. J Clin Oncol. 2010;28:2512-2519.
47. Kelland L. The resurgence of platinum-based cancer chemotherapy. Nat Rev Cancer. 2007;7:573-584.
48. Nakayama N, Nakayama K, Shamima Y. Gene amplification CCNE1 is related to poor survival and potential therapeutic target in ovarian cancer. Cancer. 2010;
116:2621-2634.

49. Wiedemeyer WR, Beach JA, Karlan BY. Reversing platinum resistance in high-grade serous ovarian carcinoma: targeting BRCA and the homologous re-
combination system. Front Oncol. 2014;4:34.
50. Quigley D, Alumkal JJ, Wyatt AW, et al. Analysis of circulating cell-free DNA identifies multiclonal heterogeneity of BRCA2 reversion mutations associated with
resistance to PARP inhibitors. Cancer Discov. 2017;7:999-1005.
51. Lheureux S, Oaknin A, Garg S. Evolve: A Post PARP Inhibitor Clinical Translational Phase II Trial of Cediranib-Olaparib in Ovarian Cancer—A Princess Margaret
Consortium – GCIG Phase II Trial. J Clin Oncol. 2019;37:15s (suppl; abstr 5521).
52. Lheureux S, Mirza M, Coleman R. The DNA repair pathway as a target for novel drugs in gynecologic cancers. J Clin Oncol. 2019;37:2449-2459.
53. Veneris JT, Matulonis UA, Liu JF, et al. Choosing wisely: selecting PARP inhibitor combinations to promote anti-tumor immune responses beyond BRCA
mutations. Gynecol Oncol. 2019;156:488-497.
54. Murai J, Feng Y, Yu GK, et al. Resistance to PARP inhibitors by SLFN11 inactivation can be overcome by ATR inhibition. Oncotarget. 2016;7:76534-76550.
55. Kim H, George E, Ragland R, et al. Targeting the ATR/CHK1 axis with PARP inhibition results in tumor regression in BRCA-mutant ovarian cancer models. Clin
Cancer Res. 2017;23:3097-3108.
56. Lallo A, Frese KK, Morrow CJ, et al. The combination of the PARP inhibitor olaparib and the Wee1 inhibitor AZD1775 as a new therapeutic option for small cell
lung cancer. Clin Cancer Res. 2018;24:5153-5164.
57. Haynes B, Murai J, Lee J-M. Restored replication fork stabilization, a mechanism of PARP inhibitor resistance, can be overcome by cell cycle checkpoint
inhibition. Cancer Treat Rev. 2018;71:1-7.
59. Do KT, Hill SJ, Kochupurakkal B, et al. Phase I combination study of the CHK1 inhibitor prexasertib (LY2606368) and olaparib in patients with high-grade serous
ovarian cancer and other advanced solid tumors. Cancer Res. 2019;79 (13 Suppl; abstr CT232).
60. Pilié PG, Tang C, Mills GB, et al. State-of-the-art strategies for targeting the DNA damage response in cancer. Nat Rev Clin Oncol. 2019;16:81-104.
61. Kaplan AR, Gueble SE, Liu Y, et al. Cediranib suppresses homology-directed DNA repair through down-regulation of BRCA1/2 and RAD51. Sci Transl Med.
2019;11:eaav4508.
62. Mirza MR, Åvall Lundqvist E, Birrer MJ, et al; AVANOVA Investigators. Niraparib plus bevacizumab versus niraparib alone for platinum-sensitive recurrent
ovarian cancer (NSGO-AVANOVA2/ENGOT-ov24): a randomised, phase 2, superiority trial. Lancet Oncol. 2019;20:1409-1419.
63. Migden MR, Rischin D, Schmults CD, et al. PD-1 blockade with cemiplimab in advanced cutaneous squamous-cell carcinoma. N Engl J Med. 2018;
379:341-351.
65. Strickland KC, Howitt BE, Shukla SA, et al. Association and prognostic significance of BRCA1/2-mutation status with neoantigen load, number of tumor-
infiltrating lymphocytes and expression of PD-1/PD-L1 in high grade serous ovarian cancer. Oncotarget. 2016;7:13587-13598.
66. Sen T, Rodriguez BL, Chen L, et al. Targeting DNA damage response promotes antitumor immunity through STING-mediated T-cell activation in small cell lung
cancer. Cancer Discov. 2019;9:646-661.
67. Jiao S, Xia W, Yamaguchi H, et al. PARP inhibitor upregulates PD-L1 expression and enhances cancer-associated immunosuppression. Clin Cancer Res. 2017;
23:3711-3720.
68. Chabanon RM, Muirhead G, Krastev DB, et al. PARP inhibition enhances tumor cell-intrinsic immunity in ERCC1-deficient non-small cell lung cancer. J Clin
Invest. 2019;129:1211-1228.
69. Pulliam N, Fang F, Ozes AR, et al. An effective epigenetic-PARP inhibitor combination therapy for breast and ovarian cancers independent of BRCA mutations.
Clin Cancer Res. 2018;24:3163-3175.
70. Karakashev S, Fukumoto T, Zhao B, et al. EZH2 inhibition sensitizes CARM1-high, homologous recombination proficient ovarian cancers to PARP inhibition.
Cancer Cell. 2020;37:157-167.e6.
71. Flynn RL, Zou L. ATR: a master conductor of cellular responses to DNA replication stress. Trends Biochem Sci. 2011;36:133-140.
72. Jin MH, Oh DY. ATM in DNA repair in cancer. Pharmacol Ther. 2019;203:107391.
73. Smith J, Tho LM, Xu N, et al. The ATM-Chk2 and ATR-Chk1 pathways in DNA damage signaling and cancer. Adv Cancer Res. 2010;108:73-112.
74. Curtin NJ. DNA repair dysregulation from cancer driver to therapeutic target. Nat Rev Cancer. 2012;12:801-817.
75. Shrivastav M, De Haro LP, Nickoloff JA. Regulation of DNA double-strand break repair pathway choice. Cell Res. 2008;18:134-147.
76. Choi M, Kipps T, Kurzrock R. ATM mutations in cancer: therapeutic implications. Mol Cancer Ther. 2016;15:1781-1791.
77. Cancer Genome Atlas Research Network. Integrated genomic analyses of ovarian carcinoma. Nature. 2011;474:609-615.
78. Weber AM, Drobnitzky N, Devery AM, et al. Phenotypic consequences of somatic mutations in the ataxia-telangiectasia mutated gene in non-small cell lung
cancer. Oncotarget. 2016;7:60807-60822.
79. Batey MA, Zhao Y, Kyle S, et al. Preclinical evaluation of a novel ATM inhibitor, KU59403, in vitro and in vivo in p53 functional and dysfunctional models of
human cancer. Mol Cancer Ther. 2013;12:959-967.
80. Hickson I, Zhao Y, Richardson CJ, et al. Identification and characterization of a novel and specific inhibitor of the ataxia-telangiectasia mutated kinase ATM.
Cancer Res. 2004;64:9152-9159.

81. Mak JPY, Ma HT, Poon RYC. Synergism between ATM and PARP1 inhibition involves DNA damage and abrogating the G2 DNA damage checkpoint. Mol
Cancer Ther. 2020;19:123-134.
82. Mei L, Zhang J, He K, Zhang J. Ataxia telangiectasia and Rad3-related inhibitors and cancer therapy: where we stand. J Hematol Oncol. 2019;12:43.
83. Schmitt A, Knittel G, Welcker D, et al. ATM deficiency is associated with sensitivity to PARP1- and ATR inhibitors in lung adenocarcinoma. Cancer Res. 2017;
77:3040-3056.
84. Durant ST, Zheng L, Wang Y, et al. The brain-penetrant clinical ATM inhibitor AZD1390 radiosensitizes and improves survival of preclinical brain tumor models.
Sci Adv. 2018;4:eaat1719.
85. Riches LC, Trinidad AG, Hughes G, et al. Pharmacology of the ATM inhibitor AZD0156: potentiation of irradiation and olaparib responses pre-clinically. Mol
Cancer Ther. 2020;19:13-25.
86. Bartkova J, Horejsı́ Z, Koed K, et al. DNA damage response as a candidate anti-cancer barrier in early human tumorigenesis. Nature. 2005;434:864-870.
87. Lee KY, Chung KY, Koo HS. The involvement of FANCM, FANCI, and checkpoint proteins in the interstrand DNA crosslink repair pathway is conserved in C.
elegans. DNA Repair (Amst). 2010;9:374-382.
88. Patil M, Pabla N, Dong Z. Checkpoint kinase 1 in DNA damage response and cell cycle regulation. Cell Mol Life Sci. 2013;70:4009-40021.
89. Gorgoulis VG, Vassiliou LV, Karakaidos P, et al. Activation of the DNA damage checkpoint and genomic instability in human precancerous lesions. Nature. 2005;
434:907-913.
91. Enders GH. Gauchos and ochos: a Wee1-Cdk tango regulating mitotic entry. Cell Div. 2010;5:12.
92. Lee J, Kumagai A, Dunphy WG. Positive regulation of Wee1 by Chk1 and 14-3-3 proteins. Mol Biol Cell. 2001;12:551-563.
93. O’Connell MJ, Raleigh JM, Verkade HM, Nurse P. Chk1 is a wee1 kinase in the G2 DNA damage checkpoint inhibiting cdc2 by Y15 phosphorylation. EMBO J.
1997;16:545-554.
94. Hill SJ, Decker B, Roberts EA, Horowitz NS, et al. Prediction of DNA repair inhibitor response in short-term patient-derived ovarian cancer organoids. Cancer
Discovery. 2018;8:1404-1421.
95. Parmar K, Kochupurakkal BS, Lazaro JB, et al. The CHK1 inhibitor prexasertib exhibits monotherapy activity in high-grade serous ovarian cancer models and
sensitizes to PARP inhibition. Clin Cancer Res. 2019;25:6127-6140.
96. Young LA, O’Connor LO, de Renty C, et al. Differential activity of ATR and WEE1 inhibitors in a highly sensitive subpopulation of DLBCL linked to replication
stress. Cancer Res. 2019;79:3762-3775.
97. Cerami E, Gao J, Dogrusoz U, et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discovery.
2012;2:401-404.
98. Ceccaldi R, O’Connor KW, Mouw KW, et al. A unique subset of epithelial ovarian cancers with platinum sensitivity and PARP inhibitor resistance. Cancer Res.
2015;75:628-634.
99. Mouw KW, D’Andrea AD, Konstantinopoulos PA. Nucleotide excision repair (NER) alterations as evolving biomarkers and therapeutic targets in epithelial
cancers. Oncoscience. 2015;2:942-943.
100. Huntoon CJ, Flatten KS, Wahner Hendrickson AE, et al. ATR inhibition broadly sensitizes ovarian cancer cells to chemotherapy independent of BRCA status.
Cancer Res. 2013;73:3683-3691.
101. Josse R, Martin SE, Guha R, et al. ATR inhibitors VE-821 and VX-970 sensitize cancer cells to topoisomerase i inhibitors by disabling DNA replication initiation
and fork elongation responses. Cancer Res. 2014;74:6968-6979.
102. Thomas A, Redon CE, Sciuto L, et al. Phase I study of ATR inhibitor M6620 in combination with topotecan in patients with advanced solid tumors. J Clin Oncol.
2018;36:1594-1602.
103. Zou L. Ataxia telangiectasia-mutated and Rad3-related inhibition and topoisomerase I trapping create a synthetic lethality in cancer cells. J Clin Oncol. 2018;
36:1628-1630.
104. Angius G, Tomao S, Stati V, et al. Prexasertib, a checkpoint kinase inhibitor: from preclinical data to clinical development. Cancer Chemother Pharmacol. 2020;
85:9-20.
105. Lowery CD, Dowless M, Renschler M,B et al. Broad spectrum activity of the checkpoint kinase 1 inhibitor prexasertib as a single agent or chemopotentiator
across a range of preclinical pediatric tumor models. Clin Cancer Res. 2019;25:2278-2289.
106. Mani C, Jonnalagadda S, Lingareddy J, et al. Prexasertib treatment induces homologous recombination deficiency and synergizes with olaparib in triple-
negative breast cancer cells. Breast Cancer Res. 2019;21:104.
107. Morimoto Y, Takada K, Takeuchi O, et al. Prexasertib increases the sensitivity of pancreatic cancer cells to gemcitabine and S-1. Oncol Rep. 2020;43:689-699.
108. D’Andrea AD. Mechanisms of PARP inhibitor sensitivity and resistance. DNA Repair (Amst). 2018;71:172-176.
110. Jin J, Fang H, Yang F, et al. Combined inhibition of ATR and WEE1 as a novel therapeutic strategy in triple-negative breast cancer. Neoplasia. 2018;20:478-488.
111. Qi W, Xu X, Wang M, et al. Inhibition of Wee1 sensitizes AML cells to ATR inhibitor VE-822-induced DNA damage and apoptosis. Biochem Pharmacol. 2019;
164:273-282.

112. Mouw KW, Konstantinopoulos PA. From checkpoint to checkpoint: DNA damage ATR/Chk1 checkpoint signalling elicits PD-L1 immune checkpoint activation.
Br J Cancer. 2018;118:933-935.
113. Montano R, Chung I, Garner KM, et al. Preclinical development of the novel Chk1 inhibitor SCH900776 in combination with DNA-damaging agents and
antimetabolites. Mol Cancer Ther. 2012;11:427-438.
114. Lee JM, Nair J, Zimmer A, et al. Prexasertib, a cell cycle checkpoint kinase 1 and 2 inhibitor, in BRCA wild-type recurrent high-grade serous ovarian cancer:
a first-in-class proof-of-concept phase 2 study. Lancet Oncol. 2018;19:207-215.
115. Lheureux S, Cabanero M, Cristea MC, et al. A randomized double-blind placebo-controlled phase II trial comparing gemcitabine monotherapy to gemcitabine in
combination with adavosertib in women with recurrent, platinum resistant epithelial ovarian cancer: a trial of the Princess Margaret, California, Chicago and
Mayo Phase II Consortia. J Clin Oncol. 2019;37 (suppl; abstr 5518).
116. Oza AM, Weberpals JI, Provencher DM, et al. An international, biomarker-directed, randomized, phase II trial of AZD1775 plus paclitaxel and carboplatin (P/C)
for the treatment of women with platinum-sensitive, TP53-mutant ovarian cancer. J Clin Oncol. 2015;33 (suppl; abstr 5506).
117. Leijen S, van Geel RM, Sonke GS, et al. Phase II study of WEE1 inhibitor AZD1775 plus carboplatin in patients with TP53-mutated ovarian cancer refractory or
resistant to first-line therapy within 3 months. J Clin Oncol. 2016;34:4354-4361.
118. Konstantinopoulos P, Hendrickson A, Penson R, et al. LBA60: Randomized phase II (RP2) study of ATR inhibitor M6620 in combination with gemcitabine
versus gemcitabine alone in platinum-resistant high grade serous ovarian cancer (HGSOC). Ann Oncol. 2019;30 (Suppl 5; abstr v897).

GYNECOLOGIC CANCER
Immune Therapy Opportunities in Ovarian Cancer

Lana E. Kandalaft, PharmD, PhD, MTR1; Kunle Odunsi, MD, PhD2,3; and George Coukos, MD, PhD1
overview
Immunotherapy has emerged as a highly promising approach in the treatment of epithelial ovarian cancer
(EOC). Immune checkpoint blockade (ICB) therapy, PARP inhibitors (PARPis), neoantigen vaccines, and
personalized T-cell therapy have been associated with encouraging clinical activity in a small subset of
patients. To increase the proportion of patients who are likely to derive benefit, it will be important not only to
generate sufficient numbers of antitumor T cells but also to overcome multiple inhibitory networks in the
ovarian tumor microenvironment (TME). Therefore, a major direction is to develop biomarkers that would
predict responsiveness to different types of immunotherapies and allow treatment selection based on the
results. Moreover, such biomarkers would allow rational combination of immunotherapies while minimizing
toxicities. In this review, we provide progress on immune therapies and future directions for maximally
exploiting immune-based strategies for the treatment of ovarian cancer.
INTRODUCTION and mucin protein-3 are implicated in immune

Despite advances in surgery and chemotherapy, the checkpoints (reviewed elsewhere4). The CTLA-4 and
5-year survival rate is only 47% for all EOC types.1 PD-1 receptors have by far generated the most interest,
Standard first-line treatment for patients with EOC and monoclonal antibodies targeting CTLA-4 (binding
consists of cytoreductive surgery and taxane/platinum- to CD80/86) or PD-1 or its ligand (PD-L1) have been
based chemotherapy. Unfortunately, most patients investigated in several clinical trials. The activity of
relapse and almost invariably develop resistance to anti–CTLA-4 was first demonstrated in patients with
platinum agents. Life expectancy for patients whose melanoma,3 with higher toxicity and lower response
disease is refractory or resistant to platinum-based rates than anti–PD-1/PD-L1 monoclonal antibodies,5
therapy does not exceed 1 year, and new treatment which are now approved as monotherapy in melanoma
options are needed to significantly increase their re- and various other tumors.6-8
sponse rate and survival. Patients sensitive to platinum- EOC is considered a potentially immunoreactive tumor
based therapy (when relapse occurs more than because the presence of tumor-infiltrating lymphocytes
6 months after first-line treatment) would also benefit (TILs) correlates with improved clinical outcomes.9-12 An
from new therapies with higher curative rates because immunoreactive molecular subtype with an enrichment
most of them will eventually relapse even after a com- of genes and signaling pathways associated with im-
plete response (CR). The use of PARPis has renewed mune cells has also been identified, and it correlates with
the therapeutic opportunities for patients with EOC,2 longer overall survival (OS).13-15 T cells infiltrating ovarian
and immunotherapy has attracted attention and been tumors appear to have impaired effector functions
granted accelerated approval for the treatment of nu-
and express multiple inhibitory receptors, such as
merous cancer types,3 yet not for EOC. In this article, we
lymphocyte activation gene-3, PD-1, and CTLA-4.16-19
review the clinical development of immunotherapy in
Interferon-γ (IFN-γ) produced by tumor-infiltrating
EOC and highlight the emerging therapeutic opportu-
CD8-positive T cells upregulates PD-L1 expression on
nities for future development.
Author affiliations EOC cells or macrophages, which in turn suppresses the
and support IMMUNE CHECKPOINT BLOCKADE THERAPY effector function of PD-1–positive TILs,20 contributing to
information (if
The blockade of immune checkpoints is of major interest the immunosuppression in the TME.16,17 Because the
applicable) appear
at the end of this in cancer research and a very promising approach to presence of TILs is a prerequisite to response to ICB,21
article. improve antitumoral immunity in patients with EOC. immunoreactive ovarian tumors are expected to respond
Accepted on March Immune checkpoints are critical to maintain self- to PD-1 blockade, and several clinical studies have been
13, 2020 and tolerance and also protect tissues after a response to conducted. In a phase II trial of nivolumab (anti–PD-1),
published at the best overall response rate (ORR) was 15% and
pathogens. However, in the TME they drastically sup-
ascopubs.org on May
14, 2020: DOI https://
press the intensity, quality, and duration of immune disease control rate (DCR) was 45%, with a 10% durable
doi.org/10.1200/ responses. Several molecules such as PD-1, CTLA-4, lym- CR rate for 20 assessable patients with recurrent EOC.6
EDBK_280539 phocyte activation gene-3 (LAG-3), T-cell immunoglobulin, In the phase II KEYNOTE-100 study (NCT02674061) of

Immunotherapy of Ovarian Cancer
transforming growth factor-β (TGF-β) also contribute to the

overall immunosuppression in the ovarian TME by estab-
lishing inhibitory cell–cell contacts with effector T cells.50 A
• Ovarian cancer is an immunologic cancer. higher concentration of immature myeloid cells47 has also
• Response rates with ICB therapy are quite low. been observed in the ovarian TME, and myeloid-derived
• Combinatorial immunotherapies have promise. suppressor cells express high levels of arginase and pro-
duce nitric oxide,48 which inhibit T-cell and natural killer cell
• Cellular therapy has potential in ovarian cancer.
function,51 thereby also contributing to tumor progression.52
The paucity of effective T-cell responses in EOC is also caused
376 patients with advanced recurrent EOC, pembrolizumab by escape mechanisms such as decreased major histo-
(anti–PD-1) as monotherapy resulted in an ORR of only 8%. compatibility complex expression53 and antigenic loss.54 Tu-
Higher PD-L1 expression as measured by a combined pos- mor antigen presentation is also impaired by the fact that the
itive score of at least 10 resulted in a higher ORR (17.1%) than tumor-infiltrating DCs are skewed toward a tolerogenic phe-
a combined positive score of 1 to less than 10 (10.2%) or less notype because of the high levels of immunoregulatory
than 1 (5%).22 It should be noted that the ORRs reported in mediators present in the EOC TME.55-57 Tumor-associated
these studies were lower than what was reported in a meta- macrophages (TAMs) represent another important immune
analysis of patients with EOC displaying intraepithelial TILs.11 cell type in EOC,58,59 and whereas a high concentration of
Nonetheless, these studies showed only modest activity of CD163-positive TAMs is associated with poor prognosis,
PD-1 blockade as a single agent in EOC. a high M1/M2 macrophage ratio predicts favorable out-
One approach to improve the efficacy of PD-1 blockade is comes.58 The polarization of macrophages toward the cancer-
combinatorial ICB, which should, in principle, synergize to supporting M2 phenotype is promoted by cytokines such as
release immune responses and thereby increase biologic and VEGF, IL-6, IL-10, TGF-β, and arachidonic acid34,60-62; ovarian
clinical responses.23 The combination of PD-1/PD-L1 block- cancer cells63,64; and platinum agents.65 Yet TAMs can pro-
ade with CTLA-4 blockade has been shown to be more ef- mote resistance to cisplatin, thereby inducing cellular stem-
ficacious than PD-1/PD-L1 blockade alone in murine models ness properties in cisplatin-sensitive cells.66
of melanoma and EOC,24-26 and ipilimumab plus nivolumab OPPORTUNITIES FOR COMBINATION IMMUNOTHERAPY
has shown some efficacy in metastatic melanoma27 and lung
cancer,8,28 although with greater toxicity than PD-1 blockade Multiple immunosuppressive factors and inhibitory signals
alone.29 The combination of ipilimumab with nivolumab is suppress T-cell activation in the TME, and various points of
being investigated in EOC (NCT03342417, NCT02498600, therapeutic intervention are under investigation. Current
and NCT03355976), and significantly higher response rates challenges are represented by the choice of rational immu-
(31.4%) compared with nivolumab alone (12.2%) were re- nomodulatory combinations with the most promising activity.
ported in a phase II study of 100 patients with persistent or Antiangiogenesis Drugs
recurrent EOC.30 Bevacizumab, the antiangiogenic monoclonal antibody that
targets VEGF, has been shown to be as active as mono-
IMMUNOSUPPRESSIVE MICROENVIRONMENT OF therapy for patients with advanced recurrent EOC,67 and it
OVARIAN CANCER is approved for use in combination with carboplatin and
The immunosuppressive TME contributes to the progression paclitaxel as first-line treatment of advanced EOC.68 Pre-
of cancer and metastasis, and the blockade of immune clinical studies indicated that anti-VEGF improves re-
checkpoints is not sufficient on its own. Additional pathways sponses to ICB.69 In a murine model of EOC overexpressing
that hamper an effective antitumor immune response must be VEGF, blockade of VEGF and PGE2 increased T-cell homing
targeted in EOC.31 Several immunosuppressive factors, such to tumors,46 and in another model of melanoma, disruption
as VEGF,32,33 interleukin 6 (IL-6),34,35 IL-10,36,37 prostaglandin of the VEGF/VEGF receptor 2 axis increased the extrava-
E2 (PGE2),38,39 and the overexpression of the tryptophan- sation of adoptively transferred T cells from tumor vessels
metabolizing enzyme indoleamine 2,3-dioxygenase,40-44 into tumors.70 VEGF blockade has also been shown to in-
contribute to tumor progression and poor prognosis.21 The crease the efficacy of a granulocyte–macrophage colony-
inflammatory cytokines IFN-γ and tumor necrosis factor-α stimulating factor–secreting tumor vaccine, which was
(TNF-α) synergize to induce COX2,45 a key enzyme of PGE2 associated with an intratumoral reduction of Tregs and
synthesis that together with VEGF limits T-cell homing to increase of effector T cells.71 In a phase II trial conducted
tumors.46 VEGF has also been shown to inhibit dendritic cell in women with recurrent EOC, the combination of bev-
(DC) maturation and function47 and upregulate PD-L1 ex- acizumab with nivolumab (anti–PD-1) resulted in a con-
pression on myeloid DCs.48 The increased number of acti- firmed ORR of 21% and a median progression-free survival
vated regulatory T cells (Treg),49 which produce IL-10 and (PFS) of 9.4 months.72 Furthermore, in a dose-escalation

Kandalaft, Odunsi, and Coukos
phase I study, the combination of durvalumab (anti–PD-L1) with cyclophosphamide combined with pembrolizumab and
cediranib (a tyrosine kinase inhibitor with broad antiangiogenic bevacizumab resulted in an ORR of 37.5% (all partial re-
activity) resulted in an ORR of 50% and a DCR of 75%.69 sponses) and a 6-month PFS of 70% (59% in patients with
platinum-resistant EOC).90
Chemotherapy
PARP Inhibitors
Cisplatin has been shown to increase tumor cell susceptibility
to CD8-positive T-cell–mediated cytotoxicity,73 suggesting that PARPs are proteins involved in DNA damage repair, an
responses to immunotherapies could be increased in combi- essential pathway especially in cells with defects in their
nation with chemotherapy.74 In a study of non–small cell lung homologous recombination or double-strand DNA repair
cancer (NSCLC), the efficacy of PD-1 blockade75 was increased functions.91 Tumors harboring BRCA1 or BRCA2 mutations
when combined with carboplatin/paclitaxel chemotherapy, are particularly sensitive to PARPis,92 which have emerged
supporting the exploration of cisplatin with immunotherapy as as an important therapeutic option for patients with EOC.93
frontline treatment for patients with EOC. Of note, CD8-positive Three PARPis are currently approved as maintenance
T-cell function in patients with EOC during carboplatin/ therapy after platinum-based chemotherapy for recurrent
paclitaxel chemotherapy appears to be initially suppressed, yet platinum-sensitive EOC.94-96 Because of their mechanism
they recover with improved antigen-specific reactivity.76 of action, PARPis may increase the mutational burden of
ovarian tumors in treated patients, a biomarker predictive of
Other chemotherapeutic agents active in platinum-resistant
response to ICB.97 In murine models, PARPis have also
EOC (e.g., doxorubicin or oxaliplatin) that may induce im-
been shown to activate a cascade of events leading to IFN-γ
munogenic tumor cell death77-79 could increase the efficacy
production in BRCA-deficient tumors and synergize with
of immunotherapy. Pegylated liposomal doxorubicin (PLD)
PD-1 or CTLA-4 blockade.98-100 Of note, PARPis can also
is a standard option for patients with EOC, and it has been
stimulate the production of type I IFNs, which is further
shown to increase tumor antigen uptake by myeloid DCs
increased by ICB independently of BRCA status.101
and presentation to T cells.80-82 The combination of PLD
with immunotherapy is currently under investigation in EOC. PARPis are currently under investigation in combination
In a phase II study of 297 patients with recurrent EOC, the with CTLA-4 blockade (NCT02571725 and NCT02485990)
addition of PLD to motolimod (a TLR8 agonist stimulating and PD-1/PD-L1 blockade (NCT03522246, NCT03642132,
innate immunity) did not increase OS or PFS compared with NCT02484404, NCT02657889, NCT03602859, NCT03737643,
placebo.83 In the phase III JAVELIN Ovarian 200 study, and NCT03740165). In the phase II MEDIOLA study, olaparib
which compared avelumab (anti–PD-L1) alone or in com- combined with durvalumab was well tolerated and resulted in
bination with PLD to PLD alone,84 the combination therapy an ORR of 72%, with a 19% rate of CR among 32 patients with
did not significantly increase PFS (HR, 0.78) or OS (HR, BRCA-mutated tumors (Y. Drew et al, personal communi-
0.89) compared with PLD alone for patients with platinum- cation, March 24-27, 2018).102 For patients with BRCA wild-
resistant, refractory EOC. It should be noted that the ORR type tumors, the combination therapy resulted in an ORR of
was 13.3% with the combination therapy compared with 17% and a DCR of 83% at 6 months.69 Final results of the
4.2% with PLD alone and 3.7% with single-agent avelu- phase I/II TOPACIO trial investigating the combination of
mab.85 The combination of PLD with another anti–PD-L1 niraparib with pembrolizumab are awaited with interest.103
(durvalumab) has also been investigated in a phase I/II study
of patients with platinum-resistant recurrent EOC. In this Other Immunomodulatory Agents
study, a promising 6-month PFS of 47.7% and a best ORR of The immunomodulatory effects of radiation therapy (RT)
15% with 5% CR were observed (O’Cearbhaill et al, personal combined with immunotherapy have been reported in several
communication, 2018).86 Of note, in the phase Ib/II FOR- preclinical and clinical studies.104,105 RT produces an interesting
WARD II trial (NCT02606305) investigating the combination response at the site of treatment and on “out-of-field” tumor
of mirvetuximab soravtansine, a folate receptor-α–targeting deposits (the so-called abscopal effect), yet the optimal dosage
antibody–drug conjugate with pembrolizumab (anti–PD-1) and fractionation that would trigger optimal immune-mediated
in heavily pretreated folate receptor-α–positive patients with responses has not been established. The combination of RT
platinum-resistant EOC, preliminary signs of efficacy were with ICB is still in early development (reviewed elsewhere106). RT
observed, supporting further investigation in this setting.87 for patients with EOC should be planned with extreme caution
Low-dose intravenous or oral metronomic chemotherapy because abdominal stereotactic RT may produce severe se-
may also increase the efficacy of immunotherapy. It has quelae such as intestinal obstruction and ureteral stenosis or
been shown that low-dose cyclophosphamide suppresses fistulae. Clinical studies investigating the combination of RT with
Tregs and improves vaccination as well as adoptive cell ICB in EOC are ongoing (NCT03277482 and NCT03312114).
therapy (ACT),88 and it increases CD8-positive TILs in Cytokine immunotherapy has been extensively investigated
combination with ICB.89 In a phase II trial, oral metronomic in EOC. No significant clinical benefit was observed with

type I IFNs and recombinant IFN-γ in combination with would induce a potent antitumor immune response. NY-
or after frontline chemotherapy.107-109 IL-12 is a proin- ESO-1 is one of the most interesting target antigens because
flammatory cytokine playing a key role in the differentiation it is not expressed in normal tissues other than the testis,
of T helper 1cells, which triggers the proliferation and and it was found to be expressed by 40.7% of tumors in
production of IFN-γ by activated T and natural killer cells.110 1,002 patients with EOC.137 Using various approaches,
In tumors, IL-12 can modulate the immunocompromising several clinical trials have investigated the efficacy of NY-
environment by reactivating anergic TILs and weakening IL- ESO-1–based vaccines and shown that overall, NY-ESO-1
10, TGF-β, and Treg-mediated suppression.111 IL-12 de- vaccines were able to extend the OS of patients by at least
livered intraperitoneally in mice is less toxic than systemic 2 years.54,138-141 However, the long-term efficacy of these
administration, and it delayed tumor progression and ex- vaccines targeting a self-antigen aberrantly expressed by
tended survival.112 In a dose-escalating phase I trial of 13 tumor cells is challenged by the fact that tumors will develop
patients with chemotherapy-resistant recurrent EOC, an in- mechanisms such as antigen loss via immunoediting to
traperitoneal IL-12 plasmid/lipopolymer complex was shown escape recognition by the immune system.54 Other vaccines
to be safe and well tolerated.113 In another phase I study, the are therefore needed to overcome escape mechanisms.
combination of an intraperitoneal IL-12 plasmid with PLD
Neoantigens arise from somatic DNA mutations occurring
resulted in an ORR of 25% (three patients), with a DCR of
because of the inherent genetic instability of tumors.
57.1% among 12 patients who received the highest plasmid
Truncal and clonal somatic mutations may generate neo-
dosage (dose-limiting toxicity was not reached).114
antigens, which are expressed by tumor cells as neo-
IL-2 was one of the first cytokines investigated in cancer peptides in the context of major histocompatibility complex
immunotherapy, and for patients with metastatic melanoma molecules. From an immunologic perspective, these neo-
the administration of high doses of IL-2 resulted in higher antigens are ideal vaccine targets because as their ex-
long-term survival115 yet with some severe toxicity. Low pression is restricted to tumors, they are recognized as
dosages of intraperitoneal IL-2 have been studied as non–self-antigens by the host T cells. Neoantigens have
monotherapy in two consecutive trials of 49 patients with the potential to activate tumor-specific T-cell responses of
EOC and residual disease after platinum-based chemo- high avidity, comparable to that observed with antiviral
therapy (positive second-look surgery), resulting in a 25% T-cell responses and in sharp contrast to the known low
pathologic overall response and a 17% pathologic CR (me- avidity T-cell responses to self-antigens.142 Tumors with the
dian survival of 2 to 10 or more years had not been highest mutational burden, thought to generate the highest
reached).116 New formulations of IL-2 (e.g., mutated forms, IL- neoantigen burden, appear to correlate with the strongest
2 fusion proteins, pegylated IL-2) with reduced toxicities and clinical response to immunotherapies.143 Yet neoantigens
more potent effects are undergoing clinical development.117 are naturally private,144,145 which complicates vaccine ap-
IL-18 is a proinflammatory cytokine that stimulates innate proaches. Based on a murine ovarian tumor model with low
and adaptive immunity and has been shown to synergize mutational burden, EOC was initially considered unsuitable
with time-sensitive chemotherapy and PLD in preclinical for neoantigen-specific vaccination.146 However, in contrast
models of EOC.118 In a phase I study of patients with to what was observed in this model, our group has identified
platinum-resistant EOC, the combination of IL-18 and PLD more than 1,300 nonsynonymous somatic mutations in
was well tolerated,119 but discontinuation of the molecule total, with about 69 mutations per tumor in a study of
precluded further clinical investigation. 19 immunotherapy-naive, chemotherapy-pretreated pa-
tients with recurrent advanced early EOC. Neoantigen-
specific CD8-positive T cells were retrieved in the periph-
VACCINES
eral blood or in the tumor of most patients evaluated.147
Several tumor-associated antigens have been identified in These encouraging results demonstrated that neoantigen-
EOC. These include EGFR2/neu,120,121 p53,122 amino en- specific CD8-positive T cells are spontaneously generated
hancer of split protein,123 folate-binding protein,124 sialyl in patients with EOC. In addition, we also found that
Tn,125 mucin antigen 1,126 mucin antigen 16/cancer neoantigen-specific T-cell responses could be amplified in
antigen,125,127 NY-ESO-1,128-130 and mesothelin.131 Other patients who received autologous DCs pulsed with oxidized
antigens that are universal have also been identified in EOC autologous tumor cell lysate. The boosted immune response
and include the human telomerase reverse transcription consisted of a mix of newly generated T cells directed
expressed by more than 80% of EOCs,132 CYP450, against previously unrecognized neoantigens and pre-
CYP1B1,133 and survivin134 (reviewed elsewhere135).Tumor- existing neoantigen-specific T cells that had expanded,
associated antigens are interesting potential targets for driven by the emergence of new T-cell clones with markedly
vaccines, but the molecular heterogeneity of the disease136 higher avidities and higher T-cell receptor (TCR) affinities
has challenged the choice of an adequate candidate that for previously recognized neoantigens.148 The use of

neoantigen vaccines is a promising personalized approach T-cell therapy has been demonstrated in hematologic
that has not yet been tested in EOC but is undergoing in- cancers,171 and it is being investigated in solid tumors. In
vestigation in other cancer indications with encouraging EOC, after positive preclinical results,172-175 CAR T-cell
early phase I study results.149-153 In the NT-001 phase Ib therapy is now in clinical development176 (reviewed
study, NEO-PV-01, a personalized neoantigen vaccine elsewhere177). In early studies of a small number of patients,
containing up to 20 peptide epitopes based on the genomic the use of TCR-engineered T cells resulted in objective
profile of each patient’s tumor, was given in combination responses, with some unexpected toxicities.178 Interestingly,
with nivolumab to patients with metastatic melanoma, three clinical studies have investigated the safety and
NSCLC, and bladder cancer. Of 19 patients who completed efficacy of autologous T cells engineered to express an
the full vaccine course, the ORR was 52.6% (10 patients), affinity-enhanced TCR recognizing NY-ESO-1 adoptively
with two patients with melanoma and one with NSCLC transferred in patients with melanoma, synovial sarcoma,
having response conversions.153 and multiple myeloma. In these studies, ORRs were
high, ranging from 50% to 80%.179-181 Similar studies are
ongoing in EOC with engineered T cells targeting NY-ESO-
ADOPTIVE T-CELL THERAPY 1 (NCT01567891, NCT02457650, and NCT03017131),
Another cell-based approach is ACT with TILs (TIL-ACT). This MUC16 (NCT02498912), MAGE-A4c1032 (NCT03132922),
personalized immunotherapy consists of collecting autologous and neoantigens (NCT03412877).
TILs that are expanded ex vivo before being transferred back
into the patient in combination with high-dose IL-2 to boost the PERSPECTIVES FOR FUTURE CLINICAL DEVELOPMENT
antitumor T-cell response in vivo.154-156 TIL-ACT is preceded by The variety of immunotherapies provides interesting op-
a nonmyeloablative lymphodepleting chemotherapy that in- portunities for the treatment of advanced recurrent EOC. As
duces a transient lymphopenia and leukopenia, which is the choice of drugs and combinations continues to grow, the
a determining factor to ensure the engraftment of the trans- development of conceptual frameworks of objectives and
ferred T cells.157 Although the use of TIL-ACT has been in- rational study designs are needed to accelerate clinical
vestigated in metastatic melanoma, resulting in high and development. Most of the progresses in EOC in the recurrent
reproducible clinical responses with approximately 50% of setting have been achieved with survival, and the recent
ORRs and up to 25% of long-lasting CRs, its use has been successes of immunotherapy raise hope that curative
scarcely investigated in other tumor types.158 In EOC (Table 1) therapies will be found. Clinical development should focus
the TIL-ACT approach has not yet given results. The lack of on the frontline setting, where maximal benefit can be
efficacy of TIL-ACT in this setting may result from several expected. The emergence of PARPis and strong signals that
factors such as inefficient expansion of TILs ex vivo, suboptimal their combination with ICB could yield important results in
lymphodepleting conditioning or IL-2 support, and charac- tumors with homologous recombination deficiency open the
teristics of the disease. Methods to expand TILs ex vivo have door for personalizing therapy and becoming ambitious with
improved,159 which raises expectations with ongoing clinical respect to making an important impact in the frontline
trials.160 In a pilot study161 of six heavily pretreated patients with setting. For these patients, the combination of PARPis with
advanced recurrent EOC, 18 to 91 109 TILs were admin- PD-1/PD-L1 blockade should be investigated as a frontline
istered with IL-2, resulting in four with stable disease as best approach because this would be a paradigm shift that could
response at 3 to 5 months. In this study, disease progression produce deep and durable responses and possibly replace
was caused primarily by the emergence of new lesions, current chemotherapy. Evidence that hypoxia suppresses
whereas targeted lesions had regressed or remained stable. the expression of BRCA1 and RAD51, hence attenuating
This result illustrates the challenge of using TIL-ACT in re- the homologous recombination pathway,182,183 and that
current EOC because of intratumoral heterogeneity,162 the hypoxic tumor cells are sensitized to PARPis184 suggests
exhausted phenotype of TILs,161 or the low frequency of tumor- that antiangiogenic drugs could cooperate with PARPis in
reactive TILs in the infusion.159 Of note, in a study conducted in a contextual synthetic lethality, which sensitizes BRCA wild-
the 1990s of 13 patients who received consolidation TIL-ACT type tumors to respond to PARPis185 and sets the stage for
(1–4.4 109 cells) without IL-2 support after a CR to first-line successful combinatorial immunotherapy. Given that both
platinum-based therapy, disease-free survival and OS rates at antiangiogenic drugs and PARPis are synergistic partners
3 years were 82% and 100%, respectively, and significantly for ICB, such combinations could yield important results in
longer than what was reported for patients who did not receive homologous recombination–proficient EOC and, if con-
TILs (67.5% and 54.5%, respectively).163 firmed, should also be tested in the frontline setting.
Adoptive T-cell therapies using T cells engineered to ex- Consolidation approaches have long been pursued in EOC
press TCRs with redirected specificity or chimeric antigen considering the high rate of disease relapse after frontline
receptors (CARs) are of great interest. The efficacy of CAR responses. Immunotherapy could increase curative response

TABLE 1. Clinical Studies Investigating TIL-ACT in Patients With EOC

No. of
First Author Patient Population Patients Phase Interventions and Combinations
Aoki164
Advanced or recurrent 17 Pilot Cy (200 mg; day 2), TIL (1 1010, 7 patients) or alternated
cisplatin-containing CT with TIL-ACT (10 patients)
Freedman165 Advanced, refractory to platinum- 8 of 11 Pilot IP TILs expanded in rIL-2 and low-dose rIL-2 IP
based CT
Ikarashi166 Stage II, III, or IV, after debulking 12 of 22 1/2 TILs (approximately 5 108) after cisplatin-based CT
surgery
Fujita163 Stage II, III, and IV, disease free 13 of 24 Pilot Day 1: Cy (350 mg/m2), doxorubicin (40 mg/m2), cisplatin
after surgery and cisplatin- (50 mg/m2); 5-FU (350 mg/m2; days 1–5), 3–5 cycles
based CT
After recovery from CT: 5 108 TILs infused
Freedman167 Advanced, refractory to platinum- Unknown Pilot Unknown
based CT
Hua168 Unknown 25 Randomized Unknown
double-blind
study
Freedman169 Stage III or IV 2 of 22 Pilot TILs on day 2 during a cycle consisting of IP rIFN-γ (days 1 and 4)
and IP rIL-2 (days 2–5)
Pedersen170 Progressive or recurrent 3 of 5 Pilot High-dose CT (60 mg/kg Cy for 2 days and 25 mg/m2 fludarabine
for 5 days) followed by T-cell administration and subsequent
decrescendo IL-2 for 5 days
Abbreviations: ACT, adoptive cell therapy; CT, chemotherapy; Cy, cyclophosphamide; EOC, epithelial ovarian cancer; 5-FU, 5-fluorouracil; IP, intraperitoneal;
rIFN-γ, recombinant interferon-γ; rIL-2, recombinant interleukin-2; TIL, tumor-infiltrating lymphocyte.
rate and survival in patients with minimal residual disease after myeloid-derived suppressor cells, Tregs, tumor-associated
frontline therapy. The combination of vaccines, especially mesenchymal cells, and the vasculature constitute possible
those targeting neoepitopes, combined with immunomodu- therapeutic targets for drug development. Clinical develop-
lation in the form of low-dose cyclophosphamide or ICB, could ment must progress with the validation of adaptive early-phase
be proposed as low-toxicity therapeutic approaches. For study designs, in which drug candidates for combination
patients with persistent disease after frontline therapy, ACT therapies could be tested and rapidly retained or discarded,
with either TILs or targeted T cells could also be a rational and during which biomarkers could be incorporated. The
approach given its single administration modality. latter requires obtaining tumor biopsy specimens, which could
Besides the approaches mentioned herein, drugs targeting be partly obviated by molecular imaging techniques and liquid
immunosuppressive factors predominant in EOC should biopsies. Finally, clinical studies testing one same-drug
be tested alone and in combination. Immunosuppressors combination and conducted simultaneously in the thera-
such as TGF-β, IL-10, indoleamine 2,3-dioxygenase-1, and peutic setting, in the neoadjuvant setting, and in tumor sur-
PGE2; the adenosine pathway; and proinflammatory factors rogate models should quickly provide the critical knowledge
such as tumor necrosis factor-α, IL-6, M2 macrophages, needed to accelerate clinical development.

1
Ludwig Institute for Cancer Research, University of Lausanne, and George Coukos, MD, PhD, Oncology Department, University of Lausanne,
Department of Oncology, Lausanne University Hospital, Lausanne, Rue du Bugnon 46, Office BH09_701, 1011 Lausanne, Switzerland;
Switzerland email: [email protected].
2
Center for Immunotherapy, Roswell Park Comprehensive Cancer Center,
Buffalo, NY
3
Department of Gynecologic Oncology, Roswell Park Comprehensive
Cancer Center, Buffalo, NY
EDBK_280539.

REFERENCES
1. International Agency for Research on Cancer. Estimated Number of Incident Cases from 2018 to 2040, Ovary, Females, All Ages. https://gco.iarc.fr/tomorrow/
graphic-isotype?type=0&type_sex=0&mode=population&sex=0&populations=900&cancers=39&age_group=value&apc_male=0&apc_female=0&single_
unit=500000&print=0. Accessed May 6, 2020.
2. Parkes E, McCabe N, Kennedy R. Development of PARP inhibitors for BRCA-deficient epithelial ovarian cancer. In Dammacco F and Silvestris F (eds).
Oncogenomics. New York: Academic Press, 2019;521-532.
4. Nirschl CJ, Drake CG. Molecular pathways: coexpression of immune checkpoint molecules: signaling pathways and implications for cancer immunotherapy.
Clin Cancer Res. 2013;19:4917-4924.
5. Quirk SK, Shure AK, Agrawal DK. Immune-mediated adverse events of anticytotoxic T lymphocyte-associated antigen 4 antibody therapy in metastatic
melanoma. Transl Res. 2015;166:412-424.
6. Hamanishi J, Mandai M, Ikeda T, et al. Safety and antitumor activity of anti-PD-1 antibody, nivolumab, in patients with platinum-resistant ovarian cancer. J Clin
Oncol. 2015;33:4015-4022.
7. Weber JS, D’Angelo SP, Minor D, et al. Nivolumab versus chemotherapy in patients with advanced melanoma who progressed after anti-CTLA-4 treatment
(CheckMate 037): a randomised, controlled, open-label, phase 3 trial. Lancet Oncol. 2015;16:375-384.
8. Hellmann MD, Callahan MK, Awad MM, et al. Tumor mutational burden and efficacy of nivolumab monotherapy and in combination with ipilimumab in small-
cell lung cancer. Cancer Cell. 2018;33:853-861 e4.
9. Sato E, Olson SH, Ahn J, et al. Intraepithelial CD8+ tumor-infiltrating lymphocytes and a high CD8+/regulatory T cell ratio are associated with favorable prognosis
in ovarian cancer. Proc Natl Acad Sci USA. 2005;102:18538-18543.
10. Zhang L, Conejo-Garcia JR, Katsaros D, et al. Intratumoral T cells, recurrence, and survival in epithelial ovarian cancer. N Engl J Med. 2003;348:203-213.
11. Hwang WT, Adams SF, Tahirovic E, et al. Prognostic significance of tumor-infiltrating T cells in ovarian cancer: a meta-analysis. Gynecol Oncol. 2012;
124:192-198.
12. Odunsi K. Immunotherapy in ovarian cancer. Ann Oncol. 2017;28:viii1-viii7.
13. Bonner JA, Willey CD, Yang ES, et al. Treatment of small cell lung cancer with TRA-8 in combination with cisplatin and radiation. Radiother Oncol. 2011;
101:183-189.
14. Tothill RW, Tinker AV, George J, et al; Australian Ovarian Cancer Study Group. Novel molecular subtypes of serous and endometrioid ovarian cancer linked to
clinical outcome. Clin Cancer Res. 2008;14:5198-5208.
15. Verhaak RGW, Tamayo P, Yang J-Y, et al; Cancer Genome Atlas Research Network. Prognostically relevant gene signatures of high-grade serous ovarian
carcinoma. J Clin Invest. 2013;123:517-525.
16. Maine CJ, Aziz NH, Chatterjee J, et al. Programmed death ligand-1 over-expression correlates with malignancy and contributes to immune regulation in ovarian
cancer. Cancer Immunol Immunother. 2014;63:215-224.
17. Wang X, Teng F, Kong L, et al. PD-L1 expression in human cancers and its association with clinical outcomes. OncoTargets Ther. 2016;9:5023-5039.
18. Matsuzaki J, Gnjatic S, Mhawech-Fauceglia P, et al. Tumor-infiltrating NY-ESO-1-specific CD8+ T cells are negatively regulated by LAG-3 and PD-1 in human
19. Huang RY, Eppolito C, Lele S, et al. LAG3 and PD1 co-inhibitory molecules collaborate to limit CD8+ T cell signaling and dampen antitumor immunity in a murine
ovarian cancer model. Oncotarget. 2015;6:27359-27377.
20. Abiko K, Matsumura N, Hamanishi J, et al. IFN-γ from lymphocytes induces PD-L1 expression and promotes progression of ovarian cancer. Br J Cancer. 2015;
112:1501-1509.
21. Rodriguez GM, Galpin KJC, McCloskey CW, et al. The tumor microenvironment of epithelial ovarian cancer and its influence on response to immunotherapy.
Cancers (Basel). 2018;10:E242.
22. Matulonis UA, Shapira-Frommer R, Santin A, et al. Antitumor activity and safety of pembrolizumab in patients with advanced recurrent ovarian cancer: interim
results from the phase 2 KEYNOTE-100 study. J Clin Oncol. 2018;36:5511.
23. Huang RY, Francois A, McGray AR, et al. Compensatory upregulation of PD-1, LAG-3, and CTLA-4 limits the efficacy of single-agent checkpoint blockade in
metastatic ovarian cancer. OncoImmunology. 2016;6:e1249561.
24. Korman A, Chen B, Wang C, et al. Activity of anti-PD-1 in murine tumor models: role of “host” PD-L1 and synergistic effect of anti-PD-1 and anti-CTLA-4.
J Immunol. 2007;178:S82.
25. Duraiswamy J, Kaluza KM, Freeman GJ, et al. Dual blockade of PD-1 and CTLA-4 combined with tumor vaccine effectively restores T-cell rejection function in
tumors. Cancer Res. 2013;73:3591-3603.
26. Curran MA, Montalvo W, Yagita H, et al. PD-1 and CTLA-4 combination blockade expands infiltrating T cells and reduces regulatory T and myeloid cells within
B16 melanoma tumors. Proc Natl Acad Sci USA. 2010;107:4275-4280.
27. Turajlic S, Gore M, Larkin J. First report of overall survival for ipilimumab plus nivolumab from the phase III Checkmate 067 study in advanced melanoma. Ann
Oncol. 2018;29:542-543.
378:2093-2104.

29. Shoushtari AN, Friedman CF, Navid-Azarbaijani P, et al. Measuring toxic effects and time to treatment failure for nivolumab plus ipilimumab in melanoma. JAMA
Oncol. 2018;4:98-101.
30. Burger R, Sill M, Zamarin D, et al. NRG Oncology phase II randomized trial of nivolumab with or without ipilimumab in patients with persistent or recurrent
ovarian cancer. Presented at the 17th Biennial Meeting of the International Gynecological Cancer Society, Kyoto, Japan, September 14–16, 2018.
31. Mellman I, Coukos G, Dranoff G. Cancer immunotherapy comes of age. Nature. 2011;480:480-489.
32. Li YL, Zhao H, Ren XB. Relationship of VEGF/VEGFR with immune and cancer cells: staggering or forward? Cancer Biol Med. 2016;13:206-214.
33. Duncan TJ, Al-Attar A, Rolland P, et al. Vascular endothelial growth factor expression in ovarian cancer: a model for targeted use of novel therapies? Clin Cancer
Res. 2008;14:3030-3035.
34. Lo CW, Chen MW, Hsiao M, et al. IL-6 trans-signaling in formation and progression of malignant ascites in ovarian cancer. Cancer Res. 2011;71:424-434.
35. Kolomeyevskaya N, Eng KH, Khan AN, et al. Cytokine profiling of ascites at primary surgery identifies an interaction of tumor necrosis factor-α and interleukin-6
in predicting reduced progression-free survival in epithelial ovarian cancer. Gynecol Oncol. 2015;138:352-357.
36. Mustea A, Könsgen D, Braicu EI, et al. Expression of IL-10 in patients with ovarian carcinoma. Anticancer Res. 2006;26(2C):1715-1718.
37. Rabinovich A, Medina L, Piura B, et al. Expression of IL-10 in human normal and cancerous ovarian tissues and cells. Eur Cytokine Netw. 2010;21:122-128.
38. Rask K, Zhu Y, Wang W, et al. Ovarian epithelial cancer: a role for PGE2-synthesis and signalling in malignant transformation and progression. Mol Cancer.
2006;5:62.
39. Qiu X, Cheng JC, Chang HM, et al. COX2 and PGE2 mediate EGF-induced E-cadherin-independent human ovarian cancer cell invasion. Endocr Relat Cancer.
2014;21:533-543.
40. Inaba T, Ino K, Kajiyama H, et al. Role of the immunosuppressive enzyme indoleamine 2,3-dioxygenase in the progression of ovarian carcinoma. Gynecol Oncol.
2009;115:185-192.
41. Okamoto A, Nikaido T, Ochiai K, et al. Indoleamine 2,3-dioxygenase serves as a marker of poor prognosis in gene expression profiles of serous ovarian cancer
cells. Clin Cancer Res. 2005;11:6030-6039.
42. Qian F, Villella J, Wallace PK, et al. Efficacy of levo-1-methyl tryptophan and dextro-1-methyl tryptophan in reversing indoleamine-2,3-dioxygenase-mediated
arrest of T-cell proliferation in human epithelial ovarian cancer. Cancer Res. 2009;69:5498-5504.
43. Godin-Ethier J, Hanafi L-A, Piccirillo CA, et al. Indoleamine 2,3-dioxygenase expression in human cancers: clinical and immunologic perspectives. Clin Cancer
Res. 2011;17:6985-6991.
44. Hornyák L, Dobos N, Koncz G, et al. The role of indoleamine-2,3-dioxygenase in cancer development, diagnostics, and therapy. Front Immunol. 2018;9:151.
45. Wong JL, Obermajer N, Odunsi K, et al. Synergistic COX2 induction by IFNγ and TNFα self-limits type-1 immunity in the human tumor microenvironment.
Cancer Immunol Res. 2016;4:303-311.
46. Motz GT, Santoro SP, Wang LP, et al. Tumor endothelium FasL establishes a selective immune barrier promoting tolerance in tumors. Nat Med. 2014;
20:607-615.
47. Ibáñez-Vea M, Zuazo M, Gato M, et al. Myeloid-derived suppressor cells in the tumor microenvironment: current knowledge and future perspectives. Arch
Immunol Ther Exp (Warsz). 2018;66:113-123.
48. Gabrilovich DI, Nagaraj S. Myeloid-derived suppressor cells as regulators of the immune system. Nat Rev Immunol. 2009;9:162-174.
49. Curiel TJ, Coukos G, Zou L, et al. Specific recruitment of regulatory T cells in ovarian carcinoma fosters immune privilege and predicts reduced survival. Nat
Med. 2004;10:942-949.
50. Schmidt A, Oberle N, Krammer PH. Molecular mechanisms of treg-mediated T cell suppression. Front Immunol. 2012;3:51.
51. Pastuła A, Marcinkiewicz J. Myeloid-derived suppressor cells: a double-edged sword? Int J Exp Pathol. 2011;92:73-78.
52. Condamine T, Ramachandran I, Youn JI, et al. Regulation of tumor metastasis by myeloid-derived suppressor cells. Annu Rev Med. 2015;66:97-110.
53. Kooi S, Zhang HZ, Patenia R, et al. HLA class I expression on human ovarian carcinoma cells correlates with T-cell infiltration in vivo and T-cell expansion in vitro
in low concentrations of recombinant interleukin-2. Cell Immunol. 1996;174:116-128.
54. Odunsi K, Qian F, Matsuzaki J, et al. Vaccination with an NY-ESO-1 peptide of HLA class I/II specificities induces integrated humoral and T cell responses in
55. Tesone AJ, Rutkowski MR, Brencicova E, et al. Satb1 overexpression drives tumor-promoting activities in cancer-associated dendritic cells. Cell Rep. 2016;
14:1774-1786.
56. Kaliński P, Hilkens CM, Snijders A, et al. IL-12-deficient dendritic cells, generated in the presence of prostaglandin E2, promote type 2 cytokine production in
maturing human naive T helper cells. J Immunol. 1997;159:28-35.
57. Obermajer N, Muthuswamy R, Lesnock J, et al. Positive feedback between PGE2 and COX2 redirects the differentiation of human dendritic cells toward stable
myeloid-derived suppressor cells. Blood. 2011;118:5498-5505.
58. Yuan X, Zhang J, Li D, et al. Prognostic significance of tumor-associated macrophages in ovarian cancer: A meta-analysis. Gynecol Oncol. 2017;147:181-187.
59. Ruffell B, Coussens LM. Macrophages and therapeutic resistance in cancer. Cancer Cell. 2015;27:462-472.
60. Reinartz S, Schumann T, Finkernagel F, et al. Mixed-polarization phenotype of ascites-associated macrophages in human ovarian carcinoma: correlation of
CD163 expression, cytokine levels and early relapse. Int J Cancer. 2014;134:32-42.

61. Finkernagel F, Reinartz S, Lieber S, et al. The transcriptional signature of human ovarian carcinoma macrophages is associated with extracellular matrix
reorganization. Oncotarget. 2016;7:75339-75352.
62. Yanaihara N, Anglesio MS, Ochiai K, et al. Cytokine gene expression signature in ovarian clear cell carcinoma. Int J Oncol. 2012;41:1094-1100.
63. Deng X, Zhang P, Liang T, et al. Ovarian cancer stem cells induce the M2 polarization of macrophages through the PPARγ and NF-κB pathways. Int J Mol Med.
2015;36:449-454.
64. Zhang Q, Cai DJ, Li B. Ovarian cancer stem-like cells elicit the polarization of M2 macrophages. Mol Med Rep. 2015;11:4685-4693.
65. Dijkgraaf EM, Heusinkveld M, Tummers B, et al. Chemotherapy alters monocyte differentiation to favor generation of cancer-supporting M2 macrophages in the
tumor microenvironment. Cancer Res. 2013;73:2480-2492.
66. Mlynska A, Povilaityte E, Zemleckaite I, et al. Platinum sensitivity of ovarian cancer cells does not influence their ability to induce M2-type macrophage
polarization. Am J Reprod Immunol. 2018;80:e12996.
67. Burger RA, Sill MW, Monk BJ, et al. Phase II trial of bevacizumab in persistent or recurrent epithelial ovarian cancer or primary peritoneal cancer: a Gynecologic
Oncology Group Study. J Clin Oncol. 2007;25:5165-5171.
68. Rossi L, Verrico M, Zaccarelli E, et al. Bevacizumab in ovarian cancer: a critical review of phase III studies. Oncotarget. 2017;8:12389-12405.
69. Lee JM, Cimino-Mathews A, Peer CJ, et al. Safety and clinical activity of the programmed death-ligand 1 inhibitor durvalumab in combination with poly (ADP-
ribose) polymerase inhibitor olaparib or vascular endothelial growth factor receptor 1-3 inhibitor cediranib in women’s cancers: a dose-escalation, phase I study.
J Clin Oncol. 2017;35:2193-2202.
70. Shrimali RK, Yu Z, Theoret MR, et al. Antiangiogenic agents can increase lymphocyte infiltration into tumor and enhance the effectiveness of adoptive
immunotherapy of cancer. Cancer Res. 2010;70:6171-6180.
71. Li B, Lalani AS, Harding TC, et al. Vascular endothelial growth factor blockade reduces intratumoral regulatory T cells and enhances the efficacy of a GM-CSF-
secreting cancer immunotherapy. Clin Cancer Res. 2006;12:6808-6816.
72. Liu J, Herold C, Luo W, et al. A phase II trial of combination nivolumab and bevacizumab in recurrent ovarian cancer. Ann Oncol. 2018;29:viii332-viii358.
73. Ramakrishnan R, Assudani D, Nagaraj S, et al. Chemotherapy enhances tumor cell susceptibility to CTL-mediated killing during cancer immunotherapy in mice.
J Clin Invest. 2010;120:1111-1124.
74. Kroon P, Frijlink E, Iglesias-Guimarais V, et al. Radiotherapy and cisplatin increase immunotherapy efficacy by enabling local and systemic intratumoral T-cell
activity. Cancer Immunol Res. 2019;7:670-682.
75. Lazzari C, Karachaliou N, Bulotta A, et al. Combination of immunotherapy with chemotherapy and radiotherapy in lung cancer: is this the beginning of the end
for cancer? Ther Adv Med Oncol. 2018;10:1758835918762094.
76. Coleman S, Clayton A, Mason MD, et al. Recovery of CD8+ T-cell function during systemic chemotherapy in advanced ovarian cancer. Cancer Res. 2005;
65:7000-7006.
77. Bezu L, Gomes-de-Silva LC, Dewitte H, et al. Combinatorial strategies for the induction of immunogenic cell death. Front Immunol. 2015;6:187.
78. Gebremeskel S, Johnston B. Concepts and mechanisms underlying chemotherapy induced immunogenic cell death: impact on clinical studies and con-
siderations for combined therapies. Oncotarget. 2015;6:41600-41619.
79. Zhou J, Yang T, Liu L, et al. Chemotherapy oxaliplatin sensitizes prostate cancer to immune checkpoint blockade therapies via stimulating tumor immu-
nogenicity. Mol Med Rep. 2017;16:2868-2874.
80. Zitvogel L, Kepp O, Senovilla L, et al. Immunogenic tumor cell death for optimal anticancer therapy: the calreticulin exposure pathway. Clin Cancer Res. 2010;
16:3100-3104.
81. Casares N, Pequignot MO, Tesniere A, et al. Caspase-dependent immunogenicity of doxorubicin-induced tumor cell death. J Exp Med. 2005;202:1691-1701.
82. Zhang Z, Yu X, Wang Z, et al. Anthracyclines potentiate anti-tumor immunity: a new opportunity for chemoimmunotherapy. Cancer Lett. 2015;369:331-335.
83. Monk BJ, Brady MF, Aghajanian C, et al. A phase 2, randomized, double-blind, placebo-controlled study of chemo-immunotherapy combination using
motolimod with pegylated liposomal doxorubicin in recurrent or persistent ovarian cancer: a Gynecologic Oncology Group partners study. Ann Oncol. 2017;
28:996-1004.
84. Pujade-Lauraine E, Fujiwara K, Dychter SS, et al. Avelumab (anti-PD-L1) in platinum-resistant/refractory ovarian cancer: JAVELIN Ovarian 200 phase III study
design. Future Oncol. 2018;14:2103-2113.
85. Colombus G. Avelumab misses primary endpoints in phase III ovarian cancer trial. https://www.onclive.com/web-exclusives/avelumabmisses-primary-
endpoints-in-phase-iii-ovarian-cancer-trial. Accessed May 6, 2020.
86. O’Cearbhaill R, Wolfer A, Disilvestro P, et al. 945PA phase I/II study of chemo-immunotherapy with durvalumab (durva) and pegylated liposomal doxorubicin
(PLD) in platinum-resistant recurrent ovarian cancer (PROC). Ann Oncol. 2018;29.
87. Matulonis U, Moore K, Martin L, et al. Initial safety and activity findings from a phase IB escalation study of mirvetuximab soravtansine, a folate receptor alpha
(FRalpha targeting antibody-drug conjugate (ADC), with pembrolizumab in platinum-resistant epithelial ovarian cancer (EOC) patients. Gynecol Oncol. 2018;
149:38.
88. Madondo MT, Quinn M, Plebanski M. Low dose cyclophosphamide: mechanisms of T cell modulation. Cancer Treat Rev. 2016;42:3-9.
89. Mkrtichyan M, Najjar YG, Raulfs EC, et al. Anti-PD-1 synergizes with cyclophosphamide to induce potent anti-tumor vaccine effects through novel mechanisms.
Eur J Immunol. 2011;41:2977-2986.

90. Zsiros E, Frederick P, Akers S, et al. A phase II trial of pembrolizumab in combination with bevacizumab and oral metronomic cyclophosphamide for recurrent
epithelial ovarian, fallopian tube or primary peritoneal cancer. Presented at the Society of Gynecologic Oncology 2019 Annual Meeting, Honolulu, Hawaii, March
16–19, 2019.
91. Ashworth A. A synthetic lethal therapeutic approach: poly(ADP) ribose polymerase inhibitors for the treatment of cancers deficient in DNA double-strand break
repair. J Clin Oncol. 2008;26:3785-3790.
92. Scott CL, Swisher EM, Kaufmann SH. Poly (ADP-ribose) polymerase inhibitors: recent advances and future development. J Clin Oncol. 2015;33:1397-1406.
93. Papa A, Caruso D, Strudel M, et al. Update on poly-ADP-ribose polymerase inhibition for ovarian cancer treatment. J Transl Med. 2016;14:267.
N Engl J Med. 2016;375:2154-2164.
95. Ledermann J, Harter P, Gourley C, et al. Olaparib maintenance therapy in platinum-sensitive relapsed ovarian cancer. N Engl J Med. 2012;366:1382-1392.
96. Coleman RL, Oza AM, Lorusso D, et al; ARIEL3 investigators. Rucaparib maintenance treatment for recurrent ovarian carcinoma after response to platinum
therapy (ARIEL3): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet. 2017;390:1949-1961.
97. Lyu GY, Yeh YH, Yeh YC, et al. Mutation load estimation model as a predictor of the response to cancer immunotherapy. NPJ Genom Med. 2018;3:12.
98. Wang Z, Sun K, Xiao Y, et al. Niraparib activates interferon signaling and potentiates anti-PD-1 antibody efficacy in tumor models. Sci Rep. 2019;9:1853.
99. Ding L, Kim HJ, Wang Q, et al. PARP inhibition elicits STING-dependent antitumor immunity in Brca1-deficient ovarian cancer. Cell Rep. 2018;25:2972-2980
e5.
100. Higuchi T, Flies DB, Marjon NA, et al. CTLA-4 blockade synergizes therapeutically with PARP inhibition in BRCA1-deficient ovarian cancer. Cancer Immunol
Res. 2015;3:1257-1268.
101. Shen J, Zhao W, Ju Z, et al. PARPi triggers the STING-dependent immune response and enhances the therapeutic efficacy of immune checkpoint blockade
independent of BRCAness. Cancer Res. 2019;79:311-319.
102. Drew Y, de Jonge M, Hong S, et al. An open-label, phase II basket study of olaparib and durvalumab (MEDIOLA): Results in germline BRCA-mutated (gBRCAm)
platinum-sensitive relapsed (PSR) ovarian cancer (OC). Presented at the SGO Annual Meeting, New Orleans, LA, March 24–27, 2018.
103. Konstantinopoulos PA, Munster P, Forero-Torez A, et al. Topacio: preliminary activity and safety in patients (pts) with platinum-resistant ovarian cancer (PROC)
in a phase 1/2 study of niraparib in combination with pembrolizumab. Gynecol Oncol. 2018;149:246.
104. Formenti SC, Demaria S. Combining radiotherapy and cancer immunotherapy: a paradigm shift. J Natl Cancer Inst. 2013;105:256-265.
105. Chajon E, Castelli J, Marsiglia H, et al. The synergistic effect of radiotherapy and immunotherapy: a promising but not simple partnership. Crit Rev Oncol
Hematol. 2017;111:124-132.
106. Herrera FG, Bourhis J, Coukos G. Radiotherapy combination opportunities leveraging immunity for the next oncology practice. CA Cancer J Clin. 2017;
67:65-85.
107. Alberts DS, Hannigan EV, Liu PY, et al. Randomized trial of adjuvant intraperitoneal alpha-interferon in stage III ovarian cancer patients who have no evidence of
disease after primary surgery and chemotherapy: an intergroup study. Gynecol Oncol. 2006;100:133-138.
108. Alberts DS, Marth C, Alvarez RD, et al; GRACES Clinical Trial Consortium. Randomized phase 3 trial of interferon gamma-1b plus standard carboplatin/paclitaxel
versus carboplatin/paclitaxel alone for first-line treatment of advanced ovarian and primary peritoneal carcinomas: results from a prospectively designed analysis
of progression-free survival. Gynecol Oncol. 2008;109:174-181.
109. Lawal AO, Musekiwa A, Grobler L. Interferon after surgery for women with advanced (stage II-IV) epithelial ovarian cancer. Cochrane Database Syst Rev. 2013;
6:CD009620.
110. Weiss JM, Subleski JJ, Wigginton JM, et al. Immunotherapy of cancer by IL-12-based cytokine combinations. Expert Opin Biol Ther. 2007;7:1705-1721.
111. Zeh HJ III, Hurd S, Storkus WJ, et al. Interleukin-12 promotes the proliferation and cytolytic maturation of immune effectors: implications for the immunotherapy
of cancer. J Immunother Emphasis Tumor Immunol. 1993;14:155-161.
112. Cohen CA, Shea AA, Heffron CL, et al. Interleukin-12 immunomodulation delays the onset of lethal peritoneal disease of ovarian cancer. J Interferon Cytokine
Res. 2016;36:62-73.
113. Anwer K, Barnes MN, Fewell J, et al. Phase-I clinical trial of IL-12 plasmid/lipopolymer complexes for the treatment of recurrent ovarian cancer. Gene Ther.
2010;17:360-369.
114. Thaker PH, Brady WE, Lankes HA, et al. A phase I trial of intraperitoneal GEN-1, an IL-12 plasmid formulated with PEG-PEI-cholesterol lipopolymer, ad-
ministered with pegylated liposomal doxorubicin in patients with recurrent or persistent epithelial ovarian, fallopian tube or primary peritoneal cancers: an NRG
Oncology/Gynecologic Oncology Group study. Gynecol Oncol. 2017;147:283-290.
115. Rosenberg SA. IL-2: the first effective immunotherapy for human cancer. J Immunol. 2014;192:5451-5458.
116. Hutchinson L. Robust responses to intraperitoneal IL-2 in ovarian cancer. Nat Rev Clin Oncol. 2010;7:122.
117. Rosalia RA, Arenas-Ramirez N, Bouchaud G, et al. Use of enhanced interleukin-2 formulations for improved immunotherapy against cancer. Curr Opin Chem
Biol. 2014;23:39-46.
118. Alagkiozidis I, Facciabene A, Tsiatas M, et al. Time-dependent cytotoxic drugs selectively cooperate with IL-18 for cancer chemo-immunotherapy. J Transl Med.
2011;9:77.

119. Simpkins F, Flores A, Chu C, et al. Chemoimmunotherapy using pegylated liposomal doxorubicin and interleukin-18 in recurrent ovarian cancer: a phase I dose-
escalation study. Cancer Immunol Res. 2013;1:168-178.
120. Murray JL, Przepiorka D, Ioannides CG. Clinical trials of HER-2/neu-specific vaccines. Semin Oncol. 2000;27(6, Suppl 11):71-75, NaN-100.
121. Hellström I, Goodman G, Pullman J, et al. Overexpression of HER-2 in ovarian carcinomas. Cancer Res. 2001;61:2420-2423.
122. Shih I, Kurman RJ. Ovarian tumorigenesis: a proposed model based on morphological and molecular genetic analysis. Am J Pathol. 2004;164:1511-1518.
123. Babcock B, Anderson BW, Papayannopoulos I, et al. Ovarian and breast cytotoxic T lymphocytes can recognize peptides from the amino enhancer of split
protein of the Notch complex. Mol Immunol. 1998;35:1121-1133.
124. Peoples GE, Anderson BW, Fisk B, et al. Ovarian cancer-associated lymphocyte recognition of folate binding protein peptides. Ann Surg Oncol. 1998;
5:743-750.
125. Ghazizadeh M, Ogawa H, Sasaki Y, et al. Mucin carbohydrate antigens (T, Tn, and sialyl-Tn) in human ovarian carcinomas: relationship with histopathology and
prognosis. Hum Pathol. 1997;28:960-966.
126. Miles D, Papazisis K. Rationale for the clinical development of STn-KLH (Theratope) and anti-MUC-1 vaccines in breast cancer. Clin Breast Cancer. 2003;
3(Suppl 4):S134-S138.
127. Felder M, Kapur A, Gonzalez-Bosquet J, et al. MUC16 (CA125): tumor biomarker to cancer therapy, a work in progress. Mol Cancer. 2014;13:129.
128. Odunsi K, Jungbluth AA, Stockert E, et al. NY-ESO-1 and LAGE-1 cancer-testis antigens are potential targets for immunotherapy in epithelial ovarian cancer.
Cancer Res. 2003;63:6076-6083.
129. Stockert E, Jäger E, Chen YT, et al. A survey of the humoral immune response of cancer patients to a panel of human tumor antigens. J Exp Med. 1998;
187:1349-1354.
130. Chen YT, Scanlan MJ, Sahin U, et al. A testicular antigen aberrantly expressed in human cancers detected by autologous antibody screening. Proc Natl Acad Sci
USA. 1997;94:1914-1918.
131. Hilliard TS. The impact of mesothelin in the ovarian cancer tumor microenvironment. Cancers (Basel). 2018;10:E277.
132. Vonderheide RH, Hahn WC, Schultze JL, et al. The telomerase catalytic subunit is a widely expressed tumor-associated antigen recognized by cytotoxic T
lymphocytes. Immunity. 1999;10:673-679.
133. Luby TM, Cole G, Baker L, et al. Repeated immunization with plasmid DNA formulated in poly(lactide-co-glycolide) microparticles is well tolerated and
stimulates durable T cell responses to the tumor-associated antigen cytochrome P450 1B1. Clin Immunol. 2004;112:45-53.
134. Otto K, Andersen MH, Eggert A, et al. Lack of toxicity of therapy-induced T cell responses against the universal tumour antigen survivin. Vaccine. 2005;
23:884-889.
135. Martin Lluesma S, Wolfer A, Harari A, et al. Cancer vaccines in ovarian cancer: how can we improve? Biomedicines. 2016;4:E10.
136. Carter SL, Cibulskis K, Helman E, et al. Absolute quantification of somatic DNA alterations in human cancer. Nat Biotechnol. 2012;30:413-421.
137. Szender JB, Papanicolau-Sengos A, Eng KH, et al. NY-ESO-1 expression predicts an aggressive phenotype of ovarian cancer. Gynecol Oncol. 2017;
145:420-425.
138. Odunsi K, Matsuzaki J, Karbach J, et al. Efficacy of vaccination with recombinant vaccinia and fowlpox vectors expressing NY-ESO-1 antigen in ovarian cancer
and melanoma patients. Proc Natl Acad Sci USA. 2012;109:5797-5802.
139. Odunsi K, Matsuzaki J, James SR, et al. Epigenetic potentiation of NY-ESO-1 vaccine therapy in human ovarian cancer. Cancer Immunol Res. 2014;2:37-49.
140. Sabbatini P, Tsuji T, Ferran L, et al. Phase I trial of overlapping long peptides from a tumor self-antigen and poly-ICLC shows rapid induction of integrated
immune response in ovarian cancer patients. Clin Cancer Res. 2012;18:6497-6508.
141. Tsuji T, Sabbatini P, Jungbluth AA, et al. Effect of Montanide and poly-ICLC adjuvant on human self/tumor antigen-specific CD4+ T cells in phase I overlapping
long peptide vaccine trial. Cancer Immunol Res. 2013;1:340-350.
142. Lennerz V, Fatho M, Gentilini C, et al. The response of autologous T cells to a human melanoma is dominated by mutated neoantigens. Proc Natl Acad Sci USA.
2005;102:16013-16018.
143. Lee CH, Yelensky R, Jooss K, et al. Update on tumor neoantigens and their utility: why it is good to be different. Trends Immunol. 2018;39:536-548.
144. Alexandrov LB, Nik-Zainal S, Wedge DC, et al; ICGC PedBrain. Signatures of mutational processes in human cancer. Nature. 2013;500:415-421.
145. Lawrence MS, Stojanov P, Polak P, et al. Mutational heterogeneity in cancer and the search for new cancer-associated genes. Nature. 2013;499:214-218.
146. Martin SD, Brown SD, Wick DA, et al. Low mutation burden in ovarian cancer may limit the utility of neoantigen-targeted vaccines. PLoS One. 2016;
11:e0155189.
147. Bobisse S, Genolet R, Roberti A, et al. Sensitive and frequent identification of high avidity neo-epitope specific CD8 + T cells in immunotherapy-naive ovarian
cancer. Nat Commun. 2018;9:1092.
148. Tanyi JL, Bobisse S, Ophir E, et al. Personalized cancer vaccine effectively mobilizes antitumor T cell immunity in ovarian cancer. Sci Transl Med. 2018;
10:eaao5931.
149. Carreno BM, Magrini V, Becker-Hapak M, et al. Cancer immunotherapy. A dendritic cell vaccine increases the breadth and diversity of melanoma neoantigen-
specific T cells. Science. 2015;348:803-808.
150. Ott PA, Hu Z, Keskin DB, et al. An immunogenic personal neoantigen vaccine for patients with melanoma. Nature. 2017;547:217-221.

151. Sahin U, Derhovanessian E, Miller M, et al. Personalized RNA mutanome vaccines mobilize poly-specific therapeutic immunity against cancer. Nature. 2017;
547:222-226.
152. Sonntag K, Hashimoto H, Eyrich M, et al. Immune monitoring and TCR sequencing of CD4 T cells in a long term responsive patient with metastasized pancreatic
ductal carcinoma treated with individualized, neoepitope-derived multipeptide vaccines: a case report. J Transl Med. 2018;16:23.
153. Ott P, Govindan R, Naing A, et al. A personal neoantigen vaccine, NEO-PV-01, with anti-PD1 induces broad de novo anti-tumor immunity in patients with
metastatic melanoma, NSCLC and bladder cancer. Ann Oncol. 2018;29:viii400-viii441.
154. Dudley ME, Wunderlich JR, Robbins PF, et al. Cancer regression and autoimmunity in patients after clonal repopulation with antitumor lymphocytes. Science.
2002;298:850-854.
155. Overwijk WW, Theoret MR, Finkelstein SE, et al. Tumor regression and autoimmunity after reversal of a functionally tolerant state of self-reactive CD8+ T cells.
J Exp Med. 2003;198:569-580.
156. Palmer DC, Balasubramaniam S, Hanada K, et al. Vaccine-stimulated, adoptively transferred CD8+ T cells traffic indiscriminately and ubiquitously while
mediating specific tumor destruction. J Immunol. 2004;173:7209-7216.
157. Dudley ME, Wunderlich JR, Yang JC, et al. A phase I study of nonmyeloablative chemotherapy and adoptive transfer of autologous tumor antigen-specific T
lymphocytes in patients with metastatic melanoma. J Immunother. 2002;25:243-251.
158. Rosenberg SA, Yang JC, Sherry RM, et al. Durable complete responses in heavily pretreated patients with metastatic melanoma using T-cell transfer im-
munotherapy. Clin Cancer Res. 2011;17:4550-4557.
159. Westergaard MCW, Andersen R, Chong C, et al. Correction: tumour-reactive T cell subsets in the microenvironment of ovarian cancer. Br J Cancer. 2019;
120:870.
160. Andersen R, Donia M, Westergaard MC, et al. Tumor infiltrating lymphocyte therapy for ovarian cancer and renal cell carcinoma. Hum Vaccin Immunother.
2015;11:2790-2795.
161. Pedersen M, Westergaard MCW, Milne K, et al. Adoptive cell therapy with tumor-infiltrating lymphocytes in patients with metastatic ovarian cancer: a pilot study.
OncoImmunology. 2018;7:e1502905.
162. Zhang AW, McPherson A, Milne K, et al. Interfaces of malignant and immunologic clonal dynamics in ovarian cancer. Cell. 2018;173:1755-1769 e22.
163. Fujita K, Ikarashi H, Takakuwa K, et al. Prolonged disease-free period in patients with advanced epithelial ovarian cancer after adoptive transfer of tumor-
infiltrating lymphocytes. Clin Cancer Res. 1995;1:501-507.
164. Aoki Y, Takakuwa K, Kodama S, et al. Use of adoptive transfer of tumor-infiltrating lymphocytes alone or in combination with cisplatin-containing chemotherapy
in patients with epithelial ovarian cancer. Cancer Res. 1991;51:1934-1939.
165. Freedman RS, Edwards CL, Kavanagh JJ, et al. Intraperitoneal adoptive immunotherapy of ovarian carcinoma with tumor-infiltrating lymphocytes and low-dose
recombinant interleukin-2: a pilot trial. J Immunother Emphasis Tumor Immunol. 1994;16:198-210.
166. Ikarashi H, Fujita K, Takakuwa K, et al. Immunomodulation in patients with epithelial ovarian cancer after adoptive transfer of tumor-infiltrating lymphocytes.
Cancer Res. 1994;54:190-196.
167. Freedman RS, Platsoucas CD. Immunotherapy for peritoneal ovarian carcinoma metastasis using ex vivo expanded tumor infiltrating lymphocytes. Cancer Treat
Res. 1996;82:115-146.
168. Hua Z, Lu J, Li H. [Clinical study on immunotherapy of ovarian cancer with tumor infiltrating lymphocytes]. Zhonghua Fu Chan Ke Za Zhi. 1996;31:555-557.
169. Freedman RS, Kudelka AP, Kavanagh JJ, et al. Clinical and biological effects of intraperitoneal injections of recombinant interferon-gamma and recombinant
interleukin 2 with or without tumor-infiltrating lymphocytes in patients with ovarian or peritoneal carcinoma. Clin Cancer Res. 2000;6:2268-2278.
170. Pedersen M, Westergaard MCW, Borch TH, et al. T cell therapy for patients with advanced ovarian cancer: a pilot study in progress. Ann Oncol. 2016;27:vi375.
171. Zheng PP, Kros JM, Li J. Approved CAR T cell therapies: ice bucket challenges on glaring safety risks and long-term impacts. Drug Discov Today. 2018;
23:1175-1182.
172. Song DG, Ye Q, Carpenito C, et al. In vivo persistence, tumor localization, and antitumor activity of CAR-engineered T cells is enhanced by costimulatory
signaling through CD137 (4-1BB). Cancer Res. 2011;71:4617-4627.
173. Sun M, Shi H, Liu C, et al. Construction and evaluation of a novel humanized HER2-specific chimeric receptor. Breast Cancer Res. 2014;16:R61.
174. Lanitis E, Poussin M, Hagemann IS, et al. Redirected antitumor activity of primary human lymphocytes transduced with a fully human anti-mesothelin chimeric
receptor. Mol Ther. 2012;20:633-643.
175. Koneru M, Purdon TJ, Spriggs D, et al. IL-12 secreting tumor-targeted chimeric antigen receptor T cells eradicate ovarian tumors in vivo. OncoImmunology.
2015;4:e994446.
176. Tanyi J, Haas A, Beatty G, et al. Anti-mesothelin chimeric antigen receptor T cells in patients with epithelial ovarian cancer. J Clin Oncol. 2016;34:5511.
177. Jindal V, Arora E, Gupta S, et al. Prospects of chimeric antigen receptor T cell therapy in ovarian cancer. Med Oncol. 2018;35:70.
178. Ping Y, Liu C, Zhang Y. T-cell receptor-engineered T cells for cancer treatment: current status and future directions. Protein Cell. 2018;9:254-266.
179. Robbins PF, Kassim SH, Tran TL, et al. A pilot trial using lymphocytes genetically engineered with an NY-ESO-1-reactive T-cell receptor: long-term follow-up and
correlates with response. Clin Cancer Res. 2015;21:1019-1027.
180. Rapoport AP, Stadtmauer EA, Binder-Scholl GK, et al. NY-ESO-1-specific TCR-engineered T cells mediate sustained antigen-specific antitumor effects in
myeloma. Nat Med. 2015;21:914-921.

181. D’Angelo SP, Melchiori L, Merchant MS, et al. Antitumor activity associated with prolonged persistence of adoptively transferred NY-ESO-1 c259T cells in synovial
sarcoma. Cancer Discov. 2018;8:944-957.
182. Bindra RS, Gibson SL, Meng A, et al. Hypoxia-induced down-regulation of BRCA1 expression by E2Fs. Cancer Res. 2005;65:11597-11604.
183. Bindra RS, Schaffer PJ, Meng A, et al. Down-regulation of Rad51 and decreased homologous recombination in hypoxic cancer cells. Mol Cell Biol. 2004;
24:8504-8518.
184. Glazer PM, Hegan DC, Lu Y, et al. Hypoxia and DNA repair. Yale J Biol Med. 2013;86:443-451.
185. Olcina M, Lecane PS, Hammond EM. Targeting hypoxic cells through the DNA damage response. Clin Cancer Res. 2010;16:5624-5629.


Novel Strategies to Effectively De-escalate

Curative-Intent Therapy for Patients With
HPV-Associated Oropharyngeal Cancer: Current
and Future Directions
Katharine A. R. Price, MD1; Anthony C. Nichols, MD2; Colette J. Shen, MD, PhD3; Almoaidbellah Rammal, MD2; Pencilla Lang, MD, PhD4;
David A. Palma, MD, PhD, MSc4; Ari J. Rosenberg, MD5; Bhisham S. Chera, MD3; and Nishant Agrawal, MD6
overview
The treatment of patients with HPV-associated oropharyngeal cancer (HPV-OPC) is rapidly evolving and
challenging the standard of care of definitive radiotherapy with concurrent cisplatin. There are numerous
promising de-escalation strategies under investigation, including deintensified definitive chemoradiotherapy,
transoral surgery followed by de-escalated adjuvant therapy, and induction chemotherapy followed by de-
escalated locoregional therapy. Definitive radiotherapy alone or with cetuximab is not recommended for
curative-intent treatment of patients with locally advanced HPV-OPC. The results of ongoing phase III studies
are awaited to help answer key questions and address ongoing controversies to transform the treatment of
patients with HPV-OPC. Strategies for de-escalation under investigation include the incorporation of im-
munotherapy and the use of novel biomarkers for patient selection for de-escalation.
INTRODUCTION RT exposure is critical to effective de-escalation, be-

Since the 1980s, the incidence of HPV-OPC has cause RT exposure is an important driver of short- and
steadily increased, and it is now the eighth most long-term side effects, specifically those that have
common cancer in men in the United States.1 Because been shown to impact patient global QoL (xerostomia
the prognosis and cancer control outcomes are ex- and dysphagia).8 Although the preferred standard
cellent in HPV-OPC and patients with HPV-OPC tend to treatment of HPV-OPC varies between institutions,
be younger and healthier than patients with smoking- definitive RT with or without chemotherapy has served
induced head and neck cancer,2-4 there is great en- as the most commonly used treatment modality across
thusiasm for evaluating novel less-intensive cancer academic and community centers. Definitive RT to
treatment regimens to decrease long-term toxicities a standard dose of 70 Gy is associated with substantial
and improve quality of life (QoL). The landscape of acute and long-term toxicities, including mucositis,
clinical trials for HPV-OPC is being transformed by dysphagia, xerostomia, dysgeusia, lymphedema, and
various deintensification strategies under investigation fibrosis, and concurrent chemotherapy increases
at the institutional and cooperative group levels. Such short- and long-term side effects.9-11 Understanding
strategies include de-escalation of definitive radio- the clinical behavior of HPV-OPC can help limit RT
therapy (RT) or chemoradiotherapy (CRT), less morbid exposure, such as limiting radiation to the ipsilateral
surgical resection through transoral surgery (TOS) neck for patients with well-lateralized tonsil cancers,12,13
Author affiliations
followed by pathologic risk-adapted adjuvant therapy, underscoring the importance of patients with HPV-OPC
and support
information (if
and induction chemotherapy followed by risk-adapted being treated by experienced head and neck radiation
applicable) appear definitive therapy. In this article, we will discuss pub- oncologists.
at the end of this lished and ongoing efforts at treatment deintensification
article.
Deintensification of CRT regimens can involve re-
and potential future strategies for identifying and dein-
Accepted on
duction in the dose and/or volume of RT as well as
tensifying therapy for select patients with HPV-OPC.
February 20, 2020 reduction or omission of concurrent chemotherapy;
and published at ROLE OF DE-ESCALATION OF DEFINITIVE RT IN several phase II/III trials evaluating such regimens are
ascopubs.org on TREATMENT DEINTENSIFICATION FOR HPV-OPC summarized in Table 1. The University of North Car-
March 26, 2020:
HPV-OPC has been shown to be more responsive to olina/University of Florida approach has been to re-
10.1200/EDBK_ RT, thus making deintensification of the RT compo- duce the RT and chemotherapy doses without
280687 nent of treatment a compelling strategy.5-7 Decreasing definitive surgery or induction chemotherapy. RT is

Price et al
arms.17 The follow-up trial is NRG HN005 (NCT03952585),

which is a randomized phase II/III trial comparing three
arms for the same patient population: standard 70-Gy RT
• Cisplatin concurrent with radiotherapy remains with concurrent cisplatin, 60-Gy RT with concurrent cis-
the standard of care for treatment of patients
platin, and 60-Gy RT with concurrent nivolumab.
with locally advanced HPV-OPC, although it is
being challenged by de-escalation clinical trials. Memorial Sloan Kettering Cancer Center is utilizing hypoxia
• Multiple de-escalation strategies in patients with imaging to guide de-escalation, with the rationale that
favorable-risk HPV-OPC have excellent short- normoxic tumors may be controlled with lower doses of RT
term survival outcomes, but phase III data are compared with hypoxic tumors.18,19 The initial study eval-
needed before any of these can be adopted as uated a dose reduction of 10 Gy to metastatic lymph nodes
standard of care. (LNs) for patients with pretreatment hypoxia who had res-
• Cetuximab with radiotherapy is inferior to cis- olution of hypoxia early in treatment.20 With a median follow-
platin for definitive treatment of patients with up of 32 months, the 2-year locoregional control, distant
HPV-OPC. metastasis-free survival, and OS were 100%, 97%, and
• Improvements in quality of life, toxicity, and 100%, respectively. Their current study is evaluating
functional outcomes are evident with treatment ultra-deintensification to 30-Gy RT with concurrent cisplatin
de-escalation. in patients with HPV-OPC without hypoxia on imaging
(NCT03323463).
Advances in the science of radiation delivery have already
reduced to 60 Gy (from standard 70 Gy), and concurrent
improved long-term toxicities from head and neck cancer
cisplatin is reduced to 30 mg/m2 weekly for 6 weeks (from
treatment, with the replacement of external beam RT with
standard cumulative dose of 300 mg/m2, a 40% reduction).
intensity-modulated RT, and they may continue to play
The initial pilot phase II study (NCT01530997) enrolled 44
a role in treatment de-intensification with novel RT tech-
patients with HPV-OPC with minimal smoking history (or
niques.21 Intensity-modulated RT is now standard of care
moderate smoking history and abstinent for at least
for treatment of OPC and allows sparing of critical normal
5 years).14,15 All patients had a biopsy of the primary site and
tissues from the high-dose RT volume. Proton therapy may
supraselective neck dissection at 6 to 14 weeks after CRT.
allow additional sparing of normal tissues from low-dose
The primary endpoint of pathologic complete response (CR)
RT volumes, and it is currently being evaluated in a ran-
was 86%; with a median follow-up of 36 months, the 3-year
domized trial of intensity-modulated RT versus intensity-
locoregional control, cancer-specific survival, and overall
modulated proton therapy led by MD Anderson Cancer
survival (OS) were 100%, 100%, and 95%, respectively.
Center (NCT01893307).
The subsequent phase II trial (NCT02281955) of 114 pa-
tients, in which a 12-week post-CRT PET scan was used to ROLE OF TRANSORAL SURGERY IN TREATMENT
guide surgical evaluation (no required neck dissection), DEINTENSIFICATION FOR HPV-OPC
confirmed similar disease control and survival outcomes as
Primary surgical therapy for head and neck squamous cell
well as favorable long-term QoL, with patient-reported swal-
carcinoma (HNSCC) was the mainstay of treatment until the
lowing function returning to baseline.16 The current trial
advent of nonoperative therapy for organ preservation in the
(NCT03077243) is using tumor genomics to select which
early 1990s.22 Concerns about toxicity associated with
patients with more extensive smoking history can receive
definitive RT doses, paired with the development of mini-
deintensified treatment. In this trial, patients whose tumors
mally invasive TOS techniques, including transoral robotic
harbor a p53 mutation receive standard-dose treatment
surgery (TORS) and transoral laser microsurgery, have led to
with the hypothesis that the presence of a p53 mutation
a dramatic increase in the use of primary surgery for HPV-
may implicate tobacco as the primary driver of tumor
OPC.23 Retrospective studies looking at the use of TOS for
development.
early T-stage OPC have demonstrated that TOS provides
Based on this approach, NRG Oncology has conducted excellent QoL and functional outcomes with similar survival
HN002, a phase II randomized study (NCT02254278) outcomes to primary RT.23,24 A primary surgical approach
comparing 60-Gy RT with concurrent weekly cisplatin has the benefit of precise pathologic staging and assess-
40 mg/m2 (6 weeks) versus 60-Gy RT only (5 weeks) in ment of risk factors on which to base decisions for sub-
patients with HPV-OPC and minimal smoking history. Only sequent therapy. The use of primary TOS allows for a select
the CRT arm met the prespecified progression-free survival subset of patients with early-stage tumors (T1-2) and
(PFS) estimate, with 2-year PFS rates of 90.5% in the 60-Gy minimal nodal involvement (0–1 LNs without extranodal
RT + cisplatin arm and 87.6% in the 60-Gy RT only arm, extension [ENE]) to be treated with surgery alone, thus
although 2-year OS results were very similar at 97% in both completely avoiding the toxicity of RT or CRT.25 However,

Novel De-escalation Strategies for HPV-Associated Oropharyngeal Cancer
TABLE 1. Trials of De-escalation of Definitive RT-Based Regimens for HPV-OPC

Primary
Trial, Status, Accrual Date Trial Design Key Inclusion Criteria Treatment Arms Endpoint
UNC-UF (NCT01530997), Single-arm phase II T0-T3, N0-N2c (AJCC 7th 60-Gy RT + weekly cisplatin 30 mg/m2 Pathologic CR
completed, 2011–2014 trial (N = 44) ed.)
HPV/p16 positive
10 pack-year smoking
history, or 30 pack-
year and abstinent
5 years
UNC-UF (NCT02281955), Single-arm phase II T0-T3, N0-N2c (AJCC 7th 60-Gy RT + weekly cisplatin 30 mg/m2 (RT only for 2-year PFS
completed, 2014–2019 trial (N = 114) ed.) T0-T2, N0-N1)
HPV/p16 positive
history or 30 pack-
year and abstinent
5 years
UNC-UF (NCT03077243), Single-arm phase II T0-T3, N0-N2c (AJCC 7th As above if 10 pack-year smoking history or 2-year PFS
active trial (N = 120) ed.) . 10 pack-year and p53 wild-type
HPV/p16 positive If . 10 pack-year smoking history and p53
mutated: 70-Gy RT + weekly cisplatin,
Any smoking history
30 mg/m2
NRG HN002 (NCT02254278), Phase II randomized T1-T2, N1-N2b; T3, N0- Randomized: 60-Gy RT + weekly cisplatin, 2-year PFS
active, completed accrual, trial (1:1) (N = N2b (AJCC 7th ed.) 40 mg/m2 vs. 60-Gy RT
2014–2019 295)
HPV/p16 positive
NRG HN005 (NCT03952585), Phase II/III T1-T2, N1; T3, N0-N1 Randomized: 70-Gy RT + cisplatin, 100 mg/m2 PFS, MDADI
active randomized trial (AJCC 8th ed.) 2, or 60-Gy RT + cisplatin, 100 mg/m2 2, or global QoL
(1:1:1) (N = 711) 60-Gy RT + nivolumab, 240 mg every 2 weeks score
p16 positive
MSKCC (NCT03323463), Single-arm phase II T1-T2, N1-N2c (AJCC 7th 30-Gy RT + cisplatin, 100 mg/m2 2 cycles 2-year PFS
active trial (N = 150) ed.)
HPV positive
Hypoxia negative
MD Anderson Cancer Center Phase II/III Stage III-IVA,B (AJCC 7th Randomized: 70-Gy IMRT vs. 70-Gy IMPT; 2-year grade 3+
(NCT01893307), active randomized trial ed.) HPV/p16 positive chemotherapy as indicated toxicity
(N = 360)
Abbreviations: RT, radiotherapy; HPV-OPC, HPV-associated oropharyngeal cancer; UNC, University of North Carolina; UF, University of Florida; AJCC,
American Joint Committee on Cancer; CR, complete response; PFS, progression-free survival; MDADI, MD Anderson Dysphagia Inventory; QoL, quality of life;
MSKCC, Memorial Sloan Kettering Cancer Center; IMRT, intensity-modulated radiotherapy; IMPT, intensity-modulated proton therapy.
many patients present with multiple nodes or have adverse great interest in deintensification trials utilizing novel in-
pathologic features, which require standard-dose adjuvant duction approaches and/or decreased doses of adjuvant RT
RT/CRT after surgery.25 Therefore, the majority of patients or CRT.28 Here, we discuss the evidence for the use of TOS
undergoing primary TOS and neck dissection ultimately in HPV-OPC and review the de-escalation strategies cur-
receive dual- or triple-modality treatment.25 Because the rently being studied.
standard indications for adjuvant therapy and the doses
used were established in a predominantly HPV-negative Randomized Trials Comparing Primary Transoral Surgery
population, it is unclear whether all patients with HPV-OPC With or Without Adjuvant Therapy With Primary RT or CRT
require the same intensity of treatment after surgery.26,27 As with HPV-negative HNSCC, it remains unclear whether
Consequently, the application of these same indications to a primary surgical or RT approach is superior for the
patients with HPV-OPC has been questioned and has led to treatment of locally advanced HPV-OPC, with a paucity of

Price et al
prospective randomized trials to guide decision-making. A aggressive treatment of N1 neck disease in both arms and
summary of randomized clinical trials comparing primary the use of elective tracheostomies in a portion of patients,
TOS and RT approaches is shown in Table 2. The ORATOR thus potentially increasing surgical morbidity.30
study is the only completed randomized trial to date, having
Two phase III trials with a QoL endpoint comparing TOS and
randomly assigned 68 patients with T1-2N0-2 OPC (of
RT in the HPV-OPC population are currently ongoing. Both
which 88% were HPV positive) to TORS and ND, with or
trials attempt to minimize the use of adjuvant RT by
without adjuvant therapy, versus RT (with or without che-
selecting for patients with less advanced nodal disease. The
motherapy).29 The primary endpoint was swallowing QoL, as
“Best Of” trial limits participation to patients with clinically
measured by the MD Anderson Dysphagia Inventory
node-negative disease (NCT02984410). The QoLATI trial
(known as "MDADI") score, at 1 year. The study found
uses a staging neck dissection prior to the planned TORS for
a statistically significant, but not clinically significant, dif-
patients who are clinically node positive (NCT04124198).
ference in MDADI scores favoring the RT arm (p = .042).
Patients found to have ENE or more than two LN metastases
There were other important toxicity differences between the
on neck dissection are referred for definitive (C)RT rather
two arms, with more neutropenia, tinnitus, and hearing loss
than TORS. The results of these trials will help inform future
in the RT arm and more pain, pain medication use, and
treatment strategies and trial design.
trismus in the surgery arm. The strengths of the trial include
being the first successful randomized trial comparing TORS With the recognition of the importance of HPV status in
with primary RT, and the trial was adequately powered for its prognosis and treatment response, the research community
primary endpoint of QoL. Criticisms of the trial include the has moved toward de-escalation of RT doses in primary RT
modest sample size and relatively short follow-up as well as and adjuvant RT.31 ORATOR2 (NCT03210103) is the only
TABLE 2. Randomized Clinical Trials Comparing Primary RT and Primary TOS Approaches in HPV-OPC
Key
Trial Name, Inclusion Primary
Status, Accrual* Criteria† Trial Design Primary RT Arm‡ Primary TOS Arm¶ Endpoint
ORATOR,8 T1-2N0-2; Phase II trial (C)RT (70 Gy); TORS; adjuvant RT (60 Gy) for pT3-4 MDADI at
completed, N = 88% randomly concurrent chemotherapy if node disease, close margins (, 2 mm), 1 year
68 HPV+ assigning patients positive nodal disease, or LVI, with a higher
1:1 to primary RT dose (64 Gy) and concurrent
vs. primary TORS chemotherapy if positive margins
or ENE
ORATOR2,13 T1-2N0-2 Phase II trial De-escalated (C)RT (60 Gy): TOS 6 de-escalated adjuvant RT: OS at 2 years
recruiting, N = randomly concurrent weekly cisplatin if 60 Gy if ENE or positive margins;
140 (target) assigning patients multiple LN or single LN . 3 cm 50 Gy if close margins (, 3
1:1 to primary RT Accelerated RT if no chemotherapy mm), . 1 LN or LN . 3 cm, LVI, or
vs. primary TORS given pT3-4 disease
Best Of,10 T1-2N0 Phase III trial Accelerated RT (66-70 Gy); TOS; adjuvant (C)RT if margins MDADI at 4,5,
recruiting, N = randomly no chemotherapy remain positive after surgical 6, 9, and
170 (target) assigning patients revision 12 months
1:1 to RT vs. TOS
QoLATI,11 T1-2N0-1 Phase III trial T1N0: accelerated RT (66 Gy), TOS; patients with cN+ neck receive MDADI at 3
recruiting, N = (AJCC 8th randomly concurrent nimorazole; staging ND prior to TORS, if . 2 and
138 (target) ed.) assigning patients T2N0: accelerated RT (66 Gy), nodes or ENE, will receive (C)RT 12 months
2:1 to TORS vs. RT concurrent weekly cisplatin and instead of TORS; adjuvant RT for
nimorazole OR hyperfractionated, . 2 LN, 2 LN . 1 cm, ENE, , 10
accelerated RT (76 Gy in 56 LN after dissection, close margins
fractions), concurrent nimorazole; (, 2 cm), pT3-4 disease
N+ disease: accelerated RT (66 Gy),
concurrent nimorazole and cisplatin
Abbreviations: RT, radiotherapy; TOS, transoral surgery; HPV-OPC, HPV-related oropharyngeal cancer; HPV+, HPV positive; TORS, transoral robotic surgery;
CRT, chemoradiotherapy; LVI, lymphovascular invasion; ENE, extranodal extension; MDADI, MD Anderson Dysphagia Inventory; LN, lymph node; OS, overall
survival; AJCC, American Joint Committee on Cancer; N+, lymph node positive; cN+, clinically node positive; ND, neck dissection.
*Recruitment status as of December 29, 2019.
†Staging using the AJCC 7th edition unless otherwise specified, because the majority of trials were registered before the AJCC 8th edition was published.
‡RT given in 2-Gy fractions with five weekly fractions unless otherwise specified. Accelerated RT is defined as six fractions per week.
¶All TOS include ND.

ongoing randomized trial comparing RT and TOS with a de- up phase III trial, DART-HPV (NCT02908477), is randomly
escalated RT dose paradigm in both arms. In this phase II assigning patients with T1-3N0-3M0, p16-positive OPC with
randomized trial, patients receiving primary RT are treated any smoking history to the 2-week regimen, as established
to a dose of 60 Gy, and patients undergoing TOS receive 50 in MC1273, versus standard-of-care therapy (6 weeks of RT
or 60 Gy based on the absence or presence of ENE or or CRT with weekly cisplatin), with a primary endpoint of
positive surgical margins, respectively. The concurrent 2-year adverse event rate and a secondary endpoint of
systemic therapy is also de-escalated in both arms: con- survival.
current weekly cisplatin (40 mg/m2) is given concurrently
Extranodal Extension in the Era of HPV-OPC
with RT in the primary RT arm, and no concurrent che-
motherapy is used with adjuvant RT, even in the setting of Although ENE is a historical indication for CRT after surgery,
positive margins or ENE. By design, the trial limits each arm the data demonstrating a survival benefit are based on
to a maximum of two treatment modalities and avoids tri- a post hoc analysis of two trials completed before the HPV
modality therapy. The primary endpoint of ORATOR2 is 2- era and not limited to the oropharynx, and the strength of
year OS, and the results will be used to direct the design of this evidence has been questioned in the setting of HPV-
a definitive phase III trial. associated disease.26,27 In addition, ENE remains difficult to
predict based on preoperative radiographic and clinical
Trials of Transoral Surgery Followed by De-escalated
assessment.33 Combined, this means that some patients will
Adjuvant Therapy
have classic indications for adjuvant CRT resulting in triple-
Several randomized trials are examining the use of TOS with modality therapy, potentially leading to serious toxicity. The
de-escalated adjuvant treatment in patients with HPV-OPC significance of ENE in the HPV-OPC population has also
(Table 3). ECOG-3311 is a cooperative group phase III trial been questioned.34 ENE has lower prognostic value in
that randomly assigned patients with intermediate risk patients with HPV-OPC, and, in the revised staging system
factors after TOS (close margin, 1 mm ENE, 2–4 meta- (American Joint Committee on Cancer [AJCC] 8th edition),
static LNs or perineural invasion, and lymphovascular in- ENE is not included as a prognostic factor for p16-positive
vasion) to 50-Gy RT versus 60-Gy RT with a primary disease.35 However, a multi-institutional retrospective study
endpoint of 2-year PFS (NCT01898494). Patients with no of patients with HPV-OPC who underwent TORS and then
adverse pathologic features do not receive any adjuvant refused standard adjuvant therapy reported a 3-year relapse
treatment, and patients with high-risk factors (positive rate of 52% in patients with high-risk pathologic features,
margin, ENE . 1 mm, . 5 LNs) receive standard CRT. suggesting that the presence or absence of ENE may still be
Accrual of 511 patients was completed in 2017, and results important.36 Whether high-risk pathologic features warrant
are anticipated in 2020. Similarly, the PATHOS trial adjuvant therapy with RT and chemotherapy remains un-
(NCT02215265) randomly assigned patients with intermediate- clear, because retrospective data suggest that there may not
risk factors (close margin, advanced T or N stage, and be a survival difference between adjuvant RT and CRT in the
lymphovascular invasion/perineural invasion) to adjuvant HPV-OPC population.37 As noted above, the ADEPT trial
RT with 50 Gy or 60 Gy, and it randomly assigned patients attempted to randomly assign patients with ENE after TOS to
with high-risk factors (positive margin, ENE) to 60-Gy RT RT versus CRT; unfortunately, it closed prematurely as
with concurrent cisplatin or RT alone. The primary endpoint a result of poor accrual. Fortunately, the ECOG-3311, PA-
for the phase II component of the trial is the MDADI score at THOS, and DART-HPV trials seek to better clarify the op-
1 year, whereas the primary endpoint for the phase III timal adjuvant therapy after surgery. In the absence of
component is OS. The ADEPT trial (NCT01687413) was definitive prospective data to guide treatment decisions,
a randomized phase III trial designed to compare adjuvant the addition of cisplatin to RT after surgery for high-risk
CRT with RT alone for patients with resected HPV-OPC pathologic features still remains part of standard-of-care
found to have ENE. Unfortunately, this trial closed as a result recommendations.
of poor accrual. MC1273 (NCT01932697) was a phase II trial
of patients with resected p16-positive OPC with negative Other Ongoing Controversies and Future Directions
margins and a minimal smoking history ( 10 pack-years) Surrounding Transoral Surgery
treated with 2 weeks of adjuvant therapy (twice-daily RT with The argument that bimodality therapy is preferable to tri-
two radiosensitizing doses of docetaxel, 15 mg/m2); it had modality therapy has been used against surgical approaches
excellent toxicity, functional, QoL, and survival outcomes, for HPV-OPC, yet the toxicity of trimodality treatment with
consistent with historical controls.32 The total radiation dose surgery, followed by standard-dose RT and high-dose cis-
was based on pathologic risk factors, with patients with platin, is inherently different from the toxicity of trimodality
intermediaterisk factors receiving 30 Gy to the primary site treatment that could include surgery followed by reduced-
and neck and those with ENE receiving 36 Gy, representing dose RT and alternative systemic therapy. Ultimately,
the lowest RT dose to date in the adjuvant setting. A follow- a randomized trial of proven de-escalation strategies using

Price et al
TABLE 3. Randomized Controlled Trials of HPV-OPC Treated With Surgery Followed by De-escalated Adjuvant Therapy
Trial Name, Status, Key Inclusion Adjuvant RT (with or without
Accrual* Criteria† Trial Design Risk Group Stratification chemotherapy)‡ Primary Endpoint
14
ECOG-3311, Stage III, IVa, or IVb, Phase III Low: no adverse features None 2-year PFS, LRC,
active, not no evidence of randomized and QoL
Intermediate: close Randomized: RT with 50 Gy vs. 60 Gy
recruiting, N = extensive or fixed/ trial
margin, , 1 mm ENE,
511 matted nodes
2–4 metastatic LNs,
PNI, LVI
High: positive margin, CRT with 66 Gy with concurrent cisplatin
. 1 mm ENE, 5
positive LNs
PATHOS,15 active, T1-3N0-N2b, Phase II/III Low: No adverse features None Phase II: MDADI
recruiting, N = , 10 pack-year randomized at 12 months
242 smoking history trial
Phase II/III Intermediate: close Randomized: RT with 50 Gy vs. 60 Gy Phase III: OS
randomized margins (1–5 mm),
trial pT3 or pT4, N2a or
N2b, PNI, LVI
High: positive margins Randomized: CRT with 60 Gy with
(, 1 mm) with concurrent cisplatin (every 3 weeks)
negative marginal vs. RT alone with 60 Gy
biopsies, ENE
DART-HPV,17 active, Post-TOS with 1 of: Phase III Randomized: CRT with 60 Gy with weekly Adverse event
recruiting, N = LN . 3 cm, 2 randomized cisplatin (standard arm) vs. 30–36 Gy rate at 2 years
214 LNs, PNI, LVI, pT3, trial in 1.5-Gy fractions twice a day with (grade 3–5
ENE concurrent low-dose docetaxel toxicity)
(experimental arm)
ADEPT,16 terminated T1-4a, N+ with ENE Phase III Randomized: RT (60 Gy) vs. CRT (60 Gy) DFS and LC at
early because of present, negative randomized with concurrent cisplatin (every 3 2 years
slow accrual, N = margins trial weeks)
41
NOTE. All TOS includes neck dissection.

Abbreviations: HPV-OPC, HPV-associated oraopharyngeal cancer; RT, radiotherapy; ENE, extranodal extension; LNs, lymph nodes; PNI, perineural invasion;
LVI, lymphovascular invasion; PFS, progression-free survival; LRC, locoregional control; QoL, quality of life; CRT, chemoradiotherapy; OS, overall survival;
TOS, transoral surgery; N+, lymph node positive; DFS, disease-free survival; LC, local control.
*Recruitment status as of December 29, 2019.
†Staging using American Joint Committee on Cancer (AJCC) 7th edition unless otherwise specified, because the majority of trials were registered before the
AJCC 8th edition was published.
‡
RT given in 2-Gy fractions, with five weekly fractions, unless otherwise specified.
surgical and nonsurgical approaches with toxicity and QoL acquisition of good prospective surgical data with which to
endpoints will be necessary. It is also unclear whether guide future treatment recommendations and clinical trials.
patients with multiple involved LNs with only intermedi- Collectively, the ongoing TOS trials will provide important
ate pathologic risk factors after TORS require adjuvant information about the rate of surgical complications, in-
therapy. The reported recurrence rate of patients with cluding hemorrhage, fistula, stroke, dysphagia, and post-
HPV-OPC who underwent TORS alone for pathologic operative mortality, in the hands of experienced surgeons.
intermediate-risk features was 11.7%, with a median This is important, because treatment-related complications
follow-up of 42.9 months.36 The majority of patients had tend to be under-reported in retrospective data.38 Another
pN2 (AJCC 7th edition) disease (68%), and 32% had pN0 important advancement in the context of these studies is
or pN1 node disease. As we refine our understanding of who the development of surgical credentialing and quality-assurance
is at risk for recurrence, there may be an opportunity to processes, which have been described by Simon et al.39
expand the selection of patients who can be treated with This will help ensure that the surgical outcomes reflect those
surgery alone. of expert clinicians, because volumes have been closely tied
The era of HPV-OPC has resulted in a resurgence of the to quality metrics, including positive margins, bleeding, and
use of surgery and, consequently, opportunities for the even survival.25,40

ROLE OF SYSTEMIC THERAPY IN DE-ESCALATION The University of Chicago approach also involves induction
STRATEGIES FOR PATIENTS WITH HPV-OPC chemotherapy, followed by risk-adaptive de-escalation of
Historically, systemic therapy as part of standard treat- RT volume and/or dose. Based on an initial trial in which
ment paradigms for HNSCC has had two distinct objec- patients with favorable response to induction cisplatin,
tives. The first is to act as a radiation sensitizer to improve paclitaxel, and cetuximab with or without everolimus re-
locoregional control, a strategy that has been used for ceived RT only to gross disease (no elective nodal RT),46 the
definitive and adjuvant CRT for decades.9,10,26,27,41 The OPTIMA trial (NCT02258659) evaluated dose and volume
second potential benefit of systemic therapy is to de- de-escalation after induction chemotherapy with carbo-
crease the risk of developing distant metastatic disease. platin and nab-paclitaxel based on clinical risk (high risk
defined by T4, N2c, . 10 pack-year smoking history) and
Although the evidence for this has been controversial in
response to induction therapy (see Table 4 for treatment
an unselected population of patients with HNSCC, meta-
allocation).47 The majority of low-risk patients received 50-
analyses have shown an absolute benefit of induction
Gy RT alone after induction chemotherapy, whereas the
chemotherapy in decreasing distant metastatic disease,
majority of high-risk patients received CRT with 45-Gy RT.
albeit without improving survival.42 For OPC specifically,
With a median follow-up of 29 months, the overall 2-year
there is evidence that would suggest a benefit to in-
PFS was 94.5% (95% for low risk and 94% for high risk).
tensifying systemic therapy as part of definitive treatment.
Grade 3+ mucositis developed in 30%, 63%, and 91% of
The GETTEC trial demonstrated a decrease in distant
patients receiving 50-Gy RT only, 45-Gy CRT, and 75-Gy
metastases with induction cisplatin and 5-fluorouracil for
CRT, respectively.
patients with OPC; HPV status was not reported in this
study.43 Investigators at Mount Sinai conducted a randomized phase
II trial of 23 patients (closed early because of poor accrual/
The role of induction chemotherapy as a means to de-
lack of financial support; NCT01706939) with HPV-OPC
escalate overall treatment toxicity is an area of active in-
( 20 pack-year smoking history) treated with induction
vestigation. In the setting of HPV-OPC, a third objective has
chemotherapy with docetaxel, cisplatin, and fluorouracil.48
emerged for systemic therapy: render gross disease mi-
Twenty patients with CR or partial response were selected
croscopic to allow for deintensified locoregional therapy.
randomly, at a 2:1 ratio, to receive 56-Gy or 70-Gy RT, both
HPV-OPC has been shown to have a better response to
arms with concurrent weekly carboplatin. The 3-year PFS
neoadjuvant chemotherapy compared with HPV-negative
was not significantly different between the two arms (87.5%
OPC,3 and several trials are evaluating reduction in definitive for 70 Gy, 83.3% for 56 Gy), but analysis was limited by the
RT dose for patients with good response to neoadjuvant small sample size.
chemotherapy (summarized in Table 4). The ECOG-E1308
trial evaluated induction chemotherapy with cisplatin, The use of neoadjuvant therapy before TOS has also been
paclitaxel, and cetuximab followed by response-adjusted RT a topic of increasing interest.49-51 The NECTORS study,
with concurrent cetuximab in patients with HPV-OPC.44 using neoadjuvant cisplatin and docetaxel before TOS in
Patients with a clinical CR (cCR) received reduced-dose patients with HPV-OPC, found high pathologic CR rates
RT to 54 Gy, whereas those with less than cCR received RT (72% and 57% at the primary and nodal disease sites,
to 69.3 Gy (primary and nodal sites evaluated and dosed respectively) and 3-year disease-specific survival of
separately based on response at each site). Of 80 evaluable 94%.51,52 OPTIMAII is a phase II trial (NCT03107182) in-
patients, 70% had a cCR at the primary site, and 58% had corporating induction chemoimmunotherapy (carboplatin,
a cCR at nodal sites. With a median follow-up of 35 months, nab-paclitaxel, and nivolumab) followed by risk- and response-
the 2-year PFS was 78% and the OS was 91% for all pa- adapted locoregional treatment (TOS, RT, or CRT). The
tients, and best outcomes were in patients with 10 pack- CIAO trial randomly assigned patients with HPV-positive
year smoking history (2-year PFS was 57% for patients and -negative OPC to durvalumab with or without trem-
with . 10 pack-year smoking history). elimumab followed by TOS.53 Promising interim results
revealed that, with this strategy, the majority of patients
Investigators at University of California, Los Angeles/Uni- (17/28, 60.7%) required surgery alone, with no adjuvant
versity of California, Davis conducted a phase II trial eval- therapy.
uating response-adapted RT after induction chemotherapy
with paclitaxel and carboplatin (NCT01716195).45 Patients Strategies To De-intensify Systemic Therapy for Patients
with a CR or partial response received 54-Gy RT, whereas With HPV-OPC
patients with less than partial response received 60-Gy RT, What have been the ongoing strategies that have focused on
all with concurrent weekly paclitaxel. At a median follow-up changes in systemic therapy to drive de-escalation beyond
of 30 months, 2-year PFS was 92%, and grade 3 dysphagia induction chemotherapy? Two strategies, eliminating cis-
and mucositis were less than 10%. platin altogether and substituting cetuximab for cisplatin,

Price et al
TABLE 4. Trials of Induction Chemotherapy Followed by De-escalation of CRT for HPV-OPC

Primary
Trial/Status/Accrual Trial Design Key Inclusion Criteria Treatment Arms End Point
ECOG-1308 (NCT01084083), Single-arm phase II trial Stage III-IVA,B (AJCC 7th ed.); Paclitaxel, 90 mg/m2 + cisplatin, 75 mg/m2 + 2-year
completed (accrued HPV/p16 positive; any smoking cetuximab, 250 mg/m2 3 cycles → 54- PFS
2010–2011), history Gy RT for cCR, 69.3-Gy RT for , cCR +
N = 90 (80 evaluable patients) weekly cetuximab, 250 mg/m2
UCLA-UC Davis Single-arm phase II trial Stage III-IV (AJCC 7th ed.); p16 Paclitaxel, 175 mg/m2 + carboplatin 2-year
(NCT01716195), completed positive; any smoking history (AUC = 6) 2 cycles → 54-Gy RT for CR/ PFS
(accrued 2012–2015), N = 45 PR, 60-Gy RT for , PR + weekly
paclitaxel, 30 mg/m2
OPTIMA (University of Chicago; Nonrandomized phase T3-T4 or N2-N3 (AJCC 7th ed.); Carboplatin (AUC = 6) + nab-paclitaxel, 2-year
NCT02258659), completed II trial, parallel HPV/p16 positive; 100 mg/m2 3 cycles → PFS
(accrued 2014–2016), N = 62 assignment any smoking history; low risk: 50-Gy RT alone for low risk with 50%
T1-3N0-2B and 10 pack- response
year; high risk: T4N2C-3 or 45-Gy RT + TFHX 3 cycles for low
bulky N2B or . 10 pack-year risk with , 50% response but 30%
response and high risk with 50%
response
75-Gy RT + TFHX 5 cycles for low
risk with , 30% response, high risk
with , 50% response, or any patient with
progression
Quarterback (Mount Sinai; Phase II randomized Stage III-IV (AJCC 7th ed.); HPV/ Docetaxel, 75 mg/m2 + cisplatin, 100 3-year
NCT01706939), closed early trial (2:1) p16 positive; 20 pack-year mg/m2 + 5-fluorouracil, 750 mg/m2 3 PFS
(2012–2018), N = 23 smoking history cycles → randomly assigned (if clinical
response) to 70-Gy RT vs. 56-Gy RT +
weekly carboplatin (AUC = 1.5)
Abbreviations: CRT, chemoradiotherapy; HPV-OPC, HPV-related oropharyngeal cancer; AJCC, American Joint Committee on Cancer; RT, radiotherapy; cCR,
clinical complete response; PFS, progression-free survival; UCLA, University of California, Los Angeles; UC-Davis, University of California, Davis; CR, complete
response; PR, partial response; AUC, area under the curve; TFHX, paclitaxel, 100 mg/m2 on day 1, 5-fluorouracil continuous infusion at 600 mg/m2/day on
days 0 through 5, and hydroxyurea, 500 mg orally twice daily on days 0 through 5, with 11 doses per cycle.
have been studied in prospective trials and have failed to RT alone,56 making cetuximab an attractive candidate for
reach their primary survival endpoints. The elimination of de-intensification regimens. Unfortunately, cetuximab has
concurrent chemotherapy is understandably an attractive been found to be inferior to cisplatin as part of a CRT
strategy to de-escalate treatment of patients with HPV-OPC, regimen for HPV-OPC.57,58 RTOG 1016 randomly assigned
because the addition of chemotherapy to RT significantly patients with locally advanced HPV-OPC to cetuximab or
increases acute and long-term toxicities.10,11,54,55 Retro- cisplatin along with standard-dose RT (70 Gy), but inferior
spective data have suggested that the addition of cisplatin to survival was found in patients treated on the non-cisplatin
RT in the adjuvant setting for patients with resected HPV- arm.57 With a median follow-up of 4.5 years, 5-year OS was
OPC provides no additional benefit.28 However, prospective 84.6% in the cisplatin arm compared with 77.9% in the
data from NRG-HN002 would suggest a benefit for systemic cetuximab arm, and PFS was 78.4% and 67.3%, re-
chemotherapy, as discussed in the previous section.17 The spectively. Locoregional failure was more than doubled in
need for concurrent chemotherapy with definitive RT for the cetuximab arm. From the standpoint of toxicity, there
curative-intent treatment of patients with HPV-OPC remains was no difference in late toxicity between the cisplatin and
standard of care. cetuximab arms. The DE-ESCALATE trial conducted in
Cisplatin chemotherapy has been the standard-of-care Europe had similar results, with the cisplatin/RT arm showing
agent, along with RT, for the curative treatment of HNSCC superior 2-year OS compared with cetuximab/RT (97.5% vs.
for decades, but it has important chemotherapy-specific 89.4%); no long-term improvement in toxicity was seen with
short- and long-term toxicities, notably nausea, ototoxicity, cetuximab.58 The ICON-S study found that EGFR inhibitors
and nephrotoxicity. In 2006, the EGFR inhibitor cetuximab had no survival benefit compared with RT alone or surgery,
and RT became an option for definitive treatment in an HPV- with or without RT, for patients with HPV-OPC, providing
unselected population of patients with HNSCC, with the more evidence that cetuximab is inferior to cisplatin in this
initial publication reporting no increase in acute toxicity over patient population.35

Because data support the ongoing use of cisplatin with RT, importance of longer follow-up of disease outcomes in de-
could the dosing of cisplatin be altered to improve toxicity? escalation trials with attention to distant recurrences. As
The use of two doses of high-dose cisplatin concurrently effective de-escalation strategies for locoregional control for
with accelerated RT has been shown to be as effective as patients with HPV-OPC emerge and we better define which
three doses with standard-fractionation RT and is an ac- patient subsets are at risk for not responding to treatment
ceptable treatment alternative for patients with HPV-OPC, through the incorporation of biomarkers, such as cancer
resulting in a 33% decrease in cumulative cisplatin genetics (PIK3CA, TP53, and 3p arm loss mutations) and
dose.57,59 Decreasing the dose of cisplatin as part of circulating tumor DNA,16,67-70 novel strategies that intensify
curative-intent CRT, from standard high dose (100 mg/m2) systemic therapy for those patients at risk for locoregional
to weekly dosing (30–40 mg/m2), is widely used with en- and distant failure are crucial and should be a focus in the
couraging survival results,60,61 although a prospective ran- next generation of clinical trials.
domized comparison has not been conducted in the HPV-
Immunotherapy as Part of De-escalation for Patients
OPC population. A small retrospective study in an HPV-
With HPV-OPC
specific population found a higher rate of distant failure in
patients treated with weekly cisplatin, although OS was Immune checkpoint inhibition with PD-1 inhibitors pem-
similar, underscoring the need for careful study and the brolizumab and nivolumab is now part of standard therapy
potential importance of systemic therapy in the treatment of for recurrent or metastatic HNSCC in the first-line setting,
patients with HPV-OPC (see next section).61 In the adjuvant with or without platinum-based chemotherapy, based on
setting, long-term follow-up of patients with resected HPV- PD-L1 expression (combined positive score), and in the
OPC treated with weekly cisplatin had excellent survival second-line setting after failure of platinum chemotherapy.71-74
outcomes. 62,63 Collective evidence would suggest that Notably, immunotherapy has an important OS benefit over
weekly cisplatin is a reasonable alternative to high-dose chemotherapy alone or chemotherapy in combination with
cisplatin, recognizing that weekly cisplatin is not supported cetuximab.71 It stands to reason that incorporating immuno-
by prospective randomized data. Higher rates of hemato- therapy into the curative-intent setting is a logical step forward,
logic toxicity, nausea, and kidney dysfunction have been and the use of immune checkpoint inhibition as part of de-
reported with high-dose cisplatin,60 but whether weekly escalation strategies for patients with HPV-OPC is underway.
cisplatin ultimately results in a distinguishable reduction in The safety and feasibility of incorporating immunotherapy
long-term toxicity is not known. The RADIO trial (NCT03649048), with concurrent RT and cisplatin were demonstrated with the
a randomized trial comparing weekly (40 mg/m2) versus early results of HN003.75 There are several ongoing trials
high-dose (100 mg/m2) cisplatin with a primary endpoint of incorporating immunotherapy with definitive RT for patients
hearing loss and hearing-related QoL, will answer some of with HPV-OPC, including HN005, which is comparing stan-
these questions. dard CRT plus cisplatin with reduced-dose RT plus nivolumab
or cisplatin for nonsmoking patients with HPV-OPC
Intensification of Systemic Therapy for Select Patients
(NCT03952585); NCT03410615, which is investigating
With HPV-OPC
standard cisplatin plus RT with durvalumab plus RT, fol-
Historically, locoregional control for HNSCC has been the lowed by adjuvant durvalumab or the combination of
primary therapeutic focus, given the high rate of locore- tremelimumab and durvalumab; a phase II study combining
gional failure and lower rate of distant metastatic disease nivolumab and ipilimumab with reduced-dose RT for low-
seen in patients with HPV-negative disease. In the era of intermediate volume HPV-OPC (NCT03799445); and de-
HPV-OPC, distant metastatic disease has emerged as the escalated (C)RT after induction nivolumab with chemo-
more pressing concern, given the high success rate of therapy (NCT03107182).
locoregional therapies (radiation and surgery) and salvage
treatment options for locoregional failure.3 Indeed, across all LESSONS LEARNED TO DATE
clinical risk groups, distant failure predominates for patients The treatment of patients with HPV-OPC is rapidly
with HPV-OPC.3 Recent work has identified subsets of evolving, with numerous promising de-escalation
patients who are at high risk for distant failure, including strategies under investigation (survival results of pub-
those with T4 or N3 disease (stage III, AJCC 8th ed.), pa- lished studies are summarized in Table 5), some of which
tients with involved lower neck LNs (5-year OS, 39% vs. are likely to become part of standard-of-care recommen-
67% with positive vs. negative lower cervical LNs), and dations in the future, pending results of phase III studies. We
patients with five or more involved cervical LNs.35,64,65 Al- know that there are subgroups of patients with HPV-OPC
though the reported survival rates for patients with HPV-OPC with inferior survival based on clinical characteristics
are high,2 we may see lower survival rates with long-term that should not be de-escalated, notably patients with
follow-up of patients with HPV-OPC driven by delayed T4 tumors or N3 nodal disease.35 Definitive RT alone or
distant recurrences.66 This observation underscores the with cetuximab is not recommended for curative-intent

Price et al
TABLE 5. Comparison of Survival Outcomes for Published De-escalation Trials

Study Key Eligibility Treatment PFS, % OS, %
NRG HN002 (NCT02254278), N = T1-2N1-N2bM0 or T3N0-N2bM0; 60-Gy RT + weekly cisplatin vs. 60-Gy RT 90.5 (2 96.7 (2
295 10 pack-year years) years)
MC1273 (NCT01932697), N = 80 Cohort A: pathologic intermediate-risk TOS → RT twice a day 2 weeks + docetaxel 91.1 (2 98.7 (2
features, margin negative; cohort B: (15 mg/m2); cohort A: 30 Gy; cohort B: 36 Gy years) years)
pathologic high-risk features, margin
negative
UNC-UF (NCT02281955), N = 114 T0-3N0-N2cM0; 10 pack-year or 60-Gy RT + weekly cisplatin, 30 mg/m2 (RT only 86 (2 95 (2
30 pack-year with 5 years abstinence for T0-T2, N0-N1) years) years)
E1308 (NCT01084083), N = 90 T1-3N0-N2bM0; 10 pack-year Induction chemo → 54 Gy RT + cetuximab if CR 80 (2 94 (2
(80 evaluable patients) years) years)
UCLA-UC Davis (NCT01716195), Stage III–IV (AJCC 7th ed.); p16 positive; Induction chemo → 54-Gy RT for CR/PR 92 (2 Not
N = 45 any smoking history years) reported
OPTIMA (University of Chicago) T3-T4 or N2-N3 (AJCC 7th ed.); Induction chemo → 50 Gy if 50% PR or CR 94.5 (2 Not
(NCT02258659), N = 62 HPV/p16 positive; years) reported
any smoking history
Abbreviations: PFS, progression-free survival; OS, overall survival; RT, radiotherapy; TOS, transoral surgery; UNC, University of North Carolina; UF, University
of Florida; chemo, chemotherapy; CR, complete response; UCLA, University of California, Los Angeles; UC Davis, University of California, Davis; AJCC,
American Joint Committee on Cancer; PR, partial response.
treatment of patients with locally advanced HPV-OPC; cis- than one single approach. The results of the de-intensification
platin concurrent with RT still remains the standard of care. trials to date have been encouraging, and we await results of
De-intensified treatment does appear to improve QoL and phase III studies to establish new standard therapy for future
functional outcomes based on early results from de-escalation patients with HPV-OPC. Novel predictors of patients at risk
trials, demonstrating that even modest reductions in RT for recurrence, such as hypoxia, circulating tumor DNA, and
doses can translate into meaningful clinical benefit for genomic data (PIK3CA, p53), are likely to be important for
patients.16,32,44,76 additional risk stratification in next-generation trials.68,77,78
CONCLUSION AND FUTURE DIRECTIONS Other patient factors that might impact outcomes could also
Since the publication of the VA Larynx Trial in 1991, there be considered when designing future trials, including
have been multiple acceptable approaches to curative-intent neutrophil/lymphocyte ratio,79,80 microbiome,81 body mass
therapy for locally advanced HNSCC.22 As we continue to index,82 nutritional markers,83 and comorbidities.84 Finally,
better refine our understanding of HPV-OPC and treatment the prevention of cancer is the ultimate de-escalation strat-
approaches, it is likely that the future will hold several egy, and we should not forget to encourage our patients
standard-of-care options for therapeutic de-escalation rather and their families to get vaccinated against HPV.

1
Division of Medical Oncology, Mayo Clinic, Rochester, MN Katharine A. R. Price, MD, Division of Medical Oncology-Head and Neck
2
Department of Otolaryngology–Head and Neck Surgery, University of Cancer Team, Mayo Clinic, 200 First St. SW, Rochester, MN 55905;
Western Ontario, London, Ontario, Canada email: [email protected].
3
Department of Radiation Oncology, University of North Carolina School of
Medicine, Chapel Hill, NC
4
Division of Radiation Oncology, Department of Oncology, University of AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Western Ontario, London, Ontario, Canada AND DATA AVAILABILITY STATEMENT
5
Section of Hematology/Oncology, Department of Medicine, University of Disclosures provided by the authors and data availability statement (if
Chicago, Chicago, IL applicable) are available with this article at DOI https://doi.org/10.1200/
6
Section of Otolaryngology–Head and Neck Surgery, Department of EDBK_280687.
Surgery, University of Chicago, Chicago, IL

REFERENCES
2. Ang KK, Harris J, Wheeler R, et al. Human papillomavirus and survival of patients with oropharyngeal cancer. N Engl J Med. 2010;363:24-35.
3. Fakhry C, Westra WH, Li S, et al. Improved survival of patients with human papillomavirus-positive head and neck squamous cell carcinoma in a prospective
clinical trial. J Natl Cancer Inst. 2008;100:261-269.
4. Chaturvedi AK, Engels EA, Pfeiffer RM, et al. Human papillomavirus and rising oropharyngeal cancer incidence in the United States. J Clin Oncol. 2011;
29:4294-4301.
5. Kimple RJ, Smith MA, Blitzer GC, et al. Enhanced radiation sensitivity in HPV-positive head and neck cancer. Cancer Res. 2013;73:4791-4800.
6. Rieckmann T, Tribius S, Grob TJ, et al. HNSCC cell lines positive for HPV and p16 possess higher cellular radiosensitivity due to an impaired DSB repair capacity.
7. Cleary C, Leeman JE, Higginson DS, et al. Biological features of human papillomavirus-related head and neck cancers contributing to improved response. Clin
Oncol (R Coll Radiol). 2016;28:467-474.
8. Langendijk JA, Doornaert P, Verdonck-de Leeuw IM, et al. Impact of late treatment-related toxicity on quality of life among patients with head and neck cancer
treated with radiotherapy. J Clin Oncol. 2008;26:3770-3776.
9. Forastiere AA, Goepfert H, Maor M, et al. Concurrent chemotherapy and radiotherapy for organ preservation in advanced laryngeal cancer. N Engl J Med. 2003;
349:2091-2098.
10. Adelstein DJ, Li Y, Adams GL, et al. An intergroup phase III comparison of standard radiation therapy and two schedules of concurrent chemoradiotherapy in
patients with unresectable squamous cell head and neck cancer. J Clin Oncol. 2003;21:92-98.
11. Trotti A, Pajak TF, Gwede CK, et al. TAME: development of a new method for summarising adverse events of cancer treatment by the Radiation Therapy Oncology
Group. Lancet Oncol. 2007;8:613-624.
12. Chin RI, Rao YJ, Hwang MY, et al. Comparison of unilateral versus bilateral intensity-modulated radiotherapy for surgically treated squamous cell carcinoma of the
palatine tonsil. Cancer. 2017;123:4594-4607.
13. Rackley TP, Namelo WC, Palaniappan N, et al. Unilateral radiotherapy for surgically resected lateralized squamous cell carcinoma of the tonsil. Head Neck.
2017;39:17-23.
14. Chera BS, Amdur RJ, Tepper J, et al. Phase 2 trial of de-intensified chemoradiation therapy for favorable-risk human papillomavirus-associated oropharyngeal
squamous cell carcinoma. Int J Radiat Oncol Biol Phys. 2015;93:976-985.
15. Chera BS, Amdur RJ, Tepper JE, et al. Mature results of a prospective study of deintensified chemoradiotherapy for low-risk human papillomavirus-associated
oropharyngeal squamous cell carcinoma. Cancer. 2018;124:2347-2354.
16. Chera BS, Amdur RJ, Green R, et al. Phase II trial of de-intensified chemoradiotherapy for human papillomavirus-associated oropharyngeal squamous cell
17. Yom SS, Torres-Saavedra P, Caudell JJ, et al. NRG-HN002: a randomized phase II trial for patients with p16-positive, non-smoking-associated, locoregionally
advanced oropharyngeal cancer. Int J Radiat Oncol Biol Phys. 2019;105:684-685.
18. Janssen HL, Haustermans KM, Balm AJ, et al. Hypoxia in head and neck cancer: how much, how important? Head Neck. 2005;27:622-638.
19. Vaupel P, Mayer A. Hypoxia in cancer: significance and impact on clinical outcome. Cancer Metastasis Rev. 2007;26:225-239.
20. Lee N, Schoder H, Beattie B, et al. Strategy of using intratreatment hypoxia imaging to selectively and safely guide radiation dose de-escalation concurrent with
chemotherapy for locoregionally advanced human papillomavirus-related oropharyngeal carcinoma. Int J Radiat Oncol Biol Phys. 2016;96:9-17.
21. Wang X, Eisbruch A. IMRT for head and neck cancer: reducing xerostomia and dysphagia. J Radiat Res (Tokyo). 2016;57 (suppl 1):i69-i75.
22. Wolf GT, Fisher SG, Hong WK, et al; Department of Veterans Affairs Laryngeal Cancer Study Group. Induction chemotherapy plus radiation compared with surgery
plus radiation in patients with advanced laryngeal cancer. N Engl J Med. 1991;324:1685-1690.
23. de Almeida JR, Byrd JK, Wu R, et al. A systematic review of transoral robotic surgery and radiotherapy for early oropharynx cancer: a systematic review.
Laryngoscope. 2014;124:2096-2102.
24. Moore EJ, Olsen KD, Kasperbauer JL. Transoral robotic surgery for oropharyngeal squamous cell carcinoma: a prospective study of feasibility and functional
outcomes. Laryngoscope. 2009;119:2156-2164.
25. Zhan KY, Puram SV, Li MM, et al. National treatment trends in human papillomavirus-positive oropharyngeal squamous cell carcinoma. Cancer. Epub 2019 Dec
11.
26. Cooper JS, Pajak TF, Forastiere AA, et al; Radiation Therapy Oncology Group 9501/Intergroup. Postoperative concurrent radiotherapy and chemotherapy for
high-risk squamous-cell carcinoma of the head and neck. N Engl J Med. 2004;350:1937-1944.
27. Bernier J, Domenge C, Ozsahin M, et al; European Organization for Research and Treatment of Cancer Trial 22931. Postoperative irradiation with or without
concomitant chemotherapy for locally advanced head and neck cancer. N Engl J Med. 2004;350:1945-1952.
28. Sinha P, Lewis JS Jr., Piccirillo JF, et al. Extracapsular spread and adjuvant therapy in human papillomavirus-related, p16-positive oropharyngeal carcinoma.
Cancer. 2012;118:3519-3530.
29. Nichols AC, Theurer J, Prisman E, et al. Radiotherapy versus transoral robotic surgery and neck dissection for oropharyngeal squamous cell carcinoma
(ORATOR): an open-label, phase 2, randomised trial. Lancet Oncol. 2019;20:1349-1359.

Price et al
30. Grégoire V, Nicolai P. Choosing surgery or radiotherapy for oropharyngeal squamous cell carcinoma: is the issue definitely settled? Lancet Oncol. 2019;
20:1328-1329.
31. Wirth LJ, Burtness B, Nathan CO, et al. Point/counterpoint: do we de-escalate treatment of HPV-associated oropharynx cancer now? And how? Am Soc Clin Oncol
Educ Book. 2019;39:364-372.
32. Ma DJ, Price KA, Moore EJ, et al. Phase II evaluation of aggressive dose de-escalation for adjuvant chemoradiotherapy in human papillomavirus-associated
oropharynx squamous cell carcinoma. J Clin Oncol. 2019;37:1909-1918.
33. Kann BH, Hicks DF, Payabvash S, et al. Multi-institutional validation of deep learning for pretreatment identification of extranodal extension in head and neck
squamous cell carcinoma. J Clin Oncol. Epub 2019 Dec 9.
34. Haughey BH, Hinni ML, Salassa JR, et al. Transoral laser microsurgery as primary treatment for advanced-stage oropharyngeal cancer: a United States
multicenter study. Head Neck. 2011;33:1683-1694.
35. O’Sullivan B, Huang SH, Su J, et al. Development and validation of a staging system for HPV-related oropharyngeal cancer by the International Collaboration on
Oropharyngeal Cancer Network for Staging (ICON-S): a multicentre cohort study. Lancet Oncol. 2016;17:440-451.
36. Routman DM, Funk RK, Tangsriwong K, et al. Relapse rates with surgery alone in human papillomavirus-related intermediate- and high-risk-group oropharynx
squamous cell cancer: a multi-institutional review. Int J Radiat Oncol Biol Phys. 2017;99:938-946. Epub 2017 Jun 28.
37. Kharytaniuk N, Molony P, Boyle S, et al. Association of extracapsular spread with survival according to human papillomavirus status in oropharynx squamous cell
carcinoma and carcinoma of unknown primary site. JAMA Otolaryngol Head Neck Surg. 2016;142:683-690.
38. Martin RC II, Brennan MF, Jaques DP. Quality of complication reporting in the surgical literature. Ann Surg. 2002;235:803-813.
39. Simon C, Caballero C, Gregoire V, et al. Surgical quality assurance in head and neck cancer trials: an EORTC Head and Neck Cancer Group position paper based
on the EORTC 1420 ‘Best of’ and 24954 ‘larynx preservation’ study. Eur J Cancer. 2018;103:69-77.
40. Eskander A, Goldstein DP, Irish JC. Health services research and regionalization of care-from policy to practice: the Ontario experience in head and neck cancer.
Curr Oncol Rep. 2016;18:19.
41. Calais G, Alfonsi M, Bardet E, et al. Randomized trial of radiation therapy versus concomitant chemotherapy and radiation therapy for advanced-stage oropharynx
carcinoma. J Natl Cancer Inst. 1999;91:2081-2086.
42. Ma J, Liu Y, Huang XL, et al. Induction chemotherapy decreases the rate of distant metastasis in patients with head and neck squamous cell carcinoma but does
not improve survival or locoregional control: a meta-analysis. Oral Oncol. 2012;48:1076-1084.
43. Domenge C, Hill C, Lefebvre JL, et al; French Groupe d’Etude des Tumeurs de la Tête et du Cou (GETTEC). Randomized trial of neoadjuvant chemotherapy in
oropharyngeal carcinoma. French Groupe d’Etude des Tumeurs de la Tête et du Cou (GETTEC). Br J Cancer. 2000;83:1594-1598.
44. Marur S, Li S, Cmelak AJ, et al. E1308: Phase II trial of induction chemotherapy followed by reduced-dose radiation and weekly cetuximab in patients with HPV-
associated resectable squamous cell carcinoma of the oropharynx—ECOG-ACRIN Cancer Research Group. J Clin Oncol. 2017;35:490-497.
45. Chen AM, Felix C, Wang PC, et al. Reduced-dose radiotherapy for human papillomavirus-associated squamous-cell carcinoma of the oropharynx: a single-arm,
phase 2 study. Lancet Oncol. 2017;18:803-811.
46. Villaflor VM, Melotek JM, Karrison TG, et al. Response-adapted volume de-escalation (RAVD) in locally advanced head and neck cancer. Ann Oncol. 2016;
27:908-913.
47. Seiwert TY, Foster CC, Blair EA, et al. OPTIMA: a phase II dose and volume de-escalation trial for human papillomavirus-positive oropharyngeal cancer. Ann
Oncol. 2019;30:297-302.
48. Misiukiewicz K, Gupta V, Miles BA, et al. Standard of care vs reduced-dose chemoradiation after induction chemotherapy in HPV+ oropharyngeal carcinoma
patients: the Quarterback trial. Oral Oncol. 2019;95:170-177.
49. Weiss JM, Grilley-Olson JE, Deal AM, et al. Phase 2 trial of neoadjuvant chemotherapy and transoral endoscopic surgery with risk-adapted adjuvant therapy for
squamous cell carcinoma of the head and neck. Cancer. 2018;124:2986-2992.
50. Park YM, Keum KC, Kim HR, et al. A clinical trial of combination neoadjuvant chemotherapy and transoral robotic surgery in patients with T3 and T4 laryngo-
hypopharyngeal cancer. Ann Surg Oncol. 2018;25:864-871.
51. Sadeghi N, Li NW, Taheri MR, et al. Neoadjuvant chemotherapy and transoral surgery as a definitive treatment for oropharyngeal cancer: a feasible novel
approach. Head Neck. 2016;38:1837-1846.
52. Sadeghi N, Khalife S, Mascarella MA, et al. Pathologic response to neoadjuvant chemotherapy in HPV-associated oropharynx cancer. Head Neck. 2020;
42:417-425.
53. Ferrarotto R, Bell D, Rubin ML. Checkpoint inhibitors assessment in oropharynx cancer (CIAO): safety and interim results. J Clin Oncol. 2019;37 (suppl; abstr
6008).
54. Barrett W, Cassidy C. Long-term toxicity following radiation therapy for head and neck squamous cell carcinoma: a retrospective study analyzing the effect of
concurrent chemotherapy and advanced t-stage on late treatment-induced toxicities. J Clin Oncol. 2019;37 (suppl; abstr e17560).
55. Machtay M, Moughan J, Trotti A, et al. Factors associated with severe late toxicity after concurrent chemoradiation for locally advanced head and neck cancer: an
RTOG analysis. J Clin Oncol. 2008;26:3582-3589.
56. Bonner JA, Harari PM, Giralt J, et al. Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N Engl J Med. 2006;354:567-578.
57. Gillison ML, Trotti AM, Harris J, et al. Radiotherapy plus cetuximab or cisplatin in human papillomavirus-positive oropharyngeal cancer (NRG Oncology RTOG
1016): a randomised, multicentre, non-inferiority trial. Lancet. 2019;393:40-50.

58. Mehanna H, Robinson M, Hartley A, et al; DE-ESCALATE HPV Trial Group. Radiotherapy plus cisplatin or cetuximab in low-risk human papillomavirus-positive
oropharyngeal cancer (DE-ESCALATE HPV): an open-label randomised controlled phase 3 trial. Lancet. 2019;393:51-60.
59. Nguyen-Tan PF, Zhang Q, Ang KK, et al. Randomized phase III trial to test accelerated versus standard fractionation in combination with concurrent cisplatin for
head and neck carcinomas in the Radiation Therapy Oncology Group 0129 trial: long-term report of efficacy and toxicity. J Clin Oncol. 2014;32:3858-3866.
60. Bauml JM, Vinnakota R, Anna Park YH, et al. Cisplatin every 3 weeks versus weekly with definitive concurrent radiotherapy for squamous cell carcinoma of the
head and neck. J Natl Cancer Inst. 2019;111:490-497.
61. Perez CA, Wu X, Amsbaugh MJ, et al. High-dose versus weekly cisplatin definitive chemoradiotherapy for HPV-related oropharyngeal squamous cell carcinoma of
the head and neck. Oral Oncol. 2017;67:24-28.
62. Geiger JL, Lazim AF, Walsh FJ, et al. Adjuvant chemoradiation therapy with high-dose versus weekly cisplatin for resected, locally-advanced HPV/p16-positive
and negative head and neck squamous cell carcinoma. Oral Oncol. 2014;50:311-318.
63. Price KA, Chintakuntlawar AV, Foote RL, et al. Long-term survival of adjuvant high-dose (HDC) vs weekly cisplatin (WC) for human papilloma-virus (HPV) and
non-HPV head and neck squamous cell carcinoma (HNSCC). J Clin Oncol. 2019;37 (suppl; abstr 6071).
64. Lee NCJ, Kelly JR, Park HS, et al. Patterns of failure in high-metastatic node number human papillomavirus-positive oropharyngeal carcinoma. Oral Oncol. 2018;
85:35-39.
65. Riaz N, Setton J, Tam M, et al. Patients with low lying lymph nodes are at high risk for distant metastasis in oropharyngeal cancer. Oral Oncol. 2014;50:863-868.
66. Huang SH, Perez-Ordonez B, Weinreb I, et al. Natural course of distant metastases following radiotherapy or chemoradiotherapy in HPV-related oropharyngeal
cancer. Oral Oncol. 2013;49:79-85.
67. Routman DM, Chera BS, Jethwa KR, et al. Detectable HPV ctDNA in post-operative oropharyngeal squamous cell carcinoma patients is associated with
progression. Int J Radiat Oncol Biol Phys. 2019;105:682-683.
68. Chera BS, Kumar S, Shen C, et al. Plasma circulating tumor HPV DNA for the surveillance of cancer recurrence in HPV-associated oropharyngeal cancer. J Clin
Oncol. Epub 2020 Feb 4.
69. Morris LGT, Chandramohan R, West L, et al. The molecular landscape of recurrent and metastatic head and neck cancers: insights from a precision oncology
sequencing platform. JAMA Oncol. 2017;3:244-255.
70. Gross AM, Orosco RK, Shen JP, et al. Multi-tiered genomic analysis of head and neck cancer ties TP53 mutation to 3p loss. Nat Genet. 2014;46:939-943.
71. Burtness B, Harrington KJ, Greil R, et al; KEYNOTE-048 Investigators. Pembrolizumab alone or with chemotherapy versus cetuximab with chemotherapy for
recurrent or metastatic squamous cell carcinoma of the head and neck (KEYNOTE-048): a randomised, open-label, phase 3 study. Lancet. 2019;
394:1915-1928.
72. Chow LQM, Haddad R, Gupta S, et al. Antitumor activity of pembrolizumab in biomarker-unselected patients with recurrent and/or metastatic head and neck
squamous cell carcinoma: results from the phase Ib KEYNOTE-012 expansion cohort. J Clin Oncol. 2016;34:3838-3845.
73. Seiwert TY, Burtness B, Mehra R, et al. Safety and clinical activity of pembrolizumab for treatment of recurrent or metastatic squamous cell carcinoma of the head
and neck (KEYNOTE-012): an open-label, multicentre, phase 1b trial. Lancet Oncol. 2016;17:956-965.
74. Ferris RL, Blumenschein G Jr., Fayette J, et al. Nivolumab for recurrent squamous-cell carcinoma of the head and neck. N Engl J Med. 2016;375:1856-1867.
75. Bauman JE, Harris J, Uppaluri R, et al. NRG-HN003: phase I and expansion cohort study of adjuvant cisplatin, intensity-modulated radiation therapy (IMRT), and
MK-3475 (pembrolizumab) in high-risk head and neck squamous cell carcinoma (HNSCC). J Clin Oncol. 2019;37 (suppl; abstr 6023).
76. Pearlstein KA, Wang K, Amdur RJ, et al. Quality of life for patients with favorable-risk HPV-associated oropharyngeal cancer after de-intensified chemo-
radiotherapy. Int J Radiat Oncol Biol Phys. 2019;103:646-653.
77. Beaty BT, Moon DH, Shen CJ, et al. PIK3CA mutation in HPV-associated OPSCC patients receiving deintensified chemoradiation. J Natl Cancer Inst. Epub 2019
Nov 20.
78. Chera BS, Kumar S, Beaty BT, et al. Rapid clearance profile of plasma circulating tumor HPV type 16 DNA during chemoradiotherapy correlates with disease
control in HPV-associated oropharyngeal cancer. Clin Cancer Res. 2019;25:4682-4690.
79. So YK, Lee G, Oh D, et al. Prognostic role of neutrophil-to-lymphocyte ratio in patients with human papillomavirus-positive oropharyngeal cancer. Otolaryngol
Head Neck Surg. 2018;159:303-309.
80. Gorphe P, Chekkoury Idrissi Y, Tao Y, et al. Anemia and neutrophil-to-lymphocyte ratio are prognostic in p16-positive oropharyngeal carcinoma treated with
concurrent chemoradiation. Papillomavirus Res. 2018;5:32-37.
81. Näsman A, Bersani C, Lindquist D, et al. Human papillomavirus and potentially relevant biomarkers in tonsillar and base of tongue squamous cell carcinoma.
Anticancer Res. 2017;37:5319-5328.
82. Albergotti WG, Davis KS, Abberbock S, et al. Association of pretreatment body mass index and survival in human papillomavirus positive oropharyngeal squamous
cell carcinoma. Oral Oncol. 2016;60:55-60.
83. Arthur AE, Duffy SA, Sanchez GI, et al. Higher micronutrient intake is associated with human papillomavirus-positive head and neck cancer: a case-only analysis.
Nutr Cancer. 2011;63:734-742.
84. Schimansky S, Lang S, Beynon R, et al. Association between comorbidity and survival in head and neck cancer: results from Head and Neck 5000. Head Neck.
2019;41:1053-1062.

Recent Advances in the Development of

Biomarkers and Chemoradiotherapeutic
Approaches for Nasopharyngeal Carcinoma
Brigette B.Y. Ma, MD1; Yu-Pei Chen, MD2; Edwin P. Hui, MD1; Xu Liu, MD2; Allen K.C. Chan, MD3; Anthony T.C. Chan, MD1; and
Jun Ma, MD2
overview
Epstein-Barr virus (EBV) associated nasopharyngeal carcinoma (NPC) is endemic in Southern China, and the
prognosis of this cancer has improved in part due to advances in radiotherapy (RT) techniques, broadened
therapeutic options, and more precise prognostic stratification of patients. RT is the primary curative
treatment of NPC, and the incorporation of chemotherapy (induction, concurrent, adjuvant) to RT has
contributed to improved survival in patients with locoregionally advanced NPC. Concurrent chemo-
radiotherapy (CCRT) in combination with adjuvant or induction chemotherapy is now the standard treatment
of locoregionally advanced NPC, but the ideal CCRT therapeutic strategy for NPC remains controversial.
Plasma EBV DNA is the archetypal tumor-derived DNA in NPC, and three generations of studies have gradually
expanded its clinical applications. Recently, the advent of whole exome/genome sequencing of NPC and the
promising clinical activity of immune checkpoint inhibitors have also spurred interest in the development of
newer biomarkers. This review will focus on two clinical advances in NPC research that have made substantial
impact on the contemporary management of NPC: (1) The integration of plasma EBV DNA in an expanding
spectrum of clinical indications, and the development of promising immune-related biomarkers; (2) the
current development of CCRT with special emphasis on the use of induction and adjuvant chemotherapy, as
well as the potential applications of metronomic chemotherapy and immune checkpoint inhibitors in the
treatment of locoregionally advanced NPC.
INTRODUCTION EBV DNA in an expanding spectrum of clinical in-

According to GLOBOCAN 2018, NPC is the 23rd most dications, and the development of immune-related
prevalent cancer with 129,000 new cases diagnosed biomarkers; (2) recent developments in the systemic
worldwide.1 China has the highest incidence of NPC in intensification of CCRT with induction chemotherapy
and/or adjuvant chemotherapy, and investigations into
terms of crude rates, but the age-adjusted rates per
the use of metronomic chemotherapy and immune
100,000 are the highest in some Southeast Asian
checkpoint inhibitors in the treatment of locoregionally
countries.1,2 The biology and epidemiology of endemic
advanced NPC.
NPC have been well-described in the literature, and the
prognosis of this cancer has improved steadily in part INTEGRATING BIOMARKERS IN THE MANAGEMENT
due to advances in early detection, RT techniques, and OF NPC
systemic therapy in curative and palliative settings.3,4 Three Generations of Studies on the Clinical
Author affiliations
As most endemic NPC are EBV-associated un- Application of Plasma EBV DNA in NPC
and support differentiated carcinomas, the historical development First-generation studies Plasma EBV DNA is an ar-
information (if of biomarkers has focused mainly on circulating or chetypal tumor-derived circulating DNA that can be
applicable) appear tumor-derived components of the EBV-related proteins
at the end of this
accurately quantified in patients with NPC using the
and genome. Recent reports on the whole exome real-time polymerase chain reaction (RT-PCR) tech-
article.
Accepted on
sequencing of NPC, and the promising clinical activity nique.11 A large body of literature published over 2
February 26, 2020 of immune checkpoint inhibitors have spurred in- decades has systematically investigated the unique
and published at terests in the development of genomic and immuno- properties and clinical utilities of this biomarker in
ascopubs.org on logic biomarkers of treatment outcome.4-10 This review NPC; see Fig. 1 for a historical perspective on these
March 19, 2020:
will focus on two clinical advances in NPC research studies. The “first generation” of studies published in
10.1200/EDBK_ that have substantial impact on the contemporary the late 1990s to early 2000 laid the foundation of
280747 management of NPC: (1) The integration of plasma subsequent research by evaluating the performance of

Recent Advances in Biomarker and Multimodal Treatment of Nasopharyngeal Carcinoma
confirmed the prognostic significance of pre-RT plasma

EBV DNA regardless of the cutoff level used, which ranged
PRACTICE APPLICATIONS
from 0 to 4,000 copies/L.19 Whereas in the post-RT setting,
• For stage III–IVA NPC, CCRT plus induction/ the presence of any detectable level of plasma EBV DNA
adjuvant chemotherapy is recommended; CCRT
was strongly associated with increased hazards of death.19
alone is also an option. Induction chemotherapy
The heterogeneity of these studies is due in part to the
offers the advantages of better tolerability and
early eradication of micrometastases than adju- varying detection limits of different quantitative EBV DNA
vant chemotherapy, and the addition of induction assays. This realization is highlighted by an important col-
chemotherapy is preferred in patients with high laboration between the United States, Hong Kong, Mainland
risk of distant failure. China, and Taiwan in an effort to identify any differences in
• Recommended induction chemotherapy in- the performance of RT-PCR assays used and to promote
cludes regimens such as TPF, GP, docetaxel plus standardization of assays across different centers.20 Sub-
cisplatin, and cisplatin plus fluorouracil. The sequently, a U.S. National Cancer Institute workgroup re-
appropriate time for addition of immunotherapy leased a recommendation for harmonizing and validating
to CCRT (induction and/or concurrent and/or EBV DNA assays.21 These efforts paved the way for multi-
adjuvant) is currently under investigation.
center clinical trials in evaluating the utility of plasma EBV
• Metronomic use of adjuvant chemotherapy may
DNA in the prognostic stratification of patients with
be a promising strategy to further improve effi-
cacy with tolerable toxicities. locoregionally advanced NPC for the purpose of in-
• Plasma EBV DNA is a tumor-derived DNA in dividualizing adjuvant chemotherapy. The first randomized
EBV-associated NPC that can be quantified trial that addressed this question was the NPC0502 study
accurately, and three “generations” of studies reported by Chan et al22 from the Hong Kong Nasopha-
conducted over the past 2 decades have sig- ryngeal Cancer Study Group, where 104 patients (out of 789
nificantly broadened its clinical applications. screened) with detectable levels of plasma EBV DNA after
These include early detection, staging at di- RT were randomly assigned to either adjuvant platinum-
agnosis, posttreatment surveillance, and identi- gemcitabine or follow-up alone without adjuvant chemo-
fication of patients at higher risk of relapse who therapy. There was no statistical difference in the primary
may benefit from systemic intensification with RT.
endpoint of relapse-free survival between the two treatment
• Promising immune-related biomarkers may help
arms. To understand how plasma EBV DNA could be better
to identify patients who are more likely to benefit
from immune checkpoint inhibitors at different used in this setting, Hui et al23 analyzed data from the
clinical settings. NPC0502 study with recursive partitioning analysis (RPA) in
a collaboration between CUHK and the Sun Yat Sen Uni-
versity (SYSU). They identified three RPA risk groups—low,
intermediate, and high groups (Table 1)—by combining
RT-PCR assays and their kinetics following treatment of different ranges of plasma EBV DNA (rather than a single
NPC. Using RT-PCR against the BamHI-W repeat region of cutoff value) with overall TNM stages. RPA groups were
the EBV genome, Lo and colleagues11-13 from the Chinese found to better discriminate hazards of death when com-
University of Hong Kong (CUHK) found that plasma EBV pared with either post-RT EBV DNA or TNM stage alone.
DNA can be detected in over 95% of patients with NPC. It These data suggest that RPA should be evaluated pro-
represents a bona fide “liquid biopsy” of NPC tumors be- spectively as a marker for identifying patients for adjuvant
cause of its tumor-derived origin, correlation with tumor chemotherapy. Another important multicenter study that
burden, and predictable circulating half-life following addresses the role of plasma EBV DNA in the adjuvant
RT.11-14 During this era, the prognostic significance of post- setting is the U.S. NRG-led HN001 study (ClinicalTrials.gov
RT level of plasma EBV DNA and the possible comple- identifier: NCT02135042), which has a different design
mentary role of this marker to Tumor, Node, Metastasis from the NPC0502 study. Patients are stratified according
(TNM) staging have also been demonstrated for the first to their post-RT plasma EBV DNA level whereby those
time in clinical studies of patients undergoing radical RT.15-18 with detectable levels are randomly assigned to receive two
different adjuvant chemotherapy regimens (platinum vs.
Second-generation studies The second generation of nonplatinum regimens). Patients with undetectable levels
studies focused on validating the earlier studies and iden- are randomly assigned to receive adjuvant chemotherapy
tifying the optimal cutoff levels of plasma EBV DNA and versus follow-up alone. This study is expected to complete
sampling time points (pre-, midpoint, and post-RT) in much accrual by the end of 2020.
larger cohorts. Collectively, these studies accrued over
27,200 patients from mostly endemic regions from the Third-generation studies The third generation of studies
mid-2000 to late 2010s.19 A meta-analysis of 40 studies published since the late 2010s revolves thematically around

Ma et al
staging, early detection, and the risk-adapted approach to unique EBV DNA methylation profile.28,29 This improved
individualizing induction and adjuvant chemotherapy in assay could improve the positive predictive value from 11%
patients with locoregionally advanced NPC (Fig. 1). Leung (as reported in the screening study) to 35.1%.26,28,29
et al16 first reported in 2006 that pretreatment plasma EBV
One of the most practice-changing studies reported in the
DNA at a cutoff of 4,000 copies/mL could further differ-
last 12 months is the phase III trial led by Zhang et al,30
entiate the prognosis of patient subgroups with stage IIB–III
which showed that the addition of induction platinum-
NPC. This is clinically meaningful because some patients
gemcitabine to CCRT improved survival in patients with
could potentially be spared of concurrent chemotherapy
locoregionally advanced NPC. The positive result of this
during RT. The application of plasma EBV DNA in staging is
study has renewed interests in addressing what should be
challenging because of a lack of consensus over the optimal
the “best” time point for sampling plasma EBV DNA during
cutoff of values of plasma EBV DNA, and also the extent to
RT, for the purpose of stratifying patients with locoregionally
which the existing TNM system should be modified to allow
advanced NPC into prognostic groups. Rather than sam-
inclusion of this biomarker. Two notable studies using RPA
pling a single time point at pre- or post-RT, Lv et al31 in-
risk groups have been reported to date24,25; these studies
vestigated the pattern of plasma EBV DNA clearance by
differed from each other in terms of the study design (e.g.,
taking serial samples in 673 patients with locoregionally
use of exploratory and validating sets), patient population
advanced NPC, who were uniformly treated with induction
and treatment received (e.g., proportion of patients treated
chemotherapy and then CCRT. They identified eight dif-
with induction chemotherapy), plasma EBV DNA assays
ferent plasma EBV DNA clearance (“response”) patterns
and cutoff values, and the number of RPA risks groups used
during induction chemotherapy and CCRT, from which
(all excluding metastatic stage; Table 1). In the RPA risk
they formulated four prognostic groups (“clusters”) with
groups proposed by Lee et al24 and Guo et al,25 plasma EBV
differential disease-free survival rates—early responders,
DNA was combined with T-stage and N-stages separately
intermediate responders, late responders, and treatment-
rather than overall stage. Using 5-year overall survival (OS)
resistant groups.31 This novel approach could potentially
as a yardstick, the RPA IIA, IIB, and III groups in the study by
provide oncologists the chance of intervening earlier during
Guo et al25 are able to discriminate the OS of subpopulations
the course of RT, by individualizing patients for therapeu-
of patients with N0 to N2 NPC. A consensus needs to be
tic intensification using chemotherapy or immunotherapy.
reached regarding which staging system should be adopted
Prospective validation in larger cohorts is eagerly awaited.
in the future in terms of practicality, reproducibility, and
ability to help clinical decision-making. In summary, three generations of studies have been re-
ported over the last 2 decades, elucidating the clinical
In a seminal screening study led by CUHK investigators in
application of plasma EBV DNA at diverse settings ranging
Hong Kong, 34 people out of 20,174 ethnically Chinese
from screening in healthy populations at risk for NPC, clinical
men aged 40 to 62 years were diagnosed with NPC over
a 3-year period.26 Justification of this inclusion criteria was
based on the fact that men in this age group have the
highest incidence of NPC in Hong Kong. Men who initially
tested positive for plasma EBV DNA were retested ap-
proximately 4 weeks later, and those with persistently
positive EBV DNA results underwent nasal endoscopic
examination and MRI. Chan et al26 found that plasma EBV
DNA has a 97.1% sensitivity, 98.6% specificity, and 11%
positive predictive value in detecting NPC. Importantly, 70.
6% of the NPC cases were detected at stage I-II, and all 34
cases had a longer progression-free survival (PFS) than data
from a historical cohort.26 This study also highlighted the
complementary role of MRI in population screening, as MRI
could detect very early tumors that might be missed on
endoscopy.27 The same group of investigators have since
developed a newer EBV DNA assay algorithm based on both FIGURE 1. Three Generations of Studies on the Clinical Application of
Plasma EBV DNA in Endemic NPC
sequence-based counting and profiling of EBV DNA frag-
Abbreviations: Gen., generation; NPC, nasopharyngeal carcinoma;
ments and the EBV DNA methylation score.28,29 This is
pEBV DNA, plasma Epstein-Barr virus DNA; RPA, recursive partitioning
because false-positive plasma EBV DNA detection could analysis; RT, radiotherapy; RT-PCR, real-time polymerase chain re-
be due to EBV lytic infection, whereas “cancer-specific” action; TNM, tumor, node, metastasis; US NCI, US National Cancer
plasma EBV DNA fragments tend to be shorter and exhibit Institute.

TABLE 1. Incorporation of Plasma EBV DNA in TNM Staging Studies Using RPA Risk Groups
Staging Groups (Excluding Metastatic Groups)
Plasma EBV DNA Cutoff
Study Population—Stages Copies/mL Plasma EBV DNA (Copies per mL) 5-year OS Rates
Pretreatment EBV DNA
Lee et al24 (HKU) N = 518 (prospectively enrolled 500 • RPA-I (T1–T4 and N0–N2, and pEBV DNA , 500; 5-year
cohort) 4 RPA groups OS = 89%)*
• RPA-II (T1–T4 and N0–N2, and pEBV DNA 500; 5-year
OS = 83%)*
• RPA-III (T1–T2 and N3; 5-year OS = 83%)*
• RPA-IVA (T3–T4 and N3; 5-year OS = 60%)* (all p , .001*).
Guo et al 25
N = 979 (Exp. set) 2,000 • RI (T1N0): 5-year OS 100%
(Mainland China
N = 550 (Val. set) 5 RPA groups • RIIA (T2–T3N0 or T1–T3N1, and pEBV DNA 2,000): 5-year
sites)
OS = 94.2%*
• RIIB (T2–T3N0 or T1–T3N1, and pEBV DNA . 2,000;
T1–T3N2, and pEBV DNA 2,000): 5-yr OS = 86.5%*
• RIII (T1–T3 and N2, and pEBV DNA . 2,000; T4N0–N2): 5-
year OS = 75.5%*
• R-IVA (any T and N3): 5-year OS = 58.3%* (all p , .001*)
Post-RT EBV DNA
Hui et al23 (CUHK N = 789 (Exp. set from NPC0502 Range used: 0, 1–49, • Low risk: (II–III, pEBV DNA = 0; II and EBV DNA 1–49) 5-year
and SYSU) study22) 50–499, . 500 OS: 89.4%*
N = 1,269 (Val. set) • Intermediate risk: (IVA and pEBV DNA = 0; III–IVAB and pEBV
DNA 1–49) 5-year OS: 78.5%*
3 RPA groups • High risk: (III–IVA and pEBV DNA 50–499; Any stage and
pEBV DNA . 500) 5-year OS: 37.2%* (all p , .005*)
Abbreviations: EBV DNA, Epstein-Barr virus DNA; TNM, Tumor, Node, Metastasis staging; RPA, recursive partitioning analysis; OS, overall survival; HKU, The
University of Hong Kong; Exp. set, exploratory set; Val. set, validating set; pEBV DNA, plasma EBV DNA; RT, radiotherapy; CUHK, The Chinese University of
Hong Kong; SYSU, Sun Yat Sen University.
*The statistical difference between these survival rates are expressed in the corresponding p values. For Guo et al, the survival rates cited are derived from the
validation set.
staging, posttreatment surveillance, and individualization of metastatic NPC, Ma et al8 found that a nonstatistical trend of
systemic intensification before and during RT. With the higher response rates in patients with higher tumoral PD-L1
lowering cost of a plasma EBV DNA test, it is hoped that this expression. In a randomized phase II study, Lim et al35
biomarker could be more broadly used in NPC-prevalent found that response to spartalizumab (PD-1 inhibitor) is
regions that include many developing countries. correlated to interferon-gamma signature, TIM3, and LAG3
gene expression in a similar population of patients. Other
Immunologic and Genomic Biomarkers on the Horizon investigators have tried to construct a prognostic signature
for NPC based on the immune-histochemical expression of immune
checkpoint proteins in NPC tumors.36 The prognostic sig-
EBV-associated NPC is a typical example of an “immune-
nificance of combined positive score of PD-L1 in NPC
hot” tumor with a dense stromal inflammatory infiltrate and
should be further defined in prospective studies such as
high expression of immune checkpoint proteins such as PD-
the KEYNOTE-122 study (ClinicalTrials.gov identifier:
1 and PD-L1, lymphocyte-activation gene 3 (LAG3) protein,
NCT02611960), a multicenter randomized study of pem-
and T-cell immunoglobulin and mucin-domain containing-3
brolizumab versus chemotherapy in platinum-refractory
(TIM3)/galectin-9 pathway.8,9,32-36 Several retrospective
recurrent or metastatic NPC that has completed accrual.
studies have reported conflicting results in the prognostic
significance of PD-L1 expression in NPC, probably because Whole exome/genome sequencing of over 100 micro-
of the different assays and methods of interpretation.9,33,34 dissected NPC tumors by Lo et al11-13 has revealed some
In a phase II study of nivolumab in patients with recurrent or interesting features concerning the genomic and

Ma et al
immunologic landscape of NPC.6 In this study led by CUHK induction cisplatin and docetaxel showed that additional
investigators, low tumor mutation burden (TMB) and the induction chemotherapy significantly improved the 3-year
presence of major histocompatibility complex-class 1 gene OS (OS: 94% vs. 68%; HR, 0.24; 95% CI, 0.08–0.73; p =
alterations were associated with poorer OS.6 Although .012).47 However, subsequent randomized trials48,49 have
around 70% of the primary NPC tumors displayed a “de- shown inconsistent results, probably due to insufficient
fective DNA” gene expression, hypermutated phenotype sample size or different induction regimens (Table 3). Owing
can be found in only 1.5% of NPC, and DNA mismatch to these inconsistent results, the role of induction chemo-
repair gene mutations are extremely rare.6 The prognostic therapy is still controversial in the early stage of NPC.
and predictive utility of TMB, immune-related gene signa- Recently, two large-scale multicenter phase III trials from
tures, major histocompatibility complex-I gene alterations, Guangzhou have been reported.50-53 One trial evaluated
or HLA protein expression must be defined in larger pro- docetaxel, cisplatin, and 5FU (TPF) induction chemother-
spective cohorts of patients undergoing immune checkpoint apy plus CCRT versus CCRT alone in locoregionally ad-
inhibitors for recurrent or metastatic NPC. vanced NPC and showed that the additional induction
IS THERE AN IDEAL CHEMORADIATION TREATMENT chemotherapy significantly improved the 5-year FFS (77%
STRATEGY FOR NPC? vs. 66%, p = .019); OS (86% vs. 78%, p = .042); distant FFS
RT is the primary curative treatment of NPC. Combining (88% vs. 80%, p = .030); and locoregional FFS (91% vs.
chemotherapy with RT is a pivotal advancement in the 84%, p = .044).50,51 Similarly, another trial comparing
treatment of locoregionally advanced NPC, of which CCRT is cisplatin plus fluorouracil (as induction chemotherapy)
the backbone.37 Currently, the National Comprehensive followed by CCRT showed that the 5-year DFS rate was 73%
Cancer Network Clinical Practice Guidelines recommend in the induction chemotherapy plus CCRT group and 63%
both CCRT with adjuvant or induction chemotherapy as in the CCRT-alone group (p = .007); moreover, the 5-year
level 2A evidence, and CCRT alone as level 2B evidence for distant metastasis–free survival rate was also significantly
stage II–IVA NPC.38 The ideal CRT treatment strategy for higher in the induction chemotherapy plus CCRT group
NPC is still debatable. than in the CCRT-alone group (83% vs. 73%; p = .014;
Table 3).52,53 In other regions, a phase III trial from Tunisia
Induction and Adjuvant Chemotherapy: Which One and France54 evaluating induction TPF (docetaxel 75 mg/m2
to Choose? on day 1, cisplatin 75 mg/m2 on day 1, and 5FU 750 mg/m2
Typically, adjuvant chemotherapy consists of cisplatin on days 1–5) showed significantly improved PFS (HR,
(80–100 mg/m2) and fluorouracil (5FU; 800–1,000 mg/m2 0.44; 95% CI, 0.20–0.97), confirming the efficacy of this
daily for 4–5 days) given every 4 weeks for three cycles triplet regimen. Another trial from Taiwan reported improved
(Table 2).39-44 Chen et al44,45 showed that compared with DFS in patients receiving induction mitomycin C, epirubicin,
CCRT alone, an additional adjuvant cisplatin and 5FU regi- cisplatin, and 5FU/leucovorin (HR, 0.74; 95% CI, 0.57–
men failed to improve survival in locoregionally advanced 0.97; p = .03; Table 3).55 Considering these encouraging
NPC: the 5-year failure-free survival (FFS) rate was 75% in results, the National Comprehensive Cancer Network
the CCRT plus adjuvant chemotherapy group and 71% in the Clinical Practice Guidelines upgraded the solidity of the
CCRT-alone group (FFS: hazard ratio [HR], 0.88; 95% CI, evidence concerning induction chemotherapy followed by
0.64–1.22; p = .45).45 Late toxic effects of grade 3 or 4 CCRT for locoregionally advanced NPC from level 3 to
appeared in 27% patients in the CCRT plus adjuvant che- level 2A in 2018.38
motherapy group and in 21% patients in the CCRT only group Although induction chemotherapy followed by CCRT for
(p = .14).44 Another phase III trial selecting high-risk patients locoregionally advanced NPC has gained considerable at-
with detectable plasma EBV DNA after CCRT showed no tention, the optimal induction regimens and biomarkers
significant difference in the 5-year relapse-free survival rate predicting ideal candidates for induction chemotherapy
between arm 1 (adjuvant chemotherapy with cisplatin and remain investigational. Recently, Zhang et al56 established
gemcitabine) and arm 2 (observation; relapse-free survival: gemcitabine plus cisplatin (GP), rather than cisplatin plus
49% vs. 55%; HR, 1.09; 95% CI, 0.63–1.89; p = .75).22 fluorouracil, as the standard first-line treatment option for
Compared with adjuvant sequencing, induction chemo- patients with recurrent or metastatic NPC. In the locore-
therapy offers the advantages of better tolerability and early gionally advanced setting, He et al57 evaluated the efficacy
eradication of micrometastases.4 In addition, induction and toxicity of GP as induction chemotherapy in a phase II
chemotherapy could shrink the tumor volume to provide trial and found that the 3-year locoregional control,
better protection for vital organs such as the brainstem.46 metastasis-free rate, and OS rates were 94.9%, 86.2%, and
Emerging clinical trials have investigated the efficiency of 87.7%, respectively. To further investigate the role of GP as
induction chemotherapy in addition to CCRT (Table 3).47-55 induction chemotherapy in locoregionally advanced NPC,
A randomized phase II trial comparing CCRT with or without Zhang et al30 conducted a parallel-group multicenter phase

TABLE 2. Phase III Randomized Trials Evaluating Concurrent Chemoradiotherapy With Adjuvant Chemotherapy
Overall Survival Progression-Free Survival
Control Experimental vs. Experimental vs. HR

Sample Experimental Chemotherapy Chemotherapy Control (n-Year HR (95% CI); p Control (n-Year (95% CI);
Trials Size Regimen Regimen Results) Value Results) p Value
Al-Sarraf 147 CC: cisplatin 100 mg/m2 day 1; — 67% vs. 37% NR; p = .001 58% vs. 29% NR; p ,
et al39 Q3W for 3 cycles (5-year) (5-year) .001
AC: cisplatin 80 mg/m2 day 1;
fluorouracil 1,000 mg/m2 days
1–4; Q4W for 4 cycles
Wee 221 CC: cisplatin 25 mg/m2 days 1–4; — 67% vs. 49% 0.60 59% vs. 46% 0.67
et al40 Q3W for 3 cycles (5-year) (0.41– (5-year) (0.46–
0.87); 0.97);
AC: cisplatin 20 mg/m2 days 1–4;
p = .0077 p = .032
Lee 348 CC: cisplatin 100 mg/m2 day 1; — 68% vs. 64% 0.81 (0.58 62% vs. 53% 0.72
et al41 Q3W for 3 cycles (5-year) –1.13); (5-year) (0.53–
p = .22 0.98);
p = .035
Lee 93 (189) CC: cisplatin 100 mg/m2 day 1; — 78% vs. 66% 0.76 60% vs. 62% 0.98
et al42 * Q3W for 3 cycles (5-year) (0.38– (5-year) (0.53–
1.54); 1.81);
p = .45 p = .93
Chen 316 CC: cisplatin 40 mg/m2 day 1; — 72% vs. 62% 0.69 68% vs. 57% 0.65
et al44 Q3W for 7 cycles (5-year) (0.48– (5-year) (0.46–
0.99); 0.92);
p = .043 p = .015
fluorouracil 800 mg/m2 days
Chen 508 CC: cisplatin 40 mg/m2 day 1; CC: cisplatin 40 83% vs. 80% 0.83 75% vs. 71% 0.88
et al44 Q3W for 7 cycles mg/m2 day 1; (5-year) (0.57–1.22); (5-year) (0.64–
Q3W 7 p = .35 1.22);
p = .45
fluorouracil 800 mg/m2 days
Abbreviations: CC, concurrent chemotherapy; AC, adjuvant chemotherapy; HR, hazard ratio; NR, not reported; Q3/4 W, every 3/4 weeks.
*Ninety-three patients in the conventional fractionation radiotherapy arms; 189 patients in the whole trial.
III trial and demonstrated an improved 3-year OS of GP as with locoregionally advanced NPC. Metronomic chemotherapy
induction chemotherapy plus CCRT versus CCRT alone refers to the maintenance therapy with low doses of cyto-
(OS: 95% vs. 90%; HR, 0.43; 95% CI, 0.24–0.77; Table 3). toxic drugs repeatedly delivered over shorter intervals
The incidence of grade 3 or 4 late toxic effects was 9.2% in without interruption, with the advantages of good compli-
the induction chemotherapy group and 11.4% in the ance, low toxicities, and convenience.58 The potential
standard-therapy group, indicating the good tolerability of mechanisms of metronomic chemotherapy include anti-
GP as induction chemotherapy.30 Overall, these results angiogenic effects, induction of tumor dormancy, and ac-
suggest GP as a potentially efficient regimen with better tivation of immunity. 59,60 The metronomic use of oral
chemotherapy agents such as capecitabine or uracil plus
compliance in NPC.
tegafur as adjuvant chemotherapy may be an alternative
Although adjuvant chemotherapy with traditional regimens after CCRT. One retrospective study evaluating the impact of
may not have additional survival benefits, poor compliance accumulated oral tegafur-uracil (UFUR) as metronomic
with the current adjuvant chemotherapy regimen after chemotherapy after CCRT showed that the 5-year OS rates
CCRT may have masked its efficacy. Identifying tolerable were 92% in the UFUR group and 58% in the non-UFUR
regimens for adjuvant chemotherapy may benefit patients group (p = .004).60 Large-scale randomized trials are

Ma et al
TABLE 3. Randomized Trials Evaluating Induction Chemotherapy Plus Concurrent Chemoradiotherapy Versus Concurrent Chemoradiotherapy Alone
Overall Survival Progression-Free Survival
Control Experimental Experimental HR (95%

Sample Experimental Chemotherapy Chemotherapy vs. Control HR (95% vs. Control CI); p
Trials Size Phase Regimen Regimen (n-Year Results) CI); p Value (n-Year Results) Value
Hui et al47 68 II IC: docetaxel 75 mg/m2 day 1; CC: cisplatin 40 94% vs. 68% 0$24 88% vs. 60% 0.49
cisplatin 75 mg/m2 day 1; Q3W mg/m2 day 1; (3-year) (0.08– (3-year) (0.20–
for 2 cycles Q1W for 8 cycles 0.73); 1.19);
p = .012 p = .12
CC: cisplatin 40 mg/m2 day 1;
Q1W 8
Fountzilas 141 II IC: cisplatin 75 mg/m2 day 1; CC: cisplatin 40 70% vs. 59% 0.95 65% vs. 64% 1.40
et al48 epirubicin 75 mg/m2 day 1; mg/m2 day 1; (3-year) (0.48– (3-year) (0.71–
paclitaxel 175 mg/m2 day 1; Q1W for 8 cycles 1.89); 2.77);
Q3W for 3 cycles p = .89 p = .38
Q1W for 8 cycles
Tan et al49 172 III IC: gemcitabine 1,000 mg/m2 day CC: cisplatin 40 94% vs. 92% 1.05 75% vs. 67% 0.77
1, day 8; carboplatin AUC = 2. mg/m2 day 1; (3-year) (0–2.19); (3-year) (0.44–
5 day 1, day 8; paclitaxel 70 mg/ Q1W for 8 cycles p = .49 1.35);
m2 day 1, day 8; Q3W for 3 p = .36
cycles
Q1W for 8 cycles
Sun 480 III IC: docetaxel 60 mg/m2 day 1; CC: cisplatin 86% vs. 78% 0.65 77% vs. 66% 0.65
et al50,51 cisplatin 60 mg/m2 day 1; 100 mg/m2 day 1; (5-year) (0.43– (5-year) (0.43–
fluorouracil 600 mg/m2 days Q3W for 3 cycles 0.98); 0.98);
1–5; Q3W for 3 cycles p = .042 p = .019
Q3W for 3 cycles
Cao 476 III IC: cisplatin 80 mg/m2 day 1; CC: cisplatin 80 81% vs. 77% 0.69 73% vs. 63% 0.66
et al52,53 fluorouracil 800 mg/m2 days mg/m2 day 1; (5-year) (0.49– (5-year) (0.48–
1–5; Q3W for 2 cycles Q3W for 3 cycles 0.98); 0.89);
p = .04 p = .007
Q3W for 3 cycles
Frikha 83 III IC: docetaxel 75 mg/m2 day 1; CC: cisplatin 40 86% vs. 69% 0.40 74% vs. 57% 0.44
et al54 cisplatin 75 mg/m2 day 1; mg/m2 day 1; (3-year) (0.15– (3-year) (0.20–
fluorouracil 750 mg/m2 days Q1W for 7 cycles 1.04); 0.97);
1–5; Q3W for 3 cycles p = .059 p = .042
Q1W for 7 cycles
Hong et al55 479 III IC: mitomycin 8 mg/m2 day 1; CC: cisplatin 30 72% vs. 68% 0.92 (0.67– 61% vs. 50% 0.74
epirubicin 60 mg/m2 day 1; mg/m2 day 1; (5-year) 1.27); p = (5-year) (0.57–
cisplatin 60 mg/m2 day 1; Q1W for 1 cycle .62 0.97);
fluorouracil 450 mg/m2 day 8; p = .026
leucovorin 30 mg/m2 day 8
Q1W for 1 cycle
Zhang 480 III IC: gemcitabine 1,000 mg/m2 day CC: cisplatin 95% vs. 90% 0.43 (0. 85% vs. 77% 0.51
et al30 1, day 8; cisplatin 80 mg/m2 day 100 mg/m2 day 1; (3-year) 24–0.77); (3-year) (0.34–
1; Q3W for 3 cycles Q3W for 3 cycles p = NR 0.77);
p = .002
Q3W for 3 cycles
Abbreviations: CC, concurrent chemotherapy; IC, induction chemotherapy; AUC, area under the concentration–time curve; HR, hazard ratio; NR, not
reported; Q1/3/4W, every 1/3/4 weeks.

urgently required to confirm the efficacy and safety of EBV DNA in the early detection and stratification of pa-
metronomic chemotherapy in locoregionally advanced tients for systemic intensification with CCRT. Induction
NPC. An ongoing phase III clinical trial (ClinicalTrials.gov chemotherapy plays an increasingly important role in the
identifier: NCT02958111) is exploring the effects of single- management of locoregionally advanced NPC in the
agent capecitabine as metronomic adjuvant chemotherapy intensity-modulated radiation therapy era, which helps to
(1,300 mg/m2 daily for 1 year) in patients with stage III–IVA further improve distant control and thereby survival.
NPC (except T3-4N0, T3N1). Much work remains to be done to refine these guidelines
Novel Applications of Immune Checkpoint Inhibitors in and make them more precise through clinical trials in the
the Treatment of Locoregionally Advanced NPC next decade. For locoregionally advanced NPC, induction
chemotherapy plus CCRT has better performance than
Tumor immunotherapy—especially immune checkpoint in- CCRT alone and is also better than CCRT plus adjuvant
hibitors targeting PD-1 or its ligand PD-L1 may represent chemotherapy in view of lower toxicity and higher efficacy.
a promising therapeutic approach in NPC. First, NPC tumors Induction chemotherapy followed by CCRT may represent
are characterized by high expression of PD-L1 (up to 90% a promising treatment strategy for NPC in the intensity-
of tumor cells).9,34,61-63 Second, NPC arguably shows the modulated radiation therapy era. The GP regimen has
highest lymphocyte infiltration among all solid tumors. About demonstrated its superiority in metastatic NPC. Novel
50% of NPC samples had percentages of stromal tumor- therapies, such as induction chemotherapy plus CCRT
infiltrating cells (TILs) greater than 70% or intratumoral TILs plus metronomic chemotherapy needs further investiga-
greater than 10%.34,64-66 Therefore, NPC was formerly called tion. Immune checkpoint inhibitor therapy is expected
lymphoepithelioma-like carcinoma by pathologists. These to revolutionize the management of locoregionally ad-
two features make NPC a very attractive target for immune vanced NPC in the future but can only be done with
checkpoint inhibitor therapies. well-designed randomized trials with adequate sample
Recently, at least five single-arm trials, with a total enroll- sizes.
ment of 377 patients, have shown that anti–PD-1 mono-
Despite advances in the treatment of NPC, most patients
clonal antibodies have promising activities as a single agent
still present with locoregionally advanced NPC in NPC-
in recurrent or metastatic NPC.4,8,67,68 These trials typically
prevalent regions, and therefore population screening
included patients with heavily pretreated, platinum-resistant
could have a strong impact on lowering the overall mortality
recurrent or metastatic NPC. The response rate ranged from
of this cancer. A multidisciplinary screening program
20.5% to 34.0%. Moreover, researchers from SYSU re-
utilizing plasma EBV DNA has been shown prospectively to
ported the first attempt to use PD-1 antibody in combination
alter the stage distribution of NPC at diagnosis in a pop-
with GP in 23 patients with treatment-naı̈ve recurrent or
ulation at risk for developing NPC such as in Hong Kong.
metastatic NPC.56 The results showed that all patients had
The feasibility and cost-effectiveness of using plasma EBV
some shrinkage in tumor with an objective response rate of
DNA in screening should be evaluated as a priority area in
91%. The median duration of response and median PFS
public health policies, especially when the highest age-
was not reached, with the 1-year PFS rate was 61.4%. This
adjusted incidence rates of NPC are found in developing
number compared favorably with the observed response
countries. International consensus is needed in how to
rate (64%) and PFS (median: 7.0 months; 1-year rate: 20%)
best apply plasma EBV DNA in NPC staging, and several
in patients treated with GP alone in a large phase III trial.56
large prospective studies are now investigating the utility of
With the promising antitumor activity and predictable safety this biomarker in identifying patients with locoregionally
profile of anti–PD-1 antibodies, several randomized phase advanced NPC who might benefit from induction or ad-
III trials of anti–PD-1 antibody for recurrent or metastatic juvant chemotherapy. With the growing body of clinical
NPC are ongoing. In the curative setting, two large randomized trials on combining immune checkpoint inhibitors with
phase III trials (ClinicalTrials.gov identifiers: NCT03427827 conventional treatment in NPC, data on key predictive
and NCT03700476) are presently investigating the combi- biomarkers such as PD-L1 (combined positive score),
nation of PD-1 antibody and curative CRT in locoregionally TMB, and immune-gene signatures will be presented in
advanced NPC, with results expected shortly. the near future.
CONCLUSION
This review summarizes two important clinical milestones in ACKNOWLEDGMENT
the management of endemic NPC: the integration of sys- B. Ma, Y.-P. Chen, E.P. Hui, and X. Liu are joint first authors.
temic chemotherapy in the treatment of locoregionally A.T.C. Chan and J. Ma are joint senior authors.
advanced NPC, and the expanding applications of plasma

Ma et al

1
State Key Laboratory of Translational Oncology, Sir Y.K. Pao Centre for Brigette B.Y. Ma, MD, Department of Clinical Oncology, Prince of Wales
Cancer, Department of Clinical Oncology, Hong Kong Cancer Institute and Hospital, Shatin New Territories, Hong Kong SAR, China; email: brigette@
Li Ka Shing Institute of Health Sciences, The Chinese University of Hong clo.cuhk.edu.hk.
Kong, Hong Kong SAR, People’s Republic of China
2
Department of Radiation Oncology, Sun Yat-sen University Cancer
Center, State Key Laboratory of Oncology in South China, Collaborative AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Innovation Center for Cancer Medicine, Guangdong Key Laboratory of AND DATA AVAILABILITY STATEMENT
Nasopharyngeal Carcinoma Diagnosis and Therapy, Guangzhou, People’s Disclosures provided by the authors and data availability statement (if
Republic of China applicable) are available with this article at DOI https://doi.org/10.1200/
3
Department of Chemical Pathology, The Chinese University of Hong EDBK_279043.
Kong. Hong Kong SAR, People’s Republic of China
REFERENCES
1. Bray F, Ferlay J, Soerjomataram I, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185
countries. CA Cancer J Clin. 2018;68:394-424.
2. International Agency for Research on Cancer (IARC). Global Cancer Observatory, World Health Organization, 2020. https://gco.iarc.fr/. Accessed March 12,
2020.
3. Tsang CM, Lui VWY, Bruce JP, et al. Translational genomics of nasopharyngeal cancer. Semin. Cancer Biol. 2019. Epub 2019 Sep 12.
4. Chen YP, Chan ATC, Le QT, et al. Nasopharyngeal carcinoma. Lancet. 2019;394:64-80.
5. Zheng H, Dai W, Cheung AK, et al. Whole-exome sequencing identifies multiple loss-of-function mutations of NF-κB pathway regulators in nasopharyngeal
carcinoma. Proc Natl Acad Sci USA. 2016;113:11283-11288.
6. Li YY, Chung GT, Lui VW, et al. Exome and genome sequencing of nasopharynx cancer identifies NF-κB pathway activating mutations. Nat Commun. 2017;
8:14121.
7. Lin DC, Meng X, Hazawa M, et al. The genomic landscape of nasopharyngeal carcinoma. Nat Genet. 2014;46:866-871.
8. Ma BBY, Lim WT, Goh BC, et al. Antitumor activity of nivolumab in recurrent and metastatic nasopharyngeal carcinoma: An international, multicenter study of the
Mayo Clinic Phase 2 Consortium (NCI-9742). J Clin Oncol. 2018;36:1412-1418.
9. Fang W, Zhang J, Hong S, et al. EBV-driven LMP1 and IFN-γ up-regulate PD-L1 in nasopharyngeal carcinoma: Implications for oncotargeted therapy. Oncotarget.
2014;5:12189-12202.
10. Lu J, Chen XM, Huang HR, et al. Detailed analysis of inflammatory cell infiltration and the prognostic impact on nasopharyngeal carcinoma. Head Neck. 2018;
40:1245-1253.
11. Lam WKJ, Chan KCA, Lo YMD. Plasma Epstein-Barr virus DNA as an archetypal circulating tumour DNA marker. J Pathol. 2019;247:641-649.
12. Lo YM, Chan LY, Lo KW, et al. Quantitative analysis of cell-free Epstein-Barr virus DNA in plasma of patients with nasopharyngeal carcinoma. Cancer Res. 1999;
59:1188-1191.
13. Lo YM, Leung SF, Chan LY, et al. Kinetics of plasma Epstein-Barr virus DNA during radiation therapy for nasopharyngeal carcinoma. Cancer Res. 2000;
60:2351-2355.
14. Chan KC, Chan AT, Leung SF, et al. Investigation into the origin and tumoral mass correlation of plasma Epstein-Barr virus DNA in nasopharyngeal carcinoma.
Clin Chem. 2005;51:2192-2195.
15. Leung SF, Chan AT, Zee B, et al. Pretherapy quantitative measurement of circulating Epstein-Barr virus DNA is predictive of posttherapy distant failure in patients
with early-stage nasopharyngeal carcinoma of undifferentiated type. Cancer. 2003;98:288-291.
16. Leung SF, Zee B, Ma BB, et al. Plasma Epstein-Barr viral deoxyribonucleic acid quantitation complements tumor-node-metastasis staging prognostication in
nasopharyngeal carcinoma. J Clin Oncol. 2006;24:5414-5418.
17. Chan AT, Lo YM, Zee B, et al. Plasma Epstein-Barr virus DNA and residual disease after radiotherapy for undifferentiated nasopharyngeal carcinoma. J Natl
Cancer Inst. 2002;94:1614-1619.
18. Lin JC, Wang WY, Chen KY, et al. Quantification of plasma Epstein-Barr virus DNA in patients with advanced nasopharyngeal carcinoma. N Engl J Med. 2004;
350:2461-2470.
19. Xie X, Ren Y, Wang K, et al. Molecular prognostic value of circulating Epstein-Barr viral DNA in nasopharyngeal carcinoma: A meta-analysis of 27,235 cases in the
endemic area of Southeast Asia. Genet Test Mol Biomarkers. 2019;23:448-459.
20. Le QT, Zhang Q, Cao H, et al. An international collaboration to harmonize the quantitative plasma Epstein-Barr virus DNA assay for future biomarker-guided trials
in nasopharyngeal carcinoma. Clin Cancer Res. 2013;19:2208-2215.
21. Kim KY, Le QT, Yom SS, et al. Current state of PCR-based Epstein-Barr virus DNA testing for nasopharyngeal cancer. J Natl Cancer Inst. 2017;109.
22. Chan ATC, Hui EP, Ngan RKC, et al. Analysis of plasma Epstein-Barr virus DNA in nasopharyngeal cancer after chemoradiation to identify high-risk patients for
adjuvant chemotherapy: a randomized controlled trial. J Clin Oncol. Epub 2018 Jul 10.

23. Hui EP, Li WF, Ma BB, et al. Development and validation of a risk model integrating plasma Epstein-Barr virus DNA (EBV DNA) level and TNM stage for
stratification of nasopharyngeal cancer (NPC) to adjuvant therapy. Presented at: European Society of Medical Oncology Congress – Asia. Singapore, Singapore;
2019. Oral abstract 287O.
24. Lee VH-F, Kwong DL-W, Leung T-W, et al. The addition of pretreatment plasma Epstein-Barr virus DNA into the eighth edition of nasopharyngeal cancer TNM
stage classification. Int J Cancer. 2019;144:1713-1722.
25. Guo R, Tang LL, Mao YP, et al. Proposed modifications and incorporation of plasma Epstein-Barr virus DNA improve the TNM staging system for Epstein-Barr
virus-related nasopharyngeal carcinoma. Cancer. 2019;125:79-89.
26. Chan KCA, Woo JKS, King A, et al. Analysis of plasma Epstein-Barr virus DNA to screen for nasopharyngeal cancer. N Engl J Med. 2017;377:513-522.
27. King AD, Woo JKS, Ai QY, et al. Complementary roles of MRI and endoscopic examination in the early detection of nasopharyngeal carcinoma. Ann Oncol. 2019;
30:977-982.
28. Lam WKJ, Jiang P, Chan KCA, et al. Methylation analysis of plasma DNA informs etiologies of Epstein-Barr virus-associated diseases. Nat Commun. 2019;
10:3256.
29. Lam WKJ, Jiang P, Chan KCA, et al. Sequencing-based counting and size profiling of plasma Epstein-Barr virus DNA enhance population screening of na-
sopharyngeal carcinoma. Proc Natl Acad Sci USA. 2018;115:E5115-E5124.
30. Zhang Y, Chen L, Hu GQ, et al. Gemcitabine and cisplatin induction chemotherapy in nasopharyngeal carcinoma. N Engl J Med. 2019;381:1124-1135.
31. Lv J, Chen Y, Zhou G, et al. Liquid biopsy tracking during sequential chemo-radiotherapy identifies distinct prognostic phenotypes in nasopharyngeal carcinoma.
Nat Commun. 2019;10:3941.
32. Chen TC, Chen CH, Wang CP, et al. The immunologic advantage of recurrent nasopharyngeal carcinoma from the viewpoint of Galectin-9/Tim-3-related changes
in the tumour microenvironment. Sci Rep. 2017;7:10349.
33. Chan OS, Kowanetz M, Ng WT, et al. Characterization of PD-L1 expression and immune cell infiltration in nasopharyngeal cancer. Oral Oncol. 2017;67:52-60.
34. Zhu Q, Cai MY, Chen CL, et al. Tumor cells PD-L1 expression as a favorable prognosis factor in nasopharyngeal carcinoma patients with pre-existing intratumor-
infiltrating lymphocytes. OncoImmunology. 2017;6:e1312240.
35. Lim D, Wang D, Li S, et al. Phase II study of spartalizumab (PDR001) vs chemotherapy (CT) in patients with recurrent/metastatic nasopharyngeal cancer (NPC).
Cancer Res. 2019;79 (suppl; abstr CT150).
36. Wang YQ, Zhang Y, Jiang W, et al. Development and validation of an immune checkpoint-based signature to predict prognosis in nasopharyngeal carcinoma
using computational pathology analysis. J Immunother Cancer. 2019;7:298.
37. Blanchard P, Lee A, Marguet S, et al; MAC-NPC Collaborative Group. Chemotherapy and radiotherapy in nasopharyngeal carcinoma: an update of the MAC-NPC
meta-analysis. Lancet Oncol. 2015;16:645-655.
38. Colevas AD, Yom SS, Pfister DG, et al. NCCN Guidelines Insights: Head and Neck Cancers, Version 1.2018. J Natl Compr Canc Netw. 2018;16:479-490.
39. Al-Sarraf M, LeBlanc M, Giri PG, et al. Chemoradiotherapy versus radiotherapy in patients with advanced nasopharyngeal cancer: phase III randomized
Intergroup study 0099. J Clin Oncol. 1998;16:1310-1317.
40. Wee J, Tan EH, Tai BC, et al. Randomized trial of radiotherapy versus concurrent chemoradiotherapy followed by adjuvant chemotherapy in patients with
American Joint Committee on Cancer/International Union against cancer stage III and IV nasopharyngeal cancer of the endemic variety. J Clin Oncol. 2005;
23:6730-6738.
41. Lee AW, Tung SY, Chua DT, et al. Randomized trial of radiotherapy plus concurrent-adjuvant chemotherapy vs radiotherapy alone for regionally advanced
nasopharyngeal carcinoma. J Natl Cancer Inst. 2010;102:1188-1198.
42. Lee AW, Tung SY, Chan AT, et al. A randomized trial on addition of concurrent-adjuvant chemotherapy and/or accelerated fractionation for locally-advanced
nasopharyngeal carcinoma. Radiother Oncol. 2011;98:15-22.
43. Chen Y, Sun Y, Liang SB, et al. Progress report of a randomized trial comparing long-term survival and late toxicity of concurrent chemoradiotherapy with adjuvant
chemotherapy versus radiotherapy alone in patients with stage III to IVB nasopharyngeal carcinoma from endemic regions of China. Cancer. 2013;
119:2230-2238.
44. Chen L, Hu CS, Chen XZ, et al. Adjuvant chemotherapy in patients with locoregionally advanced nasopharyngeal carcinoma: Long-term results of a phase 3
multicentre randomised controlled trial. Eur J Cancer. 2017;75:150-158.
45. Chen L, Hu C-S, Chen X-Z, et al. Concurrent chemoradiotherapy plus adjuvant chemotherapy versus concurrent chemoradiotherapy alone in patients with
locoregionally advanced nasopharyngeal carcinoma: a phase 3 multicentre randomised controlled trial. Lancet Oncol. 2012;13:163-171.
46. Peng L, Liu JQ, Chen YP, et al. The next decade of clinical trials in locoregionally advanced nasopharyngeal carcinoma. Br J Radiol. 2019;92:20181031.
47. Hui EP, Ma BB, Leung SF, et al. Randomized phase II trial of concurrent cisplatin-radiotherapy with or without neoadjuvant docetaxel and cisplatin in advanced
nasopharyngeal carcinoma. J Clin Oncol. 2009;27:242-249.
48. Fountzilas G, Ciuleanu E, Bobos M, et al. Induction chemotherapy followed by concomitant radiotherapy and weekly cisplatin versus the same concomitant
chemoradiotherapy in patients with nasopharyngeal carcinoma: a randomized phase II study conducted by the Hellenic Cooperative Oncology Group (HeCOG)
with biomarker evaluation. Ann Oncol. 2012;23:427-435.
49. Tan T, Lim WT, Fong KW, et al. Concurrent chemo-radiation with or without induction gemcitabine, Carboplatin, and Paclitaxel: a randomized, phase 2/3 trial in
locally advanced nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys. 2015;91:952-960.

Ma et al
50. Sun Y, Li WF, Chen NY, et al. Induction chemotherapy plus concurrent chemoradiotherapy versus concurrent chemoradiotherapy alone in locoregionally
advanced nasopharyngeal carcinoma: a phase 3, multicentre, randomised controlled trial. Lancet Oncol. 2016;17:1509-1520.
51. Li WF, Chen NY, Zhang N, et al. Concurrent chemoradiotherapy with/without induction chemotherapy in locoregionally advanced nasopharyngeal carcinoma:
Long-term results of phase 3 randomized controlled trial. Int J Cancer. 2019;145:295-305.
52. Yang Q, Cao SM, Guo L, et al. Induction chemotherapy followed by concurrent chemoradiotherapy versus concurrent chemoradiotherapy alone in locoregionally
advanced nasopharyngeal carcinoma: long-term results of a phase III multicentre randomised controlled trial. Eur J Cancer. 2019;119:87-96.
53. Cao SM, Yang Q, Guo L, et al. Neoadjuvant chemotherapy followed by concurrent chemoradiotherapy versus concurrent chemoradiotherapy alone in
locoregionally advanced nasopharyngeal carcinoma: A phase III multicentre randomised controlled trial. Eur J Cancer. 2017;75:14-23.
54. Frikha M, Auperin A, Tao Y, et al; GORTEC. A randomized trial of induction docetaxel-cisplatin-5FU followed by concomitant cisplatin-RT versus concomitant
cisplatin-RT in nasopharyngeal carcinoma (GORTEC 2006-02). Ann Oncol. 2018;29:731-736.
55. Hong RL, Hsiao CF, Ting LL, et al. Final results of a randomized phase III trial of induction chemotherapy followed by concurrent chemoradiotherapy versus
concurrent chemoradiotherapy alone in patients with stage IVA and IVB nasopharyngeal carcinoma-Taiwan Cooperative Oncology Group (TCOG) 1303 Study.
Ann Oncol. 2018;29:1972-1979.
56. Zhang L, Huang Y, Hong S, et al. Gemcitabine plus cisplatin versus fluorouracil plus cisplatin in recurrent or metastatic nasopharyngeal carcinoma: a multicentre,
randomised, open-label, phase 3 trial. Lancet. 2016;388:1883-1892.
57. He X, Ou D, Ying H, et al. Experience with combination of cisplatin plus gemcitabine chemotherapy and intensity-modulated radiotherapy for locoregionally
advanced nasopharyngeal carcinoma. Eur Arch Otorhinolaryngol. 2012;269:1027-1033.
58. Bocci G, Kerbel RS. Pharmacokinetics of metronomic chemotherapy: a neglected but crucial aspect. Nat Rev Clin Oncol. 2016;13:659-673.
59. Chen YL, Chang MC, Cheng WF. Metronomic chemotherapy and immunotherapy in cancer treatment. Cancer Lett. 2017;400:282-292.
60. Chen JH, Huang WY, Ho CL, et al. Evaluation of oral tegafur-uracil as metronomic therapy following concurrent chemoradiotherapy in patients with non-distant
metastatic TNM stage IV nasopharyngeal carcinoma. Head Neck. 2019;41:3775-3782.
61. Chen BJ, Chapuy B, Ouyang J, et al. PD-L1 expression is characteristic of a subset of aggressive B-cell lymphomas and virus-associated malignancies. Clin
Cancer Res. 2013;19:3462-3473.
62. Lee VH, Lo AW, Leung CY, et al. Correlation of PD-L1 expression of tumor cells with survival outcomes after radical intensity-modulated radiation therapy for non-
metastatic nasopharyngeal carcinoma. PLoS One. 2016;11:e0157969.
63. Larbcharoensub N, Mahaprom K, Jiarpinitnun C, et al. Characterization of PD-L1 and PD-1 expression and CD8+ tumor-infiltrating lymphocyte in Epstein-Barr
virus-associated nasopharyngeal carcinoma. Am J Clin Oncol. 2018;41:1204-1210.
64. Hsu MC, Hsiao JR, Chang KC, et al. Increase of programmed death-1-expressing intratumoral CD8 T cells predicts a poor prognosis for nasopharyngeal
carcinoma. Mod Pathol. 2010;23:1393-1403.
65. Ono T, Azuma K, Kawahara A, et al. Prognostic stratification of patients with nasopharyngeal carcinoma based on tumor immune microenvironment. Head Neck.
2018;40:2007-2019.
66. Wang YQ, Chen YP, Zhang Y, et al. Prognostic significance of tumor-infiltrating lymphocytes in nondisseminated nasopharyngeal carcinoma: A large-scale cohort
study. Int J Cancer. 2018;142:2558-2566.
67. Hsu C, Lee SH, Ejadi S, et al. Safety and antitumor activity of pembrolizumab in patients with programmed death-ligand 1-positive nasopharyngeal carcinoma:
Results of the KEYNOTE-028 Study. J Clin Oncol. 2017;35:4050-4056.
68. Fang W, Yang Y, Ma Y, et al. Camrelizumab (SHR-1210) alone or in combination with gemcitabine plus cisplatin for nasopharyngeal carcinoma: results from two
single-arm, phase 1 trials. Lancet Oncol. 2018;19:1338-1350.

HEALTH SERVICES
RESEARCH AND QUALITY
IMPROVEMENT

How Do We Align Health Services Research and

Quality Improvement?
Devika Das, MD1,2,3; Lalan Wilfong, MD4; Katherine Enright, MD5,6; and Gabrielle Rocque, MD, MSPH1,2,7
overview
Quality improvement (QI) initiatives and health services research (HSR) are commonly used to target health
care quality. These disciplines are increasingly important because of the movement toward value-based health
care as alternative payment and care delivery models drive institutions and investigators to focus on reducing
unnecessary health care use and improving care coordination. QI efforts frequently target medical error and/or
efficiency of care through the Plan-Do-Study-Act methodology. Within the QI framework, strategies for data
display (e.g., Pareto charts, run charts, histograms, scatter plots) are leveraged to identify opportunities for
intervention and improvement. HSR is a multidisciplinary field of study that seeks to identify the most ef-
fective way to organize, deliver, and finance health care to maximize the quality and value of care at both the
individual and population levels. HSR uses a diverse set of quantitative and qualitative methodologies, such as
case-control studies, cohort studies, randomized control trials, and semistructured interview/focus group
evaluations. This manuscript provides examples of methodologic approaches for QI and HSR, discusses
potential challenges associated with concurrent quality efforts, and identifies strategies to successfully le-
verage the strengths of each discipline in care delivery.
INTRODUCTION The types of errors that contributed to decreased safety of

The Institute of Medicine defines “quality” as the health care delivery fell under the following domains:
degree to which health services for individuals and diagnostic, treatment, preventive, and others—including
populations increase the likelihood of desired health failures in communication.4 These reports helped create
outcomes and are consistent with current professional a national focus on health care safety by developing
knowledge.1 Both QI initiatives and HSR are commonly a nationwide public mandatory reporting system and
used to measure quality and can be leveraged con- improve oversight by raising performance standards. The
currently to target health care quality. These disciplines second report noted that, although medical science and
are increasingly important because of the movement technology were advancing at a rapid rate, our nation’s
toward value-based health care. Alternative payment health care delivery system was falling short in its ability to
and care delivery models are driving institutions and translate knowledge into safe and appropriate practice.1
investigators to focus on common themes, such as This report went on to define the six aims for health care
reducing unnecessary health care use and improving improvement, which were built around the fundamental
care coordination.2 As such, these topics are often needs of heath care to be safe, effective, patient centered,
simultaneous targets of both QI and HSR. This man- timely, efficient, and equitable.1
uscript reviews common methodologic approaches for Health care organizations strive to improve the quality
QI and HSR, discusses strengths of each methodology, of health care delivery by adapting and using the
outlines potential challenges associated with concur-
Author affiliations principles of high-reliability science, which is the study
rent quality efforts, and identifies strategies to suc-
and support of industries like aviation and nuclear power that
information (if
cessfully leverage strengths of each discipline in care
maintain high safety levels despite operating under
applicable) appear delivery.
hazardous conditions.5 The Institute for Healthcare
at the end of this
article.
QUALITY IMPROVEMENT IN HEALTH CARE Improvement defines “reliability” as the measurable
Accepted on March The field of QI was energized by two publications, capability of a process, procedure, or health service to
4, 2020 and released between 1999 and 2001, which revealed perform its intended function in the required time
published at that close to 98,000 people died in hospitals within under commonly occurring conditions.6 High-reliability
ascopubs.org on
the United States each year because of preventable organizations aim to achieve a “six sigma” perfor-
April 2, 2020:
medical errors. The initial report defined medical errors mance in which the processes result in the intended
10.1200/EDBK_ as the “failure of a planned action to be completed as result at a rate of 99.99966%, or fewer than 3.4 defects
281093 intended or the use of a wrong plan to achieve an aim.”3 or harms per million. In radiation oncology, Woodhouse

Aligning Health Services Research and Quality Imporovent
and colleagues7 used a process of continuous identification

of safety issues, coupled with improvement activities, to
reduce radiation errors and create a high-reliability orga-
• Quality improvement initiatives and health nization. They were able to substantially reduce the fre-
services research are often used to target health
quency of reportable medical events related to radiation.7
care quality, which is a common target for
Additional efforts for other disciplines in oncology and in-
value-based health care initiatives.
stitutions are needed to achieve this level of reliability.
• Although quality improvement and health ser-
vices research both focus on improving the care Another key target for QI efforts is cutting waste or adding
of patients, key differences exist in purpose, value. Lean is a set of operating philosophies, originally used
duration, project design, intervention scope, in the automobile industry, which, when applied to health
receptiveness of changing of the environment, care, help create a maximum value for patients by reducing
strategies to assess validity, and funding. waste. It involves a continuous process improvement system
• Quality improvement efforts frequently use consisting of resource management and waste reduction.8
Plan-Do-Study-Act methodology and other In a recent review, it was predicted that the estimated cost of
rapid-cycle evaluations to target medical error waste in the U.S. health care system ranged from $760
and/or efficiency of care. billion to $935 billion, accounting for approximately 25% of
• Health services research is a multidisciplinary total health care spending.9 Waste can occur with care
field of study that seeks to identify the most delivery and care coordination, overtreatment, administra-
effective way to organize, deliver, and finance tive processes, and pricing because of use, which is felt to
health care to maximize the quality and value of occur frequently in cancer care.10 QI can improve outcomes
care at both the individual and population by reducing waste, decreasing variation, and making pro-
levels. cesses more effective and efficient. Value-based care re-
• Coordination of efforts between quality im- form is based on the assumption that improving quality can
provement experts and health services re- lower costs.
searchers can leverage the strengths of each
discipline to improve care delivery. Quality Improvement Methodology
The most commonly used QI methodology is Plan-Do-
Study-Act (PDSA). It originated from industry—Walter
Shewhart and Edward Deming’s concept of iterative pro-
cesses, which is incorporated within the framework of
a PDSA cycle.
Planning encompasses the understanding of what may be
contributing to the current performance and identification of
the problem, selection of a team, conduction of a systematic
approach to identify drivers of current performance, and
identification of change strategies that may address gaps
and improve performance.
Understanding the current process often involves mapping
out all of the steps in it. An example of a simple process map
is displayed in Figure 1. It tracks the patient flow from the
point of clinic check-in to being seated in a room for the visit.
Ovals signify the start or end of the process map; rectangular
boxes are the sequential steps; diamonds signify decision
points in a process. Arrows show the direction of flow.
The “Plan” step also relies on the use of data to understand
variation and helps inform the development of solutions
most likely to address the real problem. Data can be pre-
sented in different ways based on their relation to time or
lack thereof. Types of data encountered in health care in-
clude continuous data like temperature or laboratory values,
count data like number of needlestick injuries or number of
FIGURE 1. Example Process Map for Standard Clinical Encounter patient complaints, and attribute data like percent or

Das et al
proportion of patients suffering harm. For data that are not Figures 3A and B are the run chart and the control chart
time dependent, the following tools are used to display data: representation of the number of days for a new patient clinic
(1) Pareto charts, (2) histograms, and (3) scatter plots.11 A visit after a referral is placed in the system. Although they
Pareto chart is a bar graph in which the lengths of the bars both clearly represent the improving trend, the control chart
represent frequency of the event of interest. The bars are has defined control limits, which gives us added information
arranged with the longest bars on the left and the shortest to about the inherent variation of our current clinic intake
the right to visually depict which events are more common. process. Types of variation seen in processes are catego-
Figure 2 is a Pareto chart for a project to focus on decreasing rized as “common cause” or “special cause” variation. The
emergency department (ED) visits among patients with common cause variation is an inherent part of the process
cancer. The baseline data consisted of the ED diagnoses and includes predictable, random variation. Special cause
for 50 patients, which are displayed using this format. It variation is not an expected part of the process; it is non-
is visible right away that 66% of patients came to the ED random, and the results are not predictable. In the given
with gastrointestinal symptoms and dehydration. This example of referral wait times, it would be helpful for the
would be useful information for the QI team, which might process improvement team to emphasize the fact that,
choose to implement strategies in clinic to prevent these within the limits of the existing process, patients referred to
chemotherapy/cancer-related side effects and help de- this clinic would be seen between 2.46 and 24.49 days (Fig.
crease ED use. 3B). This variation is common cause and is to be expected;
When displaying data that have a time component, run it is often referred to as the voice of the process. If the clinic
charts, Shewhart/statistical process control charts, and time finds this variation to be unacceptable, they must design an
series are used. A run chart (Fig. 3A) is a graphic display of entirely new process to be developed around the acceptable
data plotted in some type of order that includes a central wait times, which is also called the voice of the customer.
tendency.12 Run charts are most commonly used to make However, the data points that lie outside the upper control
process performance visible and analyze data for patterns limit are delays that are secondary to factors not related to
or signals, but it cannot determine if a process is stable. the intake process and should be looked at further to
Statistical control charts (Fig. 3B) are used to understand prevent them (special cause variation). Evaluating variations
and predict process variability (common cause vs. special while conducting QI projects is important for meaningful
cause) and are thus the best tools to determine if our im- interpretation of past data and thus prediction or intervention
provement strategies have the desired effect. appropriate for future performance.
FIGURE 2. Pareto Chart Displaying Reasons for Emergency Department (ED) Visits in an Oncology Clinic

Argentina, Australia, Brazil, the European Union (all

countries), India, New Zealand, Pakistan, the Philippines,
and Saudi Arabia. Several practices outside the United
States participate in this oncologist-led, practice-based
quality assessment program that helps the programs use
data-driven metrics to drive process improvement and is
available to any practice with an ASCO member with no
participation fees. 15 ASCO is also making available its
Quality Training Program in the form of workshops outside of
the United States that helps train multidisciplinary teams to
design and implement process improvement initiatives
locally.
Quality Improvement Challenges and Processes to

Overcome Them
How do oncology professionals incorporate QI within their
everyday workflow? In a recent survey-based study done
within Canadian oncology practices, more than 95% of the
oncologists agreed that QI was an important aspect of their
practice, but only 49% had participated in QI within the past
5 years.16 Unfamiliarity with QI methodologies, lack of time,
and minimal institutional support/recognition were found to
be barriers contributing to the low participation of physicians
in these projects. Value-based care payment reform re-
quires practices to undertake QI initiatives but are fraught
FIGURE 3. (A) Run Chart and (B) Control Chart for Referral Wait Times with issues regarding timely data, funding, and knowledge
of the QI process. However, multiple QI training programs
are emerging. For example, the Institute for Healthcare
Once an intervention is identified, the change is imple- Improvement Open School provides a complete catalog of
mented: “Do.” Postimplementation data are collected in online courses, including more than 35 continuing edu-
real time and analyzed under the “Study” component of this cation credits for nurses, physicians, and pharmacists as
cycle. Decisions are made about additional interventions or well as a Basic Certificate in Quality and Safety. ASCO
sustenance of a meaningful change, as evidenced by im- administers a Quality Training Program that prepares
proved process or outcome measures: “Act.” Multiple PDSA physician-led oncology teams to design, implement, and
cycles are often necessary to achieve the desired results. lead successful QI activities in their practice settings. Since
Quality Improvement in Low- and Middle-Income Countries its inception, the ASCO Quality Training Program has had
The World Health Organization’s Health System Framework multiple domestic and international teams participate. In the
includes six building blocks of a health system, in which post-program surveys, most participants noted increases
quality was classified as one of the fundamental connecting in overall knowledge and competence in the use of QI
pieces.13 Some of the recommendations for incorporating methodologies and practical tools to make changes in their
QI and HSR into policy making and practice in low- and practices.17
middle-income countries include building coalition for HEALTH SERVICES RESEARCH IN HEALTH CARE
quality in these countries by engaging national ministries,
following this up with formation of QI interests groups with During the past 20 years, HSR has been integral to un-
the local stakeholders, and finding best ways to include derstanding and defining quality cancer care delivery. HSR
rapid cycle improvements and HSR within the existing is a multidisciplinary field of study that seeks to identify the
infrastructure.14 most effective way to organize, deliver, and finance health
care to maximize the quality and value of care at both the
The Quality Oncology Practice Initiative was designed by individual and population levels. HSR can provide impor-
a group of ASCO-affiliated medical oncologists to assess and tant insights into the current delivery of cancer care, iden-
improve processes of care in oncology practices, and tify gaps in care or areas for improvement, and help guide
more than 700 oncology practices within the United States the development of interventions aimed at improving
have participated to date. This program is now available in overall care.
39 countries outside the United States and its territories:

Das et al
Health Services Methodology association between various exposures and the outcome. In
HSR uses a mixture of qualitative and quantitative meth- this study, three exposures (prior radiotherapy, underlying
odologies to describe current practice and gaps and to lung conditions, and combination therapy) were found to be
generate the evidence basis that can help inform clinicians, associated with pneumonitis.20
institutions, and systems on how to best improve the quality Qualitative research uses techniques such as observation,
of cancer care. Methods include observational studies (Fig. structured interviews, and focus groups in an attempt to
4) aimed at establishing an association between an expo- understand health behaviors. For example, a qualitative
sure and an outcome, experimental studies evaluating the study by Henson et al21 focused on exploring issues that
impact of an intervention, and qualitative studies focused on influence the decision to seek care in the ED for patients
elucidating the experience and views of participants to with advanced cancer. This study used semistructured
develop frameworks for understanding social influence.18 interviews with patients with advanced cancer who had used
Cohort studies focus on patient populations defined by the ED at a large university-based hospital in London, U.K.
exposure to an identified variable and follow the populations Analysis of the interviews identified several issues that influ-
prospectively to report associations between the exposure enced decision-making, including disease-related anxiety,
and an outcome. As an example, in a study by Enright and prior patterns of health-seeking behavior, feelings of safety at
colleagues,19 a cohort of women who received chemo- home, and difficulties accessing community health care ser-
therapy for early-stage breast cancer was identified (ex- vices. The data informed health professionals and policy
posed) and compared with a group of noncancer controls makers on how to best develop systems of care that may be
(unexposed) with respect to the number and frequency of able to reduce ED use in patients with advanced cancer.21
ED visits and hospitalizations (outcome). This study re- Randomized clinical trials provide the most rigorous eval-
ported that the proportion of patients with at least one ED uation of the impact of interventions in health care, but often
visit and/or hospitalization was higher in the patients with do so in tightly controlled conditions and, therefore, may not
breast cancer undergoing chemotherapy compared with be generalizable to routine practice. Thus, pragmatic ran-
the noncancer controls (43.3% vs. 9.5%; p , .001). Pa- domized control trials are being used in HSR in an attempt
tients with breast cancer were more likely to have multiple to rigorously test the impact of interventions aimed at im-
ED and hospital visits (17.9% vs. 2.4%; p , .001). The proving quality. Pragmatic trials are designed to evaluate
study also used multivariable analysis, a statistical tech- effectiveness of interventions in routine practice condi-
nique used to analyze multiple independent variables at tions to provide more broadly generalizable results related
once, to identify patient (comorbidity), treatment (having to quality. The ongoing Ambulatory Toxicity Management
received docetaxel), and health system (geographic region) study is an example of a pragmatic cluster randomized
factors associated with an increased risk of ED visits or study evaluating proactive symptom management for pa-
hospitalizations.19 tients receiving adjuvant chemotherapy for breast cancer
Case-control studies identify populations based on an out- during ED visits or hospitalizations.22 In this study, cancer
come and look retrospectively to determine the associa- centers were randomized to provide either nurse-led, pro-
tion of the outcome with prior exposures. In a case-control active, telephone-based toxicity management for all patients
study, Cui and colleagues20 identified a patient population receiving adjuvant chemotherapy for breast cancer (in-
who had been treated with an immune checkpoint inhibitor tervention) or standard monitoring (control). The outcome,
and divided this population based on whether they had ED visits or hospitalizations, will be evaluated using ad-
developed pneumonitis. These cases and controls were ministrative data. The advantage of such a design is that it
then reviewed retrospectively to determine the
FIGURE 4. Distinction Between

Cohort and Case-Control Designs

allows a rigorous evaluation in a real-world setting that may across a health system or institution may translate to more
be easily translatable to other regions. meaningful and attainable improvement efforts.27,28
Using Health Services Research to Understand Quality Another strategy to establish achievable benchmarks for
The applications of HSR to the field of quality of cancer care performance on quality measures is the data-driven ap-
has been extensive, considering questions such as (1) What proach using large observational cohorts. In a study by
is the current quality of cancer care? (2) What should we Powis and colleagues,29 achievable benchmarks for a panel
be aiming for? and (3) How can we best improve our 15 quality measures specific to the quality of systemic
performance? therapy delivery for patients with early breast cancer were
calculated based on the achieved performance of the top-
Measuring current performance Population-level assess- performing institutions. This approached allowed for in-
ments of quality have the potential to identify gaps in quality stitutions to be ranked relative to the benchmark for each of
and opportunities for improvement. However, they provide the quality measures and allowed for identification of factors
little guidance on where improvement efforts might be most associated with high performance across quality measures,
impactful and at what level (i.e., provider, institution, sys- such as institution size.29
tem) they should be focused. Variation in performance on
quality measures has been proposed as a method for This approach can also be helpful in adapting perfor-
quantifying and ranking quality measures based on the mance targets to different health care systems or regions.
likelihood of impact.23 This approach has been applied Performance targets for several end-of-life measures were
to population cohorts to provide evidence-driven strategies established using Surveillance, Epidemiology, and End
to help guide prioritization among quality measures.24,25 Results–Medicare data by Earle and colleagues.30 Using
Hassett and colleagues24 used the performance data for achieved performance, the benchmark for death in hospital
30 quality indicators from 9,019 patients with breast cancer was set at 17% and admission to the intensive care unit at the
to systematically prioritize the measures based on potential end of life, at 4%. These benchmarks are now widely used in
to affect quality. An algorithm incorporating the number of QI programs. The same end-of-life measures have been ex-
nonconcordant patients, concordance across institutions, amined across Canada using population data by Barbera and
the highest concordance of any single institution, and an colleagues.31 Because of variation in health system structure
estimate of the clinical impact of concordance was de- and culture between countries, direct translation of the U.S.
veloped and applied to the measures. The highest-ranking benchmarks from Surveillance, Epidemiology, and End
measures included the recommendation for chemotherapy Results–Medicare may not have been applicable. Therefore,
and hormone therapy for hormone receptor–positive breast new achievable benchmarks were established based on the
cancer and radiation therapy after breast-conserving surgery.24 top performers across Canada, establishing a target death in
Using a similar approach, Enright and colleagues25 used the hospital rate of 38% and intensive care unit admission rate of
performance of a population-based cohort of 28,427 patients 2%.31 Additional study of the systems associated with the high-
with early-stage breast cancer treated in Ontario, Canada, to performing institutions driving the benchmark performance
create a systematic approach to prioritizing quality measures may provide more insights to drive quality.
based on the potential for impact. This approach created Targeted areas for improvement Large population-level co-
a ranking based on the interinstitutional variation in perfor- horts and case-control studies reported a higher rate of ED
mances and patient volume. ER use during chemotherapy and visits or hospitalizations in patients receiving both adjuvant
timeliness of chemotherapy were identified as having the and palliative-intent chemotherapy for breast and lung
highest potential impact.25 These structured approaches can cancer than expected from clinical trial data.19,32-35 A recent
help provide data-driven insights to health systems and in- meta-analysis by Prince and colleagues36 identified 112
stitutions when setting QI priorities. observation studies, representing 308,662 patients, re-
Setting benchmarks for quality To drive improvement in porting on ED visits or hospitalizations during chemotherapy
quality across cancer care systems, it is important to es- in routine practice.32 In the meta-analysis, treatment-related
tablish how these systems should be performing. Many and all-cause hospitalizations were more common in pa-
performance targets are established subjectively using tients treated in routine practice than those treated in
consensus-driven approaches to establish best practice randomized clinical trials (29% vs. 16% and 42% vs. 28%,
standards.26 This approach often results in aspirational respectively). This was seen consistently across disease
targets to encourage QI and runs the risk of setting targets sites and treatment intent (adjuvant vs. palliative) and
that are unrealistic or unattainable. When prioritizing and across various global regions. Although some unplanned
planning health system– or practice-level QI efforts, the chosen care is anticipated in patients receiving systemic therapy,
quality measures should be meaningful, actionable, and the ratio of ED visits to hospitalizations suggests that some
achievable. A better understanding of what is achievable proportion are potentially avoidable.36

Das et al
Understanding predictors of ED visits and hospitalizations is patients at 20 Canadian sites across multiple years. At the
an important step in identifying opportunities for targeted time of study initiation, no sites had telephone-based
QI efforts. Patient and disease factors (age, comorbidity, symptom monitoring for this population. However, several
stage), treatment factors (docetaxel-based chemotherapy sites prioritized reducing hospitalizations in local QI projects,
vs. other, use of growth factor support)19,37,38 and health including some telephone-based outreach to patients re-
system factors (health region/authority, rural vs. urban ceiving chemotherapy during the course of the study. Al-
status, access to a primary care physician)19,37,39 have been though randomization minimizes the impact of confounders
identified as predictors of unplanned ED visits or hospital- that would affect all participants, such as national policy
izations while on chemotherapy treatment and may serve as changes that may be rolled out over time, this does not
targets for improvement efforts. The ability to proactively account for changes at a local level. For the Ambulatory
identify patients at increased risk for acute care visits Toxicity Management investigators, local infrastructure and
during chemotherapy may further target improvement ef- programmatic modifications, as well as the cultural shifts
forts. Using a population-based cancer cohort, Grant and secondary to institutional emphasis on hospitalizations, had
colleagues40 developed and validated a prediction model for the potential to confound both the likelihood of sites en-
acute care visits based on patient (age, history of ED visits) rolling patients in the trial and the interpretation of results.
and treatment or tumor factors. For example, the impact of the intervention would be an-
ticipated to be less in sites that have robust, standard-of-
DIFFERENCES BETWEEN QUALITY IMPROVEMENT care symptom monitoring programs compared with those
AND HEALTH SERVICES RESEARCH who did not. To mitigate this research challenge, the re-
Although QI and HSR both focus on improving the care of searchers conducted a qualitative evaluation to document
patients, key differences exist in purpose, approach, and known changes occurring at participating sites. This con-
interpretation of findings (Table 1). These differences can textual information will be leveraged in secondary analyses
be leveraged to meet the needs of investigators. aimed to evaluate the impact of the intervention at sites with
different levels of support for patients undergoing chemo-
Challenges Related to Overlap in Quality Improvement and therapy. Such analyses will be essential to understanding
Health Services Research how these HSR findings can be generalized to other health
A common challenge encountered for health services re- systems.
searchers is how to understand findings in the context of
Another challenge is the potential overlap of support ser-
a changing implementation environment, which may have
vices for patients. As an example, HSR demonstrated that
overlapping QI initiatives. This challenge was encountered
routine collection of patient-reported outcomes (PROs) im-
in the Ambulatory Toxicity Management study, which is the
proved efficiency of symptom assessment, patient-clinician
pragmatic cluster randomized trial of telephone-based,
communication and satisfaction, symptom control, and well-
proactive symptom management in patients receiving ad-
being, survival, and health care use.41-52 Furthermore, PRO
juvant or neoadjuvant chemotherapy in early-stage breast
implementation is a component of the upcoming payment
cancer, described previously.22 The primary outcome of the
model by Medicare, the Oncology Care First Model. As centers
trial is ED visits and hospitalizations. This study enrolled
consider their approach to PRO implementation, they may
encounter overlap with ongoing HSR projects. Investigators
TABLE 1. Differences Between Quality Improvement and Health should consider implementation approaches that prevent
Services Research duplication of similar surveys, which may confuse patients or
Quality Improvement Health Services Research lead to frustration related to survey burden. Dialogue between
Purpose Address a local problem Develop generalizable the QI and health services teams can help align initiative and
knowledge provide maximal, efficient evaluation of PROs for patients with
Duration Rapid cycles (months) Long duration (years) cancer.
Design Preport, interrupted time Randomized designs are
Opportunities for Synergy
series gold standard
Intervention Small, iterative changes Capacity for more complex Close alignment of health services researchers and QI ex-
interventions perts can benefit everyone. The University of Alabama at
Environment Embraces changes Minimizes changes to limit Birmingham has leveraged both QI and HSR in an alter-
confounding native approach to maximize results and optimally translate
Validity Process measurement, Statistical testing of research interventions into practice. In 2011, the University
face validity hypothesis of Alabama at Birmingham elected to pursue participation in
Funding Internal funding External funding
the Oncology Care Model,53 Medicare’s payment reform
demonstration project. One requirement for participation

was to provide patients a comprehensive treatment plan aspects of the treatment planning experience (PROs,
containing the Institute of Medicine’s 13 recommended shared decision-making) in different disease types. These
components.53 In an initial quality-oriented needs assess- projects are ongoing. If results are positive, the operation
ment, this requirement was identified by the operational team will consider adoption of these additional components
team as a substantial challenge. The administrative team as a standard of care using a QI approach.
engaged with health services researchers to study formally
Both the clinical teams and the research teams report ben-
the impact of a specific technology-based solution that
efits from a coordinated approach to changing the care de-
would meet this requirement in a small sample of breast
livery system. The team science approach56 of administrators,
cancer patients.54 The study, which was grant funded,
clinical team members, and health services researchers
demonstrated improvement in quality measures defined
participating in the discussion leads to innovative solutions.
with the ASCO Quality Oncology Practice Measures.55 This
research study was presented to the Oncology Care Model This approach fosters collaboration and inclusion of diverse
operations committee, who then initiated a large QI-based perspectives, which builds on the strengths of individuals.56
implementation of the treatment planning software using The rigorous statistical approach used by HSR helps to better
multiple PDSA cycles. This effort was supported by funding inform the selection of interventions and process changes that
from the health system, rather than by research dollars. This are likely to lead to improvement or to be beneficial. In addition,
QI effort resulted in sustainability for the intervention tested it allows for dissemination of research findings, which has the
in the initial research study, which is a known challenge potential to aid other institutions with similar goals and informs
for health services interventions. When the infrastructure the policy discussion. From the perspective of the health
for treatment planning was complete, health services re- services researcher, this approach provides a mechanism for
searchers were again engaged to leverage the infrastructure sustainability of successful interventions, which has been
created by the Oncology Care Model operation team in five a challenge of this discipline. Synergy between these teams
additional grant-funded research projects targeting specific can be impactful for health care delivery.

1
University of Alabama at Birmingham (UAB) Comprehensive Cancer Gabrielle Rocque, MD, MSPH, Department of Medicine, Division of
Center, Birmingham, AL Hematology and Oncology, University of Alabama at Birmingham,
2
UAB Division of Hematology and Oncology, Birmingham, AL 1720 2nd Ave. South, Birmingham, AL 35294; Twitter: @GRocqueMD;
3
Birmingham VA Medical Center, Birmingham, AL email: [email protected].
4
Texas Oncology, PA, Dallas, TX
5
Carlo Fidani Regional Cancer Centre, Trillium Health Partners,
Mississauga, Ontario, Canada AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
6
University of Toronto, Division of Medical Oncology, Toronto, Ontario, AND DATA AVAILABILITY STATEMENT
Canada Disclosures provided by the authors and data availability statement (if
7
UAB Division of Gerontology/Geriatrics/Palliative Care, Birmingham, AL applicable) are available with this article at DOI https://doi.org/10.1200/
EDBK_281093.
REFERENCES
1. Institute of Medicine. Committee on Quality of Health Care in America. Crossing the Quality Chasm: a New Health System for the 21st Century. Washington, DC:
National Academy Press; 2001.
2. Aviki EM, Schleicher SM, Mullangi S, et al. Alternative payment and care-delivery models in oncology: a systematic review. Cancer. 2018;124:3293-3306.
3. Kohn LT, Corrigan J, Donaldson MS. To Err Is Human: Building a Safer Health System. Washington, DC: National Academy Press; 2000.
4. Leape LL, Lawthers AG, Brennan TA, et al. Preventing medical injury. QRB Qual Rev Bull. 1993;19:144-149.
5. Chassin MR, Loeb JM. High-reliability health care: getting there from here. Milbank Q. 2013;91:459-490.
6. Institute for Healthcare Improvement. Improving the Reliability of Health Care. Cambridge, MA: Institute for Healthcare Improvement; 2004.
7. Woodhouse KD, Volz E, Maity A, et al. Journey toward high reliability: a comprehensive safety program to improve quality of care and safety culture in a large,
multisite radiation oncology department. J Oncol Pract. 2016;12:e603-e612.
8. Lawal AK, Rotter T, Kinsman L, et al. Lean management in health care: definition, concepts, methodology and effects reported (systematic review protocol). Syst
Rev. 2014;3:103.
9. Shrank WH, Rogstad TL, Parekh N. Waste in the U.S. health care system: estimated costs and potential for savings. JAMA. 2019;322:1501.
10. Lawal AK, Rotter T, Kinsman L, et al. Lean management in health care: definition, concepts, methodology and effects reported (systematic review protocol). Syst
Rev. 2014;3:103.

Das et al
11. Provost LP, Murray SK. The Health Care Data Guide: Learning From Data for Improvement. San Francisco, CA: Jossey-Bass; 2011.
12. Graves KD, Arnold SM, Love CL, et al. Distress screening in a multidisciplinary lung cancer clinic: prevalence and predictors of clinically significant distress. Lung
Cancer. 2007;55:215-224.
13. Crawford KH. Up to the Task: Fostering Quality Cancer Care Around the World. ASCO Connection. 2019. https://connection.asco.org/magazine/features/task-
fostering-quality-cancer-care-around-world. Accessed February 27, 2020.
14. Leatherman S, Ferris TG, Berwick D, et al. The role of quality improvement in strengthening health systems in developing countries. Int J Qual Health Care. 2010;
22:237-243.
15. World Health Organization. Everybody’s Business: Strengthening Health Systems to Improve Health Outcomes: WHO’s Framework for Action. 2007; https://www.
who.int/healthsystems/strategy/everybodys_business.pdf.
16. Lim C, Cheung MC, Franco B, et al. Quality improvement: an assessment of participation and attitudes of medical oncologists. J Oncol Pract. 2014;
10:e408-e414.
17. Keng MM, Cunningham G, Gilmore T, et al. ASCO Quality Training Program: five-year review. J Clin Oncol. 2019;37 (suppl; abstr 7).
18. Pope C, Mays N. Reaching the parts other methods cannot reach: an introduction to qualitative methods in health and health services research. BMJ. 1995;
311:42-45.
19. Enright K, Grunfeld E, Yun L, et al. Population-based assessment of emergency room visits and hospitalizations among women receiving adjuvant chemotherapy
for early breast cancer. J Oncol Pract. 2015;11:126-132.
20. Cui P, Liu Z, Wang G, et al. Risk factors for pneumonitis in patients treated with anti-programmed death-1 therapy: a case-control study. Cancer Med. 2018;
7:4115-4120.
21. Henson LA, Higginson IJ, Daveson BA, et al; BuildCARE. ‘I’ll be in a safe place’: a qualitative study of the decisions taken by people with advanced cancer to seek
emergency department care. BMJ Open. 2016;6:e012134.
22. Krzyzanowska MK, Julian JA, Powis M, et al. Ambulatory toxicity management (AToM) in patients receiving adjuvant or neoadjuvant chemotherapy for early-stage
breast cancer: a pragmatic cluster randomized trial protocol. BMC Cancer. 2019;19:884.
23. Selby JV, Schmittdiel JA, Lee J, et al. Meaningful variation in performance: what does variation in quality tell us about improving quality? Med Care. 2010;
48:133-139.
24. Hassett MJ, Hughes ME, Niland JC, et al. Selecting high priority quality measures for breast cancer quality improvement. Med Care. 2008;46:762-770.
25. Enright KA, Taback N, Powis ML, et al. Setting quality improvement priorities for women receiving systemic therapy for early-stage breast cancer by using
population-level administrative data. J Clin Oncol. 2017;35:3207-3214.
26. Ellis J. Sharing the evidence: clinical practice benchmarking to improve continuously the quality of care. J Adv Nurs. 2000;32:215-225.
27. Kiefe CI, Allison JJ, Williams OD, et al. Improving quality improvement using achievable benchmarks for physician feedback: a randomized controlled trial. JAMA.
2001;285:2871-2879.
28. Kiefe CI, Weissman NW, Allison JJ, et al. Identifying achievable benchmarks of care: concepts and methodology. Int J Qual Health Care. 1998;10:443-447.
29. Powis M, Sutradhar R, Gonzalez A, et al. Establishing achievable benchmarks for quality improvement in systemic therapy for early-stage breast cancer. Cancer.
2017;123:3772-3780.
30. Earle CC, Park ER, Lai B, et al. Identifying potential indicators of the quality of end-of-life cancer care from administrative data. J Clin Oncol. 2003;21:1133-1138.
31. Barbera L, Seow H, Sutradhar R, et al. Quality indicators of end-of-life care in patients with cancer: what rate is right? J Oncol Pract. 2015;11:e279-e287.
32. Prince RM, Atenafu EG, Krzyzanowska MK. Hospitalizations during systemic therapy for metastatic lung cancer: a systematic review of real world vs clinical trial
outcomes. JAMA Oncol. 2015;1:1333-1339.
33. Du XL, Osborne C, Goodwin JS. Population-based assessment of hospitalizations for toxicity from chemotherapy in older women with breast cancer. J Clin Oncol.
2002;20:4636-4642.
34. Sanoff HK, Carpenter WR, Freburger J, et al. Comparison of adverse events during 5-fluorouracil versus 5-fluorouracil/oxaliplatin adjuvant chemotherapy for stage
III colon cancer: a population-based analysis. Cancer. 2012;118:4309-4320.
35. Hassett MJ, Rao SR, Brozovic S, et al. Chemotherapy-related hospitalization among community cancer center patients. Oncologist. 2011;16:378-387.
36. Prince RM, Powis M, Zer A, et al. Hospitalisations and emergency department visits in cancer patients receiving systemic therapy: systematic review and meta-
analysis. Eur J Cancer Care (Engl). 2019;28:e12909.
37. Ruddy KJ, Van Houten HK, Sangaralingham LR, et al. Impact of treatment regimen on acute care use during and after adjuvant chemotherapy for early-stage
breast cancer. Breast Cancer Res Treat. 2017;164:515-525.
38. Barcenas CH, Niu J, Zhang N, et al. Risk of hospitalization according to chemotherapy regimen in early-stage breast cancer. J Clin Oncol. 2014;32:2010-2017.
39. Bastedo SJ, Krzyzanowska MK, Moineddin R, et al. A population-based assessment of primary care visits during adjuvant chemotherapy for breast cancer. Curr
Oncol. 2017;24:90-94.
40. Grant RC, Moineddin R, Yao Z, et al. Development and validation of a score to predict acute care use after initiation of systemic therapy for cancer. JAMA Netw
Open. 2019;2:e1912823.
41. Seow H, Sussman J, Martelli-Reid L, et al. Do high symptom scores trigger clinical actions? An audit after implementing electronic symptom screening. J Oncol
Pract. 2012;8:e142-e148.

42. Santana MJ, Feeny D, Johnson JA, et al. Assessing the use of health-related quality of life measures in the routine clinical care of lung-transplant patients. Qual
Life Res. 2010;19:371-379.
43. Kroenke K, Krebs EE, Wu J, et al. Telecare collaborative management of chronic pain in primary care: a randomized clinical trial. JAMA. 2014;312:240-248.
Oncol. 2016;34:557-565.
45. Cleeland CS, Wang XS, Shi Q, et al. Automated symptom alerts reduce postoperative symptom severity after cancer surgery: a randomized controlled clinical trial.
J Clin Oncol. 2011;29:994-1000.
46. Gilbert JE, Howell D, King S, et al. Quality improvement in cancer symptom assessment and control: the Provincial Palliative Care Integration Project (PPCIP).
J Pain Symptom Manage. 2012;43:663-678.
47. Valderas JM, Kotzeva A, Espallargues M, et al. The impact of measuring patient-reported outcomes in clinical practice: a systematic review of the literature. Qual
Life Res. 2008;17:179-193.
48. Chen J, Ou L, Hollis SJ. A systematic review of the impact of routine collection of patient reported outcome measures on patients, providers and health
organisations in an oncologic setting. BMC Health Serv Res. 2013;13:211.
49. Detmar SB, Muller MJ, Schornagel JH, et al. Health-related quality-of-life assessments and patient-physician communication: a randomized controlled trial.
JAMA. 2002;288:3027-3034.
50. Basch E, Artz D, Dulko D, et al. Patient online self-reporting of toxicity symptoms during chemotherapy. J Clin Oncol. 2005;23:3552-3561.
51. Velikova G, Booth L, Smith AB, et al. Measuring quality of life in routine oncology practice improves communication and patient well-being: a randomized
controlled trial. J Clin Oncol. 2004;22:714-724.
52. Berry DL, Blumenstein BA, Halpenny B, et al. Enhancing patient-provider communication with the electronic self-report assessment for cancer: a randomized
trial. J Clin Oncol. 2011;29:1029-1035.
53. Center for Medicare and Medicaid Services. Oncology Care Model. 2015; http://innovation.cms.gov/initiatives/Oncology-Care/. Accessed January 16, 2020.
54. Rocque GB, Cadden A. Creation of Institute of Medicine care plans with an eye on up-front care. J Oncol Pract. 2017;13:512-514.
55. Rocque GB, Williams CP, Hathaway AR, et al. Evaluating the impact of treatment care planning on quality measures. J Oncol Pract. 2019;15:e271-e276.
56. Little MM, St Hill CA, Ware KB, et al. Team science as interprofessional collaborative research practice: a systematic review of the science of team science
literature. J Investig Med. 2017;65:15-22.

Successful Strategies to Address Disparities:

Insurer and Employer Perspectives
Manali I. Patel, MD, MPH, MS1,2,3; Richard Snyder, MD4; and Otis Brawley, MD, MACP, FASCO, FACE5
overview
Disparities in cancer have been documented for decades and continue to persist despite clinical ad-
vancements in cancer prevention, detection, and treatment. Disparate cancer outcomes continue to affect
many populations in the United States and globally, including racial and ethnic minorities, populations with
low income and education, and residents of rural areas or low socioeconomic neighborhoods, among others.
Addressing cancer disparities requires approaches that are multilevel. Addressing social determinants of
health, such as removing obstacles to health (e.g., poverty, discrimination, access to housing and education,
jobs with fair pay, and health care) can reduce cancer disparities. However, to achieve cancer health equity,
multilevel approaches are required to ensure that access to high-quality cancer care and equitable receipt of
evidence-based services can reduce cancer disparities. Policy, health system interventions, and innovative
delivery and health care coverage approaches by private and public payers, employer-based payers, and labor
union organizations can assist in ensuring access to and receipt of high-quality cancer care while addressing
the high costs of care delivery. Partnerships among patients, caregivers, employers, health care providers, and
health care payers can make impactful changes in the way in which cancer care is delivered and, in turn, can
assist in reducing cancer disparities.
INTRODUCTION would require that all Americans get what is currently

The National Cancer Institute (NCI) defines cancer health available to college-educated Americans. The 122,000
disparities as adverse differences in cancer incidence, patients that form the disparate population lived
cancer prevalence, cancer mortality, cancer survivorship, throughout the entire United States, but a larger pro-
and burden of cancer or related health conditions that portion lived in the southern United States than in any
exist among specific populations among groups in the other region. The majority of the disparate population is
United States.1 white and lives in states that have not expanded
Medicaid under the Affordable Care Act. This latter fact
Epidemiologists at the American Cancer Society have suggests that state-by-state disparities that already
tried to quantify and describe those populations suf- exist are destined to grow in the future. Thus, health
fering from disparate cancer health in the United
disparities are not a racial issue but a largely sociopolitical
States.2 The impact of education on health outcomes is
issue.
tremendous for people of all races/ethnicities. It is a fact
that college-educated Americans have a much lower Health disparities involve a failure of a large portion of
risk for death from cancer compared with non–college- society to benefit from preventive activities, screening
educated Americans. In 2019, approximately 607,000 activities, and medical diagnostic and therapeutic in-
Americans died of cancer. If all Americans had the risk terventions. Overcoming health disparities is a moral
of death of the approximately 30% of Americans who are and ethical issue. It is a responsibility for all of society;
Author affiliations
and support college educated, one-fifth of the 607,000 deaths all have an obligation to address it. Elements that can
information (if (about 122,000) would not have occurred. This is the address and reduce disparities in health are at multiple
applicable) appear number of deaths that would have occurred if all levels and include local, state, and federal governments,
at the end of this educational institutions, and community organizations.
Americans had the benefits in terms of prevention,
article.
screening, early detection, diagnosis, and treatment that Health care providers, health insurers, and employers
Accepted on March
college-educated Americans have throughout their have a special advantage in reducing disparities in
2, 2020 and
published at entire lifetime. health.
ascopubs.org on
A total of 122,000 deaths is a rough estimate of how Today, an abundance of outcomes data is available,
April 2, 2020:
many people died preventable death from cancer if and it has stimulated the academic field of cancer health
10.1200/EDBK_ there were broader access to current technology. It disparities. The first academic papers to discuss dif-
279959 would not require a new treatment to obtain this. It ferences in cancer outcomes among populations were

Multilevel Approaches by Payers and Employers to Reduce Cancer Disparities
useful because it collects data on insurance status. Analysis

of the National Cancer Database and creative combinations
of SEER and Centers for Disease Control and Prevention
• Cancer disparities, which describe the mea- data with U.S. Census income data and Medicare payment
surable differences in cancer outcomes in
data helped to move the emphasis from racial differences to
population groups, have persisted for more than
3 decades, despite advancements in cancer socioeconomic differences.
care and treatment. Yet, disparities can be Disparate populations are generally poorer, have less ed-
tackled and overcome when addressed by ucation, and vary by social acculturation. These are factors
multilevel interventions.
that influence, if not determine, how or if the patient in-
• Disparities in cancer are a function of several teracts with the health care system and health care pro-
factors, including social determinants of health viders. Many of the solutions to cancer health disparities are
and lack of equitable provision of high-quality
sociopolitical. Some prefer to discuss “cancer health equity”
cancer prevention, treatment, palliation, and
survivorship care. instead of “cancer health disparities.” This is an acknowl-
edgment that some populations need more assistance than
• Addressing social determinants of health, such
others to get the parity.
as education, can reduce some cancer
disparities. The American health care system is disjointed and in-
• To achieve cancer health equity, multilevel efficient. It emphasizes diagnosis and treatment more so
approaches are needed across policy, patients, than a lifetime of prevention and risk reduction. Cancer
caregivers, providers, health systems, and prevention is an often underappreciated element in the
payer organizations. reduction in cancer health disparities. The population prone
• Novel delivery and payment approaches to disparate outcomes has higher rates of cancer risk
implemented by payer organizations, em- factors, such as smoking, obesity, and energy imbalance,
ployers, and organized labor can improve ac- compared with the population as a whole, over a lifetime. In
cess to high-quality cancer care across the this sense, the prevention of health disparities is a pediatric
cancer continuum and, thus, reduce disparities issue. Counseling and coaching to encourage healthful
in cancer outcomes.
behaviors are risk-reduction activities of benefit and should
start before birth. Lack of access to high-quality care is
a reason for population differences.
published in the 1970s. These studies largely looked at
black-white outcomes, and the initial emphasis was on Some disparities are due to difficulty with accessing quality
“minority health.” The NCI Surveillance, Epidemiology, and care. Often, the reasons for disparate receipt of care are
End Results (SEER) program was established by the NCI in logistical and nonmedical. The reasons for these disparities
1971.3 SEER began by gathering and publishing incidence, can involve the inability to afford parts of care or even lack of
mortality, and 5-year survival data for blacks and whites with transportation. The health care payer and health care pro-
cancer in nine distinct areas of the country. SEER allowed for vider are becoming more active in activities that can reduce
the NCI Black/White Cancer Survival Study, which began in cancer health disparities. Although some of the innovations
the 1980s and continued well into the 1990s.4 The Black/ by health care payers and providers may often be motivated
White Cancer Survival Study documented population differ- by reducing costs of care, the effect of these activities can
ences in cancer incidence, mortality, and 5-year survival by improve quality of care and reduce disparities. There is in-
race and area of residence. It also documented differences in creasing interest in value-based health care payment models,
patterns of care.5 in which providers enjoy a higher reimbursement when there
The SEER program began collecting data using the five U.S. are good outcomes. More and more providers are moving
Office of Management and Budget racial and ethnic cat- toward team-based health care that deals with the whole
egories in the early 1990s.4 Cancer outcomes data are patient and the patient’s realities of life. Providers commonly
gathered and published by the NCI SEER program (https:// employ social workers to help patients and families overcome
seer.cancer.gov), state registries funded by the U.S. Centers the barriers to obtaining care. More recently, some systems
for Disease Control and Prevention (https://www.cdc.gov/ are employing navigators to provide even more personal
cancer/dcpc/data/index.htm), and the American College of assistance. Navigators can be nurses acting as patient ad-
Surgeons–sponsored National Cancer Database (https:// vocates or laypersons who may not be trained medically but
www.facs.org/quality-programs/cancer/ncdb). The National can be trained on specific activities and integrated into the
Cancer Database collects data from selected hospitals and health care system. Navigators can provide education to pa-
health systems throughout the United States. It is especially tients and families, as well as help with things like scheduling

Patel, Snyder, and Brawley
appointments, arranging transportation, and overcoming lo- to health care services. Strategies must address the features
gistical barriers to obtaining health care. of rural, suburban, and urban populations.
This type of assistance is only available once the patient is in Health Plan–Funding Mechanism
the health care system. Our current health care system in
Health plans are uniquely positioned to address health care
the United States is fragmented and many pieces of health
access. Funding mechanisms differ depending on the health
care are delivered in silos. These silos and the associated
plan. In the United States, health plans provide coverage
disjointedness of health care are not helpful. Currently,
through programs and products funded by the government,
many still get access to the health care system after the
by employers, or by individuals. Government-funded pro-
diagnosis of a major illness. There is increasing migration
grams are fully insured or put health plans at risk for the cost
from fee-for-service health care to alternative payment
of care. In employer-funded programs, almost two-thirds of
models. Patients who are part of health maintenance or-
members’ care is self-funded, with health plans adminis-
ganizations or Medicare Advantage Plans frequently have
tering the benefits selected by the employer.
access to health prevention and risk-reduction interventions
before a diagnosis. This may improve outcomes and reduce Health Plan Design
disparities. Government-funded programs include managed Medicaid,
Today, health care payers can often monitor billing patterns Medicare Advantage, and exchange-based Affordable Care
and identify a patient who may have difficulty getting proper Act products with defined benefits and some flexibility
care. It is not unusual for payers to call or mail insured around incremental benefits.
patients, reminding them that the insurance has not re-
Products and features of health plans range from more
ceived a bill for breast or colon cancer screening, and,
restrictive networks and medical management at lower cost
therefore, the patient may be overdue. It is not unusual for
through exclusive provider organizations or health mainte-
a patient with chronic disease, such as hypertension, to be
nance organizations that allow health services from doctors,
counseled by the insurer that they should have regular blood
hospitals, and other care providers who are within a certain
pressure checks or for those who are prescribed diabetes
network to less restrictive network access and medical
medicine to be reminded that they should have their he-
management at higher cost through preferred provider
moglobin A1c checked regularly. The insurance may be
organization products that provide more flexibility. Member
able to identify individuals who are having difficulty adhering
costs built into these products include deductibles, copay-
to medication or treatment regimens and provide home
ments, and coinsurance. These are intended to drive
health assistance, such as visiting nurses.
accountable use of services but may have the unintended
In addition to these efforts, payers and employers have consequence of delaying access. Some products waive
implemented a variety of innovative care-delivery ap- copayments for preventive care in certain settings.
proaches to improve cancer care delivery and reduce
Health plans and self-insured employers generally recog-
cancer disparities. Clinical transformation efforts and
nize that primary prevention and early detection result in
alignment of financial incentives to reward value-based
improved outcomes and lower costs than treatment of
care, such as those led by the Independence Blue Cross
advanced disease. Benefits and incentives promote member
Foundation, and unique patient-centered employer-based
and provider behaviors that achieve the goals of prevention
initiatives can ensure receipt of value-based cancer care
and early detection.
and reduce disparities in cancer outcomes. These unique
approaches, which are described later in this article and in Increasingly, health plans are moving away from traditional
the Sidebar, along with efforts by policy makers, health fee-for-service medicine, which incents overutilization and
advocates, health systems, and health providers can help to fragmentation. Progressive health plans and the Center for
overcome some of the barriers to achieving cancer health Medicare and Medicaid Services, through the Center for
equity. Medicare and Medicaid Innovation,6 are rapidly adopting
reimbursement models featuring capitation, episodes of
STRATEGIES BY HEALTH PLANS TO REDUCE CANCER
care, global budgets, and various gainsharing and downside
CARE DISPARITIES
risk arrangements in return for driving integrated, team-
To address disparities in cancer care prevention, diagnosis, based, accessible, well-coordinated, high-quality, and af-
and treatment, one must first understand the populations fordable care with a good experience.
in question, the nature of health care coverage products,
and their rapidly changing funding mechanisms. The rate- The Business Case for Equitable Cancer Care
limiting steps to achieving optimal cancer care are strongly Many mission-driven health plans have long histories in the
influenced by culture, socioeconomic factors (social de- communities that they serve and a commitment to improve
terminants of health), education, health literacy, and access health of the community at large, which are primary drivers

SIDEBAR. Multilevel Approaches to Improve Cancer Care Delivery

PROVIDER
• Provide team-based health care to address the whole patient and social determinants of health (e.g., through navigators,
social workers, and patient advocates).
• Counsel and coach to encourage healthful behaviors.
• Participate in value-based cancer care models.
• Develop and implement novel stakeholder-engaged approaches to deliver high-value cancer care.
• Advocate policies and procedures to improve equitable cancer care delivery.
• Address unmet needs and barriers to care.
PAYER
• Determine barriers to care by monitoring billing patterns.
• Employ proactive approaches to remind patients to adhere to care (e.g., screening and treatments).
• Promote healthy lifestyles (e.g., member incentives and programs).
• Engage in care delivery (e.g., home health assistance and visiting nurses).
• Use novel payment models to incentivize high-quality care.
• Ensure provision of payment for screening, immunization, and social determinants of health.
• Integrate ancillary services (e.g., behavioral health) into cancer care delivery.
• Communicate through a variety of strategies (e.g., text messaging) to close gaps in access and care.
EMPLOYER ORGANIZATIONS
• Ensure coverage that improves health and well-being of employees and their dependents at high quality and low cost to the
employee.
• Design comprehensive cancer care benefits that include screening, genetic testing, treatment, pharmaceutical coverage,
access to 24/7 emergency care, advice for treatment side effects, behavioral medicine, physical therapy, palliation, and
survivorship, among other evidence-based cancer services.
• Encourage use of high-quality cancer care services free of charge to the employee and dependents.
• Engage in partnerships with cancer centers to ensure specialized access to high-quality cancer care providers for care
delivery, second opinions, care coordination, and ancillary cancer care services.
• Steer employees to higher-value providers through narrow networks, special access programs, or direct contracting.
• Identify and implement best practices for building comprehensive cancer service benefits to employees.
• Advocate for policies and procedures to ensure better benefits for workers and their families, including higher wages and
pensions, paid sick leave, and better health coverage at lower costs.
• Address social determinants of health through such activities as providing access to free transportation, food banks, and
housing services.
• Develop and implement innovative approaches to high-quality care delivery (e.g., through the use of health advocates,
navigators, or promoters) to ensure access to and delivery of evidence-based cancer care services.
• Partner with high-value providers and cancer centers to ensure access to high-quality patient-centered cancer care
services.
• Proactively collect data in engagement strategies, including patient-reported outcomes and other metrics that are currently
unavailable to most payer organizations and employers.
for improving primary prevention, early detection, and evidence- Although patients may not be fully aware of the relative
based treatment of cancer in all populations. In a competitive quality of care that they receive, health plans also rec-
environment for health plans and self-insured employers, there ognize that preventable or duplicative care or complica-
are strong economic reasons as well. tions can create avoidable costs. Plans are motivated to
Employers value keeping employees and their families improve outcomes and reduce disparity to optimize re-
healthy, because of outcomes that include lower costs, imbursement and marketing position under programs like
lower rates of absenteeism and presenteeism, and improved the Medicare Advantage and Affordable Care Act star
productivity. Health plans compete for employer business ratings of health plans, which gives consumers a snapshot
on price but also on member experience and retention, of how each health plan’s quality compares with that of
which are driven by access to, coordination of, and navi- other exchange health plans (five stars represents the
gation through the health care system. highest quality).7

Expanding Access Beyond Health Plan Membership Similarly, Independence seeks to engage, enable, and em-
Health plans recognize a duty to address disparities in health power members to achieve the highest possible clinical
care including cancer care for the broader community be- outcomes and eliminate disparities. A very high proportion of
yond their members. Independence Health Group and the members are engaged in an award-winning member health
Independence Blue Cross Foundation (Independence) have, portal through which they choose how they prefer to com-
for many years, provided a steady stream of financial support municate with Independence (e.g., phone, text, email, pa-
to clinics and organizations providing care to underserved per). Most members prefer secure text prompts to close care
communities in the Philadelphia region.8 These dollars gaps and access follow-up care. Switching to text prompting
support screening (e.g., mammography and cervical and has increased the rate of colorectal cancer and mammog-
colorectal cancer), immunizations (e.g., HPV vaccine), and raphy screening materially.
treatment. Funding also addresses the social determinants With some products, members have access to financial
of health (e.g., housing, education, food insecurity, trans- incentives that encourage adherence and eliminate care
portation, enrollment in federal and state programs) that gaps, including cancer screening. In accordance with federal
lead to disparities in health outcomes, especially for the requirements, Independence does not require copayments
chronically ill. Independence has partnered with the Cen- for preventive services. Independence also analyzes aggre-
ters for Disease Control and Prevention, the American gated claim-based and provider electronic health record
Cancer Society, state government, providers, and other data, coupled with real-time admission discharge and
institutions to promote healthy lifestyles (e.g., nutritious transfer data from all health systems. The results are risk
school lunches, access to prescription meals for the stratified and used to prioritize case management staff
chronically ill) and screening programs (e.g., colorectal outreach to members. The goal is to help members at a time
cancer screening and treatment). when they most need assistance to navigate the complex,
and often chaotic, health care ecosystem.
Transitioning to Value-Based Care: The
In summary, the health plan has a central and critical role in
Independence Experience
facilitating access to well-coordinated high-quality afford-
Using the Independence example again, this health plan able health care. For many local and regional health plans
has an overarching strategy to engage, enable, and em- and their affiliated foundations, the deep and caring re-
power providers. Value-based contracts engage providers lationship with the community results in a more compre-
with aligned financial incentives that include accountability hensive response to prevention, early detection, and treatment
for total cost of care (e.g., capitation, episodes of care, percent of cancer; it also addresses biologic, psychological, and
of premium), with increased gainshare for improving and social factors that impact the outcome and experience of the
maintaining high-quality outcomes. The plan enables provider patient.
performance through bidirectional data exchange, data ag-
gregation, and analytics that support role-based real-time EMPLOYERS AND LABOR UNIONS: LEVERAGING THE
reporting to providers and administrators. Joint value com- WORKFORCE TO IMPROVE EQUITABLE ACCESS TO AND
mittees, including health plan and provider leadership, identify DELIVERY OF VALUE-BASED CANCER CARE
joint interventions that empower improvement in access, co- Employers and labor unions play a significant role in re-
ordination, experience, quality, and cost (e.g., episode-based ducing health disparities by improving access to high-
radiation therapy that eliminates the need for precertification, quality care for many workers and their dependents
integration of behavioral health into physical health to address in the United States and globally. This section highlights
clinical outcomes, and increased costs for members with employer and labor union strategies that are aimed at im-
chronic illnesses, including cancer). Physicians get financial proving cancer care delivery and reducing cancer health
incentives for closing care gaps, including screening and inequities.
immunizations. To support physicians in achieving these in-
centives, Independence has facilitated screening programs. Employer-Based Insurance Coverage and Innovative
For example, the names of members with no history of co- Programs to Improve Value-Based Cancer Care Delivery
lorectal cancer screening are compiled on an order sheet that In the United States, almost half of the population is insured
is sent to the treating physician. If the physician agrees, by employer-based health care coverage. Fifty-six percent of
members are sent the less-invasive fecal immunochemical small businesses and 99% of large businesses offer health
test, which can detect hidden blood in the stool. Upon its benefits to their employees.10 Given the continued rise in
completion and return, samples are processed, and results health care costs, employers are increasingly faced with
are sent to the physician. Independence pursues positive difficult decisions about buying care and services that affect
screens to completion by monitoring claims and following up the health and well-being of their employees and their
with physicians.9 dependents. In 2019, for example, annual premiums for

employer-sponsored family health coverage reached $20,576, Group on Health, other employer organizations have
up 5% from 2018, with workers paying an average of $6,015 implemented specialized programs to educate their em-
toward the cost of their coverage.11 Increased premiums, nar- ployees about palliative care and hospice care; encourage
rower and limited health plan benefits, and larger deductibles their employees to consider their goals of care, file their
have led many employer organizations to seek ways to offer advance directives and identify surrogate decision-makers
employees and their dependents high-quality care at lower after their diagnosis; and contract with health plans and
costs. providers to ensure palliative care coverage after a diagnosis
Because employer benefits are frequently the first used for and throughout the continuum of cancer care. In smaller
employees and their dependents after a diagnosis of cancer, and/or community settings in which palliative care is not
many employer organizations are interested in how to as- available, employer organizations, such as the Northeast
sess the value of the cancer care that they purchase to Business Group on Health, have identified palliative care
ensure the greatest efficacy, as well as employees’ or de- specialists who can be consulted remotely to ensure eq-
pendents’ well-being, work productivity, and quality of life. uitable access to high-quality supportive care services. The
The Pacific Business Group on Health, for example, has Northeast Business Group on Health also published
created a specialized program, the Employers Centers of a guide, Delivering Value in Cancer Care: The Employer
Excellence Network,12 to enhance partnerships between Perspective,13 which includes actions, resources, and in-
high-performing cancer centers and local cancer care formation to assist health plans and care providers in best
teams to ensure that patients have access to specialized practices for building comprehensive cancer care benefits
center of excellence–based cancer care and home-based for employees.
cancer care across the continuum of disease. The Pacific
Organized Employment Insurance Coverage and
Business Group on Health contracts with high-value cancer
Strategies to Reduce Cancer Disparities
care centers to ensure access to early appointments,
second opinions, care coordination among specialists, care Like employer organizations, organized employment has
navigators, an emergency advice line for side effects from also led to improved cancer care delivery. Since their in-
cancer treatments, mental health services, integrated pal- ception, early worker and trade unions have played a sig-
liative care, and counseling regarding work schedules, nificant public health role through interventions that
nutrition, and financial strain. The Employers Centers of improve working conditions and promote individual, family,
Excellence Network for oncology provides a high-touch and community well-being globally.14 By collectively bar-
concierge experience while increasing evidence-based care gaining over wages, benefits, and workplace conditions,
receipt, lowering the risk of unnecessary and inappropriate engaging in lobbying activities, and participating in elec-
care, and reducing the total costs of care for patients and tions, unionized work is associated with healthier and safer
employers.12 workplace conditions, as well as higher pay and more highly
valued benefits for its members.15-23 These benefits allow
Other approaches that employers have used to improve
workers the means to purchase healthier foods, live in safer
equitable access to cancer care include exerting pressure
neighborhoods, and remain in a steady occupation without
on health plans and providers to provide data-driven
having to find additional sources of work to supplement their
reporting on quality, outcomes, and patient satisfaction;
income. Rights on the job, such as seniority and grievance
rewarding providers who meet the criteria for excellence by
procedures, provide workers more job stability and less
steering employees to those providers through narrow
stress. Thus, labor unions play a unique role in addressing
networks, special access programs, or direct contracting;
social determinants of health that, in turn, reduce cancer
and encouraging employees to seek second opinions after
health inequities.
a cancer diagnosis. Employers seek to design compre-
hensive cancer care benefits that include high-quality According to the Organization for Economic Cooperation
services and communicate these benefits widely to their and Development,24 although labor union membership
employees. Employers also seek to collect data on en- declined from 45% in 1985 to 33% in 2018, the density of
gagement with existing programs and services and actively labor union membership varies considerably by nation, with
engage with stakeholders, such as patients, physicians, Iceland maintaining the highest membership rate. Since the
hospitals, pharmaceutical industries, advocacy organiza- passing of the National Labor Relations Act in 1935, labor
tions, and policy makers, to consider novel approaches to unions remain a key component of the political and eco-
improving care delivery. Many employer-based organiza- nomic system in the United States. According to the U.S.
tions have sought to cover many cancer care services free of Bureau of Labor Statistics, roughly 1 in 10 workers, or 14.7
charge for the employees and their dependents to en- million people, are organized labor union members in
courage their use. As with the Employers Centers of Ex- a variety of industries, such as service, construction, health
cellence Network in Oncology through Pacific Business care, hospitality, and education. In the United States, the

majority of union members (54%) in 2019 was men, with with union workers, nonunion employees who have access
one-third self-identifying as black, Hispanic, and Asian, to insurance often avoid or delay care because of the in-
among other underrepresented minorities.10 creasing premiums and deductibles that they are re-
Labor unions play a key role in influencing the broader sponsible for paying directly.
public health and political and economic systems that affect Some labor unions also further reduce health disparities
health. Globally, through interventions that promote healthy through implementation of direct health care partnerships
life choices, social and community networks, and improved and delivery services. For example, with the power to col-
socioeconomic, cultural, and environmental conditions, lectively bargain, labor unions can broker partnerships with
labor unions are integral to the public health infrastructure.22 employers to directly provide health care services, such as
As framed by the Social-Ecological Theoretical Framework primary care and chronic disease–management programs,
(Fig. 1), which describes how the broader socioeconomic, can negotiate coverage options and rates, can create narrow
cultural, and environmental conditions impact health, labor networks to ensure high-quality care delivery by specific
unions are often overlooked opportunities to address health providers, and can negotiate directly with pharmaceutical
and health inequity.25 Labor union organizations, for exam- companies to purchase drugs with discounts, allowing
ple, play a key role in advocating for workplace interventions members to access effective treatments at lower costs
and community- and policy-level interventions to promote overall.
health for populations at risk for disparities. One such ex-
To address the increase in health care costs and disparate
ample involves asthma, a condition known to dispropor-
access to care while maintaining value-based health care
tionately affect low-income inner-city residents and black
delivery for their members, labor unions have also begun to
populations. Labor union organizations have also led many
participate in the redesign of care delivery to address root
community- and policy-level interventions, such as green
causes of disparate access to care and high costs of care
industry efforts, clean air campaigns, and smoke-free acts, to
for their members. UNITE HERE HEALTH, which covers
reduce asthma-related conditions and secondhand expo-
230,000 mostly low-income hourly wage workers (111,000
sures to asthmagens for workers, consumers, and the
UNITE HERE union members and their dependents), is
broader community.23-25
a multiemployer Taft-Hartley Plan. The fund is paid for by
Access to health care through medical benefits and health trust funds established between employers and unions to
care insurance coverage is an important aspect of labor provide health care coverage for unionized workers and,
union membership that is crucial to overcoming health thus, is jointly governed by the UNITE HERE union and
disparities, in addition to addressing social determinants of employers. For decades UNITE HERE HEALTH has served
health, at least in the United States. These benefits can be UNITE HERE hospitality, food service, and gaming industry
directly administered through labor unions on behalf of their union workers. UNITE HERE HEALTH’s mission is to pro-
members. In 2019, for example, 95% of union workers in vide benefits that provide the highest quality and most af-
the United States had access to employer medical benefits fordable health care at better value and service than is
compared with 68% of nonunion workers. Union employers available in the current market. To promote high-value care
pay 77.4% more (per hour worked) toward their employees’ delivery, UNITE HERE HEALTH manages health care
health coverage, providing better benefits for a greater share
of workers than comparable nonunion employers.26 Addi-
tionally, 87% of union workers have access to paid sick days
compared with 69% of nonunion workers.27 These bene-
fits—access to health care and paid sick days—are espe-
cially important in reducing health disparities, especially for
the working population. For example, labor union members,
through medical benefits and health insurance coverage,
can access high-quality preventive cancer care services,
such as cancer screening, as well as cancer treatments and
high-quality cancer care delivery after diagnosis throughout
the cancer continuum. Paid sick days are specifically im-
portant because they encourage adherence to prevention
services and treatment regimens, both of which are im-
portant drivers of cancer outcomes. Most labor union health
insurance coverage provides more benefits at much lower
deductibles with lower premium sharing than nonunion
employer–sponsored health insurance coverage. Compared FIGURE 1. Social-Ecological Theoretical Framework

benefits, as well as participates in novel care-delivery pro- Institute on Minority Health and Health Disparities, to
grams that empower UNITE HERE union members with evaluate the efficacy of trained health promoters as related
better management of their health and health care. As an to (1) receipt of evidence-based cancer care services, in-
organization focused on value-based health insurance cov- cluding goals of care and advance directive documentation,
erage for its members, UNITE HERE HEALTH strategically (2) patient comprehension of prognosis, (3) education
aims to enhance its ability to delivery high-quality medical about symptom management, (4) equitable access to
care while controlling costs, empower and improve partici- palliative and hospice care, and (5) patient quality of life and
pants’ health before they enter the health care system, re- patient activation.28,29 In Atlantic City, the study also in-
duce health care costs, and improve quality and access. cludes innovative approaches to delivering cancer care for
Over the past decade, UNITE HERE HEALTH has partici- UNITE HERE HEALTH members and their dependents,
pated in a variety of other innovative approaches to improve such as steerage of patients to high-quality providers who
care delivery and health outcomes, especially among its are known to provide better cancer care at lower costs. To
members who are primarily underrepresented minority encourage members and their dependents to seek care with
hourly-wage workers. Innovative approaches include free- these providers, UNITE HERE HEALTH provides all cancer
standing health centers staffed by primary care clinicians care at no charge, as well as free transportation to and from
with expanded hours and longer visits, as well as urgent care the cancer center. In the health center in Atlantic City,
clinics that are open 24/7, such as those available to the members and their dependents also have access to a variety
Culinary Health Fund in Las Vegas, Nevada. UNITE HERE of ancillary services, such as free pharmacy benefits, im-
HEALTH also trains members to lead workshops for their aging, mental health services, physical therapy, and cancer
coworkers on chronic disease management and provides support groups. This study29 will determine whether these
members with lay “health promoters” (i.e., nonclinical, new processes improve cancer care delivery and quality of
trained community-based personnel). These lay health life for members and will also allow for the proactive col-
promoters encourage patients to participate in their health lection and use of important patient-centered data (e.g.,
and health care and support for union workers who have patient-reported outcomes, stage of disease, goals of care
high-cost health care issues as a result of complex condi- and advance directive documentation, and performance
tions. These proactive approaches to care delivery and status) that are not available to many payer organizations.
services are free of charge to UNITE HERE HEALTH In summary, employer organizations and organized labor
members, primarily because the return on investment has organizations play an important role in insuring the working
kept costs low by reducing the use of acute care services. population. Through provision of equitable access to health
When it comes to cancer care delivery, UNITE HERE insurance coverage and the development and imple-
HEALTH has implemented a variety of approaches to im- mentation of novel initiatives to improve the value in cancer
prove outcomes for its members. Through contract agree- care, employers and organized labor groups can address
ments with capitation payments and narrow networks with some of the main drivers of cancer health disparities.
a wide range of oncologists, UNITE HERE HEALTH strives to
ensure high-quality subspecialty care delivery. Using the CONCLUSIONS
health coaching model available to union members with Disparities across the cancer continuum, from prevention,
complex conditions, the fund in Atlantic City, New Jersey, diagnosis, treatment, and palliation to survivorship, persist
and Chicago, Illinois, has expanded the roles of the “health for many populations across the globe. Addressing social
promoters” to engage members and their dependents with determinants of health through multilevel approaches is
cancer to proactively document goals of care and advance crucial in reducing cancer disparities. Health systems and
directives. This novel model of cancer care delivery, which payer organizations, including private and public payers,
utilizes “health promoters” on the frontline to educate and employer groups, and labor unions, play an increasingly
empower patients with cancer, is a unique approach to important role through the provision of equitable access to
patients who are most at risk for disparities in receipt of high-quality cancer care and novel, patient-centered,
these services. In Chicago and Atlantic City, UNITE HERE stakeholder-informed approaches to achieve cancer health
HEALTH is participating in a study, funded by the National equity.
AFFILIATIONS
4
Independence Health Group, Philadelphia, PA
5
1
Stanford University School of Medicine, Stanford, CA Johns Hopkins University School of Medicine and Bloomberg School of
2
VA Palo Alto Health Care System, Palo Alto, CA Health, Baltimore, MD
3
Center for Health Policy/Primary Care and Outcomes Research, Stanford,
CA


Manali I. Patel, MD, MPH, MS, 1701 Page Mill Dr., Palo Alto, CA 94304; AND DATA AVAILABILITY STATEMENT
Twitter: @manalipatelmd; email: [email protected]. Disclosures provided by the authors and data availability statement (if
EDBK_279959.
REFERENCES
1. Polite BN, Adams-Campbell LL, Brawley OW, et al. Charting the future of cancer health disparities research: a position statement from the American Association
for Cancer Research, the American Cancer Society, the American Society of Clinical Oncology, and the National Cancer Institute. Cancer Res. 2017;
77:4548-4555.
2. Siegel RL, Jemal A, Wender RC, et al. An assessment of progress in cancer control. CA Cancer J Clin. 2018;68:329-339.
3. National Cancer Institute (NCI) Surveillance Epidemiology and End Results. https://seer.cancer.gov/. Accessed March 11, 2020.
4. Brawley OW. Some perspective on black-white cancer statistics. CA Cancer J Clin. 2002;52:322-325.
5. Breen N, Wesley MN, Merrill RM, et al. The relationship of socio-economic status and access to minimum expected therapy among female breast cancer patients
in the National Cancer Institute Black-White Cancer Survival Study. Ethn Dis. 1999;9:111-125.
6. CMS Innovation Center. https://innovation.cms.gov. Accessed March 11, 2020.
7. Health Insurance Exchange Quality Ratings System 101 | CMS. https://www.cms.gov/newsroom/fact-sheets/health-insurance-exchange-quality-ratings-system-
101. Accessed March 11, 2020.
8. Independence Blue Cross Foundation. http://www.ibxfoundation.org/. Accessed March 15, 2020.
9. Fecal immunochemical test (FIT). https://medlineplus.gov/ency/patientinstructions/000704.htm. Accessed March 11, 2020.
10. U.S. Bureau of Labor Statistics, U.S. Department of Labor. https://www.bls.gov/opub/ted/2019/union-membership-rate-10-point-5-percent-in-2018-down-from-
20-point-1-percent-in-1983.htm. Accessed March 11, 2020.
11. Claxton G, Rae M, Damico A, et al. Health Benefits In 2019: Premiums Inch Higher, Employers Respond To Federal Policy. Health Aff (Millwood). 2019;
38:1752-1761.
12. Patel MI, Lee RV, Smithline N, et al. The Employers Centers of Excellence in Cancer Care: a collaboration of employers and providers to improve cancer care.
J Clin Oncol. 2018;36:15s (suppl; abstr 141).
13. Northeast Business Group on Health. Delivering Value in Cancer Care: The Employer Perspective. https://online.flippingbook.com/view/654153/8/. Accessed
March 11, 2020.
14. Lichtenstein A. Building with Balint: a conference that connected generations. Int J Psychiatry Med. 2015;49:117-127.
15. Budd JW, Brey AM. Unions and family leave: early experience under The Family and Medical Leave Act. Labor Stud J. 2003;28:85-105.
16. Family-Friendly Workplaces. Do Unions Make a Difference. http://laborcenter.berkeley.edu/pdf/2009/familyfriendly09.pdf. Accessed March 11, 2020.
17. Kramer A. Unions as facilitators of employment rights: an analysis of individuals’ awareness of parental leave in the National Longitudinal Survey of Youth. Ind
Relat. 2008;47:651-658.
18. Robbins JM, Ford MT, Tetrick LE. Perceived unfairness and employee health: a meta-analytic integration. J Appl Psychol. 2012;97:235-272.
19. Kivimäki M, Virtanen M, Elovainio M, et al. Work stress in the etiology of coronary heart disease--a meta-analysis. Scand J Work Environ Health. 2006;32:431-442.
20. Berliner HS, Gibson G, Devine-Perez C. Health care workers’ unions and health insurance: the 1199 story. Int J Health Serv. 2001;31:279-289.
21. Siegrist J. Adverse health effects of high-effort/low-reward conditions. J Occup Health Psychol. 1996;1:27-41.
22. Sapp AL, Kawachi I, Sorensen G, et al. Does workplace social capital buffer the effects of job stress? A cross-sectional, multilevel analysis of cigarette smoking
among U.S. manufacturing workers. J Occup Environ Med. 2010;52:740-750.
23. Keane L, Pacek A, Radcliff B. Organized labor, democracy, and life satisfaction: a cross-national analysis. Labor Stud J. 2012;37:253-270.
24. Organization for Economic Cooperation and Development. http://www.oecd.org/. Accessed March 11, 2020.
25. Bronfenbrenner U. Toward an experimental ecology of human development. Am Psychol. 1977;32:513-531.
26. Abdelsattar ZM, Hendren S, Wong SL. The impact of health insurance on cancer care in disadvantaged communities. Cancer. 2017;123:1219-1227.
27. Harrington E, McInturff B. Key Findings – National Surveys of Cancer Patients, Survivors, and Caregivers, American Cancer Society Cancer Action Network.
https://www.acscan.org/sites/default/files/ACS%20CAN%20Paid%20Leave%20Surveys%20Key%20Findings%20Press%20Memo%20FINAL.pdf. Accessed
March 11, 2020.
28. Patel M, Andrea N, Jay B, et al. A community-partnered, evidence-based approach to improving cancer care delivery for low-income and minority patients with
cancer. J Community Health. 2019;44:912-920.
29. NCT03699748. Lay Health Worker Engage, Educate, and Encourage Patients to Share (LEAPS). https://clinicaltrials.gov/ct2/show/NCT03699748. Accessed
March 11, 2020.

Direct-to-Consumer Advertising for Cancer

Centers and Institutes: Ethical Dilemmas and
Practical Implications
Fay J. Hlubocky, PhD, MA1; Daniel F. McFarland, DO2; Patricia A. Spears, BS3; Laura Smith, JD4; Bonnie Patten, JD4;
Jeffery Peppercorn, MD, MPH5; and Randall Holcombe, MD, MBA6
overview
In the United States, many cancer centers advertise their clinical services directly to the public. Although
there are potential public benefits from such advertising, including increased patient awareness of treatment
options and improved access to care and clinical trials, there is also potential for harm through mis-
information, provision of false hope, inappropriate use of health care resources, and disruption in doctor–
patient relationships. Although patient education through advertising is appropriate, misleading patients in
the name of gaining market share, boosting profits, or even boosting trial accrual is not. It is critical that
rigorous ethical guidelines are adopted and that oversight is introduced to ensure that cancer center marketing
supports good patient care and public health interests. Patients with cancer have been identified as an
especially vulnerable population because of fears and anxiety related to their diagnosis and the very real need
to identify optimal sources of care. Cancer organizations have a fiduciary duty and a moral and legal obligation
to provide truthful information to avoid deceptive, inaccurate claims associated with treatment success. In
this article, actionable recommendations are provided for both the oncologist and the cancer center’s
marketing team to promote ethical marketing of services to patients with cancer. This tailored guidance for the
oncology community includes explicit communication on (1) ensuring fair and balanced promotion of cancer
services, (2) avoiding exaggeration of claims in the context of reputational marketing, (3) providing data and
statistics to support direct and implied assertions of treatment success, and (4) defining eligible patient
groups in the context of marketing for research. These recommendations for cancer centers are designed to
promote ethical quality marketing information to patients with cancer.
INTRODUCTION research involving clinical trials.6-8 However, for many,

Cancer treatment is big business in the United States. the advertisement as presented may provide confu-
In general, patients with cancer incur four times the sion, bias, or incomplete, inaccurate information that
mean expenditures per person than patients without has the potential to generate false hope and elevated
cancer.1 In 2001, a total of $57 billion was spent on expectations for therapeutic benefit. From the in-
cancer care, but this escalated to $88.3 billion in 2011 ception of DTCA in the health care industry during
and is projected to be as high as $173 billion in 2020 the 1980s, there has been significant ethical debate
per the Agency for Healthcare Research and Quality.1 and concern surrounding this type of advertising.9
To gain market share, significant financial invest- Given the degree to which cancer care has evolved
ments are expended by pharmaceutical companies, with new treatments becoming a major focus of
hospitals, and cancer centers engaged in direct-to- health care investment and spending, it is important
Author affiliations
and support consumer advertising (DTCA) for cancer services and to review the current empirical evidence and ethi-
information (if products.2 A proportion of this specific type of mar- cal, legal, and policy implications of cancer center
applicable) appear
keting targets patients directly with patient-centered advertising.
at the end of this
article. testimonials in the form of: television or radio advertise- By definition, patients with cancer have been identified
Accepted on April ments, print advertisements (e.g., magazine, newspaper, as a vulnerable population.10-14 That is, patients with
12, 2020 and brochures, billboards, transportation), online adver- cancer face a serious diagnosis, often with mortality
published at tisements, and social media platforms (e.g., Facebook, risk, and a potentially confusing array of treatment
ascopubs.org on May
Twitter, Instagram).3-5 The purpose of DTCA is two-fold: options. Traditionally, a balanced patient–physician
7, 2020: DOI https://
doi.org/10.1200/ (1) to attract the patient and family to cancer organizations relationship leads to shared decision-making founded
EDBK_279963 and (2) to promote clinical services and/or novel on the physician’s expertise, a fiduciary duty of care,

Hlubocky et al
the prices and services they offered. As a result, in 1975,

the FTC sued the AMA on the grounds that its Code of
Ethics was attempting to stifle competition and support
• In the United States, cancer centers commonly a monopoly on health care.20 By the time the case reached
advertise their services to the public.
the U.S. Supreme Court in the early 1980s, the AMA
• Several benefits from such advertising exist that abandoned any claim restricting physicians or medical
include patient education of cancer treatment practices from marketing their services. Subsequent cases
options and improved access to care and involving advertising practices in multiple professions were
clinical trials.
decided on in a similar fashion, indicating that professional
• Overwhelmingly, there is potential for significant services should be treated similarly to advertisements for
harm through misinformation, inappropriate other types of goods and services.16 This indicates that ad-
use of health care resources, and disruption in vertising conducted by for-profit health care institutions are
doctor–patient relationships.
subject to FTC oversight under the same “reasonable stan-
• Patient testimonials using emotional appeals dards” for fairness and accuracy applied to advertisements
may intensify a patient’s hope and fear. for common consumer goods and services. However, the
• Oncologists who are aware of and identify de- FTC does not have jurisdiction over advertisements by
ceptive marketing issues may take active steps to nonprofit health care institutions, which indicates nonprofit
help protect patients from making critical health marketing is not generally subject to federal oversight. Not
care decisions based on misleading marketing surprisingly, DTCA by health care organizations has steadily
information through effective communication. increased over the last 2 decades.3,21 Although the standard
• Cancer organizations have a fiduciary duty and concepts of fairness and honesty in DTCA are applicable for
a moral and legal obligation to provide per- many subspecialties in medicine, given the unique nature of
sonalized, quality, balanced, ethical information cancer treatment, potential for toxicities, vulnerable patient
rather than hype. population, and costs of care, the marketing of cancer center
services and treatment equally raises ethical concerns.
Although there are both proponents and opponents of
and the patient’s values and preferences.15 However,
DTCA, it inevitably raises questions surrounding the in-
shared decision-making is at risk in oncology because of
terplay of society’s values with medical values, which do not
patient vulnerability and an informational, experiential im-
always align. For example, most Americans believe in a free
balance between the patient and oncologist. Treatment
market society but simultaneously desire protection from
decisions are high-stakes endeavors that impact patient
paying the high costs of health care.22,23 As a result, some
emotions, as well as clinicians, with potential for long-
Americans worry more about the where and how to get
lasting implications. Simultaneously, over the past 2 de-
treatment than they do about their diagnosis.24 A cancer
cades, new oncology drugs and treatments, including
diagnosis, of course, conjures up fears about both. Not only
targeted cancer therapies and immunotherapy, have
do Americans hear about the unfortunate outcomes asso-
proliferated to provide new life-sustaining treatment. As
ciated with cancer, but they also learn about the financial
a result, a new phenomenon centering on the business of
toxicity that patients and families may endure as a result of
cancer treatment with its bottom-line implications vies to
cancer treatments.25 In other words, a cancer diagnosis and
maximize its market share of an already overwhelmed
the prescribed treatment are fraught with emotions that
population of patients with cancer.2
place patients and their families in compromised and ex-
THE EXPLOSION OF CANCER CENTER ADVERTISING: A posed situations that lead to information-seeking behav-
HISTORICAL CONTEXT BEHIND MARKETING TO PATIENTS iors.26 This results in a unique consumer baseline that
In a previous era, the American Medical Association (AMA) differs significantly from typical consumers in a free market
publicly discouraged advertising of medical services, society.6
claiming that the practice ran contrary to the ethos and The DTCA in health care debate also parallels a larger
integrity of physicians.15,16 Specifically, the AMA argued societal issue of communication about serious health is-
“that the best way to interdict false and deceptive ad- sues, which has transitioned over time from a paternalistic
vertising and overreaching by physicians is to proscribe communication model to either providing information, or at
practically the full spectrum of advertising and solicitation its best, shared decision-making models.27 In fact, appro-
activities.”17-19 However, the Federal Trade Commission priate handling of greater health care knowledge in the
(FTC) believed that banning advertising and solicitation public is only appropriately handled with shared decision-
unreasonably restrained trade and prevented doctors and making to gain informed consent.28 This is a monumen-
medical organizations from disseminating information on tal task for physicians, however, because they have

Ethics of Cancer Center Advertising
increasingly limited time with patients (caused in part by (score 65 = readable for average person). Proportion, type
documentation requirements and financial pressures). size, and placement of benefits and risks were assessed
Nevertheless, the vast amount of information patient con- including the purpose/content of advertising appeals. Forty-
sumers obtain through DTCA, among other sources, con- nine of 284 advertisements identified were unique. DTCA
tinuously raises questions about whether an alternative was seldom presented in the popular magazines, and it
optimal model exists. targeted female readership. Equivalent amounts of text
centered on benefits and risks/adverse effects, and all text
DTCA in oncology is especially relevant and impactful be-
was challenging to read. The mean Flesch reading ease
cause a plethora of expensive drugs and treatments may be
score for benefit text was 39.7 and for risk/adverse effect text
available to patients who are in a vulnerable, life-threatening
was 38.2, which is a difference of 1.5 (95% CI, 4.0 to 7.0).
position with a serious illness and undergoing complex
The largest font size for benefits was 4.6 mm on average and
medical and financial decisions with vast amounts of in-
for risks/adverse effects was 2.4 mm, which is a difference
formation to process and conflicting sources of information.29,30
of 2.2 mm (95% CI, 1.3–3.1). Appeals for medication ef-
As such, the cancer organization has a fiduciary duty, as
ficacy were frequent (95%) and often for specific clinical
well as a moral and legal obligation, to provide accurate
trial data (61%). Oncology-print DTCA was prevalent in
information to avoid deceptive, inaccurate, biased, and
cancer-related, patient-tailored magazines yet were un-
atypical claims associated with treatment success.4,6,8 This
common in the popular press. The information presented
article presents current research evidence on cancer center
was found challenging to read and comprehend, raising
DTCA, including DTCA advantages and disadvantages in
questions regarding the potential educational value and
the cancer setting. Ethical dilemmas associated with cancer
suitability of DTCA for promotion of oncology treatment.31
center DTCA will be addressed, including actions that ex-
pose ethical concerns arising from inaccurate cancer- Similarly, Vater et al4 described the informational and
related information and actions that adhere to clinical emotional content of cancer center DTCA. Content analysis
medical ethics. Additional legal and policy implications are was conducted using leading U.S. consumer magazines
discussed. Recommendations for communication about (269 items) and television networks (44 items) in 2012.
DTCA are provided for the oncologist, and future directions Types of clinical services promoted; clinical service in-
are considered. Although the ethical implications behind formation (e.g., risks, benefits); insurance availability; use of
monitoring medical practice to ensure patient safety, reduce emotional advertising appeals; and patient testimonials
redundancies, improve quality assurance, and reduce were assessed. Results revealed that 102 cancer centers
medical errors appears fairly straightforward, issues and placed 409 clinical advertisements in top media markets.
questions with regard to addressing cancer center DTCA DTCA promoted treatments (88%) more frequently than
remain. screening (18%) or supportive services (13%). Benefits of
treatment were depicted more than risks (27% vs. 2%);
EMPIRICAL EVIDENCE TO DATE: CANCER CENTER DTCA however, they were minimally quantified (2%). Only a lim-
CONTENT AND IMPACT ON PATIENT UNDERSTANDING ited number of DTCA described treatment costs (5%), and
AND EMOTIONS none described insurance. Emotional appeals were fre-
Prior research conducted over the past decade aimed to quent (85%), prompting hope for survival (61%), depicting
investigate precisely what information cancer centers cancer treatment as a fight (41%), and inducing distress
market to the public and what patients’ experiences and (30%). Approximately 50% of DTCA included patient tes-
emotions are associated with DTCA.4,31,32 For example, in timonials, traditionally centered on survival, rarely depicted
2005, one of the first studies completed on medical center disclaimers, and never described the results that a typical
advertising assessed facilities named to the U.S. News and patient may expect. The researchers concluded that DTCA
World’s Report’s list of the United States’ best hospitals.31 by cancer centers frequently promoted cancer treatment
Results revealed that 16 of 17 hospitals reported that the with emotional appeals that prompted hope and distress,
purpose of their advertising was to attract patients and that whereas rarely providing salient treatment information about
these advertisements highlighted cancer services, used risks, benefits, or costs.4
emotional marketing appeals, promoted unproven in- An additional investigation examined the awareness of
terventions, and failed to present potential risk of treatment cancer-related DTCA of patients with breast and hemato-
or validate positive claims. Additional research involved logic cancers undergoing active treatment at a well-known
a formal content analysis of cancer-related DTCA materials cancer center.30 Patients completed a mailed questionnaire
focused on the presentation of on benefit and risk in- assessing DTCA: awareness, perceptions, impact on pa-
formation. All oncology DTCA appearing in three patient- tient, and clinician behaviors. The results revealed that 348
focused cancer magazines and selected popular magazines patients completed questionnaires (response rate, 75%).
using the Flesch reading ease score for text readability Overall, 86.2% of respondents reported an awareness of

Hlubocky et al
DTCA, with 78% from television. Awareness did not differ patient hope. It is a mechanism for patients to engage in
with clinical or sociodemographic factors except when active communication with their oncologists regarding new
patients were more likely to be aware of DTCA for specific cancer treatment or clinical trials. Finally, DTCA encour-
cancer-type products (p , .0001). Most patients aware of ages beneficial competition in the oncology health care
DTCA reported it made them “aware of treatments they did marketplace.21,34 Equally important to our understanding of
not know about” (62%), provided information in “a bal- any advantages of cancer center DTCA, is that numerous
anced manner” (65%), and assisted in initiating “better disadvantages have been recognized and discussed.35 For
discussions” with clinicians (57%). DTCA perceptions were many, DTCA information is confusing, biased, inaccurate,
significantly more favorable among those with less than misleading, or incomplete. DTCA may motivate patients to
a college education (p . .05). Only 11% reported that DTCA request treatment that is not applicable to their cancer
impacted their confidence with clinician recommendations. diagnosis or where therapeutic benefit has yet to be proven
For patients with awareness, 17% reported discussions about scientifically. In fact, recent research of hospital adver-
an advertised medication with their clinician, although less tising for cancer services has not been correlated with
than one-fifth reported receiving a prescription for the ad- patient outcomes, which therefore indicates information
vertised medication. The study investigators concluded that shared through advertising might mislead patients and
patients were highly aware of DTCA and perceived it to be generate inaccurate expectations of treatment benefit.
available and useful. Therefore, for some patients, DTCA Patients may also reject standard treatment offered
decreased their confidence in their clinician’s recommen- to them that is appropriate for their cancer because of
dation, yet DTCA prompted a modest amount of patient– this marketing. DTCA also poses a challenge to the
clinician discussion. oncologist–patient relationship because of potential conflict
over treatment and inevitably threatens the informed con-
Finally, a recent 2019 study aimed to evaluate whether DTCA
sent process for care or trial participation because of patient
spending for a hospital’s cancer services was associated with
misunderstanding.
long-term survival outcomes of patients with cancer treated in
those institutions.21 Investigators examined cancer hospital Finally, cost is rarely presented in DTCA, posing both so-
DTCA spending for services in 2014 through U.S. media cietal and personal costs to patients and others who are
outlets (e.g., television, magazines, internet), using data from misled. DTCA leads to additional hidden costs to patients
a media monitoring agency from 50 leading cancer hospitals and families who travel for treatment and take more time off
and systems. Results revealed there was considerable varia- work, for example. In addition, DTCA leads to loss of revenue
tion in advertising spending and survival outcomes. Minimal for community cancer centers or smaller practices that have
evidence existed that cancer centers that would most likely no or little budget for such advertising and are at a huge
impact patients through DTCA were those with optimal patient disadvantage compared with larger tertiary cancer centers
outcomes. Variation was found in both DTCA advertising and health care systems who often advertise for services
spending and survival outcomes.21 In summary, evidence to that are not unique to them but are also conveniently
date has extensively examined cancer center DTCA content, available at a nearby community cancer center. For ex-
finding them challenging to read; impactful on awareness, ample, immunotherapy agents are approved across several
expectations, and emotions; and an impediment to the tumor types and are delivered at both larger and smaller
oncologist–patient relationship and ultimately to an informed cancer centers. However, when larger centers boast “We
discussion regarding appropriate, optimal quality cancer care. harness your immune system to attack cancer,” the im-
plication is that other treatment centers do not in fact use
ADVANTAGES AND DISADVANTAGES OF CANCER these medications and entices patients and families to
CENTER DTCA forego a local and potentially less expensive option. In this
As described above, several advantages and disadvan- way, deceptive advertising can have long-reaching effects
tages of cancer center DTCA have been identified by re- for patients, families, and medical practices that are at
searchers, ethicists, and other stakeholders in oncology a financial disadvantage.
(Table 1).21,33-35 Proponents supporting the use of cancer In summary, both proponents and opponents provide
center DTCA offer several concrete advantages of this several reasons for and arguments against DTCA tailored to
marketing to patients.33,34 DTCA offers an opportunity to patient consumers.
provide tailored education to patients about the individual
cancer institution’s services, treatment, and research. It also ETHICAL DILEMMAS ASSOCIATED WITH DTCA FOR
raises awareness of the available cancer services, treatments, CANCER CENTERS
or new clinical research trial potential as a viable option for To critically examine the impact of the cancer center DTCA
care. Moreover, proponents argue that DTCA reduces stigma on patient decision-making and on the oncologist–patient
associated with cancer by offering care that provides the relationship, we will illuminate ethical dilemmas that arise

TABLE 1. Cancer Center DTCA Advantages and Disadvantages

Advantages of Cancer Marketing for the Patient With Cancer
Consumer Disadvantages of Cancer Marketing
Provide tailored education to patients about the individual cancer Information is confusing, biased, or incomplete
institution’s services, treatment, and research
Motivate patients to request treatment that is not applicable to their cancer
diagnosis, reject contemporary standard treatment of their cancer, or therapeutic
benefit has yet to be proven scientifically
Challenge to the oncologist–patient relationship because of conflict over treatment
Threatens the informed consent process for care or trial participation
Raise awareness of the cancer treatment or clinical research trial as Testimonials may mislead consumers who may misinterpret, and patients may
potential treatment options assume that excellent outcomes achieved by the patient giving the testimonial
represents the average or usual outcome for all patients
Reduces stigma associated with cancer by offering care that gives Marketing, especially by medical centers, is often not data driven and may
the patient hope intentionally avoid mention of outcomes
Mechanism for patients to engage in communication with their
oncologists regarding new cancer treatment or clinical trials
Encourages competition in the health care, oncology marketplace Cost is rarely mentioned in advertisements; increases in both societal costs for new
technologies and individual costs, if patients’ insurance does not cover
nonstandard treatment approaches.
because of this specific form of marketing. This analysis intentionally (or perhaps unintentionally), which is de-
will describe actions that adhere to clinical ethics and ceiving. It is also disingenuous to claim that DTCA simply
those actions that reveal ethical concerns and violations provides information because the general goal of adver-
arising from an inaccurate presentation of cancer-related tising is to persuade consumers to choose a specific good
information.14. or service. The goal of DTCA is to primarily influence rather
Impact of DTCA Influence on Autonomy than to inform, whereas the only goal of public service
announcements is to inform.37,38 Freedom of speech is
It is the right of the individual patient to control one’s own protected under autonomy. The public, pharmaceutical
medical care that guides decision-making, including companies, and health care providers alike would retreat
gaining access to information and cancer services. It is from an attempt to block marketing communications.
grounded on the concept of respect for the individual and However, deceptive commercial speech, such as false
that the decisions or values of others, including what others advertising, is not a protected form of speech. Neverthe-
may feel is in one’s own best interest, should not undermine less, the issue of what constitutes a deceptive message is
our own right to make decisions regarding what treatment to open for interpretation.
pursue, what sources of information one prioritizes, and
where or when one receives care. Violation of Justice
Autonomy requires informational access for the public, DTCA provides a mechanism for patients to become in-
which DTCA provides. The caveat is that the message must formed about what may constitute the best and adequate
be accurate, truthful, or coercive to fulfill autonomy. Is the treatment. Patients deserve and have a right to have ac-
patient aware of the fact they are intentionally influenced cess to treatments. This is justice. Justice is fairness.14 The
by DTCA and that the mechanism of that influence is this medical community calls these minimum treatment re-
form of marketing?35 The autonomy of individual organi- quirements “standard of care,” which all patients should
zations or groups of organizations (such as a cancer receive, irrespective of ability to pay or other patient
center) enables them to freely educate the public to ad- characteristics. As well, justice evokes the privilege to
vertise their services, as long as other ethical principles are obtain not what the medical community deems the right
not violated. Proponents of DTCA believe it provides society treatment but what the patient has determined as pre-
access to information that they would not otherwise have ferred treatment. Here, patient preferences may be
about a drug or treatment.33,34 However, this educational- respected by receipt of standard medical care or opting out
based argument is not without concern. DTCA often of treatment. DTCA oversimplifies this issue because there is
provides a one-sided perspective, that although not alto- the enormous hurdle of information dissymmetry.3,6 The
gether false, may create a misleading message signaled physician stance on providing patient care (e.g., pro-
nonverbally.36 DTCA may also commit errors of omission fessionalism) should allow for justice to inform individual

Hlubocky et al
treatment decisions. It is possible that DTCA leads to relationship, but to a patient’s relationship with the medical
a discussion about treatment that may align with the pa- community. Although many patients are able to decipher
tients’ goals more fairly. One must also consider that this the appropriateness of different sources of information,
scenario entails a significant amount of working together DTCA may disproportionately affect patients who would
and communication. most benefit from a trusted relationship with their cancer
Physician–Patient Conflicts and Nonmaleficence clinicians, namely those who are less health literate or have
less education and certain minority groups who are inclined
The term pimum non nocere (do no harm) is the guiding toward mistrust of the medical establishment.39
principle to choose medical treatments that inflict the least
harm to reach a beneficial outcome.14 However, the goal of Beneficence
DTCA is to influence rather than to inform, and whenever an Beyond primum non nocere, one should act with the benefit
entity is exhibiting influence in a conflictual way (they stand of the other person in mind.14 The physician–patient re-
to benefit from the influence), one could argue that the lationship exemplifies this to an extent that would not be
principle of nonmaleficence will be compromised based on practical in everyday relationships because of the amount of
conflicting priorities. For example, private entities such as self-sacrifice involved. Although this is a generally agreed on
pharmaceutical companies do not have the same direct principle of ethics, the amount of personal self-sacrifice to
fiduciary relationship with individual patients as do clini- sufficiently serve this principle can be debated. Larger
cians, and thus, they rely on the clinician to ensure that entities including health care institutions and pharmaceu-
ultimately patients are not harmed.3,39 This implicit re- tical companies should operate by the same principles, and
lationship is arguably acceptable to all agreeing parties with this is exemplified by mission statements and planning
the exception that DTCA’s influence may be subtly proposals creating patient-centric treatments and environ-
impacting some decisions. Patients can benefit from having ments. A challenge arises in evaluating the effects of
more information; however, that is not the primary purpose marketing on individual patients and what is a benefit with
of these advertisements, which imposes harm by making potential harms. For example, the provision of information,
patients feel conflicted about their treatments and guilty albeit for the purpose of influence, may be viewed a benefit
about not doing enough and expecting a miracle. Potential to the uninformed patient.3,6 However, this information may
conflict arises between the patient’s interest and the outside lead to cognitive dissonance for patients who feel contra-
interest of society. The role of the oncologist may be in dictory information is being communication or who are
conflict in balancing these interests and meeting their overwhelmed with treatment options without a clear sense of
primary obligation to the patient while considering the who to trust. Although these large corporations should not
limitations of this duty and their competing obligations to be expected to manage individual situations, one could
other parties. This dilemma surrounds the dual role of the argue that DTCA should preemptively minimize the potential
oncologist who may be asked to honor their fiduciary re- to misinterpret, distort indications, or provide false hope.
sponsibility to the patient and simultaneously consider the CANCER CENTER LEGAL OBLIGATIONS: AVOIDING
interests of society or the organization.3,6,39 Oncology pa- DECEPTIVE, PERSUASIVE ADVERTISING TO INFORM
tients may become increasingly desperate (i.e., vulnerable) PATIENT EXPECTATIONS AND DECISION-MAKING
for further treatments, and although drug X may not be
individually indicated or the patients may not qualify for the A significant concern involving DTCA to patients with cancer
treatment under usual circumstances, the advertisement is the accuracy of the information provided to patients. The
has presented information and a picture of substantial patient’s hope for benefit and survival can be fueled by
tolerability. As a patient nears end-of-life, the oncologist may a television advertisement depicting healthy-appearing
feel a willingness to meet the patient’s desires, thereby patients with cancer and caregivers looking out at the
placing the patient at greater risk than otherwise without the ocean.6-8,30,31,35 Such advertising may provide false opti-
additional hope DTCA had created. Here, DTCA may be- mism and also impact the understanding of the disease,
come a factor that pushes the bounds of nonmaleficence prognosis, and treatment. This may lead patients to make
further, creating a situation where the physician is re- unrealistic decisions based on inaccurate expectations.
sponsible for correcting the misinformation. This may be Patient consumers desire truthful information. False, mis-
particularly damaging to the physician–patient relationship, leading, biased, deceptive, or unreliable information un-
especially with patients with low health care literacy who dermines informed decision-making.40
may be unable to understand source differences between Marketing uses behavioral science and psychology meth-
their clinicians and DTCA; this ultimately sets up a situation odology to present information designed to modify the in-
of mistrust. It may not always be possible to remediate some dividual patient and group health care decisions and
of the hope and promises so often portrayed in DTCA, and behaviors.35-38,41 DTCA itself has been identified as a pre-
this could be harmful, not only to the doctor–patient dictor of attitudinal and behavioral change. Such behaviors

are specific strategies that range on a continuum between limitations. As defined, this type of misleading and de-
reason and coercion. Examples of these patient-directed ceptive marketing is against the law.
behavioral strategies include “framing and focusing effects”
centering on benefits and risks or “vivid images.” As a result,
POLICY IMPLICATIONS: REASON FOR OVERSIGHT?
deceptive marketing may become so persuasive and
harmful to consumers that it influences them to seek in- The issue of commercial speech is a complex endeavor.41 In
appropriate options for care.42 fact, one may question if DTCA should be allowed, and if so,
what limits should they be subject to given the complexity
Deception is arguably the most pervasive ethical issue in
of the information directly communicated to the patient
consumer marketing and is a direct consequence of
consumer.41 Policy statements on deceptive and unfair
a marketer’s action.37 It can be defined as attainment of
marketing practices were founded solely on economic
false impressions or beliefs that lead to unfilled expec-
considerations, ignoring the wider consequences with
tations. In the case of advertisement, if it leaves the
regard to the individual consumer, the patient with cancer.38
consumer with an impression or belief different from
Behaviors, expectations, and decision-making were not
what would normally be expected if the consumer had
taken into consideration. David Vladeck40 explored the role
reasonable knowledge, leaving belief is factually untrue,
of DTCA at length in a formal analysis, contending that
or potentially misleading, then deception exists.43 De-
it does indeed make a difference that the evaluation of
ception does not only occur through advertising but may
evaluating the risks and benefits of prescription medication
be part of each element of the marketer’s communi-
is a task fraught with complexity and generally beyond the
cation (e.g., sales advice, messaging, packaging, public
competence of consumers. It did make a difference that
relations). 37,38
empirical evidence confirmed DTC advertisements are
Specifically, for patients, deception has the potential to highly influential in consumer decision-making yet fre-
harm their health and welfare, their financial resources, quently fail to adequately disclose the risks of the drug. It
privacy, self-esteem, and trust in society. Deception un- mattered that the U.S. Food and Drug Administration lacks
dermines fair competitions and demeans the profession of the statutory tools and resources to police the DTCA mar-
marketing overall.37 Here, testimonials are an incredibly ketplace effectively.41 He concluded that it does matter,
popular and powerful tool used in health care marketing recommending guidelines to advertisers who use en-
materials, which can be deceptive inducing patients to dorsements and testimonials that will now require adver-
make decisions (Table 2). The deceptions involved can tisers to “clearly and conspicuously disclose the generally
include the promotion of anecdotal and atypical results expected results consumers can expect in the depicted
without disclosure of more typical outcomes and promotion circumstances.”44 However, for some areas in medical
of novel treatments without disclosing those therapies’ marketing, this guidance has been loosely followed, if at all.
TABLE 2. Example of Deceptive Cancer Center Marketing: Patient Testimonials

DTCA Advertisement With Patient Testimonial Inappropriate, Deceptive Marketing Elements Used
An online article has an embedded short video advertisement for a well- 1. Advances the narrative, either explicitly or implicitly, that treatment at
known Cancer Center X. This video depicts a healthy, elderly male a specific cancer center will provide patients with a therapeutic advantage,
trimming trees in his yard during a bright sunny day. He looks up to allowing them to beat the odds and live beyond 5 years.
address you/the consumer to say, “I’m a 7-year lung cancer survivor.
2. Promotes clinical trials (i.e., research endeavors with no guarantee of
I knew I had a chance because I went to Cancer Center X for their
therapeutic benefit) and novel treatments, such as immunotherapy and/or
innovative high-tech clinical trials. Look at me now. It saved my life.
experimental procedures, without clearly and conspicuously disclosing
Don’t give up hope. I didn’t.” The male then proceeds to return to his
their limitations, risks, and relative rarity.
yard work.
3. Use of persuasive terms: “innovative”; “chance”; “hope”
A billboard advertisement of a healthy, middle-aged woman standing on 1. Touting “atypical” results nor is it enough to hyperlink to survival statistics,
a tennis court with a racket in hand. The caption statement embedded as such weak disclaimers are ineffective. In patient testimonials without
within the ad states: “I am a stage IV pancreatic cancer survivor and I beat clearly and conspicuously disclosing what the generally expected results
it with treatment at Cancer Center X. I now have a new life full of activities for a patient in a similarly situation would be violates FTC law.
I enjoy.” Another caption below includes another statement: “These
2. If a cancer center does not have substantiation that a promoted patient
results are not typical for the specific cancer represented here. Do not
experience is representative of what consumers will generally achieve, the
expect similar results.”
testimonial must clearly and conspicuously disclose the generally
expected results in the depicted circumstances.

Hlubocky et al
SIDEBAR 1. RECOMMENDATIONS FOR ONCOLOGIST COMMUNICATION REGARDING CANCER CENTER TREATMENT AND
RESEARCH INFORMATION
1. Discuss balanced patient-centered information during the initial meeting with the patient and family to address any
confusion or elevated expectations for therapeutic benefit from a specific advertised treatment.
2. Effective, objective physician disclosure and patient communication about benefits, risks, and alternatives to treatment of
those considering a treatment or a new clinical trial are the hallmark elements of the informed consent process.
3. Provide personalized, quality, balanced, and ethical patient-centered information rather than hype.
4. Oncologists who are aware of and identify deceptive marketing issues may take active steps to help protect patients from
making critical health care decisions based on incomplete and misleading marketing information.
5. When discussing research and disclosing relevant clinical trial information, informing patients that these trials may not be
available to them as not all patients may qualify for the advertised trial.
6. Disclose relevant clinical trial information and inform patients that these trials may not be available to them as not all
patients may qualify for the advertised trial is critical; this helps to enhance patient understanding and facilitates trust.
RECOMMENDATIONS FOR THE ONCOLOGIST AND CANCER care and services. Recommendations include the fol-
CENTER TO ENSURE ETHICAL MARKETING OF SERVICES lowing described below.
TO PATIENTS
Because of the complexities of cancer center DTCA de- Oncologist
scribed here, it is vital that ethical guidelines are adopted Oncologists who are aware of and identify deceptive mar-
and that appropriate oversight is implemented to ensure keting issues may take active steps to help protect patients
compliance with the law. As cancer centers increasingly from making critical health care decisions based on in-
use DTCA methods, including online social media, it is complete and misleading marketing information (Sidebar
critical to ensure ethical practices include compliance with 1). Effective, objective physician disclosure and patient
Health Insurance Portability and Accountability Act privacy communication about benefits, risks, and alternatives to
regulations that protect patient data and that patient an- treatment of those considering a treatment or a new clinical
onymity is met. Confirmation that communication about trial are the hallmark elements of the informed consent
services and treatments including clinical trials is valid, process. Therefore, physicians should disclose balanced
accurate, and truthful is required. The target populations patient-centered information during the initial meeting with
for cancer center DTCA include oncology practitioners and the patient and family to address confusion, false optimism,
a vulnerable population of patients with cancer. Primary or elevated expectations of therapeutic benefit to enhance
stakeholders for cancer marketing include both pharma- understanding.39 This discussion may involve revisiting the
ceutical companies and large cancer centers vying for patient’s symptoms, prognosis, and goals of care. When
increased market share. Patients, clinicians, and stake- discussing research and disclosing relevant clinical trial
holders all have an opportunity to improve communica- information, it is critical to inform patients that these trials
tion about cancer and the marketing to patients and may not be available to them because not all patients may
families to provide quality ethical information regarding qualify for the advertised trial. This helps to enhance patient
SIDEBAR 2. RECOMMENDATIONS FOR MARKETING AND COMMUNICATING CANCER CENTER AND INSTITUTE TREATMENT AND
RESEARCH INFORMATION
1. When using patient testimonials in marketing, do not depict specific outcomes that are atypical that may mislead patients
about the actual disease prognosis.
2. Avoid deceptive, unethical information designed to induce patients to travel for a clinical trial designed to improve survival
only to have the patient realize there is no viable trial option available for them.
3. When promoting research and disclosing relevant clinical trial information, inform patients that these trials may not be
available. Inform the consumer that not all patients may qualify for the advertised trial.
4. Reputational marketing is acceptable, but exaggerated claims should be avoided.
5. Provide data and statistics about new technologies and approaches. Be transparent if these data do not exist.
6. Achieve a fair balance and ensure satisfaction of the principles of medical ethics.

understanding and facilitate trust. Patients expect trans- broadly performed by the main accrediting agency for
parency, the reasoning for the recommendations made, hospitals: the Joint Commission. If newer guidelines were
promotion of the patient’s best interest, and ultimately, introduced, it is likely that marketing practices could be
autonomy.35 modified to satisfy Joint Commission guidelines because
Cancer Centers accreditation is critical to individual hospitals. Cancer care is
big business, and there are many opportunities for unethical
Because health care expenditures attributable to cancer are marketing practices by various segments of the industry.2 It
large, it is reasonable to expect that pharmaceutical compa- is essential that ethical guidelines are adopted and that
nies and cancer centers will advertise their services to increase appropriate oversight is introduced to ensure compliance
market share (Sidebar 2). Given the vulnerability of the pop- and provide ethical quality information on cancer care.
ulation targeted, however, it is essential that marketing ac-
tivities do not take advantage of the distress that is felt by SUMMARY
a patient with cancer.2 Recommendations for marketing Ethical challenges arise regarding DTCA of cancer services
cancer include the following: (1) a fair balance should be and treatments in the United States. However, DTCA has
achieved and satisfaction of principles of medical ethics historically increased demand.41 The ethical implications of
should be ensured, (2) reputational marketing should be DTCA represent the two value systems of free market
provided without exaggerated claims, (3) data and statistics commercialism and health care as a right for all colluding
about new technologies and approaches should be provided, with one another. The activities of corporate entities ad-
with transparency if such data do not exist, (4) when promoting
vertising to gain and sustain market share reveals the un-
research and disclosing relevant information regarding clinical
derbelly of marketplace health care, where, like all other
trials, patients should be informed that these trials may not be
business, survival depends on influencing patient choice.
available and that not all patients may qualify; this is a legiti-
mate focus of advertising and supports the best interests of Although clinicians are supposed to approach patients
patients, (5) it would be deceptive and unethical to persuade impartially, DTCA exposes a challenging health care delivery
and induce patients to travel for a clinical trial believing they system with competing priorities that adversely impacts the
must, only to have them realize there is no viable trial option physician–patient relationship. Formal investigation of the
available for them, and (6) when using patient testimonials in implications of these not-so-subtle marketing influences
marketing, specific outcomes that are atypical should not be and how they truly impacting the cost of care is worth in-
provided that may mislead patients about the actual prognosis vestigating in the future. DTCA may provide some in-
of their disease. formation, yet not transparency, which is the cornerstone of
Marketing by pharmaceutical companies is regulated; obtaining adequate informed consent. We must improve our
however, the U.S. Food and Drug Administration should communication and marketing to patients and families to
consider heightened scrutiny for cancer-related DTCA.45 provide quality ethical information regarding care and
Furthermore, oversight of marketing activities could be services

1
Department of Medicine, Section of Hematology/Oncology, MacLean Fay J. Hlubocky, PhD, MA, The University of Chicago Medicine, 5841 S.
Center for Clinical Medical Ethics, University of Chicago Medicine, and Maryland Ave., MC 2115, Chicago, IL 60637; Twitter: @FayUCCancerCr;
the Cancer Research Center, Chicago, IL email: [email protected].
2
Department of Psychiatry and Behavioral Sciences, Memorial Sloan
Kettering Cancer Center, New York, NY
3
UNC Lineberger Patient Advocates for Research Council, University of AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
North Carolina at Chapel Hill, Chapel Hill, NC AND DATA AVAILABILITY STATEMENT
4
Truth in Advertising, Inc., Madison, CT Disclosures provided by the authors and data availability statement (if
5
Division of Hematology/Oncology, Massachusetts General Hospital, applicable) are available with this article at DOI https://doi.org/10.1200/
Dana-Farber Partners/Harvard Health System, Boston, MA EDBK_279963.
6
University of Hawai’i Cancer Center, Honolulu, HI
REFERENCES
1. Park J, Look KA. Health expenditure burden of cancer care in the United States. Inquiry. 2019;56:46958019880696.
2. Holcombe RF. The ethics of marketing cancer. J Cancer Policy. 2019;3:1-2.

Hlubocky et al
3. Schwartz LM, Woloshin S. Medical marketing in the United States, 1997-2016. JAMA. 2019;321:80-96.
4. Vater LB, Donohue JM, Arnold R, et al. What are cancer centers advertising to the public?: a content analysis. Ann Intern Med. 2014;160:813-820.
5. YarBorough M, Houk T, Perrault ST, et al. Marketing the research missions of acadmic medical centers: why messages blurring lines between clinical care and
research are bad for business & ethics. Camb Q Healthc Ethics. 2019;28:468-475.
6. Schenker Y, Arnold RM, London AJ. The ethics of advertising for health care services. Am J Bioeth. 2014;14:34-43.
7. Schenker Y, Arnold RM, London AJ. Response to open peer commentaries on “The ethics of advertising for health care services”. Am J Bioeth. 2014;14:W3-W4.
8. Vater LB, Donohue JM, Park SY, et al. Trends in cancer-center spending on advertising in the United States, 2005 to 2014. JAMA Intern Med. 2016;176:
1214-1216.
9. Kontos EZ, Viswanath K. Cancer-related direct-to-consumer advertising: a critical review. Nat Rev Cancer. 2011;11:142-150.
10. Palmer D, Hedberg T. The ethics of marketing to vulnerable populations. J Bus Ethics. 2013;116:403-413.
11. Brenkert G. Marketing and the vulnerable. Bus Ethics Qtly. 1998;1998:7-20.
12. Flaskerud JH, Winslow BJ. Conceptualizing vulnerable populations health-related research. Nurs Res. 1998;47:69-78.
13. MacKenzie AR, Parker I. Introduction to quality issues in vulnerable populations. J Oncol Pract. 2015;11:185-186.
14. Jonsen AR, Seigler M, Winslade WJ. Clinical Ethics, 8th ed. New York: McGraw-Hill; 2015.
15. Donohue J. A history of drug advertising: the evolving roles of consumers and consumer protection. Milbank Q. 2006;84:659-699.
16. Kwoka JE. The Federal Trade Commission and the professions. A quarter century of accomplishment and some new challenges. Antitrust Law J. 2005;
72:997-1021.
17. Burnham JC. American medicine’s golden age: what happened to it? Science. 1982;215:1474-1479.
18. American Medical Association. Principles of Medical Ethics. The Journal of the American Medical Association, Special Edition. Chicago, IL: AMA Press; 1958.
19. 638 F.2d 443: American Medical Association, Petitioner, v. Federal Trade Commission, Respondent. Connecticut State Medical Society and New Haven County
Medical Association, Inc., Petitioners, v. Federal Trade Commission, Respondent. U.S. Court of Appeals SC-Fd e, ed1980.
20. Gail A. Advertising increases health care costs and undermines medical ethics. Missouri Med. 2019;116:344-346.
21. Lipitz-Snyderman A, Vater L, Curry M, et al. Cancer hospital advertising and outcomes: trust the messenger? Lancet Oncol. 2019;20:760-762.
22. Lyles A. Direct marketing of pharmaceuticals to consumers. Annu Rev Public Health. 2002;23:73-91.
23. Wilkes MS, Bell RA, Kravitz RL. Direct-to-consumer prescription drug advertising: trends, impact, and implications. Health Aff (Millwood). 2000;19:110-128.
24. Jajich-Toth C, Roper BW. Americans’ views on health care: a study in contradictions. Health Aff (Millwood). 1990;9:149-157.
25. Carrera PM, Kantarjian HM, Blinder VS. The financial burden and distress of patients with cancer: Understanding and stepping-up action on the financial toxicity
of cancer treatment. CA Cancer J Clin. 2018;68:153-165.
26. Gray SW, Armstrong K, Demichele A, et al. Colon cancer patient information seeking and the adoption of targeted therapy for on-label and off-label indications.
Cancer. 2009;115:1424-1434.
27. Jo HS, Park K, Jung SM. A scoping review of consumer needs for cancer information. Patient Educ Couns. 2019;102:1237-1250.
28. Kaplan RM, Frosch DL. Decision making in medicine and health care. Ann Rev Clin Psych. 2005;1:525-556.
29. Gray SW, Abel GA. Update on direct-to-consumer marketing in oncology. J Oncol Pract. 2012;8:124-127.
30. Abel GA, Burstein HJ, Hevelone ND, et al. Cancer-related direct-to-consumer advertising: awareness, perceptions, and reported impact among patients
undergoing active cancer treatment. J Clin Oncol. 2009;27:4182-4187.
31. Abel GA, Lee SJ, Weeks JC. Direct-to-consumer advertising in oncology: a content analysis of print media. J Clin Oncol. 2007;25:1267-1271.
32. Larson RJ, Schwartz LM, Woloshin S, et al. Advertising by academic medical centers. Arch Intern Med. 2005;165:645-651.
33. Adams C. Direct-to-consumer advertising of prescription drugs can inform the public and improve health. JAMA Oncol. 2016;2:1395-1396.
34. Zivotofsky AZ, Zivotofsky NTS. Capitalism works for health care too. Am J Bioeth. 2014;14:56-58.
35. Blumenthal-Barby JS. Between reason and coercion: ethically permissible influence in health care and health policy contexts. Kennedy Inst Ethics J. 2012;
22:345-366.
36. DPaulo PJ. Research on deception in marketing communications: Its relevance to the study of nonverbal behavior. J Nonverbal Behav. 1988;12:253-273.
37. Aditya RN. The psychology of deception in marketing; a conceptual framework for research and practice. Psychol Market. 2001;18:735-761.
38. Held J, Germelmann CC. Deception in consumer behavior research: a literature review on objective and perceived deception. Projectics. 2018;3:119-145.
39. Schnipper LE, Abel GA. Direct-to-consumer drug advertising in oncology is not beneficial to patients or public health. JAMA Oncol. 2016;2:1397-1398.
40. Vladeck DC. The difficult case of direct-to-consumer drug advertising. Loyola Los Angel Law Rev. 2007;41:259-296.
41. Lee M, Whitehill King K, Reid LN. Factors influencing consumers’ attitudinal and behavioral responses to direct-to-consumer and over-the-counter drug
advertising. J Health Commun. 2015;20:431-444.
42. O’Donohoe S, Jack G. Deception in the marketplace: the psychology of deceptive persuasion and consumer self-protection. Int J Advert. 2010;29:141-144.

43. Gardner DM. Deception in advertising: a conceptual approach. J Mark. 1975;39:40-46.

44. Federal Trade Commission. Guides Concerning the Use of Endorsements & Testimonials in Advertising-16 CFR Part 255. https://www.ftc.gov/sites/default/files/
attachments/press-releases/ftc-publishes-final-guides-governing-endorsements-testimonials/091005revisedendorsementguides.pdf. Accessed March 2020.
45. Lexchin J, Menkes DB. Can direct-to-consumer advertising of prescription drugs be effectively regulated? N Z Med J. 2019;132:59-65.

The Changing Health Insurance Coverage

Landscape in the United States
K. Robin Yabroff, PhD1; Samuel Valdez, MA2; Mireille Jacobson, PhD3; Xuesong Han, PhD1; and A. Mark Fendrick, MD4
overview
Changes in the health insurance coverage landscape in the United States during the past decade have
important implications for receipt and affordability of cancer care. In this paper, we summarize evidence for
the association between health insurance coverage and cancer prevention and treatment. We then discuss
ongoing changes in health care coverage, including implementation of provisions of the Affordable Care Act,
increasing prevalence of high-deductible health insurance plans, and factors that affect health care delivery,
with a focus on vertical integration of hospitals and providers. We summarize the evidence for the effects of the
changes in health coverage on care and discuss areas for future research with the goal of informing efforts to
improve cancer care delivery and outcomes in the United States.
INTRODUCTION (ACA), increasing prevalence of high-deductible health

The United States leads the world in the speed of insurance plans (HDHPs), and factors that affect
adoption of innovative new therapies and state-of-the- health care delivery, with a focus on vertical integration
art health care. Many of these innovations are in cancer of hospitals and providers.
care, including chimeric antigen receptor T-cell ther- HEALTH INSURANCE COVERAGE AND CANCER CARE
apy and other forms of targeted systemic therapies, as The burden of cancer is unequal in the United States,
well as robotic surgery and proton beam therapy. with large variation in survival and mortality by state,8
These therapies are expensive, and list prices of new with the highest mortality rates among socioeco-
cancer drugs are now routinely above $100,000 nomically disadvantaged and uninsured populations.9
annually.1,2 Other components of cancer care, in- Many of these disparities are thought to be driven by
cluding in-patient hospitalizations, advanced imaging, disparities in access to cancer prevention, screening,
and supportive care, are also expensive and contribute and treatment—key components of the cancer control
to cancer being one of the most expensive medical continuum (Fig. 1). Cancer prevention, such as to-
conditions to treat.3 With the increasing prevalence of bacco cessation, can reduce the risk of developing
patients with cancer and survivors of cancer mainly cancer, and asymptomatic cancer screening and
because of an aging and growing population,4 the timely follow-up of abnormal test results can lead to
economic burden of cancer care is projected to in- early detection and earlier stage of disease at di-
crease dramatically in the future.5 agnosis, where treatment is more effective. Similarly,
Access to state-of-the-art cancer treatment is often timely access to evidence-based cancer treatment is
unattainable without health insurance, however. The associated with improved survival, and coordinated
United States does not have a single health care survivorship care is associated with better patient out-
system, nor does it provide universal health insurance comes. However, a large body of research has docu-
coverage. Because health insurance has historically mented that uninsured individuals are less likely than
been one of the strongest determinants of access to their counterparts with health insurance coverage to
Author affiliations care and health outcomes,6,7 many health policy efforts receive high-quality care across the cancer control
and support have focused on increasing health insurance coverage continuum,10-12 potentially explaining some of the dis-
information (if
and the structure, financing, and organization of health parities in cancer outcomes.
applicable) appear
at the end of this
care delivery. Uninsured individuals are less likely to receive cancer
article. In this paper, we summarize the evidence for the risk assessment and prevention, such as smoking
Accepted on May 3, association between health insurance coverage and cessation or diet, weight, and physical activity coun-
2020 and published
access to and receipt of cancer care. We then discuss seling than their counterparts with insurance cover-
at ascopubs.org on
health insurance in the United States and the chang- age.13 Uninsured individuals are also less likely to
doi.org/10.1200/ ing landscape of health care coverage, including the receive age-appropriate breast, cervical, colorectal,
EDBK_279951 implementation of provisions in the Affordable Care Act and lung cancer screening as recommended by the

growing cancer survival disparities over the past decades

between uninsured individuals and those with private in-
surance coverage.19 In the following sections, we discuss
• Having health insurance coverage is strongly different types of health insurance coverage and how as-
associated with access to care. The uninsured
pects of coverage, especially patient cost sharing, are in-
are less likely to receive care throughout the
creasing for access to care across the cancer control
cancer control continuum, including cancer
prevention, screening, diagnosis, treatment, continuum.
survivorship, and end-of-life care. The un- HEALTH INSURANCE COVERAGE OPTIONS
insured also have poorer survival following
cancer diagnosis. A key component of health care in the United States is the
type and number of health insurance coverage options,
• Implementation of insurance coverage expan-
sion provisions as part of the ACA led to declines which differ substantially by age group. The federal Medi-
in the uninsured population and improvements care program provides insurance coverage for more than
in receipt of cancer screening and earlier-stage 96% of Americans age 65 or older.20 Many (56% in 2018)
disease at diagnosis. Additional research is also have supplemental private coverage from prior em-
needed to evaluate the association of insurance ployers as part of retirement benefits or purchased sepa-
expansion on receipt of cancer treatment and rately. Some also receive coverage from the state/federal
survival. Medicaid programs for low-income populations, also known
• Trends toward increased patient cost sharing as “dually eligible” for both Medicare and Medicaid (14% in
based on the price of health care services, 2018).20
rather than their clinical value, including growth
For the adult population younger than age 65 years, about
in high deductible health plans, may limit ac-
two-thirds have some form of private health insurance
cess to care and increase financial hardship.
V-BID models may realign coverage to improve coverage, which is mainly employer based. About 24% of
the quality of care delivery. adults younger than age 65 have some form of public in-
surance coverage.20 A small proportion (3% in 2018) of the
• Hospitals and health systems are increasingly
younger population receives coverage through the federal
acquiring or otherwise aligning with oncology
practices. This vertical integration is associated Medicare program because of permanent disability. Med-
with increased in-system patient referral pat- icaid coverage in the low-income population younger than
terns. Additional research is needed to un- age 65 years varies more by state than in the elderly
derstand the association of these changes in population, with variation in eligibility thresholds, covered
referral patterns for the quality of cancer care. services, recertification requirements, and provider re-
imbursement. Other health care programs are available,
such as TRICARE for uniformed service members, retirees,
U.S. Preventive Services Task Force (USPSTF) and other and their families, and the Department of Veterans Affairs,
screening guidelines.10,13-15 Among those who do receive which offers coverage options for current and former military
cancer screening and have abnormal test findings, un- personnel. Those without employer-based or other health
insured individuals are less likely to receive timely follow-up insurance coverage options can purchase private policies
diagnostic testing, if any.16 Insurance coverage is also as- individually. Before implementation of the health insurance
sociated with stage of disease at diagnosis; uninsured in- marketplace as part of the ACA described later, individual
dividuals are less likely to be diagnosed with early-stage policies were frequently expensive and could restrict cov-
disease,7,10 where treatment can be more effective. erage for treatments of preexisting conditions or deny
Patients with cancer without health insurance are also less coverage altogether for individuals with preexisting condi-
likely to receive high-quality treatment, including surgery, tions. Individuals without private or public health insurance
radiation therapy, and chemotherapy,10 as well as newer coverage options are uninsured (10% in 2018).20
immunotherapies17 and targeted agents. After treatment, As noted previously, research has consistently shown that
survivors of cancer without health insurance coverage are uninsured individuals have poorer access to care and lower
less likely to receive recommended survivorship care,18 receipt of evidence-based cancer care.10 However, even
such as surveillance for recurrence and other recom- those with health insurance may experience barriers to care
mended preventive services. Innovations in cancer pre- because of high premiums, deductibles that patients must
vention, screening, and treatment, including genomic pay out of pocket before insurance coverage begins, and
therapies, may also be less available to uninsured pop- cost sharing from copayments and coinsurance rates that
ulations, potentially widening existing disparities in can- are usually required at the point of care after the plan
cer outcomes. Indeed, recent studies have documented deductible is met. Oral cancer drugs, typically classified as

Yabroff et al
FIGURE 1. Cancer Control

Continuum
specialty drugs on most formularies, frequently have the enrolled in Medicare Advantage,26 private managed care
highest coinsurance rates, as much as 30% of the list price plans paid with capitated rates.
in many cases. With the increase in the number of expensive In the following sections, we discuss recent and ongoing
oral drugs available for treating cancer, concerns about changes in the health care landscape in the United States,
patient out-of-pocket burden has increased as well. Insured including implementation of provisions in the ACA, in-
patients with greater cost sharing are more likely to delay or creasing prevalence of HDHPs, and factors that affect
forgo care,21-23 including necessary care. This phenomenon health care delivery, with a focus on vertical integration of
of having health insurance coverage but being unable to hospitals and providers. We also summarize current evi-
access or afford care is known as underinsurance. dence and highlight key areas for future research.
HEALTH CARE IN THE UNITED STATES THE AFFORDABLE CARE ACT AND CARE ACROSS THE
As noted previously, there is not a single health care system CANCER CONTROL CONTINUUM
in the United States. Health care is decentralized, and in- The ACA was enacted in 2010 with the goals of improving
surers, both public and private, operate independently of health insurance coverage options, quality of care, and
each other. Health care providers and practices can be patient outcomes and maintaining or lowering costs through
affiliated with multiple hospitals, networks, and insurance delivery system changes.27-29 The passage of the ACA
plans but not necessarily with each other, and these affil- legislation was highly politicized, and there are persisting
iations can change over time. Care is not explicitly co- misconceptions and a lack of understanding by the gen-
ordinated across these multiple insurers, plans, hospitals, eral public,30 many of whom benefit from its provisions.
practices, and providers. Nonetheless, the ACA can impact cancer care through
Historically, health care has been structured and delivered three main mechanisms: expansion of health insurance
through a fee-for-service model, which reimburses pro- options, coverage reform, and payment and delivery system
viders for each service they deliver. The fee-for-service reform.27,31-37 The health insurance coverage expansion
model provides incentives for greater volumes of services, provisions of the ACA include dependent coverage ex-
without consideration of quality of care, patient out- pansion (DCE), Medicaid expansion, and the establishment
comes, or cost of care. Specialty care has been generally of the Marketplace. After implementation of these pro-
reimbursed at a higher rate than primary care. The fee-for- visions, the number of uninsured individuals in the United
service model of reimbursement is thought to be associated States dropped to a historic low,38 only to increase slightly in
with greater use of technology, specialty care, and higher the past few years.38
health care costs in the United States compared with other The DCE was implemented in September 2010 and allowed
countries.24 young adults to remain on their parents’ private health in-
Starting in the 1970s, health maintenance organizations, surance coverage until age 26. Most research evaluating the
also known as managed care, provided a different model effects of the DCE on coverage and cancer care compare
than fee-for-service. They assume responsibility for all the young adults age 19 to 25 years who are age-eligible for DCE
health care for a defined population of patients for a spec- with young adults who are slightly older (e.g., age 27–34)
ified period and are paid a capitated rate, or fixed fee, for but age ineligible for DCE. Numerous studies have shown
each patient in their defined population. Managed care is that the DCE is associated with increased private coverage,31
primary care focused, with a more limited network of pro- receipt of some preventive services,31 and earlier stage at
viders and facilities, and patients require primary care re- diagnosis among those diagnosed with cancer.31,39-42 The
ferrals to specialty care. This financial structure incentivizes DCE is also associated with more timely receipt of cancer
coordination of care and elimination of service duplica- treatment.43 To date, little research has evaluated the
tion and, unlike fee-for-service, does not reward for ser- effects of DCE on survival after diagnosis; this will be an
vice volume. Managed care has grown dramatically in state important area for future research (Table 1).
Medicaid programs, and as of 2017, more than two-thirds of The ACA expanded Medicaid eligibility to individuals with
beneficiaries nationally were enrolled in managed care income 138% or less of the federal poverty line (FPL) with or
plans.25 Managed care has also grown in the Medicare without dependent children. It went into effect in January
program, and in 2017, 35% of Medicare beneficiaries were 2014 and was originally intended to be implemented

TABLE 1. Research Gaps Across the Cancer Control Continuum

Component of the Cancer Control Continuum
Component and Provision Prevention Screening Diagnosis Treatment Survivorship End of Life
Health Insurance Expansion
Medicaid + + ++ +++ +++ +++
DCE + + ++ +++ +++ +++
Marketplace +++ +++ +++ +++ +++ +++
Health Insurance Coverage Reform
Preventive services cost sharing + + +
Part D donut hole +++ +++ +++ +++ +++ +++
Payment and Delivery System Reform +++ +++ +++ +++ +++ +++
Note: Greater research gaps are indicated by more “+”.

Abbreviation: DCE, dependent coverage expansion.
Reprinted from Zhao et al.31
nationwide. However, a Supreme Court ruling made Med- benefits, bans lifetime or annual maximums, and prohibits
icaid expansion a state option, which created a natural exclusions based on preexisting conditions. To date, little
experiment, allowing comparisons of the receipt of care and research has evaluated the effects of the introduction of
outcomes among adults residing in states that expanded Marketplace on care across the cancer control continuum.31
Medicaid eligibility to those adults residing in states that did
Coverage reform under the ACA includes elimination of cost
not expand Medicaid eligibility. In states that did not expand
sharing for effective preventive services with a USPSTF “A”
Medicaid, eligibility was as low as 17% of the FPL in 2019 for
or “B” rating, including breast, cervical, colorectal, and lung
parents of a family of three in Texas.44 Adults with low income
cancer screening for nongrandfathered health plans. This
without dependent children are not eligible for Medicaid in
provision reflects evidence that, by removing patient cost
many nonexpansion states, regardless of their income.
sharing, health insurance benefits can be designed to
After Medicaid expansion, patients with newly diagnosed promote the use of evidence-based care. Research eval-
cancer in expansion states were more likely to be insured, uating the effects of eliminating cost sharing for preven-
have access to care, and be diagnosed at an earlier stage of tive services has generally found increases in cancer
disease than those in nonexpansion states.31,45-49 Addi- screening,51,52 although increases are not always statisti-
tionally, historic disparities in coverage for patients with cally significant at p , .05. Effects are frequently limited to
newly diagnosed cancer that were a minority, low income, low-income and low-socioeconomic status populations,51
and rural were reduced in Medicaid expansion states, with who might be expected to benefit most from the elimina-
little change in disparities in nonexpansion states.49 Find- tion of cost sharing. Variation in findings may reflect dif-
ings related to the effects of Medicaid expansion on cancer ferences in the years after the implementation of this
screening have been mixed, with studies reporting in- provision, data sources used in studies, and the fact that
creases in cancer screening50 and null effects.50 Mixed most studies were not able to distinguish between non-
findings potentially reflect changes in USPSTF screening grandfathered plans, subject to the provision, and grand-
recommendations during the past decade, the type of fathered plans, which were not subject to this provision.
cancer screening evaluated, and potential latency of the
Other coverage reform provisions in the ACA include the
effect of newly available insurance coverage. Data after
elimination of preexisting condition exclusions or refusals as
Medicaid expansion continue to mature; evaluation of re-
part of medical underwriting, as well as lifetime and annual
ceipt of treatment and survival will be important for future
coverage limits, coverage of routine care for clinical trial
research (Table 1).
participants, and gradual closing of the Medicare Part D
The health insurance Marketplace established under the donut hole, a coverage gap when patients are responsible
ACA provides coverage options for purchase by individuals, for the entirety of their oral prescription drugs costs, until the
families, and small businesses. Tax credit premium sub- catastrophic phase coverage begins (with a 5% coinsurance
sidies are available for those with income 100%–400% of without an annual cap on out-of-pocket spending). To date,
FPL, and cost-sharing subsidies are available to those with little research has addressed the effects of these other
income 100%–250% of FPL. The ACA also requires that coverage reform provisions of the ACA on care across the
health plans on the Marketplace cover essential health cancer control continuum.

Yabroff et al
The ACA also authorized the creation of the Centers for use of clinician visits, diagnostic tests, and medications falls.
Medicare and Medicaid Services Innovation Center to de- Survivors of cancer enrolled in HDHPs are more likely to
velop and test models of health care delivery. The goals of delay or forgo care because of cost than those enrolled
the Innovation Center are to address the triple aim of im- in low-deductible plans.57 Among women diagnosed with
proving health care—better quality, better health, and lower breast cancer, being enrolled in an HDHP was associated
cost—with greater focus on the value of care provided rather with delays in diagnostic imaging, receipt of biopsy, early-
than volume of services. Emerging models are generally stage diagnosis, and chemotherapy initiation.21
referred to as value-based payment models or alternative
Some patients with cancer who cannot afford their pre-
payment models (APMs) and feature greater provider ac-
scription medications turn to independent charitable patient
countability for higher quality of care based on specified
assistance programs to help cover out-of-pocket costs. One
quality metrics and costs of care. They also incentivize
of these organizations, the Patient Access Network Foun-
better care coordination and elimination of duplication of
dation, is busiest in January when deductibles reset each
services and use of ineffective services. The Oncology Care
year. In January 2020 alone, the Patient Access Network
Model (OCM) is a cancer-specific APM for 6-month epi-
Foundation approved more than 34,000 grants to patients
sodes of chemotherapy, which includes patient navigation,
and received nearly 60,000 calls from patients and pro-
provision of treatment summaries, and use of data for
viders throughout the nation seeking help. As shown in
continuous quality improvement. The OCM performance
Fig. 2, the requests for assistance each January dwarf other
period started in July 2016, and preliminary results suggest
months of the year. This January effect is driven not by
that patients with cancer in volunteer practices participating
clinical need but instead by an archaic benefit design that
in the OCM had fewer intensive care unit admissions and
places a new financial burden on patients at the beginning
emergency department visits compared with similar pa-
of each year.
tients treated in comparison practices.53 Additional evalu-
ations of the OCM and other APMs on cancer care and Much of the negative clinical effects of cost-related non-
outcomes are ongoing. Other changes to the health in- adherence to recommended health care (i.e., financial
surance landscape, including the rapid growth of HDHPs toxicity) impacts those who are economically disadvantaged
and vertical integration of oncology practices are described and/or diagnosed with chronic medical conditions.59 The
in more detail in the following sections. growing number of Americans, including Medicare bene-
ficiaries, facing thousands of dollars in out-of-pocket costs
HIGH-DEDUCTIBLE HEALTH INSURANCE PLANS in addition to premiums raises serious concern. Medicare
fee-for-service does not have annual spending caps on cost
Forty-three percent of working-age adults in the United
sharing, and in 2016, the average out-of-pocket spending
States with employer-sponsored private health insurance
by Medicare enrollees was $5,460,60 a significant portion of
coverage were enrolled in HDHPs in 2017, up from just
fixed annual income for many elderly adults. Approximately
15% in 2007.54 These plans offer lower premiums than
40% of all Americans would struggle to pay an unexpected
alternative plans, which are achieved by shifting costs to
bill of $400.56 Although it would be advantageous for the
enrollees who use medical care. In addition, HDHPs spe-
many individuals with moderate to significant medical
cifically enable patients to open tax-advantaged health
needs and low disposable income to avoid plans with high
savings accounts (HSAs). Partly because of these features,
deductibles, many Americans have no choice but to enroll in
more than three-fourths of adults with employer-sponsored
such plans.
health insurance have to meet some type of deductible
before their insurance starts to pay for care received.55 The A value-based insurance design (V-BID) is a clinically driven
current minimum deductibles in HDHPs are $1,350 in alternative to strategies such as deductibles that do not
individual plans and $2,700 in family plans.56 However, differentiate between evidence-based care and services that
most plans set deductibles even higher; in 2019, the av- are not clinically indicated. V-BID plans are built on the
erage individual deductible was $1,655 and the average principle of lowering or removing financial barriers to es-
family deductible was $4,779.55 sential, high-value clinical services and increasing out-of-
pocket costs for low-value care that does not improve
One of the goals of plan deductibles is to encourage patients
patient-reported outcomes. V-BID plans are designed with
to become more engaged in their health care purchases.
the tenets of clinical nuance in mind, which recognize the
However, in the current “blunt” form in which deductibles
clinical benefit derived from a specific service that depends
are applied, there is no distinction between essential high-
on the consumer using it, as well as when, where, and by
value care and lower-value care. Thus, before meeting the
whom the service is provided.
deductible, enrollees must pay full price for almost all of
their care, regardless of clinical value. A robust evidence For over a decade, numerous private and public payers,
base demonstrates that as patient out-of-pocket costs rise, employers, unions, and business coalitions nationwide have

FIGURE 2. Monthly Patterns of the Patient Access Network

Reprinted from Klein and Fendrick AM.58
implemented clinically nuanced V-BID programs, reducing high deductibles on essential chronic disease services to
consumer cost sharing for specified visits, diagnostic tests, control spending has imparted a tax on millions of Ameri-
and treatments. Evidenced by success in the private sector cans with chronic medical conditions. Innovative, cost-
and its implementation in the ACA (Section 2713 requires neutral plan designs that lower consumer out-of-pocket
nongrandfathered private insurance plans to provide first- spending on essential clinical services across the cancer
dollar coverage for preventive care services specified by the control continuum while decreasing exposure to and
USPSTF with an “A” or “B” rating), V-BID can now be spending on harmful care could also improve patient out-
applied to HSA-eligible HDHPs, Medicare Advantage, and comes, reduce disparities, and potentially lower overall
the TRICARE program. Evidence is accumulating that V-BID health care spending. Ongoing evaluation of plans with
plans increase the use of high-value services, lower patient value-based principles that are increasingly implemented
out-of-pocket costs, and reduce health care disparities, by public and private purchasers on access to care across
often without added spending, and in some circumstances the cancer control continuum will be important.
can reduce aggregate expenditures.61
Accordingly, policymakers are shifting their focus toward VERTICAL INTEGRATION OF HOSPITALS AND PROVIDERS
value-based designs in public and private insurance pro- Managed care plans, whether available through employer-
grams. The latest iteration, V-BID X, demonstrates that out- sponsored private insurance, Medicare Advantage, or pri-
of-pocket costs for high-value services can be reduced vate Medicaid plans or on the ACA Marketplace exchanges,
without increasing deductibles or premiums. Actuarial es- try to reign in high health care costs in part by restricting
timates are used to match the incremental spending on network access. That is, enrollees must access providers
high-value services dollar for dollar with cost savings within their plans’ restricted networks or face large out-of-
resulting from higher cost sharing on, and the resultant pocket costs. Perhaps to offset the continued move toward
decreased use of, unnecessary services. In February 2020, restricted networks, hospitals and health systems are in-
the U.S. Department of Health and Human Services de- creasingly acquiring or otherwise aligning with physician
clared broad support for V-BID X health plans in the pro- practices. Such vertical integration of hospitals and physi-
posed 2021 Notice of Benefit and Payment Parameters rule cians may give health systems more leverage than in-
for Federally Qualified Health Plans.62 dependent hospitals or physicians in network coverage
negotiations.64 This leverage may be particularly strong
In the last year, nearly one-third of Americans reported not
when it includes specialists such as oncologists.
following clinicians’ recommendations for care because of
cost,63 potentially leading to worse health outcomes and Although vertical integration has increased across spe-
sometimes higher total medical expenditures. Although cialties, oncology practices consistently have the highest
deductibles are entrenched in the American health in- rates of integration.65 To describe trends in vertical in-
surance landscape, the common implementation imposing tegration of oncologists, we (Jacobson and Valdez) use

Yabroff et al
2003–2017 data from SK&A, a national office-based phy- reimbursement for Part B drugs, primarily physician-
sician database developed for medical marketing purposes, administered infusion drugs including most chemother-
that has been acquired by IQVIA and rebranded as the apy drugs, generally lowered drug profit margins for
OneKey. These data are increasingly used by researchers to physicians. This decline in profit margins could have en-
study physician markets generally and specifically for ver- couraged vertical integration so that oncologists could
tical integration.66 They have been validated against several benefit from the bulk discounts that health systems may
other well-known data sources65,66 and compare well receive on chemotherapy agents. In contrast, the 340B
specifically with other data on oncologists.67 Our data in- Drug Discount Program enables qualified entities to buy
clude all physicians who had a specialty in oncology/he- drugs, including chemotherapy agents, at deeply dis-
matology, radiation oncology, or gynecologic oncology. To counted prices. Thus, in principle, the 340B Program could
study the implications of integration for patient referral provide a profit-driven incentive for newly covered hospitals
patterns, we merge the SK&A data by National Provider to expand their patient base for chemotherapy treatment by
Identifier to the Medicare’s Physician Shared Patient Pat- integrating with oncology practices.
terns data from 2009 to 2015, the only years of data publicly
Although no evidence has been found to support the role of
available at present.68 We focus on the unique Medicare
Part B reimbursement changes,68 the evidence on the 340B
beneficiaries treated by pairs of oncologists within a 30-day
Program is conflicting and depends on the period of time
period, aggregated across each calendar year.
analyzed and the source of variation in program access.
Figure 3 shows trends in both the number of practices with Expansion of access to the 340B Program under the ACA to
oncologists and the percentage of oncology practices Critical Access Hospitals, sole community hospitals, rural
owned by a hospital or health system between 2003 and referral centers, and cancer centers did not contribute to the
2017. Although the number of practices with oncologists growth of oncology practice integration.68 In contrast, one
has increased by about 10% since 2010, integration has study found that the discrete change in 340B Program
increased much more dramatically. Rates of hospital or eligibility for hospitals with Disproportionate Share Hospital
health system ownership of oncology practices declined (reflects disproportionate number of hospitalized patients
slightly between 2003 and 2010 but increased sharply who are low income) adjustment percentages at the 11.75%
thereafter. System acquisitions of oncology practices in- threshold increases the likelihood of consolidation at a point
creased quickly between 2010 and 2015 and stabilized in time.69 Other potential causes of oncology practice
thereafter at about 60% in 2017. consolidation have received less scholarly attention, in-
Some causes for the increased pace of health system ac- cluding increases in (1) health system incentives to manage
quisition of oncology practices have been suggested and the continuum of care under APMs, (2) benefits to practices
studied previously, including the role of changes to Medi- from having access to electronic infrastructure available
care’s Part B reimbursement system under the Medicare through large systems, and (3) the value of physician–
Drug Improvement and Modernization Act in 2005 and hospital alignment as payers pursue narrow networks.
the 340B Drug Payment Program. 68,69 The Medicare What is the value of alignment for oncologists and health
Drug Improvement and Modernization Act’s changes in systems in markets with narrow networks? For oncologists,
FIGURE 3. Trends in Hospital or

Health System Ownership of On-
cology Practices, 2003–2017
This figure is based on the au-
thors’ analysis of SK&A data.
Data have been modified and
imputations performed on 2015
data relative to Alpert et al, ex-
hibit 1.68

excluding those physicians, however, in-system referrals

increase to more than 40% in the 4 years after integration.
Although we have no data to test at this time, we suspect that
patients enrolled in managed care plans, whether Medicare
Advantage or others, face even starker changes in referral
patterns when their oncologists are acquired by health
systems.
CONCLUSION
In this review, we summarized the evidence for the asso-
ciation between health insurance coverage and access to
and receipt of cancer prevention and treatment in the
FIGURE 4. Oncologists’ Share of Referrals Made to Current/Future United States. Many studies have reported that uninsured
Health System Partners, by Year Relative to Integration (Balanced) individuals are less likely to receive evidence-based health
This figure is based on the authors’ analysis of SK&A and PSPP care throughout the cancer control continuum and have
data. Data cover the years 2009–2015. Abbreviation: PSPP, Medicare’s poorer survival after diagnosis. The implementation of in-
Physician Shared Patient Practice Patterns. surance coverage expansion provisions as part of the ACA
has led to historic declines in the uninsured population and
integration with a health system may increase the odds that improvements in receipt of cancer screening and earlier-
payers keep them in network. This is because payers are stage disease diagnosis. Evaluation of other ACA provisions
responsive to the fact that patients (consumers) want ac- is ongoing. Other aspects of the changing health insurance
cess to high-quality health systems. An oncologist who landscape, including the rapid increase in prevalence of
aligns with a health system creates a competitive advantage HDHPs, may limit access to care and increase financial
in their access to health system facilities. For health sys- hardship. V-BID models may realign coverage to improve
tems, alignment may better enable them to control referrals care delivery. Insurance coverage and plan benefit design
by offering incentives for oncologists (and other providers) to directly affect patients. However, the structure of provider
keep patient referrals for imaging, surgery, and other spe- relationships can also affect delivery of care. Most notably,
cialized oncology services in-house.70 We take a first look at vertical integration of hospitals and providers can affect
this possibility by analyzing how integration affects referrals where patients are referred for care. We showed here that,
by oncologists within the Medicare fee-for-service system. although beneficiaries enrolled in traditional Medicare have
Although fee-for-service Medicare beneficiaries ostensibly access to any oncologist who accepts Medicare, referral
have access to any provider accepting Medicare, where they patterns are sensitive to the relationships oncologists have
are referred to seems to change as oncologists integrate. In with hospitals and health systems. When oncologists ver-
particular, our analysis shows that the share of referrals by tically integrate with a hospital or health system, they change
oncologists to future/current health system partners in- whom they refer their patients to for care. In particular, they
creases sharply after integration (Fig. 4). Although about are much more likely to refer patients to other in-house
15% of referrals are to future health system partners in the oncologists. Ongoing work by one of the study authors
3 years before integration, that share increases sharply (Valdez) is analyzing the implications of these changes in
during the year of integration. Overall, in-system referrals referral patterns for the quality of cancer care. As health
increase to about 60%. Some of that increase is me- insurance coverage and care continue to evolve, ongoing
chanical, because of referrals to other physicians in the research can inform future efforts to improve cancer care
same practice who integrate at the same time. Even delivery and outcomes in the United States.

1
Surveillance and Health Services Research, American Cancer Society, K. Robin Yabroff, PhD, Surveillance and Health Services Research
Atlanta, GA Program, American Cancer Society, 250 Williams St., Atlanta, GA 30303;
2
Department of Economics, University of California, Irvine, CA email: [email protected].
3
Leonard Davis School of Gerontology, University of Southern California,
Los Angeles, CA
4
University of Michigan Center for Value-Based Insurance Design, Ann
Arbor, MI
EDBK_279951.

Yabroff et al
REFERENCES
1. Bach PB. Limits on Medicare’s ability to control rising spending on cancer drugs. N Engl J Med. 2009;360:626-633.
2. Memorial Sloan Kettering Cancer Center. Price and value of cancer drug. https://www.mskcc.org/research-areas/programs-centers/health-policy-outcomes/cost-
drugs. Accessed May 10, 2018.
3. Soni A. Trends in the Five Most Costly Conditions among the U.S. Civilian Noninstitutionalized Population, 2002 and 2012. Rockville, MD: Agency for Healthcare
Research and Quality, Rockville, MD. http://www.meps.ahrq.gov/mepsweb/data_files/publications/st470/stat470.shtml. Accessed May 12, 2020.
4. Miller KD, Nogueira L, Mariotto AB, et al. Cancer treatment and survivorship statistics, 2019. CA Cancer J Clin. 2019;69:363-385.
5. Mariotto AB, Yabroff KR, Shao Y, et al. Projections of the cost of cancer care in the United States: 2010-2020. J Natl Cancer Inst. 2011;103:117-128.
6. Ward E, Halpern M, Schrag N, et al. Association of insurance with cancer care utilization and outcomes. CA Cancer J Clin. 2008;58:9-31.
7. Halpern MT, Ward EM, Pavluck AL, et al. Association of insurance status and ethnicity with cancer stage at diagnosis for 12 cancer sites: a retrospective analysis.
Lancet Oncol. 2008;9:222-231.
9. Ma J, Altekruse S, Cosgrove C, et al. Educational disparities in mortality between adults aged 50-64 and 66-79 years, U.S. Am J Prev Med. 2017;52:728-734.
10. Mandelblatt JS, Yabroff KR, Kerner JF. Equitable access to cancer services: A review of barriers to quality care. Cancer. 1999;86:2378-2390.
11. Institute of Medicine. National Research Council National Cancer Policy B. In Hewitt M, Simone JV, (eds). Ensuring Quality Cancer Care. Washington, DC:
National Academies Press (US); 1999;46-78.
12. Institute of Medicine. Delivering Affordable Cancer Care in the 21st Century: Workshop Summary. Washington, DC: National Academies Press (US); 2013.
13. Ross JS, Bradley EH, Busch SH. Use of health care services by lower-income and higher-income uninsured adults. JAMA. 2006;295:2027-2036.
14. Sauer AG, Siegel RL, Jemal A, et al. Updated review of prevalence of major risk factors and use of screening tests for cancer in the United States. Cancer
Epidemiol Biomarkers Prev. 2017;26:1192-1208.
15. Zhao G, Okoro CA, Li J, et al. Health insurance status and clinical cancer screenings among U.S. adults. Am J Prev Med. 2018;54:e11-e19.
16. Yabroff KR, Washington KS, Leader A, et al. Is the promise of cancer-screening programs being compromised? Quality of follow-up care after abnormal screening
results. Med Care Res Rev. 2003;60:294-331.
17. Shih YC, Elting LS, Halpern MT. Factors associated with immunotherapy use among newly diagnosed cancer patients. Med Care. 2009;47:948-958.
18. Robin Yabroff K, Short PF, Machlin S, et al. Access to preventive health care for cancer survivors. Am J Prev Med. 2013;45:304-312.
19. Ellis L, Canchola AJ, Spiegel D, et al. Trends in cancer survival by health insurance status in California from 1997 to 2014. JAMA Oncol. 2018;4:317-323.
20. U.S. Census Bureau. 2018 American Community Survey. https://wwwcensusgov/data/tables/time-series/demo/health-insurance/acs-hihtml. Accessed March
16, 2020.
21. Wharam JF, Zhang F, Lu CY, et al. Breast cancer diagnosis and treatment after high-deductible insurance enrollment. J Clin Oncol. 2018;36:1121-1127.
22. Dusetzina SB, Winn AN, Abel GA, et al. Cost sharing and adherence to tyrosine kinase inhibitors for patients with chronic myeloid leukemia. J Clin Oncol. 2014;
32:306-311.
23. Zheng Z, Han X, Guy GP Jr., et al. Do cancer survivors change their prescription drug use for financial reasons? Findings from a nationally representative sample in
the United States. Cancer. 2017;123:1453-1463.
24. Miller HD. From volume to value: better ways to pay for health care. Health Aff (Millwood). 2009;28:1418-1428.
25. Centers for Medicare and Medicaid Services. Medicaid managed care enrollment and program characteristics, 2017. https://www.medicaid.gov/medicaid/
managed-care/downloads/enrollment/2017-medicaid-managed-care-enrollment-report.pdf. Accessed March 16, 2020.
26. Yabroff KR, Kirby J, Zodet M. Association of insurance gains and losses with access to prescription drugs. JAMA Intern Med. 2017;177:1531-1532.
27. Leopold C, Park ER, Nekhlyudov L. The impact of the Affordable Care Act on cancer survivorship. Cancer J. 2017;23:181-189.
28. Bailes JS, Kamin DY, Foster SE. The patient protection and affordable care act: exploring the potential impact on oncology practice. Cancer J. 2010;16:588-592.
29. Collins BL, Saylor J. The Affordable Care Act: Where are we now? Nursing. 2018;48:43-47.
30. Kaiser Family Foundation. Misconceptions surrounding the ACA. https://www.kff.org/health-reform/poll-finding/data-note-5-misconceptions-surrounding-the-
aca/. Accessed April 26, 2020.
31. Zhao J, Mao Z, Fedewa SA, et al. The Affordable Care Act and access to care across the cancer control continuum: A review at 10 years. CA Cancer J Clin. Epub
2020 Mar 23.
32. Graves JA, Swartz K. Effects of affordable care act marketplaces and Medicaid eligibility expansion on access to cancer care. Cancer J. 2017;23:168-174.
33. Brooks GA, Hoverman JR, Colla CH. The Affordable Care Act and cancer care delivery. Cancer J. 2017;23:163-167.
34. Parikh RB, Wright AA. The affordable care act and end-of-life care for patients with cancer. Cancer J. 2017;23:190-193.
35. Sabik LM, Adunlin G. The ACA and cancer screening and diagnosis. Cancer J. 2017;23:151-162.
36. Han X, Jemal A. The Affordable Care Act and cancer care for young adults. Cancer J. 2017;23:194-198.
37. Dixon MS, Cole AL, Dusetzina SB. Out-of-pocket spending under the Affordable Care Act for patients with cancer. Cancer J. 2017;23:175-180.

38. Garfield ROK, Damico A. The Henry J Kaiser Family Foundation. The uninsured and the ACA: a primer: key facts about health insurance and the uninsured
amidst changes to the Affordable Care Act. https://www.kff.org/uninsured/report/the-uninsured-and-the-aca-a-primer-key-facts-about-health-insurance-and-
the-uninsured-amidst-changes-to-the-affordable-care-act/. Accessed May 12, 2020.
39. Han X, Yabroff KR, Robbins AS, et al. Dependent coverage and use of preventive care under the Affordable Care Act. N Engl J Med. 2014;371:2341-2342.
40. Robbins AS, Han X, Ward EM, et al. Association between the affordable care act dependent coverage expansion and cervical cancer stage and treatment in young
women. JAMA. 2015;314:2189-2191.
41. Wallace J, Sommers BD. Effect of dependent coverage expansion of the Affordable Care Act on health and access to care for young adults. JAMA Pediatr. 2015;
169:495-497.
42. Han X, Zang Xiong K, Kramer MR, et al. The Affordable Care Act and cancer stage at diagnosis among young adults. J Natl Cancer Inst. 2016;108:djw058.
43. Nogueira L, Chawla N, Han X, et al. Colorectal cancer care among young adult patients after the dependent coverage expansion under the Affordable Care Act.
J Natl Cancer Inst. 2019;djz235.
44. Kaiser Family Foundation. Medicaid income eligibility limits for adults as a percent of the federal poverty level. https://www.kfforg/health-reform/state-indicator/
medicaid-income-eligibility-limits-for-adults-as-a-percent-of-the-federal-poverty-level/?currentTimeframe=0&sortModel=%7B%22colId%22:%22Location%
22,%22sort%22:%22asc%22%7D. Accessed March 17, 2020.
45. Jemal A, Lin CC, Davidoff AJ, et al. Changes in insurance coverage and stage at diagnosis among nonelderly patients with cancer after the Affordable Care Act.
J Clin Oncol. 2017;35:3906-3915.
46. Davidoff AJ, Guy GP Jr., Hu X, et al. Changes in health insurance coverage associated with the Affordable Care Act among adults with and without a cancer
history: population-based national estimates. Med Care. 2018;56:220-227.
47. Soni A, Simon K, Cawley J, et al. Effect of Medicaid expansions of 2014 on overall and early-stage cancer diagnoses. Am J Public Health. 2018;108:216-218.
48. Soni A, Sabik LM, Simon K, et al. Changes in insurance coverage among cancer patients under the Affordable Care Act. JAMA Oncol. 2018;4:122-124.
49. Han X, Yabroff KR, Ward E, et al. Comparison of insurance status and diagnosis stage among patients with newly diagnosed cancer before vs after implementation
of the Patient Protection and Affordable Care Act. JAMA Oncol. 2018;4:1713-1720.
50. Fedewa SA, Yabroff KR, Smith RA, et al. Changes in breast and colorectal cancer screening after medicaid expansion under the Affordable Care Act. Am J Prev
Med. 2019;57:3-12.
51. Fedewa SA, Goodman M, Flanders WD, et al. Elimination of cost-sharing and receipt of screening for colorectal and breast cancer. Cancer. 2015;
121:3272-3280.
52. Trivedi AN, Leyva B, Lee Y, et al. Elimination of cost sharing for screening mammography in Medicare Advantage Plans. N Engl J Med. 2018;378:262-269.
53. Brooks GA, Jhatakia S, Tripp A, et al. Early findings from the oncology care model evaluation. J Oncol Pract. 2019;15:e888-e896.
54. Cohen RAZE, Zammitti EP. High-deductible health plan enrollment among adults aged 18-64 with employment-based insurance coverage. NCHS Data Brief.
2018;317:1-8.
55. Kaiser Family Foundation. 2019 Employer Health Benefits Survey: section 7: employee cost sharing. https://www.kff.org/report-section/ehbs-2019-section-7-
employee-cost-sharing. Accessed January 13, 2020.
56. Board of Governors of the Federal Reserve System. Report on the economic well-being of U.S. households in 2018. https://www.federalreserve.gov/publications/
files/2018-report-economic-well-being-us-households-201905.pdf. Accessed January 13, 2020.
57. Zheng Z, Jemal A, Banegas MP, et al. High-deductible health plans and cancer survivorship: what is the association with access to care and hospital emergency
department use? J Oncol Pract. 2019;15:e957-e968.
58. Klein D, Fendrick AM. Another costly January has come and gone for Medicare Part D beneficiaries. https://morningconsult.com/opinions/another-costly-
january-has-come-and-gone-for-medicare-part-d-beneficiaries/. Accessed March 16, 2020.
59. Yabroff KR, Zhao J, Han X, et al. Prevalence and correlates of medical financial hardship in the USA. J Gen Intern Med. 2019;34:1494-1502.
60. Cubanski JKW, Damico A, Neuman T. How much do medicare beneficiaries spend out of pocket on health care? https://wwwkfforg/report-section/how-much-do-
medicare-beneficiaries-spend-out-of-pocket-on-health-care-methodology/. Accessed March 17, 2020.
61. Agarwal R, Gupta A, Fendrick AM. Value-based insurance design improves medication adherence without an increase in total health care spending. Health Aff
(Millwood). 2018;37:1057-1064.
62. HHS Notice of Benefit and Payment Parameters. Notice requirement for non-federal governmental plans. https://www.federalregister.gov/documents/2020/02/
06/2020-02021/patient-protection-and-affordable-care-act-hhs-notice-of-benefit-and-payment-parameters-for-2021. Accessed February 22, 2020.
63. Kirzinger A, Lopes L, Wu B, Brodie M KFF health tracking poll—February 2019: prescription drugs. https://www.kff.org/health-costs/poll-finding/kff-health-
tracking-poll-february-2019-prescription-drugs. Accessed January 14, 2020.
64. McCarthy I, Huang SS. Vertical alignment between hospitals and physicians as a bargaining response to commercial insurance markets. Rev Ind Organ. 2018;
53:7-29.
65. Nikpay SS, Richards MR, Penson D. Hospital-physician consolidation accelerated in the past decade in cardiology, oncology. Health Aff (Millwood). 2018;
37:1123-1127.
66. Baker LC, Bundorf MK, Kessler DP. The effect of hospital/physician integration on hospital choice. J Health Econ. 2016;50:1-8.

Yabroff et al
67. DesRoches CM, Barrett KA, Harvey BE, et al. The results are only as good as the sample: assessing three national physician sampling frames. J Gen Intern Med.
2015;30(suppl 3):S595-S601.
68. Alpert A, Hsi H, Jacobson M. Evaluating the role of payment policy in driving vertical integration in the oncology market. Health Aff (Millwood). 2017;36:680-688.
69. Desai S, McWilliams JM. Consequences of the 340B drug pricing program. N Engl J Med. 2018;378:539-548.
70. Mathews AWEM. The hidden system that explains how your doctor makes referrals. The Wall Street Journal (December 27, 2018). https://www.wsj.com/articles/
the-hidden-system-that-explains-how-your-doctor-makes-referrals-11545926166. Accessed May 12, 2020.

HEMATOLOGIC
MALIGNANCIES

ADCs, BiTEs, CARs, and Small Molecules:

A New Era of Targeted Therapy in Non-Hodgkin
Lymphoma
Jeremy S. Abramson, MD, MMSC1; Nilanjan Ghosh, MD, PhD2; and Sonali M. Smith, MD, FASCO3
overview
Novel immunotherapies and small molecular inhibitors are transforming our approach to previously treated
and newly diagnosed patients across the spectrum of non-Hodgkin lymphomas (NHLs). Anti-CD19 CAR T cells
are now indicated for the treatment of relapsed/refractory aggressive B-cell lymphomas after at least two
previous lines of therapy in which durable remissions are achieved in approximately 40% of previously in-
curable patients. Second-line chemoimmunotherapy remains the standard of care at first relapse, but poor
outcomes with conventional treatment in this setting creates an appealing rationale for earlier use of CAR
T cells, which is currently under investigation, along with even earlier use in selected high-risk patients in the
frontline setting. Other emerging immunotherapies include antibody-drug conjugates (ADCs), such as
polatuzumab vedotin for multiple-relapsed diffuse large B-cell lymphoma (DLBCL) in combination with
bendamustine-rituximab. Multiple bispecific antibodies that bring malignant B cells in contact with effector
T cells appear promising in early clinical trials and will likely emerge as off-the-shelf immunotherapy options.
Chemotherapy-free small molecule–based regimens are increasingly available for mantle cell (MCLs) and
follicular lymphomas (FLs). Bruton tyrosine kinase inhibitors (BTKi) now represent standard second-line
therapy for MCL and are being investigated in combination and as initial therapy. Lenalidomide-rituximab is
an active regimen in both FL and MCL and may be used in either relapsed/refractory or previously untreated
disease. Three PI3K inhibitors are approved for multiple-relapsed FL and can induce durable remissions in
patients with chemotherapy- and rituximab-refractory disease. Additional emerging targeted therapies include
BCL2 inhibition in MCL and EZH2 inhibition in FL.
INTRODUCTION 1 year of autologous stem cell transplant (ASCT),

The therapeutic landscape in NHLs is rapidly evolv- have particularly poor outcomes with only a 7%
ing with novel immunotherapies and small molecule chance of achieving complete response (CR) to
pathway inhibitors transforming treatment para- conventional therapy and brief OS.2 In this context,
digms for DLBCL, FL, MCL, and others. This review three anti-CD19 CAR T-cell products (axicabtagene
will consider how targeted therapies fit into cur- ciloleucel, 3 tisagenlecleucel,4 and lisocabtagene
rent standards of care for these diseases and maraleucel 5) have demonstrated high overall re-
look to the future at how this evolution is likely to sponse and CR rates in these difficult-to-treat patients,
continue. with durable remissions seen in approximately 40% of
patients. These trials demonstrated markedly improved
CURRENT AND EVOLVING ROLE OF CAR T CELLS IN OS compared with what would be expected with con-
AGGRESSIVE B-CELL LYMPHOMAS ventional therapies, which suggests a dramatic change
Author affiliations for the better in the natural history of multiple-relapsed
and support CAR T Cells for Multiple-Relapsed/Refractory or refractory patients with aggressive B-cell lymphomas.
information (if Aggressive B-Cell Lymphomas
applicable) appear
Based on the strength of these data, axicabtagene
at the end of this Anti-CD19 CAR T cells have transformed our approach ciloleucel and tisagenlecleucel have been approved
article. to multiple-relapsed/refractory aggressive B-cell lym- by the U.S. Food and Drug Administration (FDA) and
Accepted on phomas. Historically, patients who relapse after second- the European Medicine Agency for the treatment of
February 26, 2020 line chemoimmunotherapy have had a poor prognosis, patients with aggressive B-cell lymphoma who have
and published at
with a median overall survival (OS) of approximately relapsed or have refractory disease after at least two
ascopubs.org on May
18, 2020: DOI https://
4 months.1 Patients with chemotherapy-refractory previous lines of therapy. Lisocabtagene maraleucel
doi.org/10.1200/ DLBCL, which is defined as a lack of response to has been submitted to the FDA for regulatory approval
EDBK_279043 a second or later line of therapy or progression with in these patients.

Targeted Therapies in Non-Hodgkin Lymphomas
chance of durable remission with a traditional transplant-

based approach, particularly patients who relapse more than
1 year from their initial therapy, but patients with primary
• Novel immunotherapies and small molecules refractory disease, PET-positive disease after salvage therapy,
are transforming the treatment landscape of
early relapsing disease, and MYC-rearranged lymphoma13,14
NHLs.
have an extremely low chance for cure with ASCT, so improved
• Anti-CD19 CAR T cells are potentially curative second-line treatments are desperately needed.10-12,15-20 Anti-
therapy for multiple-relapsed/refractory ag- CD19 CAR T cells are currently being evaluated in patients with
gressive B-cell lymphomas and are under in-
primary refractory or early relapsed aggressive B-cell lym-
vestigation in earlier lines of therapy.
phomas in no fewer than three randomized trials compared
• Bispecific antibodies facilitate T cell mediated with traditional salvage therapy and ASCT (TRANSFORM,
killing of B-cell lymphoma cells using an “off- NCT03575351; BELINDA, NCT03570892; and ZUMA-7,
the-shelf” treatment and appear promising in
NCT03391466). The high and durable rates of CR seen with
relapsed/refractory indolent and aggressive
CAR T cells in more heavily pretreated DLBCL holds promise
B-cell lymphomas.
that these products may well transform the standard of care for
• Chemotherapy-free small molecules are avail- these high-risk patients at first relapse.
able for MCL, including BTKi and lenalidomide,
as well as emerging data for venetoclax. Management of DLBCL at First Relapse:
Transplantation-Ineligible Patients
• Chemotherapy-free small molecules in FL in-
clude lenalidomide and inhibitors of PI3K, with Novel treatment approaches are also needed for patients
EZH2 inhibitors in development. who relapse after frontline therapy and who are not con-
sidered candidates for high-dose chemotherapy and ASCT.
These patients have historically been treated with palliative
The encouraging efficacy observed for these cellular ther- intent regimens such as R-GemOx (rituximab, gemcitabine,
apies in heavily pretreated patients raises the question of oxaliplatin) or bendamustine-rituximab (BR), but durable
whether they should be used earlier in the treatment responses are rare. R-GemOx has been reported to induce
landscape, such as the time of first relapse after frontline complete remissions in 44% of transplantation-ineligible
chemoimmunotherapy, or potentially even as initial therapy subjects, but only 13% remain progression free at 1 year.21
in high-risk patients. In considering the promise of such BR has been widely used in non–transplantation-eligible pa-
approaches, we must consider the performance of currently tients because of its favorable tolerability, but the complete
available standards of care and existing areas of unmet remission rate is low at 15% and the median PFS is less than
medical need. 4 months.22 Additional options are available for select pa-
tient subsets, such as lenalidomide23 or ibrutinib24 in non-
Management of DLBCL at First Relapse: germinal center B-cell (GCB) DLBCL, or pembrolizumab in
Transplantation-Eligible Patients relapsed primary mediastinal B-cell lymphoma,25 but even
Standard rituximab, cyclophosphamide, doxorubicin hydro- among selected patients, the complete remission rates with
chloride (hydroxydaunorubicin), vincristine sulfate (oncovin), these targeted approaches are low, and durable responses
and prednisone (R-CHOP) cures most patients in the front- are uncommon. Anti-CD19 CAR T cells could therefore
line setting, although approximately 30% to 45% of pa- represent an appealing option for patients considered in-
tients will relapse and require second-line therapy.6-8 eligible for ASCT and could provide the first curative intent
Second-line, platinum-based chemotherapy followed by option for these patients who were previously offered only
high-dose chemotherapy with ASCT has long been the palliative intent therapy. Unlike ASCT, there is no upper age
standard of care for young and fit patients based on cure rates limit to undergo CAR T-cell therapy, and the lympho-
of approximately 50% in the pre-rituximab era.9 However, depleting chemotherapy required before CAR T cells is
since the introduction of rituximab as frontline therapy, the less toxic than the high-dose chemotherapy conditioning of
curative potential for traditional salvage regimens and ASCT an ASCT.3-5 Patient selection is still essential, and so can-
at first relapse has declined significantly. Few patients re- didacy for CAR T cells must be informed by performance
lapsing after R-CHOP will achieve CR to second-line regi- status, medical comorbidities, and organ and bone marrow
mens such as R-ICE (rituximab, ifosfamide, carboplatin, function. The TRANSCEND study of lisocabtagene mar-
etoposide), R-DHAP (rituximab, dexamethasone, high-dose aleucel included patients as old as 86 as well as patients
cytarabine, platinum), or R-GDP (rituximab, gemcitabine, with a performance status of 2 and moderate medical
dexamethasone, platinum), and more than 25% of patients comorbidities, including creatinine clearance as low as
will experience a durable progression-free survival (PFS).10-12 30 mL/min and left ventricular ejection fraction as low as
Subsets of patients can be identified who still have a good 40%. 5 These data suggest that anti-CD19 CAR T-cell

Abramson, Ghosh, and Smith
products can be safely administered to transplantation- immunohistochemistry, was identified as prognostic in

ineligible patients, thus validating the investigation of this the pre-rituximab era with patients with ABC/non-GCB
potentially curable treatment as second-line therapy for having significantly inferior outcomes compared with
patients considered ineligible for high-dose chemotherapy. patients with GCB. Although this classification has shown
Lisocabtagene maraleucel is currently being studied in continued prognostic value in many (but not all) series in
a phase II clinical trial for this population (TRANSCEND- the modern treatment era, R-CHOP now appears to cure
PILOT-017006, NCT03483103). Choice of the CAR T cell more than one-half of patients with ABC/non-GCB,6,28-30
warrants consideration in an older or frailer patient population which suggests that CAR T cells would best be reserved
because the incidence and severity of toxicity varies by tfor patients who relapse after conventional frontline
product, although all show similarly encouraging evidence of therapy.
efficacy. Tisagenlecleucel and lisocabtagene maraleucel,
Perhaps no aggressive B-cell lymphoma subset has inspired
both of which incorporate a 4-1BB co-stimulation domain,
as much fear in the last decade as DHL and, to a lesser
are associated with lower rates of any-grade as well as severe
extent, DELs. Initial series of DHLs demonstrated dismal
cytokine release syndrome (CRS) and neurologic toxicity
outcomes in patients treated with R-CHOP, which prompted
relative to the CD28 co-stimulated axicabtagene ciloleucel,
the use of more intensive Burkitt-like treatment strategies.
which may make them preferred for patients in whom severe
However, more recent studies showed DHLs fared better
toxicities would carry higher risks of morbidity or mortality.3-5
than previously believed, likely reflecting selection bias in
The increased toxicity associated with axicabtagene cil-
early series in which patients with aggressive clinical or
oleucel has prompted investigation of earlier introduction of
histologic features were more likely to be tested for the
corticosteroids at the time of grade 1 CRS and neurotoxicity,
presence of a MYC rearrangement. Recent studies showed
which does appear to reduce the progression to severe
that a substantial proportion of patients with DHLs could
toxicity in these patients.26 Therefore, this product also
achieve a CR with intensified treatment regimens such as
warrants ongoing investigation in transplantation-ineligible
DA-EPOCH-R (dose-adjusted etoposide, prednisone, vin-
patients.
cristine, cyclophosphamide, doxorubicin, rituximab) or
R-CODOX-M/IVAC (rituximab, cyclophosphamide, vincristine,
Prospect of CAR T Cells as Initial Therapy
doxorubicin, methotrexate/ifosfamide, etoposide, cytarabine),
R-CHOP has remained the standard of care as initial and that most patients who achieve a complete remission
treatment of DLBCL for more than 20 years based on a high will be cured of their disease.31-34 One of the most com-
rate of success and generally favorable tolerability. In that prehensive analyses to date by the Lunenburg Lymphoma
context, it seems unlikely that the additional logistical bur- Biomarker Consortium evaluated the role of MYC rear-
dens and high costs associated with autologous CAR T cells rangements with or without double-hit cytogenetics in
could be justified in the frontline setting for most patients. a DLBCL population uniformly treated with R-CHOP. Their
However, there are high-risk patients who have lower cure data showed that patients with DHL had an inferior PFS
rates in conventional chemoimmunotherapy in which CAR compared with their non-DHL counterparts, but that most
T cells may be investigated. High-risk populations with ag- patients with DHL were cured with R-CHOP. This challenges
gressive B-cell lymphoma worthy of consideration include the conventional wisdom that R-CHOP is an ineffective
patients with high-risk international prognostic index (IPI) therapy in this population and suggests that R-CHOP would
scores, patients with activated B-cell (ABC; or non-GCB) be an appropriate treatment option in a patient with DHL
subtype DLBCL, and patients with rearrangements and/or with DLBCL morphology. This study did not include patients
protein expression of MYC and BCL2 (so-called double-hit with higher grade morphologic features, which were pre-
lymphomas [DHLs] or double expressor lymphomas [DELs]). viously called Burkitt-like and are now called high-grade
The IPI was initially published in 1993 and continues to be B-cell lymphoma, for whom more intensive treatment
a valuable and reproducible prognostic tool in the modern strategies remain appropriate considerations. Because
treatment era. However, prognosis of DLBCL has improved a large proportion of patients with DHL can enjoy a favorable
dramatically in all IPI risk groups with the incorporation of outcome when treated with chemoimmunotherapy, CAR
rituximab and modern supportive care. Data from the Ger- T cells would again best be reserved for patients with
man High Grade Lymphoma Study Group showed that most suboptimal response to upfront therapy. One possibility
patients with high-risk IPI scores in the current treatment era currently being explored is using anti-CD19 CAR T cells
will be cured with standard chemoimmunotherapy,27 making early in patients with persistent PET-positive disease on an
it unlikely that IPI alone will identify a sufficiently high-risk interim PET/CT scan during initial therapy (ZUMA-12,
subgroup to justify upfront CAR T cells. NCT03761056). However, caution should be taken be-
Cell of origin classification as either ABC or GCB by cause the prognostic value of an interim PET/CT scan
gene expression profiling, or non-GCB versus GCB by has proven controversial. Although some studies have

suggested a significantly worse outcome in patients with B-NHL, clinical use is limited largely by the logistics of a 4- to
interim PET-positive disease, multiple other studies have 8-week continuous infusion required because of its short
found that most patients with interim PET-positive disease half life and the current lack of phase III data. A combination
receiving R-CHOP will still achieve durable complete re- of blinatumomab with lenalidomide in heavily pretreated
mission at the end of therapy.35,36 A randomized clinical relapsed/refractory B-NHL reported an ORR of 91% and CR
trial will therefore be needed to definitively answer the rate of 55%. Grade 3 neurologic toxicity was reported in
question of whether patients with interim PET-positive 28% of patients. The median PFS was 8.3 months for those
disease warrant a change in treatment. patients who received the combination therapy.41
BISPECIFIC ANTIBODIES AND ADCS IN B-CELL LYMPHOMAS Several bispecific CD3 and CD20 antibodies have been
developed and are under investigation in B-NHL (Table 1).
Bispecific Antibodies
These include mosunetuzumab, GEN3013, REGN1979, and
Monoclonal antibodies have revolutionized the treatment of RO7082859. Mosunetuzumab, RO7082859, and GEN3013
B-cell lymphomas since the approval of the anti-CD20 are immunoglobulin-G1 antibodies, whereas REGN1979 is
monoclonal antibody rituximab in 1997. They facilitate an immunoglobulin-G4 molecule. Although most bispecific
apoptosis of cells by targeting one antigen and mediate antibodies are administered as an intravenous infusion,
antibody-dependent cellular cytotoxicity by engaging im- GEN3013 is administered subcutaneously. RO7082859
mune effector cells. Using protein engineering techniques, contains two binding domains for CD20 and a single
researchers have developed bispecific antibodies that can binding domain for CD3 (a 2:1 structure). The additional
bind two target antigens rather than one. By targeting one CD20-binding domain provides enhanced avidity to the
antigen on malignant cells and another on T cells, bispecific molecule. This agent has been combined with obinutuzumab
antibodies can act as a bridge to bring immune effector cells to mitigate CRS. These bispecific antibodies are in early-
in close proximity to malignant cells, which results in cell- phase clinical development. CRS and neurotoxicity can be
mediated cytotoxicity. In addition, T cells undergo activation seen with these agents; however, grade 3 or above CRS and
due to CD3 cross-linking, which is associated with an increase neurotoxicity are uncommon. Use of tocilizumab for CRS has
in T-cell activation markers (CD25 and CD69), cytokine been infrequent in these studies.42-45
release (interferon-gamma, tumor necrosis factor alpha,
One of the major advantages of bispecific antibodies over
interleukin-2, -6, -10) and cytotoxic granule release (gran-
CAR T cells is that they are readily available as off-the-shelf
zyme B). Normally, T cells need an antigen-specific signal
products, unlike autologous cellular therapy products that
and a co-stimulatory signal to mount an effective immune
take weeks to manufacture. Whether these bispecific an-
response. However, bispecific antibodies do not appear to
tibodies will demonstrate long-term durable remissions as
require a co-stimulatory signal to activate T cells. Although the
observed with CAR T cells remains uncertain and requires
mechanism is not fully understood, this could be due to the
longer term follow-up. Because the toxicity associated with
recruitment of memory T cells that need less activation.37
this therapy appears to be lower than that with CAR T cells, it
The first bispecific therapy available in the clinic was bli- could be available to patients who cannot take CAR T-cell
natumomab, which is a bispecific T-cell engager (BiTE). therapy. Moreover, mosunetuzumab has recently shown
One arm of this molecule contains a single-chain variable responses after failure of CAR T-cell therapy, with a few
fragment that recognizes and binds to CD19 on tumor cells, patients (22%) achieving CR.42 The reliance on endogenous
whereas the other arm binds to CD3 on T cells. The two T cells for biologic activity could be a disadvantage in some
domains are joined by a linker. Blinatumomab potently patients. Heavily pretreated patients could have poor T-cell
triggers the signaling cascade of the T-cell receptor complex health and could make this immunotherapy less efficacious.
only when both of its binding sites are occupied.38 Blina- Antigen escape is another mechanism of resistance to
tumomab is FDA approved for the treatment of relapsed/ bispecific antibodies. Combination of bispecific antibodies
refractory B-cell precursor acute lymphoblastic leukemia with other agents and development of next-generation
and has demonstrated modest activity in relapsed/refractory bispecific antibodies that can target multiple antigens or
B-cell NHL (B-NHL). A phase I study found an overall re- incorporate co-stimulatory molecules may further enhance
sponse rate (ORR) of 69% and a CR rate of 37% in patients the potency of these agents.
treated at the dose of 60 μg/m2/d.39 The median duration of
response (DOR) for patients who achieved a CR was 17 ADCs
months. In patients with DLBCL, an ORR of 43% and a CR ADCs exploit the specific binding properties of monoclonal
rate of 19% was reported with a median PFS of 3.7 antibodies as a mechanism for selective delivery of cytotoxic
months. 40 Neurologic adverse events, as observed with agents to tumor cells. The three essential components of an
other T-cell activating therapies, were dose-limiting in 22% ADC—the antibody, cytotoxic payload, and covalent linker
of patients. Although blinatumomab has modest activity in that joins them together, as well as the biologic properties of

TABLE 1. Preliminary Results From Selected Phase I Studies of Bispecific Antibodies in B-NHL
Name Dose (mg) Efficacy CRS Neurotoxicity
Mosunetuzumab42 2.8–40.5 Aggressive NHL (n = 124) 29% (all grades) 44% (all grades)
ORR, 37%; CR, 19% 1% (grade 3) 4% (grade 3)
2.8–13.5 Indolent NHL (n = 67)
ORR, 63%; CR, 43%
REGN197943 80–320 Aggressive NHL (n = 19) 59% (all grades) Data not available
ORR, 58%; CR, 42% 6% (grade 3)
5–320 Indolent NHL (n = 22)
ORR, 95%; CR, 77%
GEN301344 0.76–6 Aggressive NHL (n = 13) 48% (all grades) No events
ORR, 15%; CR, 8% No grade 3
0.76–6 Indolent NHL (n = 5)
ORR, 100%; CR, 0%
RO7082859 plus obinutuzumab45 0.6–16 Aggressive NHL (n = 16) 57% (all grades) 25% (all grades)
ORR, 38%; CR, 31% 8% ( grade 3) 4% (grade 3)
0.6–16 Indolent NHL (n = 6)
ORR, 80%; CR, 80%
Abbreviations: B-NHL, B-cell non-Hodgkin lymphoma; CR, complete response; CRS, cytokine release syndrome; Gr, grade; NHL, non-Hodgkin
lymphoma; ORR, overall response rate.
the cell-surface target antigen—are important in designing lymphoma, in anaplastic large-cell lymphoma (ALCL), and
an effective agent that can result in target cell death. in subsets of DLBCL and peripheral T-cell lymphoma
(PTCL). BV in combination with chemotherapy has been
The target antigen should be highly expressed on the
malignant cells with minimal expression elsewhere. The approved for previously untreated systemic ALCL or other
antibody preferably has high affinity and is humanized or CD30-expressing PTCL, including angioimmunoblastic
fully human, which leads to internalization of the antigen- T-cell lymphoma and PTCL not otherwise specified. Ap-
antibody complex. The linker must be stable to prevent proval was based on the results of the ECHELON-2 trial,
systemic release of the cytotoxic payload but must also be which randomly assigned patients to receive BV plus cy-
labile to allow release of the payload in the malignant cell. clophosphamide, doxorubicin, and prednisone (CHP) or
Conjugation focuses on the payload-to-antibody ratio. If too CHOP.46 The median PFS was 48.2 months (95% CI,
little payload is conjugated to the antibody, potency can be 35.2–not estimable) for patients on the BV plus CHP arm and
lost; however, if there is too much, it can result in increased 20.8 months (95% CI, 12.7–47.6) for those on the CHOP
toxicity The two major types of payloads are (1) antimitotic arm (hazard ratio [HR], 0.71; 95% CI, 0.54–0.93; p = .011).
payloads (e.g., auristatins, monomethyl auristatin E [MMAE] The clinical trial also demonstrated improvement in OS (HR,
and monomethyl auristatin F [MMAF]), and (2) DNA- 0.66; 95% CI, 0.46 0.95; p = .024), CR rate (68% vs. 56%;
binding payloads (calicheamicin and pyrrolobenzodiaze- p = .007), and ORR (83% vs. 72%; p = .003). BV as a single
pine dimers). The antimitotic payloads bind to tubulin, agent is approved for relapsed/refractory systemic ALCL,
resulting in G2/M arrest and subsequent apoptosis. They are primary cutaneous ALCL, and CD30-expressing mycosis
frequently associated with peripheral neuropathy. The DNA- fungoides that have received previous systemic therapy. In
binding payloads induce double-stranded DNA breaks. relapsed/refractory CD30+ DLBCL, single-agent BV has an
Calicheamicin payloads are associated with thrombocyto- ORR of 44%.47 BV-RCHP is being tested for the frontline
penia and sinusoidal obstructive syndrome, especially in treatment of CD30+ DLBCL.
patients undergoing stem cell transplantation. PoV is an ADC consisting of an anti-CD79b antibody site-
Two ADCs, brentuximab vedotin (BV) and polatuzumab specifically conjugated to MMAE. CD79b expression is
vedotin (PoV), are currently FDA approved for treatment of limited to mature B cells and is a component of the BCR
NHL. BV consists of an anti-CD30 antibody, a cleavable complex that drives receptor signaling. PoV has been ap-
linker, and an MMAE payload. CD30 is a member of the proved for the treatment of relapsed/refractory DLBCL in
tumor necrosis factor receptor superfamily and is primarily combination with bendamustine and rituximab (BR) after
expressed in the Reed-Sternberg cells in classic Hodgkin more than two previous therapies. Approval was based on

a randomized clinical trial that included 80 patients with rational targets, chemotherapy-free regimens avoid the
relapsed/refractory DLBCL. 48 Patients were randomly typical toxicities associated with chemotherapy, although
assigned 1:1 to receive either PoV in combination with BR or they carry their own unique array of effects. The onus of
BR alone for six cycles. Patients who received PoV-BR had chemotherapy-free therapy is to provide a high-quality re-
a significantly higher CR rate (40.0% vs. 17.5%; p = .026) mission anddurable response, improve the duration and
and a longer median PFS (9.5 vs. 3.7 months; HR, 0.36; quality of life with the least amount of short- and long-term
95% CI, 0.21 0.63; p , .001) and OS (median 12.4 vs. 4.7 toxicity, and perhaps move us closer to cure where che-
months; HR, 0.42; 95% CI, 0.24 0.75; p = .002). Due to motherapy has failed.
the clinical activity in relapsed/refractory DLBCL, a large
randomized, double-blind, placebo-controlled study, the
MCL: Initial Therapy
POLARIX trial, randomly assigned patients with newly di-
agnosed DLBCL to receive R-CHOP versus PoV-RCHP. The MCL is an uncommon mature B-cell lymphoma with unique
study has completed accrual, with results pending. PoV is clinicopathologic features, including a skewed male pre-
also being tested in combination with lenalidomide, ven- disposition, increased incidence with age, and character-
etoclax, and checkpoint inhibitors in relapsed/refractory istic findings of t(11;14) with cyclin D1 overexpression.
DLBCL. Typically considered a more aggressive disease, up to one-
third of patients with MCL have an indolent presentation with
Other target antigens for ADCs under evaluation include
two variants: a leukemic version reminiscent of chronic
CD19, CD22, CD25, and CD27. Inotuzumab ozogamicin is
lymphocytic leukemia, and a low tumor burden version with
targeted to cells expressing CD22 and was approved by the
disseminated adenopathy in an asymptomatic patient. The
FDA in 2017 for relapsed/refractory precursor B-cell acute
leukemic variant is recognized by the World Health Orga-
lymphoblastic leukemia. It is currently being evaluated in
nization and has distinct clinicopathologic features with
relapsed/refractory DLBCL and FL. However, a randomized
splenomegaly, circulating lymphocytes, and SOX11 nega-
comparison of inotuzumab plus rituximab versus rituximab
tivity.52 Low tumor burden MCL has a low Ki67 and a slow
with investigator choice chemotherapy did not show any
clinical course.53 For both of these groups, a period of active
survival advantage for the inotuzumab-containing regi-
observation is appropriate, and aggressive intervention can
men.49 Toxicities include myelosuppression, nausea, and
be safely deferred.
fatigue. Loncastuximab tesirine, a pyrrolobenzodiazepine-
based CD19-targeted ADC, is currently being evaluated in Chemotherapy-free approaches have been mainly de-
patients with relapsed/refractory B-NHL50 with a reported veloped for the latter group, although one could argue that
ORR of 59% and CR rate of 41%. Median DOR, PFS, and success with a chemotherapy-free approach could or
OS were 4.8, 5.5, and 11.6 months, respectively. Adverse should be offered to fit patients with an otherwise incurable
events include fatigue, edema, liver test abnormalities, disease. However, no chemotherapy-free regimens are cur-
nausea, rash, and dyspnea. Camidanlumab terisine, rently FDA approved in the frontline setting. The three main
a CD25-targeted ADC, is being investigated in an ongoing families of agents tested thus far include anti-CD20 mono-
phase I/II trial in relapsed/refractory NHL.51 Toxicities in- clonal antibodies, BTKi, and immunomodulatory agents, al-
clude diarrhea, fatigue, rash, edema, thyroiditis, and ery- though venetoclax trials are also underway.
thema multiforme. The ORR for patients with PTCL treated A multicenter phase II trial tested lenalidomide plus ritux-
with at least 60 μg/kg was 50% (10 patients). imab in 38 patients with newly diagnosed MCL.54 Patients
By harnessing the antibody-antigen interactions to deliver were older (median age, 65), and one-third had a high-risk
cytotoxins preferentially to tumor cells, ADCs can maximize Mantle Cell Lymphoma International Prognostic Index
efficacy and minimize toxicity. Advancements in linker (MIPI) score. Only one-half of the patients had symptomatic
technology and conjugation strategies could lead to the disease, and Ki67 was low in 68%. Treatment consisted of
development of more potent ADCs with lower toxicity. rituximab weekly for cycle 1 and then every other month
plus lenalidomide 20 mg/day for 21 days on a 28-day cycle,
CHEMOTHERAPY-FREE APPROACHES IN FL AND MCL repeated for 12 cycles before starting a maintenance pro-
Targeted small molecule therapies are now routinely used gram for both agents without a clear end date. The ORR was
for treatment of relapsed/refractory FL and MCL, and are 92% with a CR rate more than 60% based on PET/CT. A
increasingly being evaluated in lieu of chemotherapy in the recent update showed 5-year estimated PFS and OS of 64%
frontline setting. Although FL and MCL are clearly different and 77%, respectively (Table 2).55 The regimen was well
diseases, they share the unfortunate distinction of being tolerated with no unexpected adverse effects. A compara-
incurable. Chemotherapy can provide quick and deep re- tive trial against chemoimmunotherapy is needed to de-
sponses, but there is progressive resistance by line of cy- finitively evaluate the role of lenalidomide/rituximab in this
totoxic therapy, with eventual treatment failure. By attacking population, but the favorable toxicity profile makes this an

appealing regimen and possibly a backbone for the addition active endeavor, although there are cautionary examples of
of other agents. this triplet conferring excess toxicity60; we await further data.
In addition to these categories of agents, the BCL2-targeted
Because of the success of BTKi in the relapsed setting,
BH3 mimetic venetoclax is also being tested in the frontline
there is interest in bringing this front line as well. Some
setting in combination strategies that are chemotherapy
studies are combining chemotherapy backbones with or
free. A major unanswered question in the frontline setting is
without a BTKi (A Study of Bendamustine and Rituximab
the implication of TP53 mutation, which occurs in ap-
Alone Versus in Combination With Acalabrutinib in Sub-
proximately 10% of patients at diagnosis.61 This is fortu-
jects With Previously Untreated Mantle Cell Lymphoma,
nately uncommon in newly diagnosed patients but presents
NCT02972840; A Study of the Bruton’s Tyrosine Kinase
an opportunity for nonchemotherapy regimens to be
Inhibitor Ibrutinib Given in Combination With Bendamustine
developed.
and Rituximab in Patients With Newly Diagnosed Mantle
Cell Lymphoma, NCT01776840), but there are also several MCL: Management of Relapsed/Refractory Disease
chemotherapy-free regimens in development. Jain and Despite differences in DOR and PFS, essentially all patients
colleagues reported on a single-institution phase II trial of with MCL will experience disease recurrence. Disease
ibrutinib plus rituximab in older adult patients with low risk of progression is associated with adverse biologic events, in-
disease. Among 49 patients, the median age was 71, 76% cluding frequent acquisition of TP53 mutations and cyto-
of patients had low Ki67, and there were few patients with genetic complexity. Clinically, many patients have rapid
high-risk MIPI scores.56 Treatment included induction disease progression, sometimes with central nervous sys-
(ibrutinib 560 mg/day plus rituximab weekly for cycle 1 and tem involvement. Chemotherapy alone has a limited role in
then monthly through cycle 8) and maintenance (ongoing relapsed disease and is typically a bridge to hematopoietic
ibrutinib, rituximab changed to every other month) with all stem cell transplantation.
therapy ending by 2 years. The ORR was 98% with 60%
The pivotal trial of ibrutinib monotherapy in relapsed/
complete remission. Follow-up was relatively short, and
refractory MCL was a major step forward,62 and adding other
there was promising activity, but nearly 40% of patients had
agents to BTKi is the subject of numerous ongoing trials.
to stop treatment due to adverse effects, including atrial
Despite a high ORR of 68%, complete remissions were
fibrillation and bleeding. In younger patients, the phase II
much less common at 21%; 24-month PFS and OS were
WINDOW trial (NCT02427620) evaluated up to 1 year of
31% and 47%, respectively. 63 In addition, despite im-
ibrutinib-rituximab followed by chemotherapy consolidation
pressive single-agent activity, rapid disease progression
with abbreviated R-HCVAD/R-MTX (rituximab, hyperfractionated
occurred eventually and was associated with high mortal-
cyclophosphamide, vincristine, doxorubicin, dexametha-
ity.64 Second-generation BTKi such as acalabrutinib and
sone/rituximab, methotrexate) and reported similar high
zanubrutinib were approved and appeared to have similar
efficacy of the chemotherapy-free component.59
monotherapy activity to ibrutinib,65,66 but their roles were
Combining all three categories of agents (BTKi, anti-CD20, likely as a replacement for ibrutinib rather than a salvage
immunomodulatory agents) in the frontline setting is an approach after ibrutinib failure. In contrast to chronic
lymphocytic leukemia, for which monotherapy BTKi ap-
TABLE 2. Chemotherapy-Sparing Frontline Treatment Regimens in proaches are highly successful, combination strategies are
Mantle Cell and Follicular Lymphomas needed in MCL.
Number
of
The addition of rituximab to ibrutinib improved the CR rate
Regimen Indication Patients ORR/CRR PFS as reported in a phase II trial.67 This single-center trial re-
ported a CR rate of 58%, which was higher than that re-
Lenalidomide- Mantle cell 38 92%/64% 5 years:
rituximab55 64% ported in the original ibrutinib monotherapy trial. With
4 years of follow-up, the median PFS was 43%, although
Ibrutinib- Mantle cell 49 98%/60% Median
rituximab56 (low risk) not activity was more modest in higher risk patients (TP53
reached mutations, higher Ki67). In contrast, when lenalidomide was
at 24 added to ibrutinib-rituximab, investigators found no differ-
months ence based on TP53 mutation status,68 which suggested
Lenalidomide- Follicular 513 61%/48% 3 years: that this might overcome adverse biologic features. Un-
rituximab57 77% fortunately, toxicity was increased with the triplet, and the
Lenalidomide- Follicular 90 98%/92% 2 years: overall and CR rates for the entire group were not com-
obinutuzumab58 96% pellingly different compared with doublet therapy.
Abbreviations: CRR, complete response rare; ORR, overall response Venetoclax plus ibrutinib has promising early results in
rate; PFS, progression-free survival. relapsed/refractory MCL. As monotherapy, venetoclax was

superior in ibrutinib-naı̈ve patients compared with patients (LO) found a CR rate of 92% with 2-year PFS of 96%.58
whose disease progressed after ibrutinib, in whom median Toxicity appears comparable to LR, and we await more
PFS was only 3 months.69 Tam and colleagues combined mature data on this combination. Triplet chemotherapy-free
ibrutinib and venetoclax in a high-risk population of patients combinations in the frontline setting have thus far been
with MCL, in whom one-half of the patients had aberrations quite toxic.60 There are also combination trials of PI3K
of TP53 and 75% had a high-risk prognostic score.70 inhibitors as frontline therapy (Response-Adapted Therapy
Among 24 patients, 42% achieved MRD negativity, hold- With Copanlisib and Rituximab in Untreated Follicular
ing promise that this combination might be a good option for Lymphoma, NCT03789240; CONTEMPO, NCT02391545;
some patients. Study to Assess Umbralisib Plus Ublituximab in Patients With
Overall, the management of relapsed/refractory MCL re- Treatment Naı̈ve Follicular Lymphoma, NCT03828448).
mains unsatisfactory. Each of these chemotherapy-free
FL: Management of Relapsed Disease
regimens is appealing, but to date they have been tested
entirely in phase II settings and with relatively small patient Relapsed/refractory FL is a heterogeneous disease. Out-
numbers. Patient selection may account for some of the comes are heavily influenced by patient- and disease-related
differences in these phase I trials, and only a prospective factors, including time to relapse, comorbidities, age, tumor
comparison with balanced arms will answer which com- burden, and likelihood of transformation at the time of re-
bination is superior to palliative monotherapy with BTKi. lapse. There are several chemotherapy-free agents approved,
although with few exceptions, these are monotherapy ap-
FL: Initial Therapy
proaches in the relapsed setting. Major categories of non-
As the prototypical indolent lymphoma, FL is increasingly chemotherapeutic agents approved or in development
viewed as a chronic disease with prolonged survival ap- include anti-CD20 monoclonal antibodies (including
proximating 2 decades in many patients.71 The limitations of radioimmunotherapy), PI3K delta inhibitors, and EZH2
chemotherapy are clear in that high initial activity does not inhibitors, among others.
translate into cure, making this disease an appealing area
for the development of chemotherapy-free approaches. Similar to the frontline setting, LR has been tested in the
relapsed/refractory setting with high efficacy and good
Based on data supporting the importance of the tumor tolerability. The AUGMENT trial is a randomized phase III
microenvironment in FL, the immunomodulatory agent trial of LR versus placebo-rituxumab in patients with re-
lenalidomide has been extensively evaluated in relapsed lapsed/refractory indolent lymphomas, with most patients
and frontline settings. In the frontline setting, both single- having FL.75 The regimen was limited to 12 months of
center and cooperative group trials show ORRs in excess of therapy, and the primary endpoint was PFS. Among 358
90% with CR rates of 60% to 70% for lenalidomide and randomly selected patients, LR had significantly improved
rituximab (LR),72,73 setting the stage for the recently pub- PFS and DOR (median 39.4 vs. 14.1 months). There was
lished RELEVANCE trial.57 RELEVANCE is a randomized more toxicity in the combination arm, but no increase in
phase III study comparing chemoimmunotherapy plus grade 3 to5 adverse events. The MAGNIFY trial tested the
maintenance rituximab against LR. More than 1,000 pa- optimal duration of LR, with all patients receiving 12 cycles
tients were randomly assigned, with the key results showing of treatment followed by random assignment between
equivalent complete remission rates (48% in the LR group continued LR versus maintenance rituximab.76 The initial
vs. 53% in the rituximab-chemotherapy group; p = .13) and phase II component was recently reported with a 73% ORR
similar 3-year PFS of 77% versus 78%, repectively. Toxicity and 45% CR rate with median PFS of 36 months.76 Building
profiles differed, but there was no difference in the fre- on LR as a backbone has been challenging. Both PI3K
quency of adverse events, leading to equipoise between the inhibitors and BTKi have been added to LR, with surprising
two arms. Although the RELEVANCE trial selected a 24- toxicity in indolent lymphomas.60,77 BTKi have no role in FL
month regimen, others have shown that shorter duration of to date, with monotherapy activity that is quite dismal.78,79 A
treatment may be effective as well. The CALGB/Alliance particularly high-risk group with no clear standard of care
trial delivered LR for only 12 months, and a recently pub- has been dubbed the POD24 population: patients pro-
lished phase II trial showed encouraging activity with only gressing within 24 months of initial chemoimmunotherapy
18 weeks of LR.74 Overall, there is ample evidence that LR is who have only a 50% 5-year predicted OS.80 The optimal
an active initial chemotherapy-free regimen and is now treatment of this group of patients is not established, but
listed in national guidelines. poor survival occurred in an era when only chemotherapy
There are two approaches building off the LR regimen: re- options were available. With this in mind, there is a United
place rituximab with obinutuzumab or add a third agent to States intergroup trial comparing LO versus umbralisib-
LR. Preliminary data in 90 patients with high-tumor burden obinutuzumab versus chemoimmunotherapy specifically
treatment-naı̈ve FL treated with lenalidomide-obinutuzumab in this patient population (Obinutuzumab With or Without

Umbralisib, Lenalidomide, or Combination Chemotherapy is modest, with only 5% of patients stopping treatment
in Treating Patients With Relapsed or Refractory Grade I-IIIa because of toxicity. Combination trials are ongoing, but this
Follicular Lymphoma, NCT03269669). Other trials in this is an exciting new target in FL.
space include the FRESCO trial (NCT02605694), combining CONCLUSIONS
duvelisib plus rituximab versus chemoimmunotherapy.
Novel treatment strategies are adding to our treatment ar-
PI3K inhibitors have significant activity in relapsed/ mamentarium across the spectrum of NHLs, improving
refractory FL, and three are currently approved for use in outcomes for countless patients. These include immuno-
the United States: idelalisib, copanlisib, and duvelisib. therapies like CAR T cells, bispecific antibodies, ADCs, and
Umbralisib, a highly selective PI3K delta inhibitor, met its immunomodulatory drugs, as well as small molecules tar-
primary endpoint in late 2019, and these data have been geting BTK, PI3K, BCL2, EZH2, and others. Although ini-
submitted for FDA approval. It is noteworthy that each of tially available as single agents in the relapsed/refractory
these agents were tested as monotherapy in extremely setting, these agents are increasingly being evaluated in
poor-risk relapsed/refractory indolent lymphoma; there are combinations and in the frontline treatment setting where
some differences in the details of the enrolled population, they may replace chemotherapy in certain clinical settings.
but monotherapy ORRs are between 50% and 60% with A recurrent theme in the development of noncytotoxic reg-
CR rates less than 20%.81-83 The first-generation PI3K imens is that chemotherapy free does not equal toxicity-free.
delta inhibitor idelalisib could not safely be combined with There are a number of distinct adverse effects seen with
LR,77 but there are ongoing combination trials with copan- immunotherapies and combinations of targeted agents, and
lisib, duvelisib, and umbralisib. future drug development needs to account for off-target ef-
Epigenetic deregulation occurs early in the pathogenesis of fects, drug-drug interactions, and the physical and economic
FL, with EZH2 inhibitors now in trials. Tazemetostat has consequences of these treatments. Progress has certainly
activity in both mutant and wild-type EZH2 lymphomas, been made, and we increasingly have chemotherapy-sparing
although activity is higher in the patient population with treatments available, with the hope that cure may be closer in
mutations with response rates greater than 60%.84 Toxicity reach for increasing numbers of patients.

1
Massachusetts General Hospital, Boston, MA AND DATA AVAILABILITY STATEMENT
2
Levine Cancer Institute, Atrium Health, Charlotte, NC Disclosures provided by the authors and data availability statement (if
3
University of Chicago, Chicago, IL applicable) are available with this article at DOI https://doi.org/10.1200/
EDBK_279043.
Jeremy S. Abramson, MD, MMSC, Massachusetts General Hospital, 55
Fruit St., Boston, MA 02114; email: [email protected].
References
1. Van Den Neste E, Schmitz N, Mounier N, et al. Outcome of patients with relapsed diffuse large B-cell lymphoma who fail second-line salvage regimens in the
International CORAL study. Bone Marrow Transplant. 2016;51:51-57.
2. Crump M, Neelapu SS, Farooq U, et al. Outcomes in refractory diffuse large B-cell lymphoma: results from the international SCHOLAR-1 study. Blood. 2017;
130:1800-1808.
3. Locke FL, Ghobadi A, Jacobson CA, et al. Long-term safety and activity of axicabtagene ciloleucel in refractory large B-cell lymphoma (ZUMA-1): a single-arm,
multicentre, phase 1-2 trial. Lancet Oncol. 2019;20:31-42.
4. Schuster SJ, Bishop MR, Tam CS, et al; JULIET Investigators. Tisagenlecleucel in adult relapsed or refractory diffuse large B-cell lymphoma. N Engl J Med. 2019;
380:45-56.
5. Abramson JS, Palomba ML, Gordon LI, et al. Pivotal safety and efficacy results from transcend NHL 001, a multicenter phase 1 study of lisocabtagene maraleucel
(liso-cel) in relapsed/refractory (R/R) large B cell lymphomas. Blood. 2019;134 (Supplement_1):241.
6. Vitolo U, Trněný M, Belada D, et al. Obinutuzumab or rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone in previously untreated diffuse
large B-cell lymphoma. J Clin Oncol. 2017;35:3529-3537.
7. Stephens DM, Li H, LeBlanc ML, et al. Continued risk of relapse independent of treatment modality in limited-stage diffuse large B-Cell lymphoma: final and long-
term analysis of Southwest Oncology Group Study S8736. J Clin Oncol. 2016;34:2997-3004.

8. Sant M, Minicozzi P, Mounier M, et al; EUROCARE-5 Working Group. Survival for haematological malignancies in Europe between 1997 and 2008 by region and
age: results of EUROCARE-5, a population-based study. Lancet Oncol. 2014;15:931-942.
9. Philip T, Armitage JO, Spitzer G, et al. High-dose therapy and autologous bone marrow transplantation after failure of conventional chemotherapy in adults with
intermediate-grade or high-grade non-Hodgkin’s lymphoma. N Engl J Med. 1987;316:1493-1498.
10. Gisselbrecht C, Glass B, Mounier N, et al. Salvage regimens with autologous transplantation for relapsed large B-cell lymphoma in the rituximab era. J Clin Oncol.
2010;28:4184-4190.
11. Crump M, Kuruvilla J, Couban S, et al. Randomized comparison of gemcitabine, dexamethasone, and cisplatin versus dexamethasone, cytarabine, and cisplatin
chemotherapy before autologous stem-cell transplantation for relapsed and refractory aggressive lymphomas: NCIC-CTG LY.12. J Clin Oncol. 2014;
32:3490-3496.
12. van Imhoff GW, McMillan A, Matasar MJ, et al. Ofatumumab versus rituximab salvage chemoimmunotherapy in relapsed or refractory diffuse large B-cell
lymphoma: The ORCHARRD Study. J Clin Oncol. 2017;35:544-551.
13. Thieblemont C, Briere J, Mounier N, et al. The germinal center/activated B-cell subclassification has a prognostic impact for response to salvage therapy in
relapsed/refractory diffuse large B-cell lymphoma: a bio-CORAL study. J Clin Oncol. 2011;29:4079-4087.
14. Epperla N, Maddocks KJ, Salhab M, et al. C-MYC-positive relapsed and refractory, diffuse large B-cell lymphoma: impact of additional “hits” and outcomes with
subsequent therapy. Cancer. 2017;123:4411-4418.
15. Hitz F, Connors JM, Gascoyne RD, et al. Outcome of patients with primary refractory diffuse large B cell lymphoma after R-CHOP treatment. Ann Hematol. 2015;
94:1839-1843.
16. Sauter CS, Matasar MJ, Meikle J, et al. Prognostic value of FDG-PET prior to autologous stem cell transplantation for relapsed and refractory diffuse large B-cell
lymphoma. Blood. 2015;125:2579-2581.
17. Armand P, Welch S, Kim HT, et al. Prognostic factors for patients with diffuse large B cell lymphoma and transformed indolent lymphoma undergoing autologous
stem cell transplantation in the positron emission tomography era. Br J Haematol. 2013;160:608-617.
18. Winter A, Rybicki L, Shah SN, et al. Prognostic value of pre-transplant PET/CT in patients with diffuse large B-cell lymphoma undergoing autologous stem cell
transplantation. Leuk Lymphoma. 2018;59:1195-1201.
19. Vose JM, Carter S, Burns LJ, et al. Phase III randomized study of rituximab/carmustine, etoposide, cytarabine, and melphalan (BEAM) compared with iodine-131
tositumomab/BEAM with autologous hematopoietic cell transplantation for relapsed diffuse large B-cell lymphoma: results from the BMT CTN 0401 trial. J Clin
Oncol. 2013;31:1662-1668.
20. Vardhana SA, Sauter CS, Matasar MJ, et al. Outcomes of primary refractory diffuse large B-cell lymphoma (DLBCL) treated with salvage chemotherapy and
intention to transplant in the rituximab era. Br J Haematol. 2017;176:591-599.
21. Mounier N, El Gnaoui T, Tilly H, et al. Rituximab plus gemcitabine and oxaliplatin in patients with refractory/relapsed diffuse large B-cell lymphoma who are not
candidates for high-dose therapy. A phase II Lymphoma Study Association trial. Haematologica. 2013;98:1726-1731.
22. Vacirca JL, Acs PI, Tabbara IA, et al. Bendamustine combined with rituximab for patients with relapsed or refractory diffuse large B cell lymphoma. Ann Hematol.
2014;93:403-409.
23. Hernandez-Ilizaliturri FJ, Deeb G, Zinzani PL, et al. Higher response to lenalidomide in relapsed/refractory diffuse large B-cell lymphoma in nongerminal center
B-cell-like than in germinal center B-cell-like phenotype. Cancer. 2011;117:5058-5066.
24. Wilson WH, Young RM, Schmitz R, et al. Targeting B cell receptor signaling with ibrutinib in diffuse large B cell lymphoma. Nat Med. 2015;21:922-926.
25. Zinzani PL, Ribrag V, Moskowitz CH, et al. Safety and tolerability of pembrolizumab in patients with relapsed/refractory primary mediastinal large B-cell
lymphoma. Blood. 2017;130:267-270.
26. Topp M, Van Meerten T, Houot R, et al. Earlier steroid use with axicabtagene ciloleucel (Axi-Cel) in patients with relapsed/refractory large B cell lymphoma. Blood.
2019;134 (Supplement_1):243.
27. Ziepert M, Hasenclever D, Kuhnt E, et al. Standard International prognostic index remains a valid predictor of outcome for patients with aggressive CD20+ B-cell
lymphoma in the rituximab era. J Clin Oncol. 2010;28:2373-2380.
28. Younes A, Sehn LH, Johnson P, et al; PHOENIX investigators. Randomized phase III trial of ibrutinib and rituximab plus cyclophosphamide, doxorubicin,
vincristine, and prednisone in non-germinal center B-cell diffuse large B-cell lymphoma. J Clin Oncol. 2019;37:1285-1295.
29. Davies A, Cummin TE, Barrans S, et al. Gene-expression profiling of bortezomib added to standard chemoimmunotherapy for diffuse large B-cell lymphoma
(REMoDL-B): an open-label, randomised, phase 3 trial. Lancet Oncol. 2019;20:649-662.
30. Staiger AM, Ziepert M, Horn H, et al; German High-Grade Lymphoma Study Group. Clinical impact of the cell-of-origin classification and the MYC/BCL2 dual
expresser status in diffuse large B-cell lymphoma treated within prospective clinical trials of the German High-Grade Non-Hodgkin’s Lymphoma Study Group.
J Clin Oncol. 2017;35:2515-2526.
31. McPhail ED, Maurer MJ, Macon WR, et al. Inferior survival in high-grade B-cell lymphoma with MYC and BCL2 and/or BCL6 rearrangements is not associated
with MYC/IG gene rearrangements. Haematologica. 2018;103:1899-1907.
32. Copie-Bergman C, Cuillière-Dartigues P, Baia M, et al. MYC-IG rearrangements are negative predictors of survival in DLBCL patients treated with immu-
nochemotherapy: a GELA/LYSA study. Blood. 2015;126:2466-2474.
33. Landsburg DJ, Falkiewicz MK, Maly J, et al. Outcomes of patients with double-hit lymphoma who achieve first complete remission. J Clin Oncol. 2017;
35:2260-2267.

34. Petrich AM, Gandhi M, Jovanovic B, et al. Impact of induction regimen and stem cell transplantation on outcomes in double-hit lymphoma: a multicenter
retrospective analysis. Blood. 2014;124:2354-2361.
35. Gallamini A, Zwarthoed C. Interim FDG-PET imaging in lymphoma. Semin Nucl Med. 2018;48:17-27.
36. Mamot C, Klingbiel D, Hitz F, et al. Final results of a prospective evaluation of the predictive value of interim positron emission tomography in patients with diffuse
large B-cell lymphoma treated with R-CHOP-14 (SAKK 38/07). J Clin Oncol. 2015;33:2523-2529.
37. Huehls AM, Coupet TA, Sentman CL. Bispecific T-cell engagers for cancer immunotherapy. Immunol Cell Biol. 2015;93:290-296.
38. Löffler A, Kufer P, Lutterbüse R, et al. A recombinant bispecific single-chain antibody, CD19 x CD3, induces rapid and high lymphoma-directed cytotoxicity by
unstimulated T lymphocytes. Blood. 2000;95:2098-2103.
39. Goebeler ME, Knop S, Viardot A, et al. Bispecific T-cell engager (BiTE) antibody construct blinatumomab for the treatment of patients with relapsed/refractory
non-Hodgkin lymphoma: final results from a phase I study. J Clin Oncol. 2016;34:1104-1111.
40. Viardot A, Goebeler ME, Hess G, et al. Phase 2 study of the bispecific T-cell engager (BiTE) antibody blinatumomab in relapsed/refractory diffuse large B-cell
lymphoma. Blood. 2016;127:1410-1416.
41. Poh C, Frankel P, Ruel C, et al. Blinatumomab/lenalidomide in relapsed/refractory non-Hodgkin’s lymphoma: a phase I California Cancer Consortium Study of
Safety, Efficacy and Immune Correlative Analysis. Blood. 2019;134 (Supplement_1):760.
42. Schuster SJ, Bartlett NL, Assouline S, et al. Mosunetuzumab induces complete remissions in poor prognosis non-Hodgkin lymphoma patients, including those
who are resistant to or relapsing after chimeric antigen receptor T-Cell (CAR-T) therapies, and is active in treatment through multiple lines. Blood. 2019;134
(Supplement_1):6.
43. Bannerji R, Allan JN, Arnason JE, et al. Clinical activity of REGN1979, a bispecific human, anti-CD20 x anti-CD3 antibody, in patients with relapsed/refractory (R/
R) B-cell non-Hodgkin lymphoma (B-NHL). Blood. 2019;134 (Supplement_1):762.
44. Lugtenburg P, Mous R, Clausen MR, et al. First-in-human, phase 1/2 trial to assess the safety and clinical activity of subcutaneous GEN3013 (DuoBody®-
CD3CD20) in B-cell non-Hodgkin lymphomas. Blood. 2019;134 (Supplement_1):758.
45. Morschhauser F, Carlo-Stella C, Offner F, et al. Dual CD20-targeted therapy with concurrent CD20-TCB and obinutuzumab shows highly promising clinical
activity and manageable safety in relapsed or refractory B-cell non-Hodgkin lymphoma: preliminary results from a phase Ib trial. Blood. 2019;134 (Supplement_
1):1584.
46. Horwitz S, O’Connor OA, Pro B, et al; ECHELON-2 Study Group. Brentuximab vedotin with chemotherapy for CD30-positive peripheral T-cell lymphoma
(ECHELON-2): a global, double-blind, randomised, phase 3 trial. Lancet. 2019;393:229-240.
47. Jacobsen ED, Sharman JP, Oki Y, et al. Brentuximab vedotin demonstrates objective responses in a phase 2 study of relapsed/refractory DLBCL with variable
CD30 expression. Blood. 2015;125:1394-1402.
48. Sehn LH, Herrera AF, Flowers CR, et al. Polatuzumab vedotin in relapsed or refractory diffuse large B-cell lymphoma. J Clin Oncol. 2020;38:155-165.
49. Dang NH, Ogura M, Castaigne S, et al. Randomized, phase 3 trial of inotuzumab ozogamicin plus rituximab versus chemotherapy plus rituximab for relapsed/
refractory aggressive B-cell non-Hodgkin lymphoma. Br J Haematol. 2018;182:583-586.
50. Kahl BS, Hamadani M, Radford J, et al. A phase I study of ADCT-402 (loncastuximab tesirine), a novel pyrrolobenzodiazepine-based antibody-drug conjugate, in
relapsed/refractory B-cell non-Hodgkin lymphoma. Clin Cancer Res. 2019;25:6986-6994.
51. Collins GP, Horwitz SM, Davies A, et al. Adct-301 (Camidanlumab Tesirine), a novel pyrrolobenzodiazepine-based CD25-targeting antibody drug conjugate, in
a phase 1 study of relapsed/refractory non-Hodgkin lymphoma shows activity in T-Cell lymphoma. Blood. 2018;132 (Supplement 1):1658.
52. Swerdlow S, Campo E, Harris NL, et al. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, FR: IARC Press; 2016.
53. Martin P, Leonard J. Is there a role for “watch and wait” in patients with mantle cell lymphoma? Semin Hematol. 2011;48:189-193.
54. Ruan J, Leonard JP. Lenalidomide plus rituximab for mantle-cell lymphoma. N Engl J Med. 2016;374:793.
55. Ruan J, Martin P, Christos P, et al. Five-year follow-up of lenalidomide plus rituximab as initial treatment of mantle cell lymphoma. Blood. 2018;132:2016-2025.
56. Jain P, Lee HJ, Steiner R, et al. Frontline treatment with ibrutinib with rituximab (IR) combination is highly effective in elderly (65 years) patients with mantle cell
lymphoma (MCL) - results from a phase II trial. Blood. 2019;134 (Supplement_1):3988.
57. Morschhauser F, Fowler NH, Feugier P, et al; RELEVANCE Trial Investigators. Rituximab plus lenalidomide in advanced untreated follicular lymphoma. N Engl
J Med. 2018;379:934-947.
58. Nastoupil L, Westin J, Hagemeister F, et al. Results of a phase II study of obinutuzumab in combination with lenalidomide in previously untreated, high tumor
burden follicular lymphoma (FL). Blood. 2019;134 (Supplement_1):125.
59. Wang ML, Jain P, Lee HJ, et al. Frontline treatment with ibrutinib plus rituximab (IR) followed by short course r-hypercvad/mtx is extremely potent and safe in
patients (age 65 years) with mantle cell lymphoma (MCL) - Results of phase-II Window-1 clinical trial. Blood. 2019;134 (Supplement_1):3987.
60. Ujjani CS, Jung SH, Pitcher B, et al. Phase 1 trial of rituximab, lenalidomide, and ibrutinib in previously untreated follicular lymphoma: Alliance A051103. Blood.
2016;128:2510-2516.
61. Eskelund CW, Dahl C, Hansen JW, et al. TP53 mutations identify younger mantle cell lymphoma patients who do not benefit from intensive chemo-
immunotherapy. Blood. 2017;130:1903-1910.
62. Wang ML, Rule S, Martin P, et al. Targeting BTK with ibrutinib in relapsed or refractory mantle-cell lymphoma. N Engl J Med. 2013;369:507-516.

63. Wang ML, Blum KA, Martin P, et al. Long-term follow-up of MCL patients treated with single-agent ibrutinib: updated safety and efficacy results. Blood. 2015;
126:739-745.
64. Martin P, Maddocks K, Leonard JP, et al. Postibrutinib outcomes in patients with mantle cell lymphoma. Blood. 2016;127:1559-1563.
65. Syed YY. Zanubrutinib: first approval. Drugs. 2020;80:91-97.
66. Wang M, Rule S, Zinzani PL, et al. Acalabrutinib in relapsed or refractory mantle cell lymphoma (ACE-LY-004): a single-arm, multicentre, phase 2 trial. Lancet.
2018;391:659-667.
67. Jain P, Romaguera J, Srour SA, et al. Four-year follow-up of a single arm, phase II clinical trial of ibrutinib with rituximab (IR) in patients with relapsed/refractory
mantle cell lymphoma (MCL). Br J Haematol. 2018;182:404-411.
68. Jerkeman M, Eskelund CW, Hutchings M, et al. Ibrutinib, lenalidomide, and rituximab in relapsed or refractory mantle cell lymphoma (PHILEMON): a mul-
ticentre, open-label, single-arm, phase 2 trial. Lancet Haematol. 2018;5:e109-e116.
69. Davids MS, Roberts AW, Seymour JF, et al. Phase I first-in-human study of venetoclax in patients with relapsed or refractory non-Hodgkin lymphoma. J Clin Oncol.
2017;35:826-833.
70. Tam CS, Anderson MA, Pott C, et al. Ibrutinib plus venetoclax for the treatment of mantle-cell lymphoma. N Engl J Med. 2018;378:1211-1223.
71. Tan D, Horning SJ, Hoppe RT, et al. Improvements in observed and relative survival in follicular grade 1-2 lymphoma during 4 decades: the Stanford University
experience. Blood. 2013;122:981-987.
72. Fowler NH, Davis RE, Rawal S, et al. Safety and activity of lenalidomide and rituximab in untreated indolent lymphoma: an open-label, phase 2 trial. Lancet Oncol.
2014;15:1311-1318.
73. Martin P, Jung SH, Pitcher B, et al. A phase II trial of lenalidomide plus rituximab in previously untreated follicular non-Hodgkin’s lymphoma (NHL): CALGB
50803 (Alliance). Ann Oncol. 2017;28:2806-2812.
74. Zucca E, Rondeau S, Vanazzi A, et al; Swiss Group for Clinical Cancer Research and the Nordic Lymphoma Group. Short regimen of rituximab plus lenalidomide
in follicular lymphoma patients in need of first-line therapy. Blood. 2019;134:353-362.
75. Leonard JP, Trneny M, Izutsu K, et al; AUGMENT Trial Investigators. AUGMENT: a phase III study of lenalidomide plus rituximab versus placebo plus rituximab in
relapsed or refractory indolent lymphoma. J Clin Oncol. 2019;37:1188-1199.
76. Andorsky DJ, Coleman M, Abdulraheem A, et al. MAGNIFY: phase IIIb interim analysis of induction R2 followed by maintenance in relapsed/refractory indolent
non-Hodgkin lymphoma. J Clin Oncol. 2019;37:15 (suppl; abstr 7513).
77. Smith SM, Pitcher BN, Jung SH, et al. Safety and tolerability of idelalisib, lenalidomide, and rituximab in relapsed and refractory lymphoma: the Alliance for
Clinical Trials in Oncology A051201 and A051202 phase 1 trials. Lancet Haematol. 2017;4:e176-e182.
78. Gopal AK, Schuster SJ, Fowler NH, et al. Ibrutinib as treatment for patients with relapsed/refractory follicular lymphoma: results from the open-label, multicenter,
phase II DAWN Study. J Clin Oncol. 2018;36:2405-2412.
79. Bartlett NL, Costello BA, LaPlant BR, et al. Single-agent ibrutinib in relapsed or refractory follicular lymphoma: a phase 2 consortium trial. Blood. 2018;
131:182-190.
80. Casulo C, Byrtek M, Dawson KL, et al. Early relapse of follicular lymphoma after rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone
defines patients at high risk for death: an analysis from the National LymphoCare Study. J Clin Oncol. 2015;33:2516-2522.
81. Gopal AK, Kahl BS, de Vos S, et al. PI3Kδ inhibition by idelalisib in patients with relapsed indolent lymphoma. N Engl J Med. 2014;370:1008-1018.
82. Flinn IW, O’Brien S, Kahl B, et al. Duvelisib, a novel oral dual inhibitor of PI3K-δ,γ, is clinically active in advanced hematologic malignancies. Blood. 2018;
131:877-887.
83. Dreyling M, Santoro A, Mollica L, et al. Phosphatidylinositol 3-kinase inhibition by copanlisib in relapsed or refractory indolent lymphoma. J Clin Oncol. 2017;
35:3898-3905.
84. Morschhauser F, Tilly H, Chaidos A, et al. Phase 2 multicenter study of tazemetostat, an EZH2 inhibitor, in patients with relapsed or refractory follicular lymphoma.
Blood. 2019;134 (Supplement_1):123.

Clinical Controversies in the Management of

Smoldering Multiple Myeloma
Oliver C. Lomas, MD, PhD,1 and Irene M. Ghobrial, MD1
overview
Forty years ago this year, smoldering multiple myeloma was defined as a clinical entity that identifies a group of
patients with a substantial burden of disease but with a relatively indolent natural history compared with
symptomatic disease. Since then, there has been a revolution in the therapeutic options for multiple myeloma.
The aim of this article is to describe recent advances in the identification of those patients who are at the
highest risk of progression and whether they may benefit from therapy. Treatment of smoldering myeloma is an
area of active debate and we present contrasting interpretations of the available trial data. We conclude by
identifying the priorities for research that will help to clarify the management of this condition, which can be
challenging for physicians and patients alike.
INTRODUCTION as a plasma cell percentage greater than 10% and/or

Plasma cells are terminally differentiated B lympho- a serum paraprotein greater than 3.0 g/dL.7 However,
cytes whose role is to produce abundant quantities of it was noted that a proportion of these patients expe-
antibody as part of the humoral immune system.1 The rienced a clinical course that was comparatively in-
malignant transformation of plasma cells may result in dolent, resembling MGUS. This indeterminate area
the symptomatic disease of multiple myeloma (MM). between premalignancy and symptomatic disease gave
The characteristic features of the disease comprise rise to the concept of smoldering multiple myeloma
any one of the following, referred to as "CRAB": ele- (SMM).8 A reason for defining this clinical subgroup was
vated serum calcium concentration, renal impair- to spare asymptomatic patients from the toxic chemo-
ment, anemia, and lytic bone lesions. Reflective of therapy of the time.
their premalignant function, transformed plasma Compared with the early 1980s, the treatment of
cells continue to secrete antibody in its entirety or in myeloma has been revolutionized by several novel
a partial form comprising the light chains of the im- classes of therapy, which have improved survival and
munoglobulin. The detection of this paraprotein has lowered morbidity.9 Because the distinction between
allowed the identification of an asymptomatic pre- smoldering and symptomatic myeloma is largely defined
cursor that is known to precede MM, amyloidosis, or by the risks of therapy versus symptomatic end-organ
other lymphoproliferative disorders.2-4 This gave rise damage by the disease, there has been a recurrent
to the terminology of a monoclonal gammopathy of desire to evaluate the optimal timing of initiation of
undetermined significance (MGUS) because not all therapy. In 2014, the International Myeloma Working
cases of a serum paraprotein necessarily progress to Group (IMWG) updated the diagnostic criteria for MM
a symptomatic disease.5 MGUS is found in approxi- to include certain laboratory and imaging markers to
mately 3% of the population older than age 50, with identify patients who would benefit from therapy de-
Author affiliations
a median age of presentation of 72. The cumulative spite being asymptomatic (Table 1). These so-called
and support risk of progression to MM was 10% at 10 years, in- myeloma-defining events comprise the following: bone
information (if creasing to 36% at 40 years, which is approximately marrow plasma cell infiltration of 60% or greater, a
applicable) appear equivalent to 1% per year.6 Therefore, most patients serum involved free light chain ratio of 100 or greater,
at the end of this
with MGUS do not progress to symptomatic disease. or two or more focal lesions on cross-sectional imaging
article.
MM and MGUS represent opposite ends of a spec- of the whole body.10 These IMWG criteria reclassified
Accepted on
February 25, 2020
trum of disease caused by the clonal expansion of a small proportion of patients with ultra–high-risk SMM
and published at malignant plasma cells in the bone marrow. to MM that would be considered for therapy. The ra-
ascopubs.org on
The asymptomatic nature of MGUS, with the overall low tionale behind this revised classification is based on
March 31, 2020:
rate of progression to symptomatic disease, means that retrospective cohort analysis rather than specific path-
10.1200/EDBK_ therapy was traditionally reserved for patients with ophysiologic characteristics of the disease during
278911 a substantial burden of disease. This state was defined monitoring. The aim of this article is to discuss the

Controversies in Smoldering Myeloma
TABLE 1. 2014 International Myeloma Working Group Diagnostic

Criteria for MGUS, SMM, and MM
PRACTICAL APPLICATIONS Disease Criterion
• Smoldering multiple myeloma is a disease entity MGUS Serum M protein , 3 g/dL and/or abnormal FLC ratio
that is defined by the therapies available. (, 0.26 or . 1.65) with an increased level of the
• Randomized controlled trials have shown that appropriate involved light chain
disease biology may be altered by early Clonal BM plasma cells , 10%
intervention. Absence of MDEs or amyloidosis
• However, improvements in overall survival have SMM Serum M protein . 3 g/dL
not been demonstrated in smoldering myeloma
according to current diagnostic criteria. And/or clonal BM plasma cells . 10% and , 60%
And/or urinary monoclonal protein . 500 mg per
• Current risk stratification relies on clinical
24 hours
markers, although the integration of molecular
markers to predict progression is under Absence of MDEs or amyloidosis
development. MM Evidence of end-organ damage
• Improved models of risk stratification are re- Hypercalcemia: serum calcium . 0.25 mM above the
quired to allow for patient counseling and fo- upper limit of normal or . 2.75 mM
cused trials. Renal insufficiency: serum creatinine . 173 mM
Anemia: hemoglobin value . 2 g/dL below the lower limit of
normal or a hemoglobin value , 10 g/dL
controversies in the assessment of risk of progression to Bone lesions: one or more lytic lesions on skeletal
myeloma and subsequent management of this challenging radiography or CT (including PET)
medical problem. SLiM criteria
Clonal BM plasma cell percentage 60%
RISK ASSESSMENT IN SMOLDERING MULTIPLE MYELOMA
Involved/uninvolved serum FLC ratio 100
The first step in counseling patients with smoldering mye-
More than one focal lesion on MRI
loma is to provide a meaningful assessment of risk of dis-
ease progression. However, SMM is both clinically and Abbreviations: MGUS, monoclonal gammopathy of undetermined
biologically heterogeneous. The largest retrospective cohort significance; SMM, smoldering multiple myeloma; MM, multiple
study of the clinical course and prognosis in SMM was myeloma; FLC, free light chain; BM, bone marrow; MDE, myeloma-
performed based on criteria that stipulated a bone marrow defining event; SLiM, sixty percent light chains MRI.
infiltrate of clonal plasma cells of 10% or greater and/or
a serum paraprotein of 30% or greater in the absence of a similar time to progression to MM of 23 months compared
symptoms.7 Based on a 26-year follow-up, patients with with the Mayo observations in 2007, the concordance
elevations of both criteria were in a group with the highest between the two parameters was less than one-third in
risk of progression than if only one criterion were met. a direct comparison.13 After refinement of the IMWG di-
Patients with the greatest burden of disease showed agnostic criteria for MM in 2014, a revised risk stratification
a median time to progression to MM of 2 years compared model was described by the Mayo group.14 It is important to
with 8 years among patients with the lowest burden of note that the consensus diagnostic criteria had changed by
disease. Overall, patients with SMM are at a 10% annual risk this time to reclassify a small number of patients with
of progression to MM compared with 1% for patients with ultra–high-risk SMM as having MM.10 The 2018 Mayo re-
MGUS. Even in the highest risk group, 13% of patients still vision used plasma cell marrow infiltration greater than
had not progressed to myeloma by 15 years.11 These data 20%, an involved free light chain ratio greater than 20, and
based on plasma cell infiltration and paraprotein concen- a paraprotein concentration of greater than 2 g/L to generate
tration demonstrate the notable heterogeneity in the rate of a 20-2-20 risk prediction model.
disease progression, which individual patients may find
These criteria continued to benefit from simple application
difficult to comprehend.
in the clinic with improved identification of high-risk SMM.
The PETHEMA group developed a model that uses pa- Using the 2003 criteria, high and intermediate groups
rameters that are not so readily accessible in the clinic. The showed a mean time to progression of 45 and 23 months,
model incorporates the proportion of plasma cells with an respectively. However, consistent with the heterogenous
aberrant immunophenotype on flow cytometry of bone nature of SMM, the confidence intervals were so large as to
marrow aspirate and serum immunoparesis to stratify pa- overlap. This meant that for some patients informed of being
tients by risk.12 Despite the high-risk group displaying at intermediate risk, they could progress to myeloma more

Lomas and Ghobrial
quickly than a patient designated as being at high risk. In the requires cross-sectional imaging with CT or MRI and a bone
newer prediction model, the high-risk group has a mean marrow trephine biopsy with cytogenetic analysis.10 Once
time to progression of 29 months compared with 68 months MM has been excluded, the standard of care is regular
for the intermediate group, without overlapping 95% con- observation even up to every 3 months for patients con-
fidence intervals by 10 years of follow-up. At the other end of sidered to be at high risk of progression. As discussed, one
the observation period, there is little separation between the of the aims of risk stratification has been to identify patients
intermediate and high-risk groups by 12 months of follow- at high risk of progression to MM and for whom earlier
up, which was similarly observed when using the conven- intervention may improve overall survival. Such an approach
tional stratification criteria from 2003.14 is reasonable given that early mortality after diagnosis of MM
remains a substantial problem.19
Other parameters such as immunoglobulin A monoclonal
gammopathy and tumor genetics have been implicated in Treatment of an asymptomatic, precursor malignant con-
the risk of progression from SMM to active disease.15 Many dition requires a different discussion with a patient com-
of the retrospective cohorts were commenced at a time pared with symptomatic disease. For patients with newly
when cytogenetic, not to mention genomic variants, were diagnosed MM, a key aim of therapy is to induce a deep
not reported as well as prior to the widespread use of cross- remission and alleviate the symptoms of disease. In the
sectional imaging. Subgroup analysis of patients in the Mayo context of SMM, improved overall survival is paramount
cohort who had cytogenetic test data available suggested because the patient will have to tolerate the inconvenience
a notable benefit to prognostic modeling. However, the and side effects of therapy in the absence of symptoms. The
observations were underpowered based on the available original definition of smoldering myeloma was made by Kyle
cohort size for the most recent model.14 and Greipp8 in 1980 specifically to protect patients from the
The current standard of care for SMM is to monitor for the toxic side effects of chemotherapy. The last 2 decades have
development of myeloma-defining events at which point the witnessed a proliferation of efficacious therapeutic options
balance of risk of drug toxicity is outweighed by the benefits for MM. These agents have dramatically improved survival
of therapy. The purpose of the risk stratification models in with fewer therapy-related side effects.9 Given that the
SMM has its origins in helping to identify patients at di- definition of disease is at least in part determined by
agnosis for interventional clinical trials. The Mayo cohort therapeutic tolerability and efficacy, it is important that the
analysis revealed that risk of progression decreased from optimal timing of therapy in SMM is readdressed.
diagnosis, particularly for the high-risk subgroup, such that The first major trial to assess intervention in smoldering
the risk of progression in all groups converged at a 3% to 5% myeloma was performed by the PETHEMA group.20 A
annual risk of progression. Clinically, this is a very important treatment arm of lenalidomide and dexamethasone was
observation that supports the practice of monitoring a pa- compared with observation alone in a cohort of patients
tient over time and then considering intervention accord- considered to be at high risk of progression to myeloma. Not
ingly. Only by identifying patients at high risk of rapid only did the study reveal a 2-year progression-free survival
progression to disease can informed decisions be made (PFS) of 92% for the treatment arm compared with 30% for
between patient and provider when the question of early the observation arm, but overall survival increased from
intervention may be addressed. 76% for the observation arm to 93% with treatment. For the
A broader unmet need in the care of patients with MM is the first time, the study showed that the natural history of
difference in incidence of disease between different ethnic precursor states to myeloma could be modified by starting
groups. Black Americans are disproportionately affected, therapy prior to the onset of end-organ damage. However,
with an incidence of MM and MGUS that is at least double the study has not changed mainstream practice. As sug-
the incidence of white Americans.16,17 Much of our un- gested by the mortality rate of approximately 25% for the
derstanding of the epidemiology of the spectrum of MM is observation arm, the trial may, in part, reflect disease that
derived from cohorts in which the population has been would be classified as MM using current diagnostic criteria.
overwhelmingly white.11,18 This disparity between research
To address these concerns, the ECOG (Eastern Co-operative
and clinical need follows through into under-representation
Oncology Group) performed a randomized, open-label,
in clinical trials and must be addressed so that health care
multicenter phase III trial that compared oral lenalidomide
services provide for the populations they serve.
(25 mg) to observation across all groups at risk of pro-
gression in SMM.21 The trial design used the 2014 IMWG
THERAPEUTIC STRATEGIES IN SMOLDERING diagnostic criteria for myeloma and the primary endpoint
MULTIPLE MYELOMA was PFS from randomization to symptomatic myeloma.
Once SMM is suspected, active MM needs to be excluded. Lenalidomide therapy was safe but not always well toler-
According to the 2014 IMWG consensus criteria, this ated. Of 88 patients, 29 discontinued therapy as a result of

adverse events or patient preference. By 3 years of follow- with MGUS compared with healthy matched controls,. This
up, PFS was 91% for the lenalidomide arm compared with effect appeared independent of paraprotein concentration
66% for the observation arm. To put these findings another and, importantly, of likelihood to progression to MM.23 In
way that may be clearer to patients, this is a cumulative SMM, prophylaxis with bisphosphonate therapy has been
incidence of myeloma of 7% for the lenalidomide arm associated with a reduced risk of fractures but without effect
compared with 32% for the observation arm at 3 years from of progression or survival.24,25 Nevertheless, the avoidance
randomization. In terms of overall survival by 35 months of of fractures is an important goal for patients and therefore
follow-up, no benefit was observed, in contrast to the earlier should be considered in all cases of SMM, irrespective of
trial. The greater PFS for the observation arm of the ECOG perceived risk of progression. This approach is consistent
trial compared with the PETHEMA group suggest a cohort with the overall aim to improve quality of life for patients
with overall lower-risk SMM. The data presented by Lonial with SMM.
et al21 build on earlier observations to show that the natural
TO TREAT OR NOT TO TREAT?
history of myeloma may be altered by early intervention. The
evidence from this randomized trial may be used in the The authors of this article represent the range of views held
clinic to counsel patients who are presented with the un- about how best to interpret the inconclusive data that are
settling diagnosis of SMM. The trial by Lonial et al21 provides currently available to counsel patients. The lack of a re-
evidence that early therapy delays progression to active ported benefit on overall survival prevents a change to the
disease for some patients with SMM. standard of care of close observation. Even if overall survival
were shown on a cohort basis, the numbers needed to treat
However, many questions remain before we change our to achieve this may be unacceptable to patients in terms of
standard of practice to treat patients with high-risk SMM. In tolerability of lenalidomide therapy. Improved risk stratifi-
particular, the effects of overall survival must be measured cation with serial reassessment may help to identify patients
before changes to the standard of care should be recom- with truly high risk while sparing patients who will follow
mended. It is not known whether therapy with single-agent a more benign natural history.14 If this were achieved, the
lenalidomide drives the selection of insensitive subclones of moniker of SMM could be dropped, as it could be argued
malignant plasma cells and, if so, what the effect on overall that it is not a helpful term for patients. Many patients,
survival may be.22 Another question is whether response especially those who ultimately follow a progression rate of
rate and minimal residual disease are critical as endpoints in 3% to 5% per year, can feel unduly afflicted by their sense
SMM. In the PETHEMA studies, PFS correlates with re- of an underlying malignancy waiting to “ignite” into MM.
sponse but the study by Lonial et al21 indicated that single- Finally, it may be argued that clonal eradication at a high-
agent lenalidomide may act as an immunomodulator that risk, low-burden stage of disease is a logical therapeutic
controls the disease burden and improves PFS in the ab- target. This may be achieved with high-intensity immuno-
sence of a deep response. Furthermore, both studies used therapy, which may be better tolerated with the lower
very different inclusion criteria for SMM, and patients and burdens of disease of SMM. On the other hand, a clear
their doctors need more predictive models of progression. increase in PFS was observed with the use of single-agent
These two large randomized trials have not proved de- lenalidomide for patients with SMM, especially those at high
finitively that lenalidomide with or without dexamethasone is risk of progression. With a PFS at 3 years of 91% for the
a therapy that prolongs overall survival in SMM as it is treatment group compared with 66% for the observation
defined in 2020. Until a more successful approach is shown group, it is reasonable to discuss with patients whether they
in clinical trials, the current standard of care in SMM re- would consider 2 years of lenalidomide therapy to delay
mains observation with initiation of therapy at the earliest progression to myeloma. Trials such as these provide clear
opportunity with triplets or even quadruplets of drugs. data whereby informed decisions can be made rather than
Given the different therapeutic goals in SMM compared with relying on speculation.
MM, this asymptomatic but high-risk precursor may benefit CONCLUSION
from therapy with the goal of a prolonged maintenance of an
Preventative therapy in the absence of symptomatic
asymptomatic MGUS-like state or even eradication of all
disease is a challenging clinical question that applies to
malignant clones. To achieve such aims, novel therapeutic
many pathologies. The goal of therapy in SMM is to im-
strategies may be required rather than just attenuated
prove overall survival with a minimal impact on quality of
therapies for MM.
life. The future of prevention of SMM progression may
A myeloma-defining event is the development of osteolytic require an entirely novel approach with interventions to
bone lesions.10 At the other end of the spectrum, MGUS may restore immune surveillance to maintain long-term con-
be not entirely asymptomatic. A 1.6-fold increased risk of trol of the malignant clones of plasma cells or perhaps
any fracture was found at 10 years for a cohort of patients even more potent therapy to eradicate the disease at

Lomas and Ghobrial
its earliest progenitor state. Further improvement in risk definition of smoldering myeloma 40 years ago this year,
stratification across all patient groups is required to bring to spare patients from treatment that may do more harm
therapeutic benefit to patients and, in the spirit of Kyle’s than good.

1
Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard AND DATA AVAILABILITY STATEMENT
Medical School, Boston, MA Disclosures provided by the authors and data availability statement (if
EDBK_278911.
Irene Ghobrial, MD, Department of Medical Oncology, Dana-Farber
Cancer Institute, Harvard Medical School, 450 Brookline Ave., Boston,
MA 02215; email: [email protected].
REFERENCES
1. Chu VT, Berek C. The establishment of the plasma cell survival niche in the bone marrow. Immunol Rev. 2013;251:177-188.
2. Weiss BM, Abadie J, Verma P, et al. A monoclonal gammopathy precedes multiple myeloma in most patients. Blood. 2009;113:5418-5422.
3. Landgren O, Kyle RA, Pfeiffer RM, et al. Monoclonal gammopathy of undetermined significance (MGUS) consistently precedes multiple myeloma: a prospective
study. Blood. 2009;113:5412-5417.
4. Goldin LR, McMaster ML, Caporaso NE. Precursors to lymphoproliferative malignancies. Cancer Epidemiol Biomarkers Prev. 2013;22:533-539.
5. Kyle RA. Monoclonal gammopathy of undetermined significance. Natural history in 241 cases. Am J Med. 1978;64:814-826.
6. Kyle RA, Larson DR, Therneau TM, et al. Long-term follow-up of monoclonal gammopathy of undetermined significance. N Engl J Med. 2018;
378:241-249.
7. International Myeloma Working Group. Criteria for the classification of monoclonal gammopathies, multiple myeloma and related disorders: a report of the
International Myeloma Working Group. Br J Haematol. 2003;121:749-757.
8. Kyle RA, Greipp PR. Smoldering multiple myeloma. N Engl J Med. 1980;302:1347-1349.
9. Thorsteinsdottir S, Dickman PW, Landgren O, et al. Dramatically improved survival in multiple myeloma patients in the recent decade: results from a Swedish
population-based study. Haematologica. 2018;103:e412-e415.
10. Rajkumar SV, Dimopoulos MA, Palumbo A, et al. International Myeloma Working Group updated criteria for the diagnosis of multiple myeloma. Lancet Oncol.
2014;15:e538-e548.
11. Kyle RA, Remstein ED, Therneau TM, et al. Clinical course and prognosis of smoldering (asymptomatic) multiple myeloma. N Engl J Med. 2007;
356:2582-2590.
12. Pérez-Persona E, Vidriales MB, Mateo G, et al. New criteria to identify risk of progression in monoclonal gammopathy of uncertain significance and smoldering
multiple myeloma based on multiparameter flow cytometry analysis of bone marrow plasma cells. Blood. 2007;110:2586-2592.
13. Cherry BM, Korde N, Kwok M, et al. Modeling progression risk for smoldering multiple myeloma: results from a prospective clinical study. Leuk Lymphoma. 2013;
54:2215-2218.
14. Lakshman A, Rajkumar SV, Buadi FK, et al. Risk stratification of smoldering multiple myeloma incorporating revised IMWG diagnostic criteria. Blood Cancer J.
2018;8:59.
15. Weber DM, Dimopoulos MA, Moulopoulos LA, et al. Prognostic features of asymptomatic multiple myeloma. Br J Haematol. 1997;97:810-814.
16. Singh J, Dudley AW Jr., Kulig KA. Increased incidence of monoclonal gammopathy of undetermined significance in blacks and its age-related differences with
whites on the basis of a study of 397 men and one woman in a hospital setting. J Lab Clin Med. 1990;116:785-789.
17. Howlader NN, Noone AM, Krapcho M, et al. SEER Cancer Statistics Review, 1975-2016. Bethesda, MD: National Cancer Institute.
18. Kyle RA, Larson DR, Therneau TM, et al. Clinical course of light-chain smouldering multiple myeloma (idiopathic Bence Jones proteinuria): a retrospective cohort
study. Lancet Haematol. 2014;1:e28-e36.
19. Costa LJ, Gonsalves WI, Kumar SK. Early mortality in multiple myeloma. Leukemia. 2015;29:1616-1618.
20. Mateos MV, Hernández MT, Giraldo P, et al. Lenalidomide plus dexamethasone for high-risk smoldering multiple myeloma. N Engl J Med. 2013;369:438-447.
21. Lonial S, Jacobus S, Fonseca R, et al. Randomized trial of lenalidomide versus observation in smoldering multiple myeloma. J Clin Oncol. 2020;38:1126-1137.
22. Jones JR, Weinhold N, Ashby C, et al; NCRI Haemato-Oncology CSG. Clonal evolution in myeloma: the impact of maintenance lenalidomide and depth of
response on the genetics and sub-clonal structure of relapsed disease in uniformly treated newly diagnosed patients. Haematologica. 2019;
104:1440-1450.

23. Kristinsson SY, Tang M, Pfeiffer RM, et al. Monoclonal gammopathy of undetermined significance and risk of skeletal fractures: a population-based study. Blood.
2010;116:2651-2655.
24. Musto P, Petrucci MT, Bringhen S, et al; GIMEMA (Italian Group for Adult Hematologic Diseases)/Multiple Myeloma Working Party and the Italian Myeloma
Network. A multicenter, randomized clinical trial comparing zoledronic acid versus observation in patients with asymptomatic myeloma. Cancer. 2008;
113:1588-1595.
25. D’arena G, Gobbi PG, Broglia C, et al; Gimema (Gruppo Italiano Malattie Ematologiche Dell’Adulto); Multiple Myeloma Working Party; Gisl (Gruppo Italiano Studio
Linfomi) Cooperative Group. Pamidronate versus observation in asymptomatic myeloma: final results with long-term follow-up of a randomized study. Leuk
Lymphoma. 2011;52:771-775.

Current Perspectives on Therapy for Chronic

Lymphocytic Leukemia
Farrukh T. Awan, MD1; Othman Al-Sawaf, MD2; Kirsten Fischer, MD2; and Jennifer A. Woyach, MD3
overview
Therapy for chronic lymphocytic leukemia has improved dramatically over the past decade with the in-
troduction of new targeted therapies and a paradigm shift toward targeted therapies for the majority of
patients. Better understanding of prognostic factors has helped tailor therapy for individual patients, and work
continues to identify optimal therapy for each patient. When therapy is required, most patients will be treated
with targeted therapies, either the Bruton tyrosine kinase (BTK) inhibitors ibrutinib or acalabrutinib or the
BCL-2 inhibitor venetoclax in combination with obinutuzumab. Without head-to-head comparisons showing
differential efficacy among these options, considerations regarding safety, patient preference, and ability to
sequence therapy currently influence treatment decisions. Also, clinical trials investigating combinations of
these therapies have the potential to further change the standard of care. In this review, we cover the currently
available options for the frontline treatment of chronic lymphocytic leukemia (CLL) and discuss safety
considerations and toxicity management with each agent as well as novel combination strategies currently
under investigation.
INTRODUCTION All of these factors had previously been shown to have

Chronic lymphocytic leukemia (CLL) is a heteroge- independent prognostic value in multiple prospective
neous disease with a natural history ranging from an trials. Historically, response to treatment, defined as
indolent clinical course in which patients do not require partial or complete response, has been defined by
therapy for many years to an aggressive disease for clinical response assessment according to the In-
which treatment is necessary soon after diagnosis.1 On ternational Workshop on CLL criteria.15 More recently,
the basis of our evolving knowledge of biologic factors minimal residual disease (MRD) assessment has been
of CLL, patients can be stratified into subgroups with incorporated into response assessments in clinical
distinct biologic features. Consequently, given the trials. Although this approach is not yet recommended
availability of numerous therapies, it is important to for general clinical practice, the presence of un-
develop a tailored treatment strategy for the individual detectable MRD—by detection of malignant cells with
patient that considers a balance of treatment efficacy a sensitivity of 1 tumor cell in 10,000 cells (10 4), with
and toxicity.2,3 Recent developments in the manage- undetectable MRD defined as less than 10 4—at the
ment of CLL have shifted the standard of care from end of therapy has emerged as the most important
chemoimmunotherapy to highly effective targeted predictor of progression-free survival and overall sur-
agents, such as oral kinase inhibitors or BCL-2 in- vival with time-limited therapies.16-18
hibitors alone or in combination with anti-CD20 anti- Initiating Frontline Therapy
bodies. Two different treatment paradigms have
emerged4-7: continuous indefinite treatment and fixed- There is currently no evidence for a potential benefit
Author affiliations
duration combination regimen.8-14 of early intervention for patients with asymptomatic
and support CLL.19-23 Therapy initiation should be postponed until
information (if
CURRENT FRONTLINE THERAPY OPTIONS FOR CLL active disease is observed, defined according to In-
applicable) appear
at the end of this
ternational Workshop on CLL guidelines15 (Sidebar 1).
Biologic and Clinical Factors Assessing Prognosis and Clinical trials evaluating the early use of novel inhibitors
article.
Response to Therapy are currently ongoing, but none of these has so far
Accepted on
February 21, 2020 At present, of the multiple clinical and biologic factors provided evidence that alters the current standard of
and published at that may prognosticate the patient’s individual course, care: watch and wait24,25 (Tables 1 and 2).
ascopubs.org on
four distinct factors guide individualized treatment:
April 2, 2020: Chemoimmunotherapy
TP53 mutation and/or deletion, age, the presence and
10.1200/EDBK_ level of comorbidities, and immunoglobulin heavy- For more than a decade, patients with CLL have been
279099 chain variable region gene (IGHV) mutational status. treated with chemoimmunotherapy (e.g., fludarabine

enrolled had a score greater than 6 on the cumulative illness

rating scale, suggesting that the patient cohort was relatively
fit as opposed to other CLL trials for unfit patients. Long-
• Watch and wait remains the standard of care term follow-up of both trials demonstrated an increase in
outside clinical trials for patients with early-
progression-free survival, indicating that ibrutinib can maintain
stage, asymptomatic CLL, regardless of biologic
long-term remissions in patients, including those with TP53
risk factors.
mutation and/or deletion and/or unmutated IGHV mutational
• Clinical stage, physical fitness, immunoglobulin status, with no new toxicities identified.8,33 The benefit of
heavy-chain variable region gene mutational
adding an anti-CD20 antibody to ibrutinib is unclear, as two
status, and TP53 mutation and/or deletion are
randomized comparisons of ibrutinib alone versus ibrutinib
the most distinct clinical and biologic factors to
be considered for frontline therapy choice. plus rituximab demonstrated similar progression-free survival
in both groups.8,33 However, the addition of anti-CD20 anti-
• Most patients with standard-risk CLL have
bodies to ibrutinib is generally well tolerated and is supported
similar expectations of efficacy with current
by data at this time.10,11
frontline therapies; therefore, safety consider-
ations and management of toxicities are of great The Alliance A041202 trial (Table 1) was conducted in
importance with these medications. previously untreated patients older than age 65 who were
• Studies are ongoing to determine whether randomly assigned to receive bendamustine plus rituximab
combinations of novel therapies are superior to (BR), ibrutinib plus rituximab, or ibrutinib alone. The trial
single novel therapy regimens. demonstrated longer progression-free survival for patients
• Clinical trials remain critical to determine the who were treated with either ibrutinib-containing regimen,
optimal therapy for patients with CLL. including those with unfavorable risk factors, such as TP53
mutation and/or deletion or unmethylated ZAP-70 status (as
a surrogate for unmutated IGHV status). Of note, no progression-
and cyclophosphamide, chlorambucil, or bendamustine in free survival difference was observed between ibrutinib
combination with anti-CD20 monoclonal antibodies).5,6,28 plus rituximab and ibrutinib alone.10
Long-term follow-up data showed that fit patients with
For previously untreated fit patients, the phase III ECOG
disease and mutated IGHV mutational status can achieve
1912 trial evaluated ibrutinib in combination with rituximab
long-term disease control and survival prolongation after
against the standard FCR regimen (Table 1). Results of
the fludarabine, cyclophosphamide, and rituximab (FCR)
ECOG 1912 showed improved overall survival for patients
regimen.4,29,30 The observed benefit of FCR for patients with
who were treated with ibrutinib; however, patients with TP53
disease and mutated IGHV mutational status encourages
mutation and/or deletion were not included. With current
the assessment of IGHV mutational status before initiating
follow-up, there is no difference regarding progression-free
a frontline therapy in fit patients with CLL. However, al-
survival and overall survival in patients with disease with
though chemoimmunotherapy likely still has a role in the
a mutated IGHV mutational status.26,34
treatment of CLL, results of recent phase III trials comparing
chemoimmunotherapy with targeted agents are shifting the In elderly patients and comorbid patients, the iLLUMINATE
standard of care away from these chemoimmunotherapy trial tested the combination of ibrutinib and obinutuzumab
regimens (Tables 1 and 2). against the standard of chlorambucil and obinutuzumab
and showed a benefit in progression-free survival for the
Targeted Therapies combination of ibrutinib and obinutuzumab versus chlor-
Targeted agents that have made the greatest impact on ambucil and obinutuzumab, leading to the U.S. Food and
frontline CLL therapy thus far include Bruton tyrosine kinase Drug Administration and European Medicines Agency ap-
(BTK) inhibitors ibrutinib and acalabrutinib and the BCL-2 proval of this combination. As the trial did not include an
inhibitor venetoclax. ibrutinib monotherapy arm, the benefit of adding obinutu-
zumab to ibrutinib remains unclear.11
Ibrutinib has been approved in the United States and
Europe for use as frontline therapy.31 The phase III RES- Most recently, the phase III ELEVATE TN trial reported on
ONATE-2 trial demonstrated improvement in median the clinical outcomes of a randomized three-arm trial
progression-free survival and overall survival in patients age comparing continuous treatment of acalabrutinib with
65 years and older without del(17p) who were treated in- continuous treatment of acalabrutinib monotherapy and
definitely with ibrutinib compared with patients treated for obinutuzumab with 6 months of chlorambucil and obinu-
up to 48 weeks with chlorambucil alone8,32 (Table 1). The tuzumab for the frontline therapy of older patients with CLL
trial led to approval for ibrutinib monotherapy in patients with comorbidities. Outcomes in both experimental groups
with previously untreated CLL. Of note, one-third of patients were superior to chlorambucil plus obinutuzumab, and both

Awan et al
TABLE 1. Selected Published Trials: Phase II and III Trials in Frontline CLL
Experimental Agent/s and Type of Patient Submission ClinicalTrials.
Stage Study ID Comparator Trial Trial Design No. Trial Reference gov Identifier
Early stage CLL7 Fludarabine-cyclophosphamide- Phase Multicenter, 800 No Herling et al23 NCT00275054
rituximab vs. watch and wait in III open label
early-stage high-risk patients
with CLL
CLL12 Ibrutinib vs. placebo in early-stage Phase Multicenter, 363 No Langerbeins NCT02863718
high-risk patients with CLL III open label et al25
Advanced CLL14 Venetoclax-obinutuzumab vs. Phase Multicenter, 432 FDA and EMA Fischer et al12 NCT02242942
stage chlorambucil-obinutuzumab III open label approval
Alliance Bendamustine-rituximab vs. Phase Multicenter, 547 No Woyach et al10 NCT01886872
A041202 ibrutinib vs. ibrutinib-rituximab III open label
E1912 Ibrutinib-rituximab vs. Phase Multicenter, 529 FDA and EMA Shanafelt NCT02048813
fludarabine- III open label approval et al26
cyclophosphamide-rituximab
iLLUMINATE Ibrutinib-obinutuzumab vs. Phase Multicenter, 229 FDA and EMA Moreno et al11 NCT02264574
chlorambucil-obinutuzumab III open label approval
ELEVATE TN Acalabrutinib-obinutuzumab vs. Phase Multicenter, 535 FDA and EMA Sharman NCT02475681
acalabrutinib vs. chlorambucil- III open label approval et al27
obinutuzumab
Abbreviations: CLL, chronic lymphocytic leukemia; FDA, U.S. Food and Drug Administration; EMA, European Medicines Agency.
seem to be similar to what is expected with ibrutinib. At treatment as an alternative option to continuous indefinite
current follow-up, progression-free survival is longer for BTK inhibitor treatment (Table 1). Most recently, longer
patients who were treated with acalabrutinib plus obinu- follow-up confirmed a sustained benefit of fixed-duration
tuzumab versus acalabrutinib alone, with the caveat that treatment with venetoclax and obinutuzumab.38
this comparison was not a preplanned analysis. It is notable
that the outcome in the standard arm with chlorambucil and Recommendations for Frontline Therapy in CLL
obinutuzumab is inferior to that in the current literature.12,35 1. IGHV mutational status, fluorescence in situ hybridiza-
As a result, acalabrutinib is now available as an alternative to tion for del(17p13.1), and sequencing for TP53 mutations
ibrutinib27 (Table 1). should be performed before the initiation of therapy
Venetoclax is a BH3-mimetic compound that selectively in CLL.
antagonizes BCL-2 and induces apoptosis of CLL cells. Its 2. FCR chemoimmunotherapy can be considered for young,
efficacy as monotherapy was previously described in pa- fit patients with IGHV-mutated disease. All other patients
tients with relapsed/refractory CLL, including those with should preferentially be treated with targeted therapies
del(17p).36,37 In frontline therapy, the phase III CLL14 trial when possible.
evaluated venetoclax and obinutuzumab followed by ven- 3. BTK inhibitors can be considered in all patients in need
etoclax monotherapy in patients with previously untreated of therapy, including those with high-risk genomic fea-
CLL and coexisting medical conditions compared with tures, such as TP53 abnormalities.
chlorambucil and obinutuzumab (Table 1). Results dem- 4. Discussion of toxicities and frequency of therapy may
onstrated superiority in progression-free survival of the fixed- help make the decision between treatment with ibrutinib
duration treatment regimen of venetoclax in combination or acalabrutinib. Patients who are intolerant of ibrutinib
with obinutuzumab over that of chlorambucil and obinu- or who have relative contraindications to ibrutinib may
tuzumab.12 After 12 cycles of venetoclax and obinutuzumab still tolerate acalabrutinib.
treatment, patients demonstrated remission with un- 5. Approval of venetoclax and obinutuzumab is based on
detectable MRD across all subgroups with prolonged data in unfit patients and can therefore be considered in
progression-free survival. On the basis of these results, these patients, including those with high-risk genomic
venetoclax in combination with obinutuzumab was ap- features, such as TP53 abnormalities; however, it is likely
proved by the U.S. Food and Drug Administration for the that data in unfit patients can be generalized to all pa-
treatment of previously untreated patients with CLL or small tients. Discussion of toxicities and duration of therapy
lymphocytic lymphoma in 2019. These results influence the may help make the decision between treatment with
choice of frontline therapy by establishing fixed-duration venetoclax and obinutuzumab and BTK inhibitors.

APPROACHES TO SEQUENCING OF INDIVIDUAL REGIMENS Ibrutinib/venetoclax has also been studied in a number of
At this time, limited data exist to inform the optimal se- trials with encouraging responses to date. In a study by MD
quence of BTK inhibitors and BCL-2 inhibitors in clinical Anderson Cancer Center, 80 previously untreated patients
practice. Within BTK inhibitors, acalabrutinib can be suc- were treated with 24 total cycles of combination therapy. At
cessfully used in patients with intolerance to ibrutinib; cycle 18, 69% had undetectable MRD in the marrow.13
however, their mechanism of action is identical, so pro- Also, the phase II CLARITY study tested this combination in
gressive disease cannot be treated by switching BTK in- 50 patients with relapsed CLL. By month 26, 44% of pa-
hibitors. In terms of sequencing these different mechanisms, tients had achieved undetectable MRD in the bone marrow.
the most comprehensive data exist for the sequence of Finally, in the ongoing CAPTIVATE study, of 164 patients
a BTK inhibitor (ibrutinib) and then venetoclax, as this who were enrolled to ibrutinib/venetoclax therapy, 73%
was investigated in a prospective study. In a cohort of were MRD negative after 12 cycles of treatment.42
91 patients who were treated with ibrutinib and who then Although these results are encouraging, data that show
received venetoclax after experiencing disease progres- prolonged progression-free survival will be required before
sion, the overall response rate was 65%, with a median these are considered standard treatment regimens. Multiple
progression-free survival of approximately 22 months.39 ongoing clinical trials are addressing the question of com-
A caveat to these data is that the patients had a median bination versus single targeted therapy strategies and are
of four prior therapies, so outcomes in patients in whom outlined in Table 2.
ibrutinib is the first therapy may be far different. The se-
quence of a BTK inhibitor after venetoclax was addressed TOXICITY CONSIDERATIONS
recently by a collaborative group in a retrospective anal- Tolerability of Novel Targeted Agents in Comparison
ysis. In this analysis, 44 patients who were BTK inhibitor With Chemoimmunotherapy
naı̈ve discontinued venetoclax and were subsequently Whereas novel targeted agents and their combinations have
treated with either ibrutinib or acalabrutinib. The response been shown to significantly improve outcomes compared
rate was 84%, with a median progression-free survival of with chemoimmunotherapy, as highlighted above,10 their
32 months.40 These data support the efficacy of the se- advantage is credited in part to the decreased incidence of
quence of venetoclax and subsequent BTK inhibitor therapy; adverse events (AEs) and generally good tolerability. In the
however, additional data on this subject will be helpful. Alliance A041202 trial of older patients, the incidence of
COMBINATIONS OF BTK INHIBITORS AND BCL-2 INHIBITORS grade 3 or higher nonhematologic AEs was significantly
IN CLINICAL TRIALS higher with BR than with ibrutinib or ibrutinib plus rituximab
(61% vs. 41% vs. 39%, respectively; p .001).10 Moreover,
With the success of both BTK inhibitors and BCL-2 inhibitors
the incidence of febrile neutropenia was also higher in the
in CLL, much current attention is focused on leveraging
BR arm compared with the ibrutinib-containing arms. Of
combinations of these agents to achieve deeper remissions
interest, however, the incidence of all AEs was higher in the
and potentially allow for long periods of therapy discontinu-
ibrutinib-containing arms compared with BR (74% vs.
ation. Although data regarding combinations are exciting,
63%). Similar trends were also observed from the ECOG
especially with high rates of undetectable MRD, given the
1912 trial of younger patients, with an increased risk of
lack of long-term follow up, these combinations should still be
cytopenias and infectious complications with the use of FCR
considered only in the context of a clinical trial.
compared with ibrutinib plus rituximab.26
The longest follow-up data available from ibrutinib and With short follow-up, the incidence of secondary malig-
venetoclax combinations comes from The Ohio State Uni- nancies seems to be similar across arms. Although these
versity phase Ib and II study of the triplet ibrutinib, ven- reports are encouraging, a substantial proportion of patients
etoclax, and obinutuzumab. Long-term data are available undergoing treatment with novel agents discontinue therapy
for the first 62 patients treated (treatment naı̈ve, 25 patients; as a result of intolerable AEs, with an incidence of 16% in
relapsed/refractory, 37 patients). Among treatment-naı̈ve large multi-center clinical trials to as high as more than 40%
patients, the overall response rate was 96%, with 28% in some real-world series.26,43 These discontinuations are
achieving an MRD-negative complete response. Unde- primarily related to unique AEs attributable to off-target,
tectable MRD was observed in the bone marrow of 67% of undesired effects of novel agents. Some of these AEs of
patients across cohorts. 41 With median follow-ups of special interest are detailed below.
24 months for treatment-naı̈ve patients and 21 months for
patients who experienced relapse (12 and 9 months off Risk of Bleeding With BTK Inhibitors
treatment, respectively), only one patient had developed Both ibrutinib and acalabrutinib are associated with an
disease progression but was successfully retreated with increased risk of minor and major bleeding events.31,32,44
ibrutinib. This is a result of inhibiting BTK- and TEC kinase–mediated

Awan et al
TABLE 2. Selected Ongoing/Planned Phase III Trials in Frontline CLL

Experimental Agent/s and Type of Patient Submission
Stage Study ID Comparator Trial Trial Design No. Trial Reference Registration
Ongoing FLAIR Fludarabine-cyclophosphamide- Phase Multicenter, 1,516 No Recruiting ISRCTN01844152
rituximab vs. ibrutinib vs. III open label
ibrutinib-venetoclax
CLL13 Fludarabine/cyclophosphamide/ Phase Multicenter, 920 No Recruitment NCT02950051
rituximab or bendamustine/ III open label completed
rituximab vs. venoclax/rituximab
vs. venetoclax/obinutuzumab vs.
obinutuzumab/ibrutinib/
venetoclax
ECOG Ibrutinib/obinutuzumab vs. Phase Multicenter, 720 No Recruiting NCT03701282
9161 obinutuzumab/ibrutinib/ III open label
venetoclax untreated younger
patients
Alliance Ibrutinib/obinutuzumab vs. Phase Multicenter, 454 No Recruiting NCT03737981
041702 obinutuzumab/ibrutinib/ III open label
venetoclax untreated older
patients
GLOW Ibrutinib/venetoclax vs. Phase Multicenter, 211 Yes Recruiting NCT03462719
chlorambucil/obinutuzumab for III open label
patients with CLL and SLL
ACE-CL- Acalabrutinib/venetoclax with and Phase Multicenter, 780 Yes Recruiting NCT03836261
311 without obinutuzumab vs. III open label
chemoimmunotherapy
(investigator’s choice)
Planned early EVOLVE Venetoclax and obinutuzumab vs. Phase Multicenter, 247 No In preparation NCT04269902
stage CLL delayed therapy with venetoclax III open label
and obinutuzumab
Planned CLL17 Ibrutinib vs. venetoclax/ Phase Multicenter, 882 No In preparation EudraCT 2019-
advanced obinutuzumab vs. ibrutinib/ III open label 003854-99
stage venetoclax
Abbreviations: CLL, chronic lymphocytic leukemia; SLL, small lymphocytic leukemia.
platelet activation downstream of the collagen receptor gly- Although ibrutinib is primarily known to cause issues with
coprotein VI, and adhesion with von Willebrand factor along bleeding, acute venous and arterial thrombotic events have
with inhibition of collagen-induced platelet aggregation.45-49 also been described in a retrospective series of patients
receiving this agent, albeit with low incidences of 0.6 and 1.1
Limited data are currently available about the safety of the
per 100 person-years, respectively.54 Venous thromboem-
combination of ibrutinib or acalabrutinib with anticoagu-
bolic events happened at a median of 7.5 months and arterial
lants. This is primarily because early studies of ibrutinib
thromboses develop at a median of 24.6 months and in-
demonstrated an increased risk of major hemorrhage, in-
cluded both deep venous thrombosis and acute cerebro-
cluding subdural hematomas, when used in conjunction
vascular accidents. In addition, older age and prior history of
with warfarin. Subsequent studies therefore prohibited the
a thrombotic event were identified as possible risk factors for
concomitant use of vitamin K antagonists, such as warfarin8,31,32;
these events.54 Subsequent anticoagulation for thromboembolic
however, multiple reports and meta-analyses have pro-
events with concomitant ibrutinib was also associated with
vided evidence that suggests a higher risk of minor and
a substantial risk of recurrent major and minor bleeding events.
major bleeding episodes when ibrutinib is combined with
anticoagulants and especially with antiplatelet agents.50-53 Acalabrutinib is also known to increase the risk of bleeding
This is of particular importance in patients with atrial fi- events, and both major and minor bleeding events have
brillation, which is a common AE of ibrutinib and is usually been reported with this agent.55 In vitro data suggest a lower
treated with anticoagulation. In addition, anticoagulation impact on platelet aggregation with acalabrutinib compared
should be used with extreme caution when combining with with ibrutinib in a subset of healthy patients. This effect is
antiplatelet agents in patients with pre-existing cardiac presumably a result of the higher effectiveness of ibrutinib in
issues.51 inhibiting BTK, which serves as an adapter protein in the

patients older than age 65 and with a previous history of

atrial fibrillation as being at the highest risk for developing
SIDEBAR 1. CRITERIA FOR THE INITIATION OF
atrial fibrillation.51,59-61 In addition, although rare, the risk of
THERAPY IN CHRONIC LYMPHOCYTIC LEUKEMIA
ventricular arrhythmias and sudden cardiac death has also
• Marrow failure: hemoglobin , 10 g/dL or platelet been shown to be higher in patients treated with ibrutinib
count , 100 109/L (relative risk, 12.4; p , .001).62 Ibrutinib use was also
• Massive, progressive, or symptomatic splenomegaly
observed to result in the development of new or worsened
• Massive, progressive, or symptomatic
lymphadenopathy hypertension in 78.3% of patients during a median follow-
• Progressive lymphocytosis with 50% increase up of 30 months. This was associated with the development
over a 2-month period, or lymphocyte doubling of major adverse cardiac events, including arrhythmias,
time , 6 months myocardial infarction, stroke, congestive heart failure, and
• Autoimmune complications poorly responsive to cardiovascular death (hazard ratio, 2.17; 95% CI, 1.08–
corticosteroids 4.38).63 Early reports on the use of acalabrutinib suggest
• Symptomatic extranodal involvement a possibly lower incidence of atrial fibrillation, but long-term
• Disease-related symptoms: weight loss 10% over and comparative data are currently lacking.64 Moreover,
6 months, significant fatigue, fevers for 2 weeks there is limited mechanistic understanding of the etiology of
without infection, night sweats 1 month without this AE, precluding formal characterization of this AE as
infection
a class effect versus a drug effect.
Adapted from International Workshop on CLL 2018
guidelines.15 RECOMMENDATIONS FOR THE MANAGEMENT OF
CARDIOVASCULAR ISSUES WITH BTK INHIBITORS
1. All patients should undergo a detailed baseline clinical
glycoprotein VI pathway of platelet aggregation, but de- evaluation for cardiovascular comorbidities and a review
finitive clinical comparative data are currently lacking.48,49,56 of concomitant medications, particularly if they have an
Moreover, the interaction of these agents with CYP3A4 established history of cardiovascular issues.
inducers or inhibitors can significantly affect drug con- 2. Patients should be monitored closely for drug interactions
centrations, thus making it difficult to predict the risk of and AEs, including palpitations and chest pain while on
bleeding and guide optimal management practices in treatment with BTK inhibitors.
routine clinical practice.57,58 3. Blood pressure should be monitored serially and man-
aged closely with primary care physicians or cardiology.
RECOMMENDATIONS FOR THE MANAGEMENT OF BLEEDING Special attention should be paid to the use of con-
RISKS WITH BTK INHIBITORS comitant CYP3A4 inhibitors commonly used in this
setting, including diltiazem, verapamil, and amiodarone.
1. Evaluate the need for anticoagulation and the use of
antiplatelet agents for patients who are being considered Other AEs of Special Interest With BTK Inhibitors
for treatment with BTK inhibitors, with an attempt to
discontinue these agents in appropriate clinical scenarios. Infectious complications are a major cause of morbidity
2. Avoid the use of concomitant medications that can affect and mortality in patients with CLL.65 Whereas ibrutinib
drug concentrations for both ibrutinib and acalabrutinib. may potentially result in the potentiation of T-cell–mediated
This includes but are not limited to medications for other immune effects66 and improved normal B-cell function by
medical conditions or fungal infections, supplements, decreasing overall disease burden, its use is still associated
vitamin E, fish oil, nonsteroidal anti-inflammatory drugs, with a risk of infectious complications, although it is not clear
and CYP3A4 inducers and inhibitors. that this is elevated over the typical population of patients
3. Patients and providers must be educated on the im- with CLL. Risk of varicella-zoster virus reactivation and
portance of holding BTK inhibitors for 3 to 7 days before Pneumocystis jiroveci pneumonia infections may also be
a minor or major surgical procedure, respectively. Treat- increased, but data are conflicting. Although there have
ment can be resumed after the surgical procedure once been small reports about the risk of these events in pre-
hemostasis has been achieved. viously untreated patients, other reports have contradicted
these findings.67-69 BTK inhibitors also seem to elevate the
risk of Aspergillus infections because of the importance of
Cardiovascular Issues With BTK Inhibitors BTK in the macrophage response to Aspergillus.70,71 The
Ibrutinib use has been shown to substantially increase the risk absolute risk to individual patients, however, remains
of atrial fibrillation (relative risk, 4.69; 95% CI, 2.17–7.64). small.69 It is anticipated that earlier use of these agents and
Multiple reports have established this risk and identified use of prophylactic strategies in high-risk patients will result

Awan et al
in a decline of the incidence of these AEs in the future. most patients experiencing laboratory tumor lysis only. A
Nevertheless, patients must be observed closely for the slow dose ramp-up schedule with supportive care, including
development of infectious complications and treatment uricosuric agents and increased hydration, can significantly
should be initiated promptly. mitigate the risk of tumor lysis syndrome in some patients.74
RECOMMENDATION FOR THE MANAGEMENT OF INFECTIOUS Patients with high disease burden, including those with
COMPLICATIONS WITH BTK INHIBITORS bulky lymphadenopathy greater than 5 cm and significant
leukocytosis (. 25,000/mL), should be monitored closely
1. Patients should be monitored closely for the devel- with frequent laboratories and supportive care in an in-
opment of infectious complications while on therapy. patient setting during the initial dose escalation weeks of the
Although prophylaxis for varicella-zoster virus and first cycle ramp up. Additional monitoring and supportive
Pneumocystis jiroveci pneumonia are not universal care may be required in elderly patients with a compromised
recommendations, it should be considered for patients renal and cardiac function.
with a history of these infections or those considered to
be at high risk. RECOMMENDATIONS FOR THE MANAGEMENT OF TUMOR
2. Routine use of intravenous immunoglobulin is not rec- LYSIS SYNDROME IN PATIENTS TREATED WITH VENETOCLAX
ommended except in select patients with documented
hypogammaglobulinemia and multiple infectious com- 1. Patients should be risk stratified before the initiation of
plications requiring hospital admissions or repeat anti- venetoclax. Appropriate outpatient and inpatient man-
biotic treatments. agement strategies should be used, including the use of
uricosuric agents, hydration, and routine laboratory moni-
Arthralgias, myalgias, and headaches are other frequently toring per established guidelines.
observed AEs in patients receiving treatment with BTK in-
hibitors. Arthralgias and myalgias can be particularly de- With early recognition and prompt management, and by
bilitating and can result in discontinuation of ibrutinib in using dose holds and reductions, it is anticipated that most
more than 40% of patients.43 These can be controlled with of the toxicities associated with novel agents can be man-
the use of anti-inflammatory drugs and short courses of aged successfully to decrease the incidence of drug dis-
corticosteroids.72 Headaches can be debilitating with aca- continuations and improve compliance and, ultimately,
labrutinib but are usually of low intensity and respond nicely patient outcomes.
to oral caffeine.55,64 Skin rashes are also frequently ob- CONCLUSION
served, but they can be managed fairly easily with sup-
portive care and topical emollients and corticosteroids. CLL therapy has undergone dramatic transformations
Similarly, hair thinning and brittle nails are also peculiar AEs during the past few years. Although the choices available
that have been described in these patients and can be may make discussions with patients more complicated,
managed with topical treatments without requiring ibrutinib most patients will achieve a durable remission with their first
dose reductions or discontinuation.73 line of therapy. Differing toxicities and efficacy in different
subgroups will continue to become more evident as follow-
Toxicities Associated With BCL-2 Inhibitors up with these agents increases and will likely lead to more
Venetoclax has been an extremely effective drug for patients specific recommendations for sequencing on the basis of
with CLL that inhibits BCL-2 and promotes apoptosis of fitness, comorbidities, and genomic risk. Also, new studies
malignant B cells. Venetoclax use is commonly associated of combination therapies are of great interest and may
with cytopenias, which can be managed with growth factors further evolve the standard of care. For patients with re-
or dose reductions as needed.12,74 Short courses of growth lapsed disease, new targeted agents with differing mech-
factors can be used for grade 4 neutropenia to minimize anisms of action and newer strategies, like chimeric antigen
dose interruptions and reductions. Tumor lysis syndrome is receptor T (CAR-T) cells and other immunotherapies, offer
a well-recognized and potentially life-threatening AE asso- additional promise to extend quality of life. We expect that
ciated with the use of venetoclax; however, as a result of the therapy will continue to evolve during the next decade
adherence to guidelines on the prevention of tumor lysis, the and continue to improve the quantity and quality of life for
incidence of clinical tumor lysis syndrome is fairly low, with our patients with CLL.
AFFILIATIONS
3
Division of Hematology, Department of Internal Medicine, The Ohio State
1
Harold C. Simmons Comprehensive Cancer Center, University of Texas University Comprehensive Cancer Center, Columbus, OH
Southwestern Medical Center, Dallas, TX
2
Department of Internal Medicine and Center of Integrated Oncology
Cologne Bonn, University Hospital, Cologne, Germany


Jennifer A. Woyach, MD, The Ohio State University Comprehensive Cancer AND DATA AVAILABILITY STATEMENT
Center, 455D Wiseman Hall CCC, 410 W 12th Ave, Columbus, OH Disclosures provided by the authors and data availability statement (if
43210; email: [email protected]. applicable) are available with this article at DOI https://doi.org/10.1200/
EDBK_279099.
REFERENCES
1. Shanafelt TD, Rabe KG, Kay NE, et al. Age at diagnosis and the utility of prognostic testing in patients with chronic lymphocytic leukemia. Cancer. 2010;
116:4777-4787.
2. Stilgenbauer S, Zenz T. Understanding and managing ultra high-risk chronic lymphocytic leukemia. Hematology (Am Soc Hematol Educ Program). 2010;
2010:481-488.
3. Parikh SA, Strati P, Tsang M, et al. Should IGHV status and FISH testing be performed in all CLL patients at diagnosis? A systematic review and meta-analysis.
Blood. 2016;127:1752-1760.
4. Fischer K, Bahlo J, Fink AM, et al. Long-term remissions after FCR chemoimmunotherapy in previously untreated patients with CLL: updated results of the CLL8
trial. Blood. 2016;127:208-215.
5. Hallek M, Fischer K, Fingerle-Rowson G, et al; German Chronic Lymphocytic Leukaemia Study Group. Addition of rituximab to fludarabine and cyclophos-
phamide in patients with chronic lymphocytic leukaemia: a randomised, open-label, phase 3 trial. Lancet. 2010;376:1164-1174.
6. Goede V, Fischer K, Busch R, et al. Obinutuzumab plus chlorambucil in patients with CLL and coexisting conditions. N Engl J Med. 2014;370:1101-1110.
7. Eichhorst B, Fink AM, Bahlo J, et al; German CLL Study Group (GCLLSG). First-line chemoimmunotherapy with bendamustine and rituximab versus fludarabine,
cyclophosphamide, and rituximab in patients with advanced chronic lymphocytic leukaemia (CLL10): an international, open-label, randomised, phase 3, non-
inferiority trial. Lancet Oncol. 2016;17:928-942.
8. Byrd JC, Hillmen P, O’Brien S, et al. Long-term follow-up of the RESONATE phase 3 trial of ibrutinib vs ofatumumab. Blood. 2019;133:2031-2042.
9. Coutre SE, Byrd JC, Hillmen P, et al. Long-term safety of single-agent ibrutinib in patients with chronic lymphocytic leukemia in 3 pivotal studies. Blood Adv. 2019;
3:1799-1807.
10. Woyach JA, Ruppert AS, Heerema NA, et al. Ibrutinib regimens versus chemoimmunotherapy in older patients with untreated CLL. N Engl J Med. 2018;
379:2517-2528.
11. Moreno C, Greil R, Demirkan F, et al. Ibrutinib plus obinutuzumab versus chlorambucil plus obinutuzumab in first-line treatment of chronic lymphocytic
leukaemia (iLLUMINATE): a multicentre, randomised, open-label, phase 3 trial. Lancet Oncol. 2019;20:43-56.
12. Fischer K, Al-Sawaf O, Bahlo J, et al. Venetoclax and obinutuzumab in patients with CLL and coexisting conditions. N Engl J Med. 2019;380:2225-2236.
13. Jain N, Keating M, Thompson P, et al. Ibrutinib and venetoclax for first-line treatment of CLL. N Engl J Med. 2019;380:2095-2103.
14. Kater AP, Levin MD, Niemann CU. Ibrutinib and venetoclax for first-line treatment of CLL. N Engl J Med. 2019;381:788-789.
15. Hallek M, Cheson BD, Catovsky D, et al. iwCLL guidelines for diagnosis, indications for treatment, response assessment, and supportive management of CLL.
Blood. 2018;131:2745-2760.
16. Böttcher S, Ritgen M, Fischer K, et al. Minimal residual disease quantification is an independent predictor of progression-free and overall survival in chronic
lymphocytic leukemia: a multivariate analysis from the randomized GCLLSG CLL8 trial. J Clin Oncol. 2012;30:980-988.
17. Kovacs G, Robrecht S, Fink AM, et al. Minimal residual disease assessment improves prediction of outcome in patients with chronic lymphocytic leukemia (CLL)
who achieve partial response: comprehensive analysis of two phase III studies of the German CLL Study Group. J Clin Oncol. 2016;34:3758-3765.
18. Moreton P, Kennedy B, Lucas G, et al. Eradication of minimal residual disease in B-cell chronic lymphocytic leukemia after alemtuzumab therapy is associated
with prolonged survival. J Clin Oncol. 2005;23:2971-2979.
19. Hoechstetter MA, Busch R, Eichhorst B, et al. Early, risk-adapted treatment with fludarabine in Binet stage A chronic lymphocytic leukemia patients: results of the
CLL1 trial of the German CLL Study Group. Leukemia. 2017;31:2833-2837.
20. Dighiero G, Maloum K, Desablens B, et al; French Cooperative Group on Chronic Lymphocytic Leukemia. Chlorambucil in indolent chronic lymphocytic
leukemia. N Engl J Med. 1998;338:1506-1514.
21. CLL Trialists’ Collaborative Group. Chemotherapeutic options in chronic lymphocytic leukemia: a meta-analysis of the randomized trials. J Natl Cancer Inst. 1999;
91:861-868.
22. The French Cooperative Group on Chronic Lymphocytic Leukemia. Effects of chlorambucil and therapeutic decision in initial forms of chronic lymphocytic
leukemia (stage A): results of a randomized clinical trial on 612 patients. Blood. 1990;75:1414-1421.
23. Herling CD, Cymbalista F, Groß-Ophoff-Müller C, et al. Early treatment with FCR versus watch and wait in patients with stage Binet A high-risk chronic lymphocytic
leukemia (CLL): a randomized phase 3 trial. Leukemia. Epub 18 Feb 2020.
24. Langerbeins P, Bahlo J, Rhein C, et al. The CLL12 trial protocol: a placebo-controlled double-blind phase III study of ibrutinib in the treatment of early-stage
chronic lymphocytic leukemia patients with risk of early disease progression. Future Oncol. 2015;11:1895-1903.
25. Langerbeins P, Bahlo J, Rhein C, et al. Ibrutinib versus placebo in patients with asymptomatic, treatment-naive early stage CLL: primary endpoint results of the
phase 3 double-blind randomized CLL12 trial. Hematol Oncol. 2019;37:38-40.

Awan et al
26. Shanafelt TD, Wang XV, Kay NE, et al. Ibrutinib-rituximab or chemoimmunotherapy for chronic lymphocytic leukemia. N Engl J Med. 2019;381:432-443.
27. Sharman JP, Banerji V, Fogliatto LM, et al. ELEVATE TN: phase 3 study of acalabrutinib combined with obinutuzumab (O) or alone vs O plus chlorambucil (Clb) in
patients (Pts) with treatment-naive chronic lymphocytic leukemia (CLL). Blood. 2019;134 (suppl 1):31.
28. Hillmen P, Robak T, Janssens A, et al; COMPLEMENT 1 Study Investigators. Chlorambucil plus ofatumumab versus chlorambucil alone in previously untreated
patients with chronic lymphocytic leukaemia (COMPLEMENT 1): a randomised, multicentre, open-label phase 3 trial. Lancet. 2015;385:1873-1883.
29. Thompson PA, Tam CS, O’Brien SM, et al. Fludarabine, cyclophosphamide, and rituximab treatment achieves long-term disease-free survival in IGHV-mutated
chronic lymphocytic leukemia. Blood. 2016;127:303-309.
30. Rossi D, Terzi-di-Bergamo L, De Paoli L, et al. Molecular prediction of durable remission after first-line fludarabine-cyclophosphamide-rituximab in chronic
lymphocytic leukemia. Blood. 2015;126:1921-1924.
31. Byrd JC, Brown JR, O’Brien S, et al; RESONATE Investigators. Ibrutinib versus ofatumumab in previously treated chronic lymphoid leukemia. N Engl J Med. 2014;
371:213-223.
32. Burger JA, Tedeschi A, Barr PM, et al; RESONATE-2 Investigators. Ibrutinib as initial therapy for patients with chronic lymphocytic leukemia. N Engl J Med. 2015;
373:2425-2437.
33. Byrd JC, Hillmen P, O’Brien S, et al. Long-term follow-up of the RESONATE phase 3 trial of ibrutinib vs ofatumumab. Blood. 2019;133:2031-2042.
34. Shanafelt T, Wang V, Kay N, et al. Ibrutinib and rituximab provides superior clinical outcome compared to FCR in younger patients with chronic lymphocytic
leukemia (CLL): extended follow-up from the E1912 trial. Blood. 2019;134:33.
35. . Goede V, Fischer K, Dyer MJS, et al. Overall survival benefit of obinutuzumab over rituximab when combined with chlorambucil in patients with chronic
lymphocytic leukemia and comorbidities: final survival analysis of the CLL11 study. Presented at: EHA 23. Stockholm, Sweden; 2018. Abstract S151.
36. Souers AJ, Leverson JD, Boghaert ER, et al. ABT-199, a potent and selective BCL-2 inhibitor, achieves antitumor activity while sparing platelets. Nat Med. 2013;
19:202-208.
37. Stilgenbauer S, Eichhorst B, Schetelig J, et al. Venetoclax in relapsed or refractory chronic lymphocytic leukaemia with 17p deletion: a multicentre, open-label,
phase 2 study. Lancet Oncol. 2016;17:768-778.
38. Fischer K, Ritgen M, Al-Sawaf O, et al. Quantitative analysis of minimal residual disease (MRD) shows high rates of undetectable MRD after fixed-duration
chemotherapy-free treatment and serves as surrogate marker for progression-free survival: a prospective analysis of the randomized CLL14 trial. Blood. 2019;
134:36.
39. Jones JA, Mato AR, Wierda WG, et al. Venetoclax for chronic lymphocytic leukaemia progressing after ibrutinib: an interim analysis of a multicentre, open-label,
phase 2 trial. Lancet Oncol. 2018;19:65-75.
40. Mato A, Sail K, Sarraf Yazdy M, et al. Treatment sequences and outcomes of patients with CLL treated with venetoclax and other novel agents post introduction of
novel therapies. Blood. 2019;134:1756.
41. Rogers K, Huang Y, Stark A, et al. Initial results of the phase 2 treatment naive cohort in a phase 1b/2 study of obinutuzumab, ibrutinib, and venetoclax in chronic
lymphocytic leukemia. Blood. 2017;130:431.
42. Tam C, Siddiqi T, Allan JN, et al. Ibrutinib (Ibr) plus venetoclax (Ven) for first-line treatment of chronic lymphocytic leukemia (CLL)/small lymphocytic lymphoma
(SLL): results from the MRD cohort of the phase 2 CAPTIVATE study. Blood. 2019;134:35.
43. Mato AR, Nabhan C, Thompson MC, et al. Toxicities and outcomes of 616 ibrutinib-treated patients in the United States: a real-world analysis. Haematologica.
2018;103:874-879.
44. Jones JA, Hillmen P, Coutre S, et al. Use of anticoagulants and antiplatelet in patients with chronic lymphocytic leukaemia treated with single-agent ibrutinib. Br
J Haematol. 2017;178:286-291.
45. Levade M, David E, Garcia C, et al. Ibrutinib treatment affects collagen and von Willebrand factor-dependent platelet functions. Blood. 2014;124:3991-3995.
46. Lipsky AH, Farooqui MZ, Tian X, et al. Incidence and risk factors of bleeding-related adverse events in patients with chronic lymphocytic leukemia treated with
ibrutinib. Haematologica. 2015;100:1571-1578.
47. Kamel S, Horton L, Ysebaert L, et al. Ibrutinib inhibits collagen-mediated but not ADP-mediated platelet aggregation. Leukemia. 2015;29:783-787.
48. Nicolson PLR, Hughes CE, Watson S, et al. Inhibition of Btk by Btk-specific concentrations of ibrutinib and acalabrutinib delays but does not block platelet
aggregation mediated by glycoprotein VI. Haematologica. 2018;103:2097-2108.
49. Denzinger V, Busygina K, Jamasbi J, et al. Optimizing platelet GPVI inhibition versus haemostatic impairment by the Btk inhibitors ibrutinib, acalabrutinib, ONO/
GS-4059, BGB-3111 and evobrutinib. Thromb Haemost. 2019;119:397-406.
50. Mock J, Kunk PR, Palkimas S, et al. Risk of major bleeding with ibrutinib. Clin Lymphoma Myeloma Leuk. 2018;18:755-761.
51. Wiczer TE, Levine LB, Brumbaugh J, et al. Cumulative incidence, risk factors, and management of atrial fibrillation in patients receiving ibrutinib. Blood Adv.
2017;1:1739-1748.
52. Caron F, Leong DP, Hillis C, et al. Current understanding of bleeding with ibrutinib use: a systematic review and meta-analysis. Blood Adv. 2017;1:772-778.
53. Brown JR, Moslehi J, Ewer MS, et al. Incidence of and risk factors for major haemorrhage in patients treated with ibrutinib: an integrated analysis. Br J Haematol.
2019;184:558-569.
54. Kander EM, Zhao Q, Bhat SA, et al. Venous and arterial thrombosis in patients with haematological malignancy during treatment with ibrutinib. Br J Haematol.
2019;187:399-402.

55. Byrd JC, Harrington B, O’Brien S, et al. Acalabrutinib (ACP-196) in relapsed chronic lymphocytic leukemia. N Engl J Med. 2016;374:323-332.
56. Series J, Garcia C, Levade M, et al. Differences and similarities in the effects of ibrutinib and acalabrutinib on platelet functions. Haematologica. 2019;
104:2292-2299.
57. de Jong J, Skee D, Murphy J, et al. Effect of CYP3A perpetrators on ibrutinib exposure in healthy participants. Pharmacol Res Perspect. 2015;3:e00156.
58. Zhou D, Podoll T, Xu Y, et al. Evaluation of the drug-drug interaction potential of acalabrutinib and its active metabolite, ACP-5862, using a physiologically-based
pharmacokinetic modeling approach. CPT Pharmacometrics Syst Pharmacol. 2019;8:489-499.
59. Caldeira D, Alves D, Costa J, et al. Ibrutinib increases the risk of hypertension and atrial fibrillation: systematic review and meta-analysis. PLoS One. 2019;
14:e0211228.
60. Leong DP, Caron F, Hillis C, et al. The risk of atrial fibrillation with ibrutinib use: a systematic review and meta-analysis. Blood. 2016;128:138-140.
61. Brown JR, Moslehi J, O’Brien S, et al. Characterization of atrial fibrillation adverse events reported in ibrutinib randomized controlled registration trials.
Haematologica. 2017;102:1796-1805.
62. Guha A, Derbala MH, Zhao Q, et al. Ventricular arrhythmias following ibrutinib initiation for lymphoid malignancies. J Am Coll Cardiol. 2018;72:697-698.
63. Dickerson T, Wiczer T, Waller A, et al. Hypertension and incident cardiovascular events following ibrutinib initiation. Blood. 2019;134:1919-1928.
64. Byrd JC, Wierda WG, Schuh A, et al. Acalabrutinib monotherapy in patients with relapsed/refractory chronic lymphocytic leukemia: updated phase 2 results.
Blood. Epub 2019 Dec 26.
65. Morrison VA. Infectious complications of chronic lymphocytic leukaemia: pathogenesis, spectrum of infection, preventive approaches. Best Pract Res Clin
Haematol. 2010;23:145-153.
66. Dubovsky JA, Beckwith KA, Natarajan G, et al. Ibrutinib is an irreversible molecular inhibitor of ITK driving a Th1-selective pressure in T lymphocytes. Blood.
2013;122:2539-2549.
67. Ahn IE, Jerussi T, Farooqui M, et al. Atypical Pneumocystis jirovecii pneumonia in previously untreated patients with CLL on single-agent ibrutinib. Blood. 2016;
128:1940-1943.
68. Byrd JC, Furman RR, Coutre SE, et al. Three-year follow-up of treatment-naı̈ve and previously treated patients with CLL and SLL receiving single-agent ibrutinib.
Blood. 2015;125:2497-2506.
69. Rogers KA, Mousa L, Zhao Q, et al. Incidence of opportunistic infections during ibrutinib treatment for B-cell malignancies. Leukemia. 2019;33:2527-2530.
70. Herbst S, Shah A, Mazon Moya M, et al. Phagocytosis-dependent activation of a TLR9-BTK-calcineurin-NFAT pathway co-ordinates innate immunity to
Aspergillus fumigatus. EMBO Mol Med. 2015;7:240-258.
71. Bercusson A, Colley T, Shah A, et al. Ibrutinib blocks Btk-dependent NF-ĸB and NFAT responses in human macrophages during Aspergillus fumigatus
phagocytosis. Blood. 2018;132:1985-1988.
72. Stephens DM, Byrd JC. How I manage ibrutinib intolerance and complications in patients with chronic lymphocytic leukemia. Blood. 2019;133:1298-1307.
73. Bitar C, Farooqui MZ, Valdez J, et al. Hair and nail changes during long-term therapy with ibrutinib for chronic lymphocytic leukemia. JAMA Dermatol. 2016;
152:698-701.
74. Roberts AW, Davids MS, Pagel JM, et al. Targeting BCL2 with venetoclax in relapsed chronic lymphocytic leukemia. N Engl J Med. 2016;374:311-322.

Recent Advances in Managing Acute

Lymphoblastic Leukemia
Daniel J. DeAngelo, MD, PhD1; Elias Jabbour, MD2; and Anjali Advani, MD3
overview
Acute lymphoblastic leukemia (ALL) is characterized by chromosomal translocations and somatic mutations
that lead to leukemogenesis. The incorporation of pediatric-type regimens has improved survival in young
adults, and the incorporation of tyrosine kinase inhibitors for patients with Philadelphia chromosome–positive
disease has led to further improvements in outcomes. However, older patients often have poor-risk biology and
reduced tolerance to chemotherapy, leading to lower remission rates and overall survival. Regardless of age,
patients with relapsed or refractory ALL have extremely poor outcomes. The advent of next-generation se-
quencing has facilitated the revolution in understanding the genetics of ALL. New genetic risk stratification
together with the ability to measure minimal residual disease, leukemic blasts left behind after cytotoxic
chemotherapy, has led to better tools to guide postremission approaches—that is, consolidation chemo-
therapy or allogeneic stem cell transplantation. In this article, we discuss the evolving and complex genetic
landscape of ALL and the emerging therapeutic options for patients with relapsed/refractory ALL and older
patients with ALL.
INTRODUCTION Regardless of age, relapsed disease is an insur-

ALL is a rare disease, with 5,930 new cases diagnosed mountable problem for many patients. The higher rate
in the United States in 2019 (0.3% of all cancers).1 ALL of relapse among adults with ALL is probably attrib-
is increasingly recognized as a genetically heteroge- utable to higher-risk disease. For example, more adult
neous disease, and a flurry of new genetic subtypes has patients have adverse cytogenetics and molecu-
led to a refinement in risk stratification. Although most lar features (i.e., the Philadelphia chromosome or
cases of ALL are diagnosed in the pediatric population, Philadelphia-like signature) and persistent MRD. Sal-
the incidence follows a bimodal pattern, with the first vage chemotherapy regimens have shown only modest
peak occurring in children younger than age 5 and activity for patients with relapsed or refractory ALL.14,15
a second peak occurring at approximately age 50.2 Monoclonal antibodies, novel immunotherapies, and
Measurement of minimal residual disease (MRD) has kinase inhibitors for specific genetic subtypes are
led to prompt risk-adapted interventions and improved entering the clinic in the upfront and relapsed/
outcomes.3,4 Tremendous progress has been made in refractory (R-R) setting, with promising results.
the treatment of children with ALL, with complete re-
mission (CR) rates of 95% and estimated 5-year event- HOW DO I PROFILE ALL WITH CURRENT
free survival (EFS) rates of 80% to 85%.5-7 Unfortunately, MOLECULAR DATA?
the results for adult patients have not kept up with Risk stratification by cytogenetics and molecular ge-
those of their pediatric counterparts. Despite CR rates netics reflects the prognostic heterogeneity of ALL that
of approximately 85% with adult regimens, the 3-year determines which patients are at a high risk of relapse
Author affiliations EFS and overall survival (OS) rates remain below and should be considered for more intensive treat-
and support
45%.8,9 Pediatric-inspired regimens that rely heavily ment strategies, including allogeneic stem cell trans-
information (if
applicable) appear
on intensification of corticosteroids, pegylated aspar- plantation (ASCT). Risk assessment at diagnosis based
at the end of this aginase, vinca alkaloids, and antimetabolites are on immunophenotypic and genetic factors will allow
article. currently being used for young adult patients, leading stratification of patients into standard-risk and high-risk
Accepted on March to improvements in the EFS and OS rates when categories. In addition, patient characteristics, in-
2, 2020 and compared with historical controls.10-13 However, these cluding advanced age, poor performance status, and
published at
agents are poorly tolerated by older patients, resulting hyperleukocytosis, are all recognized adverse risk
ascopubs.org on May
18, 2020: DOI https://
in poor survival rates. Therefore, newer and lower- factors predicting poor outcomes of chemotherapy. In
doi.org/10.1200/ intensity therapies incorporating novel immunother- both pediatric and adult studies, persistence of MRD,
EDBK_280175 apies and now being tested for older patients with ALL. defined as more than 10 4 blasts, has proven to be

somatic mutations typical of myeloid malignancies.27 In

a retrospective analysis, 17% of adult patients with T-ALL or
lymphoblastic lymphoma were found to have the charac-
• Cytogenetics and molecular profiling are nec- teristic ETP-ALL immunophenotype. These patients, when
essary for appropriate risk stratification in pa-
treated with hyper-CVAD (i.e., hyperfractionated cyclo-
tients with ALL.
phosphamide, vincristine, doxorubicin, and dexametha-
• Assessment of the Philadelphia-like signature is sone) had significantly lower remission rates (73% vs. 91%;
essential for decisions regarding postremission p = .03) and a shorter median OS (20 months vs. not
therapy and treatment at relapse.
reached; p = .008).28 In contrast, in a cohort of pediatric and
• The assessment of minimal residual disease adolescent patients treated with a pediatric regimen (Chil-
remains the most important assessment for risk dren’s Oncology Group AALL0434), ETP-ALL was not as-
stratification in ALL. sociated with adverse outcomes, but postremission therapy
• Novel immunotherapy agents are more effective was based on MRD status.29
than conventional therapy for patients with re-
lapsed or refractory disease. Cytogenetic Classification
• The treatment of older patients with ALL re- The most common chromosomal abnormalities observed in
mains a difficult endeavor with high therapy-
children with B-cell ALL are hyperdiploidy (. 50 chromo-
related toxicity and poor long-term survival, and
somes) and the ETV6-RUNX1 (t12;21) subtypes, both of
therefore novel approaches are needed.
which are associated with favorable outcomes (Table 1).30
Adolescent and young adult patients with ALL have a much
a major independent risk factor for relapse.16 Thus, MRD lower incidence of hyperdiploidy and ETV6-RUNX131 and
assessment by either molecular- or flow-based approaches a higher incidence of unfavorable cytogenetics, including
has become standard clinical practice not only for pediatric Philadelphia-positive ALL, hypodiploidy, and complex kar-
but also for adult patients.17 yotype. The most important cytogenetic abnormality is the
Philadelphia chromosome. It is seen in only 3% of pediatric
Immunophenotypic Classification patients but approximately 25% of adults, and it increases
Determining the cell of origin (T-cell vs. B-cell ALL) and with age, representing approximately half of the cases in
specific immunophenotypes (e.g., Pro-B and Pre-B ALL) patients older than age 60.32 Although the outcome of
were traditionally of prognostic importance in ALL18-20; patients with Philadelphia chromosome positive–ALL has
however, more recent reports dispute their prognostic improved with the addition of a tyrosine kinase inhibitor
value.21 It remains essential to separate precursor B-cell (TKI), it remains unclear whether young adults need an
ALL from mature B-cell ALL (Burkitt) because these entities intensive chemotherapy (IC) backbone and whether ASCT
are typically treated with different chemotherapy strategies. is still necessary for patients who achieve molecular re-
Mature B-cell tumors typically express surface immu- mission. Single-agent imatinib or dasatinib plus cortico-
noglobulin and are negative for CD34 and terminal steroid therapy, pioneered by the Gruppo Italiano Malattie
deoxytransferase markers of immaturity. In B-cell ALL, Ematologiche dell’Adulto (GIMEMA),33,34 induced CR in
the expression of CD20 (. 20%) has consistently been almost all patients without risk of induction death. With TKI
shown to be associated with disease resistance and poorer plus chemotherapy combinations, CR rate exceeded 95%,
outcomes,22,23 which can be ameliorated with the adminis- but death occurred in 2% to 7% of the cases. In a ran-
tration of anti-CD20 immunotherapies, such as rituximab.24 domized trial from the Group for Research on Adult Acute
T-cell ALL accounts for approximately 15% of ALL, and with Lymphoblastic Leukemia,35 a combination of de-escalated
pediatric-based regimens adopting an MRD risk-stratified chemotherapy and TKI resulted in less induction toxicity
approach, the outcomes of T-cell ALL may be superior to and noninferior CR and survival results compared with
those of B-cell ALL. Subgroup analysis from the large standard chemotherapy plus TKI. In an MD Anderson
Medical Research Council UKALLXII/Eastern Cooperative Cancer Center study, ponatinib combined with hyper-CVAD
Oncology Group (ECOG) 2993 trial demonstrated that, in led to an excellent 83% 2-year OS, even without ASCT.36
T-cell ALL, CD1a positivity with the absence of CD13 ex- KMT2A (also known as the mixed lineage leukemia
pression was associated with improved survival. 25 Early gene [MLL] on 11q23)–rearranged ALL carries a poor
T-cell precursor (ETP) ALL is a recently recognized prognosis.30,37 The incidence of KMT2A ranges from 60% to
immunophenotypic high-risk subgroup of T-cell ALL. ETP 80% in infants with ALL. In children and adults, the in-
ALL has a characteristic immunophenotype characterized cidence of KMT2A gene rearrangements ranges from 4.5%
by CD1aneg, CD8neg, and CD5weak with aberrant expression of to 5.7%. One of the most common 11q23 abnormalities,
myeloid markers.26 Interestingly, ETP-ALL often harbors t(4;11), occurs in 2% of children and adults with ALL.

DeAngelo, Jabbour, and Advani
TABLE 1. Common Chromosomal and Molecular Abnormalities in B-Cell ALL

Cytogenetics Gene Incidence in Adults Incidence in Children
Hyperdiploidy (. 50 chromosomes) — 7% 25%–30%
t(12;21)(p13;q22) ETV6-RUNX1 (TEL-AML1) 2% 22%–25%
t(9;22)(q34;q11): Philadelphia chromosome BCR-ABL1 25% 2%–4%
t(4:11)(q21;q23) and other KNMT2A translocations KNMT2A 8%–10% 2%–3% (60%–80% in infants)
Low hypodiploid near triploid TP53 in low hypodiploid 8%–10% 2%–3%
t(1;19)(q23;p13) TCF3-PBX1 3% 4%
t(11;14)(q11), e.g., (p13;q11), (p15;q11) TCRα and TCRδ 20%–25% 10%–20%
BCR-ABL1-like, Philadelphia-like Various 10%–30% 15%
Ikaros IKZF1 25%–35% 12%–17%
KMT2A rearrangements seldom express CD10, and patients with Philadelphia-like ALL have a poorer outcome
therefore the blasts have a common immunophenotype of compared with other B-cell ALLs, and it remains unclear
CD19+, CD10 , CD15+, or CD65+. Interestingly, they are whether they should receive ASCT up front or based on
associated with high levels of H3K79 methylation catalyzed MRD persistence.45 The Children’s Oncology Group is
by the DOTL1 enzyme, suggesting a novel therapeutic testing ruxolitinib in patients with CRLF2-rearranged and
target.38,39 In addition, multiple studies have shown high JAK-STAT dysregulated disease (NCT02723994).
levels of FLT3 expression in patients with KMT2A rear- The most common mutation in classic T-cell ALL is
rangements; therefore, inhibitors of FLT3 (a tyrosine kinase) NOTCH1, whereas DNMT3A and other myeloid-specific
may also prove beneficial.40 In some patients, KMT2A gene mutations are more commonly seen in ETP-ALL.27 NOTCH1
rearrangements are not detected by conventional cytoge- can be targeted by the gamma-secretase inhibitors, which
netics, and commercially available dual-color KMT2A have been tested in patients with advanced late-stage
probes can detect translocations in metaphase chromo- disease, with some durable responses but at the expense of
somes by fluorescence in situ hybridization techniques. gut toxicity.48
TCF3-PBX1, associated with t(1;19) translocation, is char-
Minimal Residual Disease
acterized by favorable outcomes with intensive treat-
ment. Hypodiploid ALL is a poor prognostic subtype, which Patients with detectable residual blasts after chemotherapy
includes near-haploid (24–31 chromosomes), low-hypodiploid (MRD positive, typically defined as more than 10 4 blasts)
(32–39 chromosomes), and high-hypodiploid (40–43 chro- are seldom cured with chemotherapy alone. In pro-
mosomes) groups. RAS and PI3K pathways are frequently spective trials,17,49,50 the OS was between 60% and 80%
mutated in near-haploid ALL, whereas TP53 and IKZF are with chemotherapy alone in patients with MRD-negative
often mutated in low-hypodiploid ALL.41 Therefore, germline status, even in high-risk subsets and Ph-positive ALL.
mutational screening of TP53 should always be performed in Conversely, patients with MRD-positive status benefit par-
these cases. tially from ASCT but with OS rates of 50% or less in intention-
to-treat analyses because of the cumulative effects of
Molecular Classification pretransplantation and post-transplantation relapse and
The Philadelphia-like subgroup, initially identified by means transplant-related deaths.4,51,52
of gene expression profiling, accounts for approximately Monitoring MRD with real-time quantitative polymerase
20% of young adult and adult B-cell ALL cases (Table 1). chain reaction, flow cytometry, or high-throughput–based
These cases are characterized by a transcriptional profile platforms should be considered a standard practice for the
similar to that of Philadelphia-positive ALL but lack the t(9;22)/ treatment of patients with ALL because it is the best pre-
BCR-ABL1 rearrangement.42-45 Instead, the underlying ge- dictor of OS. Similar robust prognostic data are seen in both
nomic lesions are heterogeneous, making its recognition B-cell and T-cell ALL. In the analysis of standard-risk pa-
difficult and uneven between trials. Rearrangements in tients enrolled on the German Multicenter Study Group for
CRLF2 are seen in approximately 50% of cases; rear- Acute Lymphoblastic Leukemia study, early MRD clearance
rangements in the ABL class genes (ABL1, ABL2, CSF1R, before day 24 was associated with a favorable outcome
PDGFRA, PDGFRB), in roughly 10%; and JAK/STAT genes (relapse rate of 0% at 3 years), whereas MRD positivity
(JAK1-3, IL7R, and CRLF2 mutations), less often (, 10%). lasting beyond week 16 predicted a relapse rate of 94%.53
Deletions in IKZF1 occur in up to 80% of cases and are Therefore, patients who achieve CR but fail to achieve
associated with a particularly poor prognosis.46,47 In general, a negative MRD status remain at high risk of early relapse

and should seek alternative therapy.4 Similarly, MRD status binding region.62,63 In the confirmatory phase II study of 189
before SCT is also a powerful indicator for post-transplantation patients with R-R Philadelphia-negative ALL, blinatumomab
outcomes and disease resistance.54 was associated with a CR-plus–CR with partial hematologic
MRD is such a powerful predictor for relapse, so it is cur- recovery rate of 43%. The median response duration was 9
rently used to guide postinduction decisions. Bassan et al55 months, and the median OS was 6 months.64
allocated patients to maintenance or SCT based on MRD After these results, a phase III trial (TOWER study) was
status at the end of intensive therapy (approximately conducted: more than 400 patients with R-R Philadelphia-
5 months from attaining CR) regardless of their cytogenetic negative ALL were randomly assigned (2:1) to blinatumo-
and clinical risk score at diagnosis, resulting in an im- mab (n = 271) or standard of care (n = 134; Table 2).59 The
pressive 75% 5-year OS rate for nontransplanted patients response rates were 44% and 25%, respectively (p , .001).
attaining MRD-negative status. This approach is being used Molecular remission rates among responders, defined as
by other large cooperative groups with similar success.56 fewer than 10 4 blasts in the first 12 weeks, were 76% and
Taken together, these data suggest that MRD may indeed 48%, respectively. Blinatumomab prolonged survival, the
be the most important predictor of OS for patients with ALL. primary study endpoint; the median OS was 7.7 months
Future Directions (95% CI, 5.6–9.6) versus 4.0 months (95% CI, 2.9–5.3; p =
.01; hazard ratio, 0.71). A total of 24% of the patients in
We are entering an intensive phase of clinical investigations each group underwent ASCT.65
using next-generation sequencing data for the exploration of
individualized or subset-specific treatment algorithms. A recent phase III study demonstrated the superiority of
Therefore, it will be crucial to design prospective clinical blinatumomab for 208 children and adolescent young
studies with modular data to evaluate optimal strategies adults in salvage 1 and randomly assigned (1:1) to bli-
for specific patient subtypes, or “personalized” medicine, natumomab or standard of care.66 Blinatumomab induced
testing immunotherapy and combinations of molecularly a higher rate of measurable residual disease negativity (79%
targeted drugs or drug combinations. vs. 21%; p , .001) and a higher 2-year OS (79% vs. 59%;
p = .005).
TREATING RELAPSED-REFRACTORY ALL: TOO MANY
OPTIONS, WHICH ONE TO PICK? Blinatumomab was evaluated in the phase II ALCANTARA
trial with 45 patients who had R-R Philadelphia-positive
Therapies targeted toward specific transcripts (e.g., BCR- ALL.67 Thirty-six percent achieved a response. The median
ABL1 tyrosine kinase oncoprotein by TKIs) and specific relapse-free survival and OS were 6.7 months and 7.1
leukemic cell surface antigens (e.g., CD20, CD22, and months, respectively; 44% of patients underwent ASCT.
CD19 monoclonal antibodies) are major breakthroughs in Furthermore, blinatumomab was given with TKIs to 20 pa-
the management of ALL.57 Historically, relapsed or re- tients with R-R Philadelphia-positive ALL or chronic myeloid
fractory (R-R) ALL was associated with a dismal prognosis, leukemia in the lymphoid blast phase. The response rate was
with a cure rate of less than 10%.14,58 The CR rates with 65%. The median survival was 14 months.68
standard chemotherapy regimens are 30% to 40% in first
relapse and 20% to 25% in second relapse. Accordingly, Gökbuget et al69 assessed blinatumomab in 113 patients
only 10% to 30% adult patients with relapsed ALL proceed with ALL in MRD-positive CR. Approximately 78% of patients
to ASCT, which is the only curative option in this setting.14 achieved MRD negativity after one cycle and 80 after two
Recent developments of monoclonal antibodies, bispecific cycles. With a median follow-up of 29 months, the median OS
antibody constructs, and CAR T-cell therapies have revo- was 36.5 months and the relapse-free survival was 18.9
lutionized the treatment of ALL, resulting in U.S. Food and months. The median OS with achievement of MRD negativity
Drug Administration (FDA) approvals of blinatumomab in was 38.9 months compared with 12.5 months for patients with
2014 and inotuzumab and tisagenlecleucel in 2017 as ALL persistent MRD. These findings were confirmed when com-
salvage strategies.59-61 Their use in combination with other pared with historical data via propensity score matching.70
treatment modalities in the salvage and frontline settings are Blinatumomab is being evaluated in the frontline setting
under investigation. in combination with chemoimmunotherapy in Philadelphia-
negative ALL and with TKIs in Philadelphia-positive ALL.
Anti-CD19 Bispecific T-Cell Engager:
Blinatumomab Anti-CD22 Antibody-Drug Conjugate:
Blinatumomab is an anti-CD19–directed CD3 bispecific T-cell Inotuzumab Ozogamicin
engager constructed with BiTE antibody technology.62-64 It Inotuzumab ozogamicin is a novel anti-CD22 monoclonal
consists of a recombinant monoclonal antibody com- antibody conjugated to the toxin calicheamicin.71 In a single-
posed of an anti-CD19 fragment antigen-binding region institution, phase II study of patients with R-R ALL, inotu-
joined by a short link to an anti-CD3 fragment antigen- zumab was administered at a starting intavenous dosage of

TABLE 2. Relapsed and Refractory Trials in ALL

Median Overall Survival Minimal Residual Disease
Clinical Trial Treatment Setting Patients Response (months) Negativitya
Blinatumomab
TOWER4 R-R Ph ALL 271 ORR: 44% 7.7 76%
12
ALCANTARA R-R Ph+ ALL 45 ORR: 36% 7.1 88%b
Inotuzumab Ozogamicin
INO-VATE5 R-R Ph– and Ph+ 164 ORR: 74% 7.7 78%
ALL
Inotuzumab plus mini-hyper-CVD 6 R-R Ph– ALL 89 ORR: 79% S1: 25 82%
blinatumomab23
S2: 6
S3: 7
CAR T
ELIANA82 R-R ALL 79 ORR: 81% NR 100%
Abbreviations: ALL, acute lymphoblastic leukemia; mini-hyper-CVD, mini-hyperfractionated cyclophosphamide, vincristine, and dexamethasone; NR, not
reached; Ph, Philadelphia chromosome; ORR, overall response rate; R-R, relapsed or refractory; S1, salvage 1; S2, salvage 2; S3, salvage 3.
a
Measured by flow cytometry.
b
Within two cycles.
1.3 to 1.8 mg/m2 every 3 to 4 weeks. Forty-nine patients were 65%, respectively. The 12-month survival rate was 40%.
treated.72 The objective response rate was 57%, and the The VOD rate was 10.4%; it was 26% after ASCT.
median survival was 5.1 months. Nearly half of the patients Inotuzumab was also evaluated with a dose-reduced mini-
treated with inotuzumab proceeded to ASCT. To minimize hyper-CVD regimen (i.e., mini-hyperfractionated cyclo-
toxicities and based on pharmacokinetic and pharmaco- phosphamide, vincristine, and dexamethasone) with or
dynamic data, inotuzumab was administered intravenously without blinatumomab in 89 patients with R-R ALL.77 The
on a weekly basis at 0.8 mg/m2 on day 1, followed by 0.5 mg/ response rate was 79%, with 82% of responders achieving
m2 on days 8 and 15, every 3 to 4 weeks in 40 patients. The MRD negativity. The 2-year progression-free survival and OS
study yielded a similar response rate to inotuzumab given rates were 52% and 39%, respectively. Among patients
every 3 to 4 weeks (59% vs. 57%), with a median survival of treated in salvage 1, the 2-year survival rate was 51%. The
9.5 months.73 Weekly administration of inotuzumab resulted VOD rate was 6%.
in fewer adverse events, including lower rates of veno-
occlusive disease (VOD). In a separate multicenter phase Better outcomes were obtained in patients who received
II trial in heavily pretreated patients with R-R ALL, inotu- inotuzumab- or blinatumomab-based therapies in first
zumab therapy resulted in a remission rate of 66%, with 78% salvage.65,78-81 This was particularly notable in patients who
of patients who achieved CR becoming MRD negative. The achieved a negative MRD status and received subsequent
median survival was 7.4 months.74 ASCT. ASCT performed in subsequent salvages did not
confer a survival advantage.80
These results led to a randomized trial comparing inotu-
zumab with physician’s choice of chemotherapy in R-R CAR T-Cell Therapies
ALL.60,74 The response rate was 88% (CR, 81%) with ino- CAR T Cells are genetically modified autologous T lym-
tuzumab and 32% (CR, 29%) with standard of care (p , phocytes engineered to express binding sites of specific
.0001). Among responders, the MRD negativity rates were antibodies, such as a receptor against CD19. These T cells,
78% and 28% (p , .0001), respectively. The median harnessed from the patients’ own immune systems, target
survival was 7.7 versus 6.7 months (p = .02; hazard ratio, the malignant cells.
0.77); The 2-year survival rate was 23% versus 10%.75 In a phase II multicenter study, 79 children and young
Serious toxicities included VOD after ASCT, mainly in pa- adults with relapsed or refractory CD19+ B-ALL received
tients who received double alkylators in pretransplantation a single infusion of tisagenlecleucel, CD19 CAR T-cells.61,82
conditioning. Age was also a risk factor for VOD. Among evaluable patients, the overall response rate was
Inotuzumab was assessed in 48 children, adolescents, and 81%, with all patients whose disease responded achieving
young adults (median age, 9; range, 1–21 years).76 The negative MRD. The 24-month relapse-free survival and OS
objective response and MRD negativity rates were 62% and rates were 62% and 66%, respectively. The CAR T cells

persisted for a median of 168 days (range, 20–617 days), agents have demonstrated single-agent and synergistic cell
suggesting that immunosurveillance with these products killing when combined with chemotherapy because of the
may be long lasting. Grade 3 and 4 cytokine release syn- upregulation of Bcl-2 in these cell lines.89,90 In a phase I trial,
drome (CRS) and neurotoxicities were encountered in 49% 36 patients with R-R ALL (median age, 27; range, 6–72
and 40%, respectively, with 48% needing an intensive care years) were treated with venetoclax and navitoclax in
unit stay. This led to the approval of tisagenlecleucel for R-R combination with chemotherapy. The objective response
ALL for patients up to age 25 after failure of two previous rate was 50%; the 6-month CR duration rate was 43%.92
therapies. Preliminary results of the combination of venetoclax with
low-intensity chemotherapy in newly diagnosed older pa-
For 53 adults with heavily pretreated B-ALL who received
tients unfit for IC are promising, with objective response and
the CAR T-cell infusion, the CR rate was 83%, including 32
MRD negativity rates of 91% and 100%, respectively.93
(67%) of 48 evaluable patients achieving negative MRD.83
Better outcomes were observed for patients with low disease Future Directions
burden ( 5% bone marrow blasts) at the time of CAR T-cell
Given the encouraging results achieved with monoclonal
infusion, with a median EFS of 10.6 months compared with
antibodies, bispecific antibody constructs, and CAR T cells,
5.3 months and a median OS of 20.1 months compared
the therapeutic tools necessary to improve outcomes of
with 12.4 months for patients with a high disease burden.
patients with adult ALL may now be available. These
Several strategies to optimize CAR T-cell activities and treatment modalities are not competitive but rather com-
minimize their toxicities have been explored. The use of plementary, and they could be administered sequentially to
CD19 CAR T cells with a lower-affinity CD19 binder with produce the deepest remissions possible. Their rational
a fast off-rate would lead to physiologic T-cell activation, combination in the frontline setting is ongoing and may
reduced toxicity, improved engraftment, and potential long- reduce the need for long-term IC and obviate ASCT for many
term persistence to deliver sustained responses.84 In a pilot patients.
trial of 16 adults treated, the objective response and MRD
negativity rates were 87% each. The 6-month OS rate was THE OLDER PATIENT WITH ALL
66%. No grade 3 cases of CRS were reported; three patients The definition of “older” or “elderly” adults with ALL has
experienced reversible grade 3 neurotoxicity.85 typically included patients age 60 and older. Historically,
their prognosis has been dismal, with a 5-year OS of ap-
To circumvent CD19 escape as a cause of relapse after
proximately 20%.94,95 This prognosis has been attributed to
CD19 CAR T-cell therapy, CD22-targeted CAR T-cell ther-
adverse biology and the inability of older patients to tolerate
apy has been developed.86 Of the 15 children and adults
IC. However, the advent of new immunotherapies and TKIs,
with R-R B-ALL, most of whom were previously treated with
incorporation of MRD to guide therapy, and the increasing
CD-19–directed immunotherapy, 11 (73%) achieved CR
safety of ASCT for older patients are leading to improved
after treatment with at least 1 106/kg body weight of CAR
outcomes. However, these new approaches make man-
T cells. Encouraging results from a phase I trial of dual
agement of these cases more complicated. In this section,
CD19/CD22 bispecific CAR T-cells in 21 children and adults
we focus on disease biology, standard treatment, and new
with R-R ALL reported grade 3 to 4 CRS and neurotoxicity in
approaches to therapy in the older patient with ALL.
only two patients (10%), with CR and MRD negativity rates
of 86% and 81%, respectively.87 Biologic Concerns for Older Patients With ALL
Current therapies use autologous lymphocytes, which can The incidence of high-risk genetic abnormalities among
be scarce and difficult to expand. New platforms provide an older adults with ALL is high. The Philadelphia chromosome
off-the-shelf approach, with cells derived from healthy is present in approximately 50% of these patients.94,96
volunteer donors. In the phase I dose escalation trial in Before the advent of TKIs, the Philadelphia chromo-
children and adults with R-R ALL, the objective response some was associated with a poor prognosis; however, TKIs
and MRD negativity rates were 82% and 71%, re- have dramatically improved outcomes.97 More recently,
spectively.88 Treatment was tolerated, with only moderate the Philadelphia-like signature has been identified in
CRS. Off-the-shelf products targeting CD22 and allogeneic approximately 24% of older adults and has been asso-
cord blood–derived natural killer cells are being developed. ciated with a poor outcome.43 The incidence of specific
fusions varies with age, with older adults expressing high
BH3 Mimetics levels of CRLF2.43 Low hypodiploidy or near triploidy,
Preclinical studies have demonstrated activity of the BH3 complex cytogenetics, IKZF1 mutations, chromosome 17
mimetics venetoclax and navitoclax in B-cell and T-cell ALL abnormalities, and KMT2A rearrangements are more
cell lines.89-91 Venetoclax has demonstrated particularly common, and high hyperdiploidy is less common, in older
strong activity in MLL-rearranged B-cell ALL, and both adults with ALL.98-103 Finally, older patients have a higher

incidence of the B-cell (89%) compared with T-cell (66%) removing asparaginase and cyclophosphamide.94,110 Al-
phenotype.98,104 though targeting CD20 with rituximab in combination with
chemotherapy has demonstrated benefit in patients younger
Outcomes of Standard Treatment for Patients With than age 60 with B-ALL, incorporating this agent for patients
Philadelphia Chromosome–Negative ALL who are at least age 60 has not demonstrated a benefit.94,111
Standard ALL treatment includes intensive induction and This finding may be related to more deaths in CR. Ribera
postremission therapy. This can be challenging for older et al112 evaluated a pediatric-inspired IC in a younger group of
adults because of their comorbidities: diabetes, cardiac older adults (age 55–65) to see whether results were more
dysfunction, pulmonary abnormalities, neuropathy, poly- favorable in this subgroup. Despite an improvement in out-
pharmacy, and decline in performance status.94,98 Although come, the 2-year EFS was only 37%,112 emphasizing the
poor outcomes have been reported in clinical trials, out- need for novel strategies.
comes are almost certainly worse when evaluating “all
comers” because many patients do not receive treatment. Patients With Philadelphia Chromosome–Positive ALL
In a Surveillance, Epidemiology, and End Results data- TKI treatment has improved the outcomes for Philadelphia
base of patients age 60 or older with ALL, the median OS chromosome–positive ALL substantially. First-generation
was 4 months, and the 3-year OS was 12.8%.98,105 In the TKIs, such as imatinib, have demonstrated higher rates
largest study conducted with IC for older adults with of remission than chemotherapy alone and low rates of
Philadelphia-negative ALL, the German Multicenter toxicity.98,113 However, remission durations have typically
Study Group for Acute Lymphoblastic Leukemia trial,98,106 been short (median duration, 8 months with imatinib).34
268 patients older than age 55 were treated with a modified Later-generation TKIs, such as dasatinib, have demon-
intensive Berlin-Frankfurt-Munster (i.e., BFM) regimen strated improved results.33 However, despite all patients
followed by postremission therapy and maintenance. Rit- achieving a hematologic remission, only 20% achieved
uximab was administered to patients with CD20-positive a major molecular remission, and the 20-month OS and
disease. The 5-year OS was only 23% despite a relatively disease-free survival rates were 69.2% and 51.1%, re-
young median age (67 years) and 62% of patients having an spectively.33 Twelve of the 17 patients who experienced
ECOG performance status of 0 to 1. An amendment was relapse had the resistant T315I Bcr-Abl mutation,33 the
incorporated into the trial, adding triple intrathecal che- dominant mechanism of relapse among patients treated
motherapy, adding asparaginase during postremission with earlier-generation TKIs.33,98,114 Ponatinib is the only
therapy, and increasing the dosages of methotrexate and active TKI in this setting. The GIMEMA group evaluated
cytarabine. The 2-year OS increased from 33% to 52%; ponatinib in combination with steroids.98,115 The rates of CR,
however, long-term outcomes remained disappointing.98,106 complete molecular remission, and 1-year OS were 95%,
With most IC, the rates of CR have decreased with ad- 46%, and 87.5%, respectively. Thirteen serious adverse
vancing age and the rates of early death have increased events related to ponatinib occurred (in 44 patients).94,115
(Table 3).106 Increased age has been associated with Therefore, caution is necessary, given the cardiovascular
a higher risk of infections during induction (81% vs. 70%) risk profile of ponatinib, and the drug dose should be re-
and more dose reductions (46% vs. 28%).107 In patients age duced once complete molecular remission is achieved.
60 or older treated with hyper-CVAD, the death rate in CR These results are encouraging, however, the median follow-
has been approximately 34%.94,108 Because of the in- up is short (11.4 months). Longer follow-up is needed.
creased rates of early deaths, most IC regimens have used
For fit older adults, TKI therapy can be combined with IC.
a standard backbone with dosage reductions to avoid tox-
Imatinib has been combined effectively with various che-
icities. The Cancer and Leukemia Group B reduced the
motherapy regimens: UKALLXII/ECOG 2993, the North
number of days of steroids to 7 days (from 21 days) during
Italian Leukemia Group multiagent chemotherapy back-
induction for older patients.109 In a PETHEMA trial, re-
bone, GRAAPH-2005 with reduced-intensity induction, and
searchers increased the 2-year OS from 39% to 52% by
hyper-CVAD.35,116-119 Although the CR rates have been more
than 90%, higher than those of chemotherapy alone, the
TABLE 3. Rates of Complete Remission and Early Death by Age Group rates of EFS and OS have varied (EFS, 33%–42%; OS,
in the German Multicenter Study Group for Acute Lymphoblastic 38%–48%).35,116-119 This variation may be related to the
Leukemia Study14 regimen, median patient age, and percentage of patients
Age Range (years) Complete Remission Rate Early Death Rate ultimately undergoing ASCT. Notably, the median age of
55–65 84% 7% patients in most of these trials was in the mid-40s, and OS
66–75 74% 14%
was significantly shorter in patients age 60 or older,
compared with those younger than age 60, stressing the
. 75 52% 37%
need for more effective approaches for older adults.116

The second-generation TKIs dasatinib and nilotinib (63% vs. 34%) compared with hyper-CVAD.125 A phase III
have been combined with chemotherapy, with EFS and trial of mini-hyper-CVD with or without inotuzumab is planned.
OS rates in the range of 42% to 55% and 46% to 72%,
Blinatumomab, an anti-CD19 bispecific T-cell engaging
respectively.114,116,120-122 Finally, a phase II trial evaluated
antibody, has been FDA approved for R-R B-ALL and has
the third-generation TKI ponatinib with hyper-CVAD.36,98
also demonstrated superiority to standard chemotherapy.59
This trial enrolled 66 patients, with 20 patients younger
This drug’s unique toxicity profile has included CRS and
than age 60. The rates of CR and complete molecular re-
neurotoxicity reminiscent of that seen with CAR T cells but at
mission were 100% and 77%, respectively, with a 3-year OS
a much lower incidence. In a phase II trial (SWOG1318),
of 77%. These results are impressive and demonstrated
patients at least age 65 with newly diagnosed Philadelphia-
increased EFS and OS compared with hyper-CVAD and
negative B-ALL were treated with blinatumomab for in-
other TKI-based therapies in a propensity analysis.98,123 This
duction, then received postremission therapy, followed by
analysis, in addition to the single-agent data, suggests
maintenance therapy with prednisone, vincristine, metho-
improved outcomes with successive-generation TKIs, al-
trexate, and mercaptopurine. Twenty-nine eligible patients
though no randomized trials have been reported. In addi-
were treated (median age, 75).126 The median bone marrow
tion, among patients with deep molecular remissions, there
blast count was 86.5%, and 34% of patients had poor-risk
were long-term survivors without ASCT. However, the dose
cytogenetics. The rate of response (CR and CR with in-
intensity of ponatinib correlated with adverse events, and
complete count recovery) was 66%, with no early deaths. At
the trial was amended for risk-adapted dosing, because
a median follow-up of 1.3 years, the estimated 1-year
three of 39 patients had a myocardial infarction at the
disease-free survival and OS were 58% and 67%, re-
continuous dose of 45 mg.36 Given the potential cardio-
spectively.126 Further follow-up will be needed to determine
vascular risks of ponatinib and the intensive nature of hyper-
the durability of these responses. A second clinical trial,
CVAD, this regimen is suitable only for truly fit older patients;
A041703, is sequencing inotuzumab and blinatumomab
strict organ function criteria were included in this trial.
treatment.
Given the short duration of remissions with TKIs alone and
For fit older adults, a recent U.S. Intergroup trial (ECOG
toxicities when combining with IC, other approaches are
1910) completed accrual (ages 30–70). Patients received
being evaluated, including combining dasatinib with im-
an intensive induction and postremission therapy, which
munotherapy, such as blinatumomab (SWOG 1318,
was modified based on age. There was a random assign-
NCT02143414), or combining various TKIs with low-
ment for patients who were MRD negative to either continue
dose chemotherapy.
standard therapy or receive blinatumomab followed by
Ongoing and Recent Clinical Trials for Patients With further chemotherapy. The results of this trial are eagerly
Philadelphia-Negative ALL awaited.
In consideration of the toxicities with standard IC, many Finally, the Bcl-2 inhibitor venetoclax is being evaluated in
current trials are reducing chemotherapy and adding novel both B-ALL and T-ALL. Bcl-2 is overexpressed in ALL.127 An
agents. Inotuzumab, an anti-CD22 antibody conjugated to ongoing trial evaluated the Bcl-2 inhibitor navitoclax plus
the cytotoxic agent calicheamicin, demonstrated higher venetoclax in combination with chemotherapy in R-R ALL.
rates of CR, MRD, and OS compared with standard che- Despite a heavily pretreated population, the response rate
motherapy in patients with R-R B-ALL due for salvage was encouraging at 56%.128 Another trial is evaluating
chemotherapy.60 This led to FDA approval of inotuzumab for venetoclax in combination with low-intensity chemotherapy
the treatment of R-R B-ALL. The drug is generally well for older patients with newly diagnosed ALL.129
tolerated, with a unique toxicity of hepatic VOD. Sub-
sequently, inotuzumab was combined with low-intensity Future Directions for Older Patients With ALL
chemotherapy (mini-hyper-CVD) for patients with Philadelphia- Multiple novel agents are being evaluated in Philadelphia-
negative B-ALL who were at least age 60.124 Fifty-two positive and -negative ALL. Incorporation of these agents
patients were treated (median age, 68). The incidence of with less-intensive chemotherapy is likely to improve out-
VOD was 8%, with a 12% incidence of treatment-related comes of older adults, and clinical trials are evaluating these
deaths.124 At a median follow-up of 29 months, the 2-year approaches. The presence of Philadelphia-like alterations
progression-free survival was encouraging at 59%.124 In and MLL rearrangements may make TKI-based therapy and
a propensity analysis comparing mini-hyper-CVD/inotuzumab MLL inhibitors in combination with low-dose therapy an-
with or without blinatumomab with hyper-CVAD, the anti- other option for these patients. Although CAR T cells are not
body plus low-intensity treatment demonstrated a higher yet FDA approved for adults older than age 26, these agents
response rate, lower rates of early death, lower rates of death are being used in clinical trials; as their toxicities are opti-
in remission, and higher 3-year EFS (64% vs. 34%) and OS mized, they are likely to become an option for older adults,

given their impressive clinical activity.83 The use of MRD to increasing availability of donors, transplantation is becoming
guide therapy will probably be incorporated into future trials, another option for patients who previously would not have
particularly because blinatumomab is now FDA approved in been candidates. Recent studies demonstrate encouraging
the MRD-positive setting. Finally, as outcomes with results for appropriate “elderly” candidates.130
reduced-intensity ASCT are improving and there is an

1
Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, Daniel J. DeAngelo, MD, PhD, Dana-Farber Cancer Institute, 450
MA Brookline Ave., Boston, MA 02215; email: daniel_deangelo@dfci.
2
The University of Texas MD Anderson Cancer Center, Houston, TX harvard.edu.
3
Cleveland Clinic Taussig Cancer Institute, Cleveland, OH
EDBK_280175.
REFERENCES
1. National Cancer Institute Surveillance, Epidemiology, and End Results Program. Cancer Stat Facts: Leukemia - Acute Lymphocytic Leukemia (ALL). https://
seer.cancer.gov/statfacts/html/alyl.html. Accessed February 10, 2020.
2. Sant M, Allemani C, Tereanu C, et al; HAEMACARE Working Group. Incidence of hematologic malignancies in Europe by morphologic subtype: results of the
HAEMACARE project. Blood. 2010;116:3724-3734.
3. Coustan-Smith E, Gajjar A, Hijiya N, et al. Clinical significance of minimal residual disease in childhood acute lymphoblastic leukemia after first relapse.
Leukemia. 2004;18:499-504.
4. Gökbuget N, Kneba M, Raff T, et al; German Multicenter Study Group for Adult Acute Lymphoblastic Leukemia. Adult patients with acute lymphoblastic
leukemia and molecular failure display a poor prognosis and are candidates for stem cell transplantation and targeted therapies. Blood. 2012;120:1868-1876.
5. Pui CH, Relling MV, Downing JR. Acute lymphoblastic leukemia. N Engl J Med. 2004;350:1535-1548.
6. Pui CH, Campana D, Pei D, et al. Treating childhood acute lymphoblastic leukemia without cranial irradiation. N Engl J Med. 2009;360:2730-2741.
7. Larsen EC, Devidas M, Chen S, et al. Dexamethasone and high-dose methotrexate improve outcome for children and young adults with high-risk B-acute
lymphoblastic leukemia: a report from Children’s Oncology Group study AALL0232. J Clin Oncol. 2016;34:2380-2388.
8. Kantarjian H, Thomas D, O’Brien S, et al. Long-term follow-up results of hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone
(Hyper-CVAD), a dose-intensive regimen, in adult acute lymphocytic leukemia. Cancer. 2004;101:2788-2801.
9. Larson RA, Dodge RK, Linker CA, et al. A randomized controlled trial of filgrastim during remission induction and consolidation chemotherapy for adults with
acute lymphoblastic leukemia: CALGB study 9111. Blood. 1998;92:1556-1564.
10. DeAngelo DJ, Silverman LB, Couban S, et al. A multicenter phase II study using a dose intensified pediatric regimen in adults with untreated acute lymphoblastic
leukemia. Blood. 2006;108 (abstr 1858).
11. DeAngelo DJ, Stevenson KE, Dahlberg SE, et al. Long-term outcome of a pediatric-inspired regimen used for adults aged 18-50 years with newly diagnosed
acute lymphoblastic leukemia. Leukemia. 2015;29:526-534.
12. Huguet F, Leguay T, Raffoux E, et al. Pediatric-inspired therapy in adults with Philadelphia chromosome–negative acute lymphoblastic leukemia: the GRAALL-
2003 study [published correction appears in J Clin Oncol. 2009;27(15):2574]. J Clin Oncol. 2009;27:911-918.
13. Stock W, Luger SM, Advani AS, et al. A pediatric regimen for older adolescents and young adults with acute lymphoblastic leukemia: results of CALGB 10403.
Blood. 2019;133:1548-1559.
14. Fielding AK, Richards SM, Chopra R, et al; Eastern Cooperative Oncology Group. Outcome of 609 adults after relapse of acute lymphoblastic leukemia (ALL); an
MRC UKALL12/ECOG 2993 study. Blood. 2007;109:944-950.
15. Thomas DA, Kantarjian H, Smith TL, et al. Primary refractory and relapsed adult acute lymphoblastic leukemia: characteristics, treatment results, and prognosis
with salvage therapy. Cancer. 1999;86:1216-1230.
16. Berry DA, Zhou S, Higley H, et al. Association of minimal residual disease with clinical outcome in pediatric and adult acute lymphoblastic leukemia: a meta-
analysis. JAMA Oncol. 2017;3:e170580.
17. van Dongen JJ, van der Velden VH, Brüggemann M, et al. Minimal residual disease diagnostics in acute lymphoblastic leukemia: need for sensitive, fast, and
standardized technologies. Blood. 2015;125:3996-4009.
18. Gökbuget N, Hoelzer D. Treatment of adult acute lymphoblastic leukemia. Semin Hematol. 2009;46:64-75.
19. Hoelzer D, Thiel E, Löffler H, et al. Prognostic factors in a multicenter study for treatment of acute lymphoblastic leukemia in adults. Blood. 1988;71:123-131.

20. Rowe JM, Buck G, Burnett AK, et al; MRC/NCRI Adult Leukemia Working Party. Induction therapy for adults with acute lymphoblastic leukemia: results of more
than 1500 patients from the international ALL trial: MRC UKALL XII/ECOG E2993. Blood. 2005;106:3760-3767.
21. Rowe JM. Prognostic factors in adult acute lymphoblastic leukaemia. Br J Haematol. 2010;150:389-405.
22. Maury S, Huguet F, Leguay T, et al; Group for Research on Adult Acute Lymphoblastic Leukemia. Adverse prognostic significance of CD20 expression in adults
with Philadelphia chromosome–negative B-cell precursor acute lymphoblastic leukemia. Haematologica. 2010;95:324-328.
23. Thomas DA, O’Brien S, Jorgensen JL, et al. Prognostic significance of CD20 expression in adults with de novo precursor B-lineage acute lymphoblastic
leukemia. Blood. 2009;113:6330-6337.
24. Maury S, Chevret S, Thomas X, et al; for GRAALL. Rituximab in B-Lineage adult acute lymphoblastic leukemia. N Engl J Med. 2016;375:1044-1053.
25. Marks DI, Paietta EM, Moorman AV, et al. T-cell acute lymphoblastic leukemia in adults: clinical features, immunophenotype, cytogenetics, and outcome from
the large randomized prospective trial (UKALL XII/ECOG 2993). Blood. 2009;114:5136-5145.
26. Coustan-Smith E, Mullighan CG, Onciu M, et al. Early T-cell precursor leukaemia: a subtype of very high-risk acute lymphoblastic leukaemia. Lancet Oncol.
2009;10:147-156.
27. Zhang J, Ding L, Holmfeldt L, et al. The genetic basis of early T-cell precursor acute lymphoblastic leukaemia. Nature. 2012;481:157-163.
28. Jain N, Lamb AV, O’Brien S, et al. Early T-cell precursor acute lymphoblastic leukemia/lymphoma (ETP-ALL/LBL) in adolescents and adults: a high-risk subtype.
Blood. 2016;127:1863-1869.
29. Dunsmore KP, Winter S, Devidas M, et al. COG AALL0434: a randomized trial testing nelarabine in newly diagnosed T-cell malignancy. J Clin Oncol. 2018;36
(suppl; abstr 10500).
30. Moorman AV, Ensor HM, Richards SM, et al. Prognostic effect of chromosomal abnormalities in childhood B-cell precursor acute lymphoblastic leukaemia:
results from the UK Medical Research Council ALL97/99 randomised trial. Lancet Oncol. 2010;11:429-438.
31. Aguiar RC, Sohal J, van Rhee F, et al; M.R.C. Adult Leukaemia Working Party. TEL-AML1 fusion in acute lymphoblastic leukaemia of adults. Br J Haematol.
1996;95:673-677.
32. Iacobucci I, Mullighan CG. Genetic basis of acute lymphoblastic leukemia. J Clin Oncol. 2017;35:975-983.
33. Foà R, Vitale A, Vignetti M, et al; GIMEMA Acute Leukemia Working Party. Dasatinib as first-line treatment for adult patients with Philadelphia
chromosome–positive acute lymphoblastic leukemia. Blood. 2011;118:6521-6528.
34. Vignetti M, Fazi P, Cimino G, et al. Imatinib plus steroids induces complete remissions and prolonged survival in elderly Philadelphia chromosome–positive
patients with acute lymphoblastic leukemia without additional chemotherapy: results of the Gruppo Italiano Malattie Ematologiche dell’Adulto (GIMEMA)
LAL0201-B protocol. Blood. 2007;109:3676-3678.
35. Chalandon Y, Thomas X, Hayette S, et al; Group for Research on Adult Acute Lymphoblastic Leukemia (GRAALL). Randomized study of reduced-intensity
chemotherapy combined with imatinib in adults with Ph-positive acute lymphoblastic leukemia [published correction appears in Blood. 2015;126:1261].
Blood. 2015;125:3711-3719.
36. Short NJ, Kantarjian HM, Ravandi F, et al. Frontline hyper-CVAD plus ponatinib for patients with Philadelphia chromosome–positive acute lymphoblastic
leukemia: updated results of a phase II study. J Clin Oncol. 2017;35 (suppl; abstr 7013).
37. Moorman AV, Harrison CJ, Buck GA, et al; Adult Leukaemia Working Party, Medical Research Council/National Cancer Research Institute. Karyotype is an
independent prognostic factor in adult acute lymphoblastic leukemia (ALL): analysis of cytogenetic data from patients treated on the Medical Research Council
(MRC) UKALLXII/Eastern Cooperative Oncology Group (ECOG) 2993 trial. Blood. 2007;109:3189-3197.
38. Armstrong SA, Staunton JE, Silverman LB, et al. MLL translocations specify a distinct gene expression profile that distinguishes a unique leukemia. Nat Genet.
2002;30:41-47.
39. Stein EM, Garcia-Manero G, Rizzieri DA, et al. The DOT1L inhibitor pinometostat reduces H3K79 methylation and has modest clinical activity in adult acute
leukemia. Blood. 2018;131:2661-2669.
40. Armstrong SA, Mabon ME, Silverman LB, et al. FLT3 mutations in childhood acute lymphoblastic leukemia. Blood. 2004;103:3544-3546.
41. Holmfeldt L, Wei L, Diaz-Flores E, et al. The genomic landscape of hypodiploid acute lymphoblastic leukemia. Nat Genet. 2013;45:242-252.
42. Herold T, Schneider S, Metzeler KH, et al. Adults with Philadelphia chromosome–like acute lymphoblastic leukemia frequently have IGH-CRLF2 and JAK2
mutations, persistence of minimal residual disease and poor prognosis. Haematologica. 2017;102:130-138.
43. Roberts KG, Gu Z, Payne-Turner D, et al. High frequency and poor outcome of Philadelphia chromosome–like acute lymphoblastic leukemia in adults. J Clin
Oncol. 2017;35:394-401.
44. Roberts KG, Li Y, Payne-Turner D, et al. Targetable kinase-activating lesions in Ph-like acute lymphoblastic leukemia. N Engl J Med. 2014;371:1005-1015.
45. Roberts KG, Pei D, Campana D, et al. Outcomes of children with BCR-ABL1–like acute lymphoblastic leukemia treated with risk-directed therapy based on the
levels of minimal residual disease. J Clin Oncol. 2014;32:3012-3020.
46. Mullighan C, Downing J. Ikaros and acute leukemia. Leuk Lymphoma. 2008;49:847-849.
47. Mullighan CG, Miller CB, Radtke I, et al. BCR-ABL1 lymphoblastic leukaemia is characterized by the deletion of Ikaros. Nature. 2008;453:110-114.
48. Zweidler-McKay PA, DeAngelo DJ, Douer D, et al. The safety and activity of BMS-906024, a gamma secretase inhibitor (GSI) with anti-notch activity, in patients
with relapsed T-cell acute lymphoblastic leukemia (T-ALL): initial results of a phase 1 trial. Blood. 2014;124:968.

49. Bassan R. Using minimal residual disease to improve treatment response definitions and hematopoietic cell transplantation strategy in acute leukemia. J Clin
Oncol. 2016;34:300-302.
50. Brüggemann M, Raff T, Kneba M. Has MRD monitoring superseded other prognostic factors in adult ALL? Blood. 2012;120:4470-4481.
51. Bassan R, Spinelli O, Oldani E, et al. Different molecular levels of post-induction minimal residual disease may predict hematopoietic stem cell transplantation
outcome in adult Philadelphia-negative acute lymphoblastic leukemia. Blood Cancer J. 2014;4:e225.
52. Dhédin N, Huynh A, Maury S, et al; GRAALL group. Role of allogeneic stem cell transplantation in adult patients with Ph-negative acute lymphoblastic leukemia.
Blood. 2015;125:2486-2496, quiz 2586.
53. Brüggemann M, Raff T, Flohr T, et al; German Multicenter Study Group for Adult Acute Lymphoblastic Leukemia. Clinical significance of minimal residual
disease quantification in adult patients with standard-risk acute lymphoblastic leukemia. Blood. 2006;107:1116-1123.
54. Spinelli O, Peruta B, Tosi M, et al. Clearance of minimal residual disease after allogeneic stem cell transplantation and the prediction of the clinical outcome of
adult patients with high-risk acute lymphoblastic leukemia. Haematologica. 2007;92:612-618.
55. Bassan R, Spinelli O, Oldani E, et al. Improved risk classification for risk-specific therapy based on the molecular study of minimal residual disease (MRD) in
adult acute lymphoblastic leukemia (ALL). Blood. 2009;113:4153-4162.
56. Ribera J-M, Oriol A, Morgades M, et al. Post-remission treatment with chemotherapy or allogeneic hematopoietic stem cell transplantation (alloHSCT) in adult
patients with high-risk (HR) Philadelphia chromosome–negative (Ph-neg) acute lymphoblastic leukemia (ALL) according to their minimal residual disease
(MRD). Final results of the Pethema ALL-HR-11 Trial. Blood. 2019;134 (suppl 1; abstr 826).
57. Jabbour E, Pui CH, Kantarjian H. Progress and innovations in the management of adult acute lymphoblastic leukemia. JAMA Oncol. 2018;4:1413-1420.
58. Gökbuget N, Dombret H, Ribera JM, et al. International reference analysis of outcomes in adults with B-precursor Ph-negative relapsed/refractory acute
lymphoblastic leukemia. Haematologica. 2016;101:1524-1533.
59. Kantarjian H, Stein A, Gökbuget N, et al. Blinatumomab versus chemotherapy for advanced acute lymphoblastic Leukemia. N Engl J Med. 2017;376:836-847.
60. Kantarjian HM, DeAngelo DJ, Stelljes M, et al. Inotuzumab ozogamicin versus standard therapy for acute lymphoblastic leukemia. N Engl J Med. 2016;375:740-753.
62. Baeuerle PA, Reinhardt C. Bispecific T-cell engaging antibodies for cancer therapy. Cancer Res. 2009;69:4941-4944.
63. Bargou R, Leo E, Zugmaier G, et al. Tumor regression in cancer patients by very low doses of a T cell-engaging antibody. Science. 2008;321:974-977.
64. Topp MS, Gökbuget N, Stein AS, et al. Safety and activity of blinatumomab for adult patients with relapsed or refractory B-precursor acute lymphoblastic
leukaemia: a multicentre, single-arm, phase 2 study. Lancet Oncol. 2015;16:57-66.
65. Jabbour EJ, Gökbuget N, Kantarjian HM, et al. Transplantation in adults with relapsed/refractory acute lymphoblastic leukemia who are treated with bli-
natumomab from a phase 3 study. Cancer. 2019;125:4181-4192.
66. Brown PA, Ji L, Xu X, et al. A randomized phase 3 trial of blinatumomab vs. chemotherapy As post-reinduction therapy in high and intermediate risk (HR/IR) first
relapse of B-acute lymphoblastic Leukemia (B-ALL) in children and adolescents/Young adults (AYAs) demonstrates superior efficacy and tolerability of
blinatumomab: a report from Children’s Oncology Group study AALL1331. Blood. 2019;134 (suppl; abstr LBA-1).
67. Martinelli G, Boissel N, Chevallier P, et al. Complete hematologic and molecular response in adult patients with relapsed/refractory Philadelphia
chromosome–positive B-precursor acute lymphoblastic leukemia following treatment with blinatumomab: results from a phase II, single-arm, multicenter study.
J Clin Oncol. 2017;35:1795-1802.
68. Assi R, Kantarjian H, Short NJ, et al. Safety and efficacy of blinatumomab in combination with a tyrosine kinase inhibitor for the treatment of relapsed
Philadelphia chromosome–positive leukemia. Clin Lymphoma Myeloma Leuk. 2017;17:897-901.
69. Gökbuget N, Dombret H, Bonifacio M, et al. Blinatumomab for minimal residual disease in adults with B-cell precursor acute lymphoblastic leukemia [published
correction appears in Blood. 2019;133:2625]. Blood. 2018;131:1522-1531.
70. Gökbuget N, Dombret H, Giebel S, et al. Blinatumomab vs historic standard-of-care treatment for minimal residual disease in adults with B-cell precursor acute
lymphoblastic leukaemia. Eur J Haematol. 2020;104:299-309.
71. Raponi S, De Propris MS, Intoppa S, et al. Flow cytometric study of potential target antigens (CD19, CD20, CD22, CD33) for antibody-based immunotherapy in
acute lymphoblastic leukemia: analysis of 552 cases. Leuk Lymphoma. 2011;52:1098-1107.
72. Kantarjian H, Thomas D, Jorgensen J, et al. Inotuzumab ozogamicin, an anti-CD22-calecheamicin conjugate, for refractory and relapsed acute lymphocytic
leukaemia: a phase 2 study. Lancet Oncol. 2012;13:403-411.
73. Kantarjian H, Thomas D, Jorgensen J, et al. Results of inotuzumab ozogamicin, a CD22 monoclonal antibody, in refractory and relapsed acute lymphocytic
leukemia. Cancer. 2013;119:2728-2736.
74. DeAngelo DJ, Stock W, Stein AS, et al. Inotuzumab ozogamicin in adults with relapsed or refractory CD22-positive acute lymphoblastic leukemia: a phase 1/2
study. Blood Adv. 2017;1:1167-1180.
75. Kantarjian HM, DeAngelo DJ, Stelljes M, et al. Inotuzumab ozogamicin versus standard of care in relapsed or refractory acute lymphoblastic leukemia: Final
report and long-term survival follow-up from the randomized, phase 3 INO-VATE study. Cancer. 2019;125:2474-2487.
76. O’Brien MM. A phase 2 trial of inotuzumab ozogamicin (InO) in children and young adults with relapsed or refractory (R/R) CD22+ B-acute lymphoblastic
leukemia (B-ALL): results from Children’s Oncology Group Protocol AALL1621. Blood. 2019;134 (suppl 1; abstr 741).
77. Jabbour E, Ravandi F, Kebriaei P, et al. Salvage chemoimmunotherapy with inotuzumab ozogamicin combined with mini-hyper-CVD for patients with relapsed
or refractory Philadelphia chromosome–negative acute lymphoblastic leukemia: a phase 2 clinical trial. JAMA Oncol. 2018;4:230-234.

78. Jabbour E, Gökbuget N, Advani A, et al. Impact of minimal residual disease status in patients with relapsed/refractory acute lymphoblastic leukemia treated with
inotuzumab ozogamicin in the phase III INO-VATE trial. Leuk Res. 2020;88:106283.
79. Jabbour E, Sasaki K, Ravandi F, et al. Chemoimmunotherapy with inotuzumab ozogamicin combined with mini-hyper-CVD, with or without blinatumomab, is
highly effective in patients with Philadelphia chromosome–negative acute lymphoblastic leukemia in first salvage. Cancer. 2018;124:4044-4055.
80. Jabbour E, Short NJ, Jorgensen JL, et al. Differential impact of minimal residual disease negativity according to the salvage status in patients with relapsed/
refractory B-cell acute lymphoblastic leukemia. Cancer. 2017;123:294-302.
81. Dombret H, Topp MS, Schuh AC, et al. Blinatumomab versus chemotherapy in first salvage or in later salvage for B-cell precursor acute lymphoblastic leukemia.
Leuk Lymphoma. 2019;60:2214-2222.
82. Grupp SA, Maude SL, Rives S, et al. Updated analysis of the efficacy and safety of tisagenlecleucel in pediatric and young adult patients with relapsed/refractory
(r/r) acute lymphoblastic leukemia. Blood. 2018;132:895.
84. Ghorashian S, Kramer AM, Onuoha S, et al. Enhanced CAR T cell expansion and prolonged persistence in pediatric patients with ALL treated with a low-affinity
CD19 CAR. Nat Med. 2019;25:1408-1414.
85. Roddie C, O’Reilly MA, Pinto JDA, et al. AUTO1, a novel fast off CD19CAR delivers durable remissions and prolonged CAR T cell persistence with low CRS or
neurotoxicity in adult ALL. Blood. 2019;134 (suppl 1; abstr 226).
86. Fry TJ, Shah NN, Orentas RJ, et al. CD22-targeted CAR T cells induce remission in B-ALL that is naive or resistant to CD19-targeted CAR immunotherapy. Nat
Med. 2018;24:20-28.
87. Schultz LM, Graham C, Yallop D, et al. Phase I trial using CD19/CD22 bispecific CAR T cells in pediatric and adult acute lymphoblastic leukemia (ALL). Blood.
2019;134 (suppl 1, abstr 744).
88. Benjamin R, Graham C, Yallop D, et al. Preliminary data on safety, cellular kinetics and anti-Leukemic activity of UCART19, an allogeneic anti-CD19 CAR T-cell
product, in a pool of adult and pediatric patients with high-risk CD19+ relapsed/refractory B-cell acute lymphoblastic leukemia. Blood. 2018;132 (suppl 1; abstr
896).
89. Benito JM, Godfrey L, Kojima K, et al. MLL-rearranged acute lymphoblastic leukemias activate BCL-2 through H3K79 methylation and are sensitive to the BCL-
2-specific antagonist ABT-199. Cell Rep. 2015;13:2715-2727.
90. Khaw SL, Suryani S, Evans K, et al. Venetoclax responses of pediatric ALL xenografts reveal sensitivity of MLL-rearranged leukemia. Blood. 2016;
128:1382-1395.
91. Frismantas V, Dobay MP, Rinaldi A, et al. Ex vivo drug response profiling detects recurrent sensitivity patterns in drug-resistant acute lymphoblastic leukemia.
Blood. 2017;129:e26-e37.
92. Lacayo NJ, Pullarkat VA, Stock W, et al. Safety and efficacy of venetoclax in combination with navitoclax in adult and pediatric relapsed/refractory acute
lymphoblastic leukemia and lymphoblastic lymphoma. Blood. 2019;134 (suppl 1; abstr 285).
93. Jain N, Stevenson KE, Winer ES, et al. A multicenter phase I study combining venetoclax with mini-hyper-CVD in older adults with untreated and relapsed/
refractory acute lymphoblastic leukemia. Blood. 2019. 134 (suppl 1; abstr 3867).
94. Chiaretti S, Foà R. Chemotherapy-free and reduced intensity approaches in elderly patients with B-lineage acute lymphoblastic leukemia. Eur J Intern Med.
2018;58:22-27.
95. Gökbuget N. Treatment of older patients with acute lymphoblastic leukemia. Hematology (Am Soc Hematol Educ Program). 2016;2016:573-579.
96. Chiaretti S, Vitale A, Cazzaniga G, et al. Clinico-biologic features of 5202 patients with acute lymphoblastic leukemia enrolled in the Italian AIEOP and GIMEMA
protocols and stratified in age cohorts. Haematologica. 2013;98:1702-1710.
97. Ribera JM, Garcı́a O, Fernández-Abellán P, et al; PETHEMA Group. Lack of negative impact of Philadelphia chromosome in older patients with acute
lymphoblastic leukaemia in the tyrosine kinase inhibitor era: comparison of two prospective parallel protocols. Br J Haematol. 2012;159:485-488.
98. Short NJ, Kantarjian H, Jabbour E, et al. Novel therapies for older adults with acute lymphoblastic leukemia. Curr Hematol Malig Rep. 2018;13:91-99.
99. Ribera J, Morgades M, Zamora L, et al; Spanish PETHEMA Group and the Spanish Society of Hematology. Prognostic significance of copy number alterations in
adolescent and adult patients with precursor B acute lymphoblastic leukemia enrolled in PETHEMA protocols. Cancer. 2015;121:3809-3817.
100. Khoral P, Atenafu EG, Craddock KJ, et al. Prognostic effect of complex karyotype, monosomal karyotype, and chromosome 17 abnormalities in B-cell acute
lymphoblastic leukemia. Clin Lymphoma Myeloma Leuk. 2017;17:215-219.
101. Issa GC, Kantarjian HM, Yin CC, et al. Prognostic impact of pretreatment cytogenetics in adult Philadelphia chromosome–negative acute lymphoblastic
leukemia in the era of minimal residual disease. Cancer. 2017;123:459-467.
102. Moorman AV, Chilton L, Wilkinson J, et al. A population-based cytogenetic study of adults with acute lymphoblastic leukemia. Blood. 2010;115:206-214.
103. Tasian SK, Hurtz C, Wertheim GB, et al. High incidence of Philadelphia chromosome–like acute lymphoblastic leukemia in older adults with B-ALL. Leukemia.
2017;31:981-984.
104. Thomas X, Olteanu N, Charrin C, et al. Acute lymphoblastic leukemia in the elderly: the Edouard Herriot Hospital experience. Am J Hematol. 2001;67:73-83.
105. Geyer MB, Hsu M, Devlin SM, et al. Overall survival among older US adults with ALL remains low despite modest improvement since 1980: SEER analysis.
Blood. 2017;129:1878-1881.

106. Goekbuget N, Beck J, Bruggeman M, et al. Moderate intensive chemotherapy including CNS prophylaxis with liposomal cytarabine is feasible and effective in
older patients with Ph-negative acute lymphoblastic leukemia (ALL): results of a prospective trial from the German Multicenter study group for adult ALL
(GMALL). Blood. 2012;120:1493.
107. Sive JI, Buck G, Fielding A, et al. Outcomes in older adults with acute lymphoblastic leukaemia (ALL): results from the international MRC UKALL XII/ ECOG 2993
trial. Br J Haematol. 2012;157:463-471.
108. O’Brien S, Thomas DA, Ravandi F, et al. Results of the hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone regimen in elderly
patients with acute lymphocytic leukemia. Cancer. 2008;113:2097-2101.
109. Larson RA, Dodge RK, Burns CP, et al. A five-drug remission induction regimen with intensive consolidation for adults with acute lymphoblastic leukemia:
cancer and leukemia group B study 8811. Blood. 1995;85:2025-2037.
110. Sancho JM, Ribera JM, Xicoy B, et al; PETHEMA Group. Results of the PETHEMA ALL-96 trial in elderly patients with Philadelphia chromosome–negative acute
lymphoblastic leukemia. Eur J Haematol. 2007;78:102-110.
111. Thomas DA, O’Brien S, Faderl S, et al. Chemoimmunotherapy with a modified hyper-CVAD and rituximab regimen improves outcome in de novo Philadelphia
chromosome–negative precursor B-lineage acute lymphoblastic leukemia. J Clin Oncol. 2010;28:3880-3889.
112. Ribera JM, Garcı́a O, Gil C, et al; PETHEMA Group. Comparison of intensive, pediatric-inspired therapy with non-intensive therapy in older adults aged 55-
65 years with Philadelphia chromosome–negative acute lymphoblastic leukemia. Leuk Res. 2018;68:79-84.
113. Ottmann OG, Wassmann B, Pfeifer H, et al; GMALL Study Group. Imatinib compared with chemotherapy as front-line treatment of elderly patients with
Philadelphia chromosome–positive acute lymphoblastic leukemia (Ph+ALL). Cancer. 2007;109:2068-2076.
114. Ravandi F, O’Brien SM, Cortes JE, et al. Long-term follow-up of a phase 2 study of chemotherapy plus dasatinib for the initial treatment of patients with
Philadelphia chromosome–positive acute lymphoblastic leukemia. Cancer. 2015;121:4158-4164.
115. Martinelli G, Piciocchi A, Papayannidis C, et al. First report of the Gimema LAL1811 phase 2 prospective study of the combination of steroids with ponatinib as
frontline therapy of elderly or unfit patients with Philadelphia chromosome–positive acute lymphoblastic leukemia. Blood. 2017; 130 (suppl 1; abstr 99).
116. Wieduwilt MJ. How should we treat older adults with Ph+ adult ALL and what novel approaches are being investigated? Best Pract Res Clin Haematol. 2017;
30:201-211.
117. Fielding AK, Rowe JM, Buck G, et al. UKALLXII/ECOG2993: addition of imatinib to a standard treatment regimen enhances long-term outcomes in Philadelphia-
positive acute lymphoblastic leukemia. Blood. 2014;123:843-850.
118. Bassan R, Rossi G, Pogliani EM, et al. Chemotherapy-phased imatinib pulses improve long-term outcome of adult patients with Philadelphia
chromosome–positive acute lymphoblastic leukemia: Northern Italy Leukemia Group protocol 09/00. J Clin Oncol. 2010;28:3644-3652.
119. Daver N, Thomas D, Ravandi F, et al. Final report of a phase II study of imatinib mesylate with hyper-CVAD for the front-line treatment of adult patients with
Philadelphia chromosome–positive acute lymphoblastic leukemia. Haematologica. 2015;100:653-661.
120. Ravandi F, O’Brien S, Thomas D, et al. First report of phase 2 study of dasatinib with hyper-CVAD for the frontline treatment of patients with Philadelphia
chromosome–positive (Ph+) acute lymphoblastic leukemia. Blood. 2010;116:2070-2077.
121. Kim DY, Joo YD, Lim SN, et al; Adult Acute Lymphoblastic Leukemia Working Party of the Korean Society of Hematology. Nilotinib combined with multiagent
chemotherapy for newly diagnosed Philadelphia-positive acute lymphoblastic leukemia. Blood. 2015;126:746-756.
122. Ravandi F, Othus M, O’Brien SM, et al. U.S. Intergroup study of chemotherapy plus dasatinib and allogeneic stem cell transplant in Philadelphia
chromosome–positive ALL. Blood Adv. 2016;1:250-259.
123. Sasaki K, Jabbour EJ, Ravandi F, et al. Hyper-CVAD plus ponatinib versus hyper-CVAD plus dasatinib as frontline therapy for patients with Philadelphia
chromosome–positive acute lymphoblastic leukemia: a propensity score analysis. Cancer. 2016;122:3650-3656.
124. Kantarjian H, Ravandi F, Short NJ, et al. Inotuzumab ozogamicin in combination with low-intensity chemotherapy for older patients with Philadelphia
chromosome–negative acute lymphoblastic leukaemia: a single-arm, phase 2 study. Lancet Oncol. 2018;19:240-248.
125. Jabbour EJ, Sasaki K, Ravandi F, et al. Inotuzumab ozogamicin in combination with low-intensity chemotherapy (mini-HCVD) with or without blinatumomab
versus standard intensive chemotherapy (HCVAD) as frontline therapy for older patients with Philadelphia chromosome–negative acute lymphoblastic leu-
kemia: a propensity score analysis. Cancer. 2019;125:2579-2586.
126. Advani AS, Moseley A, O’Dwyer KM, et al. Results of SWOG 1318: a phase 2 trial of blinatumomab followed by POMP (prednisone, vincristine, methotrexate, 6-
mercaptopurine) maintenance in elderly patients with newly diagnosed Philadelphia chromosome–negative B-cell acute lymphoblastic leukemia. Blood. 2018;
132 (suppl 1; abstr 33).
127. Alford SE, Kothari A, Loeff FC, et al. BH3 inhibitor sensitivity and bcl-2 dependence in primary acute lymphoblastic leukemia cells. Cancer Res. 2015;
75:1366-1375.
128. Lacayo N, Pullarkat V, Stock W, et al. Safety and efficacy of venetoclax in combination with navitoclax in adult and pediatric relapsed/refractory acute
lymphoblastic leukemia and lymphoblastic lymphoma. Blood. 2019;134 (suppl; abstr 285).
129. Jain N, Stevenson K, Winder ES, et al. A multicenter phase 1 study combining venetoclax with mini-hyper-CVD in older adults with untreated and relapsed/
refractory acute lymphoblastic leukemia. Blood. 2019 (abstr 3867).
130. Rosko A, Wang HL, de Lima M, et al. Reduced intensity conditioned allograft yields favorable survival for older adults with B-cell acute lymphoblastic leukemia.
Am J Hematol. 2017;92:42-49.

The Changing Landscape of Treatment in Acute

Myeloid Leukemia
Kristin Koenig, MD1; Alice Mims, MD, MSCR2; Mark J. Levis, MD, PhD3; and Mary M. Horowitz, MD, MS4
overview
The treatment of acute myeloid leukemia is evolving, with increased understanding of molecular pathogenesis
allowing better risk stratification and development of new therapies. Tests to identify and drugs to target
specific molecular abnormalities are improving remission rates and prolonging survival in patients with high-
risk disease. Allogeneic hematopoietic stem cell transplantation remains an important curative therapy, with
advances in donor availability and approaches to reduce transplant-related mortality making it applicable in
many more patients. Considerations in identifying appropriate patients for targeted therapy and trans-
plantation are presented.
INTRODUCTION In current models of the pathogenesis of AML, the

Acute myeloid leukemia (AML) is the most common disease develops from sequential acquisition of key
acute leukemia in adults, with more than 20,000 cases mutations within the hematopoietic stem cell com-
diagnosed in the United States annually. Increased partment. An AML genome has fewer mutations than
understanding of the disease and the development of other cancers, and relatively few are required to
new therapies now allow more personalized and ef- transform a normal stem cell into a leukemia stem
fective treatment strategies. This review outlines the cell.1,2 These few mutations are drawn from a larger
use of new molecular testing tools to estimate prog- group of 50 to 100, but AML pathogenesis is not as
nosis and guide therapy, describes recently approved simple as having three or four random mutations from
targeted therapies, and discusses the role of allogeneic this larger group occur in a stem cell.3 Certain muta-
hematopoietic stem cell transplantation (alloHCT) in tions tend to co-occur, and there are clear patterns of
the current era. different molecularly defined AML subtypes. In par-
ticular, mutations in so-called landscaping genes oc-
cur early, whereas mutations in proliferative genes
INTERPRETING MOLECULAR TESTS IN ACUTE
follow later.4 A common pattern, for example, is a
MYELOID LEUKEMIA
mutation in an epigenetic modifier such as ASXL1,
The past decade saw the advent of next-generation DNMT3a, or TET2 occurring as a founder mutation.
sequencing (NGS) technologies and, with it, a re- That clone might expand and persist for years (as
markable leap in our understanding of the biology of Clonal Hematopoiesis of Indeterminate Potential or
AML. New AML drugs received regulatory approval, CHIP).5,6 A mutation in the nucleophosmin 1 gene
with several directed against subtypes of the disease (NPM1) might follow, and then a mutation in a growth
that are defined by molecular tests. In the recent past, factor pathway, such as an NRAS or FLT3 mutation,
facing a new patient with AML meant that the oncol- will often be the final hit prior to the emergence of overt
ogist had to choose between intensive treatment,
AML. Conventional karyotype abnormalities certainly fit
Author affiliations nonintensive treatment, or no treatment, with a limited
into this model. The initial epigenetic mutation might
and support repertoire of mostly nonselective cytotoxic agents.
information (if be followed by a fusion involving a transcription factor
Now, however, a new patient with AML often causes
applicable) appear (e.g., CBFB/MYH11) and then a KIT D816 mutation.
at the end of this
the oncologist to freeze in place, waiting—or deciding
The NGS mutation report can be examined almost like
article. how long to wait—for results of molecular testing before
an archeological record, with the founder mutation
Accepted on March choosing a treatment plan. Indeed, now more than
recognized as being in all of the cells and each sub-
2, 2020 and ever, prognosis and treatment are dependent on the
published at sequent mutation present in a smaller fraction.
ability to accurately interpret and correctly act on the
ascopubs.org on Every patient’s AML is thus unique, but mutations tend
results of molecular testing. It is, therefore, incumbent
April 2, 2020:
upon us to understand as much as possible about the to cluster into groups, conferring predictable clinical
10.1200/EDBK_ nature of these tests and the biology of the disease to features. Understanding how mutations evolve helps
279129 which they are applied. us more effectively treat patients. For example, detecting

Koenig et al
hybridization of specific chromosomal fusions or deletions

can overcome this problem, but this technique is usually
limited to a relatively narrow set of well-defined genetic
• Obtaining and correctly interpreting the results defects, such as loss of chromosomes 5, 7, 8, 11, and 20, or
of molecular testing is now an essential com-
known fusions, like those in core-binding factor AML.
ponent in the optimal management of acute
myeloid leukemia. NGS mutation panels are important new diagnostic tools
• Molecular sequencing should be checked at for AML. However, they have notable limitations. NGS
both diagnosis and time of relapse because panels estimate mutation burden using the variant allele
acute myeloid leukemia mutations change over fraction (VAF), which is the fraction of total alleles with
the patient’s disease course and could offer new a specific mutation. For heterozygous mutations (e.g.,
therapy options. IDH1 or IDH2 mutations), the VAF should be close to 0.5.
• Transplantation should be considered early in However, the multiple different clones within the pop-
patients at high risk of leukemia relapse, in- ulation of cells are essentially ground up into a homoge-
cluding in patients as old as age 75. nous DNA mixture before sequencing. If there are
• Most patients will have a suitable donor, in- nonmalignant (or premalignant) cells in the mix, which is
cluding related (fully matched and haplo- often the case, even a mutation present in all leukemia
identical) and unrelated donors and umbilical cells will have a VAF less than 0.5. To correctly interpret
cord blood units. molecular testing, one must know (using microscopy and,
• Targeted therapies offer new treatment ap- usually, flow cytometry) what was submitted for testing.
proaches and allow options for older patients Was it a homogenous population of blasts or a sample
(age 75) or other patient populations who containing 10% blasts in a background of normal he-
may not have been considered treatment matopoietic cells (or dysplastic cells)? In general, when
candidates in the past. requesting an NGS panel, it is best to analyze as pure
a sample of leukemia as possible. If the patient has a low
white blood cell count with few circulating blasts, marrow
the persistence of a founding landscape gene mutation in
should be used. If the patient has a high white blood cell
a bone marrow sample that looks morphologically normal
count with virtually all blasts, peripheral blood is ap-
tells us the patient still has clonal hematopoiesis but not
propriate (and quicker to obtain). Because of the rela-
leukemia.7 On the other hand, detecting a persisting FLT3
tively high error rate inherent in the polymerases used for
or RAS mutation—a late mutation—indicates residual
NGS, single-nucleotide mutations generally must be
disease. Of course, any interpretation of the molecular data
present in at least 5% of reads to be believable, which
and its implications for treatment must be overlaid on the
defines the sensitivity of this approach. This must be
clinical context: Is the patient newly diagnosed or re-
factored into the clinician’s interpretation of the results
lapsed? Is the patient older or younger, fit or less fit? What
and correlated with how much leukemia was present in
comorbid conditions does the patient have? In other words,
the sample analyzed. Another limitation to be considered
decisions about treatment should not be based solely on
is the time frame for turn-around of NGS results, which
the molecular profile, especially persistent “founder mu-
can be 7 days or longer. For some patients, such as those
tations.”
with rapidly progressing disease, this delay in treatment is
Mutation Burden unacceptable.
Each patient’s disease is molecularly heterogeneous, with FLT3 mutations are another matter entirely. NGS panels do
multiple subclones present at the time of diagnosis.2 The not routinely detect internal tandem duplication (ITD)
founding mutation, if it can be identified, is expected to be in mutations (the most common type) because of their variable
100% of the leukemia cells, but different subclones can length and the fact that they consist of normal (though
harbor different cooperating mutations. Our ability to dis- duplicated) sequences.8 Although NGS platforms can be
cern the burden of mutations within a sample of leukemia adapted to detect these mutations, most FLT3-ITD muta-
cells is hampered by the nature of the available assays. For tions are detected using conventional polymerase chain
example, a chromosomal translocation at 11q23 might be reaction. The mutation burden for FLT3-ITD mutations is
present in all malignant cells, but the technique of kar- usually expressed not as a VAF but as an allelic ratio,
yotyping leukemia with G-banding of metaphase cells calculated by dividing the amount of mutant (peak height)
relies on obtaining dividing cells. Some nonmalignant cells by the amount of wild type. The sensitivity of polymerase
undergoing cell division may be included in the analy- chain reaction for FLT3-ITD mutations is hampered by
sis, leading to a karyotype report showing fewer than competition from the wild-type allele such that estimates of
100% of cells with the translocation. Fluorescence in situ FLT3-ITD allelic burden can be inaccurate.9,10 However, the

Changing Treatment Landscape in AML
mutation burden of FLT3-ITD mutations is highly prognostic.11,12 “Know your enemy” is a quote often attributed to a famous
The higher the burden (however expressed, either as allelic ratio general from the distant past. In the modern era, molecular
or VAF), the worse the clinical outcome. testing allows us to reconstruct the development of this
deadly disease to truly “know the enemy” of each patient.
Molecular Testing Indicated for Acute Myeloid Leukemia With that knowledge, we are beginning to find ways to
Table 1 lists the tests that can be considered standard of explore its weaknesses.
care for evaluating newly diagnosed and relapsed AML. The TARGETED THERAPIES IN ACUTE MYELOID LEUKEMIA
molecular status of AML must be reassessed on disease Until recently, treatment of AML had changed little for
progression because new mutations may emerge that offer decades, comprised of an intensive induction-based ap-
potential for different targeted therapies. proach with cytarabine and an anthracycline (e.g., 7 + 3)
with or without alloHCT. This strategy is typically only tol-
Summary of Molecular Testing erated by younger (younger than age 60) and/or “fit” pa-
The complexity of molecular tests for AML can be bewil- tients, with complete remission (CR) rates of 60% to 80%
dering even for experienced clinicians, but the correct in- but 5-year survival rates of only about 40%.15 Table 2 de-
terpretation of their results is fast becoming the foundation fines current response definitions in AML. “Unfit” and/or
for optimal therapy. It is essential that the therapeutic and older patients (age 60 and older) often receive less intensive
prognostic impact of any given mutation be interpreted in approaches, including hypomethylating agents (HMAs),
the context of other mutations.3 A patient whose AML low-dose cytarabine (LDAC), and supportive care/hospice.
harbors a FLT3-ITD mutation has a better prognosis if that The most commonly used of the latter approaches in the
leukemia also has an NPM1 mutation.12,13 The presence of United States is HMA therapy; median survival rates are
a TP53 mutation by itself portends a poor prognosis, but the 7–12 months.15,16 There is no current standard definition of
situation is considerably worse when that mutation co- “fitness,” and exclusion of all patients older than age 60
occurs with a complex karyotype.14 from intensive therapy and alloHCT is probably not justified
TABLE 1. Assay for Molecular Typing in Acute Myeloid Leukemia

Assay Method Molecular Result Clinical Utility
G-banding Karyotype Prognostic
Fluorescence in situ hybridization Partial or complete loss of Prognostic
chromosomes 5, 7, 11, and 20
BCR-ABL1 Potentially can alter therapy choices (addition of BCR-ABL inhibitor)
Core-binding factor fusions: CBFB- Prognostic; potentially can alter therapy choices (addition of gemtuzumab
MYH11, RUNX1-RUNX1T1 ozogamicin)
Next-generation sequencing NPM1 Prognostic; can be used to monitor for minimal residual disease
TP53 Prognostic, particularly in the setting of a complex karyotype; potentially can
alter therapy choices (avoidance of intensive chemotherapy; use of
a hypomethylating agent)
IDH1 Potentially can alter therapy choices (use of ivosidenib in patients unfit for
intensive therapy); potential therapy if patient’s disease is refractory or
relapsed
IDH2 Potential therapy if patient’s disease is refractory or relapsed
CEBPa Prognostic
ASXL1 Prognostic
RUNX1 Prognostic
c-Kit Prognostic in patients with core-binding factor acute myeloid leukemia
RAS pathway Potential source of resistance to FLT3 and IDH inhibitors
FLT3-TKD Alters therapy (addition of FLT3 inhibitor)
Polymerase chain reaction FLT3-ITD, FLT3-TKD Prognostic; alters therapy (addition of FLT3 inhibitor)
Reverse transcription–polymerase Core-binding factor fusions: CBFB- Prognostic; potentially can alter therapy choices (addition of gemtuzumab and
chain reaction MYH11, RUNX1-RUNX1T1 ozogomicin)
Abbreviation: TKD, tyrosine kinase domain.

Koenig et al
(see below). Even among patients who are young and fit, cells with the toxic calicheamicin “payload” attached to the
there are subgroups with low probabilities of cure due to antibody. More than 80% of AMLs express CD33. GO was
high-risk disease features.17 Fortunately, advances in our initially approved in 2000 in older ( age 60) patients with r/r
understanding of the molecular basis of this disease have AML but was voluntarily withdrawn in 2010 due to safety
led to the development of new targeted agents, with ap- concerns; revised dosing has been associated with less
proval of five therapies for newly diagnosed and three for toxicity.22 It was reapproved September 2017 for both newly
relapsed/refractory (r/r) AML since 2017. These treatments diagnosed and r/r CD33-positive AML.22,23
and a general algorithm for their use are outlined in Fig. 1
Results from the ALFA-0701 study were key to this reap-
and are further described below.
proval. This was a phase III, multicenter, randomized, open-
Approved Therapies for Newly Diagnosed Acute label trial of 280 de novo, untreated patients with AML (age
Myeloid Leukemia 50–70) that compared 7 + 3 and consolidation alone versus
Midostaurin Midostaurin is a multikinase inhibitor approved in combination with GO (3 mg/m2 intravenously on days 1, 4,
in April 2017 for used in combination with standard 7 + 3 and 7 of induction and on day 1 of both consolidation
induction and high-dose cytarabine (HiDAC) consolidation courses).23 Complete remission and complete remission with
for adult patients with AML with fms-related tyrosine kinase incomplete platelet recovery (CRp) rates were 81% with GO
3 gene (FLT3) mutations (both FLT3-ITD and tyrosine ki- compared with 75%, respectively, in controls; this difference
nase domain).18,19 Up to 30% of newly diagnosed AMLs was not statistically significant. However, 2-year EFS and OS
harbor these mutations, with FLT3-ITD associated with poor were significantly higher with GO, with HRs of mortality and
prognosis as mentioned above.20 The RATIFY study (Cancer treatment failure of 0.58 (95% CI, 0.43–0.78; p = .0003) and
and Leukemia Group B 10603) was a randomized, double- 0.69 (0.49–0.98; p = .0368), respectively. Currently, GO is
blind, placebo-controlled trial of 717 patients, age 18–59; recommended only for patients with favorable risk cytoge-
midostaurin was given with 7 + 3 and HiDAC consolidation netics per the National Comprehensive Cancer Network
followed by continued midostaurin as single-agent main- guidelines because this was the only group demonstrating
tenance.18 Among patients receiving midostaurin, both the survival benefit in the earlier MRC AML 15 trial.24
median event-free survival (EFS) and overall survival (OS) The EORTC-GIMEMA AML-19 Trial investigated GO, with
times were 74.7 months (95% CI, 31.5 months to not a modified dose schedule, as a single agent for patients
reached) in the midostaurin group and 3.0 months (95% CI, unable to tolerate intensive chemotherapy.25 This trial
1.9–5.9) in the placebo group (one-sided p = .002 by compared GO with best supportive care in 237 patients
stratified log-rank test). The HR for death with versus older than age 61 with de novo or secondary AML. It was an
without midostaurin was 0.78 (95% CI, 0.63–0.96). CR
open-label, phase III trial performed in 35 sites in three
rates with midostaurin were slightly higher in the trial and
European countries. Patients receiving GO had a CR plus
higher compared with historical controls, but the differences
CR with incomplete count recovery (CRi) rate of 24.3%;
were not statistically significant.21 Single-agent midostaurin
their median overall survival was 4.9 months (4.2–6.8
maintenance therapy was not approved given the small
months) versus 3.6 months (2.6–4.2 months; p = .005) with
sample size available for analysis because most patients
best supportive care.
with CR received alloHCT.
The major toxicities associated with GO are hepatoxicity
Toxicities associated with midostaurin are QTc prolongation,
(black box warning), infusion reactions including anaphy-
liver enzyme elevation, rash/desquamation, and drug-
laxis, tumor lysis syndrome (TLS), and hemorrhage due to
induced pneumonitis.18,19
prolonged thrombocytopenia. It should be used with caution
Gemtuzumab ozogamicin Gemtuzumab ozogamicin (GO) is in patients for whom alloHCT is planned because the risk of
an anti-CD33 antibody-drug conjugate that targets leukemia veno-occlusive disease may be increased.22,26
TABLE 2. Response Definitions in Acute Myeloid Leukemia

Response Bone Marrow Blasts (%) ANC (/µL) Platelets (/µL)
Complete remission ,5 . 1,000 . 100,000
CR with hematologic improvement ,5 . 500 . 50,000
CR with incomplete count recovery ,5 , 1,000 . 100,000
CR with incomplete platelet recovery ,5 . 1,000 , 100,000
Morphologic leukemia free state ,5 — —
Partial remission 5 25 (. 50% reduction) . 1,000 . 100, 000

FIGURE 1. Decision Algorithm

for How to Treat Newly Diagnosed
and Relapsed Refractory AML
Based on Patient and Disease
Characteristics
Abbreviations: AML-MRC, acute
myeloid leukemia with myelodys-
plastic-related changes; t-AML
treatment-related acute myeloid
leukemia; HMA, hypomethylating
agent; LDAC, low-dose cytarabine;
7 + 3 cytarabine plus daunorubi-
cin or other intensive induction
chemotherapy regimen.
Venetoclax Venetoclax is an oral, potent, selective B-cell diagnosed AML in patients age 75 and older or in patients
lymphoma 2 (BCL-2) inhibitor.27 BCL-2 is an antiapoptotic who are unable to tolerate intensive chemotherapy.28
protein that sequesters the proapoptotic protein BAX; The clinical trial of venetoclax in combination with HMA
inhibiting BCL-2 leads to BAX release and subsequent was an international, open-label, phase Ib dose-escalation/
apoptosis. Venetoclax was granted accelerated U.S. Food expansion study that enrolled 145 patients age 65 and
and Drug Administration (FDA) approval in November 2018 older.27 Patients previously treated for myelodysplastic syn-
for use in combination with HMA therapy or LDAC for newly drome (MDS) or myeloproliferative neoplasm were excluded.

Koenig et al
Venetoclax was administered daily in a 28-day cycle with combination compared with 4.3 months with LDAC alone
either decitabine (20 mg/m2 intravenously on days 1–5) or (HR, 0.46; 80% CI, 0.35–0.62; p = .0002). Corresponding
azacitidine (75 mg/m2 intravenously or subcutaneously on ORRs (CR + CRi + MLFS) were 26.9% and 5.3%.
days 1–7) with a venetoclax initial ramp up followed by Toxicities of note for glasdegib are febrile neutropenia, QTc
continued daily dose of 400 mg. In the intention-to-treat prolongation, hyponatremia, and fatigue.31
analysis, the CR rate was 37%, and the CRi rate was 30%,
with an overall response rate (ORR; CR + CRi + partial Ivosidenib Ivosidenib is an oral IDH1 inhibitor. Mutated
remission [PR] + morphologic leukemia free state [MLFS]) IDH1, seen in 6% to 10% of patients with AML, leads to
of 83%. Median OS was 17.5 months (95% CI, 12.3–not catalyzing isocitrate to alpha-ketoglutarate and produces
reached); 21 patients received alloHCT. This combina- D-2-hydroxyglutarate, an oncometabolite that causes epi-
tion demonstrated rapid disease control, with time to first genetic alternations preventing hematopoietic differentia-
response and best response (CR) of 1.2 and 2.1 months, tion.33 Ivosidenib was approved in May 2019 for patients
respectively. A randomized phase III study of 400 mg with newly diagnosed AML age 75 or older or who were
venetoclax versus placebo in combination with azaciti- unable to tolerate intensive chemotherapy and harbor
dine will determine full FDA approval for this regimen a susceptible IDH1 mutation (R132*).33,34
(NCT02993523). Approval in the newly diagnosed setting came from a subset
An open-label, multicenter, international phase Ib/II study analysis of 34 patients with newly diagnosed AML who were
using venetoclax with LDAC in 82 patients with AML (age unable to receive standard chemotherapy taken from
60) led to this combination’s accelerated approval.29 This a phase I, multicenter, open-label, dose-escalation/expan-
trial included patients previously treated with HMAs for sion study evaluating ivosidenib usage in patients age 18 or
MDS. Venetoclax was administered at 600 mg daily in older with r/r and newly diagnosed AML (NCT02074839);
a 28-day cycle with LDAC (20 mg/m2 subcutaneously, days prior HMA therapy was allowed.33 Ivosidenib was admin-
1–10), with a CR rate of 26%, CRi of 28%, and median OS of istered at 500 mg daily. The CR rate was 30.3%, the ORR
10.1 months (95% CI, 5.7–14.2). The median time to first (CR + CRi + CR with hematologic improvement [CRh] + PR +
CR/CRi was 1.4 months. Patients without prior HMA therapy MLFS) was 54.5%, and median OS was 12.6 months.
performed better, with a CR/CRi rate of 62% and median OS Median times to response and to CR were 1.9 and 2.8
of 13.5 months, compared with a CR/CRi rate of 33% and months, respectively. Patients with previous HMA therapy
median OS of 4.1 months with prior exposure. A phase III fared worse, with CR rates of 20.0% and CR + CRh rates
placebo-controlled trial is ongoing to determine full FDA of 26.7% compared with 38.9% and 55.6% in HMA-naive
approval in this setting (NCT03069352). patients.
Major toxicities of ivosidenib are QTc prolongation, cyto-
Major toxicities of venetoclax are gastrointestinal toxicities,
penias, and differentiation syndrome.35
myelosuppression, and TLS (risk mitigated by administering
once white blood cell count is 25 109 L).30 Venetoclax is Liposomal daunorubicin-cytarabine Although not targeting
also a major substrate of CYP3A4, and dose adjustments are a specific molecule, liposomal daunorubicin-cytarabine is
required for concomitant strong or moderate CYP3A in- a “tailored” AML therapy in that it is approved for only a
hibitors or P-gp inhibitors, which include antifungals such specific subset of AML. The drug is a liposomal encapsu-
as posaconazole, voriconazole, and isavuconazole. lation of cytarabine and daunorubicin in a 5:1 ratio. It was
approved in August 2017 for adults with newly diagnosed
Glasdegib Glasdegib is a potent, oral, selective inhibitor of therapy-related AML and AML with myelodysplasia-related
Smoothened (part of the Hedgehog signaling pathway) changes (AML-MRC).36,37 The formulation preferentially
with inhibition decreasing leukemia stem cell survival and delivers drug into leukemia cells versus normal cells in the
leading to increased sensitivity to chemotherapy.31 It was bone marrow.37 The pivotal phase III, open-label, ran-
granted FDA approval in November 2018 for use with LDAC domized trial leading to approval enrolled 309 patients,
in newly diagnosed AML in patients age 75 or older or in age 60–75 (although the therapy is approved for adults of
patients unable to tolerate intensive chemotherapy.32 all ages), with de novo AML-MRC, therapy-related AML,
The BRIGHT AML 1003 phase II, open-label, randomized, or secondary AML (from prior MDS or chronic myelomo-
multicenter, international trial enrolled 132 patients age 55 nocytic leukemia) and compared liposomal daunorubicin-
and older with AML or high-risk MDS; 116 patients had cytarabine with standard 7 + 3 therapy.37 Median OS was
AML. Eighty-eight and 44 patients received glasdegib + significantly longer, at 9.56 months in the liposomal
LDAC or LDAC alone, respectively.31 Prior therapy with daunorubicin-cytarabine arm compared with 5.95 months
HMA was allowed. Patients received glasdegib 100 mg daily in the 7 + 3 arm (HR, 0.69; 95% CI, 0.52–0.90). Overall
with LDAC 20 mg/m2 subcutaneously twice daily days 1–10 remission rates were 47.7% with liposomal daunorubicin-
in 28-day cycles. Median OS was 8.3 months with the cytarabine versus 33.3% with 7 + 3 (p = .016). More

patients on the liposomal daunorubicin-cytarabine went on 2018. This was based on the ADMIRAL trial, a phase III,
to alloHCT, and their postHCT survival was better than those randomized, multicenter, international study of 625 patients
whose alloHCT occurred after 7 + 3 (HR of mortality, 0.46; with r/r AML.41 The trial compared gilteritinib 120 mg daily to
95% CI, 0.24–0.89). investigator’s choice of salvage therapy (HiDAC-based
Toxicities of liposomal daunorubicin-cytarabine are similar regimens, LDAC, or HMA). Only 5.7% of patients had prior
to standard 7 + 3 except for having more prolonged delayed midostaurin exposure. CR and CR/CRh rates were 21.1%
neutrophil and platelet count recovery compared with 7 + and 34.0%, respectively, with gilteritinib compared with
3.36,37 10.5% and 15.3% with salvage therapy. Median OS was
9.3 months with gilteritinib and 5.6 months with controls,
Approved Therapies for Relapsed/Refractory Acute with an HR of mortality of 0.64 (95% CI, 0.49–0.83; p , .001).
Myeloid Leukemia Median EFS was 2.8 and 0.7 months with gilteritinib versus
Ivosidenib Ivosidenib was initially approved for r/r AML in salvage, respectively. More patients with gilteritinib sub-
July 2018 based on the phase I trial described above.35 sequently received alloHCT (25.5% vs. 15.3%); survival
There were 125 patients included in the primary efficacy benefit was observed even with censoring for alloHCT.
population (r/r IDH1 mutated AML with at least 6 months Major toxicities of gilteritinib are febrile neutropenia, ele-
follow-up) who received 500 mg ivosidenib daily. The CR vated transaminases, QTc prolongation, and, rarely (but of
rate was 21.6%, CRh was 30.4%, ORR (CR + CRi + CRp + note), reversible posterior encephalopathy, and differenti-
PR + MLFS) was 41.6%, and median OS was 8.8 months. ation syndrome.41
The number of prior treatment regimens affected the CR +
CRi rate, which was 46.0% with one prior regimen but only Quizartinib and crenolanib are other FLT3 inhibitors under
15.0% with at least three prior regimens. investigation for r/r FLT3-mutated AML; quizartinib is ap-
proved in Japan but not in the United States due to efficacy
Enasidenib Enasidenib is an oral IDH2 inhibitor. IDH2 concerns in the phase III trial.42-44
mutations occur at an incidence of 12% and result in the
generation of the oncometabolite 2-HG. However, the IDH2 Summary of Targeted Therapies
enzyme is localized to the mitochondria (unlike IDH1, The introduction of targeted therapies has revolutionized our
which is cytosolic), and the consequences of this difference approach to many patients with AML previously facing poor
to leukemogenesis is not yet clear.33 Enasidenib was ap- prognoses with few treatment options. Although there is still
proved for r/r IDH2 (R140* and R172*) mutated AML in room for improvement and uncertainty about which, if any,
August 2017; it is not currently approved in the upfront of these treatments is curative, results are promising and
setting.38,39 The study leading to approval was a phase I/II provide potential for success in the treatment of this deadly
dose-escalation/expansion trial with efficacy assessed in disease. Mature, long-term data are needed to determine if
176 patients age 18 and older with r/r AML.39 The CR rate targeted therapy is curative for select populations, as long-
was 19.3%, the ORR (CR + CRi + CRp + PR + MLFS) was term responders have been observed. Questions remain
40.3%, and median OS was 9.3 months. Of these patients, that need further exploration, including if alloHCT is nec-
109 received the recommended dose of 100 mg daily, with essary for all patients in first remission and comparison of
similar response rates. More than half (53%) of patients had post-transplant outcomes in patients treated with targeted
received at least two prior regimens; their median OS was therapeutics alone versus those with cytotoxic chemother-
8.0 months. apy exposure.
Major toxicities of enasidenib are hyperbilirubinemia, nausea, CURRENT ROLE OF ALLOHCT IN ACUTE MYELOID LEUKEMIA
and differentiation syndrome.39
AlloHCT is an established curative therapy for AML, and
GO The FDA approved GO in the r/r setting based on the AML is the most common indication for alloHCT. About
MyloFrance-1 study, a phase II, open-label, single-arm, 3,500 patients undergo HCT for AML in the United States
multicenter study of 57 patients with CD33-positive AML in annually, accounting for 35% to 40% of all alloHCT.45
first relapse who received GO as monotherapy induction Overall survival rates have increased since transplants
(3 mg/m2 intravenously on days 1, 4, and 7).40 The CR rate were first done in the 1970s; 3-year survival rates now
was 26%, the ORR (CR + CRp) was 33.3% at day 43 after exceed 50% overall and are more than 60% in patients
GO, and median OS was 8.4 months. transplanted in first CR (Fig. 2).43 The major cause of
Gilteritinib Gilteritinib is a tyrosine kinase inhibitor and treatment failure is relapse, but graft-versus-host disease
a potent, selective FLT3 inhibitor. In contrast to midostaurin, (GVHD), infection, and organ toxicity also contribute to
which has not shown clinical activity as a single agent, mortality.
gilteritinib was approved as a single agent for FLT3-mutated Although use of HCT has increased over the past 10 years,
(ITD and tyrosine kinase domain) r/r AML in November only about 25% of patients with AML under age 75 receive

Koenig et al
FIGURE 2. Probability of Survival

After HLA-Identical Sibling (A) or
Unrelated Donor (B) Transplantation
for Acute Myeloid Leukemia in
2007–2017 as Reported to the
Center for International Blood and
Marrow Transplant Research
a transplant. Reasons cited include lack of suitable donors status, which predict TRM. Risks of relapse are highest in
and the high risk of transplant-related mortality (TRM) transplant recipients with active disease and in those with
causing many physicians to defer referral for HCT, espe- high-risk cytogenetic and molecular abnormalities. How-
cially in older patients. The median age at HCT is 55, in ever, even among patients with high-risk features, there is
contrast to a median age at AML diagnosis of 68 (Center for a meaningful chance of cure. The association of specific
International Blood and Marrow Transplant Registry, un- disease risk features with survival has been translated into
published data, February 2020). A recent survey by the a Disease Risk Index tool, developed by Armand et al49 and
National Marrow Donor Program/Be The Match revealed validated by the Center for International Blood and Marrow
that more than half of community oncologists think that HCT Transplant Registry. The Disease Risk Index classifies pa-
is not appropriate for patients over age 50 (Steve Devine, tients as low (AML with favorable cytogenetics in CR), in-
personal communication, February 2020). However, the termediate (AML with intermediate risk cytogenetics in CR),
strong antileukemia efficacy of alloHCT and recent ad- high (AML with favorable or intermediate cytogenetics not in
vances in transplant strategies to decrease TRM make this CR or with unfavorable cytogenetics in CR), or very high
curative therapy an appropriate option in many older (AML with adverse cytogenetics not in CR) risk, with esti-
patients. mated post-transplant 2-year survival rates of 66%, 51%,
Selection of Patients 33%, and 23%, respectively.49 Given that cytogenetics now
represent only a fraction of the sum of prognostic data on
The decision to perform HCT represents a balance between
a patient’s disease, it will be important to update this
the estimated likelihood of cure and survival with non-HCT
classification system to incorporate the information provided
therapy versus the estimated likelihood with HCT. Although
by more detailed molecular tests such as NGS. Patients with
treatment-related deaths occur with non-HCT therapy, the
primary induction failure deserve special mention because
primary driver of survival is disease control, whereas survival
they are generally incurable with conventional therapy, but
after HCT is driven by both disease control and TRM. The
long-term survival can be achieved in about 20% of these
balance then is generally between the likelihood of cure with
patients with alloHCT, especially those with low numbers of
non-HCT therapy and the likelihood of both cure and TRM
circulating blasts and good performance scores.50
with HCT. Some of the considerations in estimating the
likelihood of cure with non-HCT are discussed above, but TRM is higher in patients who are older; however, HCT can
patients for whom HCT is likely to be a preferred therapy are be successful in patients in their 70s with use of reduced in-
those with high white blood cell counts at diagnosis, un- tensity conditioning regimens.51,52 Consideration of comor-
favorable cytogenetics, and high-risk molecular abnormal- bidities aids risk assessment and is more reliable than age
ities; those who fail initial induction therapy; and those with alone. The Sorror HCT Comorbidity Index is a validated tool for
hematologic remission but minimal residual disease after evaluating risk.53 Additional geriatric assessment can add
induction and consolidation.3,20,46,47 Of special note is that value in patients over the age of 60.54 It is important to note that
age itself is an adverse risk factor, with only rare patients older patients in their 50s, 60s, and 70s with good perfor-
over the age of 50 being cured with conventional therapy.48 mance status and few comorbidities have outcomes similar to
Regardless of cytogenetic and molecular features, relapse patients in their 40s, and, given the dismal outcomes of
after an initial CR is an indication for HCT. conventional therapy in these patients, they should be referred
early for consideration of HCT.55,56
HCT outcomes are strongly influenced by remission state at
time of transplant and other disease-related factors, which The ability to identify a donor and perform HCT at an optimal
predict post-transplant relapse, and by age and clinical time depends on having adequate lead time to make the

necessary preparations. Ideally, patients should be referred and radiation without concern for marrow toxicity and an
for evaluation early in their disease course so that their op- immune-mediated graft-versus-leukemia effect of donor
tions can be evaluated and a plan can be made, including cells.61,62 The toxicity of high-dose preparative regimens,
optimal timing. precluding their use in older and sicker patients, prompted
Selection of Donors studies evaluating whether regimens sufficiently intensive
to allow donor cell engraftment but not high enough to
The gold standard for alloHCT is, traditionally, an HLA- cause myeloablation could deliver a graft-versus-leukemia
identical sibling. Such donors account for approximately effect with meaningful clinical benefit. This is now well
25% of transplants done for AML.45 Alternatives include accepted, and reduced intensity regimens are used in
HLA-matched and mismatched unrelated volunteer donors, about 45% of HCTs done for AML and in about 60% of
HLA-mismatched (haploidentical) relatives, and banked those done in patients over the age of 50.45 The debate as
unrelated umbilical cord blood. Survival after HCT from to whether myeloablative therapy, if possible, gives a better
unrelated donors matched at high resolution for HLA-A, outcome because of reduced relapse continues, with di-
HLA-B, HLA-C, and DRB1 is equivalent to survival after verse results being reported by multiple groups. However,
HLA-identical sibling HCT; matched unrelated donors ac- the largest study, from the U.S. Blood and Marrow
count for 45% of HCTs for AML.57,58 Data regarding the Transplant Clinical Trials Network, suggests that, among
relative outcomes of HCT from other donor sources are still patients who can safely receive high-intensity regimens,
emerging, with recent improvements in HLA-mismatched leukemia-free survival is higher with higher intensity.63 An
adult donor and cord blood transplants reported with novel important ancillary study to that trial was recently pub-
GVHD prevention regimens and better supportive care and, lished by Hourigan et al64 and suggests that this benefit is
in the case of cord blood transplants, better selection of largely due to lower relapse in patients with minimal re-
units. It is not possible with currently available data to define sidual disease prior to conditioning. For now, however,
the “best” graft source for an individual patient; however, whether to use high-intensity versus reducing-intensity
the differences in outcomes with an HLA-identical sibling, conditioning depends on the estimated risk of TRM with
HLA-matched, or single-locus mismatched unrelated do- myeloablative regimens, which are the approach of choice
nor, haploidentical donor, or single or double umbilical cord in patients who can tolerate them.
blood are likely to be less than 10%, making the HCT versus
no-HCT decision similar whether or not the patient has an Newer agents, such as treosulfan, recently approved in
HLA-identical sibling. The specific choice of donor is best Europe, may allow myeloablation with lower toxicity and
made by an experienced transplant physician considering new approaches, such as post-transplant maintenance
not only the degree of HLA match but also the urgency of therapy with targeted drugs for particular subsets of AML or
HCT as well as the speed of donor availability, age of the AML-directed cell therapies, that may improve the relapse
donor, adequacy of the anticipated cell dose (influenced by rate after reduced-intensity regimens. Recent de-
recipient weight and especially important for cord blood velopments that address nonrelapse mortality and mor-
HCT), donor-recipient cytomegalovirus and ABO status, and bidity include new approaches to preventing and treating
other factors.59 Referral for HCT evaluation should be made GVHD, including the use of post-transplant cyclophos-
on the basis of patient factors because few patients will lack phamide for both matched and mismatched transplants
a suitable donor in the current era. Other graft decisions and selective T-cell depletion and the use of antiviral drugs
include whether to use bone marrow versus peripheral and virus-specific cytotoxic T-cell therapies for post-
blood grafts from adult donors. Current data suggest that, in transplant infections.
the related donor setting, peripheral blood grafts may offer
an advantage, particularly in patients with advanced disease, SUMMARY
but in the unrelated donor setting survival is not increased
while risk of significant chronic GVHD is dramatically in- Recent advances in defining the molecular characteristics
creased with peripheral blood grafts, with impaired long-term of AML have allowed refinement of prognosis estimates and
quality of life.60 development of targeted therapies that can be applied as
part of initial therapy or to treat relapse. AlloHCT remains the
Minimizing TRM and Maximizing Cure curative option of choice for patients at high risk of disease
The efficacy of alloHCT in AML derives from two factors: the recurrence and acceptable risks of TRM. The role of tar-
ability to deliver high, myeloablative doses of chemotherapy geted therapies in HCT regimens remains to be defined.

Koenig et al

1
Division of Hematology, Department of Medicine, The Ohio State Mary M. Horowitz, MD, MS, Froedtert and the Medical College of
University and The Ohio State University Comprehensive Cancer Center, Wisconsin Clinical Cancer Center, CIBMTR | Milwaukee Campus, 9200 W.
Columbus, OH Wisconsin Ave., Suite C5500, Milwaukee, WI 53226; email: marymh@
2
Division of Hematology, Department of Medicine, The Ohio State mcw.edu.
University and The Ohio State University Comprehensive Cancer Center,
Columbus, OH
3
Hematologic Malignancies and Bone Marrow Transplant Program, Sidney AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Kimmel Comprehensive Cancer Center, Baltimore, MD AND DATA AVAILABILITY STATEMENT
4
Center for International Blood and Marrow Transplant Research, Disclosures provided by the authors and data availability statement (if
Department of Medicine, Medical College of Wisconsin, Milwaukee, WI applicable) are available with this article at DOI https://doi.org/10.1200/
EDBK_279129.
REFERENCES
1. Kandoth C, McLellan MD, Vandin F, et al. Mutational landscape and significance across 12 major cancer types. Nature. 2013;502:333-339.
2. Welch JS, Ley TJ, Link DC, et al. The origin and evolution of mutations in acute myeloid leukemia. Cell. 2012;150:264-278.
3. Papaemmanuil E, Gerstung M, Bullinger L, et al. Genomic classification and prognosis in acute myeloid leukemia. N Engl J Med. 2016;374:2209-2221.
4. Corces-Zimmerman MR, Hong WJ, Weissman IL, et al. Preleukemic mutations in human acute myeloid leukemia affect epigenetic regulators and persist in
remission. Proc Natl Acad Sci USA. 2014;111:2548-2553.
5. Shlush LI, Zandi S, Mitchell A, et al; HALT Pan-Leukemia Gene Panel Consortium. Identification of pre-leukaemic haematopoietic stem cells in acute leukaemia
[published correction appears in Nature. 508(7496):420]. Nature. 2014;506:328-333.
6. Steensma DP, Bejar R, Jaiswal S, et al. Clonal hematopoiesis of indeterminate potential and its distinction from myelodysplastic syndromes. Blood. 2015;
126:9-16.
7. Jongen-Lavrencic M, Grob T, Hanekamp D, et al. Molecular minimal residual disease in acute myeloid leukemia. N Engl J Med. 2018;378:1189-1199.
8. Levis M. FLT3 mutations in acute myeloid leukemia: what is the best approach in 2013? Hematology (Am Soc Hematol Educ Program). 2013;2013:220-226.
9. Murphy KM, Levis M, Hafez MJ, et al. Detection of FLT3 internal tandem duplication and D835 mutations by a multiplex polymerase chain reaction and capillary
electrophoresis assay. J Mol Diagn. 2003;5:96-102.
10. Polz MF, Cavanaugh CM. Bias in template-to-product ratios in multitemplate PCR. Appl Environ Microbiol. 1998;64:3724-3730.
11. Thiede C, Steudel C, Mohr B, et al. Analysis of FLT3-activating mutations in 979 patients with acute myelogenous leukemia: association with FAB subtypes and
identification of subgroups with poor prognosis. Blood. 2002;99:4326-4335.
12. Gale RE, Green C, Allen C, et al; Medical Research Council Adult Leukaemia Working Party. The impact of FLT3 internal tandem duplication mutant level,
number, size, and interaction with NPM1 mutations in a large cohort of young adult patients with acute myeloid leukemia. Blood. 2008;111:2776-2784.
13. Schlenk RF, Döhner K, Krauter J, et al; German-Austrian Acute Myeloid Leukemia Study Group. Mutations and treatment outcome in cytogenetically normal
acute myeloid leukemia. N Engl J Med. 2008;358:1909-1918.
14. Rücker FG, Schlenk RF, Bullinger L, et al. TP53 alterations in acute myeloid leukemia with complex karyotype correlate with specific copy number alterations,
monosomal karyotype, and dismal outcome. Blood. 2012;119:2114-2121.
15. Dombret H, Gardin C. An update of current treatments for adult acute myeloid leukemia. Blood. 2016;127:53-61.
16. Kantarjian HM, Thomas XG, Dmoszynska A, et al. Multicenter, randomized, open-label, phase III trial of decitabine versus patient choice, with physician advice,
of either supportive care or low-dose cytarabine for the treatment of older patients with newly diagnosed acute myeloid leukemia. J Clin Oncol. 2012;
30:2670-2677.
17. Acute Myeloid Leukemia - Cancer Stat Facts. https://seer-cancer-gov.proxy.lib.ohio-state.edu/statfacts/html/amyl.html. Accessed January 31, 2020.
18. Stone RM, Mandrekar SJ, Sanford BL, et al. Midostaurin plus chemotherapy for acute myeloid leukemia with a FLT3 mutation. N Engl J Med. 2017;377:454-464.
19. U.S. Food and Drug Administration. Midostaurin. https://www.fda.gov/drugs/resources-information-approved-drugs/midostaurin. Accessed January 20, 2020.
20. Döhner H, Estey E, Grimwade D, et al. Diagnosis and management of AML in adults: 2017 ELN recommendations from an international expert panel. Blood.
2017;129:424-447.
21. Kottaridis PD, Gale RE, Frew ME, et al. The presence of a FLT3 internal tandem duplication in patients with acute myeloid leukemia (AML) adds important
prognostic information to cytogenetic risk group and response to the first cycle of chemotherapy: analysis of 854 patients from the United Kingdom Medical
Research Council AML 10 and 12 trials. Blood. 2001;98:1752-1759.
22. U.S. Food and Drug Administration. FDA approves gemtuzumab ozogamicin for CD33-positive AML. https://www.fda.gov/drugs/resources-information-
approved-drugs/fda-approves-gemtuzumab-ozogamicin-cd33-positive-aml. Accessed January 20, 2020.
23. Castaigne S, Pautas C, Terré C, et al; Acute Leukemia French Association. Effect of gemtuzumab ozogamicin on survival of adult patients with de-novo acute
myeloid leukaemia (ALFA-0701): a randomised, open-label, phase 3 study. Lancet. 2012;379:1508-1516.

24. Burnett AK, Hills RK, Milligan D, et al. Identification of patients with acute myeloblastic leukemia who benefit from the addition of gemtuzumab ozogamicin:
results of the MRC AML15 trial. J Clin Oncol. 2011;29:369-377.
25. Amadori S, Suciu S, Selleslag D, et al. Gemtuzumab ozogamicin versus best supportive care in older patients with newly diagnosed acute myeloid leukemia
unsuitable for intensive chemotherapy: results of the randomized phase III EORTC-GIMEMA AML-19 trial. J Clin Oncol. 2016;34:972-979.
26. Lambert J, Pautas C, Terré C, et al. Gemtuzumab ozogamicin for de novo acute myeloid leukemia: final efficacy and safety updates from the open-label, phase III
ALFA-0701 trial. Haematologica. 2019;104:113-119.
27. DiNardo CD, Pratz K, Pullarkat V, et al. Venetoclax combined with decitabine or azacitidine in treatment-naive, elderly patients with acute myeloid leukemia.
Blood. 2019;133:7-17.
28. U.S. Food and Drug Administration. FDA approves venetoclax in combination for AML in adults. https://www.fda.gov/drugs/fda-approves-venetoclax-
combination-aml-adults. Accessed January 20, 2020.
29. Wei AH, Strickland SA Jr., Hou JZ, et al. Venetoclax combined with low-dose cytarabine for previously untreated patients with acute myeloid leukemia: results
from a phase Ib/II study. J Clin Oncol. 2019;37:1277-1284.
30. Konopleva M, Pollyea DA, Potluri J, et al. Efficacy and biological correlates of response in a phase II study of venetoclax monotherapy in patients with acute
myelogenous leukemia. Cancer Discov. 2016;6:1106-1117.
31. Cortes JE, Heidel FH, Hellmann A, et al. Randomized comparison of low dose cytarabine with or without glasdegib in patients with newly diagnosed acute myeloid
leukemia or high-risk myelodysplastic syndrome. Leukemia. 2019;33:379-389.
32. Norsworthy KJ, By K, Subramaniam S, et al. FDA approval summary: glasdegib for newly diagnosed acute myeloid leukemia. Clin Cancer Res. 2019;
25:6021-6025.
33. Roboz GJ, DiNardo CD, Stein EM, et al. Ivosidenib induces deep durable remissions in patients with newly diagnosed IDH1-mutant acute myeloid leukemia.
Blood. 2019;135:463-471.
34. U.S. Food and Drug Administration. FDA approves ivosidenib as first-line treatment for AML with IDH1 mutation. https://www.fda.gov/drugs/resources-
information-approved-drugs/fda-approves-ivosidenib-first-line-treatment-aml-idh1-mutation. Accessed January 20, 2020.
35. DiNardo CD, Stein EM, de Botton S, et al. Durable remissions with ivosidenib in IDH1-mutated relapsed or refractory AML. N Engl J Med. 2018;378:2386-2398.
36. Krauss AC, Gao X, Li L, et al. FDA approval summary: (Daunorubicin and Cytarabine) liposome for injection for the treatment of adults with high-risk acute myeloid
leukemia. Clin Cancer Res. 2019;25:2685-2690.
37. Lancet JE, Uy GL, Cortes JE, et al. CPX-351 (cytarabine and daunorubicin) liposome for injection versus conventional cytarabine plus daunorubicin in older
patients with newly diagnosed secondary acute myeloid leukemia. J Clin Oncol. 2018;36:2684-2692.
38. U.S. Food and Drug Administration. FDA granted regular approval to enasidenib for the treatment of relapsed or refractory AML | FDA. https://www.fda.gov/drugs/
resources-information-approved-drugs/fda-granted-regular-approval-enasidenib-treatment-relapsed-or-refractory-aml. Accessed January 20, 2020.
39. Stein EM, DiNardo CD, Pollyea DA, et al. Enasidenib in mutant IDH2 relapsed or refractory acute myeloid leukemia. Blood. 2017;130:722-731.
40. Taksin AL, Legrand O, Raffoux E, et al. High efficacy and safety profile of fractionated doses of Mylotarg as induction therapy in patients with relapsed acute
myeloblastic leukemia: a prospective study of the alfa group. Leukemia. 2007;21:66-71.
41. Perl AE, Martinelli G, Cortes JE, et al. Gilteritinib or chemotherapy for relapsed or refractory FLT3-mutated AML. N Engl J Med. 2019;381:1728-1740.
42. U.S. Food and Drug Administration. FDA briefing document, Oncologic Drugs Advisory Committee (ODAC) Meeting: NDA 212166 Quizartinib.
43. Daiichi-Sankyo. Daiichi Sankyo’s VANFLYTA® Receives Approval in Japan for the Treatment of Relapsed/Refractory FLT3-ITD AML. https://www.daiichisankyo.
com/media_investors/media_relations/press_releases/detail/007030.html.
44. Galanis A, Ma H, Rajkhowa T, et al. Crenolanib is a potent inhibitor of FLT3 with activity against resistance-conferring point mutants. Blood. 2014;123:94-100.
45. D’Souza A, Fretham C. Current Uses and Outcomes of Hematopoietic Cell Transplantation (HCT): CIBMTR Summary Slides, 2018. http://www.cibmtr.org/
ReferenceCenter/SlidesReports/SummarySlides/pages/index.aspx. Accessed January 20, 2020.
46. Terwijn M, van Putten WL, Kelder A, et al. High prognostic impact of flow cytometric minimal residual disease detection in acute myeloid leukemia: data from the
HOVON/SAKK AML 42A study. J Clin Oncol. 2013;31:3889-3897.
47. Freeman SD, Hills RK, Virgo P, et al. Measurable residual disease at induction redefines partial response in acute myeloid leukemia and stratifies outcomes in
patients at standard risk without NPM1 mutations. J Clin Oncol. 2018;36:1486-1497.
48. Webster JA, Pratz KW. Acute myeloid leukemia in the elderly: therapeutic options and choice. Leuk Lymphoma. 2018;59:274-287.
49. Armand P, Kim HT, Logan BR, et al. Validation and refinement of the disease risk index for allogeneic stem cell transplantation. Blood. 2014;123:3664-3671.
50. Duval M, Klein JP, He W, et al. Hematopoietic stem-cell transplantation for acute leukemia in relapse or primary induction failure. J Clin Oncol. 2010;
28:3730-3738.
51. Muffly L, Pasquini MC, Martens M, et al. Increasing use of allogeneic hematopoietic cell transplantation in patients aged 70 years and older in the United States.
Blood. 2017;130:1156-1164.
52. Ringdén O, Boumendil A, Labopin M, et al. Outcome of allogeneic hematopoietic stem cell transplantation in patients age .69 years with acute myelogenous
leukemia: on behalf of the Acute Leukemia Working Party of the European Society for Blood and Marrow Transplantation. Biol Blood Marrow Transplant. 2019;
25:1975-1983.

Koenig et al
53. Sorror ML, Storb RF, Sandmaier BM, et al. Comorbidity-age index: a clinical measure of biologic age before allogeneic hematopoietic cell transplantation. J Clin
Oncol. 2014;32:3249-3256.
54. Muffly LS, Boulukos M, Swanson K, et al. Pilot study of comprehensive geriatric assessment (CGA) in allogeneic transplant: CGA captures a high prevalence of
vulnerabilities in older transplant recipients. Biol Blood Marrow Transplant. 2013;19:429-434.
55. McClune BL, Weisdorf DJ, Pedersen TL, et al. Effect of age on outcome of reduced-intensity hematopoietic cell transplantation for older patients with acute
myeloid leukemia in first complete remission or with myelodysplastic syndrome. J Clin Oncol. 2010;28:1878-1887.
56. Nikolousis E, Nagra S, Pearce R, et al. Impact of pre-transplant co-morbidities on outcome after alemtuzumab-based reduced intensity conditioning allo-SCT in
elderly patients: a British Society of Blood and Marrow Transplantation study. Bone Marrow Transplant. 2015;50:82-86.
57. Saber W, Opie S, Rizzo JD, et al. Outcomes after matched unrelated donor versus identical sibling hematopoietic cell transplantation in adults with acute
myelogenous leukemia. Blood. 2012;119:3908-3916.
58. Schlenk RF, Döhner K, Mack S, et al. Prospective evaluation of allogeneic hematopoietic stem-cell transplantation from matched related and matched unrelated
donors in younger adults with high-risk acute myeloid leukemia: German-Austrian trial AMLHD98A. J Clin Oncol. 2010;28:4642-4648.
59. Dehn J, Spellman S, Hurley CK, et al. Selection of unrelated donors and cord blood units for hematopoietic cell transplantation: guidelines from the NMDP/
CIBMTR. Blood. 2019;134:924-934.
60. Lee SJ, Logan B, Westervelt P, et al. Comparison of patient-reported outcomes in 5-year survivors who received bone marrow vs peripheral blood unrelated donor
transplantation: long-term follow-up of a randomized clinical trial. JAMA Oncol. 2016;2:1583-1589.
61. Horowitz MM, Gale RP, Sondel PM, et al. Graft-versus-leukemia reactions after bone marrow transplantation. Blood. 1990;75:555-562.
62. Weisdorf D, Zhang MJ, Arora M, et al. Graft-versus-host disease induced graft-versus-leukemia effect: greater impact on relapse and disease-free survival after
reduced intensity conditioning. Biol Blood Marrow Transplant. 2012;18:1727-1733.
63. Scott BL, Pasquini MC, Logan BR, et al. Myeloablative versus reduced-intensity hematopoietic cell transplantation for acute myeloid leukemia and myelo-
dysplastic syndromes. J Clin Oncol. 2017;35:1154-1161.
64. Hourigan CS, Dillon LW, Gui G, et al. Impact of conditioning intensity of allogeneic transplantation for acute myeloid leukemia with genomic evidence of residual
disease. J Clin Oncol. Epub 2019 Dec 20.

The Role of Early Intervention in High-Risk

Smoldering Myeloma
Nisha S. Joseph, MD1; Madhav V. Dhodapkar, MBBS1; and Sagar Lonial, MD, FACP1
overview
Smoldering multiple myeloma (SMM) is a precursor disease state that precedes the development of
symptomatic myeloma. As we have learned more about the disease biology of SMM and risk factors for
progression, updated risk stratification models, such as the Mayo 2018 model, or 20/2/20, have been
developed. More accurate risk stratification and the development of effective and well-tolerated therapeutic
agents have led to the investigation of early treatment of select patients with high-risk SMM with the aim of
delaying time to progression to multiple myeloma. Ongoing debate surrounds which subset of patients with
SMM to target, as well as the best treatment approach: preventative versus curative. Phase III data from the
Spanish Myeloma Group/PETHEMA as well as the Eastern Cooperative Oncology Group (ECOG) E3A06 trial
have shown the efficacy of lenalidomide with and without dexamethasone in high-risk SMM in delaying
progression to symptomatic disease. Conversely, there exists an alternate strategy attempting to cure the
disease prior to progression utilizing more intensive regimens similar to what is used for patients with newly
diagnosed myeloma. However, our understanding of the disease biology of SMM and the role of immune
regulation in preventing malignant transformation provides a strong rationale for an interventional strategy.
Here, we review the definition of SMM, the current models for risk stratification, and the current data available
supporting the early treatment of patients with high-risk SMM.
SMM is an intermediate clinical stage between mono- from early intervention, but how do we identify those
clonal gammopathy of undetermined significance that would? There are currently no molecular markers
(MGUS) and symptomatic myeloma, defined as that enable us to distinguish between these patients,
the presence of a serum monoclonal protein 3 g/dL and in the past, this delineation has largely relied on
or urinary monoclonal protein 500 mg per 24 hours, the extent of tumor burden. The challenge has be-
and/or 10%–60% bone marrow plasma cells in the come how to appropriately identify patients at this
absence of myeloma defining events or amyloidosis. increased risk for progression, with the intent to
Until recently, the standard of care for all patients with intervene prior to the development of end-organ
SMM has been observation; however, recent data have damage. In 2014, the International Myeloma Work-
challenged this norm and proposed early intervention ing Group (IMWG) updated the diagnostic criteria to
as an effective strategy for select subsets of patients include ultra–high-risk disease with an 80% risk of
with SMM. As more data have emerged on treatment of progression at 2 years that previously would have been
high-risk SMM, this has fueled debate about which classified as SMM. These criteria include bone marrow
patients to target and how best to do so. We propose plasmacytosis 60%, a serum-free light chain ratio
that patients with high-risk SMM derive the most 100, and . 1 focal lesion on whole-body MRI or PET-
benefit from early intervention, and that a preventative CT.3-7 These modifications ultimately only affected
Author affiliations strategy is superior in safely promoting long-term dis- a small proportion of patients with SMM and neces-
and support ease control. sitated reconsideration of how we stratify the remainder
information (if
of patients with SMM.
applicable) appear SMM is a biologically heterogeneous disorder encom-
at the end of this passing disease that behaves similarly to MGUS and The Mayo 2018 criteria, also referred to as the 20/2/20
article. disease that is essentially malignancy without overt criteria, incorporates three independent risk factors of
Accepted on CRAB features (i.e., elevated calcium, renal failure, progression (serum monoclonal protein . 2 g/dL, an
February 24, 2020
anemia, bone lesions). Statistics on the risk of pro- involved to uninvolved serum-free light chain ratio
and published at
ascopubs.org on gression reflect this heterogeneity, with some patients . 20, and bone marrow plasma cells . 20%) to
March 17, 2020: with SMM progressing in the first 2 years, whereas categorize patients as either low risk (0 factors), in-
DOI https://doi.org/ others remain stable many years from diagnosis.1,2 termediate risk (1 factor), or high risk (2–3 factors).
10.1200/EDBK_ Clearly, not all patients with SMM would then benefit These risk groups correspond to a 2-year rate of
278915

Joseph, Dhodapkar, and Lonial
risk of progression, with one study utilizing receiver oper-

ating characteristic analysis to identify a cutoff value of
74.4 ng/mL to be predictive of progression to myeloma at
• SMM is a precursor state to symptomatic 5 years, whereas lower levels of B-cell maturation antigen
myeloma defined as the presence of a serum
correlated with MGUS-like disease.13
monoclonal protein 3 g/dL or urinary
monoclonal protein 500 mg per 24 hours With a model to now identify patients with SMM at highest
and/or 10%–60% bone marrow plasma cells risk for progression, should we treat these patients? And if
in the absence of myeloma defining events. so, how should we approach their care? To address these
• SMM is a heterogeneous disorder including questions, we must first review our understanding of what
patients with varying rates of progression to truly differentiates SMM from symptomatic myeloma. What
symptomatic melanoma, with some patients we have learned, largely through cytogenetics and gene
progressing quickly, whereas others remain expression profiling, is that the difference between myeloma
clinically stable several years out from and its precursor states, surprisingly, is not rooted in ge-
diagnosis. netics. MGUS and SMM are genetically advanced states,
• MGUS and SMM are genetically advanced just like symptomatic myeloma. The intraclonal heteroge-
lesions, and gene-expression profiling and neity and major cytogenetic anomalies in symptomatic
cytogenetics have revealed that malignant melanoma are established early in the disease course and
transformation is not because of the acquisition remain the same as the disease undergoes malignant
of new mutations, but rather changes in the transformation. 12,14,15 Though some of these genetic
relationship between the tumor cell and the
changes may ultimately confer increased risk of progres-
bone marrow microenvironment.
sion, these changes do not evolve at the time of that pro-
• The Spanish Myeloma Group QuiReDex study gression. Small prospective studies have also confirmed that
and the ECOG E3A06 trial demonstrated sig- this genomic complexity is determined early in disease
nificant improvements in PFS with lenalidomide
evolution, and there was no correlation with the develop-
with and without dexamethasone in patients
ment of new mutations and disease progression.16,17 The
with SMM, and this benefit was most
pronounced in the high-risk cohort. idea then that MGUS is an advanced state, but is typically
clinically indolent, implies this stability may be conferred by
• There are ongoing clinical trials exploring an
interactions with the tumor microenvironment. Myeloma
intensive treatment approach with potential
cells reside in the bone marrow and interact with other
curative intent, but to date, there is insufficient
evidence that this approach is superior in terms immune cells, bone, stromal, and endothelial cells. Both
of disease control and long-term survival. intrinsic factors within the tumor cells, as well as extrinsic
factors, play an important role in suppressing tumor ex-
pansion. As the disease transitions from asymptomatic to
progression of 5%, 17%, and 46%, respectively.8,9 The 20/ symptomatic, there is a loss of the normal bone marrow
2/20 model is easily applicable and most commonly used microenvironment and thus a shift in the extent of immune
in the clinical setting, but it is limited to these three factors regulation of the malignant clone. We also know that loss of
and does not address the dynamic nature of the disease. effector function in T- and natural killer cells is associated
Other risk factors for progression have also been identified with progression to symtomatic melanoma.12 The balance of
but are not incorporated in the current model, such as IgA permissive and restrictive signaling of the clone is key, and
isotype, immunoparesis of uninvolved immunoglobulins, the interruption of this balance can lead to malignant
increased circulating plasma cells, and high-risk cytoge- transformation.18 Indeed, MGUS cells grow progressively in
netic features including t(4;14), +1q21, del(17p), and humanized murine models with permissive microenviron-
hyperdiploidy.10,11 The IMWG validated the 20/2/20 model ments.19 Recent application of single cell tools has begun to
in a retrospective multicenter study, also including cyto- yield novel insights into the mechanisms underlying attrition
genetic abnormalities t(4;14), t(14;16), 1q gain, and/or of immune surveillance with malignant transformation.20
del13q corresponding with four risk categories (low risk, Understanding the importance of immune regulation in
low-intermediate risk, intermediate-high risk, and high SMM provides a rationale for a preventative strategy that
risk) corresponding with a 2-year risk of progression of seeks to promote long-term disease stability via enhance-
3.7%, 25%, 49%, and 72%, respectively (Table 1).9 A ment of immune function and highlights the potential long-
prospective study of 179 patients with SMM observed term drawbacks of disrupting immune regulation with
a correlation with high-risk gene expression profiling sig- aggressive regimens early in the disease course.
nature and increased risk of progression at 2 years.12 B-cell We are currently seeing two broad strategies in the
maturation antigen level has also been linked with increased management of high-risk SMM: (1) an immune-based

Preventative Treatment of High-risk Smoldering Myeloma
TABLE 1. Validated Risk Stratification Models for Smoldering Multiple Myeloma

Model Risk Factors Risk Group Risk of Progression
5-year PD risk
Mayo 2008 Model 1
M-protein 3 g/dL, clonal bone marrow plasma cells Low (1 factor) 25%
10%, FLCr , 0.125 or . 8
Intermediate (2 factors) 51%
High (3 factors) 76%
2-year PD risk
Spanish Myeloma Group Model18 95% of aberrant BMPCs by MFC plus immunoparesis Low 5%
95% of aberrant BMPCs by MFC or immunoparesis Intermediate 35%
None of the above High 50%
2-year PD risk
Mayo 2018 Model8 M-protein . 2 g/dL, FLCr . 20, BMPCs . 20% Low (0 factors) 5%
Intermediate (1 factor) 17%
High (2–3 factors) 46%
2-year PD risk
IMWG 2019 Model M-protein . 2 g/dL, FLCr . 20, BMPCs . 20%, and high- Low (0 factors) 8%
risk cytogenetics [t(4;14), r(14;16), +1q, del(13q)]
Low-intermediate (1 factor) 21%
Intermediate (2 factors) 37%
High ( 3 factors) 59%
Abbreviations: PD, progressive disease; FLCr, free light chain ratio; BMPC, bone marrow plasma cell; MFC, multiparametric flow cytometry; IMWG,
International Myeloma Working Group.
preventative approach favoring disease control as just de- cells and at least 25% reduction in at least one uninvolved
scribed, or (2) a “curative” approach seeking disease immunoglobulin level.21 Patients on the treatment arm re-
eradication through upfront intensive therapy (Fig. 1). In ceived an induction regimen with lenalidomide 25 mg on days
SMM, sustained minimal residual disease (MRD) negativity 1 to 21 and dexamethasone 20 mg on days 1 to 4 and 12 to
has been proposed as a marker of “operational cure,” al- 15 for nine 28-day cycles. Induction was then followed by
though the follow-up on currently ongoing studies is too maintenance with single-agent lenalidomide dosed at 10 mg
short to draw any definitive conclusions. Which approach on days 1 to 21 of a 28-day cycle for 2 years. The primary
will prove superior is a subject of ongoing debate, and, endpoint was time to progression to symptomatic disease. After
ultimately, only time will provide this answer in terms of a median follow-up of more than 6 years, the lenalidomide/
survival benefit without undue consequences. We would dexamethasone (Rd) arm demonstrated a significant
argue, though, that the important role immune regulation progression-free survival (PFS) benefit with a median time to
plays in the transition from SMM to myeloma, combined progression of not reported versus 23 months (hazard ratio
with recently published phase III data utilizing single-agent [HR], 0.24; 95% CI, 0.14–0.41; p , .0001). Median overall
lenalidomide, supports the former. Although we may see survival (OS) has not yet been reached in either group, though
a small number of patients cured with more intensive at 6 years, 86% of patients in the treatment arm versus 62% of
therapy up front, a strategy aimed at disease control that patients in the observation arm were still alive, and there was
yields similar survival outcomes without the toxicity of tra- a 57% reduction in the risk of death for the treatment arm
ditional myeloma therapies is preferable. versus the observation arm (HR, 0.43; 95% CI, 0.2–0.9; p =
Several studies to date support this immune-based pre- .02).22 Of note, the authors also looked at survival from the start
ventative strategy. The Spanish Myeloma Group (PETHEMA/ of the subsequent line of therapy after progression on study,
GEM) conducted an open-label, randomized phase III trial and there was no difference in survival between the two arms
(QuiReDex) of lenalidomide and dexamethasone versus suggesting that early exposure to Rd does not produce
observation of 119 patients with high-risk SMM. High-risk a more resistant disease phenotype.23 Therapy was overall well
SMM was defined as bone marrow plasma cells 10% and tolerated. Grade 3 adverse events during induction were un-
a monoclonal protein (specified as IgG 3 g/dL, IgA 2 g/dL, common, with the most frequently reported being infection in
or urine paraprotein . 1 g/24 hours), or one of these criteria 6% of patients, asthenia in 6% of patients, neutropenia in 5%
plus 95% phenotypically aberrant bone marrow plasma of patients, and rash in 3% of patients. Infection was the most

too early to detect an OS benefit, given a similar

magnitude of PFS benefit seen in the Spanish
Myeloma Group trial, it is reasonable to expect we
will see this survival benefit in the E3A06 data with
time. Given the median duration of therapy was
23 months in the phase III portion of the trial,
treatment was recommended for 2 years to limit the
risk of long-term adverse events. In the lenalido-
mide arm, grade 3 and 4 nonhematologic toxicities
occurred in 28% of patients, most commonly hy-
FIGURE 1. Current Approaches to the Treatment of High-Risk Smoldering pertension and infections. However, notably, there
Multiple Myeloma
was no difference in reported quality of life between
the two arms. It is important to point out that though
common nonhematologic adverse event, but these were a few patients did achieve deep responses ( very good
predominantly grade 1 and 2 in severity and not at a signifi- partial response [VGPR]), even a limited exposure to
cantly increased rate when compared with the observation lenalidomide translated to a PFS benefit, suggesting that
arm. Ten patients required dose reductions of lenalidomide achieving deep responses may not be as critical in SMM as
during induction, and six patients required dose reductions in with symptomatic myeloma, and that prolonged disease
dexamethasone. Though this study demonstrated both an control is feasible without a large reduction in tumor burden.
improved PFS and OS for patients with high-risk SMM with
Recent phase II data investigating single-agent isatuximab,
treatment, Rd was not widely adopted at the time in SMM
an anti-CD38 IgG1 monoclonal antibody, in patients with
largely due to the use of a combination regimen making it
high-risk SMM for a fixed duration of 30 cycles was pre-
difficult to assess the efficacy of lenalidomide alone, the lack of
sented at the 2019 American Society of Hematology Annual
modern imaging to ensure occult myeloma was definitively
Meeting. This study also adopted more of an immune-based
excluded, and the criteria by which high-risk was defined.
approach as with ECOG E3A06. Isatuximab was dosed
Nonetheless, it has served as an important benchmark sup-
intravenously at 20 mg/kg on days 1, 8, 15, and 22 during
porting early treatment in high-risk SMM and has been used as
cycle, and subsequently dosed on days 1 and 15 of cycles
the basis for several studies focused on delaying progression to
2–6, and then monthly for cycles 7–30. Twenty-four patients
symptomatic melanoma.
were enrolled with a primary endpoint of overall response
Building on the Spanish data and incorporating modern rate (ORR). Outcomes and response rates were similar to
imaging techniques, ECOG designed the largest random- that seen with lenalidomide with an ORR of 62.5%, with
ized phase III trial to date (E3A06) assessing the efficacy of a VGPR or better of 22%. There were five grade 3 treatment-
single-agent lenalidomide compared with observation of related adverse events, which all resolved to baseline, and
182 patients with intermediate- and high-risk SMM. At the most common grade 1–2 events were largely gastro-
a median follow-up of 35 months, there was improvement in intestinal toxicities. Overall, isatuximab was well tolerated,
PFS with a 72% reduction in the risk of progression to and patients actually reported improved quality of life by the
symptomatic disease. Notably, though this benefit was seen end of cycle six of treatment due to reduced anxiety and
across risk groups, it was most pronounced in patients with concern about progression to myeloma.25 Ongoing genomic
high-risk SMM by both the Mayo 2008 and the Mayo 2018 and biomarker analysis may provide further insight on which
risk models. For the overall cohort, the 1-, 2-, and 3-year patients benefit most from early intervention. In the CEN-
PFS in the lenalidomide arm was 98%, 93%, and 91% TAURUS study, single-agent daratumumab was evaluated
compared with 89%, 76%, and 66% in the observation arm, in incremental dosing schedules (short, intermediate, and
respectively. Although the use of lenalidomide was asso- long) and demonstrated improved ORR and 12-month PFS
ciated with some grade 3 and 4 adverse events, as well as favoring the long and intermediate dosing schedule versus
a few cases of second primary malignancies, there were also the more intensive short duration schedule (93%, 75%, and
second primary malignancies in the observation arm with 56%, respectively). These data again suggest less intensive
a cumulative incidence of 5.2% in the lenalidomide arm and longer duration of therapy is more effective in SMM.26
versus 3.5% in the observation arm, suggesting that some of
An intermediate approach is also being tested using pre-
this risk may be associated with MGUS and SMM itself and
24
dominantly combination regimens. Elotuzumab, an anti-
not solely attributable to treatment.
CS1 monoclonal antibody, has been evaluated both as a
The conclusion from the E3A06 data was patients with high- single agent and in combination with Rd in high-risk SMM
risk SMM, as defined by IMWG 2019 or Mayo 2018 criteria, showing a clinical benefit rate of 100%.27-29 Ixazomib in
benefit from early treatment with lenalidomide. Though it is combination with Rd has also shown an ORR of 93.1%

with a VGPR of more than 50%, and carfilzomib in a potential curative strategy. Patients are treated with six
combination has similarly shown high response rates in cycles of induction therapy followed by six additional cy-
high-risk SMM.30,31 Importantly, these combinations have cles of consolidation with daratumumab-KRD (16 mg/kg of
shown good tolerability as well as efficacy (Table 2). intravenous daratumumab on days 1, 8, 15, and 22 of
On the other side of the coin, do we have sufficient evidence cycles 1–2, days 1 and 16 of cycles 3–6, and day 1 of
currently to support the curative strategy for patients with cycles 7–12), carfilzomib (56 mg/m2 on days 1, 8, and 15),
SMM? There are two ongoing studies examining this in- lenalidomide (25 mg on days 1–21), and dexamethasone
tensive treatment approach for SMM with the intent of (40 mg weekly for cycles 1–6, 20 mg weekly for cycles
cure: the GEM-CESAR and ASCENT trials. GEM-CESAR is 7–12) on 28-day cycles. Patients then continue on
a phase II single-arm clinical trial that uses a carfilzomib, maintenance with lenalidomide (dose decreased to 10 mg
lenalidomide, and dexamethasone (KRD) backbone for on days 1–21) and daratumumab (administered on day 1 of
induction followed by high-dose therapy and autologous odd cycles for cycles 13–24) alone for 12 cycles. The
stem cell transplant and maintenance, essentially treating primary endpoint is stringent complete response, and
high-risk SMM as symptomatic myeloma. Ninety transplant- secondary endpoints are MRD status after each treatment
eligible patients were treated with KRD (carfilzomib 36 phase, OS, PFS, and toxicity. Results of this trial are awaited,
mg/m2 twice weekly, lenalidomide 25 mg on days 1–21, and but it is reasonable to expect high rates of MRD negativity
dexamethasone 40 mg weekly) for six 28-day cycles, fol- and good response rates. As with the GEM-CESAR trial,
lowed by autologous stem cell transplant conditioning with long-term follow-up will be important.
melphalan 200 mg/m2, consolidation with KRD for two 28- CONCLUSION
day cycles (carfilzomib 36 mg/m2 twice weekly, lenalido- SMM is a heterogeneous disease, and therefore the chal-
mide 10 mg on days 1–21, and dexamethasone 20 mg lenge lies in correctly identifying patients who will benefit
weekly), and maintenance therapy with the same KRD from early intervention, as well as utilizing what we un-
dosing used in consolidation for 2 years. High risk was again derstand about the disease biology of SMM to offer the most
defined by the same criteria used in the Rd trial ineffective and safe treatment to a patient population that
corporating both the Spanish and Mayo models; 21% of otherwise would be offered observation. Updated risk
patients were classified as high risk per the Mayo model, stratification models, such as the 20/2/20 model, allow us to
52% high risk per the Spanish model, and 27% were high accurately identify those patients, and discussions revolving
risk by both models. The primary endpoint was MRD how best to treat these patients are ongoing. The goal of
negativity rate by next-generation flow postinduction and early treatment in SMM is the prevention of end-organ
posttransplant. At 30 months follow-up, 93% of patients damage, namely renal dysfunction and bone disease,
were alive and free from progression. When evaluating the and improvement in long-term survival. There is no evi-
83 patients who had completed consolidation, the ORR was dence that early treatment with lenalidomide produces
100% with a complete response rate of 76%, and the a more resistant disease at relapse, and data from E3A06
MRD negativity (10 6) rate postconsolidation was 63%. The and others have shown a decrease in the risk of progression
most common grade 3–4 adverse events during induction with similar PFS data that we are seeing even with the use of
included infection (10%), rash (9%), thrombocytopenia more intensive strategies up front. In terms of depth of
(5%), neutropenia (3%), and heart failure (1%). During the response, the GEM-CESAR approach is clearly more ef-
transplant phase, two stem cell mobilizations failed and 11 fective than treatment with either lenalidomide with or
patients required a second mobilization; however, there was without dexamethasone or isatuximab. However, with
no transplant-related mortality. Despite this more intensive similar survival outcomes, it is challenging to justify the
approach and significantly improved responses and MRD toxicities and potential long-term consequences of this in-
negativity rates compared with lenalidomide alone, the tensive therapy at present. We have seen that achievement
3-year PFS in GEM-CESAR was essentially identical to that of MRD negativity in these patients is not necessary to attain
seen in E3A06. Again, this may be showing us that depth of similar results. It is possible that more intensive strategies
response and ORR may not be as important in SMM as it is will provide greater benefit in the long term, but there simply
in symptomatic melanoma, and that disease control can be is not sufficient data as of yet to support this claim. Longer
achieved without major responses.32 Longer follow-up is follow-up and maturation of survival data are needed on the
needed to draw more definitive conclusions. previously discussed ongoing studies, and there are nu-
Similarly, the International Myeloma Foundation–sponsored merous ongoing studies in this space, which will provide
multicenter phase II ASCENT trial is currently evaluat- further insight with time. There are several ongoing studies
ing the efficacy and safety of KRD with the addition of with combination regimens using the Rd backbone with
daratumumab in attempts to deepen responses and rates a third agent including ixazomib (ClinicalTrials.gov identi-
of MRD negativity, provide sustained remissions, and offer fier: NCT02916771) and carfilzomib. There is also an

360
TABLE 2. Completed and Ongoing Clinical Trials and Results in Smoldering Multiple Myeloma
Follow-Up
Clinical Trial Phase Treatment n (Months) Results Safety Profile (Grade ‡ 3 AEs)
Completed Enrollment/Ongoing Studies (Data Available)
Spanish Myeloma III Len/dex vs. obs: Induction: Len 25 mg on days 119 75 Median TTP: NR vs. 23 months (HR, Grade 3: Infection (6%), asthenia (6%),
Group/PETHEMA22 1–21/dex 20 mg on days 1–4 and 12–15 on 0.24; 95% CI, 0.14–.41; p , neutropenia (5%), rash (3%); Grade 4: none
28-day cycles x 9 cycles vs. obs; .0001)
Maintenance: len 10 mg on days 1–21 on 28-
day cycles x 2 years vs. obs
ECOG E3A0624 II/III Len vs. obs: Len 25 mg on days 1–21 of 28-day 182 28 ORR: 48.9% Len arm vs. 0%; 1-, 2-, Grade 3/4 nonhematologic: 28% (fatigue, n = 5);
cycles until progression 3-year PFS: 98%, 93%, 91% vs. Grade 4 hematologic: neutropenia 5.7%; SPMs:
89%, 76%, and 66% for the obs 5.2% (len) vs. 3.5% (obs)
arm
E-PRISM II Elo/len/dex: Induction: Elo 10 mg/kg IV on days 50 29 ORR: 84%; 29-month PFS: 95% Grade 3: hypophosphatemia (34%), neutropenia
(NCT03379394)27 1, 8, 15, 22 cycles 1–2, days 1 and 15 on (26%), lymphopenia (22%); Grade 4:
cycles 3–8; Len 25 mg on days 1–21 cycles hypophosphatemia (6%), cholecystitis (2%),

1–8; Dex 40 mg on days 1, 8, 15, 22 cycles cataract (2%), hyperglycemia (2%), neutropenia
1–2, days 1, 8, 15 cycles 3–8; Maintenance: (2%), thrombocytopenia (2%)
Elo 20 mg/kg IV on day 1 and len on days
1–21 of a 28-day cycle x 24 cycles
NCT0291677130 II Ixa/len/dex: Induction: Ixa 4 mg days 1, 8, 15, 53 14.4 RR ( 1 cycle): ORR 91.1%, CR 31. Grade 3: hypertension (6.3%), hypophosphatemia
len 25 mg days 1–21, dex 1, 8, 15, 22 on 28- 1%, VGPR 20%, PR 40%, MR (4.2%), rash (4.2%); Grade 4: TCP (4.4%),
day cycles x 9; Maintenance: ixa (4 mg) and 10% neutropenia (4.4%), hyperglycemia (2.2%)
len (15 mg) x 24 cycles
NCT0296055525 II Isatuximab: Isatuximab 20 mg/kg IV days 1, 8, 24 CBR 79%; ORR 62.5%, CR MRD neg Grade 3: dyspnea (n = 2), headache (n = 1),
15, 22 cycle 1; days 1 and 15 cycles 2–6, day 10-5 5%, VGPR 17%, PR 42%, neutropenia (n = 1), UTI (n = 1)
1 cycles 7–30 MR 18%, SD 21%
CENTAURUS II Daratumumab IV: Arm A (long): weekly cycle 1, 41* 15.8 ORR 56%/54%/38%; 12-month PFS Infection (, 5% in all arms)
(NCT02316106)26 Q2W cycles 2–3, Q4W cycles 4–7, day 1 95%/88%/81%
cycles 8–20, Q8W C20+; Arm B
(intermediate): weekly cycle 1, day 1 cycles
2–20, Q8W C20+; Arm C (short): weekly cycle
1
NCT0157248031 II Carfilzomib/len/dex: Car 20/36 mg/m2 on days 12 15.9 CR 100%, MRD negativity 92% Rash (33%), neutropenia (17%), anemia (17%),

1, 2, 8, 9, 15, 16; len 25 mg days 1–21, dex infection (8%), cardiac (8%)
20/10 mg (cycles 1–4/5–8) on days 1, 2, 8, 9,

15, 16, 22, 23 on 28-day cycles x 8 followed
by len x 24 cycles
GEM-CESAR II Induction: KRD x 6 cycles (carfilzomib IV 20/ 90 28 OS 98%, 30 months PFS 93%; Induction (grade3/4): infections (10%), rash (9%),
(NCT02415413)32 36 mg/m2 on days 1, 2, 8, 9, 15, 16; len 25 mg following consolidation and 1 year TCP (5%), neutropenia (3%), heart failure (1%);
days 1–21; dex 40 mg on days 1, 8, 15, 22) maintenance: CR 85%, VGPR consolidation and maintenance (grade 3/4):
followed by ASCT. Consolidation: KRD x 2 10%, PR 5%, MRD negativity rate neutropenia (8%), TCP (8%), infections (5%)
cycles; maintenance: Rd x 12 cycles 68%
TABLE 2. Completed and Ongoing Clinical Trials and Results in Smoldering Multiple Myeloma (Continued)
Follow-Up
Clinical Trial Phase Treatment n (Months) Results Safety Profile (Grade ‡ 3 AEs)
Ongoing Studies (Data Not Yet Available)
ASCENT II Dara + car/len/dex: Induction: KRD + dara x 6 Not available Not available
(NCT03289299) cycles; followed by consolidation x 6 cycles
and maintenance with dara-len x 12 cycles
DETER-SMM III Dara/len/dex vs. len/dex: Dara 16 mg/kg IV on Not available Not available
(NCT03937635) days 1, 8, 15, 22 cycles 1–2, days 1 and 15
cycles 3–6, day 1 cycles 7–24; len 25 mg days
1–21, dex on days 1, 8, 15, 22 on 28-day
cycles 1–12 x 24 cycles vs. len/dex at the
same doses x 24 cycles
Abbreviations: AEs, adverse events; Len, lenalidomide; Dex, dexamethasone; Obs, observation; TTP, time to progression; NR, not reported; HR, hazard ratio; ORR, overall response rate; PFS,
progression-free survival; SPM, second primary malignancy; Elo, elotuzumab; Ixa, ixazomib; RR, response rate; CR, complete response; VGPR, very good partial response; PR, partial response,
MR, minimal response; TCP, thrombocytopenia; CBR, clinical benefit rate; MRD, minimal residual disease; SD, stable disease; UTI, urinary tract infection; IV, intravenous; Q2W, every 2 weeks;
Q4W, every 4 weeks; Q8W, every 8 weeks; Car, carfilzomib; KRD, carfilzomib, lenalidomide, and dexamethasone; ASCT, autologous stem cell transplant; OS, overall survival; Dara, daratumumab.
*In each arm.

361
ongoing ECOG-ACRIN study (ClinicalTrials.gov identifier: and dexamethasone, an approved regimen for symptom-
NCT03937635) in high-risk SMM that could further shed atic myeloma, versus lenalidomide and dexamethasone,
light on the issue comparing daratumumab, lenalidomide, a tested preventative strategy in SMM.

1
Winship Cancer Institute of Emory University, Atlanta, GA AND DATA AVAILABILITY STATEMENT
EDBK_278915.
Sagar Lonial, MD, FACP, Winship Cancer Institute of Emory University,
1365C Clifton Rd., Atlanta, GA 30322; Twitter: @SagarLonialMD; email:
[email protected].
REFERENCES
1. Kyle RA, Remstein ED, Therneau TM, et al. Clinical course and prognosis of smoldering (asymptomatic) multiple myeloma. N Engl J Med. 2007;356:2582-2590.
2. Rajkumar SV, Merlini G, San Miguel JF. Haematological cancer: redefining myeloma. Nat Rev Clin Oncol. 2012;9:494-496.
3. Rajkumar SV, Larson D, Kyle RA. Diagnosis of smoldering multiple myeloma. N Engl J Med. 2011;365:474-475.
4. Larsen JT, Kumar SK, Dispenzieri A, et al. Serum free light chain ratio as a biomarker for high-risk smoldering multiple myeloma. Leukemia. 2013;27:941-946.
5. Kastritis E, Terpos E, Moulopoulos L, et al. Extensive bone marrow infiltration and abnormal free light chain ratio identifies patients with asymptomatic myeloma at
high risk for progression to symptomatic disease. Leukemia. 2013;27:947-953.
6. Kastritis E, Moulopoulos LA, Terpos E, et al. The prognostic importance of the presence of more than one focal lesion in spine MRI of patients with asymptomatic
(smoldering) multiple myeloma. Leukemia. 2014;28:2402-2403.
7. Hillengass J, Fechtner K, Weber MA, et al. Prognostic significance of focal lesions in whole-body magnetic resonance imaging in patients with asymptomatic
multiple myeloma. J Clin Oncol. 2010;28:1606-1610.
2018;8:59.
9. San Miguel J, Mateos MV, Gonzalez V, et al. Updated risk stratification model for smoldering multiple myeloma (SMM) incorporating the revised IMWG diagnostic
criteria. J Clin Oncol. 2019;37:15s (suppl; abstr 8000).
10. Rajkumar SV, Gupta V, Fonseca R, et al. Impact of primary molecular cytogenetic abnormalities and risk of progression in smoldering multiple myeloma.
Leukemia. 2013;27:1738-1744.
11. Neben K, Jauch A, Hielscher T, et al. Progression in smoldering myeloma is independently determined by the chromosomal abnormalities del(17p), t(4;14), gain
1q, hyperdiploidy, and tumor load. J Clin Oncol. 2013;31:4325-4332.
12. Dhodapkar MV, Sexton R, Waheed S, et al. Clinical, genomic, and imaging predictors of myeloma progression from asymptomatic monoclonal gammopathies
(SWOG S0120). Blood. 2014;123:78-85.
13. Dispenzieri A, Soof C, Rajkumar SV, et al. Serum BCMA levels to predict outcomes for patients with MGUS and smoldering multiple myeloma (SMM). J Clin Oncol.
2019;37:15s (suppl; abstr 8020).
14. Morgan GJ, Walker BA, Davies FE. The genetic architecture of multiple myeloma. Nat Rev Cancer. 2012;12:335-348.
15. Zhan F, Hardin J, Kordsmeier B, et al. Global gene expression profiling of multiple myeloma, monoclonal gammopathy of undetermined significance, and normal
bone marrow plasma cells. Blood. 2002;99:1745-1757.
16. Walker BA, Wardell CP, Melchor L, et al. Intraclonal heterogeneity is a critical early event in the development of myeloma and precedes the development of clinical
symptoms. Leukemia. 2014;28:384-390.
17. Zhao S, Choi M, Heuck C, et al. Serial exome analysis of disease progression in premalignant gammopathies. Leukemia. 2014;28:1548-1552.
18. Dhodapkar MV. MGUS to myeloma: a mysterious gammopathy of underexplored significance. Blood. 2016;128:2599-2606.
19. Das R, Strowig T, Verma R, et al. Microenvironment-dependent growth of preneoplastic and malignant plasma cells in humanized mice. Nat Med. 2016;
22:1351-1357.
20. Bailur JK, McCachren SS, Doxie DB, et al. Early alterations in stem-like/resident T cells, innate and myeloid cells in the bone marrow in preneoplastic
gammopathy. JCI Insight. 2019;5:5.
21. Pérez-Persona E, Vidriales MB, Mateo G, et al. New criteria to identify risk of progression in monoclonal gammopathy of uncertain significance and smoldering
22. Mateos MV, Hernández MT, Giraldo P, et al. Lenalidomide plus dexamethasone for high-risk smoldering multiple myeloma. N Engl J Med. 2013;369:438-447.

23. Mateos M, Hernandez JM, Giraldo P, et al. Sustained overall survival benefit with lenalidomide plus dexamethasone versus no treatment in patients with
smoldering myeloma at high risk of progression to myeloma: long term analysis. Blood. 2016;128:3308-3008.
24. Lonial S, Jacobus S, Fonseca R, et al. Randomized trial of lenalidomide versus observation in smoldering multiple myeloma. J Clin Oncol. Epub 2019 Oct 25.
25. Manasanch E, Jagannath S. A multicenter phase II single arm trial of isatuximab in patients with high risk smoldering multiple myeloma (HRSMM). Blood. 2019;
184 (suppl 1):8118.
26. Hofmeister C, Chari A, Cohen Y, et al. Daratumumab monotherapy for patients with intermediate or high-risk smoldering multiple myeloma (SMM): Centaurus,
a randomized, open-label, multicenter phase 2 study. Blood. 2017;130 (suppl 1):510.
27. Ghobrial IM, Badros AZ, Vredenburgh JJ, et al. Phase II trial of combination of elotuzumab, lenalidomide, and dexamethasone in high-risk smoldering multiple
myeloma. Blood. 2016;128:976.
28. Jagannath S, Laubach J, Wong E, et al. Elotuzumab monotherapy in patients with smouldering multiple myeloma: a phase 2 study. Br J Haematol. 2018;
182:495-503.
29. Liu CJ, Ghobrial IM, Bustoros M, et al. Phase II trial of combination of elotuzumab, lenalidomide, and dexamethasone in high-risk smoldering multiple myeloma.
Blood. 2018;132 (suppl 1):154.
30. Bustoros M, Liu CJ, Reyes K, et al. Phase II trial of the combination of ixazomib, lenalidomide, and dexamethasone in high-risk smoldering multiple myeloma.
Blood. 2018;132 (suppl 1):804.
31. Korde N, Roschewski M, Zingone A, et al. Treatment with carfilzomib-lenalidomide-dexamethasone with lenalidomide extension in patients with smoldering or
newly diagnosed multiple myeloma. JAMA Oncol. 2015;1:746-754.
32. Mateos M. Curative strategy (GEM-CESAR) for high-risk smoldering myeloma (SMM): carfilzomib, lenalidomide, and dexamethasone (KRd) as induction followed
by HDT-ASCT, consolidation with Krd and maintenance with Rd. Blood. 2019;134 (suppl 1):781.

Treatment of Smoldering Multiple Myeloma:

Expectant Observation Should Still Be
the Standard
Rafael Fonseca, MD1 and Miguel Gonzalez-Velez, MD1
overview
Recent clinical trials have addressed the notion of early treatment of smoldering multiple myeloma (SMM).
The results evidence improvement in progression-free survival and, in one study, overall survival. Although the
treatment of SMM can be considered under specific circumstances, we propose here that careful interpretation
of the clinical trials and the patient-specific data are needed before recommending therapy. In particular, many
questions remain regarding the best regimen to be used as well as how to adapt based on the underlying disease
biology. Hematologists should have a very thorough understanding of models designed to predict the progression
from SMM to multiple myeloma, because their correct interpretation is paramount to establish proper care.
Although there is no doubt that treatment should be started before overt end-organ damage, we do not believe
that the current data support the widespread treatment of all SMM.
INTRODUCTION converted to the same risk of progression as MGUS.2

Smoldering multiple myeloma (SMM) is an intermediate This inflection, with decreased risk of progression over
category among the plasma cell neoplasms that de- time, has been taken to imply that a better definition of
notes a stage more advanced than that of the what constitutes SMM versus MM is needed. Specif-
monoclonal gammopathy of undetermined significance ically, one can only wonder whether those progressing
(MGUS) but not yet with the clinical consequences of early, in fact, had incipient MM.
(active) multiple myeloma (MM). Although initially Because of the possibility of no progression to MM in
recognized in a categorical fashion, we now recognize many cases, it was advised to be conservative with
a continuous spectrum that evolves from minimal the management of SMM and only initiate therapy
states of clonal plasmacytosis (MGUS) to patients with when overt evidence of end-organ damage occurred.
a malignant expansion of clonal plasma cells and In a meeting held in 2002 at the National Institutes of
subsequent clinical consequences. However, this Health to discuss Waldenström macroglobulinemia,
current paradigm of understanding would not be a conversation ensued regarding SMM. Richard Fisher,
possible without the recognition and definition of SMM. MD, of Fox Chase Cancer Center, made the astute
The word smoldering denotes a dangerous “meta- observations that, although MM specialists cher-
stable” situation in which cells can rapidly expand and ished a mantra of not treating too early, they should
lead to active MM.1 ask themselves if they sometimes treated too late.
SMOLDERING MYELOMA DEFINITION AND SLIM-CRAB Over the next several years, many studies were
conducted to create a better definition of what consti-
In the past, a diagnosis of SMM was sometimes wel-
tutes true SMM versus MM. A number of factors
Author affiliations comed with a sense of relief because the patient did not
and support predicting an increased risk of progression were pro-
have MM, but this failed to account for the real pos-
information (if posed, but they were not widely used as the mantra
sibility of progression. Patients were often told not to
applicable) appear remained that one should not treat SMM, because it
at the end of this
worry, as sometimes the disease did not progress for
needed to declare itself as MM via evidence of end-
article. many years. However, published series demonstrate
organ damage.3
Accepted on that patients with SMM have a substantially higher risk
February 28, 2020 of progression to MM than MGUS. In a seminal paper The MM community proposed the acronym CRAB as
and published at from the Mayo Clinic, the progression of SMM to MM a mnemonic of what elements constitute end-organ
ascopubs.org on
was estimated to be approximately 10% per year for the damage: C for hypercalcemia, R for renal dysfunction/
March 24, 2020:
first 5 years after diagnosis (cumulative at 5 years of failure, A for anemia, and B for bone disease.4 These
10.1200/EDBK_ 50%) and 3% per year for the next 5 years (cumulative complications of MM can be easily recognized while
280179 total of 65% at 10 years’ risk of progression); it then florid, but many subtle variations exist. It is often said

Most Smoldering Multiple Myeloma Should Be Observed and Not Treated
facets existent in the spine. In contrast, fractures of long

bones resulting from MM, although immediately painful
and serious, can often be successfully treated with surgery.
• Proper management of the early stages of the Bone lesions in the skull and ribs are rarely of major
plasma cell tumors requires a careful analysis of
significance; thus, specifically preventing and predicting
factors indicating imminent progression.
vertebral compression fractures is of paramount impor-
• Although much progress has been made in the tance for the optimal management of SMM. We un-
ability to predict progression, the practical ap- derstand much better the molecular mechanisms leading
plication at the individual level still needs to be
to bone destruction in MM9,10 but lack good predictive
done with caution, particularly as it relates to the
biomarkers. Soluble markers lack specificity or technical
serum free light chains.
simplicity. Imaging is by necessity retroactive. Because of
the aforementioned considerations, predicting bone damage
remains one of the most important gaps for surveillance
that it is easier to teach about CRAB than it is to fully un- in SMM.
derstand it in the clinic. Of the various aspects related to MM To enhance our ability to identify patients at risk for im-
progression to CRAB, two are noteworthy because of their mediate progression to MM from SMM, the International
serious nature: renal failure and bone disease, the so-called Myeloma Working Group convened a group of experts that
“claws of the CRAB.”5 Renal failure as a consequence of in 2014 proposed a new set of criteria that would allow
MM is sometimes irreversible and condemns the patient clinicians to start therapy, even in the absence of overt
to lifelong renal replacement therapy. Fortunately, we now CRAB.11 These criteria were broadly selected by identifying
have a biomarker capable of predicting the risk of cast two studies in the literature that could propose biomarkers
nephropathy (i.e., myeloma kidney), the serum free light capable of predicting an 80% risk of progression 2 years
chains (sFLCs). Substantial elevations (usually . 100 after detection in SMM. These criteria are having a plas-
mg/dL) of the absolute sFLC create challenges for the macytosis of 60% or greater, having two or more bone le-
management of their excretion into the urine and place the sions on an magnetic resonance imaging scan, and having
patient at risk for renal damage.6,7 This occurs because the an sFLC ratio greater than 20 (involved/uninvolved). It is now
intact immunoglobulin has a molecular size that prevents it customary to propose therapy initiation for patients who
from filtering in the glomerular units, but the sFLC can pass harbor these features. Although the criteria have been largely
into the urine. This, previously referred to as Bence-Jones accepted, some questions and reservations remain. For in-
proteinuria, is simply a description of the free light chains stance, most clinicians consider the presence of two or more
detectable in the urine. Therefore, our ability to predict renal lesions in a PET-CT scan as equivalent, although no formal
damage associated with MM has been greatly enhanced. On validation has occurred.
the contrary, patients with a normal or low level of the sFLC
The most problematic aspect, and perhaps the most dis-
can be reassured, because their risk of progression to cast
puted, is the validity of the sFLC ratio as a marker of pro-
nephropathy is minimal. Thus, the sFLC can be used to
gression. In many practices, patients have abnormal and
predict the risk of progression to renal damage, one of the
excessive ratios and do not exhibit progression. The ab-
claws!
normal ratio can be said to be predictive because of at least
It has proven to be much more difficult to predict bone three putative mechanisms: its effect on the risk of renal
damage in MM. In most cases, bone damage is detected by damage, its effect of it being a marker of a more expansive
imaging, and only after extensive damage as occurred. plasmacytosis, and its effect because of its association with
Alternatively, patients can report evidence of bone damage more aggressive biology. The effect as a marker of renal
in the form of pain that is exacerbated by weight bearing or damage can be discerned by the absolute serum con-
movement. Bone damage is highly problematic when it centration of the sFLC.6,7 The effect related to it being
leads to structural changes that will result in persistent pain a marker of tumor burden can be checked against and
and worse overall survival.8 Compression fractures of the validated against other markers of tumor burden (bone
spine are perhaps the most problematic aspect of MM bone marrow plasmacytosis, quantitative immunoglobulins, or M
disease. When they occur, they can lead to pain and dif- spike sometimes) or consequence of tumor burden ex-
ficulties with mobility, loss of height, and even respiratory pansion (anemia, incipient bone disease, or imaging ab-
difficulty. Even among patients successfully treated for MM, normalities). Last, its effect because of more aggressive
pain associated with pre-existing bone damage may be biology may be more difficult to elucidate, because the
enduring and important consequences to the quality of life usual genetic testing cannot fully describe the intricacies
of patients. This is perhaps a consequence of the intricacies and complexities of plasma cell biology. Also, it is worth
of the anatomy of the vertebral bodies and the many articular remembering that the ratio is dependent on wide fluctuation

Fonseca and Gonzalez-Velez
and variations of the uninvolved sFLC, often creating much of progression. At the Mayo Clinic, we used to measure the
greater fluctuations in the sFLC ratio that could be ascribed plasma cell labeling, a simple slide-based test that mea-
to true progression to MM. Careful interpretation of the ratio sured clonal proliferation as an indicator of mitotic rate and
is essential, as these criteria alone could be a trigger for a higher risk of progression.21 More recently, this is de-
therapy initiation, given ongoing research. These expanded termined via flow cytometry in the form of S phase de-
criteria are often referred to as “SliM-CRAB.” termination. Because of the lack of large series, most of
these factors have not been routinely incorporated into
The other features of CRAB, anemia and hypercalcemia,
clinical use.
can usually be corrected and are not enduring. We refer to
them as the not-so-dangerous part of the CRAB. Two models have prevailed in providing risk stratification for
progression to MM.6,13 The Spanish have proposed that an
Prognostic Models to Predict Progression excess abnormal plasmacytosis ( 95%) is indicative of
In an effort to better understand the possibility of pro- a high risk of progression and thus can be integrated into the
gression, several risk models have been proposed.12,13 Most clinic.13 More recently the Mayo Clinic proposed a system
of these models have been based on two determinants: that incorporates three markers and is called the 20-2-20
tumor bulk and disease biology. Although undoubtedly model: greater than 20% plasmacytosis, serum M spike
other mechanisms may be operative in allowing progres- greater than 2 g/dL, or an sFLC ratio greater than 20 to
sion, such as loss of immune surveillance, so far most of the stratify patients.6 Patients with two or more of these criteria
models focus on the two aforementioned factors. are said to have high-risk SMM. Patients with one of these
criteria are said to have intermediate risk, and those without
Tumor burden can be measured in multiple ways, including
any of them are classified as low risk. Based on the results of
direct measurement of bone marrow involvement as well as
recent clinical trials, this classification is of great impor-
an indirect approach via measurement of good biologic
tance, because some of these studies would propose
surrogates. Measuring the monoclonal protein serum con-
treatment of SMM when the features indicate high-risk
centration, the quantitative immunoglobulins, and the serum
disease. This assignation can be made incorporating the
concentration of the sFLC, in fact, constitutes an indirect
sFLC ratio and its attendant nuances in measurement, so
measurement of tumor burden. Arguably, even measuring
one must remain cautious in overinterpreting data and
circulating numbers of plasma cells might be indirectly in-
hastily proposing treatment.
dicative of tumor burden.14,15 Novel markers to measure
tumor burden, such as the serum concentration of BCMA, Recent Clinical Trials Addressing the Treatment of SMM
are also being proposed for the disease.16,17 The prevailing Several clinical trials were conducted previously to address
hypothesis is that the risk of progression has a substantial the potential benefit of the treatment of SMM. These trials
stochastic event: the greater the number of clonal cells, the are mostly of historic interest, as they used agents we would
more likely SMM will progress to MM. no longer consider of benefit in the clinic as treatments
Although the measurement of tumor burden in the bone for SMM.
marrow is fraught with both biologic (patchy nature) and The first randomized phase III trial to address the early
technical difficulties (obtaining a good sample), the overall treatment of SMM was published by Mateos and colleagues.22
approximation is usually considered a good indicator of net In that trial, patients were assigned to receive therapy with
clonal expansion. The Spanish group augmented the in- the combination of lenalidomide and dexamethasone
terpretation of high-bulk SMM and incorporated it into its versus observation alone. The trial was the first of its kind to
models by measuring the fraction of plasma cells that are demonstrate an improvement in progression-free and
clonally expanded.18 Indirectly, and in an imprecise and overall survivals for patients who received therapy.23 In-
subjective manner, imaging can also provide information terestingly, despite these findings, the trial did not become
regarding tumor burden. Marrow signal abnormalities in the practice changing and was seen only as of the first of its
magnetic resonance imaging scan and PET-CT scans often kind. Reservations about the design and also about the
correlate with other laboratory and pathology measurements relatively poor outcome reported by those in the obser-
of tumor burden. vation arm clouded the adoption of this practice in the
Unraveling disease biology has been more complicated and clinic. Most MM experts wanted additional assurances from
has not been widely used to predict the risk of progression to other similar clinical trials.
active MM. Several markers, such as genetic category More recently, the Eastern Cooperative Oncology Group
determined by fluorescence in situ hybridization and gene published the randomized phase III trial E3A06.24 In this
expression profiling, have been reported to have some trial, patients were randomly assigned to receive treatment
predictive ability.19,20 Indirect markers of more aggressive with single-agent lenalidomide versus observation alone.
biology often percolate as potential indicators of higher risk The trial reached its primary endpoint, progression-free

survival, and follow-up will continue to later address overall First, defining what constitutes SMM and what features best
survival. This trial adds more data in support of the concept predict progression is now challenged with the advent of
of early treatment of SMM. This trial performed a subset more modern imaging and laboratory techniques. The
analysis to determine which patients seem to most benefit original studies describing SMM lacked the detailed imaging
from therapy. In particular, patients deemed to have high- analysis that comes from the use of tools such as PET-CT
risk SMM, as determined by the Mayo criteria (20-2-20), scans.2 Absent this information, it becomes more difficult to
were those thought to benefit the most from early therapy. It ascertain in which cases SMM exists versus in which cases its
is worth remembering that high risk meant having two of detection is merely detection of a transient state of clonal
three of the aforementioned markers, including the sFLC evolution that will ultimately result in progression to MM. The
ratio greater than 20 (Fig. 1).24 notion that 10% of patients progress per year for the first
Other clinical trials are ongoing and addressing the role of 5 years and then the risk diminishes should be considered as
more involved therapy for SMM. In some instances, these empiric evidence in favor of this hypothesis.
trials include treatment paradigms that mimic those used for Second, the criteria defining what constitutes high-risk SMM
the treatment of active MM: triplets followed by stem cell can be used in extreme situations when the values obtained
transplant and then more therapy.25 Sometimes, they test greatly diverge from the cutoff points proposed for the 20-2-
the addition of daratumumab to lenalidomide and dexa- 20 system. But proximity to this boundary can easily cat-
methasone (EAA173, DETER-SMM). The rationale for these egorize a patient as having high- versus intermediate-risk
trials is that, if this therapy appears to confer advantages disease and hence trigger an algorithmic cascade that leads
when used for the treatment of MM, it might creative cu- to therapy initiation. Creating categoric values in prognostic
rative regimens when used for the treatment of SMM. The models served us well in a time when the information was
results of such trials will read into the future. mainly transmitted in the printed form or needed to be
memorized. The beauty of its simplicity can create difficulty
Limitations of the Data when widely proposing treatment guidelines. Creating elec-
Many considerations apply when interpreting the data re- tronic tools that accept a number of values, all treated as
garding the definition of the disease, prognostic models, and continuous variables, and provide an estimated risk of pro-
applicability of recently reported trials. They will be discussed gression would be better. Unless the situation is urgent,
in greater detail in the following section. repeated measurements and changes over time would seem
FIGURE 1. Progression-Free
Survival for High-Risk SMM
in E3A06 Study
Kaplan-Meier estimates of
progression-free survival by
treatment arm within Mayo 2008
risk subgroups in phase III: (A)
high risk, (B) intermediate risk,
and (C) low risk.24
Copyright © 2020 American
Society of Clinical Oncology.

to be more important than a single set of determinations. As immunomodulatory imide drugs) are essential for the
mentioned before, we have the ability to predict in a large proper treatment of active MM. Having MM cells become
number of cases the risk of renal damage by the sFLC. A resistant to lenalidomide might create important hurdles at
major step forward would be to develop adequate markers the time of disease progression. Then the question be-
predicting bone disease development. comes, should we treat SMM with the best available therapy
for MM? Those clinical trials are being conducted but will not
Third, all of these prognostic models are static and do not
ready for some time.
consider evolution over time. In most situations, there is
ample clinical opportunity to monitor patients and decide if Sixth, if indeed most models that predict progression in-
progression to MM is occurring. Arguably, more frequent corporate measurements of tumor burden, then one pos-
testing would allow interception before end-organ damage. sibility is that reducing tumor burden is intrinsically beneficial
Also, models that include determinations of which changes by numerically reducing this risk; indeed, this is how lena-
over time are meaningful would be more dynamic and likely lidomide might work. What about other novel therapeutic
to predict early progression. Such models have begun to be agents, such as daratumumab? Fortunately, such clinical
used in other hematologic conditions, such as B-cell trials are ongoing. One could propose that the ideal treatment
chronic lymphocytic leukemia. That would require a con- strategy for SMM (and not simply a situation of “pseudo-
certed effort, not only by clinicians directly caring for pa- SMM,” but really early evolving MM) should use agents that are
tients but also by health systems and electronic medical best tolerated and that engender the lower risk of resistance.
records. Doing so with proficiency would be an important Seventh, the treatment of different biologic subtypes of SMM
advance but also a challenging undertaking. should be considered when deciding on therapy. Although
Fourth, these authors have substantial reservations about lenalidomide might be beneficial at large, we hypothesize
recommending the sFLC ratio as a variable to decide on the that treating SMM with a t(4,14) is not ideal with this agent
risk of progression to MM. Although the literature has re- alone and that the introduction of a proteasome inhibitor
ported the ratio as predictor, there are many examples might be desirable. Correlative clinical trials have not shed
found in the clinic on how this can be misleading. In- information because of the limited number of patients who
terpretation of the meaning of sFLC requires substantial have samples collected and associated with these trials.
clinical expertise, usually not fully developed unless the Likewise, what is one to do in cases of SMM with a t(11,14)?
practice is devoted to the care of patients with MM. In Clinical trials with venetoclax combinations should be
particular, small absolute fluctuations of the serum con- considered and perhaps will form part of a more person-
centration of the uninvolved sFLC, great in relative value, alized approach in the future for the treatment of MM.26
can easily create major changes in the actual ratio. The Recommendations Into the Future
authors have experienced anxiety when patients read ratio Although there are some situations in which early therapy may
variations as a consequence of this and have also seen be appropriate, we suggest therapy for SMM should be
effective therapy changed because of irrelevant changes in considered, still, with a conservative stance. In cases in which
the ratio. Although the ratio is important to understand the there is doubt in diagnostic allocation between true SMM and
clonal nature of an sFLC elevation, it has no biologic value. early evolving MM, treatment can be considered. In cases
It is only the absolute elevation of the sFLC that carries when the high-risk tenets are satisfied and without a doubt of
consequences to patients. The ratio can be used to adjust borderline interpretations (e.g., in those with not only a high
new levels of sFLC when both (involved and uninvolved sFLC ratio but also a noteworthy elevation of the absolute sFLC
sFLC) increase in the setting of renal dysfunction. Additional serum concentration), therapy should be considered. But even
markers that do not depend on renal clearance would be in these cases, uncertainty remains as to the best regimen to
desirable. be used, given the aforementioned considerations.
Fifth, what to use for therapy is difficult to decide. The Additional clinical trials are needed, and investigators con-
relevant clinical literature would indicate that lenalidomide ducting or having completed trials are to be commended,
alone or in combination with dexamethasone should be the because the interpretations of such trials are fraught with
treatment of choice. However, several reasons for concern uncertainty. Perhaps there is a future when the therapy of
must be pointed out. The use of single/doublet regimens is MM is done with abbreviated and safer regimens, such that
more likely to engender resistance to anti-MM agents. Al- the conversation of whether SMM should be treated will
though this resistance has not been reported after exposure be moot, and treatment will be given in most cases. And
to lenalidomide, one should recall that absence of evidence perhaps then we will hold conversations about which
is not evidence of absence. Given standard evolutionary MGUS patients should be treated. But until then, we believe it
principles, we can assume that, in time, resistance will be is appropriate to recommend expectant observation for most
reported. Lenalidomide and related medications (the so-called patients with SMM.

ACKNOWLEDGMENT coined the term smoldering multiple myeloma, developed

This article is dedicated to the memory of Dr. Phillip Greipp, the plasma cell labeling index and the International Staging
who died in February 2020. Along with Dr. Robert Kyle, he System, and was a mentor to many in the field.

1
Mayo Clinic, Phoenix, AZ AND DATA AVAILABILITY STATEMENT
EDBK_280179.
Rafael Fonseca, MD, 13400 E. Shea Blvd., Mayo Clinic, Scottsdale, AZ
85259; Twitter: @rfonsi1; email: [email protected].
REFERENCES
1. Kyle RA, Greipp PR. Smoldering multiple myeloma. N Engl J Med. 1980;302:1347-1349.
2. Kyle RA, Remstein ED, Therneau TM, et al. Clinical course and prognosis of smoldering (asymptomatic) multiple myeloma. N Engl J Med. 2007;356:2582-2590.
3. Cherry BM, Korde N, Kwok M, et al. Modeling progression risk for smoldering multiple myeloma: results from a prospective clinical study. Leuk Lymphoma. 2013;
54:2215-2218.
4. Kyle RA, Gertz MA, Witzig TE, et al. Review of 1027 patients with newly diagnosed multiple myeloma. Mayo Clin Proc. 2003;78:21-33.
5. Fonseca R, Jain T. Bone disease in myeloma: the claws of CRAB. Clin Cancer Res. 2016;22:1301-1303.
6. Dispenzieri A, Kyle RA, Katzmann JA, et al. Immunoglobulin free light chain ratio is an independent risk factor for progression of smoldering (asymptomatic)
multiple myeloma. Blood. 2008;111:785-789.
7. Leung N, Bridoux F, Batuman V, et al. The evaluation of monoclonal gammopathy of renal significance: a consensus report of the International Kidney and
Monoclonal Gammopathy Research Group. Nat Rev Nephrol. 2019;15:45-59.
8. Sonmez M, Akagun T, Topbas M, et al. Effect of pathologic fractures on survival in multiple myeloma patients: a case control study. J Exp Clin Cancer Res. 2008;
27:11.
9. Bataille R, Chappard D, Marcelli C, et al. Mechanisms of bone destruction in multiple myeloma: the importance of an unbalanced process in determining the
severity of lytic bone disease. J Clin Oncol. 1989;7:1909-1914.
10. Alsina M, Boyce B, Devlin RD, et al. Development of an in vivo model of human multiple myeloma bone disease. Blood. 1996;87:1495-1501.
11. Rajkumar SV, Dimopoulos MA, Palumbo A, et al. International Myeloma Working Group updated criteria for the diagnosis of multiple myeloma. Lancet Oncol.
2014;15:e538-e548.
2018;8:59.
13. Pérez-Persona E, Vidriales M-B, Mateo G, et al. New criteria to identify risk of progression in monoclonal gammopathy of uncertain significance and smoldering
14. Bianchi G, Kyle RA, Larson DR, et al. High levels of peripheral blood circulating plasma cells as a specific risk factor for progression of smoldering multiple
myeloma. Leukemia. 2013;27:680-685.
15. Gonsalves WI, Rajkumar SV, Dispenzieri A, et al. Quantification of circulating clonal plasma cells via multiparametric flow cytometry identifies patients with
smoldering multiple myeloma at high risk of progression. Leukemia. 2017;31:130-135.
16. Sanchez E, Li M, Kitto A, et al. Serum B-cell maturation antigen is elevated in multiple myeloma and correlates with disease status and survival. Br J Haematol.
2012;158:727-738.
17. Dispenzieri A, Soof CM, Rajkumar SV, et al. Serum BCMA levels to predict outcomes for patients with MGUS and smoldering multiple myeloma (SMM). J Clin
Oncol. 2019;37:15s (suppl; abstr 8020).
18. San Miguel J, Mateos M-V, Gonzalez V, et al. Updated risk stratification model for smoldering multiple myeloma (SMM) incorporating the revised IMWG diagnostic
criteria. J Clin Oncol. 2019;37:15s (suppl; abstr 8000).
19. Khan R, Dhodapkar M, Rosenthal A, et al. Four genes predict high risk of progression from smoldering to symptomatic multiple myeloma (SWOG S0120).
Haematologica. 2015;100:1214-1221.
20. López-Corral L, Mateos MV, Corchete LA, et al. Genomic analysis of high-risk smoldering multiple myeloma. Haematologica. 2012;97:1439-1443.
21. Madan S, Kyle RA, Greipp PR. Plasma cell labeling index in the evaluation of smoldering (asymptomatic) multiple myeloma. Mayo Clin Proc. 2010;85:300.
22. Mateos M-V, Hernández M-T, Giraldo P, et al. Lenalidomide plus dexamethasone for high-risk smoldering multiple myeloma. N Engl J Med. 2013;369:438-447.

23. Mateos M-V, Hernández M-T, Giraldo P, et al. Lenalidomide plus dexamethasone versus observation in patients with high-risk smouldering multiple myeloma
(QuiRedex): long-term follow-up of a randomised, controlled, phase 3 trial. Lancet Oncol. 2016;17:1127-1136.
24. Lonial S, Jacobus S, Fonseca R, et al. Randomized trial of lenalidomide versus observation in smoldering multiple myeloma. J Clin Oncol. Epub 2019 Oct 25.
25. Mateos MV. Curative strategy (GEM-CESAR) for high-risk smoldering myeloma: carfilzomib, lenalidomide and dexamethasone (KRd) as induction followed by
HDT-ASCT consolidation with KRd and maintenance with Rd. Presented at: 24th Congress of EHA. Amsterdam, Netherlands; 2019. Abstract S871.
26. Harrison S, Cavo M, De La Rubia J, et al. T(11;14) and high BCL2 expression are predictive biomarkers of response to venetoclax in combination with bortezomib
and dexamethasone in patients with relapsed/refractory multiple myeloma: biomarker analyses from the phase 3 Bellini Study. Blood. 2019;134:142.

Newly Diagnosed Myeloma in 2020

Philippe Moreau, MD1; Cyrille Touzeau, MD, PhD1; Ravi Vij, MD2; Scott R. Goldsmith, MD2; and Ashley E. Rosko, MD3
overview
Over the last few years, there has been great progress in the treatment of multiple myeloma (MM), with many
new agents and combinations having been approved and being now routinely incorporated into treatment
strategies for newly diagnosed patients. As a result, patients are experiencing benefits in terms of survival and
better tolerance. However, the multitude of treatment options also presents a challenge to select the best
options tailored to the specific patient situation. Frontline autologous stem cell transplantation (ASCT) is the
standard of care for fit patients younger than age 71 who are newly diagnosed with MM, and triplet com-
binations are the backbone of induction therapy before ASCT. Post-transplant consolidation and prolonged
lower-intensity maintenance are two strategies that have been used to deepen responses and delay pro-
gression. For older patients not eligible for ASCT, lenalidomide (len) is increasingly being used as part of
frontline therapy, and current approaches are now targeting combinations of anti-CD38 antibodies. Strategies
for selecting therapeutic regimens for older adults newly diagnosed with MM can be augmented with use of
predictive tools to better capture physiologic age and estimate treatment tolerance. Here we review a decade
of trials identifying clinical endpoints and toxicities relevant for the frontline treatment of younger patients and
older adults.
INDUCTION STRATEGIES FOR NEWLY DIAGNOSED drugs may be added.1,2 Thalidomide has been com-
MYELOMA PROCEEDING TO TRANSPLANT bined with VD in the bortezomib, thalidomide, and
According to international guidelines, outside clinical dexamethasone (VTD) regimen, which showed a su-
trials, frontline ASCT is the standard of care for fit perior response rate compared with VD alone in
patients younger than age 71 who are newly diagnosed a phase III trial.3 VTD has also shown better efficacy
with MM.1,2 This intensive strategy, which is associated compared with thalidomide and dexamethasone (TD),
with high response rates and prolonged progression- establishing this triplet combination as one of the
free survival (PFS) and overall survival (OS) consists of standard of care regimens before ASCT.4 Nevertheless,
different phases: induction therapy; high-dose che- VTD is associated with a higher rate of neurotoxicity
motherapy, usually entailing melphalan 200 mg/m2; because of the combination of two neurotoxic agents,
and ASCT, followed by maintenance treatment aimed bortezomib and thalidomide, and is not widely used in
at prolonging the duration of response. the United States. Cyclophosphamide has also been
added to VD in the bortezomib, cyclophosphamide,
Induction therapy, given before ASCT, has several goals: and dexamethasone (VCD) regimen. High response
(1) to achieve a fast control of the disease; (2) to induce rates can be achieved with this triplet combination,
high response rates (if possible, high rates of minimal with less neurotoxicity compared with VTD, making this
residual disease [MRD] negativity), with minimal toxicity, combination a preferred option in many countries
so that patients can proceed to ASCT; and (3) to allow worldwide. Nevertheless, a prospective comparison of
adequate stem cell harvesting. International guidelines four cycles of VTD versus four cycles of VCD before
recommend triplet induction regimens, combining a pro- ASCT clearly showed significant improvements in re-
Author affiliations
teasome inhibitor (PI) and an immunomodulatory drug sponse rates for the thalidomide-containing combi-
and support
information (if
(IMiD) with dexamethasone for four to six cycles before nation, suggesting a synergy between IMiDs and PIs.5
applicable) appear ASCT. Recently, the introduction of monoclonal anti- This has also been confirmed in a pair-matched
at the end of this bodies targeting CD38 has led to the use of quadruplet analysis conducted by the Italian myeloma study
article. combinations for induction, thereby deepening overall group.6
Accepted on March response rates (ORRs) before ASCT without adding rel-
31, 2020 and
evant toxicities. Here, we will review new data on induction Recently, thalidomide has been replaced by len in the
published at bortezomib, len, and dexamethasone (VRD) regimen.
ascopubs.org on
regimens before ASCT for transplant-eligible patients.
This triplet combination yielded high response rates
April 21, 2020: Triplet Induction Regimens
and reduced rates of neuropathy compared with the
10.1200/EDBK_ The international guidelines propose bortezomib- VTD regimen. No phase III data comparing VTD versus
280221 dexamethasone (VD) as a backbone to which several VRD are available, but indirect comparisons have been

Newly Diagnosed Myeloma
Carfilzomib has also been combined with thalidomide and

dexamethasone in the KTD regimen in a phase II trial
conducted by the HOVON group.13 Again, the efficacy was
• Triplet combinations are the backbone of in- satisfactory, and more than 90% of the patients could
duction therapy before ASCT; VRD is currently
proceed to ASCT.
considered the optimal regimen.
• Quadruplet combinations including CD38 an- In summary, triplet combinations are now the backbone of
tibodies are undergoing investigation in clinical induction therapy before ASCT. VRD is well established and
trials and may become a treatment of choice in at least as effective and less toxic compared with VTD. VCD,
the future. although less effective compared with VTD, but very active
• Post-transplant maintenance with len is the as well, remains widely used and represents a cost-effective
standard of care. combination. KRD is potentially the more active regimen,
but its availability remains limited because of cost and lack
• Combination therapies using IMiDs and PIs or
of approval in Europe.
CD38 antibodies are tested in clinical trials after
transplant.
Quadruplet Induction Regimens
• For older patients not eligible for ASCT, current
strategies are now targeting combinations of Two monoclonal antibodies targeting CD38, daratumumab
anti-CD38 antibodies. and isatuximab, were developed in the last decade. These
two agents have recently been added to triplet induction
regimens before ASCT. Isatuximab is currently being pro-
made. The Spanish PETHEMA group conducted two con- spectively evaluated with VRD in a phase III trial (VRD +/
secutive trials with six cycles of VTD before ASCT in one trial isatuximab before ASCT) conducted by the German GMMG
and six cycles of VRD before ASCT in the second trial.7,8 The group, but no data have yet been presented. More data are
integrated analysis of these two studies involving a large available with daratumumab. This agent has been com-
number of patients showed that VRD is associated with bined with VTD, and the prospective phase III CASSIOPEIA
higher response rates and longer 2-year PFS, with a re- trial conducted by the Intergroupe Francophone du Mye-
duced neurotoxicity rate compared with VTD.9 Therefore, lome (IFM) and Hemato-Oncologie voor Volwassenen
VRD is currently considered the optimal regimen before Nederland (HOVON) cooperative groups investigated the
ASCT in many countries, including the United States. outcome of transplant-eligible patients treated with VTD with
Nevertheless, because of the lack of phase III data com- or without daratumumab administered both before (in-
paring VRD to other triplet combinations prospectively, this duction, four cycles) and after (consolidation, two cycles)
regimen is not yet approved by the European authorities. single ASCT prepared by melphalan 200 mg/m2.14 A total of
Recently, carfilzomib, the second-in-class PI, has been 1,085 patients younger than age 66 were randomly
used instead of bortezomib in the carfilzomib, len, and assigned. The addition of daratumumab to VTD during
dexamethasone (KRD) regimen. This triplet combination induction induced significantly higher response rates and
has been evaluated in different phase II trials,10 and high VGPR and CR rates but higher MRD negativity rates (34.6%
response rates, including complete response (CR) or very in the VTD-daratumumab arm vs. 23.1% in the VTD arm;
good partial response (VGPR), were achieved after four odds ratio, 1.76; p , .0001). The high response rates
cycles before ASCT. In one phase III randomized study, achieved after induction, but also after consolidation and
KRD was compared with KCD (cyclophosphamide instead before maintenance, translated into a significant improve-
of len), and KRD showed superiority in terms of response ment in PFS in the daratumumab arm of the study. Based
and MRD negativity rates versus KCD.11,12 No prospective on these results, VTD plus daratumumab was recently
evaluation of KRD versus VRD is yet available. This study is approved by the U.S. Food and Drug Administration and
ongoing in the United States, but cross-trial comparisons European Medicines Agency, and this combination may be
seem to indicate that KRD yields higher rates of response considered as one of the standards of care before ASCT.
compared with VRD. The toxicity of both regimens is Because VRD is a superior regimen compared with VTD, the
manageable in routine practice, and only a few patients are quadruplet combination of VRD and daratumumab must be
not able to proceed to high-dose therapy and ASCT after evaluated. This analysis is ongoing in the prospective phase
induction using these two regimens. A careful assessment III PERSEUS study conducted by the European Myeloma
of cardiac and renal functions is necessary before using Network consortium, in which VRD with or without dar-
KRD. Because of the lack of published phase III data, KRD, atumumab both before (induction, four cycles) and after
although recommended by the most recent National ASCT (consolidation, two cycles) is being examined. A
Comprehensive Cancer Network guidelines along with other similar study, the randomized phase II GRIFFIN trial, which
regimens,2 is not yet approved by the regulatory authorities. was recently reported, included 104 patients treated with

Moreau et al
daratumumab and VRD (D-VRD) and 103 patients treated outcome of patients with high-risk disease was not different
with VRD alone.15 At the end of induction, 19.2% and when they were treated with VRD or VRD-daratumumab
71.7% of the patients had reached CR or better and at least followed by ASCT, but the comparison of outcome in the
VGPR in the D-VRD arm versus 13.4% and 56.7% in the high-risk versus standard-risk disease in the daratumumab
VRD arm, respectively. The higher rates of response and arm of the study was not reported.15 No data are yet
MRD negativity for D-VRD were also confirmed after con- available for KRD-daratumumab in high-risk patients.
solidation. These preliminary results, with a median follow-
Induction Summary
up of 13.5 months, must be confirmed with a longer follow-
up, including PFS and OS data. Nevertheless, D-VRD is Induction based on VRD followed by ASCT is the standard of
a promising induction combination, which can safely be care in the United States and in many countries in Europe.
proposed to patients eligible for ASCT. VTD or VCD are sound alternatives when VRD is not
available. Quadruplet combinations including CD38 anti-
KRD has also been combined with daratumumab in several
bodies are undergoing investigation in clinical trials and may
phase II trials. Early results from trials conducted by
become a treatment of choice in the future. Furthermore,
Landgren et al16 and Costa et al17 indicate that this qua-
the impressive MRD negativity rates achieved with KRD-
druplet combination might potentially be the most effective
daratumumab may challenge frontline ASCT in the future.
regimen before ASCT in terms of response and MRD
negativity rates. Carfilzomib was administered intravenously CONSOLIDATION AND MAINTENANCE THERAPY IN MM
weekly on days 1, 8, and 15 of 28-day cycles at the dose of ASCT with high-dose melphalan remains a cornerstone of
56 mg/m2. According to Costa et al, based on 67 patients, therapy for MM.18-21 Nevertheless, most patients will inevi-
the MRD negativity rate after four cycles of KRD- tably relapse. Post-transplant consolidation and prolonged
daratumumab was 40% at a detection level of 10 5 lower-intensity maintenance are two strategies that have been
by next-generation sequencing, and the VGPR rate or better used to deepen responses and delay progression.
was 90%.17 The weekly KRD-daratumumab regimen was
associated with low toxicity, and stem cell harvest was Historically, consolidation protocols used conventional
adequate. The rate of MRD negativity further improved after chemotherapy, such as Total Therapy 2, in which patients
single ASCT, up to 73%. Because of the short follow-up at received four cycles of dexamethasone, cisplatin, doxoru-
the time of presentation (median, 7.9 months), no PFS data bicin, cyclophosphamide, and etoposide after tandem
were presented. Landgren et al reported the results of eight ASCT.22,23 Recombinant interferon α was used as mainte-
weekly KRD-daratumumab cycles without ASCT in a small nance therapy initially after induction chemotherapy and
phase II study on 41 patients.16 With a short follow-up of 8.6 later after ASCT; although the benefits were marginal, it
months, the investigators showed that the MRD negativity served as a comparator for some early trials of thalidomide
rate was 75% for patients who completed all eight cycles, maintenance.22,24-27
including a VGPR rate or better of 92% and an ORR of Although these earlier practices sparked interest in post-
100%. No death on study was seen. At the time of the ASCT therapy, novel therapies including IMiDs, PIs, and,
report, no patient with MRD-negative disease had pro- more recently, monoclonal antibodies now comprise the
gressed. Despite the short follow-up, based on the high rate consolidation and maintenance arsenals. This section fo-
of MRD negativity and the 0% relapse rate achieved thus far cuses on the evidence from large and pivotal trials exam-
with this quadruplet combination, the authors of this small ining such therapies for post-ASCT consolidation and
phase II series now propose to systematically delay ASCT in maintenance, with additional emphases on MRD as an
patients with standard-risk disease. This provocative rec- endpoint, personalization for cytogenetic risk, and on-
ommendation requires validation in a phase III randomized going clinical trials.
trial comparing frontline versus delayed ASCT in patients
with MRD-negative disease after induction, a trial that is IMiD Monotherapy
being planned by the French cooperative group IFM. IMiDs have been considered ideal for maintenance therapy
An important issue is the outcome of patients with high-risk because of their known antimyeloma efficacy, oral ad-
disease (i.e., poor-risk cytogenetics) in the context of ministration, and modular dosing. Thalidomide was in-
quadruplet induction regimens. Very few data are available vestigated initially as IMiD-based maintenance therapy
to date. In the CASSIOPEIA trial, the PFS in patients with (Table 1). In some studies, it was found to provide superior
high-risk disease at diagnosis, defined by 17p deletion and PFS relative to observation or interferon α. However, toxicity,
t(4;14), treated with VTD-daratumumab and ASCT was often neuropathy, limited its adoption.
inferior to that of patients with standard-risk disease treated Recently, len has become the standard of care for main-
with the same regimen (P. Sonneveld, MD, written com- tenance therapy in the United States. Table 2 summarizes
munication, December 2020). In the GRIFFIN trial, the the phase III trials of len maintenance after ASCT. In a phase

III study by IFM (IFM 05-02), two 28-day cycles of len Myeloma Study Group randomly assigned patients to re-
consolidation were given to all patients who were then ceive 20 doses (six cycles) of bortezomib after ASCT or
randomly assigned to receive len maintenance or placebo observation only, and in the two studies published by
thereafter.34 The consolidation significantly improved the Einsele et al,40,41 consolidation consisted of four 35-day
ORR, and those who received len maintenance had su- cycles of bortezomib. Both studies demonstrated an im-
perior PFS relative to placebo, with comparable OS at provement in ORR with consolidation, which translated to
4 years. The CALGB 100104 trial randomly assigned pa- improvement in PFS, although neither saw an OS benefit.
tients to receive post-ASCT len maintenance or placebo, Peripheral neuropathy was significantly higher in the bor-
although in contrast, there was not a consolidation phase.20 tezomib arms. However, two of the studies excluded pa-
Unlike the IFM 05-02 study, there a significant difference in tients who had bortezomib-based induction, potentially
not only PFS but also OS. Although the underlying cause of limiting their applicability to modern treatment paradigms.
the OS discrepancy between the two trials remains unclear,
The oral PI ixazomib was compared with placebo as post-
the consolidation phase and an imbalance in risk stratifi-
ASCT maintenance in the phase III TOURMALINE-MM3
cation in IFM 05-02 have been suggested. Two other trials,
trial.44 Patients were randomly assigned to receive 2 years of
GIMEMA RV-MM-PI209 and Myeloma XI, compared len
either ixazomib or placebo. Ixazomib improved PFS relative
maintenance with observation.35,36 In both, len mainte-
to placebo and without a difference in SPMs. However, OS
nance provided superior PFS without an OS benefit.
was comparable, and the improvement in PFS was less
McCarthy et al38 conducted a meta-analysis that included impressive than that seen in the studies with len mainte-
IFM 05-02, CALGB 100104, and GIMEMA RV-MM-PI209 to nance. Therefore, until direct comparative data are avail-
address the lack of power to assess OS in the individual trials. able, it is generally used as single-agent maintenance for
This analysis of 1,208 patients demonstrated that len main- those intolerant to len.
tenance was associated with a 25% reduction in the risk of
death relative to placebo/observation (hazard ratio [HR], 0.75; PIs Compared With IMiDs
95% CI, 0.63–0.90; p = .001) and a 52% reduction in the risk In the HOVON-65/GMMG-HD4 trial, patients were randomly
of progression or death (HR, 0.48; 95% CI, 0.41–0.55). assigned to receive bortezomib-based induction followed
by ASCT and bortezomib post-ASCT maintenance versus
In all trials, len maintenance was associated with a signifi-
conventional chemotherapy induction followed by ASCT
cant incidence of neutropenia. More importantly, in the IFM
and thalidomide maintenance (Table 3).19,43 PFS was sig-
05-02, CALGB 100104, and Myeloma XI trials, len main-
nificantly better in the bortezomib arm, and a landmark
tenance was associated with a significantly higher incidence
analysis following ASCT suggested that bortezomib main-
of secondary primary malignancies (SPMs). In the meta-
tenance was associated with superior PFS compared with
analysis of McCarthy et al38 SPMs occurred in 6.1% and
thalidomide. Among patients with renal impairment or 17p
2.8% of patients in the len and control arms, respectively.
deletion, bortezomib contributed to improved OS relative
Ongoing trials will hopefully define an ideal duration that
to thalidomide, underscoring a role for PIs in patients with
balances risk of progression with SPMs, cost, and other
high-risk cytogenetics.
toxicities, as well as the use of sustained MRD negativity as
a signal for maintenance discontinuation.39 In a phase II trial by the Multiple Myeloma Research
Consortium (MMRC) following ixazomib, len, and dexa-
PI Monotherapy methasone (IRD) consolidation,45 patients were randomly
Three phase III studies have evaluated post-ASCT consol- assigned to receive ixazomib (94 patients) or len (95 pa-
idation bortezomib versus observation (Table 3). The Nordic tients) as maintenance therapy until progression or
TABLE 1. Phase III Trials Evaluating Thalidomide Maintenance Therapy

Regimens Control PFS Reference
PAM + thal 400 mg/day PAM or OBS 52% (3 years); p , .009 Attal et al28
Thal 200 mg/day + pred 50 mg QOD Pred 42% (3 years); p , .001 Spencer et al29
Thal 50 mg/day IFN 31% (5 years); p , .001 Van De Donk et al30
Thal 100 mg/day OBS 50% (30 months); p = .003 Morgan et al31
Thal 200 mg/day + dex 40 mg days 1-4 Dex 64% (2 years); p = .002 Maiolino et al32
Thal 200 mg/day + pred 50 QOD OBS 32% (4 years); p , .001 Stewart et al33
Abbreviations: PFS, progression-free survival; PAM, pamidronate; thal, thalidomide; OBS, observation; pred, prednisone; QOD, every other day; IFN,
interferon α; dex, dexamethasone

Moreau et al
TABLE 2. Phase III Trials Evaluating Lenalidomide Monotherapy Consolidation/Maintenance Therapy

Outcomes
Study Comparison Planned Duration PFS OS

34
Attal et al (IFM 05-02) Len vs. placebo (after two cycles of len Until PD (terminated Median PFS (41 vs. 23 4-year OS (73% vs.
consolidation) early for SPM) months; p , .001) 75%; p = NS)
McCarthy et al20,37 (CALGB Len vs. placebo Until PD Median PFS (46 vs. 27 3-year OS (88% vs.
100104) months; p , .001) 80%; p = .03)
Palumbo et al35 (GIMEMA MPR vs. tandem ASCT followed by len vs. Until PD Median PFS (41.9 vs. 21.6 3-year OS (88% vs.
RV-MM-PI209) OBS months; p , .001) 79.2%; p = NS)
Jackson et al36 (Myeloma Len vs. OBS (after randomization for Until PD Median PFS (39 vs. 20 3-year OS (78.6% vs.
XI) induction and intensification) months; p , .001) 75.8%; p = NS)
Abbreviations: PFS, progression-free survival; OS, overall survival; len, lenalidomide; PD, progressive disease; SPM, secondary primary malignancies; NS,
nonsignificant; MPR, melphalan/prednisone/lenalidomide; ASCT, autologous stem cell transplantation; OBS, observation.
intolerance. Ixazomib was tolerated well in comparison with occurred in the KRD12 arm among those with Revised
len, but conclusions regarding efficacy could not be made, International Staging System stage II or III disease.12
and after an interim analysis, the randomization was IRD consolidation has shown promise in two phase II studies
stopped because of futility. in improving ORR.45,51 The aforementioned MMRC trial
Combination Regimens With PIs and IMiDs demonstrated a significant improvement in the primary
endpoint of MRD negativity after four cycles of IRD
The GIMEMA Myeloma Network conducted a phase III
consolidation.
study to assess the efficacy of VTD versus TD as induction
before tandem ASCT; both cohorts received the same MRD
regimen as consolidation therapy (Table 4).4 PFS was The phase III IFM/DFCI 2009 trial demonstrated that ASCT
significantly better in the VTD arm, with notable improve- straddled with RVD induction and consolidation was as-
ment in those with high-risk cytogenetics. OS was similar. sociated with improved MRD negativity and PFS relative to
The phase III EMN02/HOVON-95 trial evaluated the impact eight cycles of RVD alone, before the start of len mainte-
of both intensification and post-ASCT consolidation.48 The nance.47 MRD status by multiparametric flow cytometry and
second randomization compared consolidation with len, next-generation sequencing were the strongest predictors of
bortezomib, and dexamethasone (RVD) versus no consol- PFS and OS, regardless of the treatment arm, implying
idation, although both cohorts subsequently received len therapy could potentially be adaptive to MRD status.52
maintenance. After adjustment for the first randomization A phase II MMRC study yielded high rates of next-
(intensification), RVD consolidation was associated with generation sequencing–MRD negativity after KRD in-
a PFS benefit with comparable OS at 5 years. These results duction, ASCT, and KRD consolidation. 49,53 With the
contrast those from the Blood and Marrow Transplant addition of attenuated KRD maintenance followed by
Clinical Trials Network (BMT CTN) 0702 STAMINA trial, len, more than 90% of those patients sustained MRD
a recent phase III trial conducted in the United States that negativity, which was associated with 94% 3-year PFS, with
compared three treatment arms, each followed by len 81% in those with high-risk disease.
maintenance: (1) ASCT plus consolidation with four cycles
The CASSIOPEIA trial evaluated the efficacy of VTD with or
of RVD; (2) tandem ASCT; and (3) single ASCT. Three-year
without daratumumab as both induction and post-ASCT
PFS, OS, and conversion rates to CR were statistically
consolidation, highlighting the importance of MRD analysis
comparable across all three.
to provide more granularity as to the effect of these intensive
A number of phase II studies have or are evaluating car- quadruplet regimens. The addition of daratumumab pro-
filzomib in combination with thalidomide and dexametha- vided better PFS, which correlated more with an improved
sone (KTD) or len and dexamethasone (KRD) used before MRD negativity after consolidation rather than conventional
and after ASCT (Table 4). In the phase III FORTE trial, response.14
patients were randomly assigned to receive KRD induction-
ASCT-KRD consolidation, 12 cycles of KRD without ASCT Consolidation and Maintenance for
(KRD12), or a regimen in which cyclophosphamide High-Risk Cytogenetics
replaced len (KCD-ASCT-KCD). 11 KRD-ASCT-KRD and IMiD-only maintenance has been associated with inferior
KRD12 had comparable response rates and were both outcomes in patients with high-risk cytogenetics.54 The use
superior the KCD arm. However, earlier progression of a PI in consolidation or maintenance may abrogate some

TABLE 3. Trials of Proteasome Inhibitor Monotherapy for Consolidation/Maintenance
Consolidation/ Before After
Study Maintenance Regimen Induction Regimen Comparison Arm Duration Consolidation Consolidation PFS OS
Bortezomib Consolidation
Uy et al42 BTZ (1.3 mg/m2 weekly Two cycles of BTZ None Six VGPR: 43% VGPR: 50% 3-year PFS: 32.3% 3-year OS:
on 35-day cycle) given before cycles 62.2%
ASCT
Mellqvist et al40 BTZ (1.3 mg/m2 twice BTZ naı̈ve OBS Six BTZ: VGPR: BTZ: VGPR: Median PFS 27 vs. 3-year OS:
weekly cycles 1 and cycles 39.7%; OBS 70.9%; OBS 20 months 80% vs.
2, weekly cycles VGPR: 39.1% VGPR: 57.4% 80%
3–6)
Einsele et al41 BTZ (1.6 mg/m2 weekly 51% BTZ based/ OBS Four BTZ: VGPR: BTZ: VGPR: Median PFS 33.6 vs. 5-year OS:
on 35-day cycle) 49% BTZ naı̈ve cycles 55%; OBS 62%; OBS 27.8 months 67% vs.
VGPR: 58% VGPR: 48% 68%
Bortezomib Maintenance
Sonneveld et al19,43 BTZ (1.3 mg/m2 every PAD Thal 50 mg 2 years NA NA Median PFS (from ASCT) 5-year OS:
(HOVON-65/GMMG- 2 weeks) (received VAD 31 vs. 26 months 61% vs.
HD4) induction) 55%
Ixazomib Maintenance
Dimopolous et al44 Ixazomib (3 → 4 mg PI/IMiD-based Placebo 2 years NA NA Median PFS 26.5 vs. NR
(TOURMALINE-MM3) weekly) 21.3 months
Vij et al45 Ixazomib (4 mg weekly) PI/IMiD-based; all Lenalidomide PD CR/sCR 47%; CR/sCR 63%; Estimated median PFS NR
received post- VGPR 81%; VGPR 86%; (from randomization):
ASCT IRD MRD –ve 22% MRD –ve 30% ixa, 28.2, len,
consolidation 36.6 months

Abbreviations: PFS, progression-free survival; OS, overall survival; BTZ, bortezomib; ASCT, autologous stem cell transplantation; VGPR, very good partial response; OBS, observation; PAD,

bortezomib/doxorubicin/dexamethasone; thal, thalidomide; VAD, vincristine/doxorubicin/dexamethasone; NA, not applicable; PI, proteasome inhibitor; IMiD, immunomodulatory drug; NR, not
reported; IRD, ixazomib/lenalidomide/dexamethasone; PD, progressive disease; CR, complete response; sCR, stringent CR; MRD –ve, minimal residual disease negative; ixa, ixazomib; len,
lenalidomide.

e149
e150
TABLE 4. Proteasome Inhibitor and Immunomodulatory Drug Combinations for Consolidation and Maintenance
Consolidation/
Maintenance Before
Study Regimen Induction Regimen Comparison Arm Duration Consolidation After Consolidation PFS OS
Bortezomib + IMiD Consolidation
Cavo et al4 VTD (160 pts) VTD 3 TD (161 pts) Two cycles CR: 48.7%; CR: 60.6%; 3-year PFS: 60% 3-year OS:
VGPR: 86.2% VGPR: 91.9% 90%
Nooka RVD (45 pts) RVD None 3 years sCR: 21%; VGPR: sCR: 51%; VGPR: Median PFS: 32 months 3-year OS:
et al46 89% 96% 93%
Attal et al47 RVD (350 pts) RVD No transplant (350 pts) Two cycles VGPR: 47% VGPR: 79% (after Median PFS: 50 months 4-year OS:
(before ASCT ASCT and 81%
consolidation)
Sonneveld RVD (455 pts) VCD No consolidation (437 Two cycles VGPR: 67% NR 5-year PFS: 48% 5-year OS:
et al48 pts) 77%

Stadtmauer RVD (254 pts) RVD (54%), VCD Tandem transplant Four cycles NR NR 38-month PFS: 57.8% 38-month
et al21 (14%) (247 pts); no OS: 85.4%
consolidation (257
pts)
Carfilzomib + IMiD Consolidation
Wester KTD (91 pts) KTD None Four cycles CR: 18%; VGPR: CR: 63%; VGPR: Median PFS: 58 months Median OS:
Moreau et al
et al13 65% 86% 83 months

Gay et al11 KRD (158 pts) KRD KRD for 12 cycles Four cycles CR: 47%; VGPR: CR: 62%; VGPR: NR NR
without ASCT (157 81% 86%
pts), KCD (159 pts)
Zimmerman KRD (71 pts) KRD (extended None Four cycles CR: 27%; CR: 86%; 3-year PFS: 86% Three-year
et al49 consolidation and VGPR: 90% VGPR: 94% OS: 95%
maintenance)
Ixazomib + IMiD Consolidation
Vij et al50 IRD (236 pts) PI/IMiD combination None Four cycles CR/sCR 47%; CR/sCR 63%; Estimated median PFS NR

(83%) VGPR 81%; VGPR 86%; (from randomization): ixa,
MRD –ve 22% MRD –ve 30% 28.2, len 36.6 months

Roussel IRD followed by IR IRD (2 weekly None IRD, two CR/sCR 8%; CR/sCR 30.2%; NR NR
et al51 (43 pts) ixazomib) cycles; IR, VGPR 60.5% VGPR 58.1%
six cycles
Abbreviations: PFS, progression-free survival; OS, overall survival; IMiD, immunomodulatory drug; VTD, bortezomib/thalidomide/dexamethasone; TD, thalidomide/dexamethasone; CR, complete
response, VGPR, very good partial response; RVD, lenalidomide/bortezomib/dexamethasone; sCR, stringent CR; ASCT, autologous stem cell transplantation; VCD, bortezomib/cyclophosphamide/
dexamethasone; NR, not reported; KTD, carfilzomib/thalidomide/dexamethasone; KRD, carfilzomib/lenalidomide/dexamethasone; IRD, ixazomib/lenalidomide/dexamethasone; PI, proteasome
inhibitor; MRD –ve, minimal residual disease negative; ixa, ixazomib; len, lenalidomide; IR, ixazomib/lenalidomide.
of the risk. In the aforementioned GIMEMA trial, VTD induction therapy. Here we review a decade of random-
consolidation eliminated the negative impact of t(4;14) ized controlled trials (Table 6) identifying toxicities and
relative to those receiving TD.4 In the HOVON-65/GMMG- clinical endpoints relevant for older adults and identifying
HD4 trial, bortezomib maintenance profoundly improved tools to personalize approaches to older adults with NDMM.
PFS and OS compared with thalidomide among patients The original PETHEMA56 strategy evaluated bortezomib,
with del(17p).19,43,55 Nooka et al46 evaluated prolonged RVD melphalan, prednisone (VMP) versus VTD, followed by
consolidation/maintenance in high-risk myeloma, admin- maintenance therapy (bortezomib-prednisone [VP] vs. VT).
istering up to 3 years of post-ASCT therapy and achieving This study concluded that triplet therapy followed by
notably high response rates and PFS compared with prior maintenance therapy was well tolerated with favorable re-
monotherapy trials. sponse profiles, with ORRs of 80% and greater among older
Ongoing Clinical Trials adults in their eighth decade of life. Quadruplet therapy,
followed by maintenance therapy, was first introduced with
The next wave of clinical trials builds on these novel con-
the GIMEMA 0305 randomized controlled trial,57 with im-
cepts, tailoring therapy to inherent disease risk, or altering
proved PFS using bortezomib, melphalan, prednisone,
therapy based on MRD status measured at multiple junc-
thalidomide (VMPT-VT) over VMP and with an updated
tures, such as the SWOG DRAMMATIC study (Table 5; Fig.
analysis demonstrating improved 5-year OS with VMPT-VT
1). Many of these trials incorporate monoclonal antibodies
(61%) versus VMP (51%) (HR, 0.70; p = .01).58 However,
in quadruplet induction regimens that are also used as post-
nonhematologic toxicities were more prevalent, specifically
ASCT consolidation, sometimes in an MRD-adaptive fash-
cardiac adverse events and neuropathy (both 11%). Ex-
ion. The use of such intensive upfront therapy before and
ploring cyclophosphamide as induction therapy is alluring;
after ASCT, guided by MRD analytics, could yield person-
given the ease of dosing and perceived tolerance, the MRC-
alized, curative approaches to MM.
IX explored cyclophosphamide, thalidomide, and dexa-
THE NEXT DECADE OF THERAPY FOR NEWLY DIAGNOSED MM methasone (CTD) versus melphalan and prednisone (MP),
Therapy for older adults with newly diagnosed MM (NDMM) with improved ORR using cyclophosphamide; both treat-
has evolved over the past 10 years. Chronologically, over the ment arms had similar but low PFS and OS.59 The strategy
decade, randomized controlled trials focused on triplet then shifted to explore continuous therapy, in the FIRST
therapy over doublet therapy, evaluating the roles of novel trial, which is the largest study of older adults with NDMM,
therapy using IMiDs (len) and PIs (bortezomib) in various comparing len-dexamethasone continuously (RD) versus
combinations in contrast to a traditional melphalan back- 18 cycles (RD18) versus melphalan-prednisone-thalidomide
bone. Nearly all the triplet-based strategies resulted in (MPT), where PFS proved to be superior with RD continu-
improved efficacy over doublet strategies. Favorable toxicity ously, and the OS was significantly improved in the two RD
profiles of novel therapy and subsequent incorporation of treatment arms.60 Older age was an adverse risk factor in this
maintenance strategies yielded the next influx of random- study, where PFS and OS were inferior compared with pa-
ized controlled trials. Studies then focused on variable in- tients younger than age 75 (PFS, 20.3 vs. 28.1 months; OS,
duction therapies and embedding maintenance strategies 52.3 vs. 60.9 months).67 In clinical practice, the availability
compared with nonmaintenance strategies, resulting in and use of IMiDs is different by region. The HOVON-87 trial62
improved PFS using the maintenance approach. We have evaluated IMiDs comparing nine cycles of melphalan,
now come full circle, evaluating the next generation of prednisone, thalidomide followed by maintenance thalido-
targeted agents, sequencing therapies such as carfilzomib mide (MPT-T) to melphalan, prednisone, len followed by len
and targeted immunotherapy in upfront NDMM strategies. maintenance (MPR-R). Efficacy was comparable; however,
Current strategies are now targeting combinations of anti- toxicity profiles were different, with more pronounced neu-
CD38 antibodies (daratumumab and isatuximab) resulting rotoxicity (thalidomide) and hematologic toxicity (len) among
in frontline comparisons of triple therapy versus quadruplet treatment arms.
therapy. The pace of drug discovery for myeloma has Importantly, although PFS is a common primary endpoint
resulted in queries regarding ideal drug therapy combina- for these studies, understanding survival advances,
tion, duration of therapy, depth and characterization of changes in health-related quality of life, and adverse events
response, immunotherapy maintenance strategies, and are equally as important. The Upfront study evaluated three
personalization of therapy based on individual character- different bortezomib-based combinations (VD, VTD, and
istics (e.g., frailty, toxicity) and disease characteristics (e.g., VMP followed by bortezomib maintenance), resulting in
high-risk mutations, stage). This has resulted in a paradox substantial response rates but lower than expected PFS and
for NDMM, where patients are deemed ineligible for in- OS.61 Toxicity-related discontinuations were common,
tensive therapy such as autologous hematopoietic stem cell particularly peripheral neuropathy, and health-related
transplantation but are deemed fit for quadruplet-based quality of life decreased initially during induction and then

Moreau et al
TABLE 5. Selected Ongoing Clinical Trials and Their Objectives

Study Phase NCT Identifier Objective/Endpoints Investigational Arm Control Arm
IFM 2018-04 II NCT03606577 Intensive, prolonged, quadruplet therapy before Dara-KRD induction, tandem None
and after tandem ASCT for patients with high-risk ASCT, dara-KRD consolidation,
cytogenetics len maintenance
Endpoints:
• Primary: % of pts receiving second ASCT
• Key secondary: MRD conversion rate, ORR,
PFS, OS
S1803 III NCT04071457 Dara + len vs. len as post-ASCT maintenance using Dara + len Len
DRAMMATIC MRD to direct therapy duration (Fig. 1)
Study
Endpoints:
• Primary: OS
• Key secondary: PFS, ORR, MRD negativity
AURIGA III NCT03901963 Dara + len vs. len maintenance in patients who are Dara + len Len
MRD positive after ASCT
Endpoints:
• Primary: NGS-MRD negativity conversion rate
• Key secondary: PFS, OS, CR/sCR rate
DSMM XVII III NCT03948035 Elo + KRD vs. KRD for induction followed by elo + Elo-KRD induction, Elo-len KRD induction, len
len vs. len maintenance, with special attention to maintenance maintenance
those with high-risk cytogenetics
Endpoints:
• Primary: induction, MRD –ve rate by MFC;
maintenance, PFS
• Secondary: ORR, OS, QoL
Abbreviations: ASCT, autologous stem cell transplantation; MRD, minimal residual disease: ORR, overall response rate; PFS, progression-free survival; OS,
overall survival; dara, daratumumab; KRD, carfilzomib/lenalidomide/dexamethasone; len, lenalidomide; NGS, next-generation sequencing; CR, complete
response; sCR, stringent CR; elo, elotuzumab; MRD –ve, minimal residual disease negative; MFC, multiparametric flow cytometry; QoL, quality of life.
FIGURE 1. Schema for

NCT04071457: S1803 DRAM-
MATIC Trial
Abbreviations: dara, daratumumab;
Len, lenalidomide; MRD, mini-
mal residual disease.

TABLE 6. Newly Diagnosed Myeloma: Transplant Ineligible
Median Median Age,
Patients; Follow-Up Range
Reference Total/Arms (months) (years) Regimen PFS OS Primary Endpoint/Comments
PETHEMA: 260 patients; 32 73 (68–77) VMP vs. VTP 31 months, all patients 3-year OS: 70% ORR: VTP 81%; VMP 80%
Mateos et al56 130/130 (64%–76%), all patients VTP more serious AEs (40 [31%]
73 (69–76) Maintenance :
2010 vs. 20 [15%], p = .01) and drug
VT or VP
discontinuations
GIMEMA 0305: 511 patients; 23.2 71 VMPT-VT vs. VMP 3-year PFS: VMPT-VT 56%; 5-year OS: VMPT-VT PFS
Palumbo et al57 254/257 VMP 41%; HR 0.67 61%; VMP 51%; VMPT-VT arm: more frequent
2010, (95% CI, 0.50–0.90; HR 0.70 grade 3/4 AEs: neutropenia
Palumbo et al58 p = .008) (p = .01) (38%), thrombocytopenia
2014 (22%), peripheral neuropathy
(11%), cardiologic events (11%)
MRC-IX: 859 patients; 44 73 MP vs. CTDa MP 12.4 mos; CTDa 13 mos; MP 30.6 mos; CTDa 33.2 ORR/PFS/OS
Morgan et al59 423/426 HR 0.82 mos; HR 0.89 ORR: MP 32.6%; CTDa 63.8%
2011 (95% CI, 0.70–0.96; (95% CI, 0.74–1.08; (p , .0001)
p = .01) p = .24) CTDa arm: higher rates of
thromboembolic events,
constipation, infection, and
neuropathy than MP arm
FIRST: 1,623 patients; 67 73 (40-92) RD vs. RD18 vs. MPT RD 25.5 mos; RD18 4-year OS: RD 59%; PFS
Benboubker et al60 535/541/ 20.7 mos; MPT 21.2 mos; RD18 56%; MPT 51% Grade 3/4 AEs more frequent
2014 547 HR 0.72 RD vs. MPT, with RD than with MPT
HR 0.70 RD vs. RD18; (70% vs. 78%)
p , .001 for both.
RD was superior to MPT for
all secondary efficacy
endpoints, including OS
UPFRONT: 502 patients; 42.7 74.5 (67-79) VD + maintenance V VD 14.7 mos; VTD 15.4 mos; VD 49.8 mos; PFS
Niesvizky et al63 168/167/ VMP 17.3 mos (p = NS) VTD 51.5 mos; Peripheral neuropathy
73 (66-77) VTD + maintenance V
2015 167 VMP 53.1 mos approached 50% all arms.
72 (68-77) VMP + maintenance V (p = NS) Early drug discontinuation
(29-38%)
QoL scores decreased during

induction and improved or
stabilized thereafter

HOVON-87: 668 patients; 36 72 (60-91) MPT-T vs. MPR-R MPT-T 20 mos; MPR-R 4-year OS: MPT-T 52% vs. PFS
Zweegman et al62 318/319 23 mos; HR 0.87 MPR-R 56% Grade 3/4 hematologic toxicity
2016 (95% CI, 0.72–1.04; with MPR-R vs. clinically
p = .12) significant neuropathy with
MPT-T
SWOG S0777: 525 patients; 54–56 43% . 65 VRD vs. RD VRD 43 mos; RD 30 mos; VRD 75 mos; RD 64 mos; PFS
Durie et al64 264/261 HR 0.712, (95% CI, HR 0.709 VRD (23%) and RD (10%)
2017 0.56–0.906; one-sided (95% CI, 0.524–0.959; discontinued induction

p = .0018) two-sided p = .025) treatment due to AE
e153
e154
TABLE 6. Newly Diagnosed Myeloma: Transplant Ineligible (Continued)
Median Median Age,
Patients; Follow-Up Range
Reference Total/Arms (months) (years) Regimen PFS OS Primary Endpoint/Comments
MAIA: 737 patients; 28 73 (50-90) D-RD continuous 30-month PFS: D-RD NA PFS
Facon et al66 368/369 70.6%; RD 55.6%; Median duration on continuous

74 (45-89) RD continuous
2019 HR 0.56 (95% CI, treatment 25.3 mos (D-Rd) and
0.43–0.73; p , .001) 21.3 mos (RD)
Pneumonia most common
serious AE more common in
dara exposed pts (13.2%)
HOVON-126/NMSG: 143 patients 23.4 73 (64-90) ITD vs. maintenance PFS-R: ITD-I 9.5 mos; OS-R at 18 mos, all PFS
Zweegman et al67 ixazomib vs. ITD-P 8.4 mos patients 96% Early mortality 8% . age 75
Moreau et al
2020 placebo (88% -99%) 55% randomly assigned to

maintenance therapy
ALCYONE: 706 patients; 40.1 71 (40-93) D-VMP-D D-VMP-D 36.4 mos; 3-year OS: D-VMP-D 78%; PFS
Mateos et al65 350/356 VMP 19.3 mos; VMP 67.9%; HR 0.60 Common daratumumab
71 (50-91) VMP
2020 HR 0.42 (95% CI 0.46–0.80; respiratory infections (bronchitis
(95% CI 0.34–0.51; p = .0003) (15%), viral upper respiratory
p , .0001) tract, cough (12%)
Diarrhea 10%

stabilized. The current landscape of VRD induction as International Myeloma Working Group used a simplified tool
frontline therapy for patients with NDMM is based on the based on age, comorbidities, activities of daily living, and
SWOG S0777 study, where VRD resulted in superior PFS instrumental activities of daily living for patients with NDMM
and OS in comparison with RD with acceptable toxicity enrolled in nontransplant frontline clinical trials.68 The In-
profiles.63 More recently, the ALCYONE study recently re- ternational Myeloma Working Group classified patients as
ported the OS advantages of D-VMP followed by dar- fit, intermediately fit, and frail, and scores were predictive of
atumumab maintenance (D-VMP-D) in comparison with death, progression, treatment discontinuation, and non-
VMP, with a 40% reduction in the risk of death with D- hematologic toxicities.68 The Revised Myeloma Comorbidity
VMP followed by daratumumab matinenance.66 Toxicities Index includes a scoring system to estimate myeloma
such as pulmonary infections were more common in dar- prognosis and therapy-related complications using the fol-
atumumab trials. Continuous treatment, as in MAIA trial, lowing variables: Karnofsky Performance status, lung dys-
evaluated D-RD compared with RD and demonstrated function, estimated glomerular function, frailty (Fried’s frailty
improved PFS and ORR with D-RD; with more lymphopenia phenotype), age, and cytogenetics, where frailty was prev-
and pneumonia in daratumumab-exposed patients.64 Pa- alent among patients with NDMM and mortality was highest
tients were on continuous therapy for approximately 2 years among patients with frailty.69 A simplified frailty scale70 ex-
(D-RD: median, 25.3 months; RD group: median, 21.3 plored the relationship of age, Charlson comorbidity index,
months). Although not all novel therapy combinations have and the Eastern Cooperative Oncology Group performance
yielded increasing benefits, the HOVON-126/NMSG eval- status to predict myeloma survival and was subsequently
uated ixazomib, thalidomide, dexamethasone induction validated in a second population of patients with NDMM not
followed by ixazomib versus placebo maintenance, resulting eligible for transplant.71 The Geriatric Assessment in He-
in similar PFS and OS after maintenance randomization.65 matology scoring system focuses on geriatric dimensions
Significantly, this study measured frailty phenotypes rec- (polypharmacy, gait speed, mood, activities of daily living,
ognizing that frail patients were less likely to reach ran- health status, nutrition, mental status, comorbidities), where
domization to maintenance therapy (45% of patients were increasing deficits were associated with survival in adults
not randomly assigned), and the lack of PFS from ran- older than age 65 who were newly diagnosed with hema-
domization between arms was independent of age, frailty, tologic malignancy.72,73 Myeloma-specific frailty tools are
cytogenetics, and response to induction therapy. a step forward to advance the care for older adults; however,
Strategies for selecting therapeutic regimens for older adults streamlining decision making and improving outcomes for
newly diagnosed with MM can be augmented with use of older adults require consensus of frailty indices, systematic
predictive tools to better capture physiologic age and esti- inclusion of frailty assessment metrics in clinical trials, and
mate treatment tolerance. Examples of myeloma-specific meaningful endpoints for clinical trials (e.g., health-related
protocols evaluating geriatric variables include the In- quality of life, functional status). This rigor will ensure that
ternational Myeloma Working Group frailty scale, the Re- frailty prediction tools are not simply an abstraction of
vised Myeloma Comorbidity Index, simplified Frailty Scale, chronologic age and will result in meaningful differences for
and the Geriatric Assessment in Hematology score. The older adults newly diagnosed with MM.

1
Department of Hematology, University Hospital, Nantes, France AND DATA AVAILABILITY STATEMENT
2
Washington University School of Medicine, St. Louis, MO Disclosures provided by the authors and data availability statement (if
3
Division of Hematology, The Ohio State University, Columbus, OH applicable) are available with this article at DOI https://doi.org/10.1200/
EDBK_280221.
Philippe Moreau, MD, Hematology Department, University Hospital Hotel-
Dieu, Place Ricordeau, 44093, Nantes, France; email: philippe.moreau@
chu-nantes.fr
REFERENCES
1. Moreau P, San Miguel J, Sonneveld P, et al; ESMO Guidelines Committee. Multiple myeloma: ESMO Clinical Practice Guidelines for diagnosis, treatment and
follow-up. Ann Oncol. 2017;28(suppl_4):iv52-iv61.
2. Kumar SK, Callander NS, Hillengass J, et al. NCCN Guidelines insights: multiple myeloma, version 1.2020. J Natl Compr Canc Netw. 2019;17:1154-1165.

Moreau et al
3. Moreau P, Avet-Loiseau H, Facon T, et al. Bortezomib plus dexamethasone versus reduced-dose bortezomib, thalidomide plus dexamethasone as induction
treatment before autologous stem cell transplantation in newly diagnosed multiple myeloma. Blood. 2011;118:5752-5758, quiz 5982.
4. Cavo M, Pantani L, Petrucci MT, et al; GIMEMA (Gruppo Italiano Malattie Ematologiche dell’Adulto) Italian Myeloma Network. Bortezomib-thalidomide-
dexamethasone is superior to thalidomide-dexamethasone as consolidation therapy after autologous hematopoietic stem cell transplantation in patients with
newly diagnosed multiple myeloma. Blood. 2012;120:9-19.
5. Moreau P, Hulin C, Macro M, et al. VTD is superior to VCD prior to intensive therapy in multiple myeloma: results of the prospective IFM2013-04 trial. Blood. 2016;
127:2569-2574.
6. Cavo M, Pantani L, Pezzi A, et al. Bortezomib-thalidomide-dexamethasone (VTD) is superior to bortezomib-cyclophosphamide-dexamethasone (VCD) as
induction therapy prior to autologous stem cell transplantation in multiple myeloma. Leukemia. 2015;29:2429-2431.
7. Rosiñol L, Oriol A, Teruel AI, et al; Programa para el Estudio y la Terapéutica de las Hemopatı́as Malignas/Grupo Español de Mieloma (PETHEMA/GEM) group.
Superiority of bortezomib, thalidomide, and dexamethasone (VTD) as induction pretransplantation therapy in multiple myeloma: a randomized phase 3
PETHEMA/GEM study. Blood. 2012;120:1589-1596.
8. Rosiñol L, Oriol A, Rios R, et al. Bortezomib, lenalidomide, and dexamethasone as induction therapy prior to autologous transplant in multiple myeloma. Blood.
2019;134:1337-1345.
9. Rosinol L, Hebraud B, Oriol A, et al. Integrated Analysis of Randomized Controlled Trials Evaluating Bortezomib + Lenalidomide + Dexamethasone or Bortezomib
+ Thalidomide + Dexamethasone Induction in Transplant-Eligible Newly Diagnosed Multiple Myeloma. Blood. 2018;132 (Supplement_1):3245.
10. Korde N, Roschewski M, Zingone A, et al. Treatment with carfilzomib-lenalidomide-dexamethasone with lenalidomide extension in patients with smoldering or
newly diagnosed multiple myeloma. JAMA Oncol. 2015;1:746-754.
11. Gay F, Cerrato C, Rota Scalabrini D, et al. Carfilzomib-lenalidomide-dexamethasone (KRd) induction-autologous transplant (ASCT)-Krd consolidation vs KRd 12
cycles vs carfilzomib-cyclophosphamide-dexamethasone (KCd) induction-ASCT-KCd consolidation: analysis of the randomized Forte trial in newly diagnosed
multiple myeloma (NDMM). Blood. 2018;132 (suppl 1):121.
12. Gay F, Cerrato C, Petrucci M, et al. Efficacy of carfilzomib lenalidomide dexamethasone (KRd) with or without transplantation in newly diagnosed myeloma
according to risk status: results from the FORTE trial. J Clin Oncol. 2019;37;15s (suppl; abstr 8002).
13. Wester R, van der Holt B, Asselbergs E, et al. Phase II study of carfilzomib, thalidomide, and low-dose dexamethasone as induction and consolidation in newly
diagnosed, transplant eligible patients with multiple myeloma; the Carthadex trial. Haematologica. 2019;104:2265-2273.
14. Moreau P, Attal M, Hulin C, et al. Bortezomib, thalidomide, and dexamethasone with or without daratumumab before and after autologous stem-cell trans-
plantation for newly diagnosed multiple myeloma (CASSIOPEIA): a randomised, open-label, phase 3 study. Lancet. 2019;394:29-38.
15. Voorhees P, Kaufman J, Laubach J, et al. Depth of response to daratumumab (DARA), lenalidomide, bortezomib, and dexamethasone (RVd) improves over time
in patients (pts) with transplant-eligible newly diagnosed multiple myeloma (NDMM): Griffin study update. Blood. 2019;134 (suppl 1):691.
16. Landgren O, Hultcrantz M, Lesokhin AM, et al. Weekly carfilzomib, lenalidomide, dexamethasone, and daratumumab combination therapy provides un-
precedented MRD negativity rates in newly diagnosed multiple myeloma: a clinical and correlative phase 2 study. Blood. 2019;134 (suppl 1):862.
17. Costa LJ, Chhabra S, Godby KN, et al. Daratumumab, carfilzomib, lenalidomide, and dexamethasone (Dara-KRd) induction, autologous transplantation and post-
transplant, response-adapted, measurable residual disease–based Dara-KRd consolidation in patients with newly diagnosed multiple myeloma. Blood. 2019;
134 (suppl 1):860.
18. Attal M, Harousseau JL. Standard therapy versus autologous transplantation in multiple myeloma. Hematol Oncol Clin North Am. 1997;11:133-146.
19. Goldschmidt H, Lokhorst HM, Mai EK, et al. Bortezomib before and after high-dose therapy in myeloma: long-term results from the phase III HOVON-65/GMMG-
HD4 trial. Leukemia. 2018;32:383-390.
20. McCarthy PL, Owzar K, Hofmeister CC, et al. Lenalidomide after stem-cell transplantation for multiple myeloma. N Engl J Med. 2012;366:1770-1781.
21. Stadtmauer EA, Pasquini MC, Blackwell B, et al. Autologous transplantation, consolidation, and maintenance therapy in multiple myeloma: results of the BMT
CTN 0702 trial. J Clin Oncol. 2019;37:589-597.
22. Barlogie B, Jagannath S, Desikan KR, et al. Total therapy with tandem transplants for newly diagnosed multiple myeloma. Blood. 1999;93:55-65.
23. Barlogie B, Tricot G, Rasmussen E, et al. Total therapy 2 without thalidomide in comparison with total therapy 1: role of intensified induction and post-
transplantation consolidation therapies. Blood. 2006;107:2633-2638.
24. Ludwig H, Cohen AM, Polliack A, et al. Interferon-alpha for induction and maintenance in multiple myeloma: results of two multicenter randomized trials and
summary of other studies. Ann Oncol. 1995;6:467-476.
25. Salmon SE, Crowley JJ, Grogan TM, et al. Combination chemotherapy, glucocorticoids, and interferon alfa in the treatment of multiple myeloma: a Southwest
Oncology Group study. J Clin Oncol. 1994;12:2405-2414.
26. Mandelli F, Avvisati G, Amadori S, et al. Maintenance treatment with recombinant interferon alfa-2b in patients with multiple myeloma responding to conventional
induction chemotherapy. N Engl J Med. 1990;322:1430-1434.
27. Lokhorst HM, van der Holt B, Zweegman S, et al; Dutch-Belgian Hemato-Oncology Group (HOVON). A randomized phase 3 study on the effect of thalidomide
combined with adriamycin, dexamethasone, and high-dose melphalan, followed by thalidomide maintenance in patients with multiple myeloma. Blood. 2010;
115:1113-1120.
28. Attal M, Harousseau J-L, Leyvraz S, et al; Inter-Groupe Francophone du Myélome (IFM). Maintenance therapy with thalidomide improves survival in patients with
multiple myeloma. Blood. 2006;108:3289-3294.

29. Spencer A, Prince HM, Roberts AW, et al. Consolidation therapy with low-dose thalidomide and prednisolone prolongs the survival of multiple myeloma patients
undergoing a single autologous stem-cell transplantation procedure. J Clin Oncol. 2009;27:1788-1793.
30. van de Donk NW, van der Holt B, Minnema MC, et al. Thalidomide before and after autologous stem cell transplantation in recently diagnosed multiple myeloma
(HOVON-50): long-term results from the phase 3, randomised controlled trial. Lancet Haematol. 2018;5:e479-e492.
31. Morgan GJ, Gregory WM, Davies FE, et al; National Cancer Research Institute Haematological Oncology Clinical Studies Group. The role of maintenance
thalidomide therapy in multiple myeloma: MRC Myeloma IX results and meta-analysis. Blood. 2012;119:7-15.
32. Maiolino A, Hungria VTM, Garnica M, et al; Brazilian Multiple Myeloma Study Group (BMMSG/GEMOH). Thalidomide plus dexamethasone as a maintenance
therapy after autologous hematopoietic stem cell transplantation improves progression-free survival in multiple myeloma. Am J Hematol. 2012;87:948-952.
33. Stewart AK, Trudel S, Bahlis NJ, et al. A randomized phase 3 trial of thalidomide and prednisone as maintenance therapy after ASCT in patients with MM with
a quality-of-life assessment: the National Cancer Institute of Canada Clinicals Trials Group Myeloma 10 Trial. Blood. 2013;121:1517-1523.
34. Attal M, Lauwers-Cances V, Marit G, et al; IFM Investigators. Lenalidomide maintenance after stem-cell transplantation for multiple myeloma. N Engl J Med. 2012;
366:1782-1791.
35. Palumbo A, Cavallo F, Gay F, et al. Autologous transplantation and maintenance therapy in multiple myeloma. N Engl J Med. 2014;371:895-905.
36. Jackson GH, Davies FE, Pawlyn C, et al; UK NCRI Haemato-oncology Clinical Studies Group. Lenalidomide maintenance versus observation for patients with
newly diagnosed multiple myeloma (Myeloma XI): a multicentre, open-label, randomised, phase 3 trial. Lancet Oncol. 2019;20:57-73.
37. Holstein SA, Jung S-H, Richardson PG, et al. Updated analysis of CALGB (Alliance) 100104 assessing lenalidomide versus placebo maintenance after single
autologous stem-cell transplantation for multiple myeloma: a randomised, double-blind, phase 3 trial. Lancet Haematol. 2017;4:e431-e442.
38. McCarthy PL, Holstein SA, Petrucci MT, et al. Lenalidomide maintenance after autologous stem-cell transplantation in newly diagnosed multiple myeloma:
a meta-analysis. J Clin Oncol. 2017;35:3279-3289.
39. Roussel M, Lauwers-Cances V, Robillard N, et al. Front-line transplantation program with lenalidomide, bortezomib, and dexamethasone combination as
induction and consolidation followed by lenalidomide maintenance in patients with multiple myeloma: a phase II study by the Intergroupe Francophone du
Myélome. J Clin Oncol. 2014;32:2712-2717.
40. Mellqvist U-H, Gimsing P, Hjertner O, et al; Nordic Myeloma Study Group. Bortezomib consolidation after autologous stem cell transplantation in multiple
myeloma: a Nordic Myeloma Study Group randomized phase 3 trial. Blood. 2013;121:4647-4654.
41. Einsele H, Knop S, Vogel M, et al. Response-adapted consolidation with bortezomib after ASCT improves progression-free survival in newly diagnosed multiple
myeloma. Leukemia. 2017;31:1463-1466.
42. Uy GL, Goyal SD, Fisher NM, et al. Bortezomib administered pre-auto-SCT and as maintenance therapy post transplant for multiple myeloma: a single institution
phase II study. Bone Marrow Transplant. 2009;43:793-800.
43. Sonneveld P, Schmidt-Wolf IGH, van der Holt B, et al. Bortezomib induction and maintenance treatment in patients with newly diagnosed multiple myeloma:
results of the randomized phase III HOVON-65/ GMMG-HD4 trial. J Clin Oncol. 2012;30:2946-2955.
44. Dimopoulos MA, Gay F, Schjesvold F, et al; TOURMALINE-MM3 study group. Oral ixazomib maintenance following autologous stem cell transplantation
(TOURMALINE-MM3): a double-blind, randomised, placebo-controlled phase 3 trial. Lancet. 2019;393:253-264.
45. Vij R, Martin TG III, Nathwani N, et al. Ixazomib or lenalidomide maintenance following autologous stem cell transplantation and ixazomib, lenalidomide, and
dexamethasone (IRD) consolidation in patients with newly diagnosed multiple myeloma: results from a large multi-center randomized phase II trial. Blood. 2019;
134 (suppl 1):602.
46. Nooka AK, Kaufman JL, Muppidi S, et al. Consolidation and maintenance therapy with lenalidomide, bortezomib and dexamethasone (RVD) in high-risk myeloma
patients. Leukemia. 2014;28:690-693.
47. Attal M, Lauwers-Cances V, Hulin C, et al; IFM 2009 Study. Lenalidomide, bortezomib, and dexamethasone with transplantation for myeloma. N Engl J Med.
2017;376:1311-1320.
48. Sonneveld P, Beksac M, van der Holt B, et al. Consolidation followed by maintenance therapy versus maintenance alone in newly diagnosed, transplant eligible
patients with multiple myeloma (mm): a randomized phase 3 study of the European Myeloma Network (EMN02/HO95 MM Trial). Blood. 2016;128:242.
49. Zimmerman T, Raje NS, Vij R, et al. Final results of a phase 2 trial of extended treatment (tx) with carfilzomib (CFZ), lenalidomide (LEN), and dexamethasone
(KRd) plus autologous stem cell transplantation (ASCT) in newly diagnosed multiple myeloma (NDMM). Blood. 2016;128:675.
50. Vij R, Nathwani N, Martin TG, et al. Ixazomib-lenalidomide-dexamethasone (IRd) consolidation following autologous stem cell transplantation in patients with
newly diagnosed multiple myeloma: a large multi-center phase II trial. Blood. 2018;132 (suppl 1):123.
51. Roussel M, Hebraud B, Hulin C, et al. Twice weekly induction with ixazomib-lenalidomide-dexamethasone (IRd) combination followed by extended IRD
consolidation and lenalidomide maintenance in transplant eligible patients with newly diagnosed multiple myeloma (NDMM): a phase 2 study from the
Intergroupe Francophone Du Myelome (IFM 2014-03). Blood. 2019; 134 (suppl 1):3159.
52. Perrot A, Lauwers-Cances V, Corre J, et al. Minimal residual disease negativity using deep sequencing is a major prognostic factor in multiple myeloma. Blood.
2018;132:2456-2464.
53. Jakubowiak AJ, Raje N, Vij R, et al. High rate of sustained minimal residual disease negativity predicts prolonged survival for the overall patient population in the
phase 2 KRd plus autologous stem cell transplantation MMRC trial. Blood. 2017;130(suppl 1):4533.
54. Nooka AK, Lonial S. New Targets and New Agents in High-Risk Multiple Myeloma. Am Soc Clin Oncol Educ Book. 2016;35:e431-e441.
55. Sonneveld P, Avet-Loiseau H, Lonial S, et al. Treatment of multiple myeloma with high-risk cytogenetics: a consensus of the International Myeloma Working
Group. Blood. 2016;127:2955-2962.

Moreau et al
56. Mateos MV, Oriol A, Martı́nez-López J, et al. Bortezomib, melphalan, and prednisone versus bortezomib, thalidomide, and prednisone as induction therapy
followed by maintenance treatment with bortezomib and thalidomide versus bortezomib and prednisone in elderly patients with untreated multiple myeloma:
a randomised trial. Lancet Oncol. 2010;11:934-941.
57. Palumbo A, Bringhen S, Rossi D, et al. Bortezomib-melphalan-prednisone-thalidomide followed by maintenance with bortezomib-thalidomide compared with
bortezomib-melphalan-prednisone for initial treatment of multiple myeloma: a randomized controlled trial. J Clin Oncol. 2010;28:5101-5109.
58. Palumbo A, Bringhen S, Larocca A, et al. Bortezomib-melphalan-prednisone-thalidomide followed by maintenance with bortezomib-thalidomide compared with
bortezomib-melphalan-prednisone for initial treatment of multiple myeloma: updated follow-up and improved survival. J Clin Oncol. 2014;32:634-640.
59. Morgan GJ, Davies FE, Gregory WM, et al; NCRI Haematological Oncology Study Group. Cyclophosphamide, thalidomide, and dexamethasone (CTD) as initial
therapy for patients with multiple myeloma unsuitable for autologous transplantation. Blood. 2011;118:1231-1238.
60. Benboubker L, Dimopoulos MA, Dispenzieri A, et al; FIRST Trial Team. Lenalidomide and dexamethasone in transplant-ineligible patients with myeloma. N Engl
J Med. 2014;371:906-917.
61. Niesvizky R, Flinn IW, Rifkin R, et al. Community-based phase IIIB trial of three UPFRONT Bortezomib-based myeloma regimens. J Clin Oncol. 2015;
33:3921-3929.
62. Zweegman S, van der Holt B, Mellqvist UH, et al. Melphalan, prednisone, and lenalidomide versus melphalan, prednisone, and thalidomide in untreated multiple
myeloma. Blood. 2016;127:1109-1116.
63. Durie BGM, Hoering A, Abidi MH, et al. Bortezomib with lenalidomide and dexamethasone versus lenalidomide and dexamethasone alone in patients with newly
diagnosed myeloma without intent for immediate autologous stem-cell transplant (SWOG S0777): a randomised, open-label, phase 3 trial. Lancet. 2017;
389:519-527.
64. Facon T, Kumar S, Plesner T, et al; MAIA Trial Investigators. Daratumumab plus lenalidomide and dexamethasone for untreated myeloma. N Engl J Med. 2019;
380:2104-2115.
65. Zweegman S, Stege CAM, Haukas E, et al. Ixazomib-thalidomide-low dose dexamethasone induction followed by maintenance therapy with ixazomib or placebo
in newly diagnosed multiple myeloma patients not eligible for autologous stem cell transplantation; results from the randomized phase II HOVON-126/NMSG 21.
13 trial. Haematologica. Epub 2020 Feb 13.
66. Mateos MV, Cavo M, Blade J, et al. Overall survival with daratumumab, bortezomib, melphalan, and prednisone in newly diagnosed multiple myeloma
(ALCYONE): a randomised, open-label, phase 3 trial. Lancet. 2020;395:132-141.
67. Hulin C, Belch A, Shustik C, et al. Updated outcomes and impact of age with lenalidomide and low-dose dexamethasone or melphalan, prednisone, and
thalidomide in the randomized, phase III FIRST trial. J Clin Oncol. 2016;34:3609-3617.
68. Palumbo A, Bringhen S, Mateos M-V, et al. Geriatric assessment predicts survival and toxicities in elderly myeloma patients: an International Myeloma Working
Group report. Blood. 2015;125:2068-74.
69. Engelhardt M, Domm AS, Dold SM, et al. A concise revised myeloma comorbidity index as a valid prognostic instrument in a large cohort of 801 multiple myeloma
patients. Haematologica. 2017;102:910-921.
70. Facon T, Dimopoulos MA, Meuleman N, et al. A simplified frailty scale predicts outcomes in transplant-ineligible patients with newly diagnosed multiple myeloma
treated in the FIRST (MM-020) trial. Leukemia. 2020;34:224-233.
71. Stege CAM, van der Holt B, Dinmohamed AG, et al. Validation of the FIRST simplified frailty scale using the ECOG performance status instead of patient-reported
activities. Leukemia. Epub 2020 Jan 22.
72. dela Rubia BJG, Hernández Rivas J, Valcárcel D, et al. GAH scale is a simple, comprehensive assessment tool in older patients with hematological malignancies
that shows mortality prediction capacities. Clin Lymphoma Myeloma Leuk. 2015;15:e99.
73. Bonanad S, De la Rubia J, Gironella M, et al; GAH Group. Development and psychometric validation of a brief comprehensive health status assessment scale in
older patients with hematological malignancies: the GAH scale. J Geriatr Oncol. 2015;6:353-361.

LUNG CANCER

LUNG CANCER
Checkpoint Blockade in Lung Cancer With Driver

Mutation: Choose the Road Wisely
Antonio Calles, MD, MSc1; Jonathan W. Riess, MD, MS2; and Julie R. Brahmer, MD, FASCO3
overview
Immune checkpoint blockade with PD-(L)1 antibodies has revolutionized the treatment of advanced non–
small cell lung cancer (NSCLC). Similarly, the identification and targeting of oncogene drivers in metastatic
NSCLC has dramatically improved patient outcomes with an expanding list of potentially actionable alter-
ations and targeted therapies. Many of these molecular aberrations are more common in patients with little or
no smoking history and adenocarcinoma histology. Certain molecular subsets of NSCLC, though gaining
greatly from targeted therapy approaches, may derive less benefit from immune checkpoint blockade. The
optimal identification, targeting, and sequencing of targeted therapies, immunotherapy, and chemotherapy
are essential to continue to improve patient outcomes in advanced NSCLC. Herein, we review the role of
immunotherapy in locally advanced and metastatic disease for patients with actionable driver alterations.
Never-smoking patients have a high probability of having lung cancer that harbors one of these molecular
aberrations that can be matched to a tyrosine kinase inhibitor with greatly improved clinical outcomes. Some
of these patients with driver mutations may derive less benefit from immune checkpoint inhibitor approaches
(either alone or combined with chemotherapy), especially compared with smoking-associated NSCLC. Given
that PD-1 blockade alone or with platinum-based chemotherapy is the de facto first-line therapy (depending
on level of PD-L1 expression) for nontargetable metastatic NSCLC, we also review treatment in never-smoking
patients for whom molecular testing results are pending and the likelihood of identifying a driver mutation
is high.
IMMUNOTHERAPY IN LOCALLY ADVANCED NSCLC WITH ALK translocations were excluded, but genetic
ACTIONABLE DRIVER MUTATIONS testing was not mandatory. However, in the up-
Although guidelines do not routinely endorse next- coming Lung Cancer Mutation Consortium 4 study,
generation sequencing of early-stage NSCLC, there patients with driver mutations or fusions will be
are several adjuvant studies of tyrosine kinase in- treated on a neoadjuvant targeted therapy arm for
hibitors (TKIs) specifically for patients with early-stage those specific abnormalities and not neoadjuvant
NSCLC whose disease harbors driver mutations such immunotherapy.
as epidermal growth factor receptor (EGFR) mutations Currently, the PACIFIC study is the only study for locally
or ALK translocations, such as the ALCHEMIST advanced disease that uses durvalumab, a PD-L1
trial (NCT02194738). However, the studies for neo- antibody, and that allowed patients with tumors har-
adjuvant immunotherapy or adjuvant immunotherapy boring driver mutations to enroll, although an ongoing
have generally excluded patients whose tumors single-institution study combining nivolumab and ipi-
harbor these abnormalities, such as the ANVIL trial of limumab concurrently with radiation does include
adjuvant nivolumab (NCT02595944). This exclusion is patients with EGFR driver mutations or ALK trans-
Author affiliations based on data from early studies in the metastatic locations (NCT04013542).5,6 This study randomly
and support
setting that have largely demonstrated that single- assigned patients with locally advanced NSCLC
information (if
applicable) appear
agent checkpoint inhibitors do not benefit patients after chemotherapy and radiation whose disease
at the end of this who have tumors harboring these abnormalities and was stable or responding to receive either placebo or
article. have stage IV disease.1-4 In phase III trials of neo- durvalumab every 2 weeks for 1 year. This study is the
Accepted on adjuvant chemotherapy combined with PD-1 or PD- first to show a survival advantage to consolidation
February 27, 2020 L1 inhibitors, only IMpower030 (NCT03456063) immunotherapy after concurrent chemotherapy and
and published at
excludes patients with EGFR or ALK abnormalities but radiation for an unselected patient population. The
ascopubs.org on May
18, 2020: DOI https://
not in KEYNOTE 671 (NCT03425642), in which tumor patients with EGFR mutation–positive cancers made
doi.org/10.1200/ genotyping is not mandatory for inclusion. In CheckMate up approximately 6% of patients on each arm. The
EDBK_280795 816 (NCT02998528), patients with EGFR mutations or patients with unknown EGFR mutation status made up

IO in Lung Cancer With Driver Mutations
who received checkpoint inhibitors either just before EGFR

TKI use or in combination on clinical trials. Investigators
reported increased pulmonary toxicities in patients treated
• The use of immunotherapy in locally advanced with osimertinib or other EGFR TKIs after receiving check-
NSCLC with actionable mutations remains
point inhibitors (Table 1).13,14 Phase I studies of durvalumab
controversial.
and osimertinib were stopped early because of an elevated
• With some exceptions, immune checkpoint incidence of pneumonitis.12 Phase I studies of durvalumab
inhibitors given alone have low efficacy in the and gefitinib also reported increased toxicities from the
treatment of patients with metastatic NSCLC
combination, specifically hepatotoxicity (liver function test
with oncogenic-driven tumors.
abnormalities).15 Although other studies of an earlier-
• The use of combination regimens of immune generation EGFR TKI, erlotinib, did not report such
checkpoint inhibitors with tyrosine kinase in- concerns of overlapping toxicities.16
hibitors has led to excess toxicities with no
additional efficacy, whereas immunotherapy in Based on what we know about the limited efficacy of single-
combination with chemotherapy, and especially agent PD-1 or PD-L1 inhibition in the metastatic setting and
with antiangiogenic drugs, looks promising for non–statistically significant increase in PFS for durvalumab
previously treated patients with targeted in patients with EGFR-mutated NSCLC, it is difficult to
therapies. recommend routine use of durvalumab after definitive
• Waiting for molecular testing results before treatment with concurrent chemotherapy in these patients
initiating treatment in never-smoking patients with locally advanced EGFR-mutant NSCLC. However,
with advanced NSCLC is recommended given based on the survival advantage seen in the overall pop-
the high probability of detecting an actionable ulation on the PACIFIC trial with the use of consolidation
molecular alteration (such as an epidermal durvalumab, an extended discussion is warranted with
growth factor receptor activating mutation, ALK patients whose lung cancer harbors an EGFR mutation if
or ROS1 fusion) that can be matched to a ty-
durvalumab use is planned.
rosine kinase inhibitor with robust clinical
activity. IMMUNO-ONCOLOGY IN METASTATIC NSCLC WITH
• Initiating immunotherapy shortly before a tyro- ACTIONABLE DRIVER MUTATIONS
sine kinase inhibitor, if a targetable mutation is Current systemic treatment of stage IV NSCLC for patients
eventually detected (such as EGFR, ALK, or with good performance status is categorized according to
ROS1) and treatment is switched to a tyrosine the presence or absence of oncogenic driver mutations.
kinase inhibitor, may increase immune related Targeted therapy is the recommended treatment of pa-
toxicity.
tients with driver mutations and subsequent selective TKIs
beyond progression. For those patients without driver
mutations, immune checkpoint inhibitors (ICIs) either
approximately 24% to 28% of patients on each arm of the alone or in combination with chemotherapy, often se-
trial. Although this was a small subset of patients with lected by PD-L1 expression, are the recommended initial
EGFR mutation–positive disease, the investigators retreatment if there are no formal contraindications to
ported a trend toward improved progression-free survival immunotherapy.
(PFS) in this group, with a hazard ratio (HR) of 0.76 (95% CI,
0.35–1.64), which was not statistically significant. Because Whether immunotherapy has a role in the treatment of
of the small numbers of patients, no observations on survival patients with lung cancer and oncogenic driver mutations is
with this therapy can be made. In a post hoc analysis of the controversial. Most of the evidence currently available from
PACIFIC trial, patients with negative (0%) PD-L1 expression the use of immunotherapy in patients with NSCLC with
did not have a survival benefit from durvalumab consoli- actionable driver mutations comes from either subgroup
dation. Extrapolating from the metastatic setting, in the analysis of clinical trials, small phase I or II noncontrolled
ATLANTIC trial, even in patients with EGFR-mutant NSCLC trials with combination regimens (generally including a tar-
treated with durvalumab as third-line or later treatment with geted TKI as backbone), or retrospective analysis from real
high PD-L1 expression (defined as at least 25%), PFS was clinical practice. Consequently, the level of evidence of the
short (median 1.9 months; 95% CI, 1.8–3.6), although the use of immunotherapy in patients with NSCLC and driver
response rate was higher in EGFR/ALK-positive lung can- mutations is low, and clinical decisions require a rigorous
cers with high PD-L1 expression (12% vs. 4%).7 A practical judgment balancing the benefits and harms to the patient
concern regarding the routine use of PD-1 or PD-L1 in- (Table 2 and Fig. 1).
hibitors before the need for EGFR TKIs is the concern about In this regard, the creation of multicenter, global registries
heightened toxicities noted in patients with stage IV disease such as the IMMUNOTARGET, a global multicenter registry

Calles, Riess, and Brahmer
TABLE 1. EGFR-TKI in Combination With ICI for Patients With Nonsquamous, EGFR-Mutant NSCLC
TKI Erlotinib Erlotinib Gefitinib Erlotinib/Gefitinib Osimertinib
ICI Atezolizumab Nivolumab Durvalumab Pembrolizumab Durvalumab

Trial phase Ib I I I/II Ib
Line of First or second First or second First First First or second
treatment
No. of 28 20 56 19 44
patients
ORR 75% Second line: 15% 63% 41.7% First line: 70%
Second line: T790M-
positive: 67%
Second line: T790M-
negative: 21%
PFS, months 15.4 10.1
Halted No No No No Yes
Grade 3–5 43% 25% 70% 71.4% (gefitinib 38%
toxicity arm)
Main toxicity ALT increase, pyrexia, Liver enzyme, ALT/AST increase; . 50% treatment- Liver toxicity, rash Interstitial lung
rash, diarrhea diarrhea, weight emergent adverse events leading to disease
loss discontinuation
8 9 10 11 12
Reference
Abbreviations: ALT, alanine aminotransferase; AST, aspartate transaminase; ICI, immune checkpoint inhibitor; ORR, objective response rate; PFS,
progression-free survival; TKI, tyrosine kinase inhibitor.
built to retrospectively collect the clinical experience of retrospective series that build the current evidence available
patients with NSCLC and driver mutations treated with on the use of immunotherapy in patients with NSCLC and
immunotherapy in the real world, is noteworthy. This study driver mutations.17-21
comprised 551 patients and is the largest series to date, With some exceptions, single anti–PD-(L)1 agents have
including information from KRAS, EGFR, BRAF, MET, offered low efficacy in the treatment of patients with
HER2, ALK, RET, and ROS1. Limitations include the bias of oncogenic-driven tumors.22 There are several explanations
the retrospective nature of the study and the limited number for the modest activity of immunotherapy when a driver
of cases for some groups. In addition to IMMUNOTARGET, mutation is present in the tumor. First, actionable driver
there are other single-center or multicenter collaborative mutations in lung cancer are enriched in patients who never
TABLE 2. Best Overall Response Rate Among Patients With NSCLC by Driver Mutation by Treatment Options and Levels of Evidence
EGFR ALK ROS1 BRAF KRAS HER2 MET RET NTRK
Targeted therapy 80% a
83% 77% 64% 54% b
55% 71% 68% 75%
ICI 11% 4% 14% 24% 57%c 15% 23% 11% NA
25%
ICI + targeted therapy 75%d 81%d
Chemotherapy + ICI 81%a NA 41%
NOTE. The following key refers to data source: bold, randomized phase III clinical trials; italic, phase I or II single-arm clinical trials; bold and italic, subgroup
analysis of phase III clinical trials; bold and underline, pooled analysis of retrospective series. For EGFR and ALK, data from phase II noncomparative trials and
subgroup analysis of phase III clinical trials are also available. For KRAS, data from subgroup analysis of phase II or III clinical trials are also available.
Abbreviations: ICI, immune checkpoint inhibitor.
a
In sensitizing mutations.
b
Specific KRAS G12C inhibitor.
c
In first line.
d
Increased grade 3 toxicities.

expression in 5%. PFS for patients with first-line EGFR-

mutant NSCLC treated with osimertinib remained un-
affected by PD-L1 expression status.25
ICI monotherapy In a meta-analysis of CheckMate 057,
KEYNOTE 010, POPLAR, and OAK trials, the use of single-
agent nivolumab, pembrolizumab, or atezolizumab was not
associated with better outcomes than in the wild-type
population in previously treated EGFR-mutant NSCLC.26
This was similar to what was observed in KEYNOTE 001,
where pembrolizumab obtained worse outcomes in the
FIGURE 1. Comparative Role of Immunotherapy in NSCLC With Driver EGFR-mutant subgroup than in wild-type patients.27 In the
Mutations phase II ATLANTIC trial, durvalumab was evaluated as third-
Abbreviation: ICI, immune checkpoint inhibitor. line or later in patients with EGFR/ALK-mutant NSCLC with
tumor PD-L1 expression greater than or equal to 25% or
less than 25% (n = 111).7 Although EGFR/ALK-mutant
smoked or had infrequent smoking history. Lung adeno- NSCLC with PD-L1 expression 25% or more had a higher
carcinomas with driver mutations from never-smokers have response rate (12.2% compared with 3.6% in PD-L1 ,
lower tumor mutational burden (TMB) than lung adeno- 25%), overall responses achieved were lower than in the
carcinomas from smokers. Second, the tumor microenvi- wild-type population and did not translate into survival gain.
ronment in lung cancer with driver mutations is considered Moreover, EGFR mutations have been found in some
a “cold tumor” based on lack of CD8+ tumor-infiltrating studies but not others to be a risk factor for hyper-
lymphocytes (TILs). Both low TMB and low CD8+ TILs are progressive disease when treated with immunotherapy.28
related to the limited efficacy of ICIs for these patients. In
In treatment-naive patients with EGFR-mutant PD-L1–positive
a recent multivariable model of response to immunotherapy in
tumors ( 1%, 22C3 antibody), pembrolizumab did not show
NSCLC, TMB (high), activating mutations (absence), molec-
any responses in the first 11 patients enrolled, which in-
ular smoking signature, and HLA status predicted response.23
cluded 73% of patients with strong PD-L1 expression of
PD-L1 expression is a continuous variable that has been 50% or higher.29 Of note, 86% (6 of 7) patients sub-
used as a predictor of immunotherapy benefit in NSCLC. sequently treated with EGFR TKI developed adverse events,
However, PD-L1 expression in tumors harboring driver including one patient with grade 3 transaminitis and
mutations does not necessarily correlate with response to grade 5 pneumonitis.29
ICIs. High PD-L1 expression may represent a constitutive
Overall, the benefit of immuno-oncology (IO) monotherapy
activation of PD-L1 signal and not a marker of adaptive
is poor, and the reasons remain to be elucidated. In a ret-
immune response. This may confound the predictive value
rospective single-institution experience at MGH, 58 EGFR/
of PD-L1 expression in tumors with driver mutations not
ALK-positive patients treated with immunotherapy were
necessarily linked to clinical response to immune check-
identified.18 Responses were observed only in one of 28
point inhibitors. For instance, differential expression of PD-
patients (3.6%). In paired pre- and post-TKI biopsies,
L1 has been described across NSCLC with driver mu-
EGFR/ALK-mutant patients had both low PD-L1 expression
tations, highlighting the importance of the genetic back-
and low presence of CD8+ TILs. This “cold” tumor mi-
ground in the PD-L1 expression: KRAS and MET alterations
croenvironment may explain the limited efficacy of IO in
associated with higher tumor PD-L1 expression, opposite to
EGFR/ALK-positive tumors. In another retrospective study,
EGFR mutations.24 A point of debate is the extent of mo-
those EGFR-mutant tumors with acquired resistance to
lecular testing to be performed as part of routine practice for
EGFR-TKI had different benefit from nivolumab depending
a patient without driver mutations and the tumor expression
on the T790M status. In T790M-negative tumors, nivolu-
levels of PD-L1 for decision making.
mab had better outcomes than in those with T790M-positive
EGFR tumors, and the reason may be that higher CD8+ TILs and
The role of ICI in the clinical treatment of patients with TMB were found in T790M-negative tumors.30
EGFR-mutant NSCLC remains controversial. Importantly, Not all EGFR mutations are equal, which may explain why
the efficacy of osimertinib for patients with EGFR-mutant certain EGFR mutation subtypes respond differently to ICI.
NSCLC is not influenced by the level of tumor PD-L1 ex- In a multicenter, retrospective study in U.S. centers (Yale
pression. A retrospective analysis from the FLAURA trial Cancer Center, Memorial Sloan Kettering Cancer Center
demonstrated PD-L1 expression of at least 1% in 51% of [MSKCC], University of California–Los Angeles, and Dana-
EGFR-mutant tumors and at least 50% PD-L1 tumor Farber Cancer Institute) involving 171 EGFR-mutant

patients treated with ICI, EGFR exon 19 deletion tumors were phase III randomized clinical trials in the first-line treatment
shown to have lower TMB than EGFR L858R despite a similar in which patients with NSCLC and EGFR mutation or ALK
smoking history. In addition, patients with EGFR exon 19– translocation who progressed to approved targeted thera-
deleted tumors had worse outcomes than patients with EGFR pies were allowed to enroll. However, the EGFR/ALK pop-
L858R–mutant lung cancer. Responses or survival out- ulation represented only 13% and 6% of the population in
comes were not affected by EGFR T790M status or PD-L1 these trials, respectively, and was not included in the pri-
expression.31 In the IMMUNOTARGET registry, 125 EGFR- mary analysis of any trial. As a consequence, all information
mutant patients treated with ICI obtained a 12.2% response provided from these trials is exploratory, and further pro-
rate with 2.1 months PFS, which confirms the limited benefit spective confirmation is needed.
of ICI in this population. Of note, exon 21 EGFR mutations
The IMpower150 trial included 124 patients with EGFR-
showed longer PFS than exon 19 deletions or T790M muta-
mutant NSCLC, 91 (73%) corresponding to sensitizing
tions.17 These results may have implications for designing clinical
exon 19 deletions or exon 21 L858R point mutations. In
trials to stratify patients according to EGFR mutation type.
EGFR-positive patients, atezolizumab + bevacizumab +
Combinational strategies have been developed to increase carboplatin + paclitaxel (ABCP; n = 34) compared with
the likelihood of response to immunotherapy. bevacizumab + carboplatin + paclitaxel (BCP; n = 45) yielded
ICI in combination with EGFR-TKIs Several early-phase a higher response rate (71% vs. 42%), duration of response
clinical trials have not clearly shown either additive or (11.1 vs. 4.7 months), median PFS (10.2 vs. 6.9 months; HR,
synergistic activity of the combination of targeted EGFR TKI 0.61; 95% CI, 0.36–1.03), and median overall survival (OS)
with ICIs (Table 1). On the contrary, reports of increased (not evaluable [NE] vs. 18.7 months; HR, 0.61; 95% CI,
grade 3 or higher toxicities have led to discontinuation of this 0.29–1.28). In patients with EGFR-sensitizing mutations,
therapeutic approach.32 In the phase Ib TATTON trial, ABCP efficacy seemed to be higher, with an objective re-
osimertinib plus durvalumab was associated with a high sponse rate (ORR) of 81%, median duration of response
incidence of grade 3 and 4 interstitial lung disease (38% of 11.1 months, PFS (10.3 vs. 6.1 months; HR, 0.41; 95% CI,
compared with . 3% reported for each single agent alone), 0.23–0.75), and OS (NE vs. 17.5 months; HR, 0.31; 95% CI,
and recruitment was discontinued.12 0.11–0.83). The efficacy of ABCP in the EGFR-mutant pop-
ulation was not related to PD-L1 expression. Interestingly, the
A retrospective analysis through the U.S. Food and Drug
better outcomes described in the ABCP arm for patients with
Administration adverse event reporting system database
EGFR/ALK mutations in the IMpower150 trial were not ob-
found a higher proportion of interstitial pneumonitis for
served for atezolizumab in combination with chemotherapy
nivolumab in combination with EGFR-TKI (25.7%) com-
without bevacizumab, either in arm A of IMpower150 or in the
pared with either drug alone, which were less than 5%,
IMpower130 trial, raising the question of the role of VEGF
with an odds ratio of 5.09 (95% CI, 2.87–9.03).14
inhibition and angiogenesis as a key component of EGFR/ALK
In addition, sequential use of EGFR TKI after ICI therapies cancer immunity.
has also been related to an increased risk of severe
In terms of safety, there were no differences in the number
immune-related adverse events (irAEs). In 41 patients
or grade of adverse events among patients with EGFR
treated with PD-(L)1 followed by osimertinib, six patients
mutations regardless of treatment arm in the IMpower150
(15%) developed severe irAEs.13 This included four cases of
trial. However, adverse events of special interest, which
grade 3 pneumonitis, one case of grade 3 colitis, and one
include irAEs and infusion-related reactions, were higher in
case of grade 4 hepatitis. The onset of irAEs occurred during
both atezolizumab-containing arms compared with the
the first week after the start of osimertinib, but latency could
bevacizumab plus chemotherapy arm: ABCP, 54.5%
be up to 3 months after the last dose of immunotherapy. The
(grades 3 and 4, 9.1%), atezolizumab + carboplatin +
extended pharmacodynamic life of anti–PD-(L)1 antibodies
paclitaxel, 52.3% (grades 3 and 4, 9.1%), and BCP, 22.7%
may explain why these irAEs occurred even several weeks
(grades 3 and 4, 2.3%).
after ICI discontinuation. By contrast, none of the 29 pa-
tients who were treated with osimertinib followed by PD-(L)1 Ultimately, prospective evidence of the use of chemo-
developed irAEs. These findings have important clinical immunotherapy in patients with EGFR-mutant NSCLC will
implications when considering immunotherapy approaches come from trials such as the KEYNOTE 789 trial, a phase III
for patients with driver mutations. randomized, double-blind, placebo-controlled study of
pemetrexed/platinum combined with pembrolizumab or
ICI in combination with chemotherapy The only evidence
placebo for patients with EGFR-TKI–resistant EGFR exon 19
available to date comes from subgroup analysis of the
deletion or L858R mutations (NCT03515837).
IMpower150 and IMpower130 studies, evaluating the
combination of atezolizumab with or without bevacizumab Another setting to explore the use of chemo-immunotherapy
with chemotherapy, respectively.33,34 These are the only for patients with EGFR-mutant NSCLC is the histologic

transformation to small-cell lung cancer after targeted limited number of ALK-positive patients in both trials limit
therapies, where platinum/etoposide in combination with our ability to draw formal conclusions. In 34 ALK-positive
atezolizumab or durvalumab has recently demonstrated evaluable patients in IMpower150, no statistically significant
a survival benefit in patients with treatment-naive small-cell differences were observed in PFS for the quadruplet
lung cancer. combination compared with the bevacizumab/chemother-
In summary, the use of ICIs for patients with EGFR-mutant apy arm (8.3 vs. 5.9 months; HR, 0.65; nonsignificant). In
tumors should be considered only after other effective the IMpower130 trial, 44 patients with NSCLC had ALK-
therapies fail in this population. The combination of im- rearranged tumors, and the EGFR/ALK group obtained
munotherapy with chemotherapy or antiangiogenic similar outcomes in both treatment groups of chemotherapy
agents must demonstrate efficacy and safety for pre- with and without atezolizumab.
viously treated patients with EGFR-mutant NSCLC in pro- KRAS
spective clinical trials. KRAS is the most common driver mutation found in patients
ALK with lung adenocarcinoma and occurs in both smokers and
never-smokers. ICIs in KRAS-mutant NSCLC have consis-
ICI monotherapy The evidence of immunotherapy efficacy
tently demonstrated clinical activity either as single agents
for patients with ALK-rearranged NSCLC is scant and in-
or in combination with chemotherapy for patients with
dicates low activity when ICI is used alone.
previously treated or first-line NSCLC. The genetic back-
In the phase II ATLANTIC trial, no responses were seen in ground of KRAS-mutant tumors may modify the response to
the 15 ALK-positive patients treated with durvalumab, re- immunotherapy. The presence of comutations in the tumor
gardless of PD-L1 expression.7 In the phase III OAK trial, suppressor gene STK11 confers poor survival and re-
only four patients were treated with atezolizumab, and no sistance to immunotherapy. KRAS/STK11-mutant tumors
formal evaluation was made.35 In the IMMUNOTARGET correlate with low expression of immune genes and PD-L1,
study, 23 ALK-positive patients were treated with ICI, with no and they regulate innate immune response to generate an
responses reported, achieving a PFS of 2.5 months and OS immunosuppressive tumor microenvironment. More re-
of 17 months.17 In the MGH experience, no responses were cently, type-specific KRAS inhibitors have shown for the
seen in the six ALK-positive patients evaluated.18 In a ret- very first time clinically meaningful efficacy for patients
rospective multicenter French study, only two of eight ALK- with KRAS-mutant NSCLC. In preclinical models, the use
positive patients achieved responses.19 of G12C KRAS inhibitors has resulted in increased
Importantly, a higher risk of hepatic toxicities has been proinflammatory antitumor activity either alone or in
reported with sequential use of ALK inhibitors after im- combination with ICIs, a finding that warrants clinical
munotherapy for the treatment of patients with ALK- validation.41
rearranged NSCLC. In a single-institution study at MGH, BRAF
453 patients were treated with crizotinib for ALK/ROS1/MET
BRAF mutations are found in 5% of patients with non-
alterations. Five of 11 patients treated with crizotinib after
squamous NSCLC. Class I BRAF V600 mutations represent
receiving previous treatment with an ICI developed grade 3
50% of BRAF mutations in lung cancer and arise in both
or 4 liver enzyme elevation, corresponding to an incidence
smokers and never-smokers. Class II and III non-V600
of 45.5% compared with 8% for those receiving crizotinib
BRAF mutations are more likely to be found in smokers
alone, leading to permanent discontinuation of crizotinib in
and are found together with comutations in genes of the
four of five patients.
RAS signaling pathway. Whereas a BRAF inhibitor in
ICI in combination with ALK-TKIs In several phase I and II combination with an MEK inhibitor is the therapy of choice
noncontrolled clinical trials, several ALK TKIs have been for patients with BRAF V600E–mutated tumors, no targeted
combined with different anti–PD-(L)1 monoclonal anti- therapy is available for non-V600 BRAF mutations.
bodies, resulting in general to response rates similar to
BRAF-mutant lung cancer is characterized by high levels of
historic ones for ALK TKIs (Table 3). In the absence of
tumor PD-L1 expression, low or intermediate TMB, and
a clear survival gain, the greater toxicity reported with these
microsatellite-stable status.42 The only evidence of the ac-
combinational regimens preclude additional research in
tivity of ICIs in patients with BRAF-mutant NSCLC comes
this setting.
from retrospective series, with limited numbers of patients
ICI in combination with chemotherapy As in EGFR-mutant and a heterogeneous population (Table 4).17,20,42-44 Re-
lung cancer, the only available information derives from sponse rate to single anti–PD-(L)1 agent is reported to be in
subgroup analysis of phase III trials IMpower150 and the range of 10% to 30%, with a median PFS of 3 months,
IMpower130. However, use of the combined group with which is to be expected of a second-line ICI in wild-type
EGFR/ALK-positive patients to perform analyses and the NSCLC. Both responses and PFS were found to be higher in

TABLE 3. ALK-TKI in Combination With ICI for Patients With ALK-Rearranged NSCLC
TKI Crizotinib Ceritinib Alectinib Lorlatinib Crizotinib
ICI Nivolumab Nivolumab Atezolizumab Avelumab Pembrolizumab

Trial phase I and II Ib Ib Ib and II Ib
Line of First First and beyond First Beyond first First
treatment
No. of 13 33 21 28 9
patients
ORR 38% First line: 81% 46% 56%
450 mg: 83%
300 mg: 60%–80%
Second line:
450 mg: 50%
300 mg: 25%
PFS, months 21.7
Halted Yes (toxicity) No but DLTs No DLTs No DLTs Yes (low accrual)
Grade 3–5 38% ALT/GGT elevation 67%; grade 54%; hypertriglyceridemia 33%; ALT/AST
toxicity (25%), rash, amylase/lipase 3 rash 19% and GGT elevation elevation,
elevation (11%) pneumonitis
Grade 3 ALT
9.5%
Pneumonitis
4.8%
Main toxicity Hepatic toxicity (includes Rash (64%) Rash (52%) Hypertriglyceridemia, ALT/AST elevation,
2 deaths) hypercholesterolemia fatigue
Notes ORR correlates to PD-L1 No grade 4
expression toxicity
36 37 38 39 40
Reference
Abbreviations: TKI, tyrosine kinase inhibitor; ICI, immune checkpoint inhibitor; ORR, objective response rate; PFS, progression-free survival; DLT, dose-
limiting toxicity; ALT, alanine aminotransferase; GGT, gamma-glutamyl transferase; AST, aspartate transaminase.
non-V600E mutations compared with V600E mutations. analysis.17 Interestingly, all the tumors expressed PD-L1,
Smoking but not PD-L1 expression was also found to be with a strong PD-L1 expression of at least 50% in 60% of
related to better outcomes. Overall, these data suggest the cases.
activity of ICI in patients with BRAF-mutant lung cancer.
MET
There are no data on ICI efficacy either alone or in com-
MET exon 14 skipping–mutant NSCLC tumors are associ-
bination with chemotherapy in subgroup analysis from
ated with high PD-L1 expression.24,45 In contrast, TMB is
pivotal phase III randomized clinical trials, in the first or
lower in patients with MET exon 14–altered NSCLCs com-
subsequent lines. Furthermore, there are no ongoing pro-
pared with other NSCLCs and did not correlate to PD-L1
spective trials to compare BRAF-targeted therapy with
expression. In a retrospective series from MSKCC and Dana-
immunotherapy-based strategies in BRAF-mutant lung
Farber Cancer Institute, 24 patients with lung cancer and
cancer.
MET alterations received immunotherapy (11 in first-line, 6
Novel pan-RAF inhibitors (inhibiting not only mutant B-RAF in second-line, and 7 in third-line). The ORR to ICI was 17%
but also wild-type B-RAF and C-RAF) and dual BRAF/MEK (95% CI, 6–36), and median PFS was 1.9 months, re-
inhibitors are being explored for safety and efficacy in gardless of PD-L1 expression or TMB. In the IMMUNO-
combination with anti–PD-(L)1 drugs in patients with TARGET study, ICIs yielded an ORR of 15.6% and a
NSCLC. 3.4-month PFS for 36 patients with MET-mutant NSCLC.17
In another retrospective series involving 30 patients with
ROS1 MET-mutant NSCLC, immunotherapy obtained 35.7% of
Responses to single-agent ICI have been reported in responses, a 10.4-month of duration of responses, and
one of seven patients (16.7%) from a retrospective a 4.9-month PFS.20

TABLE 4. ICI Efficacy in BRAF-Mutant NSCLC in Retrospective Series

Study BRAF No. of Patients ORR (%) PFS (months) OS (months)
Global All 43 24 3.1 13.6
IMMUNOTARGET (n = 35)17
V600E 17 1.8 8.2
Non-V600 18 4.1 17.2
United States V600E 10 10 1.4 26
Memorial Sloan Kettering Cancer Center (n = 46)43
Non-V600 36 22 3.2 24
Israel V600E 12 25 3.7 NR
ILCG (n = 39)42
Non-V600E 10 33 4.1
Italy NA 11 9 NA 10.3
EAP nivolumab (n = 11)44
France V600 26 26 5.3 22.5
GFPC 01-2018 (n = 44)20
Non-V600 18 35 4.9 12
Abbreviations: ICI, immune checkpoint inhibitor; ORR, objective response rate; PFS, progression-free survival; OS, overall survival.
RET progression to receive the selective RET TKI of the study, so we

Limited efficacy of ICI in tumors with RET fusions has been can also detect patterns of toxicity of the treatment sequence
previously reported.46 Conflicting results of the activity of ICI and infer the impact of sequencing in OS.
in patients with RET-rearranged NSCLC are available from HER2
retrospective data. In a single-institution experience, 13
Evidence on the use of ICIs for patients with lung cancer and
patients with RET-rearranged lung cancer had both low PD-
HER2 mutations is derived from retrospective studies. In
L1 expression and low TMB.47 No responses were achieved
a retrospective study at MSKCC, 122 patients with HER2-
for patients with either anti–PD-(L)1 or combination with
mutant lung cancer were identified.48 PD-L1 expression was
anti-CTLA4. However, there was no representation of tu-
lower but TMB was similar to those in unselected lung
mors with high PD-L1 expression of at least 50%, which
cancers. In the 26 patients with NSCLC and HER2 muta-
limits our ability to draw formal conclusions in this particular
tions treated with ICI, a response rate of 12% was reported,
subgroup of patients. Data from registry studies showed
with a median duration of response of 3.4 months. None of
disparate results: Only one patient out of 16 with RET-
the responders had an HER2 exon 20 YVMA mutation. PFS
positive NSCLC responded (ORR 6%), and PFS was 2.1
and OS were 1.9 and 10.4 months, respectively. In the
months in IMMUNOTARGET17; an ORR of 37.5% and a
IMMUNOTARGET registry,17 29 patients with NSCLC had
7.6-month PFS were reported in nine patients from the
HER2 mutations, reporting a 7.4% response rate and 2.5-
GFPC 01-2018 study.20 In the latter, 56% of cases were PD-
month PFS. PFS was noted to be longer in smokers (3.4 vs.
L1 negative, and 22% had PD-L1 of at least 50%. All ICIs
2 months). In another retrospective study involving 23
were single agents used in the second line or beyond.
patients with HER2-mutant NSCLC, responses were seen in
In contrast to the suggested low activity of IO in NSCLC with six patients (27.3%), achieving a PFS of 2.2 months and OS
RET fusions, both targeted therapy and pemetrexed-based of 20.4 months.20
chemotherapy are highly active in these patients. The
benefit of combining chemotherapy with immunotherapy is Other Driver Mutations (NTRK, NRG1, Others)
unclear in this situation but merits evaluation. The highly There are not enough cases to provide clinical guidance.21
selective and effective RET inhibitors selpercatinib (LOXO-
292) and pralsetinib (BLU-667) are being evaluated as first- Brain Metastasis
line treatment of RET fusion–positive NSCLC in multicenter, A particular clinical feature in patients with NSCLC har-
randomized, open-label, phase III clinical trials compared boring oncogenic driver mutations is the increased in-
with standard-of-care platinum-based and pemetrexed cidence of brain metastasis. Brain metastasis can be
chemotherapy with or without pembrolizumab (NCT04194944, present in 20% to 40% of patients with lung cancer and
NCT04222972). Although these trials are not formally driver mutations at the time of diagnosis, increasing to 80%
designed to compare TKI efficacy with a chemo-IO arm, we over the course of the disease. Brain metastases are as-
can certainly gain indirect evidence of the efficacy of the sociated with decreased quality of life and poorer survival
immunotherapy combined with chemotherapy in RET- compared with patients without brain metastases. Thus,
translocated NSCLC. Moreover, in both trials patients ran- the management of brain metastases is a continuous
domly assigned to the control arm may cross over upon challenge for patients with lung cancer and oncogenic driver

mutations based on the heterogeneity of the presentation, Among patients who have never smoked and have NSCLC
different access to radiotherapy techniques, or surgical adenocarcinoma, there is a high incidence of actionable
capabilities. In addition, new targeted therapies have high molecular drivers that can drastically change the disease
central nervous system penetration and have demonstrated course and prognosis; in some studies the rate is as high as
high efficacy for treating brain metastases and preventing 90% in detecting a potential oncogenic driver that can be
the onset of new ones. matched to an appropriate targeted therapy.58-60 In a meta-
Immunotherapy has demonstrated activity for patients with analysis of 167 studies, patients with lung cancer who never
NSCLC and previously treated brain metastases, as per smoked had higher odds of EGFR and ALK mutations than
subgroup analysis for pivotal trials, but also for patients with ever smokers.61 As another example, in several studies
untreated brain metastases.49 However, limited information looking at the frequency of EGFR and ALK mutations in
is available for patients with NSCLC with driver mutations never-smokers conducted mainly in Asia, EGFR mutations
and brain metastases treated with immunotherapy. More- were very common (48% to 76%), and, in one large U.S.
over, prospective data on the efficacy and safety of radiation study, EGFR mutations occurred at a similar frequency in
therapy and ICIs for patients with NSCLC and driver mu- never-smokers (49%).62
tations are lacking.50 It is hypothesized that the combination With these advances, molecular testing for approved tar-
of ICIs and radiation therapy to the brain may increase the geted therapies (EGFR mutation, ALK fusion, ROS1 fusion,
antitumor activity of the immune system, challenging the BRAF V600E/K) is essential, and broad molecular profiling
belief that the brain is an immunologically privileged organ. including RET, MET exon 14, Her2 mutations, NTRK fusion,
and other potentially actionable alterations is strongly en-
In a recent meta-analysis, treatment of brain metastases
couraged. Guidelines recommend generating results within
with stereotactic radiosurgery and ICI was associated with
2 weeks (10 business days) to provide timely information for
better outcomes for concurrent rather than sequential ad-
treatment.63
ministration, with respective 1-year local control rates of
89.2% and 67.8% and respective 1-year OS of 64.6% and PD-(L)1 ICIs have similarly revolutionized advanced lung
51.6%.51 The overall incidence of radionecrosis was esti- cancer treatment. For patients with high PD-(L)1 expression
mated at 5.3%. Most studies in this meta-analysis included ( 50% TPS), pembrolizumab is approved as first-line
patients with melanoma brain metastases, but two included treatment based on higher PFS and OS compared with
studies of patients with NSCLC confirmed their separate platinum-based chemotherapy.64 Similarly, platinum-based
findings.52,53 chemotherapy (KN189 platinum/pemetrexed for non-
squamous; KN407 platinum-(nab)paclitaxel for squamous)
The optimal timing, dosing, and sequencing of radiation
in combination with pembrolizumab is approved regardless
therapy in combination with immunotherapy remains
of PD-L1 expression based on PFS and OS benefit
a subject of continued investigation, particularly in the
compared with platinum-based chemotherapy alone.65,66
population of patients with NSCLC and oncogenic driver
However, patients with EGFR mutations and ALK fusions,
mutations.
representing the highest proportion of molecular aberrations
in never-smokers, were excluded from these trials and do
NEVER-SMOKER WITH LUNG CANCER PENDING MUTATION not appear to derive benefit from PD-(L)1 blockade, with
STATUS: WAIT OR TREAT? multiple trials and meta-analysis not demonstrating im-
The proportion of actionable molecular driver mutations in proved outcomes with single-agent PD-(L)1 inhibitors1,29,67
advanced NSCLC where initiation of approved targeted (Table 2). Moreover, patients with lung cancers with
therapies substantially improves clinical outcomes is in- smoking-associated signatures may benefit more from
creasing. Most (EGFR-activating mutation, ALK fusion, checkpoint immunotherapy, which generates more neo-
ROS1 fusion) but not all (BRAF V600E/K) are associated antigens that can prime and activate neoantigen-specific
mainly with patients with little or no smoking history, al- T cells in the presence of PD-(L)1 blockade.68,69 The ad-
though these molecular aberrations can still occur in dition of the PD-L1 antibody atezolizumab concurrent with
smokers or former smokers.54-56 These mutations are often the VEGF antibody bevacizumab to carboplatin/paclitaxel
associated with adenocarcinoma histology, but they can still improved PFS and OS compared with carboplatin/paclitaxel/
occur in squamous and other histologic subtypes (i.e., large bevacizumab in the IMpower150 study.70 In a subset
cell, MET exon 14 skipping mutations in sarcomatoid his- analysis of patients with EGFR mutations and ALK trans-
tology), with increased frequency among patients with little locations who progressed on previous TKI, PFS was longer
or no smoking history. Moreover, other driver mutations and ORR was robust at 71% in patients with EGFR muta-
(RET fusions, HER2 mutations, MET exon 14 skipping tions, with hypothesized benefit from VEGF inhibition further
mutations) are increasingly being matched to effective potentiating clinical activity of PD-L1 blockade.33,71 How-
targeted therapies in clinical trials.57 ever, this was only in a subset analysis and requires further

validation, and the regimen is not currently approved by the blockade and the length of time it remains bound to the
U.S. Food and Drug Administration for EGFR-mutated or receptor13,14,74 (Tables 1 and 3). Because of this concern, if
ALK-positive NSCLC. systemic treatment must be initiated because of pressing
symptoms, one could consider platinum-based chemo-
Thus, given the dramatic benefits of TKIs as approved
therapy without PD-L1 immune checkpoint blockade while
targeted therapies and the potential for clinical trials in
waiting for molecular testing results.
certain molecular subsets of never-smoking NSCLC, the
recommendation is to wait for molecular testing results To conclude, in a never-smoking patient pending mutation
before initiating therapy. For example, the median PFS of status, in the absence of symptomatic deterioration, waiting
first-line alectinib is on the order of almost 3 years, and for molecular test results is best. At minimum, waiting for
osimertinib as first-line treatment of EGFR-mutant NSCLC test results is warranted for EGFR mutation and ALK and
provides a substantial PFS benefit and an OS benefit even ROS1 fusions, where highly active targeted treatments are
approved and available and where benefit of PD-L1
over first-generation EGFR-TKI.72,73
blockade either alone or in combination with chemo-
Moreover, if treatment is started with an immunotherapy- therapy is unclear. Given the high likelihood of an ac-
containing regimen and an oncogenic driver is discovered, tionable driver mutation that can be matched to a more
there appears to be greater risk of irAEs, including pneu- effective therapy and considering potentially higher
monitis, in combined TKI/PD-L1 inhibitor trials and for immune-related toxicity with ICIs and TKI, the brief delay
patients who received PD-L1 inhibitors before TKI as part in starting treatment while molecular testing results are
of standard treatment, given the long half-life of PD-L1 pending is often worthwhile.
AFFILIATIONS 28007 Madrid, Spain; Twitter: @Tony_Calles; email: antonio.calles@

1
Medical Oncology Department, Hospital General Universitario Gregorio live.com.
Marañón, Madrid, Spain
2
UC Davis Comprehensive Cancer Center, Sacramento, CA AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
3
Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, AND DATA AVAILABILITY STATEMENT
Baltimore, MD Disclosures provided by the authors and data availability statement (if
EDBK_280795.
Antonio Calles, MD, MSc, Medical Oncology Department, Hospital
General Universitario Gregorio Marañón, 46 Doctor Esquerdo Street,
REFERENCES
1. Gainor JF, Shaw AT, Sequist LV, et al. EGFR mutations and ALK rearrangements are associated with low response rates to PD-1 pathway blockade in non-small
cell lung cancer: a retrospective analysis. Clin Cancer Res. 2016;22:4585-4593.
2. Herbst RS, Baas P, Kim DW, et al. Pembrolizumab versus docetaxel for previously treated, PD-L1-positive, advanced non-small-cell lung cancer (KEYNOTE-010):
a randomised controlled trial. Lancet. 2016;387:1540-1550.
3. Borghaei H, Paz-Ares L, Horn L, et al. Nivolumab versus docetaxel in advanced nonsquamous non-small-cell lung cancer. N Engl J Med. 2015;373:1627-1639.
4. Rittmeyer A, Barlesi F, Waterkamp D, et al; OAK Study Group. Atezolizumab versus docetaxel in patients with previously treated non-small-cell lung cancer (OAK):
a phase 3, open-label, multicentre randomised controlled trial. Lancet. 2017;389:255-265.
5. Antonia SJ, Villegas A, Daniel D, et al; PACIFIC Investigators. Durvalumab after chemoradiotherapy in stage III non-small-cell lung cancer. N Engl J Med. 2017;
377:1919-1929.
6. Antonia SJ, Villegas A, Daniel D, et al; PACIFIC Investigators. Overall survival with durvalumab after chemoradiotherapy in stage III NSCLC. N Engl J Med. 2018;
379:2342-2350.
7. Garassino MC, Cho BC, Kim JH, et al; ATLANTIC Investigators. Durvalumab as third-line or later treatment for advanced non-small-cell lung cancer (ATLANTIC):
an open-label, single-arm, phase 2 study. Lancet Oncol. 2018;19:521-536.
8. Rudin C, Cervantes A, Dowlati A, et al. MA15.02 long-term safety and clinical activity results from a phase Ib study of erlotinib plus atezolizumab in advanced
NSCLC. J Thorac Oncol. 2018;13:S407.
9. Gettinger S, Hellmann MD, Chow LQM, et al. Nivolumab plus erlotinib in patients with EGFR-mutant advanced NSCLC. J Thorac Oncol. 2018;13:1363-1372.

10. Creelan BC, Yeh T, Kim SW, et al. 84O: Phase I study of gefitinib (G) + durvalumab (D) for locally advanced/metastatic non-small cell lung cancer (NSCLC)
harbouring epidermal growth factor receptor (EGFR) sensitising mutations. Ann Oncol. 2019;30:ii31-ii37.
11. Yang JC-H, Gadgeel SM, Sequist LV, et al. Pembrolizumab in combination with erlotinib or gefitinib as first-line therapy for advanced NSCLC with sensitizing EGFR
mutation. J Thorac Oncol. 2019;14:553-559.
12. Ahn M, Yang J, Yu H, et al. 136O: Osimertinib combined with durvalumab in EGFR-mutant NSCLC: results from the Tatton phase 1b trial. J Thorac Oncol. 2016;
11:S115.
13. Schoenfeld AJ, Arbour KC, Rizvi H, et al. Severe immune-related adverse events are common with sequential PD-(L)1 blockade and osimertinib. Ann Oncol.
2019;30:839-844.
14. Oshima Y, Tanimoto T, Yuji K, et al. EGFR-TKI-associated interstitial pneumonitis in nivolumab-treated patients with non-small cell lung cancer. JAMA Oncol.
2018;4:1112-1115.
15. Gibbons DL, Chow LQ, Kim DW, et al. Efficacy, safety and tolerability of MEDI4736 (durvalumab), a human IgG1 anti-programmed cell death-ligand-1 (PD-L1)
antibody, combined with gefitinib (G): a phase 1 expansion in TKA naive patients (pts) with EGFR mutant NSCLC. J Thorac Oncol. 2016;11:S79.
16. Ma BBY, Rudin CM, Cervantes A, et al. Preliminary safety and clinical activity of erlotinib plus atezolizumab from a phase 1b study in advanced NSCLC. Ann
Oncol. 2016;27 (suppl_9):ix139-ix156.
17. Mazieres J, Drilon A, Lusque A, et al. Immune checkpoint inhibitors for patients with advanced lung cancer and oncogenic driver alterations: results from the
IMMUNOTARGET registry. Ann Oncol. 2019;30:1321-1328.
18. Gainor JF, Shaw AT, Sequist LV, et al. EGFR mutations and ALK rearrangements are associated with low response rates to PD-1 pathway blockade in non-small
cell lung cancer: a retrospective analysis. Clin Cancer Res. 2016;22:4585-4593.
19. Bylicki O, Guisier F, Monnet I, et al. Efficacy and safety of programmed cell-death-protein-1 and its ligand inhibitors in pretreated patients with epidermal growth-
factor receptor-mutated or anaplastic lymphoma kinase-translocated lung adenocarcinoma. Medicine (Baltimore). 2020;99:e18726.
20. Guisier F, Dubos-Arvis C, Viñas F, et al. Efficacy and safety of anti-PD-1 immunotherapy in patients with advanced non small cell lung cancer with BRAF, HER2 or
MET mutation or RET-translocation: GFPC 01-2018. J Thorac Oncol. 2020;15:628-636.
21. Dudnik E, Bshara E, Grubstein A, et al. Rare targetable drivers (RTDs) in non-small cell lung cancer (NSCLC): outcomes with immune check-point inhibitors
(ICPi). Lung Cancer. 2018;124:117-124.
22. Huang Q, Zhang H, Hai J, et al. Impact of PD-L1 expression, driver mutations and clinical characteristics on survival after anti-PD-1/PD-L1 immunotherapy versus
chemotherapy in non-small-cell lung cancer: a meta-analysis of randomized trials. OncoImmunology. 2018;7:e1396403.
23. Anagnostou V, Niknafs N, Marrone K, et al. Multimodal genomic features predict outcome of immune checkpoint blockade in non-small-cell lung cancer. Nat
Cancer. 2020;1:99-111.
24. Schoenfeld AJ, Rizvi H, Bandlamudi C, et al. Clinical and molecular correlates of PD-L1 expression in patients with lung adenocarcinomas. Ann Oncol. Epub
2020 Feb 6.
25. Brown H, Vansteenkiste J, Nakagawa K, et al. Programmed cell death ligand 1 expression in untreated EGFR mutated advanced NSCLC and response to
osimertinib versus comparator in FLAURA. J Thorac Oncol. 2020;15:138-143.
26. Lee CK, Man J, Lord S, et al. Clinical and molecular characteristics associated with survival among patients treated with checkpoint inhibitors for advanced non-
small cell lung carcinoma: a systematic review and meta-analysis. JAMA Oncol. 2018;4:210-216.
27. Garon EB, Hellmann MD, Rizvi NA, et al. Five-year overall survival for patients with advanced non‒small-cell lung cancer treated with pembrolizumab: results
from the phase I KEYNOTE-001 study. J Clin Oncol. 2019;37:2518-2527.
28. Kato S, Goodman A, Walavalkar V, et al. Hyperprogressors after immunotherapy: analysis of genomic alterations associated with accelerated growth rate. Clin
Cancer Res. 2017;23:4242-4250.
29. Lisberg A, Cummings A, Goldman JW, et al. A phase II study of pembrolizumab in EGFR-mutant, PD-L1+, tyrosine kinase inhibitor naı̈ve patients with advanced
NSCLC. J Thorac Oncol. 2018;13:1138-1145.
30. Haratani K, Hayashi H, Tanaka T, et al. Tumor immune microenvironment and nivolumab efficacy in EGFR mutation-positive non-small-cell lung cancer based
on T790M status after disease progression during EGFR-TKI treatment. Ann Oncol. 2017;28:1532-1539.
31. Hastings K, Yu HA, Wei W, et al. EGFR mutation subtypes and response to immune checkpoint blockade treatment in non-small-cell lung cancer. Ann Oncol.
2019;30:1311-1320.
32. Yang JC-H, Shepherd FA, Kim D-W, et al. Osimertinib plus durvalumab versus osimertinib monotherapy in EGFR T790M-positive NSCLC following previous EGFR
TKI therapy: CAURAL brief report. J Thorac Oncol. 2019;14:933-939.
33. Reck M, Mok TSK, Nishio M, et al; IMpower150 Study Group. Atezolizumab plus bevacizumab and chemotherapy in non-small-cell lung cancer (IMpower150):
key subgroup analyses of patients with EGFR mutations or baseline liver metastases in a randomised, open-label phase 3 trial. Lancet Respir Med. 2019;
7:387-401.
34. West H, McCleod M, Hussein M, et al. Atezolizumab in combination with carboplatin plus nab-paclitaxel chemotherapy compared with chemotherapy alone as
first-line treatment for metastatic non-squamous non-small-cell lung cancer (IMpower130): a multicentre, randomised, open-label, phase 3 trial. Lancet Oncol.
2019;20:924-937.
35. Fehrenbacher L, von Pawel J, Park, K, et al. Updated efficacy analysis including secondary population results for OAK: a randomized phase III study of
atezolizumab versus docetaxel in patients with previously treated advanced non-small cell lung cancer [published correction appears in J Thorac Oncol. 2018;13:
1800]. J Thorac Oncol. 2018; 13: 1156-1170.

36. Spigel DR, Reynolds C, Waterhouse D, et al. Phase 1/2 study of the safety and tolerability of nivolumab plus crizotinib for the first-line treatment of anaplastic
lymphoma kinase translocation - positive advanced non-small cell lung cancer (CheckMate 370). J Thorac Oncol. 2018; 13: 682-688.
37. Felip E, de Braud FG, Maur M, et al. Ceritinib plus nivolumab in patients with advanced ALK-rearranged non–small cell lung cancer: results of an open-label,
multicenter, phase 1B study. J Thorac Oncol. 2020;15:392-403.
38. Kim D-W, Gadgeel SM, Gettinger SN, et al. Safety and clinical activity results from a phase Ib study of alectinib plus atezolizumab in ALK+ advanced NSCLC
(aNSCLC). J Clin Oncol. 2018;36:15s (suppl; abstr 9009).
39. Shaw AT, Lee S-H, Ramalingam SS, et al. Avelumab (anti–PD-L1) in combination with crizotinib or lorlatinib in patients with previously treated advanced NSCLC:
phase 1b results from JAVELIN Lung 101. J Clin Oncol. 2018;36:15s (suppl; 9008).
40. Patel SP, Pakkala S, Pennell NA, et al. Phase Ib study of crizotinib plus pembrolizumab in patients with previously untreated advanced non-small cell lung cancer
with ALK translocation. Oncologist. Epub 2020 Feb 12.
41. Canon J, Rex K, Saiki AY, et al. The clinical KRAS(G12C) inhibitor AMG 510 drives anti-tumour immunity. Nature. 2019;575:217-223.
42. Dudnik E, Peled N, Nechushtan H, et al.; Israel Lung Cancer Group. BRAF mutant lung cancer: programmed death ligand 1 expression, tumor mutational
burden, microsatellite instability status, and response to immune check-point inhibitors. J Thorac Oncol. 2018;13:1128-1137.
43. Offin M, Pak T, Mondaca S, et al. P1.04-39 molecular characteristics, immunophenotype, and immune checkpoint inhibitor response in BRAF non-V600 mutant
lung cancers. J Thorac Oncol. 2019;14:S455.
44. Rihawi K, Giannarelli D, Galetta D, et al. BRAF mutant NSCLC and immune checkpoint inhibitors: results from a real-world experience. J Thorac Oncol. 2019;
14:e57-e59.
45. Sabari JK, Leonardi GC, Shu CA, et al. PD-L1 expression, tumor mutational burden, and response to immunotherapy in patients with MET exon 14 altered lung
cancers. Ann Oncol. 2018;29:2085-2091.
46. Hegde A, Huang L, Liu S, et al. Abstract 4997: responsiveness to immune checkpoint inhibitors in RET dependent cancers. Cancer Res. 2019;79:4997.
47. Offin M, Guo R, Wu SL, et al. Immunophenotype and response to immunotherapy of RET-rearranged lung cancers. JCO Precis Oncol. 2019;3:1-8.
48. Lai W-CV, Feldman DL, Buonocore DJ, et al. PD-L1 expression, tumor mutation burden and response to immune checkpoint blockade in patients with HER2-
mutant lung cancers. J Clin Oncol. 2018;36:15s (suppl; abstr 9060).
49. Goldberg SB, Gettinger SN, Mahajan A, et al. Pembrolizumab for patients with melanoma or non-small-cell lung cancer and untreated brain metastases: early
analysis of a non-randomised, open-label, phase 2 trial. Lancet Oncol. 2016;17:976-983.
50. Lehrer EJ, McGee HM, Peterson JL, et al. Stereotactic radiosurgery and immune checkpoint inhibitors in the management of brain metastases. Int J Mol Sci.
2018;19:3054.
51. Lehrer EJ, Peterson J, Brown PD, et al. Treatment of brain metastases with stereotactic radiosurgery and immune checkpoint inhibitors: an international meta-
analysis of individual patient data. Radiother Oncol. 2019;130:104-112.
52. Chen L, Douglass J, Kleinberg L, et al. Concurrent immune checkpoint inhibitors and stereotactic radiosurgery for brain metastases in non-small cell lung cancer,
melanoma, and renal cell carcinoma. Int J Radiat Oncol Biol Phys. 2018;100:916-925.
53. Ahmed KA, Kim S, Arrington J, et al. Outcomes targeting the PD-1/PD-L1 axis in conjunction with stereotactic radiation for patients with non-small cell lung cancer
brain metastases. J Neurooncol. 2017;133:331-338.
54. Li T, Kung HJ, Mack PC, et al. Genotyping and genomic profiling of non-small-cell lung cancer: implications for current and future therapies. J Clin Oncol. 2013;
31:1039-1049.
55. Govindan R, Ding L, Griffith M, et al. Genomic landscape of non-small cell lung cancer in smokers and never-smokers. Cell. 2012;150:1121-1134.
56. Paik PK, Arcila ME, Fara M, et al. Clinical characteristics of patients with lung adenocarcinomas harboring BRAF mutations. J Clin Oncol. 2011;29:2046-2051.
57. Paik PK, Drilon A, Fan PD, et al. Response to MET inhibitors in patients with stage IV lung adenocarcinomas harboring MET mutations causing exon 14 skipping.
Cancer Discov. 2015;5:842-849.
58. Grosse C, Soltermann A, Rechsteiner M, et al. Oncogenic driver mutations in Swiss never smoker patients with lung adenocarcinoma and correlation with
clinicopathologic characteristics and outcome. PLoS One. 2019;14:e0220691.
59. Zhou F, Zhou C. Lung cancer in never smokers: the East Asian experience. Transl Lung Cancer Res. 2018;7:450-463.
60. Kohsaka S, Hayashi T, Nagano M, et al. Identification of novel CD74-NRG2α fusion from comprehensive profiling of lung adenocarcinoma in Japanese never or
light smokers. J Thorac Oncol. Epub 2020 Feb 6.
61. Chapman AM, Sun KY, Ruestow P, et al. Lung cancer mutation profile of EGFR, ALK, and KRAS: meta-analysis and comparison of never and ever smokers. Lung
Cancer. 2016;102:122-134.
62. Zhang Y, Sun Y, Pan Y, et al. Frequency of driver mutations in lung adenocarcinoma from female never-smokers varies with histologic subtypes and age at
diagnosis. Clin Cancer Res. 2012;18:1947-1953.
63. Ettinger DS, Wood DE, Aggarwal C, et al.; OCN. NCCN guidelines insights: non-small cell lung cancer, version 1.2020. J Natl Compr Canc Netw. 2019;
17:1464-1472.
64. Reck M, Rodrı́guez-Abreu D, Robinson AG, et al.; KEYNOTE-024 Investigators. Pembrolizumab versus chemotherapy for PD-L1-positive non-small-cell lung

65. Paz-Ares L, Luft A, Vicente D, et al.; KEYNOTE-407 Investigators. Pembrolizumab plus chemotherapy for squamous non-small-cell lung cancer. N Engl J Med.
2018;379:2040-2051.
66. Gandhi L, Garassino MC. Pembrolizumab plus chemotherapy in lung cancer. N Engl J Med. 2018;379:e18.
67. Lee CK, Man J, Lord S, et al. Checkpoint inhibitors in metastatic EGFR-mutated non-small cell lung cancer: a meta-analysis. J Thorac Oncol. 2017;12:403-407.
Science. 2015;348:124-128.
69. Kim JH, Kim HS, Kim BJ. Prognostic value of smoking status in non-small-cell lung cancer patients treated with immune checkpoint inhibitors: a meta-analysis.
Oncotarget. 2017;8:93149-93155.
70. Socinski MA, Jotte RM, Cappuzzo F, et al.; IMpower150 Study Group. Atezolizumab for first-line treatment of metastatic nonsquamous NSCLC. N Engl J Med.
2018;378:2288-2301.
71. Hegde PS, Wallin JJ, Mancao C. Predictive markers of anti-VEGF and emerging role of angiogenesis inhibitors as immunotherapeutics. Semin Cancer Biol. 2018;
52:117-124.
72. Peters S, Camidge DR, Shaw AT, et al.; ALEX Trial Investigators. Alectinib versus crizotinib in untreated ALK-positive non-small-cell lung cancer. N Engl J Med.
2017;377:829-838.
73. Ramalingam SS, Vansteenkiste J, Planchard D, et al.; FLAURA Investigators. Overall survival with osimertinib in untreated, EGFR-mutated advanced NSCLC.
N Engl J Med. 2020;382:41-50.
74. Lin JJ, Chin E, Yeap BY, et al. Increased hepatotoxicity associated with sequential immune checkpoint inhibitor and crizotinib therapy in patients with non-small
cell lung cancer. J Thorac Oncol. 2019;14:135-140.

LUNG CANCER
Making Checkpoint Inhibitors Part of Treatment of

Patients With Locally Advanced Lung Cancers:
The Time Is Now
Mark G. Kris, MD1; Corinne Faivre-Finn, MD, PhD2; Tiana Kordbacheh, MD2; Jamie Chaft, MD1; Jia Luo, MD1; Anne Tsao, MD3; and
Stephen Swisher, MD3
overview
The PACIFIC trial of durvalumab administered for 1 year to patients with stage III lung cancers has set a new
standard of care. PACIFIC established the role of immune checkpoint inhibitors (ICIs) for individuals with
inoperable and unresectable locally advanced lung cancers that achieve disease control from concurrent
chemoradiation. For patients with resectable and operable disease, ICIs administered before surgery, either
alone (JHU/MSK, LCMC3, and NEOSTAR) or in combination with chemotherapy (Columbia/MGH and
NADIM), have yielded high rates of major pathologic response in resection specimens, an outcome measure
that correlates with improved progression-free survival and overall survival. These results have brought forth
the dilemma of how to choose the optimal local therapy—either definitive concurrent chemoradiation or
surgery—to use with an ICI for patients with stage III lung cancers that are both operable and resectable.
Here, we review the data that support the use of each local therapy. Recent successes have also raised the
possibility that using ICIs in patients with earlier stages of lung cancer will enhance curability. Randomized
trials are underway; however, until they read out, physicians must choose between local and systemic
therapies on the basis of the information we have today. Research demonstrates that using surgery, ra-
diation, chemotherapy, and ICIs improve all efficacy outcomes and curability. All modalities should be
considered in every patient with locally advanced lung cancer. It is imperative that a multimodality dis-
cussion that includes the possible addition of ICIs takes place to choose the best modality and sequence of
therapies for each patient.
INTRODUCTION We can begin by rethinking our basic assumptions

For the field of thoracic oncology, the treatment of concerning early-stage lung cancers. We can ac-
patients with locally advanced lung cancers marks knowledge that all stages of lung cancers are poten-
both one of our greatest successes and one of our tially deadly and routinely explore opportunities to
greatest disappointments. Combining surgery, radia- improve outcomes. Clinically or pathologically staged
tion, and chemotherapy, we can cure some, but not all, 1-cm primary lung cancers carry a 5-year survival of
individuals with stages II to III lung cancers. Despite only 92%.1 Many physicians caring for patients with
precise imaging studies and the availability of ad- lung cancers consider this a good prognosis. Current
vanced radiation techniques and minimally invasive treatment guidelines recommend no additional in-
complete resections, both delivered with utmost pre- terventions, even for patients with tumor characteris-
cision, we fail to achieve cure in the majority of patients. tics that suggest a higher risk of recurrence (discussed
Author affiliations We attempt to enhance the individual cure rates of in detail below). In contrast, patients with HER2-driven
and support local therapies with systemic ones, usually cytotoxic breast cancers with the same risk of recurrence are
information (if offered surgery, radiation, chemotherapy, 1 year of
chemotherapy administered before, during, or after the
applicable) appear
definitive local therapy. Cisplatin-based chemotherapy trastuzumab, and an additional year of neratinib.2 If
at the end of this
article. enhances the curability of locally advanced cancers to individuals with lung cancers have clinical evidence of
Accepted on March a degree that is comparable to or that exceeds that with mediastinal nodal spread, even with a 1-cm primary
25, 2020 and the use of perioperative therapies in breast or colorectal tumor, their 5-year survival estimates decrease to
published at cancers. However, in the minds of many physicians 32%.1 Additional treatment modalities are recom-
ascopubs.org on mended with clinical stage IIIA disease, but they still fall
and patients, the benefits are too small to justify the
April 16, 2020:
disruption in lifestyle brought on by multimodality lung short, especially for patients with large primary tumors
10.1200/EDBK_ cancer therapies. How can we build on our successes and mediastinal nodal spread. Better staging can
280807 and turn our frustration into a path to progress? define the risk of recurrence, but, regardless of disease

Kris et al
A recent study suggests an even greater degree of adjuvant

therapy benefit among individuals with tumors with certain
high-risk features, including lymphatic, vascular, or visceral
• Patients with locally advanced lung cancers pleural invasion and invasive size greater than 2 cm.5
(stage III) can achieve cure by combining
However, chemotherapy benefits have generally plateaued;
surgery, radiation, and systemic therapy in
few chemotherapeutic agents are under study, and, as the
a multimodal plan of care.
population of patients with lung cancers ages, many drugs,
• Even with optimal timing and delivery of sur- especially cisplatin, are difficult or impossible to deliver safely.
gery, radiation, and chemotherapy, relapse can
Targeted therapies paired with surgery or radiation in patients
occur, primarily at distant sites. To cure more,
with tumors with oncogenic drivers hold promise, but they
we need better systemic therapies—adding ICIs
to multimodal regimens can fill that need. have yet to become part of routine management. Concurrent
chemotherapy with definitive thoracic radiation is now
• For individuals with unresectable and in-
a worldwide standard of care for individuals with inoperable or
operable stage III NSCLCs that remain con-
unresectable stage III disease and is a comparable approach
trolled after concurrent chemotherapy and
radiation, 1 year of durvalumab improves both to surgery in patients with operable and resectable stage IIIA
progression-free and overall survivals—57% lung cancers in many cases. How the optimal local modality is
overall survival rate at 3 years, a 13% chosen to pair with cytotoxic chemotherapy is an area of
improvement. scientific inquiry and daily debate at the world’s multimodality
• Neoadjuvant approaches offer many advan- tumor boards. With the tools in hand, choosing the optimal
tages, both for the care of individual patients type and sequence of each will undoubtedly result in addi-
and for research. Preoperative opportunities tional incremental gains but not the transformative im-
should be considered in patients for whom provements that are necessary to make cure the norm and not
surgery and adjuvant therapy are appropriate the exception.
according to stage.
• In resectable stage IIIA NSCLCs (T4N0-1 or Immunotherapeutics Provide Additional Benefits Beyond
nonbulky single-station N2), the standard of Surgery, Radiation, and Chemotherapy
care should include consideration of surgical After more than a century of research, we have realized the
resection after multidisciplinary review (CRT → dream of therapeutics to harness a person’s immune system
surgery or chemotherapy → surgery → 6 RT).
to fight cancers. This breakthrough has the potential to
Thoracic surgical oncologists are in the best
position to determine operability and re- overcome the barriers we have faced to cure patients with
sectability. On the basis of research to date, it is locally advanced lung cancers. Experience in patients with
likely that, in the near future, ICIs will be added metastatic lung cancers demonstrates that modulating
to multimodality regimens. T-cell function using ICIs that target PD-1, PD-L1, and
CTLA4 can lead to durable tumor regressions, better overall
survival,6-8 and 5-year disease-free survival off therapy.9 ICIs
extent, our best therapies are unable to lead to cures in
provide an additional modality that enhances a person’s
many patients.
immune system to target cancer and can work in concert
Improved Results With Multimodality Approaches for with surgery, radiation, and cytotoxic chemotherapy to
Locally Advanced Lung Cancers achieve results beyond those with other modalities ad-
In the past 2 decades, we have made tremendous strides in ministered alone or in combination. The platform has been
surgery and radiation in terms of both oncologic success set for a new wave of clinical research to define optimal
and limiting treatment-related disability. Local failure alone systemic agents, dosing, scheduling, and sequencing of
rarely occurs with optimally delivered local therapies. both systemic and local therapies—all attempting to in-
Whereas continued advancements in these areas can be crease the curability of locally advanced lung cancers.
anticipated, they cannot be expected to have a major impact Trials using neoadjuvant ICIs represent the greatest amount
on the curability of lung cancers, in which relapse occurs of data available today. Five studies that have released data
systemically in most patients who are destined to experience on ICIs used alone or with cytotoxic chemotherapy before
relapse. Guideline-recommended chemotherapy—ideally surgery are reported in Table 1.10-15 In neoadjuvant trials in
cisplatin based, administered before or after surgery— patients with lung cancers, the primary endpoint is major
consistently improves curability beyond surgery alone in pathologic response (MPR), defined as less than 10% viable
patients with any nodal spread. Guidelines recommend tumor cells in the surgical resection specimen.16 MPR has
routine chemotherapy use and, moreover, consideration in been found to predict both progression-free survival and
patients with larger primary tumors with no nodal spread.3,4 overall survival. Preclinical data have suggested that ICIs

Including Checkpoint Inhibitors for Patients With Locally Advanced Lung Cancers
administered before chemotherapy in preclinical lung can- Combining ICIs and Surgery
cer models result in better outcomes than the same ther- Much of the challenge of determining the optimal treatment
apies administered after surgery.13,17,18 Investigators have of stage III disease is the heterogeneity of the disease.
theorized that an intact primary tumor and draining lymph Available approaches, highlighting those that include sur-
nodes, sites where key processes of immune response gery, are listed in Table 2. In addition, the definition of
initiation and antigen encounter occur, may achieve optimal resectable versus unresectable is variable, making com-
results with ICIs. Neoadjuvant anti–PD-1- (nivolumab) and parisons between trials difficult. Numerous randomized
anti–PD-L1- (atezolizumab) targeted monoclonal antibodies trials during the past 20 years have tried to address the
alone produced surprisingly better results compared with optimal approach to patients with stage III disease (con-
those observed in patients with metastatic lung cancers who current chemotherapy and radiation [CRT] alone, CRT →
received ICIs.10,12,13 When neoadjuvant ICIs were combined surgery, or chemotherapy → surgery → 6 radiation therapy
with cytotoxic chemotherapies, results were nothing short of [RT]). For the most part, these trials have not demonstrated
astonishing, with MPR rates of 57% and 80%.11,14 These a clear benefit of adding surgery to CRT except in a subset of
results have led to multiple phase III trials comparing patients with stage III cancers referred to as resectable (T3/
chemotherapy alone with the combination of chemotherapy 4N0-1 or N2 nonbulky single station), in which improved
plus atezolizumab, durvalumab, nivolumab, and pem- locoregional control may be important.20 The dramatic re-
brolizumab—all agents administered before surgery. Trials sults of the PACIFIC trial with 1 year of durvalumab after CRT
administering the same agents after chemotherapy post- compared with CRT alone has created a new standard of
operatively (adjuvantly) are also in progress. By design, no care for patients with unresectable stage III disease who do
early results are available, and the readouts will take longer. not experience progression after CRT.21,22 Of interest, most
No trials comparing neoadjuvant with adjuvant use of the of the benefit in the PACIFIC trial seemed to address the
same agent have been attempted. In addition to the pre- greatest need in the field—the reduction of systemic re-
clinical data that favor neoadjuvant use of ICIs, there are lapse, with longer time to death or new metastases.21 Given
many advantages of neoadjuvant approaches (Sidebar 1), the landmark results of PACIFIC, the role of surgery in in-
not the least of which is a years-earlier demonstration of dividuals with stage III non–small cell lung cancers (NSCLCs)
long-term benefit on the basis of MPR rates.16,19 In all trials of must be re-evaluated.
neoadjuvant ICIs, no new perioperative toxicities emerged.
Accumulated data detailed below demonstrate that ICIs ad- Role of Surgery to Enhance Locoregional Control in
ministered alone or with chemotherapy lead to high rates of Patients With Stage III NSCLCs
MPR—historically associated with better progression-free Investigators at The University of Texas MD Anderson
survival and overall survival—in patients with stage IB to Cancer Center recently reviewed their experience with stage
IIIB lung cancers. We are years away from determining IIIA (N2) lung cancers treated from 2004 to 2014, in which
whether adjuvant ICIs are more effective as a perioperative 159 patients with resectable lung cancers with documented
strategy than in their neoadjuvant use. N2 nodes underwent induction chemotherapy followed by
TABLE 1. Phase II Trials to Determine pCR and MPR After Neoadjuvant ICI or ICI Plus Chemotherapy in Patients With Resectable and Operable Stage I–III
Lung Cancers
MPR, % pCR, %
Trial Author(s) Agent(s) (95% CI) (95% CI)
Johns Hopkins University/Memorial Sloan Kettering Cancer Center Forde et al10 Nivolumab 43 (21 to 14 (4 to 34)
(21 patients) 66)
LCMC3; Lung Cancer Mutation Consortium (82 patients) Kwiatkowski et al12 Atezolizumab 18 (11 to 5 (2 to 12)
28)
NEOSTAR; MD Anderson Cancer Center (44 patients) Cascone et al13 Nivolumab 6 ipilimumab 25 (14 to 18 (9 to 32)
40)
NADIM; Spanish Lung Cancer Group (30 patients) Provencio-Pulla Nivolumab + paclitaxel/ 80 (64 to 75 (4 to 76)
et al14 carboplatin 91)
Columbia University New York/MGH (30 patients) Shu et al11 Atezolizumab + paclitaxel/ 57 (36 to 33 (18 to
carboplatin 76) 52)
Duke/Dartmouth/Mayo Clinic (25 patients) Ready et al15 Pembrolizumab 28 (12 to 8 (1 to 26)
49)
Abbreviations: ICI, immune checkpoint inhibitor; MPR, major pathologic response; pCR, pathologic complete response.

Kris et al
SIDEBAR 1. FACTORS FAVORING NEOADJUVANT THERAPY

• Attacks micrometastases, the primary cause of relapse, at the earliest time
• Preservation of primary tumor and draining nodes can provide an intact immune environment when using checkpoint
inhibitors
• Ability to assess sensitivity to agents planned for adjuvant use for individual patients and as part of drug development
• Opportunity to assess remaining tumor cells at maximal response (persisters)
• Ability to assess sensitivity of agents used in induction; you can change the regimen
• Complete and major pathologic response and outcome surrogate
• Better chemotherapy drug delivery and tolerability
• Better compliance with subsequent therapies
• Time to identify unsuspected metastases and comorbidities before local therapy
surgery with or without adjuvant RT (40% of patients). The immunoreactive T cells through dendritic antigen pre-
5-year overall survival was 50%, with 15% locoregional and sentation at the organ site of antigen presentation.18,24 In
44% distant recurrences.23 During that same period, 336 preclinical models, Cascone et al13 and others demon-
individuals with unresectable stage IIIA disease underwent strated in mice that neoadjuvant immunotherapy was more
definitive CRT, with a 5-year overall survival rate of 29% and effective than adjuvant treatment.17 Surrogate endpoints,
with 30% locoregional and 41% distant recurrence.23 These such as pathologic complete response (pCR) in resection
two groups represent distinct patient populations, but the specimens or MPR (, 10% viable tumor) may also allow
data suggest that better locoregional control can be ob- a more rapid readout of long-term benefits than can be
tained by incorporating surgery with chemotherapy with or anticipated in adjuvant studies.16,25 Several phase II studies
without RT. Ongoing clinical trials will demonstrate whether listed in Table 1 have evaluated this neoadjuvant approach.
the addition of immunotherapy to a surgical strategy can Forde et al10 noted no increase in adverse events after
improve distant control, as demonstrated in patients who surgery, with MPR rates of 45% after two cycles of nivo-
were treated with concurrent CRT in the PACIFIC study. lumab. There was also evidence of biologic effect, with
increased T-cell infiltration in the tumor and peripheral
Role of Surgery With Immunotherapy blood and increased neoantigen-specific T-cell clones from
Successes with immunotherapy in metastatic disease and the primary tumor. NEOSTAR, conducted at MD Anderson
in the PACIFIC trial have led to efforts to improve survival in Cancer Center, randomly assigned patients with stage IB to
earlier-stage NSCLC by incorporating ICIs with surgery. IIIA NSCLC to three cycles of nivolumab or nivolumab and
These efforts have focused on adjuvant and neoadjuvant ipilimumab. MPR rates were higher with the combination
approaches with surgery. Adjuvant approaches to date treatment (44% vs. 19%), as was T-cell infiltration in the
include four randomized trials using immunotherapy. The tumor.13 LCMC3 is a 180-patient phase II trial evaluating
ANVIL trial (EA5142) in ALCHEMIST randomly assigned neoadjuvant atezolizumab for two cycles in patients with
patients with resected stage IB to IIIA NSCLC to adjuvant resectable stage IB to IIIA NSCLC. Preliminary results have
nivolumab—every 4 weeks for 1 year—or observation after suggested no increased toxicity. The MPR rate in the initial
standard adjuvant therapy (NCT02595944; 903 patients). 82 patients was 18%, with a pCR rate of 4%.12 The range of
This trial completed enrollment in 2019 but is not expected MPR rates from these neoadjuvant studies is 18% to 43%,
to read out its disease-free survival primary endpoint for which is encouraging when compared with MPR rates of
several years, highlighting the challenge with adjuvant 19% reported in earlier trials of neoadjuvant chemotherapy
studies that endpoints of overall survival and progression- alone.16,25,26
free survival can take up to a decade to assess. In addition,
the subset of patients with each stage of the disease may not Role of Surgery With Chemoimmunotherapy
be large enough to make clear determinations of the benefit
Chemoimmunotherapy has demonstrated superiority com-
of adjuvant ICIs by stage.
pared with chemotherapy alone in patients with metastatic
The neoadjuvant approach represents another strategy by lung cancers.7 This has led to interest in assessing the role of
which to incorporate ICIs with surgery. Biologically, this chemoimmunotherapy with surgery in earlier-stage NSCLC.
approach may be more effective than the adjuvant strategy, In the adjuvant setting, ALCHEMIST is launching a new trial
because the presence of intact tumor-draining lymph nodes designed to evaluate adjuvant chemoimmunotherapy. This
at the time of ICI treatment may allow for better neoantigen study (NCT04267848; 1,263 patients; disease-free survival
presentation to dendritic cells and the development of endpoint), set to open in the United States in 2020, will

TABLE 2. Current and Future Strategies for Patients With Stage III Adenocarcinomas and Squamous Cell Carcinomas
Current Treatment (requires multidisciplinary Future Strategies (requires additional clinical trial
Stage III Subset discussion) assessment)
Unresectable CRT → durvalumab CIRT → I
Large T and N3 or bulky, multistation or or
N2
Stage IIIA–IIIC CRT alone (ICI contraindication) IRT → I
or
CIRT → S → I
CI → S → RT (to LNs) → I
I → S → RT (to LNs) → I
or
CRT alone (immunocontraindication)
Unresectable CRT → durvalumab CI → RT → I
Small T, PD-L1 IHC or immunotherapy or or
biomarker high, multistation N2
Stage IIIA CRT alone (ICI contraindication) I → RT → I
or
I → S → RT (to LNs) → I
Unresectable CRT → durvalumab CRT → I
Small T, PD-L1 IHC or immunotherapy or
biomarker low, multistation N2
Stage III A CRT alone (ICI contraindication)
Resectable C → S → 6 RT CIRT → S → I
Large T or CI → S → RT (to LNs) → I
N0-1, single-station N2 CRT → S (sup sulcus) I → S → RT (to LNs) → I
Stage III A or or
CRT alone (poor PS, etc.) CRT → I
or
CRT alone (immunocontraindication)
Resectable C → S → 6 RT CIRT → S → I
Small T, single-station N2 or CI → S → RT (to LNs) → I
Stage IIIA CRT → S I → S → RT (to LNs) → I
or or
CRT alone (poor PS, need for pneumonectomy) CRT → I
or
CRT alone (poor PS, immunocontraindication)
Abbreviations: C, chemotherapy; CI, chemoimmunotherapy; CIRT, concurrent chemoradiation and immunotherapy; CRT, concurrent chemoradiation; I,
immunotherapy; ICI, immune checkpoint inhibitor; IHC, immunohistochemistry; IRT, immunotherapy and radiation therapy; LN, lymph node; PS,
performance status; RT, radiation therapy; S, surgery.
randomly assign patients with resected stage IB to IIIA immunologic response that targets micrometastases.18,24
NSCLCs of any histology to one of three arms: standard From this standpoint, a neoadjuvant approach may be more
adjuvant therapy with a platinum doublet, standard adjuvant efficacious than adjuvant because of the presence of the
therapy with a platinum doublet followed by pembrolizumab primary tumor during treatment with chemoimmunotherapy
(for 17 cycles total), or a standard adjuvant platinum doublet (Sidebar 1). The NADIM trial evaluated neoadjuvant
with concurrent pembrolizumab for four cycles followed by nivolumab, paclitaxel, and carboplatin in patients with
12 additional cycles of pembrolizumab (17 doses total).
stage IIIA resectable NSCLC followed by surgical resection
One potential benefit of chemoimmunotherapy may be an and adjuvant nivolumab for 1 year. Preliminary observations
increase in immunogenic tumor-specific peptides, or neo- include no increased toxicity and a high rate of MPR (80%)
antigens, released from the chemotherapy-induced cell and pCR (75%) in resected specimens.14 Another study
death of the primary tumor. This might then lead to increases from Columbia University using neoadjuvant atezolizumab
in primed neoantigen-specific T cells and a more robust and chemotherapy has also found encouraging MPR and

Kris et al
pCR rates of 57% and 33%, respectively, in resected (NCT04133337; 20 patients; MPR endpoint).28 The neo-
specimens.11 These dramatic phase II pathologic re- adjuvant platform provides the opportunity to rapidly test many
sponse results have led to worldwide interest and the combinations, using MPR or pCR as a surrogate.16 Many
development of phase III trials: IMpower 030 is evaluating investigators have proposed using ctDNA in the future to
atezolizumab and chemotherapy versus chemotherapy quantitate minimal residual disease, and its presence can
alone (NCT03456063; 374 patients; endpoints of MPR, serve as a trigger to intensify therapy if ctDNA is still detectable
event-free survival), the Aegean trial is evaluating dur- after the completion of multimodality therapy.29
valumab with chemotherapy versus chemotherapy alone
Personalized Approach to Stage III Disease
(NCT03800134; 300 patients; MPR endpoint), and
KEYNOTE-671 is evaluating pembrolizumab and chemo- Although the PACIFIC trial established a new standard of
therapy versus chemotherapy alone (NCT03425643, 786 care for patients with unresectable stage III lung cancers in
patients; endpoints of event-free survival, overall survival). general, surgery remains a critical component of multi-
The Lung Cancer Mutation Consortium is initiating a 1,000- modality care for many individuals, especially those with
patient screening trial (LCMC4) to detect eight oncogenic stage IIIA disease. Patients with an Eastern Cooperative
drivers (EGFR, ALK, MET, BRAF, RET, NTRK, ROS1, and Oncology Group performance status of 0 or 1, T3-4N0-1
HER2) in patients being evaluated for neoadjuvant thera- disease, or single-station nonbulky N2 tumors should be
pies.19 Individuals identified with early-stage tumors with considered for CRT → surgery or chemotherapy → surgery
oncogenic drivers—approximately one-third of patients— → 6 RT after multidisciplinary review. In the future, if the
will be directed to industry-sponsored trials of targeted impressive results observed in the completed phase II trials
therapies matched to the oncogenic driver detected. (Table 1) are confirmed in larger series and eventually in
phase III studies, patients with stage III lung cancers may
Role of Surgery With Immunotherapy and CRT routinely receive ICIs in addition to surgery. The neo-
Ultimately, the effectiveness of checkpoint inhibition may adjuvant approach provides the advantage of assessing
also depend on the immunogenic environment in and tumor response at the time of surgery, permitting the use of
around the primary tumor. Nonimmunogenic environments different treatments in the adjuvant setting for patients with
or cold tumors may be less likely to respond to ICIs. These nonresponding disease or those with responding disease
environments could be enhanced not only with chemo- with less than complete regressions. This success strategy
therapy but also with the addition of RT to increase the has been demonstrated recently in patients with HER2-
neoantigen load in the tumor microenvironment. This driven breast cancers with present disease after trastuzu-
concept has led to a neoadjuvant strategy that combines mab.30 One could even imagine adding in an adjuvant
immunotherapy with CRT followed by surgery and con- fashion different immune-enhancing drugs, chimeric anti-
solidative immunotherapy. The optimal timing sequence and gen receptor T cells, chimeric antigen receptor natural killer
potential morbidity of such a strategy is being evaluated in cells, or tumor-infiltrating lymphcytes derived from the
several ongoing trials. The Hoosier Cancer Research Network resected primary tumor in ICI-nonresponding disease. Many
is evaluating the feasibility of durvalumab and CRT followed groups are pursuing the development of blood or tissue
by surgery and adjuvant durvalumab (NCT03871153; 25 biomarkers to better select optimal therapies, including
patients; endpoints of safety, pCR), while Case Compre- surgical resection, in stage III disease.
hensive Cancer Center is evaluating pembrolizumab and
Patient case 1 (Sidebar 2; Fig. 1) demonstrates the con-
CRT followed by surgery and adjuvant pembrolizumab
cepts surrounding the place of surgery in the management
(NCT029879998; 20 patients; safety endpoint).
of patients with locally advanced lung cancers.
Role of Surgery With Immunotherapy and Novel This case is not straightforward with regard to the initial
Immune-Enhancing Agents decision to not include surgery and to proceed with CRT
Novel immune-enhancing agents are also being evaluated followed by durvalumab. The single N2 station is considered
in the neoadjuvant setting in the hopes of rapidly identifying by many to be borderline bulky, with a short axis of 18 mm
effective combinations. NEOCOAST is evaluating neo- and a long axis of 26 mm. Locoregional recurrence high-
adjuvant durvalumab alone or with oleclumab, danvatirsen, lights the challenge of choosing the best local therapy if ICIs
or monalizumab (NCT03794544; 160 patients; MPR are able to control distant recurrences. The decision to
endpoint). Canopy N is evaluating neoadjuvant pem- proceed with resection 8 months after CRT while still on
brolizumab and canakinumab (an anti–interleukin-1B an- durvalumab is likely associated with an increased risk of
tibody) versus canakinumab alone (NCT03968419; 110 pulmonary toxicity after chemotherapy, radiation, and
patients; MPR endpoint).27 Neoadjuvant SHR1210, an durvalumab. The increased fibrosis at the hilum and right
anti–PD-1 antibody, is also being evaluated with apatinib, upper lobe can be appreciated on the CT-PET scan
a tyrosine kinase inhibitor with selective inhibition of VEGFR2 (8 months post-CRT).

Combining ICIs and RT as well as to induce the release of undesirable cytokines and
There is a strong rationale for combining immunotherapy chemokines that further recruit suppressive immune ef-
and RT. More than 100 years after the discovery of ionizing fector cells or impair T-cell trafficking.38,39 It is prudent to
radiation, we now know that RT elicits immune interactions consider that these negative immune consequences may
that can synergize with systemic therapies, particularly outweigh the positive, leading to RT-resistant disease.
immunotherapeutics. Our knowledge and understanding of Biomarkers are needed to help in patient selection for
how RT affects the tumor microenvironment is growing—its combination therapy with ICIs.
ability to induce immunogenic cell death, release tumor Results from the PACIFIC trial mark the first change in the
antigens, induce the cGAS-STING pathway, upregulate management of inoperable and unresectable stage III lung
major histocompatibility complex-I expression, stimulate cancers in more than 2 decades.40 The addition of induction
type-I interferons, and promote CD8+ T-cell infiltration.31-33 or consolidation chemotherapy to CRT and dose escalation
These immune effects provide the rationale that RT can did not lead to improved outcomes.41 The phase III PACIFIC
deliver the kickstart required to improve immune-mediated trial has demonstrated that the ICI durvalumab, administered
tumor control with ICIs. Furthermore, the ability of RT to for 1 year after CRT, improves both median progression-free
induce PD-L1 expression on tumor cells provides an addi- survival (increase by 12 months, to 17.2 months) and 3-year
tional benefit: The combination with ICI may negate adaptive overall survival (increase by 13%, to 57%) in patients with
resistance mechanisms.34 inoperable or unresectable stage III lung cancers.21,42
In locally advanced NSCLCs, CRT forms the mainstay of Furthermore, the addition of durvalumab after chemo-
treatment. Combining ICIs with CRT in this patient pop- therapy was well tolerated. This trial was the first demon-
ulation is an attractive strategy. Preclinical studies of RT/ICI stration of improved survival compared with CRT in more
combinations have demonstrated improved outcomes, than 2 decades and has established durvalumab after CRT
primarily in the setting of the ICI being administered con- as a standard of care. It is important to highlight that patients
currently with RT.34-37 However, we are coming to realize in PACIFIC were eligible if they had unresectable disease,
that RT can also have negative immune consequences in had experienced a response to CRT, had recovered from
unselected patients. These include the ability of RT to adverse effects (Common Terminology Criteria for Adverse
promote the introduction of myeloid-derived suppressor Events grade 2), and had a performance status of 0 to 1.
cells and regulatory T cells in the tumor microenvironment An important consideration, therefore, is to use state-of-the-
SIDEBAR 2. PATIENT CASE 1: STAGE IIIA, RIGHT UPPER LOBE SQUAMOUS CELL LUNG CANCER WITH A SINGLE BULKY
IPSILATERAL MEDIASTINAL LYMPH NODE (N2)
• 64-year-old man with a history of smoking
• Presented with right upper lobe mass noted on chest x-ray
• Past medical history: high blood pressure, no coronary artery disease
• WHO performance status of 1
• Pulmonary function tests: FEV1, 81% predicted; DLCO, 86% predicted
• CT of thorax, abdomen: right upper lobe tumor and enlarged station 4R (25.6 mm 17.4 mm)
• EBUS: positive 4R, 11R; negative 7, 4L, 10R
• PET-CT: FDG right hilar adenopathy lymph node, T2N2M0
• MRI of brain: no metastases
• Treatment: intensity-modulated RT, 66 Gy, concurrent with paclitaxel and carboplatin
• 4 weeks post-CRT:
• Decision to start durvalumab 5 weeks after the completion of CRT for a duration of 12 months
• 2 months post-CRT (on durvalumab):
• PET-CT: decreased avidity right upper lobe mass and right hilar adenopathy
• 8 months post-CRT (on durvalumab):
• PET-CT: no FDG avidity in right upper lobe mass, increased avidity right hilar adenopathy, no metastases
• EBUS: positive 4R; negative 7, 4L
• Referred for surgical evaluation
• Pulmonary function tests: FEV1, 71% predicted; DLCO, 74% predicted
• MRI brain: no metastases
• Treatment: robotic resection

Kris et al
FIGURE 1. Patient Case 1: CT of Thorax and CT-PET

(A) Initial CT and CT-PET. (B) CT-PET 2 months post-chemoradiotherapy (CRT) on durvalumab. (C) CT-PET 8 months post-CRT on durvalumab.
art radiation treatment to reduce the dose delivered to durvalumab, as the patient would have met the inclusion
normal tissues and, consequently, the adverse effects of criteria of the PACIFIC study. However, a number of
CRT. This can be achieved by using four-dimensional CT questions remain unanswered regarding the application of
scanning for RT planning, which results in the use of smaller this treatment (Sidebar 4).
margins around the gross tumor volume; using intensity-
Several important points should be considered when
modulated RT as a treatment delivery technique that adds
interpreting the results of the PACIFIC trial. First, the
fluence modulation to beam shaping, which improves RT
PACIFIC population may not be fully representative of the
dose conformity around the tumor and spares surrounding
real world, as only 45% of patients were older than age
normal structures; and using daily online imaging for
65 years. Second, no data were collected at the time of
treatment verification.
random assignment on disease volume, radiation dose
Patient case 2 (Sidebar 3; Fig. 2) is a typical case of a patient delivered to organs at risk, and RT techniques used. There
who is suitable for durvalumab after CRT. This case is is uncertainty about whether patients who present with
straightforward with regard to the decision to treat with large-volume disease and with radiation doses to the
SIDEBAR 3. PATIENT CASE 2

• 71-year-old woman
• Presented with shortness of breath
• Past medical history: high blood pressure, ischemic heart disease 10 years ago
• WHO performance status of 1
• Pulmonary function tests: FEV1, 80% predicted; KCO, 105% predicted
• CT of thorax, abdomen: right upper lobe tumor and enlarged station 4R, 7 lymph nodes T3N2M0
• Mediastinoscopy: station 4R adenocarcinoma PD-L1 10%
• PET-CT: FDG-avid right supraclavicular lymph node, T3N3M0
• MRI of brain: clear
• Treatment: intensity-modulated RT, 66 Gy in 33 fractions, concurrent with cisplatin/etoposide for two cycles starting day 1 of
radiotherapy
• Common Toxicity Criteria for Adverse Events, v5.0, grade 3 esophagitis, performance status of 2
• Common Toxicity Criteria for Adverse Events, v5.0, grade 2 esophagitis, performance status of 1
• Decision to start durvalumab 5 weeks after the completion of CRT for a duration of 12 months

organs at risk at the limit of tolerance can be safely ad- treated with sequential chemotherapy followed by RT also
ministered durvalumab. Third, subgroup analyses sug- remains to be determined and is currently under in-
gested that progression-free survival and overall survival vestigation in PACIFIC-6 (NCT03693300).
were superior if durvalumab was delivered fewer than
To make additional progress, we need a better under-
14 days after the completion of CRT. However, this does
standing of the biology of lung cancers to identify which
not mean that the patient should be treated as soon as
patients will benefit from CRT/ICI combinations. One key
possible after completion of CRT in routine practice. It is
issue to address is the timing of RT. When is RT likely to have
likely that these patients who had recovered from the
a detrimental immune-priming effect, preventing appro-
adverse effects of CRT within 2 weeks had small-volume
priate T-cell infiltration into the tissue microenvironment and
disease at some distance from the mediastinum. Finally,
instead promoting a suppressive immunophenotype? Per-
tissue collection was not mandatory for trial entry. Con-
haps, in patients whose tumor biopsies suggest an already
sequently, tissue samples were available for only 63% of
suppressive tissue microenvironment (low T-cell and high
patients, with only 148 of 713 patients having PD-L1 ex-
myeloid infiltrate), an alternative RT/-ICI approach is
pression of less than 1%. There is no guarantee, therefore,
needed, whereby ICIs help reprogram immune effector cells
that the tumor samples are missing completely at random,
to allow improved T-cell infiltration and response to ICIs
potentially biasing the interpretation of the results on the
(e.g., TLR and CD40 agonist antibodies).
basis of PD-L1 expression.43 PACIFIC also left many
questions unanswered (Sidebar 4) for individuals with To explore these issues, biomarker evaluations should be
stage III lung cancers. incorporated into future clinical trials. Minimal residual
disease as assessed by ctDNA can predict survival in pa-
Other trials have evaluated the administration of the ICIs
tients undergoing curative-intent treatment and has also
pembrolizumab, atezolizumab, and nivolumab with CRT, as
been linked to improved response to ICIs.47,48 In addition,
outlined in Table 3.35,36 Investigators have theorized that
although we know that tumor mutational burden correlates
cellular apoptosis from CRT leads to increased tumor-
to response to ICIs, the ratio of clonal-to-subclonal muta-
derived neoantigens, increased antigen presentation, and
tions may be important for long-term response. RT is ca-
improved response to ICIs. Preclinical data suggest that PD-
pable of generating tumor antigens and immunogenic cell
1 blockade delivered either concurrently with RT or just
death, but recent data suggest that the generation of po-
afterward is superior to sequential PD-1 blockade.34 Both of
tential subclonal versus clonal mutations after RT needs
these studies have demonstrated that the approach is
careful study.49
feasible, with no dramatic increase in the rate of treatment-
related pneumonitis compared with CRT alone (Table 3).
Whether this latter approach improves outcomes compared CONCLUSION
with ICIs administered after radiation requires additional The exciting data that emerged clearly establish a role for ICIs
study. The benefit of adding an ICI in patients who were with surgery, radiation, and chemotherapy. In individuals for
SIDEBAR 4. QUESTIONS NOT ANSWERED BY THE PACIFIC TRIAL (TESTING DURVALUMAB AFTER THE COMPLETION OF
CONCURRENT CHEMORADIATION IN PATIENTS WITH UNRESECTABLE AND INOPERABLE STAGE III LUNG CANCERS)
• Applicability of the PACIFIC data to other population of patients?
• Resectable disease
• Performance status of 2
• Elderly
• Individuals receiving sequential chemoradiotherapy
• Patients with EFGR L858R and exon 19 deletion mutations
• Can durvalumab be safely prescribed in patients who have received a high radiation dose to thoracic organs at risk (e.g.,
lung)?
• Can we create precise definitions of operable and resectable? How is this assessment best obtained?
• How soon should durvalumab start after CRT?
• What is the optimal duration of durvalumab?
• What is the optimal timing durvalumab and CRT: concomitant or sequential?
• Should we treat patients with durvalumab if PD-L1 is , 1%?
• Do biomarkers exist to better select patients?

Kris et al
FIGURE 2. Patient Case 2

(A) CT thorax with intravenous
(IV) contrast. (B) Radiotherapy
plan.
TABLE 3. Trials With ICIs During or After CRT in Stage III Lung Cancer
Author and Trial Name
Antonia et al22; Antonia et al22; Durm et al44; Peters et al45; ETOP Lin et al46;
Variable PACIFIC PACIFIC HCRN NICHOLAS DETERRED
Agent Durvalumab Placebo Pembrolizumab Nivolumab Atezolizumab
No. of patients 473 236 93 80 40
Timing of ICI After CRT — After CRT During CRT During CRT
18-month progression-free 44 27 50 Not reported Not reported
survival, %
Any pneumonitis, % 34 25 Not reported 43 25
Grade 3 pneumonitis, % 3 3 6 10 3
Deaths on study, % 4 6 3 9 5
Abbreviations: CRT, concurrent chemoradiation; ICI, immune checkpoint inhibitor.
whom data support the use of either surgical resection or resection), radiation technique (intensity-modulated RT vs.
CRT, we find ourselves in a virtually data-free zone, with no protons), and radiation dose. Our best advice is to consider all
direct comparisons between the two local approaches. In approaches in a frank and open multimodality discussion
addition to the local therapy questions, we have only limited that includes a careful assessment of individual patient at-
data to assist practitioners caring for patients with locally tributes and preferences.
advanced lung cancers on the selection of specific che-
motherapeutic agents, the selection of ICIs, and—when ACKNOWLEDGMENT
surgery is the chosen modality—which treatments should be Supported by a grant from the National Institute of Health
administered adjuvantly or neoadjuvantly. Additional ques- Research Manchester Biomedical Research Centre (C.F.-F.)
tions remain after deciding between surgery or CRT, such as and in part through National Cancer Institute Cancer Center
the extent of surgical resection (lobectomy or limited Support Grant No. P30-CA008748 (M.G.K., J.C., and J.L.).

1
Memorial Sloan Kettering Cancer Center, Weill Cornell Medical College, Mark G. Kris, MD, Memorial Sloan Kettering Cancer Center, 530 East
New York, NY 74th St., New York, NY 10021; email: [email protected].
2
The University of Manchester, The Christie NHS Foundation Trust,
Institute of Cancer Sciences, Manchester, United Kingdom
3
The University of Texas MD Anderson Cancer Center, Houston, TX

REFERENCES
1. Rami-Porta R, Asamura H, Travis WD, et al. Lung cancer: major changes in the American Joint Committee on Cancer eighth edition cancer staging manual. CA
Cancer J Clin. 2017;67:138-155.
3. Kris MG, Gaspar LE, Chaft JE, et al. Adjuvant systemic therapy and adjuvant radiation therapy for stage I to IIIA completely resected non–small cell cancers:
American Society of Clinical Oncology/Cancer Care Ontario clinical practice guideline update. J Clin Oncol. 2017;35:2960-2974.
4. Ettinger DS, Wood DE, Aggarwal C, et al; OCN. NCCN guidelines insights: non–small cell lung cancer, version 1.2020. J Natl Compr Canc Netw. 2019;
17:1464-1472.
5. Tsutani Y, Imai K, Ito H, et al. Adjuvant chemotherapy for pathological stage I non–small cell lung cancer with high-risk factors for recurrence: a multicenter study.
6. Reck M, Rodrı́guez-Abreu D, Robinson AG, et al; KEYNOTE-024 Investigators. Pembrolizumab versus chemotherapy for PD-L1–positive non–small cell lung
7. Gandhi L, Rodrı́guez-Abreu D, Gadgeel S, et al; KEYNOTE-189 Investigators. Pembrolizumab plus chemotherapy in metastatic non–small cell lung cancer.
N Engl J Med. 2018;378:2078-2092.
8. Hellmann MD, Paz-Ares L, Bernabe Caro R, et al. Nivolumab plus ipilimumab in advanced non–small cell lung cancer. N Engl J Med. 2019;381:2020-2031.
9. Garon EB, Hellmann MD, Rizvi NA, et al. Five-year overall survival for patients with advanced non–small cell lung cancer treated with pembrolizumab: results from
the phase I KEYNOTE-001 study. J Clin Oncol. 2019;37:2518-2527.
11. Shu CA, Gainor JF, Awad MM, et al. Neoadjuvant atezolizumab and chemotherapy in patients with resectable non–small cell lung cancer: a single-arm, phase 2
trial. Lancet Oncol. In press.
12. Kwiatkowski DJ, Rusch VW, Chaft JE, et al. Neoadjuvant atezolizumab in resectable non–small cell lung cancer (NSCLC): interim analysis and biomarker data
from a multicenter study (LCMC3). J Clin Oncol. 2019;37 (suppl; abstr 8503).
13. Cascone T, William WN, Weissferdt A, et al. Neoadjuvant nivolumab (N) or nivolumab plus ipilimumab (NI) for resectable non–small cell lung cancer (NSCLC):
clinical and correlative results from the NEOSTAR study. J Clin Oncol. 2019;37 (suppl; abstr 8504).
14. Provencio-Pulla M, Nadal-Alforja E, Cobo M, et al. Neoadjuvant chemo/immunotherapy for the treatment of stages IIIA resectable non–small cell lung cancer
(NSCLC): a phase II multicenter exploratory study—NADIM study-SLCG. J Clin Oncol. 2018;36 (suppl; abstr 8521).
15. Ready N, Tong B, Clarke J, et al. Neoadjuvant pembrolizumab in early stage non–small cell lung cancer (NSCLC): toxicity, efficacy, and surgical outcomes.
J Thorac Oncol. 2019;14 (suppl; abstr P2.04).
16. Hellmann MD, Chaft JE, William WN Jr., et al; University of Texas MD Anderson Lung Cancer Collaborative Group. Pathological response after neoadjuvant
chemotherapy in resectable non–small cell lung cancers: proposal for the use of major pathological response as a surrogate endpoint. Lancet Oncol. 2014;
15:e42-e50.
17. Liu J, Blake SJ, Yong MC, et al. Improved efficacy of neoadjuvant compared to adjuvant immunotherapy to eradicate metastatic disease. Cancer Discov. 2016;
6:1382-1399.
18. Topalian SL, Taube JM, Pardoll DM. Neoadjuvant checkpoint blockade for cancer immunotherapy. Science. 2020;367:eaax0182.
19. Blumenthal GM, Bunn PA Jr., Chaft JE, et al. Current status and future perspectives on neoadjuvant therapy in lung cancer. J Thorac Oncol. 2018;13:1818-1831.
20. Majem M, Hernandez-Hernandez J, Hernando-Trancho F, et al. Multidisciplinary consensus statement on the clinical management of patients with stage III
non–small cell lung cancer. Clin Transl Oncol. 2020;22:21-36.
21. Antonia SJ, Villegas A, Daniel D, et al; PACIFIC Investigators. Overall survival with durvalumab after chemoradiotherapy in stage III NSCLC. N Engl J Med. 2018;
379:2342-2350.
22. Antonia SJ, Villegas A, Daniel D, et al; PACIFIC Investigators. Durvalumab after chemoradiotherapy in stage III non–small cell lung cancer. N Engl J Med. 2017;
377:1919-1929.
23. Rajaram R, Correa AM, Xu T, et al. Locoregional control, overall survival, and disease-free survival in stage IIIA (N2) non–small cell lung cancer: analysis of
resected and unresected patients. Clin Lung Cancer. Epub 2020 Jan 27.
24. Sharma P, Allison JP. The future of immune checkpoint therapy. Science. 2015;348:56-61.
25. Pataer A, Kalhor N, Correa AM, et al. Histopathologic response criteria predict survival of patients with resected lung cancer after neoadjuvant chemotherapy.
J Thorac Oncol. 2012;7:825-832.
26. William WN Jr., Pataer A, Kalhor N, et al. Computed tomography RECIST assessment of histopathologic response and prediction of survival in patients with
resectable non–small cell lung cancer after neoadjuvant chemotherapy. J Thorac Oncol. 2013;8:222-228.
27. Ridker PM, MacFadyen JG, Thuren T, et al; CANTOS Trial Group. Effect of interleukin-1β inhibition with canakinumab on incident lung cancer in patients with
atherosclerosis: exploratory results from a randomised, double-blind, placebo-controlled trial. Lancet. 2017;390:1833-1842.
28. Xu JM, Zhang Y, Jia R, et al. Anti-PD-1 antibody SHR-1210 combined with apatinib for advanced hepatocellular carcinoma, gastric or esophagogastric junction
cancer: an open-label, dose escalation and expansion study. Clin Cancer Res. 2019;25:515-523.

Kris et al
29. Chaudhuri AA, Chabon JJ, Lovejoy AF, et al. Early detection of molecular residual disease in localized lung cancer by circulating tumor DNA profiling. Cancer
Discov. 2017;7:1394-1403.
2019;380:617-628.
31. Deng L, Liang H, Xu M, et al. STING-dependent cytosolic DNA sensing promotes radiation-induced type I interferon-dependent antitumor immunity in im-
munogenic tumors. Immunity. 2014;41:843-852.
32. Reits EA, Hodge JW, Herberts CA, et al. Radiation modulates the peptide repertoire, enhances MHC class I expression, and induces successful antitumor
immunotherapy. J Exp Med. 2006;203:1259-1271.
33. Gupta A, Probst HC, Vuong V, et al. Radiotherapy promotes tumor-specific effector CD8+ T cells via dendritic cell activation. J Immunol. 2012;189:558-566.
34. Dovedi SJ, Adlard AL, Lipowska-Bhalla G, et al. Acquired resistance to fractionated radiotherapy can be overcome by concurrent PD-L1 blockade. Cancer Res.
2014;74:5458-5468.
35. Twyman-Saint Victor C, Rech AJ, Maity A, et al. Radiation and dual checkpoint blockade activate non-redundant immune mechanisms in cancer. Nature. 2015;
520:373-377.
36. Herter-Sprie GS, Koyama S, Korideck H, et al. Synergy of radiotherapy and PD-1 blockade in Kras-mutant lung cancer. JCI Insight. 2016;1:e87415.
37. Gong X, Li X, Jiang T, et al. Combined radiotherapy and anti–PD-L1 antibody synergistically enhances antitumor effect in non–small cell lung cancer. J Thorac
Oncol. 2017;12:1085-1097.
38. Barker HE, Paget JT, Khan AA, et al. The tumour microenvironment after radiotherapy: mechanisms of resistance and recurrence [published correction appears
in Nat Rev Cancer. 2015;15:509]. Nat Rev Cancer. 2015;15:409-425.
39. Sampath S, Won H, Massarelli E, et al. Combined modality radiation therapy promotes tolerogenic myeloid cell populations and STAT3-related gene expression in
head and neck cancer patients. Oncotarget. 2018;9:11279-11290.
40. Aupérin A, Le Péchoux C, Rolland E, et al. Meta-analysis of concomitant versus sequential radiochemotherapy in locally advanced non–small cell lung cancer.
J Clin Oncol. 2010;28:2181-2190.
41. Bradley JD, Paulus R, Komaki R, et al. Standard-dose versus high-dose conformal radiotherapy with concurrent and consolidation carboplatin plus paclitaxel with
or without cetuximab for patients with stage IIIA or IIIB non–small cell lung cancer (RTOG 0617): a randomised, two-by-two factorial phase 3 study. Lancet Oncol.
2015;16:187-199.
42. Gray JE, Villegas A, Daniel D, et al. Three-year overall survival with durvalumab after chemoradiotherapy in stage III NSCLC: update from PACIFIC. J Thorac Oncol.
2020;15:288-293.
43. Peters S, Dafni U, Boyer M, et al. Position of a panel of international lung cancer experts on the approval decision for use of durvalumab in stage III non–small cell
lung cancer (NSCLC) by the Committee for Medicinal Products for Human Use (CHMP). Ann Oncol. 2019;30:161-165.
44. Durm GA, Althouse SK, Sadiq AA, et al. Phase II trial of consolidation pembrolizumab in patients with unresectable stage III non–small cell lung cancer: Hoosier
Cancer Research Network LUN 14-179. J Clin Oncol. 2018;36(suppl; abstr 8500).
45. Peters S, Felip E, Dafni U, et al. Safety evaluation of nivolumab added concurrently to radiotherapy in a standard first line chemo-radiotherapy regimen in stage III
non–small cell lung cancer: the ETOP NICOLAS trial. Lung Cancer. 2019;133:83-87.
46. Lin SH, Lin Y, Yao L, et al. Phase II trial of concurrent atezolizumab with chemoradiation for unresectable NSCLC. J Thorac Oncol. 2020;15:248-257.
47. Moding EJ, Liu Y, Nabet BY, et al. Circulating tumor DNA dynamics predict benefit from consolidation immunotherapy in locally advanced non–small cell lung
cancer. Nature Cancer. 2020;1:176-183.
48. Hellmann MD, Nabet BY, Rizvi H, et al. Circulating tumor DNA analysis to assess risk of progression after long-term response to PD-(L)1 blockade in NSCLC. Clin
Cancer Res. Epub 2020 Feb 11.
49. McGranahan N, Furness AJ, Rosenthal R, et al. Clonal neoantigens elicit T cell immunoreactivity and sensitivity to immune checkpoint blockade. Science. 2016;
351:1463-1469.

LUNG CANCER
Promising Immuno-Oncology Options for the

Future: Cellular Therapies and Personalized
Cancer Vaccines
Benjamin J. Solomon, MBBS, PhD1; Paul A. Beavis, PhD1; and Philip K. Darcy, PhD1
overview
A common pathway for an effective immune anticancer response involves recognition of tumor neoantigens
and subsequent targeting of cancer cells by T cells. In this article, we provide an overview of the current
status of two approaches to directly enhance this interaction using either adoptive cell therapy or personalized
cancer vaccines with focus on recent advances in solid tumors, including lung cancer.
INTRODUCTION This approach has also been explored in other solid

A common pathway for an effective immune anti- tumors where occasional responses have been iden-
cancer response involves recognition of tumor neo- tified and correlated with specificity of the infused
antigens and subsequent targeting of cancer cells by lymphocytes to tumor neoantigens. 6,7 In a patient
T cells. In this article, we provide an overview of the with colorectal cancer, TILs targeting an epitope on
current status of two approaches to directly enhance KRASG12D led to tumor regression,8 and TILs tar-
this interaction using either adoptive cell therapy or geting a specific mutation in the erbb2-interacting
personalized cancer vaccines with focus on recent protein (ERBB2IP) resulted in responses in a patient
advances in solid tumors, including lung cancer. Adop- with biliary duct cancer.7 In another study, infusion of
tive cell therapy involves the administration of immune TILs including HPV-directed T cells resulted in re-
cells with direct anticancer activity to patients. Adop- gression in two of nine patients with HPV-related
tive T-cell therapy using endogenous tumor-infiltrating cervical cancer,9,10 presumably because of T cells
lymphocytes (TILs) extracted from tumors from pa- specifically directed at viral antigens.
tients, or alternatively genetically engineered T cells In a preliminary report of a trial conducted in patients
collected from peripheral blood involving T-cell re- with advanced non–small cell lung cancer progressing
ceptor (TCR) or chimeric antigen receptor (CAR) T cells,1 on nivolumab, the use of TILs combined with the PD-1
has been investigated preclinically and in clinical trials inhibitor nivolumab seemed to have acceptable tox-
in patients with cancer. icity and demonstrated persistence of the infused
lymphocytes and encouraging activity with a reduc-
TARGETING CANCERS WITH ADOPTIVE TRANSFER tion in the size of lesions from seven of nine patients
OF TILS whose tumors progressed on single-agent nivolumab
(NCT03215810).11
TILs may be isolated from resected tumor tissue, se-
lected for tumor specificity, and expanded ex vivo in There are, however, significant challenges to the broad
media containing interleukin (IL)-2 and other cytokines applicability of TIL therapy. It can be challenging to
before reinfusion into patients. Lymphodepletion using collect, extract, and culture tumor-specific lympho-
Author affiliations a nonmyeloablative chemotherapy regimen (such as flu- cytes from many tumor types; IL-2 toxicity is prob-
and support lematic for many patients; there may be issues of poor
darabine/cyclophosphamide) before infusion of cells and
information (if
applicable) appear
subsequent IL-2 has been shown to increase the effec- localization and persistence of T cells after adoptive
at the end of this tiveness of this approach. The infused lymphocytes rec- transfer; and loss of expression of tumor peptide/
article. ognize specific antigens in autologous tumors from which major histocompatibility class (MHC) I may lead to re-
Accepted on April they are derived and can directly target tumors, causing sistance. Furthermore, TIL therapy requires significant
20, 2020 and regression.2 TIL therapy has been pioneered in patients time, expense, and resources, restricting the broad
published at
with advanced melanoma, where the approach has been applicability of this approach. To overcome some of
ascopubs.org on May
14, 2020: DOI https://
effective, resulting in objective responses in 55% of these limitations, strategies have been developed to
doi.org/10.1200/ patients3,4 and durable complete regressions in 24% of genetically engineer peripheral T cells to direct antigen
EDBK_281101 patients, with median survival of greater than 3 years.5 specificity using TCR or CAR T cells.1

Solomon, Beavis, and Darcy
responses, and resistance was seen in the context of defects

in the antigen presentation pathway, without dose-limiting
toxicities.16,17
• Cellular therapies using TILs as well as genet-
ically engineered T cells collected from ENGINEERED CAR T CELLS
peripheral blood (TCR or CAR T cells) are CAR T cells, which are autologous T cells transduced
under clinical trial evaluation in patients with
with CARs, have demonstrated remarkable efficacy in
solid cancers including lung cancer.
hematologic malignancies and are under investigation
• Novel personalized cancer vaccines targeting in solid cancers.1,18 CARs consist of an extracellular im-
tumor neoantigens are being tested alone or munoglobulin-derived heavy and light chain, which enable
with checkpoint inhibitors.
antigen targeting, linked to the TCR-activating domain (CD3ζ)
• These approaches provide strategies to via a transmembrane domain and costimulatory domains,
personalize immunotherapies for patients such as CD28 or 41BB, which enhance activation and in-
with cancer. crease persistence of CAR T cells.18 These constructs are
transduced into T cells collected from patients’ peripheral
blood after leukapheresis and expanded in vitro to produce
CAR T cells. The CAR T cells are reinfused into the pa-
TCR IMMUNOTHERAPY tient after administration of lymphodepleting chemotherapy,
TCR immunotherapy involves the use of lymphocyctes where they proliferate and are able to target cell surface
collected from a patient’s peripheral blood that are genet- antigens present on tumor cells that they are directed at
ically manipulated by insertion of α- and β-chain TCRs, independent of MHC restriction. The CAR T cells are capable
which are able to recognize specific antigens that are of self-amplifying after recognition of the target antigen, and
processed and presented by one of the patient’s own MHC each CAR T cell can identify, target, and kill many tumor
molecules. Unlike TIL therapy, where endogenous tumor- cells19 and persist, promoting immune surveillance to prevent
directed lymphocytes are selected, specific antigens (which tumor recurrence.
may be either extracellular or intracellular) are selected to In 2017, the U.S. Food and Drug Administration approved
be targeted by the modified TCR. These TCR T cells are two CD19-targeting CAR T cells, axicabtagene ciloleucel
reinfused into the patient and then proliferate and secrete and tisagenlecleucel, for relapsed and refractory acute
cytokines on antigenic stimulation and cause tumor cell lymphoblastic leukemia and for diffuse large B-cell lym-
lysis. The approach is limited by MHC restriction and phoma. These CAR T cells target CD19, a molecule that is
therefore can only be used in patients with specific HLA highly expressed on B-cell malignancies and B-cell lineages
haplotypes. Tumors may also develop resistance to TCR but not in pleuripotent stem cells or other tissues. The
T cells by downregulating MHC expression. toxicities observed in trials with CD19 CAR T cells are well
A significant challenge for TCR therapy, particularly in solid characterized20 and include cytokine release syndrome,
tumors, has been the choice of an appropriate antigen that which comprises fevers, hypotension, hypoxia, and eleva-
allows tumor-specific targeting. Trials with TCR T cells dition of serum cytokines (including IL-6 and interferon-γ).
rected at cancer-germline antigens such as the HLA*0201 This temporally correlates with activation and expansion of
epitope of NY-ESO-1 in melanoma and synovial cell the CAR T cells and is more frequent in the setting of high
carcinoma12 or MAGE-A-3 in melanoma13 or at differenti- tumor burden. Cytokine release syndrome may be severe or
ation antigens such as MART-1 and gp-10014 have been life-threatening and may be managed using IL-6 receptor
conducted. Efficacy in these trials was limited by neurologic antagonists such as tocilizumab. Neurologic toxicity has
toxicity13 and toxicity to skin, eyes, and ears14 caused by been observed in studies with CD19 CAR T cells, and al-
expression of the targeted antigen in normal tissues. An- though reversible, its cause is not clearly understood. Fi-
other strategy, with similar limitations, is targeting antigens nally, on-target, off-tumor toxicity was confined to B cells,
that are overexpressed in tumor tissue compared with and the anticipated B-cell aplasia is manageable with in-
normal tissue. Trials with TCR T cells directed at carci- travenous immunoglobulin.
noembryonic antigen have resulted in significant colitis The significant activity in hematologic malignancies has led
caused by carcinoembryonic antigen expression on colonic to interest in exploring CAR T cells in solid cancers. How-
mucosa.15 Targeting viral antigens with TCR therapies, ever, efficacy to date has been limited, with some notable
however, obviates issues of normal tissue toxicity. A trial of exceptions, as a result of significant barriers present in the
autologous genetically engineered T cells expressing a TCR context of solid cancers to effective CAR T-cell therapy.21,22
directed against HPV16 E6 in HPV-related cancers dem- A major challenge in solid tumors has been selection of
onstrated that two of 12 patients had tumors with objective the right antigen to target. Ideally, expression would be

Promising Immuno-Oncology Options for the Future
restricted to tumor tissue and be uniform with limited or no CAR natural killer cells, which are not limited by antigen
intra- and intertumor antigen heterogeneity, and there specificity and do not cause graft-versus-host disease or
would be no expression in normal tissue (to avoid on-target, autoimmune toxicity.32
off-tumor toxicity). These challenges are exemplified by PERSONALIZED CANCER VACCINES
a trial of CAR T cells directed at carbonic anhydrase IX, an
antigen overexpressed in renal cancers. Unexpectedly, the Cancer vaccines provide an opportunity to sensitize the host
trial resulted in significant liver toxicity caused by expression immune system to the tumor by enhancing existing antigen-
of carbonic anhydrase IX in biliary duct epithelium,23 in- specific T-cell responses and potentially generating new
dicative of on-target, off-tumor toxicity. A variety of target responses. A challenge has been the selection of appro-
antigens has been identified for CAR T-cell therapy in solid priate tumor-specific antigen targets that result in a potent
cancers, which includes thoracic malignancies such as and tumor-specific response while minimizing responses
mesothelin, glypican-3, epithelial cell adhesion molecule, to normal tissue. Initial approaches focused on nonmutated
prostate stem cell antigen, aberrantly glycosylated proteins self-antigens such as NY-ESO1, MAGE-A3, or MUC1, with
(MUC1, Lewis Y, receptor tyrosine kinase-like orphan re- generally disappointing results. More recent approaches
ceptor 1, and FAP), which have differing degrees of ex- have been directed at unique tumor neoantigens not found
pression in normal tissue.24 in normal tissue 33 or at viral antigens in viral-related
cancers. 34-36
Other barriers that have been identified in solid tumors
include T-cell migration and infiltration into tumors, over- There exists a strong rationale for targeting neoantigens,
coming immunosuppression within the tumor microenvi- which may be considered the key target for an effective
ronment, and limited persistence of CAR T cells. Several antitumor immune response. Increased tumor mutation
approaches have been developed to overcome these bar- burden and frequency of tumor neoantigens have been
riers.22 Novel delivery strategies have, for example, been correlated with stronger T-cell responses37 or responses to
used to enhance trafficking of CAR T cells into tumors, checkpoint blockade38,39; neoantigen-specific T cells are ex-
including engineering CAR T cells to express chemokine re- panded in patients deriving benefit from immunotherapy38,40;
ceptors or other factors that enhance tissue permeability25 and neoantigen-specific T cells have been demonstrated to
and local or regional administration of CAR T cells. Intra- be cytolytic to tumor cells presenting the mutant peptide.7
pleural administration of mesothelin-targeting CAR T cells For mutations to be appropriate targets, they must be
has resulted in improved efficacy and T-cell persistence in expressed, and the altered amino acid sequence must be
preclinical models26 and significantly in a clinical trial in processed into short peptide fragments and presented on
patients with mesothelioma.27 Furthermore, intracranial the cell surface in the context of MHC molecules to be
delivery of IL-13Rα2–targeted CAR T cells resulted in effi- recognized by the immune system. Approaches targeting
cacy in a patient with glioblastoma.28 A rationale also exists neoantigens have been facilitated by the advent and ac-
for combining PD-1 or PD-L1 inhibitors or other checkpoint cessibility of rapid whole-exome and transcriptome se-
inhibitors with CAR T cells.29,30 Combinations at least with quencing of tumor and normal tissue, coupled with
PD-1 or PD-L1 inhibitors seem clinically feasible and are bioinformatic algorithms to predict and prioritize potential
associated with encouraging activity in hematologic ma- neoepitopes based on prediction of binding affinity of
lignancies and solid tumors.27 Other approaches that may peptides to MHC. Additional approaches to enhance ac-
overcome the immunosuppressive tumor microenviron- curacy include mass spectrometric analysis of peptides
ment include targeting the adenosine 2A receptor,31 the use eluted from peptide-MHC complexes from resected tumors.
of immune agonists (e.g., 4-1BB agonists) or modulating Much effort and technical refinement have gone into de-
the cytokine and/or chemokine milieu (e.g., engineering sign and delivery strategies for vaccines, which include
T cells to counteract immunosppresive factors, such as IL- peptide-, DNA-, and RNA-based vaccines, some using viral
15 expression), or the use of CRISPR (clustered regularly vectors and others using dendritic cells and other delivery
interspaced short palindromic repeats) to knock out PD-1 strategies and immune adjuvants with the goal of eliciting
from CAR T cells.22 effective CD8+ and CD4+ T-cell antitumor responses. Fur-
Important factors limiting the broad applicability of CAR thermore, vaccines have potential to mediate upregulation of
T-cell therapy are that generation of autologous CAR T cells PD-L1 in the tumor microenvironment and turn “cold” tu-
is time consuming and requires specialized facilities and mors into “hot” tumors, providing a rationale for combi-
personnel, and this is associated with significant expense. A nations of vaccines with PD-1/PD-L1 inhibitors.34
potential solution to this is the use of off-the-shelf alloge- Phase I trials of tumor neoantigen vaccines conducted
neic CAR T cells made possible through gene editing initially in patients with melanoma41-43 and glioblastoma44,45
approaches or the use of induced pluripotent stem CAR showed that the vaccines were safe and could induce ro-
T cells. Another promising approach is the use of allogeneic bust neoantigen-specific responses. Personalized neoantigen

vaccine trials are now ongoing in other tumor types in- vaccines; an approach has already entered clinical trials
cluding non–small cell lung cancer; studies of vaccines (NCT03953235).
alone or in combination with immune checkpoint inhibitors CONCLUSION
are ongoing in patients with advanced disease, and studies
are planned in patients with early-stage resected non–small Cellular therapies and neoantigen vaccines, at least in the
cell lung cancer. context of solid tumors including lung cancer, remain ex-
perimental and have shown limited efficacy in clinical trials
Vaccine trials with vaccines directed at viral antigens have to date. It is expected that advances in T-cell engineering,
shown encouraging results in HPV-related cancers. Vac- opportunities for gene editing, and cell manufacturing ad-
cination with peptides directed at HPV antigens (HPV-16 vances, as well as the use of combination strategies, will
viral oncoproteins E6 and E7) in patients with premalignant make cell therapies more accessible and potentially over-
vulvar intraepithelial neoplasia (HPV-16–positive grade III come the significant barriers to efficacy in solid tumors.
vulvar intraepithelial neoplasia)46 and cervical intraepithelial Similarly, technological advances in recent personalized
neoplasia47 resulted in responses in some patients that neoantigen vaccines potentially in combination with check-
correlated with induction of HPV-16–specific immunity, point inhibitors have raised expectations that ongoing trials
and studies are ongoing that combine the vaccine with will have different outcomes than earlier cancer vaccine
a checkpoint inhibitor. studies. Advances in these cellular or vaccine strategies
In addition, the identification of common shared mutated tailored to individual patients and targeted at unique neo-
epitopes such as those found in some KRAS mutations6 antigens provides the opportunity for true personalization of
or TP53 mutations48 provide the basis for off-the-shelf immune therapy.

1
Peter MacCallum Cancer Centre, Melbourne, Australia AND DATA AVAILABILITY STATEMENT
EDBK_281101.
Benjamin J. Solomon, MBBS, PhD, Peter MacCallum Cancer Centre, 305
Grattan Street Melbourne 3000, Australia; email: ben.solomon@
petermac.org.
REFERENCES
1. Rosenberg SA, Restifo NP. Adoptive cell transfer as personalized immunotherapy for human cancer. Science. 2015;348:62-68.
2. Zhou J, Dudley ME, Rosenberg SA, et al. Persistence of multiple tumor-specific T-cell clones is associated with complete tumor regression in a melanoma patient
receiving adoptive cell transfer therapy. J Immunother. 2005;28:53-62.
3. Rosenberg SA, Packard BS, Aebersold PM, et al. Use of tumor-infiltrating lymphocytes and interleukin-2 in the immunotherapy of patients with metastatic
melanoma. A preliminary report. N Engl J Med. 1988;319:1676-1680.
4. Dudley ME, Wunderlich JR, Yang JC, et al. Adoptive cell transfer therapy following non-myeloablative but lymphodepleting chemotherapy for the treatment of
patients with refractory metastatic melanoma. J Clin Oncol. 2005;23:2346-2357.
5. Goff SL, Dudley ME, Citrin DE, et al. Randomized, prospective evaluation comparing intensity of lymphodepletion before adoptive transfer of tumor-infiltrating
lymphocytes for patients with metastatic melanoma. J Clin Oncol. 2016;34:2389-2397.
6. Tran E, Ahmadzadeh M, Lu YC, et al. Immunogenicity of somatic mutations in human gastrointestinal cancers. Science. 2015;350:1387-1390.
7. Tran E, Turcotte S, Gros A, et al. Cancer immunotherapy based on mutation-specific CD4+ T cells in a patient with epithelial cancer. Science. 2014;344:641-645.
8. Tran E, Robbins PF, Lu YC, et al. T-cell transfer therapy targeting mutant KRAS in cancer. N Engl J Med. 2016;375:2255-2262.
9. Stevanović S, Draper LM, Langhan MM, et al. Complete regression of metastatic cervical cancer after treatment with human papillomavirus-targeted tumor-
infiltrating T cells. J Clin Oncol. 2015;33:1543-1550.
10. Stevanović S, Helman SR, Wunderlich JR, et al. A phase II study of tumor-infiltrating lymphocyte therapy for human papillomavirus-associated epithelial cancers.
Clin Cancer Res. 2019;25:1486-1493.
11. Creelan B, Teer J, Toloza E, et al. Safety and clinical activity of adoptive cell transfer using tumor infiltrating lymphocytes (TIL) combined with nivolumab in NSCLC.
J Thorac Oncol. 2018;13:S330-S331.
12. Robbins PF, Kassim SH, Tran TL, et al. A pilot trial using lymphocytes genetically engineered with an NY-ESO-1-reactive T-cell receptor: long-term follow-up and
correlates with response. Clin Cancer Res. 2015;21:1019-1027.

Promising Immuno-Oncology Options for the Future
13. Morgan RA, Chinnasamy N, Abate-Daga D, et al. Cancer regression and neurological toxicity following anti-MAGE-A3 TCR gene therapy. J Immunother. 2013;
36:133-151.
14. Johnson LA, Morgan RA, Dudley ME, et al. Gene therapy with human and mouse T-cell receptors mediates cancer regression and targets normal tissues
expressing cognate antigen. Blood. 2009;114:535-546.
15. Parkhurst MR, Yang JC, Langan RC, et al. T cells targeting carcinoembryonic antigen can mediate regression of metastatic colorectal cancer but induce severe
transient colitis. Mol Ther. 2011;19:620-626.
16. Draper LM, Kwong ML, Gros A, et al. Targeting of HPV-16+ epithelial cancer cells by TCR gene engineered T cells directed against E6. Clin Cancer Res. 2015;
21:4431-4439.
17. Doran SL, Stevanović S, Adhikary S, et al. T-cell receptor gene therapy for human papillomavirus-associated epithelial cancers: a first-in-human, phase I/II study.
J Clin Oncol. 2019;37:2759-2768.
18. June CH, Sadelain M. Chimeric antigen receptor therapy. N Engl J Med. 2018;379:64-73.
19. Davenport AJ, Jenkins MR, Cross RS, et al. CAR-T cells inflict sequential killing of multiple tumor target cells. Cancer Immunol Res. 2015;3:483-494.
20. Neelapu SS, Tummala S, Kebriaei P, et al. Toxicity management after chimeric antigen receptor T cell therapy: one size does not fit ‘ALL’. Nat Rev Clin Oncol.
2018;15:218.
21. Newick K, O’Brien S, Moon E, et al. CAR T cell therapy for solid tumors. Annu Rev Med. 2017;68:139-152.
22. Mardiana S, Solomon BJ, Darcy PK, et al. Supercharging adoptive T cell therapy to overcome solid tumor-induced immunosuppression. Sci Transl Med. 2019;
11:aaw2293.
23. Lamers CH, Sleijfer S, van Steenbergen S, et al. Treatment of metastatic renal cell carcinoma with CAIX CAR-engineered T cells: clinical evaluation and
management of on-target toxicity. Mol Ther. 2013;21:904-912.
24. Kiesgen S, Chicaybam L, Chintala NK, et al. Chimeric antigen receptor (CAR) T-cell therapy for thoracic malignancies. J Thorac Oncol. 2018;13:16-26.
25. Slaney CY, Kershaw MH, Darcy PK. Trafficking of T cells into tumors. Cancer Res. 2014;74:7168-7174.
26. Adusumilli PS, Cherkassky L, Villena-Vargas J, et al. Regional delivery of mesothelin-targeted CAR T cell therapy generates potent and long-lasting CD4-
dependent tumor immunity. Sci Transl Med. 2014;6:261ra151.
27. Adusumilli PS, Zauderer MG, Rusch VW, et al. Abstract CT036: a phase I clinical trial of malignant pleural disease treated with regionally delivered autologous
mesothelin-targeted CAR T cells: Safety and efficacy. Cancer Res. 2019;79(13 suppl):CT036.
28. Brown CE, Alizadeh D, Starr R, et al. Regression of glioblastoma after chimeric antigen receptor T-cell therapy. N Engl J Med. 2016;375:2561-2569.
29. John LB, Devaud C, Duong CP, et al. Anti-PD-1 antibody therapy potently enhances the eradication of established tumors by gene-modified T cells. Clin Cancer
Res. 2013;19:5636-5646.
30. Grosser R, Cherkassky L, Chintala N, et al. Combination immunotherapy with CAR T cells and checkpoint blockade for the treatment of solid tumors. Cancer Cell.
2019;36:471-482.
31. Beavis PA, Henderson MA, Giuffrida L, et al. Targeting the adenosine 2A receptor enhances chimeric antigen receptor T cell efficacy. J Clin Invest. 2017;
127:929-941.
32. Liu E, Marin D, Banerjee P, et al. Use of CAR-transduced natural killer cells in CD19-positive lymphoid tumors. N Engl J Med. 2020;382:545-553.
33. Schumacher TN, Schreiber RD. Neoantigens in cancer immunotherapy. Science. 2015;348:69-74.
34. Sahin U, Türeci Ö. Personalized vaccines for cancer immunotherapy. Science. 2018;359:1355-1360.
35. Hu Z, Ott PA, Wu CJ. Towards personalized, tumour-specific, therapeutic vaccines for cancer. Nat Rev Immunol. 2018;18:168-182.
36. Li L, Goedegebuure SP, Gillanders WE. Preclinical and clinical development of neoantigen vaccines. Ann Oncol. 2017;28(suppl 12):xii11-xii17.
Science. 2015;348:124-128.
41. Sahin U, Derhovanessian E, Miller M, et al. Personalized RNA mutanome vaccines mobilize poly-specific therapeutic immunity against cancer. Nature. 2017;
547:222-226.
42. Ott PA, Hu Z, Keskin DB, et al. Correction: corrigendum: an immunogenic personal neoantigen vaccine for patients with melanoma. Nature. 2017;547:217-221.
43. Carreno BM, Magrini V, Becker-Hapak M, et al. Cancer immunotherapy. A dendritic cell vaccine increases the breadth and diversity of melanoma neoantigen-
specific T cells. Science. 2015;348:803-808.
44. Keskin DB, Anandappa AJ, Sun J, et al. Neoantigen vaccine generates intratumoral T cell responses in phase Ib glioblastoma trial. Nature. 2019;565:234-239.

45. Hilf N, Kuttruff-Coqui S, Frenzel K, et al. Published correction: actively personalized vaccination trial for newly diagnosed glioblastoma. Nature. 2019;
565:240-245.
46. Kenter GG, Welters MJ, Valentijn AR, et al. Vaccination against HPV-16 oncoproteins for vulvar intraepithelial neoplasia. N Engl J Med. 2009;361:1838-1847.
47. Kim TJ, Jin HT, Hur SY, et al. Clearance of persistent HPV infection and cervical lesion by therapeutic DNA vaccine in CIN3 patients. Nat Commun. 2014;5:5317.
48. Lo W, Parkhurst M, Robbins PF, et al. Immunologic recognition of a shared p53 mutated neoantigen in a patient with metastatic colorectal cancer. Cancer
Immunol Res. 2019;7:534-543.


And Now for Something Completely Different:

Immunotherapy Beyond Checkpoints
in Melanoma
Isabella Claudia Glitza, MD, PhD1; Stephanie L. Goff, MD2; Merrick Ross, MD3; and Kim Margolin, MD4
overview
Advances in the understanding of biology and therapy of melanoma have occurred at an astonishing pace over
the past approximately 15 years, and successful melanoma therapy has led the way for similar advances in
many other solid tumors that are continuing to improve outcomes for all patients with cancer. Although the
2018 Nobel Prize was awarded to two investigators who discovered that therapeutic targeting of immune
checkpoints held the key to major patient benefits, there are many additional immunotherapeutic strategies
that warrant further study and discussion at scientific and medical meetings. This article provides the newest
information on three areas of immunotherapy that have been successfully applied to melanoma and continue
to pave the way for new developments: cytokines, adoptive cell therapies (ADTs), and intratumoral injection of
immunomodulatory agents.
BACK TO THE FUTURE: THE MODERN ROLE OF those that target T cells and promote their growth and
CYTOKINE THERAPY cytotoxicity against tumor cells, have reopened the
Cytokines are critical signaling molecules in the innate field to the design of novel cytokine structures with
and adaptive immune systems, transmitting signals of improved function and potentially reduced toxicity that
activation and inhibition, as well as providing growth are currently in clinical studies.
and differentiation activities to many immune and in- IL-2
flammatory cells. Interferon-α (IFN-α) was the first
Recombinant human IL-2 aldesleukin was approved
cytokine developed for the treatment of malignancy,
by the U.S. Food and Drug Administration in 1992 for
with modest antiproliferative effects in chronic mye-
patients with advanced renal cell cancer, with objective
logenous leukemia, hairy cell leukemia, indolent B-cell
response rates (ORRs) of approximately 20%, about
lymphomas, and selected solid tumors.1,2 However,
one-half of which are durable. 6 IL-2, which stim-
the once-popular use of IFN-α in the adjuvant setting
ulates its targets through a cytokine-specific α receptor
for patients with resected high-risk melanoma was
(CD25), an inactive β-receptor (CD122), and a signal-
eclipsed in recent years by the more powerful and
ing receptor identical to that of several common γ
better-tolerated immune checkpoint blocking (ICB) receptor (γc) cytokines (CD132), is a type 1 immune-
antibodies, and cytokines found no role in adjuvant response cytokine that is secreted predominantly by
therapy for any solid tumor or hematologic malignancy. CD4 cells and stimulates other T cells and natural killer
However, interleukin-2 (IL-2), in combination with (NK) cells, with additional cascades of cells and in-
ex vivo–generated cytotoxic effector cells and then as flammatory molecules (e.g., TNF-α, IL-1β, and IFN-γ)
Author affiliations a single agent with comparable activity and better resulting from the initial action of IL-2.9 IL-2 activates
and support tolerance, was the only agent with meaningful activity, and promotes the proliferation of CD8 T cells to be-
information (if even inducing some durable complete responses come potent antitumor effector cells but has similar
applicable) appear (CRs) in advanced melanoma and renal cell cancer in
at the end of this
effects on suppressive regulatory CD4 T cells through
article.
the 1990s to early 2000s.3-6 Thereafter, a series of their constitutively active high-affinity αβγc receptor. IL-
Accepted on
small molecule kinase inhibitors (mainly for renal cell 2 also promotes antibody-dependent cellular cytotox-
February 25, 2020 cancer) and, soon thereafter, ICB (initially for mela- icity mediated by many different immune cells but also
and published at noma, subsequently for many solid tumors and se- prompts activation-induced cell death, another phys-
ascopubs.org on lected hematologic malignancies) demonstrated excellent iologic response to curb excessive immune stimulation
March 31, 2020:
therapeutic indices, justifying their place as first- and and protect host tissues, as well as limit antitumor
10.1200/EDBK_ even later-line therapy for these diseases.7,8 Never- responses. With the development of ICB, the use of
79437 theless, the powerful roles played by cytokines, particularly aldesleukin nearly disappeared as a result of the much

Immunotherapy Beyond Checkpoints in Melanoma
development.10 A limited course of high-dose IL-2 remains

part of many ACT regimens to expand and maintain the
function of the antitumor effector cells.
• The mainstay of immunotherapy for advanced
melanoma and melanoma at high risk of relapse Engineered IL-2 Molecules Designed to Improve
is immune checkpoint blockade, using either Therapeutic Index
single-agent PD-1 blockade or PD-1 antibody in
combination with CTLA-4 blockade. Bempegaldesleukin Bempegaldesleukin is a prodrug of
conjugated IL-2, in which the IL-2 retains the same amino
• Unmet needs remain in the therapy of mela-
acid sequence as aldesleukin, modified with six non-
noma, including the approximately 50% or
more of patients with cutaneous melanoma who enzymatically releasable polyethylene glycol (PEG) chains
do not achieve a durable remission from im- that dissociate slowly, providing a very long half-life at
mune checkpoint blockade and the even higher physiologic pH to an active form retaining one or two PEG
proportion of patients with acral, mucosal, and moieties that partially block the α-chain binding site.11
particularly uveal melanoma who do not re- Murine models confirmed substantial enhancement of tu-
spond well to these agents. mor exposure to bioactive IL-2 compared with aldesleukin.12
• IL-2 provides a low level of activity and a high Initial phase I trials showed proliferation and activation of
risk of toxicity when used as a single agent in CD4+, CD8+, and NK cells in peripheral blood, with in-
advanced melanoma, but new engineering creased infiltration of CD8+ and NK cells and an improved
approaches have provided alternative IL-2 ratio of CD8 effector/regulatory T cells in tumor.13,14
structures with the potential for improved im-
munotherapeutic properties that may synergize Bempegaldesleukin can also increase tumor-infiltrating
with immune checkpoint blockade and/or other lymphocytes (TILs) and PD-1 expression (now considered
therapeutic agents and provide an improved an early activation marker) on immune cells and, thus,
therapeutic index. complement the benefits of PD-1/PD-L1 blockade. The
• The most promising adoptive T-cell therapy for PIVOT-02 trial is currently testing bempegaldesleukin plus
melanoma is the use of TILs expanded from nivolumab in several solid tumors (NCT02983045). Initial
fragments of autologous tumor and potentially reports were encouraging, with higher than expected ORRs
selected for the recognition of tumor antigens or in patients with PD-L1–negative tumors predicted to have
for functional characteristics such as activation a low ORR.15 Importantly, toxicities were very modest, with
state or T-cell subset. The role of this form of only about 17% of patients experiencing a grade 3 or 4
therapy and its place in the therapeutic rep- event. A phase III trial is now comparing the combination
ertoire for various subsets of melanoma are with nivolumab in patients with advanced untreated mela-
likely to emerge with ongoing studies. noma; other trials will test bempegaldesleukin in other com-
• Melanoma often features cutaneous or nodal binations and tumor types (NCT03635983, NCT04209114,
metastases that can be injected with thera- NCT03138889, NCT03729245).
peutic agents to induce a strong local immune
response with the potential for developing Other engineered IL-2 molecules and immunocytokines
systemic antitumor immunity. A modified her- L19IL2 is a recombinant human L19 antibody fused to IL-2.
pesvirus engineered to express GM-CSF L19 has high affinity to the spliced extradomain B of fi-
(T-VEC) has modest activity as a single agent bronectin, which is associated in angiogenesis, and provides
and is now under investigation in combination
for selective delivery into bioactive IL-2 into the tumor.16
with immune checkpoint blockade. Several
Clinical testing of this agent is ongoing (NCT03567889,
other immunomodulatory agents are also in
development for intralesional therapy as single NCT01058538).17 Other anticancer IL-2–based cytokines
agents or in various immunotherapy under clinical investigation include fibroblast activation pro-
combinations. tein (FAP):IL-2, which concentrates IL-2 in FAP-expressing
tumors, including some adenocarcinomas, melanomas, and
sarcomas, resulting in activation of immune effector CD8 T
and NK lymphocytes but reduced stimulation of regulatory
T cells.18,19
more favorable outcomes and tolerance of ICB.8 However,
a recent retrospective analysis in patients who progressed Additional PEGylated IL-2s are also in development (THOR-
on ICB and then received high-dose IL-2 suggested simi- 707, also PEGylated at residues that block the IL-2α
lar activity; although the data are limited, this observa- receptor20), as well as other engineered “not alpha” struc-
tion supports further studies of IL-2 in selected patients tures (e.g., ALKS 423021) that enhance IL-2’s βγc binding,
and supports combination immunotherapeutic strategies in half-life, and other therapeutic features.

Glitza et al
Homeostatic γc Cytokines IL-15 and IL-7 and secretion of IL-12 lead to local and distant immune and
IL-15 IL-2 and IL-15 stimulate the βγc receptor, promoting antitumor responses.31,32 Further studies of this and other
CD8 T cell and NK cell activation and proliferation, but key engineered IL-12 molecules will assess the overall potential
differences include the lack of stimulation of regulatory of IL-12 in immunotherapy strategies.33
T cells and activation-induced cell death by IL-15, which KILLER T CELLS: THE HISTORY AND FUTURE OF
plays a homeostatic role in the maintenance of normal CELL THERAPY
numbers of circulating CD8 and NK cells.22,23 The differ-
The ways in which T cells can be applied to the treatment of
ences between IL-2 and IL-15 are based on the α-receptor,
cancer fall into three broad categories: (1) nonspecific
which, for IL-15, is cell bound on nontarget cells, primarily
stimulation of all of the T cells in the body (e.g., cytokines,
antigen-presenting cells, such as dendritic cells, whereas
ICB), (2) stimulation of specific cells in vivo (e.g., vaccines),
for IL-2, it is a subunit of the high-affinity αβγc trimeric
and (3) generation and stimulation of cells ex vivo to be
receptor. IL-15 is critical in the generation of memory CD8+
administered in vivo, which is the principle of ACT, generally
T cells, and it also enhanced the efficacy of ACT in the
featuring the passive transfer of large numbers of activated
murine model. IL-15 can expand NK and CD8 cells at doses
T cells. ACT regimens consist of preparative lymphode-
that cause less prominent capillary leak and hemodynamic
pleting chemotherapy, followed by infusion of one of a va-
compromise than the IL-2 doses associated with clinical
riety of cell products: TILs grown to large numbers from
benefits, but IL-15 has shown little clinical single-agent
tumor resections or biopsies, genetically engineered T cells
activity. However, modifications of IL-15 that prolong its
expressing a CAR designed to couple the cytolytic capacity
circulating half-life and deliver the cytokine in complexes
of T-cell signaling with the antibody recognition of surface
containing the cytokine-specific α-receptor mimic its physi-
proteins, or peripheral blood genetically modified to express
ologic structure:function and have already undergone pre-
human or transgenic T-cell receptors that recognize im-
liminary testing in the clinic, where combinations are now
munogenic epitopes presented by HLA molecules on tu-
being tested with ICB, antitumor antibodies, ACT, and other
mor. NK cells, autologously derived from patients, as well
immunomodulators.24-26
as allogeneic NK cells or cell lines with highly cytotoxic
IL-7 This cytokine is produced mainly by stromal cells of the properties, are being investigated for their potential as
thymus and lymph nodes and plays an important role in the a form of ACT but will not be further detailed here. For the
survival of naive and memory T cells by tightly moderating treatment of melanoma, patient-derived autologous TILs
the transition from effector cell to memory cell.27 IL-7 is have been studied extensively, prior to and since the ad-
a homeostatic cytokine that maintains normal circulating vent of present-day highly active immunotherapies, partic-
lymphocyte numbers and promotes the recovery of lym- ularly ICB.
phocytes after lymphodepleting therapies.28 Initial clinical
The history and current status of IL-2–based therapies for
assessments showed very little toxicity at doses associated
advanced melanoma were detailed earlier in this article.
with immunomodulatory effects but without antitumor ac-
Early murine research suggested that NK cells and
tivity as a single agent. However, the promotion by IL-7 of
lymphokine-activated killer cells generated by IL-2 could
increased T-cell receptor diversity and enhancement of
mediate tumor regression, but a randomized human clinical
memory responses have led to the development of com-
trial of IL-2, with and without lymphokine-activated killer cell
binations for further investigation, including ICB, cell-based
infusions, showed no difference in outcomes.34 However,
vaccines (sipuleucel-T for prostate cancer), postchemotherapy
the likelihood that other T-cell subsets mediated these IL-
restoration of lymphocyte counts and repertoire, and following
2–induced remissions was explored; further studies dem-
sepsis (NCT03821038).
onstrated that TILs within the tumor stroma could be cul-
IL-12, the Prototypic Type 1 Cytokine tured from resected melanoma metastases and were more
Similar to IL-15, but structurally unrelated, IL-12 promotes potent than lymphokine-activated killer cells in a murine
NK cell cytotoxicity via strong induction of the type 1 cy- metastasis model.35
tokine response, mainly IFN-γ, the predominant mediator of Human TILs are generated by the sterile processing of
antitumor cytotoxicity.29,30 Although systemic IL-12 and a freshly resected tumor, prior to fixation and further
IFN-γ are excessively toxic and can also induce the se- pathologic analysis. Although techniques now vary based on
cretion of counterregulatory type 2 cytokines at doses re- the research and clinical needs of each institution, initial
quired for immunotherapeutic effects, IL-12 can be given cultures were based on fragments (approximately 2–3 mm3)
locally in one of several engineered forms. The most ex- placed into media supplemented with high concentrations
tensively studied IL-12–based immunotherapy consists of of IL-2.36 Common alternative techniques include creating
a plasmid-encoded human IL-12 gene injected and elec- an enzymatic tumor digest for isolation of lymphocytes
troporated into accessible tumors, where gene expression rather than sharp dissection into fragments, as well as

inclusion of additional cytokines or other pharmacologic alone, with a 24% CR rate in each arm and increased
agents to drive cell differentiation and diversity. TILs are toxicity with the combined-modality regimen.43
nonadherent and can be harvested from media over the
Other groups have made valuable contributions to the de-
course of 2 to 3 weeks.37
velopment of autologous TIL transfer for metastatic mela-
The first human experience of ACT using TILs occurred in noma, demonstrating that tumor regression is possible
the Surgery Branch of the National Cancer Institute and using different ex vivo cell growth conditions44 with differ-
included 20 patients with metastatic melanoma.38 The ent culture selection criteria,45 selecting different T-cell
regimen consisted of a single dose of cyclophosphamide to subsets,46 and with administration of low-dose IL-2 after
provide lymphodepletion (now known to function in part by cell transfer. 47,48 However, the success of ICB rapidly
inducing homeostatic cytokines, detailed below and also changed the landscape for patients with metastatic mela-
discussed in the section on homeostatic Yc cytokines IL-15 noma and their physicians, considering the high activity and
and IL-7), intravenous infusion of autologous TILs, and high- safety of ICB, particularly with PD-1 blockade, and its rapid
dose IL-2 to expand and maintain the activity of the ad- expansion into the treatment of multiple other malignancies.
ministered TILs. Objective responses were seen in 11 of 20 Two remarkable aspects of ACT to consider are the depth
(55%) patients, including a single complete responder. and durability of response. In 226 patients treated in four
Toxicities of the regimen were transient, similar to those of trials at the Surgery Branch, 49 patients (21%) achieved
comparable doses of IL-2 alone. However, the responses CRs; only two of those patients have experienced re-
were largely of short duration, and the mechanisms of ac- currence of disease, with median potential follow-up of more
tion and of intrinsic and acquired resistance could not be than 8 years. A third patient in CR developed recurrence
studied in that era. after intense immunosuppression for the treatment of
Sequential iterations of the TIL technology were designed to a hematologic malignancy. Six patients remain in ongoing
improve the activity and broaden the population of patients partial response, with no evidence of progression with
with advanced melanoma who could benefit from TIL a minimum follow-up of 5 years, suggesting the strategy may
therapy. Re-treating patients with transient responses to be capable of eliminating micrometastatic disease, even
adoptive cell transfer often proved informative, revealing when unable to eradicate radiologically evident tumors. The
nuanced changes in efficacy or duration within the same median overall survival of the entire cohort is 20.4 months,
patient or helping to elucidate some mechanisms of tumor with 5- and 10-year survival rates of 36% and 32% (S.L.
escape (loss of antigen expression, loss of class I MHC).39,40 Goff, MD, unpublished data, December 2019). Because
The most important change was intensification of the most of these early patients who received TILs were naive to
preparative chemotherapy to provide deeper lymphode- current-day standard therapies, those who received any ICB
pletion without the toxicities and risks of myeloablation. The therapy upon relapse had a median postprogression overall
profound, but relatively transient, lymphopenia induced by survival of 18.7 months.43 Surgical resection of progressing
the combination of cyclophosphamide and fludarabine lesions in carefully selected patients can also provide
prompted a homeostatic cytokine response, consisting of secondary long-term disease control, with a 5-year post-
a surge of IL-7 and IL-15 in response to lymphodepletion operative survival rate of 57%.49
that, in turn, promoted lymphocyte expansion, including
Because ICB has become the standard first-line therapy
that of the administered TILs. In sequential studies, the use
for patients with metastatic melanoma, it is likely that the
of this more lymphodepleting nonmyeloablative regimen as
clinical equipoise no longer exists to test the strategies head-
preconditioning for TILs was associated with an overall
to-head in a randomized fashion, although the identification
response rate of 51% and a CR rate of 12%,41 which was
of patients who will not benefit from ICB and/or targeted
higher than that seen in the prior trials with the less-intensive
therapies would support the earlier use of novel therapies,
lymphodepleting regimen and in patients with similar char-
like TILs, in some of these patients. The current strategy of
acteristics, including prior therapy. In subsequent trials, the
offering TILs and other novel ACT predominantly to patients
lymphodepletion was further intensified by the addition of
whose melanoma was not controlled by ICB, by lack of an
total body irradiation (2 or 12 Gy), similar to that which has
initial response (intrinsic resistance), or by progression
been used for allogeneic hematopoietic cell transplant, to
following initial response (acquired resistance) may change
cyclophosphamide and fludarabine. In this series, ORRs
as additional data become available.
were highest (72%) in those patients who received 12 Gy
total body irradiation, an increasingly complex, now mye- Some of the more recent studies from our group at the
loablative, regimen requiring autologous hematopoietic stem National Cancer Institute Surgery Branch were able to
cell support.42 However, a prospective randomized trial dem- address these treatment patterns in part; for example, our
onstrated no improvement in outcomes from the addition of most recently published ACT trial accrued patients from
total body irradiation over chemotherapy lymphodepletion 2011 to 2013, bridging the introduction of ICB to the wider

Glitza et al
oncology community.43 Approximately 40% of patients had Studies of long-term responders in past trials of ACT for
experienced tumor progression through anti–CTLA-4 ther- melanoma have identified recognition of melanoma-specific
apy, and, when compared with the cohort as a whole, ORRs antigens (MART-1 or gp100), as well as neoantigens de-
were nearly identical (overall ORR 55%, CR,26%). There rived from nonsynonymous somatic mutations of the
were too few patients in the studies during that time period autologous tumor.53 The frequency in the TME and re-
to provide meaningful analysis of response rates after pro- quirements for identifying and expanding cells that me-
gression on anti–PD-1/PD-L1 agents. diate this autologous recognition are active areas of
investigation.
It is reasonable to predict that response rates after un-
successful or transiently successful manipulation of the PD- Commercial development of ACT with TILs for melanoma
1/PD-L1 axis may be lower. Unlike CTLA-4 and its ligands, has focused on its potential role as a post-ICB late-line
the role of PD-1/PD-L1 is to modulate immune responses in therapy. The single-treatment strategy can mediate durable
peripheral tissues, minimizing inflammation after antigen regression in patients whose tumors are refractory to ap-
recognition by effector T cells.50 TILs generated from tumors proved medications. Ongoing research may include at-
that have not responded to checkpoint blockade inhibi- tempts to generate TILs from patients prior to the start of
tion may be qualitatively or quantitatively inferior. In a pre- anti–PD-1/PD-L1 antibodies, combining ACT with ICB
liminary analysis of all melanoma tumors resected for TILs at (NCT02621021), or investigating TIL cultures for reactivity
the Surgery Branch from 2011 to 2018 (301 patients), there to potential neoantigens to enable selection of an enhanced
were fewer CD8 cells in cultures derived from anti–PD- tumor-reactive infusion product. Other future strategies
1–refractory tumors (82 patients; median, 41%; 95% CI, could include the generation of genetically engineered
39.4–49.5) than anti–PD-1–naive tumors (219 patients; lymphocytes targeting a patient’s melanoma. Prior efforts
median, 65.5%; 95% CI, 57.4–63.8). In the same analysis, targeting shared melanoma antigens (gp100, MART-1)
there were no differences in the phenotype of cultures demonstrated modest responses but also uncovered the
derived from tumors that had or had not been exposed to exquisite sensitivity of high-avidity receptors, with on-target,
anti–CTLA-4 (G. Ivey, MD, unpublished data, June 2019). off-tumor toxicities that affected normal tissues with minimal
antigen expression.54 Growing evidence suggests that the
Although there have been considerable resources applied to final common pathway for immune-mediated destruction of
the search for a biomarker to better predict responses to cancer is the recognition of unique somatic tumor muta-
ICB, those efforts are usually focused on primary tumors or tions.55 Academic institutions and immune-oncology pharma-
pretreatment metastatic disease. There are few examples of ceutical interests continue to develop unique techniques to
studying the tumor or the tumor microenvironment (TME) identify patient mutations and refine technology platforms for
after treatment with ICB. In a small sample of patients gene insertion.
undergoing serial biopsy during pembrolizumab treatment,
nonresponding patients had lower CD8+ T-cell density at all IMMUNOLOGIC CELL DEATH BY A THOUSAND INJECTIONS:
timepoints.51 It is possible that intratumoral T-cell diversity ONCOLYTIC IMMUNOTHERAPY FOR
is also affected by anti–PD-1 treatment, but by different ADVANCED-STAGE MELANOMA
parameters in patients with and without exposure to Oncolytic Therapy and the “Cancer-Immunity Cycle”
anti–CTLA-4.52 Ongoing investigations are likely to further
Developing in parallel with the amazing treatment advances
elucidate the immunologic dependencies of CTLA-4, PD-1/
of systemically adinistered immune and targeted therapies
PD-L1, and other targets for ICB that will, in turn, inform the
for patients with advanced/unresectable disease has been
design of TILs and other forms of ACT in development for
the role for direct intralesional injection of accessible me-
advanced melanoma.
tastases with a variety of agents. Although only one therapy
In currently accruing TIL trials for patients who progressed has U.S. Food and Drug Administration approval, an
on PD-1 blockade at the National Cancer Institute, pre- oncolytic herpes simplex virus, talimogene laherpar-
liminary data suggest an ORR of 23.5% (95% CI, 12.4–40). epvec (T-VEC), many other agents that span a wide
Although this appears lower than for TIL therapy in PD-1 spectrum of class and mechanism of action are under active
antibody–naive patients, any meaningful rate of durable investigation. These include additional oncolytic viruses,
tumor regression in such patients is an encouraging in- inflammatory dyes (PV-10), cytokines (IL-2), electroporation
dication that some component(s) of the adoptive transfer with IL-12 plasmids, Toll-like receptor 9 (TLR9) agonists,
strategy can overcome a tumor’s resistance to ICB-based and STING pathway–activating drugs (Fig. 1). These dif-
immunotherapy: lymphodepletion may alter the TME, the ferent agents share the common goal of local tumor ablation
process of generating TILs may expand rare tumor-reactive to achieve local disease control, as well as the generation of
clonotypes in vitro, or the bolus delivery of activated an effective systemic antitumor immune response leading to
T cells may engage different cytolytic mechanisms than ICB. tumor regression at distant metastatic sites—a genuine

without a substantial increase in treatment-related toxicities.

For patients with visceral metastases, image-guided in-
jections may be possible but remain to be proven safe and
effective enough for wide use. The following topics will be
addressed in this section and, for the most part, apply to
patients with safely accessible injectable lesion(s) that may
be dermal, subcutaneous, or nodal: (1) oncolytic mono-
therapy, (2) combination with ICB, (3) neoadjuvant onco-
lytic therapy for resectable metastases, and (4) resistance to
PD-1 blockade.
FIGURE 1. Intralesional Agents in Development in Melanoma
A spectrum of agents in terms of class and mode of action are being Intratumoral Monotherapy
investigated. Most of the clinical trial experience using oncolytic mono-
therapy was initiated prior to the more recent approvals of
systemic adjuvant effect, or “abscopal” response. Simply
ICB and targeted therapy for unresectable stage III and IV
described, the initiating step is ablation of an injected tumor
disease. T-VEC is a herpes simplex type 1 virus, genetically
leading to an inflammatory tumor cell death resulting in
modified to preferentially replicate in tumors and produce
exposure of nondenatured tumor antigens, as well as re-
granulocyte-macrophage colony-stimulating factor (GM-
lease of “danger” signals, molecules that bind receptors
CSF) during replication. In October 2015, the U.S. Food and
on antigen-presenting cells and lead to their activation of
Drug Administration approved T-VEC for the treatment of
local immune cells, creating a lymph node–like structure
recurrent melanoma with metastatic lesions in the skin and
in the TME that leads to a systemic immune response
lymph nodes, based on the phase III OncovexGM-CSF Pivotal
against tumor antigens. Tumor infiltration, followed by
Trial in Melanoma (OPTiM), which tested intratumoral
T-cell–mediated tumor destruction, augments the inflam-
T-VEC against subcutaneous GM-CSF in 436 patients with
matory response and perpetuates the immunity cycle in
unresected stage IIIB-IV melanoma who were randomly
a paracrine fashion (Fig. 2).56 In support of this mechanism
assigned to receive, in a 2:1 ratio, T-VEC or GM-CSF, re-
is evidence that many oncolytic agents generate increased
spectively. Durable responses were achieved in 16.3% of
levels of circulating CD8 cells and IFN-γ and T-cell in-
the T-VEC group and in 2.1% of the GM-CSF group (p ,
filtration in injected and uninjected tumors in association
.0001). The durable response rate was highest (33%)
with measurable responses at distant uninjected sites of
among patients with stage III and stage IVM1a disease. The
disease.57-59
overall response rate (partial response + CR) was also higher
There is strong rationale for combining oncolytics with for T-VEC (26.4%) compared with GM-CSF (5.7%; p ,
ICB. Anti–CTLA-4 (ipilimumab) appears to enhance T-cell .0001).60 An analysis of lesion-level responses among 2,116
priming at the level of the dendritic cell, and anti-PD-(L)1 lesions showed 50% or greater tumor area decreases in
acts at the level of the T cell within the TME. The goals of 64% of injected lesions, 34% of uninjected nonvisceral
oncolytic therapy are to increase the response to ICB and lesions, and 15% of uninjected visceral lesions (Fig. 3). CR
overcome innate resistance/prevent acquired resistance, was accomplished in 47% of injected lesions, 22% of
FIGURE 2. Oncolytic Immunother-

apy Designed To Produce Local and
Systemic Effects56
Cancer-immunity cycle: a proposed
cascade of events initiated by an
intratumoral injection leading to
a systemic antitumor immune T-
cell–mediated response is illustrated.
Points of synergy with checkpoint
blockade (arrows) provides the ra-
tionale for combinatory approaches.
Abbreviation: APC, antigen-presenting
cell.

Glitza et al
FIGURE 3. Lesion-Level and

Patient-Level Responses to T-VEC
Local, regional, and distant re-
sponse rates generated by local
injections of T-VEC shown in
waterfall plots. Responses ob-
served at uninjected regional
and distant sites of disease il-
lustrate an “abscopal”/systemic
immune antitumor effect.
Abbreviatons: CR, complete re-
sponse; OR, overall response.
uninjected nonvisceral lesions, and 9% of uninjected vis- subcutaneous lesions. Objective responses were robust,
ceral lesions. T-VEC was well tolerated, with cellulitis (2.1%), with 25% achieving a CR, 26% achieving a partial response,
as the only grade 3/4 adverse event, reported in 2% or more and 18% achieving stable disease; locoregional disease
of patients.61,62 control was achieved in 69%. Among 35 patients with
evaluable untreated bystander lesions, a CR was achieved
Another oncolytic virus demonstrating promising clinical
in 31%, a partial response was achieved in 9%, and stable
activity is coxsackie virus A21 (CVA21). CVA21 is a naturally
disease was achieved in 20%. Responses in injected lesions
occurring common cold intercellular adhesion molecular
correlated strongly with uninjected lesion responses.66-68
1–targeted RNA virus.63 Among a number of cancers, in-
cluding melanoma, surface intercellular adhesion molec- Other immunomodulatory agents have been injected intra-
ular 1 is upregulated, which represents the mechanism for tumorally, mostly in the historical timeframe prior to ICB and
intracellular concentration of virus, followed by viral repli- current understanding of immunotherapy mechanisms.
cation and tumor cell lysis. Tumor cells lysed by CVA21, in Although intralesional IL-2 has activity and has been ap-
animal models and in in vivo xenografts, induced a sec- proved in Canada for use in patients with unresectable in-
ondary systemic host-generated antitumor immune re- transit metastases, there is little evidence of an abscopal
sponse. The preclinical results accurately predicted the response, and it is impractical to inject all lesions in most
clinical trial activity observed in patients with melanoma. A patients.
phase II trial with CVA21 monotherapy demonstrated local Other immunomodulators with a less substantial track re-
disease control and responses at uninjected sites of disease cord and little immunologic understanding of mechanisms
at rates similar to that of T-VEC64,65 (NCT01227551). Tox- have been injected intratumorally, including type 1 (α and β)
icity profile has been favorable, because intercellular ad- and type 2 (γ) interferons, GM-CSF, and bacillus Calmette-
hesion molecular 1 levels are low in normal cells. Guérin.
PV-10 is a sterile nonpyrogenic 10% solution of Rose Combinations of Intratumoral Therapy With ICB
Bengal disodium, a small molecule fluorescein derivative
with a long history as a hepatic and ophthalmic diagnostic The rationale for combining intralesional therapy with sys-
with excellent safety. When injected directly into tumors, PV- temic therapy lies in their single-agent activity, potential
10 transits the plasmalemma of cancer cells, accumulating synergy, and nonoverlapping toxicities. The goal of frontline
in lysosomes and triggering their release with complete combinations is to provide a regimen with activity compa-
autophagy within 30 to 60 minutes. In a multicenter phase rable or superior to combined ICB with less toxicity.
II trial that included 80 patients with stage III–IV melanoma A phase Ib/II trial of T-VEC plus anti–CTLA-4 (ipilimumab)
in Australia and the United States, patients received up demonstrated little additional toxicity over that expected
to four PV-10 courses, treating up to 20 cutaneous or from ipilimumab alone. In phase II, 198 patients with stage

IIIB or IV melanoma were randomly assigned to receive ongoing PD-1 blockade in patients whose disease progressed
ipilimumab alone or with T-VEC. We found that grade 3 on PD-1 blockade alone.
toxicity was slightly increased with the combination, and A phase II trial of PV-10 in combination with pembrolizumab
nearly all grade 3 toxicities could be attributed to ipilimu- is ongoing in patients with unresectable stage III-IV mela-
mab. The added toxicity of T-VEC, in combination with noma and at least one injectable lesion (NCT02557321).
ipilimumab, was minimal and short-lived, consisting of flu- Other combination frontline trials include a phase II study of
like symptoms that were self-limited and well tolerated. CVA21 plus pembrolizumab, with encouraging preliminary
Although the response rate was higher with the combina- response rates of more than 60%.70 Comparative rates of
tion (39% vs. 18%), the randomized phase II trial had in- response and toxicity levels of the most studied immuno-
sufficient power to demonstrate statistically significant therapies for advanced melanoma are shown in Table 1.
benefits, and overall survival was not enhanced with com-
bination therapy, potentially because of the commercial The combination trials in Table 1 were primarily for treatment-
availability of T-VEC, as well as the availability of targeted naive patients (frontline). A single-arm phase II trial of the
therapies and PD-1 blockade for patients whose disease oncolytic type 1 herpes simplex virus HF10 (canerpaturev),
progressed.69 in combination with ipilimumab, treated 43 patients with
unresected stage IIIB-IVm1c disease. It included treatment-
Another study examining T-VEC plus a checkpoint inhibi- naive patients and patients whose disease did not respond
tor is MASTERKEY-265, a randomized phase III trial of to anti–PD-1 treatment, with an overall response rate of
pembrolizumab with T-VEC or with a placebo injection in 49%.71
unresected advanced melanoma. Results of the phase Ib
portion of the study, with extensive immunologic correlates, Neoadjuvant Intralesional Therapy
were published in 2017, based on 21 patients treated with A logical application of intralesional therapy is prior to
T-VEC alone, followed by T-VEC plus pembrolizumab. The surgery for resectable recurrent single-site or oligo-metastatic
ORR was 62%. Although this was a small substudy, the melanoma. Although there may be inherent risks in delaying
results were dramatic because the rate of response to surgery, injection of the metastatic tumor may stimulate
pembrolizumab alone is between 35% and 40%, and the a systemic immune response that persists after resection
rate of response to T-VEC alone is approximately 26%. The and may be superior to that achieved with postsurgical
combination was well tolerated, with no dose-limiting tox- adjuvant therapy against micrometastatic disease. The
icities. Evidence of immune activation with this combination largest completed melanoma neoadjuvant study to date was
was based on tumor biopsy specimens from before and a randomized phase II trial including 150 patients with
during therapy.70 In the phase III portion of MASTERKEY-265 resectable stage IIIB, IIIC, or IVM1a melanoma who were
(NCT02263508), 660 patients were accrued in the same randomly assigned to neoadjuvant T-VEC followed by sur-
study design, with the results for the primary endpoint— gery, or surgery alone; all patients were offered systemic
progression-free survival—expected in late 2020. If the study adjuvant therapy of the physician’s choice. In this study,
is positive, the results may support a regulatory approval for which was done prior to the widespread use of ICB
the first-ever combination of intratumoral oncolytic virother- in advanced melanoma, the pathologic CR rate to T-VEC
apy and ICB. Other combinations are under evaluation, was 21%, higher than the rate expected from T-VEC in
mainly with the approach of adding intratumoral therapy to advanced disease lesions. Interestingly, improved 1-year
TABLE 1. Toxicity and Response Rates With Immunotherapy in Advanced Unresectable Stage III/IV Melanoma
Therapy Grade ‡ 3 Adverse Events, % Response Rate, %
Monotherapy
Ipilimumab 27 6–15
Pembrolizumab 13 27–38
Nivolumab 16 34–40
T-VEC 11 26
Combination Therapy
Nivolumab + ipilimumab 55 52
T-VEC + ipilimumab 32 50
T-VEC + pembrolizumab 24 56
HF10 + ipilimumab 30 49

Glitza et al
recurrence-free survival (primary endpoint) with neoadjuvant On-Label Treatment, Off-Label Attempts, and
T-VEC was presented at the 2019 ASCO Annual Meeting.72 Future Perspectives
However, this trial was initiated and almost completely Although major advances in ICB have been made in the
accrued prior to the approval of the anti–PD-1 and targeted systemic therapy of advanced melanoma, these treatments
therapy in the adjuvant settings, so the optimal use of come with the potential for serious and irreversible toxicities.
intralesional injection therapies, especially their potential in Furthermore, several contraindications to systemic immu-
neoadjuvant regimens, remains to be ascertained.73,74 Given notherapy exist, including autoimmune disease and other
the promising activity for combinations of intralesional medical comorbidities that are more common in elderly
therapy and ICB in advanced melanoma, such combina- patients. The lower toxicities of intralesional therapies may
tions are being evaluated in the neoadjuvant setting, for allow them to be considered for such patients, as well as for
example, using pembrolizumab and intratumoral injections the substantial fraction of patients who are not cured by ICB
of a first-in-class TLR9 agonist (an oligonucleotide) en- and do not have prolonged responses to targeted therapy, if
capsulated within a virus-like particle (CMP-001) that applicable. T-VEC monotherapy, the only agent approved by
showed activity in patients with melanoma whose disease the U.S. Food and Drug Administration in class, has been
progressed on PD-1 blockade; promising preliminary data used extensively on-label with favorable results, particularly
were recently reported, showing a pathologic CR in 13 of 21 in patients with an indolent disease course and with pre-
patients.75 dominantly unresectable in-transit disease. T-VEC injections
Overcoming resistance to PD-1 blockade Although routine frequently provide needed local disease control and may, at
use of anti–PD-1 as frontline therapy for patients with the same time, reduce the risk for distant disease failure.
metastatic disease has substantially improved long-term
Given the favorable published data with T-VEC in combi-
outcomes, more than one-half of patients do not have
nation with ipilimumab or pembrolizumab, it is not sur-
long-term benefit with ICB. This group of patients represents
prising that off-label use has been attempted in patients with
the largest growing patient subset with metastatic mela-
unresectable stage III-IV disease, particularly in the absence
noma, so we are eager to discover accurate biomarkers to
of an available clinical trial. Such attempts are often un-
help identify and successfully treat those patients whose
successful because of insurance denials. The use of T-VEC
tumors are inherently resistant or who will acquire resistance
as monotherapy in the neoadjuvant setting would also be
to PD-1 blockade. Strategies that have been used include
off-label, because the approval wording states “unresect-
addition of CTLA-4 blockade or using it as second-line
able recurrent disease after initial surgery.” Although the
single-agent treatment, adding an intralesional therapy
preliminary results of neoadjuvant T-VEC are promising and
while continuing PD-1 blockade, or switching to ipilimumab
provide proof of concept that intralesional therapy can lead
in combination with an intratumoral oncolytic agent. One
to systemic immune surveillance, subsequent neoadjuvant
such example is the PISCES/KEYNOTE-695 study that is
trials of ICB have generated higher pathologic CR rates,74 so
investigating the use of electroporation with intralesional IL-
the role of intralesional monotherapy in neoadjuvant regi-
12 gene–encoding plasmids (tavokinogene telseplasmid)
mens will remain very limited; however, early results of small
plus pembrolizumab in patients with metastatic melanoma
phase II trials of intralesional therapy, in combination with
progressing on PD-1 inhibitor (NCT03132675).
anti–PD-1, demonstrating high pathologic CR rates with
In addition, several TLR9 agonists are being studied in a favorable toxicity profile,76 provide the rationale for con-
combination with ipilimumab or pembrolizumab in similar tinued clinical investigation of such combinatorial strategies
clinical settings. The most promising and furthest in dein the neoadjuvant setting.
velopment is a phase III randomized trial of second-line
ipilimumab versus ipilimumab plus the TLR9 agonist til- CONCLUSIONS
sotolimod, which is based on encouraging results from the The immunotherapy of melanoma has benefited from suc-
phase II trial that had a 30% rate of response to the cessful application of emerging insights and technology in
combination, which is higher than that expected from ipi- the field, the dedication of translational and clinical sci-
limumab alone in second-line therapy, in the range of 10% entists, and the support of patients and advocacy orga-
to 16% objective responses (NCT03445533). Additional nizations for the clinical trials that are critical to provide the
ongoing phase II second-line (PD-1 failure) trials include evidence basis for new drug approvals. Application of
combinations of ipilimumab with HF10 (NCT02272855) these approaches to the ICBs, as well as a variety of other
and with CVA21 (NCT02307149). A phase Ib trial is ongoing immunotherapeutic strategies—cytokines, ACTs and intra-
that combines CMP-001 with anti–PD-1 after progression tumoral injections—has led to additional options and new
on PD-1 antibody (ClinicalTrials.gov Identifier, NCT03084640), study designs that are expected to further the successes and
with preliminary data reported in 2019 showing 25% ORR improved outcomes for patients with melanoma and those
among 144 patients treated.76 diagnosed with other cancers in the foreseeable future.


1
Department of Melanoma Medical Oncology, University of Texas, MD Kim Margolin, MD, City of Hope Medical Center, 1500 East Duarte Rd.,
Anderson Cancer Center, Houston, TX Duarte, CA 91010; Twitter: @kmargolin; email: [email protected].
2
Surgery Branch, National Cancer Institute, National Institutes of Health,
Bethesda, MD
3
Department of Surgical Oncology, University of Texas, MD Anderson
Cancer Center, Houston, TX
4 Disclosures provided by the authors and data availability statement (if
Department of Medical Oncology and Therapeutics Research, City of
Hope National Medical Center, Duarte, CA
EDBK_79437.
REFERENCES
1. Schwarzinger I, Bettelheim P, Lechner K. Interferontherapie bei hmatologischen Neoplasien. [Interferon therapy in hematologic neoplasms] Wien Klin
Wochenschr. 1988;100:497-504.
2. Antonelli G, Scagnolari C, Moschella F, et al. Twenty-five years of type I interferon-based treatment: a critical analysis of its therapeutic use. Cytokine Growth
Factor Rev. 2015;26:121-131.
3. Foon KA, Walther PJ, Bernstein ZP, et al. Renal cell carcinoma treated with continuous-infusion interleukin-2 with ex vivo-activated killer cells. J Immunother.
1992;11:184-190.
4. Sznol M, Clark JW, Smith JW II, et al. Pilot study of interleukin-2 and lymphokine-activated killer cells combined with immunomodulatory doses of chemotherapy
and sequenced with interferon alfa-2a in patients with metastatic melanoma and renal cell carcinoma. J Natl Cancer Inst. 1992;84:929-937.
5. Whitehead RP, Kopecky KJ, Samson MK, et al. Phase II study of intravenous bolus recombinant interleukin-2 in advanced malignant melanoma: Southwest
Oncology Group study. J Natl Cancer Inst. 1991;83:1250-1252.
6. Atkins MB, Lotze MT, Dutcher JP, et al. High-dose recombinant interleukin 2 therapy for patients with metastatic melanoma: analysis of 270 patients treated
between 1985 and 1993. J Clin Oncol. 1999;17:2105-2116.
7. Mihaly Z, Sztupinszki Z, Surowiak P, et al. A comprehensive overview of targeted therapy in metastatic renal cell carcinoma. Curr Cancer Drug Targets. 2012;
12:857-872.
8. Schachter J, Ribas A, Long GV, et al. Pembrolizumab versus ipilimumab for advanced melanoma: final overall survival results of a multicentre, randomised, open-
label phase 3 study (KEYNOTE-006). Lancet. 2017;390:1853-1862.
9. Russell SM, Keegan AD, Harada N, et al. Interleukin-2 receptor gamma chain: a functional component of the interleukin-4 receptor. Science. 1993;
262:1880-1883.
10. Buchbinder EI, Dutcher JP, Daniels GA, et al. Therapy with high-dose Interleukin-2 (HD IL-2) in metastatic melanoma and renal cell carcinoma following PD1 or
PDL1 inhibition. J Immunother Cancer. 2019;7:49.
11. Sharma M, Khong H, Fa’ak F, et al. Bempegaldesleukin selectively depletes intratumoral Tregs and potentiates T cell-mediated cancer therapy. Nat Commun.
2020;11:661.
12. Charych DH, Hoch U, Langowski JL, et al. NKTR-214, an engineered cytokine with biased IL2 receptor binding, increased tumor exposure, and marked efficacy
in mouse tumor models. Clin Cancer Res. 2016;22:680-690.
13. Doberstein SK. Bempegaldesleukin (NKTR-214): a CD-122-biased IL-2 receptor agonist for cancer immunotherapy. Expert Opin Biol Ther. 2019;19:1223-1228.
14. Jenkins RW, Thummalapalli R, Carter J, et al. Molecular and genomic determinants of response to immune checkpoint inhibition in cancer. Annu Rev Med.
2018;69:333-347.
15. Diab A, Hurwitz M, Cho DC, et al. NKTR-214 (CD-122-biased agonist) plus nivolumab in patients with advanced solid tumors: preliminary phase 1/2 results of
PIVOT. J Clin Oncol. 2018;36:15s (suppl; abstr 3006).
16. Carnemolla B, Borsi L, Balza E, et al. Enhancement of the antitumor properties of interleukin-2 by its targeted delivery to the tumor blood vessel extracellular
matrix. Blood. 2002;99:1659-1665.
17. Weide B, Eigentler T, Catania C, et al. A phase II study of the L19IL2 immunocytokine in combination with dacarbazine in advanced metastatic melanoma
patients. Cancer Immunol Immunother. 2019;68:1547-1559.
18. Melero I, Alvarez EC, Mau-Sorensen M, et al. Clinical activity, safety, and PK/PD from a phase I study of RO6874281, a fibroblast activation protein (FAP) targeted
interleukin-2 variant (IL-2v). Ann Oncol. 2018;29 (suppl 8):viii133-viii148.
19. Brennen WN, Isaacs JT, Denmeade SR. Rationale behind targeting fibroblast activation protein-expressing carcinoma-associated fibroblasts as a novel
chemotherapeutic strategy. Mol Cancer Ther. 2012;11:257-266.
20. Luo D, Abdul-Karim R, Azad A, et al. Open-label, multicenter phase 1/2 dose escalation and expansion study of THOR-707 as a single agent and in combination
with a PD-1 inhibitor in adult subjects with advanced or metastatic solid tumors. J Immunother Cancer. 2019;7 (suppl 1; abstr P430).
21. Losey HC, Lopes JE, Dean RL, et al. Efficacy of ALKS 4230, a novel immunotherapeutic agent, in murine syngeneic tumor models alone and in combination with
immune checkpoint inhibitors. Cancer Res. 2017;77 (13 suppl; abstr 591).
22. Mishra A, Sullivan L, Caligiuri MA. Molecular pathways: interleukin-15 signaling in health and in cancer. Clin Cancer Res. 2014;20:2044-2050.
23. Roychowdhury S, May KF Jr., Tzou KS, et al. Failed adoptive immunotherapy with tumor-specific T cells: reversal with low-dose interleukin 15 but not low-dose
interleukin 2. Cancer Res. 2004;64:8062-8067.

Glitza et al
24. Wrangle JM, Velcheti V, Patel MR, et al. ALT-803, an IL-15 superagonist, in combination with nivolumab in patients with metastatic non-small cell lung cancer:
a non-randomised, open-label, phase 1b trial. Lancet Oncol. 2018;19:694-704.
25. Margolin K, Morishima C, Velcheti V, et al. Phase I trial of ALT-803, a novel recombinant IL15 complex, in patients with advanced solid tumors. Clin Cancer Res.
2018;24:5552-5561.
26. Romee R, Cooley S, Berrien-Elliott MM, et al. First-in-human phase 1 clinical study of the IL-15 superagonist complex ALT-803 to treat relapse after trans-
plantation. Blood. 2018;131:2515-2527.
27. Lin J, Zhu Z, Xiao H, et al. The role of IL-7 in immunity and cancer. Anticancer Res. 2017;37:963-967.
28. Silk AW, Margolin K. Cytokine therapy. Hematol Oncol Clin North Am. 2019;33:261-274.
29. Leonard JP, Sherman ML, Fisher GL, et al. Effects of single-dose interleukin-12 exposure on interleukin-12-associated toxicity and interferon-gamma production.
Blood. 1997;90:2541-2548.
30. Brunda MJ, Luistro L, Rumennik L, et al. Antitumor activity of interleukin 12 in preclinical models. Cancer Chemother Pharmacol. 1996;38 (suppl):S16-S21.
31. Daud AI, DeConti RC, Andrews S, et al. Phase I trial of interleukin-12 plasmid electroporation in patients with metastatic melanoma. J Clin Oncol. 2008;
26:5896-5903.
32. Bhatia S, Longino NV, Miller NJ, et al. Intratumoral delivery of plasmid IL12 via electroporation leads to regression of injected and noninjected tumors in merkel
cell carcinoma. Clin Cancer Res. 2020;26:598-607.
33. Garris CS, Arlauckas SP, Kohler RH, et al. Successful anti-PD-1 cancer immunotherapy requires T cell-dendritic cell crosstalk involving the cytokines IFN-γ and
IL-12. Immunity. 2018;49:1148-1161.e7.
34. Rosenberg SA, Lotze MT, Yang JC, et al. Prospective randomized trial of high-dose interleukin-2 alone or in conjunction with lymphokine-activated killer cells for
the treatment of patients with advanced cancer. J Natl Cancer Inst. 1993;85:622-632.
35. Rosenberg SA, Spiess P, Lafreniere R. A new approach to the adoptive immunotherapy of cancer with tumor-infiltrating lymphocytes. Science. 1986;
233:1318-1321.
36. Dudley ME, Wunderlich JR, Shelton TE, et al. Generation of tumor-infiltrating lymphocyte cultures for use in adoptive transfer therapy for melanoma patients.
J Immunother. 2003;26:332-342.
37. Tran KQ, Zhou J, Durflinger KH, et al. Minimally cultured tumor-infiltrating lymphocytes display optimal characteristics for adoptive cell therapy. J Immunother.
2008;31:742-751.
38. Rosenberg SA, Packard BS, Aebersold PM, et al. Use of tumor-infiltrating lymphocytes and interleukin-2 in the immunotherapy of patients with metastatic
melanoma. A preliminary report. N Engl J Med. 1988;319:1676-1680.
39. Khong HT, Wang QJ, Rosenberg SA. Identification of multiple antigens recognized by tumor-infiltrating lymphocytes from a single patient: tumor escape by
antigen loss and loss of MHC expression. J Immunother. 2004;27:184-190.
40. Rosenberg SA, Yang JC, Robbins PF, et al. Cell transfer therapy for cancer: lessons from sequential treatments of a patient with metastatic melanoma.
J Immunother. 2003;26:385-393.
41. Dudley ME, Wunderlich JR, Yang JC, et al. Adoptive cell transfer therapy following non-myeloablative but lymphodepleting chemotherapy for the treatment of
patients with refractory metastatic melanoma. J Clin Oncol. 2005;23:2346-2357.
42. Dudley ME, Yang JC, Sherry R, et al. Adoptive cell therapy for patients with metastatic melanoma: evaluation of intensive myeloablative chemoradiation
preparative regimens. J Clin Oncol. 2008;26:5233-5239.
43. Goff SL, Dudley ME, Citrin DE, et al. Randomized, prospective evaluation comparing intensity of lymphodepletion before adoptive transfer of tumor-infiltrating
lymphocytes for patients with metastatic melanoma. J Clin Oncol. 2016;34:2389-2397.
44. Itzhaki O, Hovav E, Ziporen Y, et al. Establishment and large-scale expansion of minimally cultured “young” tumor infiltrating lymphocytes for adoptive transfer
therapy. J Immunother. 2011;34:212-220.
45. Pilon-Thomas S, Kuhn L, Ellwanger S, et al. Efficacy of adoptive cell transfer of tumor-infiltrating lymphocytes after lymphopenia induction for metastatic
melanoma. J Immunother. 2012;35:615-620.
46. Radvanyi LG, Bernatchez C, Zhang M, et al. Specific lymphocyte subsets predict response to adoptive cell therapy using expanded autologous tumor-infiltrating
lymphocytes in metastatic melanoma patients. Clin Cancer Res. 2012;18:6758-6770.
47. Ellebaek E, Iversen TZ, Junker N, et al. Adoptive cell therapy with autologous tumor infiltrating lymphocytes and low-dose Interleukin-2 in metastatic melanoma
patients. J Transl Med. 2012;10:169.
48. Nguyen LT, Saibil SD, Sotov V, et al. Phase II clinical trial of adoptive cell therapy for patients with metastatic melanoma with autologous tumor-infiltrating
lymphocytes and low-dose interleukin-2. Cancer Immunol Immunother. 2019;68:773-785.
49. Klemen ND, Feingold PL, Goff SL, et al. Metastasectomy following immunotherapy with adoptive cell transfer for patients with advanced melanoma. Ann Surg
Oncol. 2017;24:135-141.
52. Riaz N, Havel JJ, Makarov V, et al. Tumor and microenvironment evolution during immunotherapy with nivolumab. Cell. 2017;171:934-949.e16.
53. Robbins PF. Tumor-infiltrating lymphocyte therapy and neoantigens. Cancer J. 2017;23:138-143.

54. Johnson LA, Morgan RA, Dudley ME, et al. Gene therapy with human and mouse T-cell receptors mediates cancer regression and targets normal tissues
expressing cognate antigen. Blood. 2009;114:535-546.
55. Tran E, Robbins PF, Rosenberg SA. ‘Final common pathway’ of human cancer immunotherapy: targeting random somatic mutations. Nat Immunol. 2017;
18:255-262.
56. Chen DS, Mellman I. Oncology meets immunology: the cancer-immunity cycle. Immunity. 2013;39:1-10.
57. Kaufman HL, Amatruda T, Reid T, et al. Systemic versus local responses in melanoma patients treated with talimogene laherparepvec from a multi-institutional
phase II study. J Immunother Cancer. 2016;4:12.
58. Ribas A, Dummer R, Puzanov I, et al. Oncolytic virotherapy promotes intratumoral T cell infiltration and improves anti-PD-1 immunotherapy. Cell. 2017;
170:1109-1119.e10.
59. Ross MI. Intralesional therapy with PV-10 (Rose Bengal) for in-transit melanoma. J Surg Oncol. 2014;109:314-319.
60. Andtbacka RH, Kaufman HL, Collichio F, et al. Talimogene laherparepvec improves durable response rate in patients with advanced melanoma. J Clin Oncol.
2015;33:2780-2788.
61. Andtbacka RH, Ross MI, Delman K, et al. Responses of injected and uninjected lesions to intralesional talimogene laherparepvec (T-VEC) in the OPTiM Study and
the contribution of surgery to response. Ann Surg Oncol. 2014;21 (suppl 1):S23.
62. Kaufman HL, Andtbacka RHI, Collichio FA, et al. Primary overall survival (OS) from OPTiM, a randomized phase Ill trial of talimogene laherparepvec (T-VEC)
versus subcutaneous (SC) granulocyte-macrophage colony-stimulating factor (GM-CSF) for the treatment (tx) of unresected stage IIIB/C and IV melanoma. J Clin
Oncol. 2014;32:15s (suppl; abstr 9008a).
63. Hersey P, Gallagher S. Intralesional immunotherapy for melanoma. J Surg Oncol. 2014;109:320-326.
64. Andtbacka RHI, Curti BD, Kaufman H, et al. CALM study: A phase II study of an intratumorally delivered oncolytic immunotherapeutic agent, coxsackievirus A21,
in patients with stage Ilk and stage IV malignant melanoma. J Clin Oncol. 2014;32:15s (suppl; abstr 3031).
65. Andtbacka RHI, Curti BD, Kaufman H, et al. Final data from CALM: A phase II study of coxsackievirus A21 (CVA21) oncolytic virus immunotherapy in patients with
advanced melanoma. J Clin Oncol. 2015;33:15s (suppl; abstr 9030).
66. Agarwala SS, Thompson JF, Smithers BM, et al. Immuno-chemoablation of metastatic melanoma with intralesional Rose Bengal (PV-10). Ann Oncol. 2012;23
(suppl 9):ix371.
67. Thompson JF, Agarwala SS, Smithers BM, et al. Phase 2 study of intralesional PV-10 in refractory metastatic melanoma. Ann Surg Oncol. 2015;22:2135-2142.
68. Sarnaik A, Crago G, Liu H, et al. Assessment of immune and clinical efficacy after intralesional PV-10 in injected and uninjected metastatic melanoma lesions. J
Clin Oncol. 2014;32:15s (suppl; abstr 9028).
69. Chesney J, Puzanov I, Collichio F, et al. Randomized, open-label phase II study evaluating the efficacy and safety of talimogene laherparepvec in combination
with ipilimumab versus ipilimumab alone in patients with advanced, unresectable melanoma. J Clin Oncol. 2018;36:1658-1667.
70. Silk A, Kaufman H, Gabrail N, et al. CAPRA: A phase 1b study of intratumoral coxsackievirus A21 (CVA21) and systemic pembrolizumab in advanced melanoma
patients. J Immunother Cancer. 2017;5 (suppl 2; abstr P245).
71. Andtbacka RHI, Ross MI, Agarwala SS, et al. Final results of a phase II multicenter trial of HF10, a replication-competent HSV-1 oncolytic virus, and ipilimumab
combination treatment in patients with stage IIIB-IV unresectable or metastatic melanoma. J Clin Oncol. 2017;35:15s (suppl; abstr 9510).
72. Andtbacka RHI, Dummer R, Gyorki DE, et al. Interim analysis of a randomized, open-label phase 2 study of talimogene laherparepvec (T-VEC) neoadjuvant
treatment (neotx) plus surgery (surgx) vs surgx for resectable stage IIIB-IVM1a melanoma (MEL). J Clin Oncol. 2018;36:15s (suppl; abstr 9508).
73. Dummer R, Gyorki DE, Hyngstrom JR, et al. One-year (yr) recurrence-free survival (RFS) from a randomized, open label phase II study of neoadjuvant (neo)
talimogene laherparepvec (T-VEC) plus surgery (surgx) versus surgx for resectable stage IIIB-IVM1a melanoma (MEL). J Clin Oncol. 2019;37:15s (suppl; abstr
9520).
74. Topalian SL, Taube JM, Pardoll DM. Neoadjuvant checkpoint blockade for cancer immunotherapy. Science. 2020;367.
75. Milhem M, Zakharia Y, Davar D, et al. Durable responses in anti-PD-1 refractory melanoma following intra-tumoral injection of a Toll-like receptor 9 (TLR9)
agonist, CMP-001, in combination with pembrolizumab. Presented at: 34th Annual Meeting of The Society for Immunotherapy of Cancer; National Harbor, MD;
2019. Abstract O85.
76. Davar D, Karunamurthy A, Hartman D, et al. Phase II trial of neoadjuvant nivolumab (Nivo) and intra-tumoral (IT) CMP-001 in high risk resectable melanoma
(MEL): preliminary results. J Immunother Cancer. 2019;7 (suppl 1; abstr O34).

Immunologic Characteristics of Nonmelanoma

Skin Cancers: Implications for Immunotherapy
Evan T. Hall, MD, MPhil1,2; Elena Fernandez-Lopez, MD3; Ann W. Silk, MD, MS4; Reinhard Dummer, MD3; and Shailender Bhatia, MD1,2
overview
In this review, we summarize the immunology of nonmelanoma skin cancers (NMSCs) and the clinical data
with immunotherapy in this heterogeneous group of cancers that include basal cell carcinoma (BCC), cu-
taneous squamous cell carcinoma (CSCC), and Merkel cell carcinoma (MCC). NMSCs are exceedingly
common, and their treatment consumes substantial health care resources. Annual global mortality from
NMSCs is comparable to that from malignant melanoma. Although the majority of NMSCs are localized at
diagnosis and are treated effectively with surgery, metastases (nodal and distant) can sometimes arise and
require systemic therapy. Given the success of immunotherapy in treating cutaneous melanoma, there has
been an increasing interest in studying the immunology of NMSCs. Immunocompromised patients have
a substantially higher risk of developing NMSCs (particularly CSCC and MCC), suggesting a role of the immune
system in the pathogenesis of these cancers. Similar to cutaneous melanoma, the pathogenesis of BCC, CSCC,
and virus-negative MCC is related to DNA damage from ultraviolet radiation exposure, and these cancers have
a very high tumor mutational burden, which likely results in higher levels of tumor neoantigens that may be
targets for the immune system. Viral antigens in virus-positive MCC are also strongly immunogenic. Emerging
data from clinical trials of immune checkpoint inhibitors in NMSCs look very promising and are rapidly
changing the treatment landscape of these cancers. Specifically, pembrolizumab and avelumab are U.S. Food
and Drug Administration–approved for treatment of metastatic MCC and cemiplimab for metastatic CSCC.
Several ongoing trials are investigating novel immunotherapies (monotherapies as well as combination) for
treatment of NMSCs.
INTRODUCTION AND EPIDEMIOLOGY NMSCs is most likely related to the aging population
Nonmelanoma skin cancers (NMSCs) represent a trends, which reflect increasing cumulative ultraviolet
heterogeneous group of diseases, including basal cell exposure over a lifetime and decreasing protection
carcinoma (BCC), cutaneous squamous cell carci- because of immunosenescence.5
noma (CSCC), and Merkel cell carcinoma (MCC), Despite their high incidence, NMSCs account for
among others. As a group, NMSCs represent the most mortality in a relatively low percentage of patients, with
commonly diagnosed malignancies in the world, with 65,000 deaths attributed to NMSCs globally in 2017.1
an estimated incidence of 7.7 million cases in 2017, Although this represents mortality in a low proportion of
representing more than 30% of all cancer diagnoses.1 NMSC cases, the absolute number of deaths is similar
Notably, BCC and CSCC are often not measured in to the mortality attributed to melanoma skin cancer
cancer registries, and incidence estimates vary widely (62,000 annual deaths worldwide in 2017).1 Most
depending on the source. It is estimated that there are BCCs are diagnosed as local tumors and are treated
Author affiliations 2 million BCC cases diagnosed annually in the United with curative surgery. However, BCC rarely can be-
and support States alone. 2 CSCC is the second most common come locally advanced or metastasize, requiring ra-
information (if NMSC, with an estimated annual incidence of ap- diation therapy and/or systemic therapy. Although
applicable) appear
at the end of this
proximately 700,000 cases in the United States.3 One CSCC rarely metastasizes (3%–4% of cases are as-
article. group estimated that the combined annual incidence sociated with nodal metastases) and is fatal in a smaller
Accepted on of BCC and CSCC may be as high as 5.4 million cases subset of patients (approximately 1.5%–3%), given the
February 25, 2020 in the United States in 2012.4 NMSC incidence has overall common incidence, it is estimated that between
and published at increased by 33% in the 10 years from 2007 to 2017.1 4,000 and 8,000 deaths were attributable to CSCC in
ascopubs.org on
MCC is much rarer than BCC and CSCC, although its the United States in 2012.3,6 Although MCC is com-
March 24, 2020:
incidence, too, has nearly doubled during the past paratively rare, it is perhaps the most aggressive
10.1200/EDBK_ 20 years and will likely exceed 3,000 cases per year in NMSC, with proportional mortality up to three times
278953 the United States by 2025.5 The rising incidence of higher than that of melanoma. More than 30% of

Immunology and Immunotherapy of NMSC
lymphocytic leukemia14). NMSCs sometimes regress after

improvement in immune function, further underscoring the
importance of immune surveillance in their pathogenesis.15,16
• Given the pathogenesis and biology of non- These observations, along with the amazing success of im-
melanoma skin cancers, there is a strong
munotherapy in melanoma, have led to a strong interest in
rationale for using immunotherapy.
studying the immunology of NMSCs, which is best exemplified
• BCC, CSCC, and VN-MCC have among the not only by the rapid advances in our understanding of MCC
highest TMB of all cancers as a result of pathogenesis but also in the emerging advances in immu-
ultraviolet radiation–induced DNA damage.
nology of CSCC and BCC.
• VP-MCC results in the production of viral anti-
genic proteins that elicit a host immune MERKEL CELL CARCINOMA
response.
Much about the biology of MCC has been discovered since
• Checkpoint inhibitor immunotherapy has the clonal integration of the Merkel cell polyomavirus
demonstrated efficacy and resulted in regula-
(MCPyV) was first described in 2008, implicating it as an
tory approvals for advanced MCC and CSCC.
oncogenic virus.17 Approximately 80% of MCC cases in the
There are case reports also demonstrating
efficacy in BCC. United States and Europe are associated with the virus and
can be classified as virus-positive MCC (VP-MCC).17 The
• More research is needed to identify novel im-
MCC polyomavirus is a small, double-stranded DNA virus,
munotherapy strategies for those patients who
which is part of the SV40 subgroup of polyomaviruses. The
do not experience a response to checkpoint
inhibitors. virus is normally harmless skin flora, and circulating anti-
bodies to the capsid protein can be detected in the majority
of healthy people.18 However, in a small fraction of people,
patients present with nodal or distant metastasis at the virus integrates into the DNA of the cell and transforms
presentation.7 the cell. The cell then begins to express the viral proteins,
including MCPyV T-antigen oncoproteins that inhibit the
Given the high incidence, treatment of NMSC consumes
tumor suppressor proteins retinoblastoma and PP2A, sta-
substantial health care resources, as the average annual
bilize the oncoprotein C-myc, and drive unregulated growth.19,20
cost to treat NMSCs in the United States between 2007 and
The remaining 20% of MCC cases in the United States and
2011 was nearly US$5 billion, an amount exceeding that
Europe (and a much higher proportion, approximately 70%,
spent on melanoma during the same time period.8 His-
in Australia) lack detectable tumor-associated MCPyV DNA
torically, clinical research in advanced NMSCs has lagged
or oncoproteins and can be classified as virus-negative MCC
far behind that in melanoma, and systemic treatment op-
(VN-MCC). The mutational burden of VN-MCC exceeds that
tions for metastatic disease were mostly cytotoxic chemo-
of melanoma, and the pattern of mutations strongly impli-
therapies. Fortunately, rapid strides are being made in our
cates ultraviolet-induced DNA damage as the causative
understanding of the biology of NMSCs that have opened up
mechanism.21,22
new avenues for investigation of rational therapies in this
important subset of skin cancers. We review the recent MCCs that have overexpression of immune response genes
discoveries in the biology of NMSCs, with a special focus on and higher intratumoral infiltration of CD8+ lymphocytes
the emerging importance of immune mechanisms in the are generally found to have more favorable prognoses.23
pathogenesis of this heterogeneous group of diseases. MCPyV-positive tumor cells continue to express viral pro-
teins, and there is evidence to suggest this is recognized by
IMMUNOLOGIC CHARACTERISTICS OF NONMELANOMA the patient’s humoral and cellular immune system.17,20,24-26
SKIN CANCERS MCPyV-specific T cells have been isolated from the pe-
Epidemiologic data suggest a strong link between NMSCs ripheral blood or tumors of affected patients, and their
and the immune system. Immunosuppression is widely presence is associated with favorable survival.26 IgG anti-
recognized as a risk factor for the development of NMSCs, bodies against the MCPyV small T antigen correlate with
and increasing use of therapeutic immunosuppression is tumor burden in patients with MCC,27 leading to the de-
likely contributing to the increasing incidence of NMSCs.9 velopment of a clinically validated assay (AMERK, run by
After solid-organ transplantation, the risk of NMSCs in- a Clinical Laboratory Improvement Amendments–certified
creases, with standardized incidence ratios of CSCC ranging laboratory) for surveillance of high-risk patients with VP-
from 15 to 357 compared with the general population.10 MCC tumors.28 In VN-MCC, the high mutational burden
Similarly, MCC risk is increased by fivefold to 50-fold in from chronic ultraviolet exposure likely leads to generation
individuals with T-cell dysfunction (solid organ transplant and expression of novel proteins and epitopes and, sub-
recipients,11,12 patients with HIV,13 or patients with chronic sequently, neoantigen-directed immune responses. 22

Hall et al
These observations readily explain the important role of rather than oncogenes,36 which has limited efforts to de-
cellular immune responses in the natural history of both VP- velop targeted therapies.
and VN-MCC.29
Interestingly, both BCC and CSCC are known to express
Despite persistent expression of viral antigens in VP-MCC or cancer-testis antigens (CTAs), which are typically only
ultraviolet-induced neoantigens in VN-MCC, MCCs that expressed in trophoblastic and male germline cells. Ex-
progress to clinical detection have demonstrated an ability pression in cancer cells is believed to arise from deme-
to evade patient immune responses. MCC tumors can thylation of their promoter.37 CTAs are frequently expressed
downregulate antigen presentation, including by major in NMSC, and more than 40% coexpress at least two CTAs.
histocompatibility complex I (MHC-I) loss, to more effec- However, no specific IgG response against CTAs are de-
tively evade immune recognition.30 MCC cells also induce tectable in BCC and CSCC. BCC expresses at least one CTA
an immunosuppressive microenvironment by producing more frequently than CSCC (81% vs. 41%, respectively) and
immunosuppressive cytokines or via recruitment of im- also expresses a wider variety of them per biopsy.32
munosuppressive cells, such as CD4+CD25+ regulatory
T cells (T regs) or myeloid-derived suppressor cells.31 In Antigen Presentation in Basal Cell and Cutaneous
response to chronic antigen exposure, antigen-specific Squamous Cell Carcinomas
CD8+ T cells in the MCC tumor microenvironment com- There is evidence that BCC has a reduced (but possibly
monly develop an exhausted phenotype with poor effector inducible) antigen presentation compared with CSCC. This
function, sustained expression of inhibitory receptors (e.g., finding is consistent with the theory that BCC has a lower
PD-1, Tim3), and a transcriptional state distinct from that immunogenicity than CSCC despite its higher TMB overall.
of functional effector or memory T cells.25 Together, these Biologic aspects felt to account for this difference include (1)
observations provided a strong rationale for investigation of impaired antigen presentation because of downregulation of
PD-1 blockade and other immunotherapies in MCC that transporter associated with antigen processing-1 (TAP-1)
have now revolutionized the treatment landscape of these and MHC-I, (2) diminished infiltration by CD4 and CD8
patients with historically poor prognosis. T cells with increased presence of T regs, and (3) immu-
BASAL CELL AND CUTANEOUS SQUAMOUS nosuppressive effects driven by interleukin (IL)-10 and Th2
CELL CARCINOMAS cytokines, among other factors.
BCCs and CSCCs derive from epidermal keratinocytes that Antigens deriving from the degradation of endogenous
are chronically exposed to ultraviolet light, a well-known proteins are presented to CD8 T cells via MHC after
mutagen. Therefore, it is not surprising that these cancers intracellular processing by TAP-1 and TAP-2 and subse-
show the highest tumor mutational burden (TMB) among all quent stabilization with β2-microglobulin.38 MHC-I and β2-
cancer types (47.3 and 45.2 mutations/Mb, respectively), microglobulin have a lower expression in BCC compared
similar to VN-MCC.32 This high TMB likely contributes to an with CSCC,32,39 although MHC-I is upregulated on BCC cells
increased immunogenicity through increased expression of in response to treatment with Hh inhibitors.40 Interestingly,
neoantigens. BCCs show a complete absence or paucity of TAP-1
expression.41 Therefore, the lack of MHC-I and β2-
Despite high TMB overall, most BCCs are driven by genetic
microglobulin expression may be because of impaired as-
alterations affecting Sonic Hedgehog (Hh) pathway signaling.
sembly caused by the lack of TAP-1. To our knowledge,
BCCs are characterized by the presence of cilia, for which
there are no studies addressing the expression of TAP-1 in
persistence is associated with the upstream activation of Hh
CSCC in the literature. However, these investigations in non-
pathway signaling. Cilia have a dual role in tumorigenesis,
CSCC observed a downregulation of TAP-1 in oral mucosal
favoring Smoothened (Smo)-dependent tumorigenesis and
squamous cell carcinoma cell lines.42 BCC shows in-
suppressing the Smo-independent growth.33 Mutations in the
consistent expression of human leukocyte antigen (HLA)–
genes related to the Wnt pathway are also frequently ob-
ABC and lacks expression of HLA-DR.43 The expression of
served in BCC, resulting in activation of the pathway and
HLA-I protein has not consistently been observed among
b-catenin overexpression. It is also known that C-myc is
CSCC, but a higher expression of these molecules in
downregulated.34 Overall, this pattern of signaling changes
comparison with BCC has been reported. CSCC tumor cells
promotes the phenotype of undifferentiated keratinocyte
also expressed high levels of HLA-II proteins, particularly of
stem cells observed in BCC tumoral cells, which may also be
HLA-DR.44
related to mechanisms of immune evasion.
A different genetic mutational profile is typically observed in Immune Cell Infiltrate in Basal Cell and Cutaneous
the case of CSCC. The most common alterations involve Squamous Cell Carcinomas
TP53, NOTCH1 (43%), and PTCH1 (11%).35 The majority There are differences in the immune infiltration among BCC
of the mutations in CSCC occur in tumor suppressor genes and CSCC tumors. The peri- and intratumoral infiltration of

CD8 T cells as well as the expression of MHC-I is drastically checkpoint axis in MCC was confirmed with the PD-1 in-
reduced in BCC compared with SCC,32 which suggest that hibitor pembrolizumab.60 The response rate was 56% (14 of
BCC escapes from immune surveillance by downregulation 25 patients) in a treatment-na ı̈ve study population, in-
of MHC-I. The hypothesis that CSCC is more subjected to cluding 24% who experienced complete responses. Ave-
immune surveillance is further supported by the observation lumab was subsequently tested in treatment-naı̈ve patients
that the risk of developing CSCC increases dramatically in as well.61 The response rate was 62%, markedly higher than
patients who received an organ transplant.45 High numbers in the trial with patients whose disease had progressed on/
of infiltrating T regs have been both reported in BCC and after chemotherapy. The fact that PD-L1 inhibition was less
CSCC, in which these T regs lead to an in vitro inhibition of effective when used after chemotherapy suggests that the
the proliferation of CD4 and CD8 T cells and interferon cytotoxic agents may be harmful to immune effector cells
gamma (IFN-γ).46,47 and blunt the response to subsequent immunotherapy. The
A strong presence of immature dendritic cells has been current standard of care for first-line MCC is pembrolizumab
reported in the stroma of BCC,48 which is both a cause and or avelumab. The drugs have not been directly compared
a consequence of the overexpression of IL-10 reported in against each other.
the tumor microenvironment of BCC. Treatment of BCC with The optimal duration of treatment of metastatic disease is
Hh inhibition can result in increased intratumoral infiltration currently unknown. Generally, treatment was given for
of CD4+ and cytotoxic CD8+ T cells.40 In contrast, CSCC are 1 year in the absence of progression or toxicity. As in
infiltrated by mature dendritic cells but are poor stimulators melanoma, responses to PD-1– and PD-L1–based therapy
of the proliferation of T cells.49 appear to be durable for months after discontinuation of
treatment; additional follow-up is needed to determine the
Cytokine Signaling in Basal Cell and Cutaneous Squamous
durability of responses and the effect of rechallenge with
Cell Carcinomas
anti–PD-1 or anti–PD-L1 therapy. The side effect profiles of
Hints of the relative immune privilege of BCC (as opposed to avelumab and pembrolizumab are acceptable and con-
CSCC) can also be observed in its stroma. Higher consistent with that of PD-1 and PD-L1 inhibitors in other
centrations of IL-4, IL-6, and IL-1B in BCC compared with cancers. Approximately 15% to 20% of patients experience
CSCC have been reported.50,51 The immunosuppressive grade 3 or 4 adverse events that are autoimmune in
cytokines transforming growth factor-B and IL-10 are nature.60,61 One notable difference between avelumab and
overexpressed in BCC and in high-grade CSCC, whereas pembrolizumab is that infusion reactions are more common
low-grade CSCCs show a lower expression.43,52,53 This high with avelumab (23% vs. 0%) but are usually mild and do not
level of IL-10 could explain the absence of upregulation of preclude additional therapy.
MHC-I and -II, ICAM-1, CD40, CD80, HLA-ABC, and HLA-
Rapid and deep responses correlated with survival.62
DR after induction of IFN-γ.43
Generally, the time to response is a few weeks, and most
IMMUNOTHERAPY OF NONMELANOMA SKIN CANCERS patients with responses demonstrated a response by the
Merkel Cell Carcinoma first on-study scan, in approximately 6 to 8 weeks. Given the
kinetics of response, nivolumab has been investigated in the
Before immunotherapy, MCC was treated like other high- neoadjuvant setting in a small study of 25 patients.63 Pa-
grade small cell neuroendocrine cancers with chemother- tients with advanced but resectable MCC were treated with
apy, such as cisplatin and etoposide. Like small cell car- two doses of nivolumab (days 1 and 15) followed by surgery.
cinoma of the lung, MCC has a high rate of initial response to Most patients (88%) had their surgery as planned during the
chemotherapy (approximately 65%), but the responses are fourth week of the trial. Response was assessed radio-
seldom durable. Survival statistics with chemotherapy were graphically and pathologically. Among 20 patients with pre/
poor, with an average overall survival of less than 1 year for post nivolumab CT scans, 45% had a radiographic partial
metastatic disease.54-56 response; among 17 patients evaluated for pathologic re-
Half of MCC cases express PD-L1 in the tumor microen- sponse, a major pathologic response, defined as 10% or
vironment. Among tumors with moderate to high intensity of less residual viable tumor cells, was achieved in 65% of
tumor-infiltrating lymphocytes, 100% of cases stained patients, including 47% who had a pathologic complete
positively for PD-L1 expression on tumor cells in one response. The authors identified characteristic histologic
study.57 This observation led to a trial using the PD-L1 in- changes associated with response to treatment, including
hibitor avelumab in patients whose disease had previously tertiary lymphoid structures and cholesterol clefts. These
progressed after chemotherapy. After at least 1 year of response criteria for immune-related pathologic response
follow-up, the response rate with avelumab treatment was are consistent across a number of cancer types.64 If major
33%, and the 1-year overall survival was 52%.58,59 The pathologic response and pathologic complete response are
impressive therapeutic effect of inhibiting the PD-1/PD-L1 proven to correlate significantly with survival outcomes, as

Hall et al
they do in breast cancer, this will clear the path for trials of metric was higher in patients with a response than without
new agents in the neoadjuvant setting. a response.67
In the adjuvant space, there are several ongoing random- Comprehensive genomic testing was performed on 317
ized trials of immune checkpoint inhibitor therapy versus MCC tumors to evaluate TMB, MCPyV genomic sequences,
placebo or observation, including studies examining ave- and the effects on clinical response.68 The most common
lumab (NCT03271372), pembrolizumab (NCT03712605), mutations were in TP53, RB1, NOTCH1, and P10 genes.
and nivolumab (NCT02196961). Low TMB was defined as fewer than six mutations per
megabase, and high TMB was defined as 20 or more
Since 2017, when avelumab became the first drug ever to
mutations per megabase. The MCC tumors clustered at the
be approved for the indication of MCC, immune checkpoint
high and low ends of the TMB spectrum, creating a bimodal
therapy has replaced chemotherapy for first-line treatment.
distribution. MCPyV genomic DNA sequences were not
Second-line options include chemotherapy or clinical trials.
identified in any of the 110 tumors that were high TMB.
Small case series suggest that alternate checkpoint in-
Upon examination of the tumors with low mutational bur-
hibitors (e.g., CTLA-4 antibodies) can result in therapeutic
den, MCPyV genomic DNA sequences were identified in
benefit in some patients with tumors that do not respond to
63% of them, or 110 of 175. In a subset of 57 patients
PD-1 or PD-L1 blockade.65 Multiple clinical trials are in-
treated with immune checkpoint inhibitors, the response
corporating radiation therapy alongside checkpoint in-
rate was similar in the TMB-high/VN-MCC tumors compared
hibitors, as MCC is exquisitely sensitive to radiation therapy.
with the TMB-low/VP-MCC tumors (50% vs. 41%; p = .63).
Other immuno-oncology approaches include other check-
This supports the immunogenicity of both subsets. In VP-
point inhibitors, oncolytic viruses, Toll-like receptor agonists,
MCC, the cell is immunogenic because it presents viral
cytokine therapy, vaccines, adoptive T-cell therapy, and
peptides. The remaining fraction of VN-MCCs are caused by
inhibition of the MDM2 protein. Selected approaches cur-
an accumulation of ultraviolet-induced mutations with
rently in human studies are summarized in Table 1.
a high TMB, and the neoantigens result in immunogenicity.
Biomarkers of response to immunotherapy The avelumab
and the pembrolizumab studies examined PD-L1 ex- Cutaneous Squamous Cell Carcinoma
pression and T-cell infiltrate as potential biomarkers, but Until recently, cytotoxic chemotherapy was commonly used
neither was predictive of response.60,66 In the avelumab for frontline treatment of metastatic CSCC with most com-
study, they also examined MCPyV status and frequency of mon regimens including cisplatin with a variety of part-
response but found no association. However, conclusions ners.69 Cetuximab, an anti-EGFR monoclonal antibody, has
are limited by underpowered analyses in these small been demonstrated to have activity against CSCC in small
studies. The proximity of PD-L1– and PD-1–bearing cells studies. A phase II study of 36 patients with unresectable or
may have higher utility in predicting response, as this metastatic CSCC showed an overall response rate (ORR) of
TABLE 1. Examples of Ongoing Investigational Immunotherapy Approaches in Nonmelanoma Skin Cancers

Disease Immunotherapy Strategy Clinical Trial
Merkel cell carcinoma Activated NK cells + IL-15 agonist NCT03853317
Gene-modified autologous T-cell therapy NCT03747484
Combination CTLA-4/PD-1 blockade 6 radiotherapy NCT03071406
Intralesional TLR agonist + PD-1 antibody NCT03684785
Radiation therapy + PD-1 antibody NCT03988647,
NCT03304639
CSCC Avelumab + cetuximab NCT03944941
PD-L1 + long-acting IL-7 compound NCT03901573
Intralesional autologous cancer cell/Streptococcal antigen vaccine (IFx-Hu2.0) NCT04160065
Intralesional Talimogene laherparepvec NCT03714828
Intralesional glatiramer acetate NCT03982212
BCC Cemiplimab NCT03132636
CTLA-4 + PD-1 combination immunotherapy NCT03521830
Abbreviations: NK, natural killer; IL, interleukin; TLR, Toll-like receptor; CSCC, cutaneous squamous cell carcinoma; BCC, basal cell carcinoma.

28% (22% partial response, 6% complete response). No- approved dose. Longer follow-up also suggested that re-
tably, the study required patients to have strong or moderate sponses to sonidegib can be durable (among patients with
EGFR expression by immunohistochemistry.70 Another anti- response, the median response was 26 months for locally
EGFR antibody, panitumumab, was studied in a single-arm advanced BCC and was 24 months for metastatic BCC).76
study of 16 patients with advanced CSCC and showed Despite the initial success of Hh inhibitors in BCC, the
similar efficacy (ORR of 31% with 19% partial response, majority of patients eventually experience relapse.77 Fur-
12% complete response).71 thermore, the toxicities (nausea, diarrhea, vomiting, hair
Based on the immunologic rationale described above, PD-1 loss, altered taste sensation, muscle spasms, and joint
blockade has been investigated. PD-1 blockade with aches) are quite substantial and frequently lead to
cemiplimab (an anti–PD-1 antibody) was studied in a phase treatment interruption and discontinuation.78 Hence, there
I/II study of patients with locally advanced or metastatic is a substantial unmet need for treatment of patients with
CSCC and produced a response rate of 47% in a cohort of advanced BCC.
59 patients (41% PR, 7% CR).72 The majority of responses Given the very high TMB in BCC, there is an emerging
were durable, exceeding 6 months in duration. Similar to the interest in using immunotherapy for Hh-refractory or -in-
MCC data, the response rates observed with cemiplimab in tolerant patients. The data on use of immune checkpoint
CSCC are among the highest seen in solid tumors. This trial inhibitors for metastatic BCC are mostly limited to case
ultimately led to the U.S. Food and Drug Administration reports at this time. A recent review found published case
(FDA) approval of cemiplimab for locally advanced or reports detailing activity with cemiplimab, nivolumab, and
metastatic CSCC on September 28, 2018. The current pembrolizumab in a total of eight patients.79 Some com-
FDA-approved dose is 350 mg every 3 weeks. The phase pleted and durable responses have been reported. Two
II clinical study of cemiplimab in patients with CSCC is patients with metastatic BCC were treated with pem-
ongoing and remains open to enrollment (NCT02760498), brolizumab and reportedly had complete responses.80
and it is also being studied in the adjuvant setting after One patient found to have BCC with biopsy-proven liver
surgery and radiation in patients with high risk of recur- metastases was treated with nivolumab and had an ongoing
rence (NCT03969004). Other checkpoint inhibitors are PR lasting at least 18 months.81 These anecdotal observa-
also being studied in advanced CSCC, including pem- tions and a strong rationale have led to a variety of ongoing
brolizumab (NCT03284424, NCT02883556) and nivolu- clinical trials with PD-1 antibodies, such as cemiplimab
mab (NCT04204837; Table 1). (NCT03132636), or CTLA-4/PD-1 combinations, such as
ipilimumab/nivolumab (NCT03521830; Table 1).
Basal Cell Carcinoma
The combination of the PD-1 antibody pembrolizumab with
For patients with locally advanced or metastatic BCC who a Hedgehog inhibitor has been assessed in a small study of
require systemic therapy, the first-line approach is usually patients with metastatic or unresectable BCC after prior Hh
Hh pathway inhibitors, two of which are currently FDA inhibitor exposure. A total of 16 patients were enrolled (nine on
approved (vismodegib and sonidegib). Vismodegib was pembrolizumab alone, seven on a pembrolizumab/vismodegib
initially approved in 2012 on the basis of a single-arm, phase combination). The ORR for all patients was 38% (44% for the
II study (ERIVANCE BCC) in patients with locally advanced pembrolizumab group and 29% for the dual-therapy group),
or metastatic BCC.73 In this study, the response rate for 63 and there were three with complete responses (one with
patients with locally advanced BCC was 43%, and the pembrolizumab monotherapy, two with dual therapy). There
complete response rate was 21%. Among 33 patients with was no sign of unexpected toxicity with the combination but no
metastatic BCC, the response rate was 30%. Overall, the clear signal that combining the two agents resulted in improved
median duration of response was 7.6 months. Long-term efficacy, albeit in small numbers.82
follow-up suggested that responses could be durable, with
median response durations of nearly 15 months in those CONCLUSION
with metastatic BCC and 26 months in those with locally The incidence of NMSC exceeds that of any other malig-
advanced BCC.74 Similarly, sonidegib was initially approved nancy. Although the overwhelming majority of BCC and
in 2015 on the basis of the BOLT study, a phase II, 1:2 CSCC cases are highly curable with surgical treatment, they
randomized trial of 230 patients who received one of two account for considerable morbidity, mortality, and public
doses of sonidegib (200 mg vs. 800 mg).75 This study health spending, given their common incidence. There are
showed ORRs for patients with locally advanced BCC of convincing clinical data that immunotherapy (particularly
58% for 200 mg sonidegib and 44% for 800 mg sonidegib, checkpoint inhibition) is a useful treatment strategy for those
but the response rates in patients with metastatic BCC were with advanced MCC and CSCC. The frequent and durable
only 8% for the 200-mg dose and 17% for the 800-mg dose. responses to PD-1/PD-L1–blocking antibodies confirm the
Ultimately, the 200-mg dose was selected as the FDA- importance of immune mechanisms in NMSC pathogenesis.

Hall et al
However, not all patients experience a response to immu- resistance are critical to uncover new rational therapies
notherapy, and some develop secondary resistance. Thus, to overcome these. Fortunately, there are a variety of in-
a key question remains as to what tumor or host charac- vestigational immunotherapy approaches currently un-
teristics might be used to predict response and/or re- derway to help improve upon the initial results of checkpoint
sistance. In addition to finding predictive biomarkers, there inhibitor immunotherapy.
is a dire unmet need for finding effective therapies in im-
munocompetent patients whose tumors do not respond to ACKNOWLEDGMENT
PD-1/PD-L1 blockade. Mechanistic studies to un- A. Silk, R. Dummer, and S. Bhatia contributed equally to this
derstand both intrinsic and acquired mechanisms of article.

1
Division of Medical Oncology, University of Washington, Seattle, WA Shailender Bhatia, MD, Medical Oncology, Department of Medicine,
2
Clinical Research Division, Fred Hutchinson Cancer Research Center, University of Washington/Fred Hutchinson Cancer Research Center, 825
Seattle, WA Eastlake Ave. E., Mailstop # CE2-102, Seattle, WA 98109; email:
3
Department of Dermatology, University Hospital of Zürich, Zürich, [email protected].
Switzerland
4
Dana-Farber Cancer Institute, Boston, MA
EDBK_278953.
REFERENCES
1. Fitzmaurice C, Abate D, Abbasi N, et al; Global Burden of Disease Cancer Collaboration. Global, regional, and national cancer incidence, mortality, years of life
lost, years lived with disability, and disability-adjusted life-years for 29 cancer groups, 1990 to 2017: a systematic analysis for the Global Burden of Disease Study.
JAMA Oncol. Epub 2019 Sep 27.
2. Asgari MM, Moffet HH, Ray GT, et al. Trends in basal cell carcinoma incidence and identification of high-risk subgroups, 1998-2012. JAMA Dermatol. 2015;
151:976-981.
3. Karia PS, Han J, Schmults CD. Cutaneous squamous cell carcinoma: estimated incidence of disease, nodal metastasis, and deaths from disease in the United
States, 2012. J Am Acad Dermatol. 2013;68:957-966.
4. Rogers HW, Weinstock MA, Feldman SR, et al. Incidence estimate of nonmelanoma skin cancer (keratinocyte carcinomas) in the U.S. population, 2012. JAMA
Dermatol. 2015;151:1081-1086.
5. Paulson KG, Park SY, Vandeven NA, et al. Merkel cell carcinoma: current US incidence and projected increases based on changing demographics. J Am Acad
Dermatol. 2018;78:457-463.e2.
6. Eigentler TK, Leiter U, Häfner HM, et al. Survival of patients with cutaneous squamous cell carcinoma: results of a prospective cohort study. J Invest Dermatol.
2017;137:2309-2315.
7. Harms PW. Update on Merkel cell carcinoma. Clin Lab Med. 2017;37:485-501.
8. Guy GP Jr., Machlin SR, Ekwueme DU, et al. Prevalence and costs of skin cancer treatment in the U.S., 2002-2006 and 2007-2011. Am J Prev Med. 2015;
48:183-187.
9. Yanik EL, Pfeiffer RM, Freedman DM, et al. Spectrum of immune-related conditions associated with risk of keratinocyte cancers among elderly adults in the
United States. Cancer Epidemiol Biomarkers Prev. 2017;26:998-1007.
10. Madeleine MM, Patel NS, Plasmeijer EI, et al; the Keratinocyte Carcinoma Consortium (KeraCon) Immunosuppression Working Group. Epidemiology of
keratinocyte carcinomas after organ transplantation. Br J Dermatol. 2017;177:1208-1216.
11. Penn I. Posttransplant malignancies. Transplant Proc. 1999;31:1260-1262.
12. Agelli M, Clegg LX, Becker JC, et al. The etiology and epidemiology of Merkel cell carcinoma. Curr Probl Cancer. 2010;34:14-37.
13. Engels EA, Frisch M, Goedert JJ, et al. Merkel cell carcinoma and HIV infection. Lancet. 2002;359:497-498.
14. Heath M, Jaimes N, Lemos B, et al. Clinical characteristics of Merkel cell carcinoma at diagnosis in 195 patients: the AEIOU features. J Am Acad Dermatol. 2008;
58:375-381.
15. Burack J, Altschuler EL. Sustained remission of metastatic Merkel cell carcinoma with treatment of HIV infection. J R Soc Med. 2003;96:238-239.
16. Muirhead R, Ritchie DM. Partial regression of Merkel cell carcinoma in response to withdrawal of azathioprine in an immunosuppression-induced case of
metastatic Merkel cell carcinoma. Clin Oncol (R Coll Radiol). 2007;19:96.
17. Feng H, Shuda M, Chang Y, et al. Clonal integration of a polyomavirus in human Merkel cell carcinoma. Science. 2008;319:1096-1100.

18. Harms PW, Harms KL, Moore PS, et al. The biology and treatment of Merkel cell carcinoma: current understanding and research priorities. Nat Rev Clin Oncol.
2018;15:763-776.
19. Dye KN, Welcker M, Clurman BE, et al. Merkel cell polyomavirus Tumor antigens expressed in Merkel cell carcinoma function independently of the ubiquitin
ligases Fbw7 and β-TrCP. PLoS Pathog. 2019;15:e1007543.
20. Houben R, Shuda M, Weinkam R, et al. Merkel cell polyomavirus-infected Merkel cell carcinoma cells require expression of viral T antigens. J Virol. 2010;
84:7064-7072.
21. Wong SQ, Waldeck K, Vergara IA, et al. UV-associated mutations underlie the etiology of MCV-negative Merkel cell carcinomas. Cancer Res. 2015;
75:5228-5234.
22. Goh G, Walradt T, Markarov V, et al. Mutational landscape of MCPyV-positive and MCPyV-negative Merkel cell carcinomas with implications for immunotherapy.
Oncotarget. 2016;7:3403-3415.
23. Paulson KG, Iyer JG, Tegeder AR, et al. Transcriptome-wide studies of Merkel cell carcinoma and validation of intratumoral CD8+ lymphocyte invasion as an
independent predictor of survival. J Clin Oncol. 2011;29:1539-1546.
24. Shuda M, Arora R, Kwun HJ, et al. Human Merkel cell polyomavirus infection I. MCV T antigen expression in Merkel cell carcinoma, lymphoid tissues and
lymphoid tumors. Int J Cancer. 2009;125:1243-1249.
25. Afanasiev OK, Yelistratova L, Miller N, et al. Merkel polyomavirus-specific T cells fluctuate with Merkel cell carcinoma burden and express therapeutically
targetable PD-1 and Tim-3 exhaustion markers. Clin Cancer Res. 2013;19:5351-5360.
26. Iyer JG, Afanasiev OK, McClurkan C, et al. Merkel cell polyomavirus-specific CD8+ and CD4+ T-cell responses identified in Merkel cell carcinomas and blood. Clin
Cancer Res. 2011;17:6671-6680.
27. Paulson KG, Carter JJ, Johnson LG, et al. Antibodies to Merkel cell polyomavirus T antigen oncoproteins reflect tumor burden in Merkel cell carcinoma patients.
Cancer Res. 2010;70:8388-8397.
28. Paulson KG, Lewis CW, Redman MW, et al. Viral oncoprotein antibodies as a marker for recurrence of Merkel cell carcinoma: a prospective validation study.
Cancer. 2017;123:1464-1474.
29. Schadendorf D, Nghiem P, Bhatia S, et al. Immune evasion mechanisms and immune checkpoint inhibition in advanced Merkel cell carcinoma. OncoIm-
munology. 2017;6:e1338237.
30. Paulson KG, Tegeder A, Willmes C, et al. Downregulation of MHC-I expression is prevalent but reversible in Merkel cell carcinoma. Cancer Immunol Res. 2014;
2:1071-1079.
31. Afanasiev OK, Nagase K, Simonson W, et al. Vascular E-selectin expression correlates with CD8 lymphocyte infiltration and improved outcome in Merkel cell
carcinoma. J Invest Dermatol. 2013;133:2065-2073.
32. Walter A, Barysch MJ, Behnke S, et al. Cancer-testis antigens and immunosurveillance in human cutaneous squamous cell and basal cell carcinomas. Clin
Cancer Res. 2010;16:3562-3570.
33. Wong SY, Seol AD, So PL, et al. Primary cilia can both mediate and suppress Hedgehog pathway-dependent tumorigenesis. Nat Med. 2009;15:1055-1061.
34. Asplund A, Gry Björklund M, Sundquist C, et al. Expression profiling of microdissected cell populations selected from basal cells in normal epidermis and basal
cell carcinoma. Br J Dermatol. 2008;158:527-538.
35. Al-Rohil RN, Tarasen AJ, Carlson JA, et al. Evaluation of 122 advanced-stage cutaneous squamous cell carcinomas by comprehensive genomic profiling opens
the door for new routes to targeted therapies. Cancer. 2016;122:249-257.
36. Pickering CR, Zhou JH, Lee JJ, et al. Mutational landscape of aggressive cutaneous squamous cell carcinoma. Clin Cancer Res. 2014;20:6582-6592.
37. Bonaventura P, Shekarian T, Alcazer V, et al. Cold tumors: a therapeutic challenge for immunotherapy. Front Immunol. 2019;10:168.
38. Agrawal S, Reemtsma K, Bagiella E, et al. Role of TAP-1 and/or TAP-2 antigen presentation defects in tumorigenicity of mouse melanoma. Cell Immunol. 2004;
228:130-137.
39. Hua LA, Kagen CN, Carpenter RJ, et al. HLA and beta 2-microglobulin expression in basal and squamous cell carcinomas of the skin. Int J Dermatol. 1985;
24:660-663.
40. Otsuka A, Dreier J, Cheng PF, et al. Hedgehog pathway inhibitors promote adaptive immune responses in basal cell carcinoma. Clin Cancer Res. 2015;
21:1289-1297.
41. Urosevic M, Maier T, Benninghoff B, et al. Mechanisms underlying imiquimod-induced regression of basal cell carcinoma in vivo. 2003;139:1325-1332.
42. Wei H, Hongya P, Linlin J, et al. IFN-γ enhances the anti-tumour immune response of dendritic cells against oral squamous cell carcinoma. Arch Oral Biol. 2011;
56:891-898.
43. Kooy AJ, Prens EP, Van Heukelum A, et al. Interferon-gamma-induced ICAM-1 and CD40 expression, complete lack of HLA-DR and CD80 (B7.1), and
inconsistent HLA-ABC expression in basal cell carcinoma: a possible role for interleukin-10? J Pathol. 1999;187:351-357.
44. Yesantharao P, Wang W, Ioannidis NM, et al. Cutaneous squamous cell cancer (cSCC) risk and the human leukocyte antigen (HLA) system. Hum Immunol.
2017;78:327-335.
45. Urosevic M, Dummer R. Immunotherapy for nonmelanoma skin cancer: does it have a future? Cancer. 2002;94:477-485.
46. Omland SH, Nielsen PS, Gjerdrum LMR, et al. Immunosuppressive environment in basal cell carcinoma: the role of regulatory T cells. Acta Derm Venereol. 2016;
96:917-921.

Hall et al
47. Lai C, August S, Albibas A, et al. Ox40+ regulatory T cells in cutaneous squamous cell carcinoma suppress effector T-cell responses and associate with metastatic
potential. Clin Cancer Res. 2016;22:4236-4248.
48. Corinti S, Albanesi C, la Sala A, et al. Regulatory activity of autocrine IL-10 on dendritic cell functions. J Immunol. 2001;166:4312-4318.
49. Bluth MJ, Zaba LC, Moussai D, et al. Myeloid dendritic cells from human cutaneous squamous cell carcinoma are poor stimulators of T-cell proliferation. J Invest
Dermatol. 2009;129:2451-2462.
50. Elamin I, Zecević RD, Vojvodić D, et al. Cytokine concentrations in basal cell carcinomas of different histological types and localization. Acta Dermatovenerol Alp
Panonica Adriat. 2008;17:55-59.
51. Yamada S, Jinnin M, Kajihara I, et al. Cytokine expression profiles in the sera of cutaneous squamous cell carcinoma patients. Drug Discov Ther. 2016;
10:172-176.
52. Gambichler T, Skrygan M, Kaczmarczyk JM, et al. Increased expression of TGF-beta/Smad proteins in basal cell carcinoma. Eur J Med Res. 2007;12:509-514.
53. Azzimonti B, Zavattaro E, Provasi M, et al. Intense Foxp3+ CD25+ regulatory T-cell infiltration is associated with high-grade cutaneous squamous cell carcinoma
and counterbalanced by CD8+/Foxp3+ CD25+ ratio. Br J Dermatol. 2015;172:64-73.
54. Tai PTH, Yu E, Winquist E, et al. Chemotherapy in neuroendocrine/Merkel cell carcinoma of the skin: case series and review of 204 cases. J Clin Oncol. 2000;
18:2493-2499.
55. Iyer JG, Blom A, Doumani R, et al. Response rates and durability of chemotherapy among 62 patients with metastatic Merkel cell carcinoma. Cancer Med. 2016;
5:2294-2301.
56. Allen PJ, Bowne WB, Jaques DP, et al. Merkel cell carcinoma: prognosis and treatment of patients from a single institution. J Clin Oncol. 2005;23:2300-2309.
57. Lipson EJ, Vincent JG, Loyo M, et al. PD-L1 expression in the Merkel cell carcinoma microenvironment: association with inflammation, Merkel cell polyomavirus
and overall survival. Cancer Immunol Res. 2013;1:54-63.
58. Kaufman HL, Russell J, Hamid O, et al. Avelumab in patients with chemotherapy-refractory metastatic Merkel cell carcinoma: a multicentre, single-group, open-
label, phase 2 trial. Lancet Oncol. 2016;17:1374-1385.
59. Kaufman HL, Russell JS, Hamid O, et al. Updated efficacy of avelumab in patients with previously treated metastatic Merkel cell carcinoma after 1 year of follow-
up: JAVELIN Merkel 200, a phase 2 clinical trial. J Immunother Cancer. 2018;6:7.
60. Nghiem P, Bhatia S, Lipson EJ, et al. Durable tumor regression and overall survival in patients with advanced Merkel cell carcinoma receiving pembrolizumab as
first-line therapy. J Clin Oncol. 2019;37:693-702.
61. D’Angelo SP, Russell J, Lebbé C, et al. Efficacy and safety of first-line avelumab treatment in patients with stage IV metastatic Merkel cell carcinoma: a preplanned
interim analysis of a clinical trial. JAMA Oncol. 2018;4:e180077.
62. D’Angelo SP, Hunger M, Brohl AS, et al. Early objective response to avelumab treatment is associated with improved overall survival in patients with metastatic
Merkel cell carcinoma. Cancer Immunol Immunother. 2019;68:609-618.
63. Topalian S, Bhatia S, Kudchadkar R, et al. Nivolumab (Nivo) as neoadjuvant therapy in patients with resectable Merkel cell carcinoma (MCC) in CheckMate 358.
J Clin Oncol. 2018;36:9505.
64. Stein JE, Lipson EJ, Cottrell TR, et al. Pan-tumor pathologic scoring of response to PD-(L)1 blockade. Clin Cancer Res. 2019;26:545-551.
65. LoPiccolo J, Schollenberger MD, Dakhil S, et al. Rescue therapy for patients with anti-PD-1-refractory Merkel cell carcinoma: a multicenter, retrospective case
series. J Immunother Cancer. 2019;7:170.
66. Shapiro I, Grote H, D’Urso V, et al. Exploratory biomarker analysis in avelumab-treated patients with metastatic Merkel cell carcinoma progressed after
chemotherapy. J Clin Oncol. 2017;35:15s (suppl; abstr 9557).
67. Giraldo NA, Nguyen P, Engle EL, et al. Multidimensional, quantitative assessment of PD-1/PD-L1 expression in patients with Merkel cell carcinoma and
association with response to pembrolizumab. J Immunother Cancer. 2018;6:99.
68. Knepper TC, Montesion M, Russell JS, et al. The genomic landscape of Merkel cell carcinoma and clinicogenomic biomarkers of response to immune checkpoint
inhibitor therapy. Clin Cancer Res. 2019;25:5961-5971.
69. Gellrich FF, Hüning S, Beissert S, et al. Medical treatment of advanced cutaneous squamous-cell carcinoma. J Eur Acad Dermatol Venereol. 2019;33:38-43.
70. Maubec E, Petrow P, Scheer-Senyarich I, et al. Phase II study of cetuximab as first-line single-drug therapy in patients with unresectable squamous cell
carcinoma of the skin. J Clin Oncol. 2011;29:3419-3426.
71. Foote MC, McGrath M, Guminski A, et al. Phase II study of single-agent panitumumab in patients with incurable cutaneous squamous cell carcinoma. Ann Oncol.
2014;25:2047-2052.
72. Migden MR, Rischin D, Schmults CD, et al. PD-1 blockade with cemiplimab in advanced cutaneous squamous-cell carcinoma. N Engl J Med. 2018;
379:341-351.
73. Sekulic A, Migden MR, Oro AE, et al. Efficacy and safety of vismodegib in advanced basal-cell carcinoma. N Engl J Med. 2012;366:2171-2179.
74. Sekulic A, Migden MR, Basset-Seguin N, et al; ERIVANCE BCC Investigators. Long-term safety and efficacy of vismodegib in patients with advanced basal cell
carcinoma: final update of the pivotal ERIVANCE BCC study. BMC Cancer. 2017;17:332.
75. Dummer R, Guminski A, Gutzmer R, et al. The 12-month analysis from Basal Cell Carcinoma Outcomes with LDE225 Treatment (BOLT): a phase II, randomized,
double-blind study of sonidegib in patients with advanced basal cell carcinoma. J Am Acad Dermatol. 2016;75:113-125.e5.

76. Lear JT, Migden MR, Lewis KD, et al. Long-term efficacy and safety of sonidegib in patients with locally advanced and metastatic basal cell carcinoma: 30-month
analysis of the randomized phase 2 BOLT study. J Eur Acad Dermatol Venereol. 2018;32:372-381.
77. Herms F, Lambert J, Grob J-J, et al. Follow-up of patients with complete remission of locally advanced basal cell carcinoma after vismodegib discontinuation:
a multicenter French study of 116 patients. J Clin Oncol. 2019;37:3275-3282.
78. Basset-Séguin N, Hauschild A, Kunstfeld R, et al. Vismodegib in patients with advanced basal cell carcinoma: primary analysis of STEVIE, an international, open-
label trial. Eur J Cancer. 2017;86:334-348.
79. Choi FD, Kraus CN, Elsensohn AN, et al. Programmed cell death 1 protein and programmed death-ligand 1 inhibitors in the treatment of nonmelanoma skin
cancer: a systematic review. J Am Acad Dermatol. 2019;82:440-459.
80. Cannon JGD, Russell JS, Kim J, et al. A case of metastatic basal cell carcinoma treated with continuous PD-1 inhibitor exposure even after subsequent initiation of
radiotherapy and surgery. JAAD Case Rep. 2018;4:248-250.
81. Cohen PR, Kato S, Goodman AM, et al. Appearance of new cutaneous superficial basal cell carcinomas during successful nivolumab treatment of refractory
metastatic disease: implications for immunotherapy in early versus late disease. Int J Mol Sci. 2017;18:E1663.
82. Chang ALS, Tran DC, Cannon JGD, et al. Pembrolizumab for advanced basal cell carcinoma: an investigator-initiated, proof-of-concept study. J Am Acad
Dermatol. 2019;80:564-566.

PEDIATRIC ONCOLOGY

PEDIATRIC ONCOLOGY
Crossing Oceans: Preclinical Collaboration to

Improve Pediatric Drug Development
Gregory Reaman, MD1; Louis Stancato, PhD2; Gilles Vassal, MD3; and John M. Maris, MD4
overview
Changes in the regulatory environment affecting pediatric cancer drug development in the United States and
the European Union provide unprecedented opportunity to advance the concept of precision medicine to
children with cancer. Increasing evidence suggests that new drugs and biologic products directed at molecular
targets presumed to be etiologically associated with many adult cancers may well provide therapeutic options
for selected subsets of children with cancer despite their histologic and biologic differences. Regulatory
requirements for early evaluation of appropriate new drugs for children based on their molecular mechanism of
action, rather than the specific clinical indications for which they are developed and/or approved, will shorten
the unacceptable time lag between first-in-human and first-in-children studies. The relative scarcity of
pediatric patients eligible for biomarker-directed studies and the ever-expanding compendium of new targeted
agents mandate rational, science-based decision-making in selecting and prioritizing appropriate drugs to
study early in development. A critical component of the evidence base in such decision-making includes
preclinical testing of relevant drugs in pediatric tumor-specific in vitro and in vivo models. Established
preclinical testing programs with academic investigator–industry collaborations are actively engaged in such
activities. International collaboration is required to address the resource constraints and increasing number of
potential products to be tested in a timely, efficient, nonduplicative, and cost-effective manner.
BACKGROUND AND CASE STATEMENT: THE EVOLVING be evaluated with children. In both the United States
REGULATORY ENVIRONMENT and the European Union, drug development for chil-
Despite the substantial advances in childhood cancer dren is subject to legislation that was intended to foster
outcomes during the past 5 decades, cancer remains and support the development and ultimately approval
the leading cause of death from disease among chil- of effective new drugs for children. The effect of this
dren, largely owing to the initial refractoriness of some legislation on the timely assessment and approval of
cancers or the development of resistance mechanisms effective new drugs for children has been markedly
and disease recurrence.1,2 Although the current overall more obvious in nearly all clinical conditions compared
5-year survival rates approach 85%, the cost of cure is with cancer.
high, because 30% to 50% of long-term survivors
The Pediatric Research Equity Act was enacted in
experience serious long-term, life-altering, and life-
2003 and authorizes the FDA to require pediatric
threatening sequelae.3,4 The need for more effective
assessment of efficacy and safety of any new drug that
and less toxic drugs for childhood cancer remains
is the subject of a New Drug or Biologics Licensing
substantially unmet. Cancer drug development for Application for any indication that also exists for chil-
children has largely leveraged adult cancer drug dis- dren unless the requirement is waived for study fea-
Author affiliations
covery and development; however, the demonstration sibility, likelihood of use, or toxicity concerns; products
and support of effectiveness and approval for use of new anticancer developed for orphan-designated indications are ex-
information (if agents for children has lagged seriously behind de- empt from the Pediatric Research Equity Act require-
applicable) appear velopment and approval of new drugs for cancers that
at the end of this
ments.6 There have never been Pediatric Research
article.
occur predominantly among adults. A recent report Equity Act–required studies of an oncology drug for
Accepted on
demonstrated a median lag time from first-in-human children. A pediatric exclusivity provision was enacted
February 28, 2020 studies of those new agents ultimately approved by the under the Food and Drug Administration Moderni-
and published at U.S. Food and Drug Administration (FDA) during the zation Act in 1997 and reauthorized as the Best
ascopubs.org on period from 1997 to 2017 to first-in-child studies of Pharmaceuticals for Children Act in 2002, which
March 24, 2020:
6.5 years (range, 0–28 years).5 The explanation for the provides a financial incentive to industry sponsors
10.1200/EDBK_ delay is multifactorial but attributable only to a minimal for voluntarily evaluating drugs among the pediatric
278893 extent by a scientific rationale for why a drug should not population through fulfillment of a Written Request.7

Reaman et al
Advances in the understanding of the molecular biology and

genetic epidemiology of human cancer have transformed
the overall paradigm for cancer drug discovery and de-
• Changes in the regulatory environment in the velopment.10 The advent of molecularly targeted drugs has
United States and European Union have the
advanced the concept of precision medicine in oncology.
potential to advance the concept of precision
Although the science of precision medicine development
medicine to children with cancer.
is less mature for children, increasing evidence implicates
• Preclinical testing of targeted drug products established molecular drivers of certain cancers prevalent
developed for cancers that occur among adults
among adults as having a causative role in childhood
and potentially relevant to cancers that occur
predominantly among children can contribute cancers across a spectrum of histologies.11,12 The evidence
to the science-based rationale for pediatric for considering that a specific molecular target may be
assessment. substantively relevant to convey the same vulnerability to
a molecule directed at that target and effective in a cancer
• Functional but insufficiently resourced
industry–academic investigator collaborations among adults varies considerably among specific targets
exist in the United States and European Union and across a variety of tumor types. The relevance of a target
to address this need. and decision-making about early investigation of drugs
directed at a target are further complicated by the differ-
• The large number of molecules in development
for which evaluation with children may be ap- ences in the degree of unmet clinical need across pediatric
propriate and the existing resource limitations cancer diagnoses. There is, therefore, a need to critically
mandate timely, efficient, and cost-effective evaluate and expand, when possible and necessary, the
preclinical testing. evidence base supporting the concept that a molecular
• International collaboration is necessary to pro- target is relevant to the etiology or progression of a pediatric
vide acceptable preclinical data packages to cancer and that a new drug product directed at such a target
inform decisions regarding the justification and may exert a deleterious effect on a pediatric cancer.
prioritization of pediatric clinical evaluations. The FDA and the National Cancer Institute (NCI) were
legislatively mandated to create an initial list of relevant
molecular targets, which was published on the FDA website
Both laws were permanently reauthorized in 2012. The in August 2018, consists of more than 200 distinct entities,
European Pediatric Regulation (EC 1901/2006), which and is likely to change with emerging scientific discovery.13
came into force in 2007,8 requires pharmaceutical com- The Relevant Pediatric Molecular Target list is intended to
panies seeking a marketing authorization for a new drug or guide rather than dictate decision-making related to de-
a new indication for adults to create a pediatric development velopment of specific targeted drugs for children based on
program in the form of a Pediatric Investigation Plan. The molecular mechanism of action. This list is essentially based
regulation allows for waivers of pediatric studies by the on peer-reviewed publications and publicly available da-
Pediatric Committee of the European Medicines Agency tabases of broad pediatric cancer genomic sequencing
based on adult indication, safety, and potential efficacy. The results demonstrating associations with both hematologic
regulation provided for a list of class waivers serially mod- and solid tumor malignant neoplasms.13 The framework
ified in 2008, 2010, 2015, and 2017. Removing class used to construct the list included the following: genomic
waivers permits changes to the requirement for pediatric data indicating an association of a specific gene pertur-
assessment from merely clinical indication to mechanism of bation/mutation, deletion, copy number variation, or protein
action. overexpression in one or more pediatric cancers irrespective
Section 504 of the Food and Drug Administration Reau- of frequency; functional genomic evidence that a gene
thorization Act of 2017 amended the Pediatric Research defect relates with another cellular pathway to result in
Equity Act in Section 505B of the Food, Drug, and Cosmetic synthetic lethality; nonclinical evidence or adult clinical
Act9 to require the sponsor of an original New Drug Ap- evidence that target modulation effects tumor cell sus-
plication or Biologics Licensing Application submitted on or tainability or growth; and the existence of predictive and/or
after August 18, 2020, to conduct an early pediatric clinical response biomarkers. Target actionability and the conse-
trial of a new drug, intended for the treatment of an adult quence of target inhibition in specific pediatric cancers is
cancer and directed at a molecular target that the FDA incomplete. The preclinical evaluation of specific target-
determines to be substantially relevant to the growth or directed therapeutics using pediatric-specific models (e.g.,
progression of a pediatric cancer to yield meaningful data cell lines and organoids for high throughput in vitro
regarding dosing, safety, and preliminary efficacy to inform screening, patient-derived xenografts [PDXs], orthotopic
potential pediatric labeling. xenografts, and human-engineered xenografts, particularly

Pediatric Preclinical Testing
for targets applicable to immune-directed therapy approaches) stocked with targeted agents with unlocked pediatric po-
can add substantially to the existing evidence base of tar- tential, must be an active participant in identifying safer and
get relevance and rational decision-making regarding the more effective therapies for children with cancer, but how?
appropriateness of early pediatric clinical assessment. Many Unfortunately, even with a supportive regulatory environ-
established targets in tumor cells, innate immune cells, and ment, the path toward the development of compelling
other cellular elements of the tumor microenvironment preclinical data packages in support of clinical development
may feasibly direct drug development strategies, including is muddled at best. Unlike counterparts in adult cancer
early evaluation, and provide a rationale for therapeutic development, there is a paucity of tools readily available to
intervention. Practically speaking, this creates many ther- the industry researcher interested in assessing drug ac-
apeutic possibilities that greatly exceed the number of tivity in the pediatric setting. Outside of the NCI-funded
children both within specific and across cancer diagnoses Preclinical Pediatric Testing Consortium (PPTC),16 which
that can be allocated to specific biomarker-enriched pop- provides access to efficacy testing in a wide range of pe-
ulations for clinical trials in which new targeted drugs can be diatric cancer in vivo models (albeit with limited budget
adequately tested for effectiveness and safety. Cancer drug and capacity), options are extremely limited, and what is
development is a global enterprise, and early assessment available is often scattered across academic centers glob-
of relevant new drugs for children mandates international ally. In a recent survey of industry conducted by the ITCC-P4
coordination and clinical trial collaboration, given the in- (Innovative Therapies for Children With Cancer Pediatric
creasingly small number of patients available for clinical Preclinical Proof-of-Concept Platform) consortium, most of
trials, particularly in the context of biomarker-directed trials. the industry representatives cited the lack of available,
Building the scientific rationale for assessing specific new well-characterized, and relevant pediatric models as a major
drugs for children requires a strengthened evidence base hurdle for pediatric development (pediatric evaluation not
achievable in part through robust preclinical testing of drug being a company priority and insufficient resources/funding
products of interest that similarly mandates international were also cited).17 Furthermore, the expertise (and frankly,
collaboration. the will) required to efficiently develop drugs for children
with cancer is often lacking across industry. As a result, it is
SETTING THE STAGE FOR INDUSTRY COLLABORATION imperative that stronger ties develop between industry and
The hurdles to developing drugs specifically for pediatric academia to effectively address the unmet medical need
patients with cancer (listed in Table 1) are well known and in pediatric oncology, particularly in the relapsed/refractory
have remained relatively unchanged for years.14 However, setting. The development of robust preclinical data pack-
as described above, the evolving regulatory environment ages matching tumor vulnerabilities with targeted agents is
creates an important opportunity for the biopharmaceutical a necessary first step in addressing this need.
industry (henceforth “industry”) to markedly increase pe- As the molecular drivers of pediatric tumors are identi-
diatric oncology research and development, commensurate fied, opportunities will continue to arise to match targeted
with its position as the major funder of biomedical research therapies with corresponding pediatric tumor drivers.18,19
in the United States.14,15 Industry, with clinical pipelines well Examples such as dasatinib in Philadelphia chromosome–
TABLE 1. Challenges to Pediatric Cancer Drug Development

Pediatric Challenge Practical Considerations
Most prevalent of the known driver mutations may be the nondruggable Bold, fresh approaches to drug targeting needed
fusion oncoproteins/transcription factors
Lack of clarity of short-term/long-term developmental consequences of Clinical data must be generated and evaluated
signaling pathway disruption in young children
Concerns about off-target toxicities of agents with less than highly specific Clinical data must be generated and evaluated
activity
Formulation concerns Begin formulation discussions much earlier in development.
Limited number of patients—must prioritize Preclinical data key to decision-making and global cooperation for clinical
evaluation
Limited number of patients—not profitable or attractive to industry Patient need, not market size, needs to be a driver; additional incentives
to sponsors?
Development expertise primarily found within academic centers; industry Stronger industry ties with academia
has little experience

Reaman et al
positive acute lymphoblastic leukemia (ALL); crizotinib drugs in industry pipelines, for which mechanisms of action
treatment of anaplastic lymphoma kinase–mutated neuro- may match known pediatric cancer vulnerabilities, remains
blastoma and anaplastic lymphoma kinase fusion–positive one of the best options for identifying newer agents for
anaplastic large cell lymphoma and inflammatory myofi- clinical development (with the potential exception of immuno-
broblastic tumor; and larotrectinib in neurotrophic tyrosine oncology agents, for which there are few relevant preclinical
receptor kinase fusion–positive pediatric tumors reduce to pediatric models).
practice the concept of precision medicine in pediatric on-
Clearly, academia is industry’s best option for the expertise
cology and offer hope for the future while also highlighting
necessary to effectively identify and target pediatric tumor
collaborative efforts between industry and academia.20-22
vulnerabilities. The payoff in terms of safer and more ef-
Unfortunately, the pace of discovery of tumor drivers is
fective medicines from years of academic research will
clearly outstripping industry’s development of potentially
most likely be realized through a merging of the in-depth
promising agents, creating a deadly clinical gap. To fill this
academic science with the drug development expertise
gap, there is a recent and steady increase in collaboration
of industry. For example, novel approaches to targeting
between industry and the public sector (both academia and
“nondruggable” targets, such as fusions (e.g., EWS-FLI1),
government) on the individual institution level all the way up
are more likely to arise out of academic rather than industry
to precompetitive public-private partnerships (PPPs), such
laboratories, where development timelines are hastened
as the ITCC-P4 consortium (explained in the next section)
and the need for quick go/no-go decisions is greater. In
and a burgeoning PPP spearheaded by the Foundation for
cases such as these, leveraging academia’s scientific know-
the National Institutes of Health. The unique challenges
how with industry’s unique ability to identify and develop
associated with pediatric drug development all but ensure
targeted agents is key to success and should result in
that multistakeholder collaboration is required for success,
compelling preclinical data packages that serve as sub-
and they provide the rationale for the importance of pre-
strates for clinical trial decisions. When used appropriately,
clinical testing using pediatric-specific models for new
preclinical data should help identify the best molecules to
drugs that may be potentially relevant (Table 1). Further-
advance to clinical trials, which is of importance in pediatric
more, fresh approaches to data sharing and transparency,
development, in which the number of patients available for
a willingness to prioritize drugs for clinical development
clinical trials is exceedingly small. From a clinical stand-
based on scientific merit, and a streamlining of the legal
point, many academic researchers are often associated with
agreement/technology transfer process are all elements
pediatric hospitals and are therefore able to move promising
critical to advancing promising therapies to the clinic.
treatment options from bench to bedside in a way not easily
In an ideal world, pediatric cancer research would be best accessible to industry; all that these researchers are lacking
served in a precompetitive setting, whereby data generated is systematic access to promising new therapies. These are
are shared publicly; however, the desire to protect current just a couple of the many reasons why industry is beginning
intellectual property while ensuring that future intellectual to develop deeper ties with the pediatric research com-
property is secured is omnipresent across both industry and munity. This deepening relationship is already contributing
academia and can be a substantial barrier to collaboration. to advances in the understanding of target biology in pe-
Unlike adult cancers, in which patient populations can run diatric cancers (e.g., “target actionability reviews”),23 led
into the hundreds and even thousands, pediatric cancers are to the development of an international consensus on the
orphan diseases, and the corresponding market for treatment depth and breadth of preclinical data required for moving
is exceptionally small. This creates an interesting paradox, a promising agent into pediatric clinical development,24 and
whereby small patient populations can deter industry from resulted in the creation of precompetitive PPPs that bring
clinical development, yet, when a positive development de- academia and industry together to identify promising drugs
cision is considered, it may be held up by perceived or for children dying from cancer. Of note regarding PPPs is
potential intellectual property considerations despite dealing the formal voice given to the critical but often overlooked
with the same small population. Industry and academia must contribution of patient advocacy, which plays a vital role in
find common ground on the issue of intellectual property and building bridges to all stakeholders while maintaining
concentrate on advancing a field that has been neglected for a patient focus. Depending on how they are structured,
too long, particularly in the generation and sharing of pre- PPPs can bring in government funding, further supporting
clinical data. Fortunately, the regulatory realities are en- the depth and breadth of the PPP (e.g., the E.U. IMI2
couraging industry to adjust pediatric development strategy [Innovative Medicines Initiative 2] program). Overall, in the
(and, in some cases, develop a strategy), which will almost past few years, there has been an increase in pediatric
certainly fuel greater collaboration with academia. From clinical development across industry as a result of the
any point of view, this development can only be viewed as growing number of collaborative efforts, and this trend must
a positive, especially as preclinical evaluation of promising continue.

Between 2000 and 2015, two drugs from Eli Lilly and ITCC-P4 is a PPP with 14 academic institutions, five
Company (“Lilly”) were evaluated in the pediatric cancer pharmaceutical companies (namely, Lilly, Roche, Pfizer,
setting. Since 2015, six targeted Lilly oncology agents Bayer, and Pharma Mar), and three preclinical contract
(nearly one dozen clinical trials) have entered the clinic, with research organizations (namely, Charles River/Oncotest,
many of the trials enabled through collaboration with both XenTech, and EPO). The aim is to establish a representative
academia and the NCI/PPTC. Underpinning many of these collection of 400 PDX models of pediatric high-risk malig-
trials were preclinical data packages that suggested com- nant solid and brain tumors, as well as leukemias, including
pelling preclinical activity and a mechanism of action deemed a substantial proportion of models from relapsed cancers
relevant to one or more pediatric cancers. A poignant ex- as well as genetically engineered mouse models for each
ample is the CHEK 1 (checkpoint kinase 1) small molecule indication (target: 15–20 models). A pilot evaluation of
inhibitor prexasertib, currently in clinical testing for multi- humanized mouse models of neuroblastoma and rhabdo-
ple pediatric malignancies (ClinicalTrials.gov identifiers: myosarcoma is ongoing, as is a pilot evaluation on the de-
NCT02808650, NCT04095221, and NCT04023669). The velopment of organoids. All PDX models and their matching
collaborative development of prexasertib was a model of primary tumor samples and germline controls are fully mo-
data sharing and transparency across multiple academic lecularly characterized (low-coverage whole-genome and
and government programs, which enabled rational, data- high-coverage whole-exome sequencing, RNA sequenc-
based decisions on progression to the clinic. 25 In the ing, methylation). Each model is further pharmacologically
absence of extensive collaboration, which entailed co- characterized by in vivo testing that includes two standard-of-
development of robust preclinical data packages and the care drugs, one standard-of-care combination, and six tar-
development of clinical protocols, it is unlikely that pre- geted agents, selected by entity experts. Testing is also being
xasertib would currently be in pediatric clinical testing. conducted in an organoid pilot study. As of December 2019,
The development of innovative therapies for children with 150 models have been established and fully characterized.
cancer begins with the identification of tumor vulnerabilities An open access repository of the data generated from the
that can be matched with corresponding targeted agents, project via the IMI-funded R2 database has been established
with preclinical research serving as the bridge to the clinic. (https://hgserver1.amc.nl/cgi-bin/r2/main.cgi).
With so many agents currently in clinical development, ITCC-P4 has also established a methodology for the sys-
preclinical pediatric data will continue to increase in im- tematic in silico evaluation of selected compounds and
portance as we search for the “best” molecules to advance corresponding actionable molecular targets and path-
to clinical trials. More than ever, success requires close, ways in the different pediatric solid tumor entities (target
transparent collaboration between industry and academia, actionability reviews) before the in vitro and in vivo testing
with completely transparent and open data sharing and phases to comprehensively assess relevant literature and
communication, a willingness to challenge the old conven- the existing (epi)genomic and transcriptomic data sets of all
tions, and a spirit of collegiality that will ultimately benefit our entities in question.23 To facilitate the design of preclinical
youngest patients with cancer. packages for regulatory purposes, a joint ITCC-P4 and PPTC
multistakeholder workshop with representatives from reg-
E.U. PERSPECTIVE ON ACADEMIA–INDUSTRY ulatory networks and industry, patient advocates, and ac-
COLLABORATION: DEVELOPMENT AND STATUS OF ademia established an international consensus on minimum
THE ITCC-P4 preclinical testing requirements for the evaluation of com-
In 2017, the 5-year ITCC-P4 project was launched as part of pounds in pediatric oncology.24
the European IMI2 PPP, with the goal to build a sustainable The ultimate goal is to deliver a sustainable high-quality
preclinical testing platform in pediatric oncology for both preclinical testing platform with unique, fully characterized
commercial and academic use. This European project was models and access to molecular information to serve the
designed to address the anticipated increased needs for needs of both the biopharmaceutical industry as part of the
pediatric preclinical evaluation to rationalize and facilitate new regulatory environment and the academic research
prioritization of anticancer compounds for pediatric devel- community in an effort to continue to gain knowledge on
opment within the new international regulatory environ- pediatric malignancies and to drive pediatric oncology drug
ment. ITCC-P4 aims to provide quality-assured, upfront development through science in a much more rigorous and
preclinical testing of novel mechanism of action–based systematic way. Companies will have access to compound
compounds in a (saturating) repertoire of molecularly and testing on ITCC-P4 models performed by contract research
pharmacologically well-characterized models to establish organizations as well as to biologic data and academic
the basis for increasing therapeutic success of new drugs for expertise to best design preclinical evaluation. Data gen-
children with solid tumors (and leukemias within a currently erated by the contract research organizations for a company
planned extension). will be owned by the company, and data generated by

Reaman et al
academic groups will be owned by those groups. In addi- been targeting specific activating mutations, allowing models
tion, companies and academic groups will be encouraged to be selected based on exome sequencing data, increasing
to share all or part of their testing results anonymously to numbers of candidate molecules are immuno-oncology agents
continuously increase the value and relevance of the panel directed against plasma membrane proteins. Hence, RNA
of models for the sake of science and relevance of any sequencing data, especially compared with normal tissues, is
additional testing. increasingly used to select models for testing. The PPTC is
currently generating reverse-phase protein array data, data for
PEDIATRIC PRECLINICAL TESTING IN A PUBLICLY FUNDED
cell membrane proteins utilizing sucrose gradient ultracen-
ACADEMIC CONSORTIUM
trifugation coupled to mass spectrometry, and whole-genome
The concept for a childhood cancers preclinical testing sequences from the osteosarcoma models.
academic collaborative research initiative was formulated
Although the list is certainly not comprehensive, major take-
during and after an NCI and Children’s Oncology Group–
home points from the PPTP/PPTC experience to date in-
sponsored meeting cochaired by Drs. Peter Houghton,
clude the following:
Malcolm Smith, and Peter Adamson in 2001.26 The major
outcome of the meeting was identification of the key sci- • PPPs to engage pharmaceutical and biotechnology
entific issues and infrastructural requirements to establish companies in testing their later-stage drugs in ultrarare
a program focused on testing novel anticancer agents in childhood cancers can be accomplished with mod-
credentialed models of childhood cancers. The Pediatric erate throughput (approximately 10 compounds evaluated
Preclinical Testing Program (PPTP) was established in 2003 annually across a wide range of cancer).
through an NCI federal contract with five testing sites in • PDX models are robust for testing drugs targeting tumor cell
the United States and one in Australia. The program in- intrinsic mechanisms on childhood cancer oncogenesis.
cluded a moderate-throughput in vitro cellular cytotoxicity • The majority of the compounds screened across PDX
screening program and PDX models of ALL, neuroblastoma, models agnostic to a mechanism of action show little if
osteosarcoma, brain cancers, and soft tissue sarcomas. The any antitumor efficacy, which allows decision-making to
PPTP evaluated 116 drugs during a decade, including limit additional time and effort in the pediatric setting for
standard-of-care drugs in the initial year of the program, to these agents so that investigators can focus on more
document sensitivity or resistance to commonly used promising agents.
chemotherapeutics. The PPTP was organized in a “top-
down” manner, in which program leaders negotiated with
TABLE 2. The Pediatric Preclinical Testing Consortium
industry to procure drugs, and testing was done in a blinded
Principal
fashion by research sites in a model similar to a contract Role Institution Investigator
research organization.
Coordinating RTI International (formerly Greg Gatto, PhD
The NCI subsequently moved from a contract to coop- Center Research Triangle
erative agreement funding mechanism, with investigative Institute)
teams competing for histology-centric research testing. The Sarcoma and Greehey Children’s Cancer Peter Houghton,
resulting PPTC was established in 2015 and is organized as Renal Cancer Research Institute PhD
shown in Table 2. Raushan
Kurmasheva,
The PPTC has tested 43 agents to date (several are still being PhD
tested). Unlike the PPTP, in which the majority of agents
Neuroblastoma Children’s Hospital of John Maris, MD
were screened for antitumor activity across all models in the Philadelphia
program, the PPTC performs hypothesis-driven testing of Yaël Mossé, MD
agents, which means that many of the agents are tested in
Osteosarcoma MD Anderson Cancer Richard Gorlick,
subsets of cases based on known molecular aberrations. To Center MD
realize this goal, the PPTC sought to dramatically expand the
Andy Kolb, MD
number of PDX models available to the consortium, and
investigators performed whole-exome sequencing, RNA se- Leukemia Children’s Cancer Institute Richard Lock,
Australia PhD
quencing, single nucleotide polymorphism arrays, and sim-
ple tandem repeat genotyping on 261 PDXs27 available in Brain Tumors Northwestern University Xiao-Nan Li, MD,
PhD
the pediatric component of the cBIO Portal, which is housed
alongside genomic data generated on 151 overlapping Sponsorship and National Cancer Institute Malcolm Smith,
Funding MD, PhD
models from the PPTP (https://pedcbioportal.kidsfirstdrc.
org/study/summary?id=pptc%2Cmixed_target_pptp). Al- Beverly Teicher,
PhD
though many of the agents tested in the PPTP/PPTC have

• Multiple agents have proceeded to pediatric clinical of molecular subsets that might be enriched in biomarker-
testing based in part on PPTP/PPTC results, including directed clinical trials.
selumetinib for low-grade gliomas; temsirolimus in com- The pediatric cancer community has a paucity of creden-
bination with chemotherapy for rhabdomyosarcoma; OBI- tialed models to properly study immuno-oncology agents.
3424 for T-cell ALL; lorvotuzumab mertansine (IMGN901); To address the anticipated increased bandwidth for pre-
SNDX-5613 in mixed-lineage leukemia, rearranged ALL, clinical testing of anticancer agents as industry plans for
and acute myelogenous leukemia; alisertib; and eribulin. the Research to Accelerate Cures and Equity for Children
The comparison of the level of preclinical activity to the Act, a broader PPP to further modernize and expand the
activity of the agents observed among patients allows PPTC model is under discussion through the Foundation
ongoing refinement of the use of preclinical data to pri- for the National Institutes of Health and may be launched
oritize agents for clinical testing. in 2020.
• Comparative pharmacokinetics and pharmacodynamics be- The challenges and opportunities provided by changes in
tween adult early-phase trials and pediatric cancer PDX the regulatory environment in both the United States and
mouse models can be critical to the success or failure of European Union and the reality of the global context of
early-phase childhood cancer trials. A noted example was cancer drug development require international coordina-
robust activity of the Aurora kinase A inhibitor alisertib in tion and collaboration at multiple levels, including both
neuroblastoma and ALL PDX models, followed by limited clinical trial design and conduct and pediatric preclinical
single-agent activity in the clinic when the same dose and investigations. Meaningful transformation of regulatory change
schedule modeled in the mice could not be achieved with to improved therapeutics for childhood cancer mandates data
patients.28-30 sharing, rational priority setting, maximum efficiency, and
• Robust proteogenomic characterization of models allows for avoidance of needless duplication. Collaboration across the
careful preclinical testing design, allowing for identification Atlantic is mandatory.

1
U.S. Food and Drug Administration, Silver Spring, MD Gregory Reaman, MD, U.S. Food and Drug Administration, 10903 New
2
Eli Lilly and Company, Indianapolis, IN Hampshire Ave., WO 22 Room 2202, Silver Spring, MD 20993; email:
3
Institute Gustave Roussy, Villejuif, France [email protected].
4
The Children’s Hospital of Philadelphia, Philadelphia, PA
EDBK_278893.
REFERENCES
2. Jemal A, Ward EM, Johnson CJ, et al. Annual report to the nation on the status of cancer, 1975-2014, featuring survival. J Natl Cancer Inst. 2017;109:djx030.
3. Landier W, Armenian S, Bhatia S. Late effects of childhood cancer and its treatment. Pediatr Clin North Am. 2015;62:275-300.
4. Bhakta N, Liu Q, Ness KK, et al. The cumulative burden of surviving childhood cancer: an initial report from the St Jude Lifetime Cohort Study (SJLIFE). Lancet.
2017;390:2569-2582.
5. Neel DV, Shulman DS, DuBois SG. Timing of first-in-child trials of FDA-approved oncology drugs. Eur J Cancer. 2019;112:49-56.
6. U.S. Food and Drug Administration. Pediatric Research Equity Act, Pub. L. No. 108–155, 117 Stat. 1936 (2003). www.fda.gov/downloads/Drugs/
DevelopmentApprovalProcess/DevelopmentResources/UCM077853.pdf. Accessed January 6, 2020.
7. U.S. Congress. Best Pharmaceuticals for Children Act (2002). www.congress.gov/107/plaws/publ109/PLAW-107publ109.pdf. Accessed January 6, 2020.
8. European Union. EU Regulation on Medicinal Products for Paediatric Use [Regulation (EC) No 1901/2006 of the European Parliament and of the Council of
December 12, 2006 on Medicinal Products for Paediatric Use and Amending Regulation (EEC) No 1768/92, Directive 2001/20/EC, Directive 2001/83/EC and
Regulation (EC) No 726/2004e]. https://op.europa.eu/en/publication-detail/-/publication/2b688364-769b-4529-ab41-ba530e266466/language-en. Accessed
January 14, 2020.
9. U.S. Food and Drug Administration. Food and Drug Administration Reauthorization Act (FDARA) of 2017. www.fda.gov/regulatory-information/selected-
amendments-fdc-act/fda-reauthorization-act-2017-fdara. Accessed January 14, 2020.
10. Vogelstein B, Papadopoulos N, Velculescu VE, et al. Cancer genome landscapes. Science. 2013;339:1546-1558.

Reaman et al
11. Gröbner SN, Worst BC, Weischenfeldt J, et al; ICGC MMML-Seq Project. The landscape of genomic alterations across childhood cancers [published correction
appears in Nature. 2018;559:E10]. Nature. 2018;555:321-327.
12. Ma X, Liu Y, Liu Y, et al. Pan-cancer genome and transcriptome analyses of 1,699 paediatric leukaemias and solid tumours. Nature. 2018;555:371-376.
13. U.S. Food and Drug Administration. OCE Pediatric Oncology Program. Relevant Pediatric Molecular Target List. www.fda.gov/about-fda/oncology-center-
excellence/pediatric-oncology. Accessed January 14, 2020.
14. Adamson PC, Houghton PJ, Perilongo G, et al. Drug discovery in paediatric oncology: roadblocks to progress. Nat Rev Clin Oncol. 2014;11:732-739.
15. Dorsey ER, de Roulet J, Thompson JP, et al. Funding of US biomedical research, 2003-2008. JAMA. 2010;303:137-143.
16. Houghton PJ, Morton CL, Tucker C, et al. The pediatric preclinical testing program: description of models and early testing results. Pediatr Blood Cancer. 2007;
49:928-940.
17. ITCC-P4. ITCC-P4 International Multistakeholder Workshop: Improving Pediatric Oncology Drug Development Through Preclinical Research. www.itccp4.eu/
files/itcc-p4_international-workshop-2018-in-amsterdam.pdf. Accessed January 15, 2020.
18. Adamson PC. Improving the outcome for children with cancer: development of targeted new agents. CA Cancer J Clin. 2015;65:212-220.
19. Jones DTW, Banito A, Grünewald TGP, et al. Molecular characteristics and therapeutic vulnerabilities across paediatric solid tumours. Nat Rev Cancer. 2019;
19:420-438.
20. Biondi A, Schrappe M, De Lorenzo P, et al. Imatinib after induction for treatment of children and adolescents with Philadelphia-chromosome-positive acute
lymphoblastic leukaemia (EsPhALL): a randomised, open-label, intergroup study. Lancet Oncol. 2012;13:936-945.
21. Bresler SC, Weiser DA, Huwe PJ, et al. ALK mutations confer differential oncogenic activation and sensitivity to ALK inhibition therapy in neuroblastoma. Cancer
Cell. 2014;26:682-694.
22. Laetsch TW, DuBois SG, Mascarenhas L, et al. Larotrectinib for paediatric solid tumours harbouring NTRK gene fusions: phase 1 results from a multicentre, open-
label, phase 1/2 study. Lancet Oncol. 2018;19:705-714.
23. Schubert NA, Lowery CD, Bergthold G, et al. Systematic target actionability reviews of preclinical proof-of-concept papers to match targeted drugs to paediatric
cancers. Eur J Cancer. , In press.
24. Vassal G, Houghton PJ, Pfister SM, et al. International consensus on minimum preclinical testing requirements for the development of innovative therapies for
children and adolescents with cancer. Mol Cancer Ther. In press.
25. Lowery CD, Dowless M, Renschler M, et al. Broad spectrum activity of the checkpoint kinase 1 inhibitor prexasertib as a single agent or chemopotentiator across
a range of preclinical pediatric tumor models. Clin Cancer Res. 2019;25:2278-2289.
26. Houghton PJ, Adamson PC, Blaney S, et al. Testing of new agents in childhood cancer preclinical models: meeting summary. Clin Cancer Res. 2002;
8:3646-3657.
27. Rokita JL, Rathi KS, Cardenas MF, et al. Genomic profiling of childhood tumor patient-derived xenograft models to enable rational clinical trial design. Cell Rep.
2019;29:1675-1689.
28. Carol H, Boehm I, Reynolds CP, et al. Efficacy and pharmacokinetic/pharmacodynamic evaluation of the Aurora kinase A inhibitor MLN8237 against preclinical
models of pediatric cancer. Cancer Chemother Pharmacol. 2011;68:1291-1304.
29. Maris JM, Morton CL, Gorlick R, et al. Initial testing of the Aurora kinase A inhibitor MLN8237 by the Pediatric Preclinical Testing Program (PPTP). Pediatr Blood
Cancer. 2010;55:26-34.
30. Mossé YP, Fox E, Teachey DT, et al. A phase II study of alisertib in children with recurrent/refractory solid tumors or leukemia: Children’s Oncology Group Phase I
and Pilot Consortium (ADVL0921). Clin Cancer Res. 2019;25:3229-3238.

PEDIATRIC ONCOLOGY
Cutting to the Front of the Line: Immunotherapy for

Childhood Acute Lymphoblastic Leukemia
Jennifer L. McNeer, MD, MS1; Rachel E. Rau, MD2; Sumit Gupta, MD, PhD3; Shannon L. Maude, MD, PhD4; and
Maureen M. O’Brien, MD, MS5
overview
Although many children and young adults with B-cell acute lymphoblastic leukemia (B-ALL) are cured
with modern, risk-adapted chemotherapy regimens, 10% to 15% of patients will experience relapse or
have refractory disease. Recent efforts to further intensify cytotoxic chemotherapy regimens in the
frontline setting have failed as a result of excessive toxicity or lack of improvement in efficacy. As
a result, novel approaches will be required to achieve cures in more newly diagnosed patients. Mul-
tiple immune-based therapies have demonstrated considerable efficacy in the setting of relapsed or
refractory (R/R) disease, including CD19 targeting with blinatumomab and tisagenlecleucel and
CD22 targeting with inotuzumab ozogamicin. These agents are now under investigation by the
Children’s Oncology Group (COG) in clinical trials for newly diagnosed B-ALL, with integration into
standard chemotherapy regimens based on clinically and biology-based risk stratification as well as
disease response.
INTRODUCTION INTENSIFICATION OF CHEMOTHERAPY IN

Acute lymphoblastic leukemia (ALL) is the most PEDIATRIC ALL
commonly diagnosed cancer in children. B-ALL ac- COG has explored various approaches to intensify
counts for 85% of cases in children younger than age therapy within these risk groups to improve outcomes.
20.1 With improvements in our understanding of The CCG-1991 trial demonstrated improved outcomes
disease biology and the development of increasingly for patients with SR disease using escalating doses of
sensitive methods to measure disease response, intravenous methotrexate (MTX) along with vincristine
more specific risk stratification and treatment allo- during IM rather than the prior IM regimen of oral
cation have led to improvements in survival from 10% mercaptopurine and MTX. 9 However, subsequent
in the 1960s to current survival rates of more than studies have not demonstrated benefit with further in-
90%.2 Despite this improvement, treatment of pa- tensification of chemotherapy. The follow-up AALL0331
tients who relapse remains challenging. Efforts are trial found no benefit to augmented consolidation in this
ongoing to improve upfront therapy to decrease rates patient population.10 Most recently, the AALL0932 trial
of R/R disease. found no improvement with intensification of oral MTX
dosing during maintenance or with vincristine/steroid
COG uses a treatment approach for B-ALL pioneered
pulses given every 4 weeks versus every 12 weeks.11,12
by the Berlin-Frankfurt-Münster (BFM) study group.3
With current event-free survival (EFS) rates of nearly
This approach consists of initial therapy (induction/
90%,10 intensification of conventional chemotherapy
protocol IA) followed by a focus on central nervous
agents seems unlikely to improve outcomes for children
Author affiliations system treatment (consolidation/protocol IB). Sub-
with SR B-ALL.
and support sequent treatment blocks include interim maintenance
information (if (IM) and then a re-induction/reconsolidation known as Similar attempts have been made to intensify treatment
applicable) appear delayed intensification. Finally, patients receive a proof patients with HR B-ALL, whose consolidation and IM
at the end of this
longed maintenance phase consisting largely of anti- phases of therapy are already more intense than those
article.
metabolite therapy. Over time, COG has modified this of SR B-ALL. The CCG-1961 study found that in-
Accepted on March
18, 2020 and backbone based on evolving variables contributing to creasing intensity of postinduction therapy, largely with
published at risk stratification. Currently, patients are determined to additional vincristine and PEG-asparaginase during
ascopubs.org on have standard-risk (SR), high-risk (HR), or very-HR times of expected myelosuppression, was beneficial,
April 21, 2020: (VHR) disease based on a combination of National whereas longer duration was not.13 The follow-up trial
Cancer Institute (NCI) criteria, leukemia biology, and (AALL0232) also successfully intensified treatment with
10.1200/EDBK_
278171 early disease response.4-8 the use of dexamethasone rather than prednisone as

Immunotherapy for Childhood Acute Lymphoblastic Leukemia
regimens. In B-ALL, these therapies have generally targeted

the nearly ubiquitously expressed B-cell antigens CD19 and
CD22. For patients with R/R B-ALL, promising response
• Immune-based therapies are effective in rates have been observed with the bispecific T-cell engager
treating patients with R/R B-ALL.
blinatumomab targeting CD19, the antibody-drug conjugate
• Immune-based therapies have unique mech- inotuzumab ozogamicin targeting CD22, and CAR T cells
anisms of action and toxicity profiles, which targeting CD19 such as tisagenlecleucel. These agents are
have implications for timing of use and com- now under evaluation as components of treatment of pa-
bination with other therapies.
tients with newly diagnosed B-ALL.
• Intensification of frontline chemotherapy regi-
mens for B-ALL with the addition of immune- Incorporation of Blinatumomab for SR B-ALL
based therapies is under investigation in Previous COG trials for the treatment of patients with NCI SR
prospective clinical trials. B-ALL (age 1–9.99 and white blood cell count , 50,000/mL
• COG trials incorporate blinatumomab for at diagnosis) have successfully identified a chemotherapy
selected standard-risk patients, inotuzumab regimen that yields excellent outcomes and acceptable
ozogamicin for selected high-risk patients, toxicity in the context of a risk classification schema in-
and tisagenlecleucel for selected very-high- cluding clinical variables, sentinel genetic lesions, and early
risk patients as defined by clinical features, treatment response.2,7,10,12,29 Nonetheless, this population
leukemia biology, and early treatment still accounts for approximately one-half of the overall burden
response.
of relapse among children with ALL. As noted previously,
recent COG studies have shown the limits of chemotherapy
the induction steroid for certain HR patients, as well as the intensification in this group of patients.10,11,29 Further im-
incorporation of high-dose MTX into the first IM phase.14 provements are thus likely to require the addition of novel
However, further attempts to intensify therapy on the recent agents. The addition of any new agent in an SR B-ALL
AALL1131 trial with the use of triple intrathecal therapy and population must balance the need to improve cure rates with
the incorporation of clofarabine, cyclophosphamide, and the recognition that baseline outcomes, although un-
etoposide into postinduction therapy did not improve out- satisfactory, are nonetheless favorable. This section will
comes. There was no survival benefit with the addition of discuss the rationale for studying blinatumomab in NCI SR
intrathecal therapy for HR patients.15 In the VHR subgroup, B-ALL given its high antileukemia activity, efficacy in a mini-
clofarabine was ultimately found to be too toxic in upfront mal residual disease (MRD) setting, and favorable toxicity
treatment, and the addition of cyclophosphamide/etoposide profile. The current COG NCI SR B-ALL trial, AALL1731
did not improve survival outcomes.16,17 It seems, therefore, (NCT03914625), will also be described.
that intensification of conventional chemotherapy has been
Blinatumomab antileukemia activity Blinatumomab is a bi-
maximized in the treatment of HR B-ALL.
specific single-chain antibody that targets the CD19 antigen
Finally, there are subgroups of patients with adverse sen- and redirects CD3+ T cells for the selective lysis of tumor cells
tinel genetic lesions or very poor disease response for whom (Fig. 1).30,31 As a result of its short in vivo half-life, blinatu-
outcomes using conventional chemotherapy are quite poor. momab is administered as a 4-week continuous infusion. Two
For patients with Philadelphia chromosome–positive (Ph+) initial adult phase II studies in R/R B-ALL demonstrated
B-ALL, targeted therapy with tyrosine kinase inhibitors has noteworthy activity with rates of complete remission with or
demonstrated improved outcomes.18,19 Based on this suc- without complete hematologic recovery (CR/CRh) of 69% (25
cess, trials are underway for subgroups of patients with out of 36) and 43% (81 out of 189), respectively.31,32 In each
Ph-like B-ALL harboring kinase-activating lesions.20-23 In of these cohorts, more than 80% of patients achieving CR/
contrast, for patients with hypodiploid B-ALL with modal CRh also achieved MRD negativity. In the subsequent phase
chromosome numbers of less than 44 and those with in- III TOWER study, adults with R/R Ph-negative (Ph–) B-ALL
duction failure (residual marrow blast percentage . 25%), were randomly assigned 2:1 to receive blinatumomab or
therapy intensification has failed.24-26 Efforts to improve standard of care chemotherapy regimens. Patients receiving
outcomes for these patients through use of allogeneic he- blinatumomab experienced CR rates double those receiving
matopoietic stem cell transplantation (HSCT) in first re- chemotherapy (34% vs. 16%; p , .001) and had a significant
mission (complete remission [CR] 1) have not demonstrated twofold improvement in median overall survival (OS; 7.7 vs.
benefit.26-28 4 months; p = .01).33
Given that further intensification of treatment has proven Blinatumomab is similarly promising in younger populations.
either too toxic or without benefit, focus has turned to in- AALL1121, a phase I/II study conducted by COG and the
corporating novel immunotherapies into frontline treatment I BFM Study Group, enrolled 70 children, adolescents,

McNeer et al
FIGURE 1. Mechanisms of Action of Blinatumomab, Inotuzumab Ozogamicin, and Tisagenlecleucel

Abbreviations: B-ALL, B-cell acute lymphoblastic leukemia; scFV, single-chain fragment variable; VH, heavy-chain variable domain; VL, light-chain variable
domain.

and young adults with R/R B-ALL.34 Among patients re- Interestingly, though MRD response rates were high
ceiving the recommended dose of blinatumomab, 39% across subgroups, the duration of response remained
achieved CR within two cycles, with 52% of these inferior among those in second or later remission, again
achieving an MRD– CR. When restricted to patients with suggesting that treatment during first remission may be
relapsed disease, the CR rate was 48%. Given the un- ideal. To date, no noteworthy comparable data exist in
favorable characteristics of the cohort (. 70% of patients pediatric cohorts in first remission. A final observation
had relapsed within 6 months of the previous treatment from the BLAST trial is that among 36 patients who did not
attempt), these data were highly encouraging. Based on receive further HSCT or chemotherapy after blinatumo-
these results, the U.S. Food and Drug Administration mab, nine (25%) remained in continuous CR at a median
(FDA) granted accelerated approval of blinatumomab for follow-up of 24 months, suggesting that blinatumomab
the treatment of Ph– R/R B-ALL in 2014, which was ex- may be capable of eradicating leukemic populations in
tended to full approval for patients with Ph+ and Ph– some patients. Based on these data, blinatumomab be-
disease in 2016. The 2016 decision also extended approval came the first FDA-approved treatment of patients with
to children with Ph– R/R disease, making blinatumomab MRD+ ALL in 2018.
the first agent to receive such approval since 2004.
Blinatumomab toxicity profile Cytokine release syndrome
Most recently, data from the COG trial for first relapse of (CRS) is a known complication of several immunotherapies,
B-ALL, AALL1331 (NCT02101853), were presented in 2019. including blinatumomab.40 In AALL1121, eight patients
Following re-induction chemotherapy, patients with HR (11%) experienced CRS of any grade, with four (6%) ex-
relapse were randomly assigned to receive either two in- periencing grade 3 or higher CRS.34 The risk of CRS depends
tensive chemotherapy blocks, based on the UKALLR3 heavily on disease burden at time of administration, as
backbone,35 or two 4-week blocks of blinatumomab.36 Both evidenced by low rates when administered to patients
arms subsequently proceeded to HSCT. Among patients with only MRD present, particularly with dexamethasone
with post–block 1 detectable MRD, rates of MRD negativity administration prior to the start of the blinatumomab
after block 2 were 21% in the chemotherapy arm versus infusion.38,39,41 Neurotoxicity is another side effect that does
79% in the blinatumomab arm (p , .0001). Two-year not seem to be associated with leukemia burden, but is
disease-free survival (DFS) was 41% versus 59% (p = .05), substantially less common among children compared with
and 2-year OS was 59% versus 79% (p = .005), both fa- adults. In AALL1121, nine patients (13%) experienced
voring blinatumomab. These results led to early randomi- treatment-related neurologic events (mainly tremor and
zation closure and suggest that blinatumomab may be the dizziness). All were grade 2 and resolved with no permanent
new standard treatment of this population. Likewise, the discontinuations because of neurologic events.34 Two pa-
international 20120215 trial (NCT02393859), a phase III tients interrupted treatment because of grade 2 seizures.
study of blinatumomab for first relapse of pediatric B-ALL,
stopped after the primary endpoint of EFS was met at There were similar neurotoxicity rates on AALL1331, with
a planned interim analysis.37 a 4% seizure rate and a 14% rate of other neurotoxicities
(e.g., tremor, ataxia, and dysarthria), with all adverse events
Blinatumomab efficacy for MRD clearance Accumulating (AEs) fully resolved.36 Of note, rates of febrile neutropenia,
data suggest that blinatumomab may be most effective at infection, sepsis, and mucositis were all strikingly lower
earlier disease stages and at lower levels of disease. In trials among patients randomly selected to receive blinatumomab
in adults with multiply relapsed disease, despite achieving as compared with standard intensive chemotherapy. The
notable rates of MRD negativity with blinatumomab, the favorable toxicity profile has led some clinicians to use
cumulative incidence of subsequent relapse remained high. blinatumomab as a successful bridge to further therapy in
Both adult and pediatric data suggest an association be- children who experience overwhelming chemotherapy-
tween lower leukemia burden at the time of blinatumomab associated toxicity. 42
administration and improved response rates. For example,
Blinatumomab trials in newly diagnosed pediatric B-ALL
adult CR/CRh rates were 73% among those with less than
With the aforementioned data in mind, the COG AALL1731
50% marrow blasts at time of blinatumomab administration
trial will identify subsets of children with NCI SR B-ALL who
compared with 29% for those with 50% or more blasts.31
have higher risk features, including unfavorable genetics or
The single-arm BLAST trial evaluated blinatumomab in adults slow MRD clearance, either measured by traditional flow
in first or later remission with persistent MRD higher than cytometry–based methods or novel high-throughput se-
0.1%.38,39 Of 113 evaluable patients, 88 (78%) achieved quencing techniques.7,43 These patients will be randomly
MRD negativity after one cycle of blinatumomab, with similar selected to receive either standard treatment backbones (of
response rates regardless of MRD level. These response rates regular or intensified therapy, based on postinduction risk
are substantially higher than those seen in R/R populations. stratification) or standard treatment plus two courses of

McNeer et al
FIGURE 2. Schema of COG AALL1731 (SR), AALL1732 (HR), and AALL1721 (VHR) B-ALL Trials
Abbreviations: B-ALL, B-cell acute lymphoblastic leukemia; COG, Children’s Oncology Group; C-MTX, Capizzi MTX; EscMTX, escalating MTX; HD-MTX,
high-dose MT; HR, high risk; HTS, high-throughput sequencing; MRD, minial residual disease; NCI, National Cancer Institute; SR, standard risk; SR-Avg,
SR-average; VHR, very high risk.
blinatumomab intercalated between chemotherapy blocks blinatumomab to the chemotherapy backbone. In the St.
(Fig. 2). Jude Children’s Research Hospital Consortium Total
Therapy Study 17 trial (NCT031177510), blinatumomab is
Of note, children with Down syndrome (DS) and B-ALL incorporated into frontline therapy for a subset of patients
continue to experience inferior outcomes compared with with residual MRD at the end of induction.
those without DS. 44-46 Both excessive treatment-related
mortality and higher rates of relapse contribute to this Other considerations Given generally favorable outcomes of
disparity. To reduce treatment-related mortality in a single- children with NCI SR B-ALL, other considerations merit
arm cohort of patients with DS and HR features, AALL1731 mention. Though blinatumomab is associated with superior
will replace intensive immunosuppressive and myelosup- health-related quality of life as compared with chemother-
pressive elements of standard chemotherapy with three apy among adults,48 continuous intravenous administration
courses of blinatumomab, which may preserve or even may present unique challenges for children and their
enhance antileukemic efficacy while reducing toxicity. Al- caregivers. AALL1731 thus includes health-related quality
though this will be the largest cohort of patients with DS of life measures. In an era of rising cancer therapy–associated
B-ALL treated with blinatumomab to date, reports of use in costs, future studies of cost effectiveness will also be war-
children with DS suggest similar safety and efficacy in ranted. Finally, CD19– relapse is a known cause of treatment
patients without DS.34,47 failure among patients receiving blinatumomab.30,49 Given
CD19-dependent options in relapse such as CAR T-cell
In addition to the active COG trial, blinatumomab is under therapy, patterns of CD19 expression in relapse will be
investigation by other pediatric cooperative groups for pa- closely monitored.
tients with newly diagnosed B-ALL. The AIEOP-BFM ALL
2017 phase III randomized trial (NCT03643276) is evalu- Incorporation of Inotuzumab Ozogamicin for
ating two cycles of postconsolidation blinatumomab com- High-Risk B-ALL
pared with two conventional, highly intensive chemotherapy NCI HR B-ALL is defined as occuring in patients age 10 or
courses in HR patients. For intermediate-risk patients, the older and/or in patients with a white blood cell count 50,000/
same trial is evaluating the addition of a single cycle of mL or higher at the time of diagnosis. For the almost 3,000

patients with NCI HR disease enrolled in COG AALL0232 remissions at dose levels ranging from 1.2 to 1.8 mg/m2/
from 2004 to 2011, the 5-year EFS and OS were 75.2% 6 cycle.54
1.1% and 85.0% 6 0.9%, respectively.14 There is con-
Subsequently, the pivotal INO-VATE phase III trial for adults
siderable room for improvement in outcomes for this pop-
with R/R B-ALL as first or second salvage therapy randomly
ulation, particularly those patients with unfavorable genetic
assigned patients to fractionated inotuzumab ozogamicin
lesions or slow clearance of MRD. Inotuzumab ozogamicin
versus best-available intensive chemotherapy (high-dose
has demonstrated impressive activity in the R/R setting and
cytarabine-based regimen).55 The CR/CRi rate was signifi-
was approved in 2017 by the FDA for adults who have
cantly higher in the inotuzumab ozogamicin arm than in the
experienced relapsed B-ALL. Compared with blinatumo-
standard therapy arm (73.8% vs. 30.9%; p , .0001).56
mab, the experience with inotuzumab ozogamicin in pe-
Among patients with CR/CRi, a higher percentage treated
diatric populations is limited, making inotuzumab ozogamicin
with inotuzumab ozogamicin had MRD less than 0.01%
more appropriate for study of a higher-risk population. The
(78.4% vs. 28.1%; p , .001) compared with standard
toxicity profile of inotuzumab ozogamicin is most notable for
chemotherapy. Most inotuzumab ozogamicin responders
risk of hepatic sinusoidal obstruction syndrome (SOS), ob-
(73%) achieved CR/CRi following one cycle of therapy.
served mainly when patients proceed to HSCT following
Patients treated with inotuzumab ozogamicin were more
inotuzumab ozogamicin treatment. To date, SOS risk in the
likely to proceed directly to HSCT after achieving CR/CRi
absence of HSCT appears low. Beyond SOS, the toxicity
(39.6% vs. 10.5%; one-sided p , .0001). Response rates
profile of inotuzumab ozogamicin has been favorable. To mit-
with inotuzumab ozogamicin were similar regardless of
igate SOS risk, it would be optimal to incorporate inotuzumab
duration of prior remission, prior HSCT, salvage attempt,
ozogamicin into treatment of patients who are not intended to
and CD22 expression (, 90% or 90% or more blasts).
receive consolidation with HSCT. This section will highlight
CR/CRi rate, MRD– rate, and OS were not different among
the efficacy and toxicity profile of inotuzumab ozogamicin
cytogenetic subgroups, though patients with KMT2A rear-
and describe AALL1732 (NCT03959085), the current COG
rangement had the lowest MRD– rate and shortest median
trial for patients with newly diagnosed NCI HR B-ALL.
progression-free survival. Notably, these patients were more
Inotuzumab ozogamicin mechanism of action Inotuzumab likely to have only partial CD22 expression.57
ozogamicin, a CD22-targeted humanized immunoglobulin
The most common toxicities related to inotuzumab ozo-
type G, subtype 4 (IgG4) antibody covalently linked to N-Ac-
gamicin include myelosuppression, infection, and hepatic
γ-calicheamicin dimethylhydrazide, is a potent antibody-
toxicity. In the INO-VATE trial, grade 3 or higher thrombo-
drug conjugate (Fig. 1).50,51 Upon binding to surface CD22
cytopenia, neutropenia, and anemia occurred in 37%,
on target cells, inotuzumab ozogamicin is rapidly internalized
46%, and 19% of patients, respectively. A total of 24% of
and traffics to the acidic lysosomal vesicular compartment
patients had grade 3 febrile neutropenia. Hepatic AEs were
of the cell, where N-Ac-γ-calicheamicin dimethylhydrazide
more frequent in the inotuzumab ozogamicin arm (50.6%
is hydrolyzed from the monoclonal antibody. N-Ac-γ-
vs. 36.4%).58 Alanine aminotransferase elevation of any
calicheamicin dimethylhydrazide undergoes reduction by
grade occurred in 14% of patients (3% grade 3), and
intracellular thiols, leading to the formation of a reactive
hyperbilirubinemia of any grade occurred in 15% (4% grade
intermediate, and binds in the minor groove of DNA. The
3). SOS was reported during or after treatment (including
resulting DNA damage results in apoptosis. Elimination is
post-HSCT follow-up) in 23 out of 164 patients (14.0%) in
primarily through the biliary system. Inotuzumab ozoga-
the inotuzumab ozogamicin arm and in three out of 143
micin does not cross the blood-brain barrier.
patients (2.1%) in the chemotherapy arm. Most patients
Efficacy and toxicity of inotuzumab ozogamicin in R/R B-ALL who developed SOS underwent HSCT either before or after
in adults In adults with R/R B-ALL, the initial early-phase treatment with inotuzumab ozogamicin; importantly, the rate
trials evaluated 1.8 mg/m2 inotuzumab ozogamicin every of SOS was low (3%) in patients who never underwent HSCT.
21 days but then adopted fractionated dosing of 0.8 mg/m2
on day 1 and 0.5 mg/m2 on days 8 and 15 based on Efficacy and toxicity of inotuzumab ozogamicin in R/R B-ALL
preclinical studies suggesting similar efficacy with dein children A retrospective multisite study reported toxicity
creased toxicity.52,53 Overall response rates (CR and CR with and response data for 51 pediatric patients treated with
complete remission with incomplete hematologic recovery inotuzumab ozogamicin through a compassionate access
[CRi]) in the phase I/II setting were similar for single-dose program from 2013 to 2016.59 In this heavily pretreated,
and fractionated-dose inotuzumab ozogamicin schedules multiply relapsed cohort, 28 out of 42 patients with overt
(57% vs. 59%, respectively), but SOS was less frequent marrow disease (67%) achieved CR/CRi, of whom 71% had
with the fractionated schedule. Additional early-phase trials MRD less than 0.01%. As in adult trials, the most common
of fractionated inotuzumab ozogamicin dosing in adults reported toxicities were febrile neutropenia and infection.
with R/R B-ALL demonstrated notable activity with MRD– Grade 3 or higher hepatotoxicity was rare, and no patient

McNeer et al
developed SOS while receiving inotuzumab ozogamicin. (younger than age 10, favorable genetics, and rapid re-
Following inotuzumab ozogamicin therapy, 21 patients sponse to induction therapy) with very favorable outcomes
underwent allogeneic HSCT, of whom 11 (51%) developed (EFS 94%) will be treated with chemotherapy only (HR-
SOS (4 mild, 2 moderate, and 5 severe, of which 2 cases Favorable, not included in Fig. 2). All patients will receive
were fatal). induction and consolidation therapy; those with CD22 ex-
pression on leukemic blasts at diagnosis and with end-of-
The first prospective early-phase studies of inotuzumab
consolidation (EOC) MRD less than 0.01% but who do not
ozogamicin in children were reported in 2019. The Euro-
meet criteria for HR-Favorable will be eligible for random-
pean Innovative Therapies for Children with Cancer Con-
ization to receive chemotherapy with or without the addition
sortium conducted a phase I trial in children with R/R B-ALL
of two cycles of inotuzumab ozogamicin (0.5 mg/m2 on days
(ITCC-059).60 This trial evaluated both 1.4 and 1.8 mg/m2/
1, 8, and 15 of each cycle). In addition to routine safety
cycle fractionated dosing, and the recommended phase II
monitoring, an initial safety phase will closely evaluate the
dose for pediatric patients was found to be the same as
first 50 randomly selected patients given the risk of hepatic
adults (1.8 mg/m2/course). The overall response rate was
toxicity and myelosuppression with intercalation of inotu-
80%, and 79% of responders had MRD less than 0.01%.
zumab ozogamicin therapy blocks between chemotherapy
Grade 3 to 4 transaminase elevations occurred in three
blocks. inotuzumab ozogamicin, like other B-cell–directed
patients (12%) and grade 3 bilirubin elevation in two pa-
therapies, results in profound transient B-cell aplasia; thus,
tients (8%) in course 1. Two cases of SOS were recorded
immune function and infectious toxicities will be moni-
after follow-up chemotherapy without HSCT.
tored.54 Inotuzumab ozogamicin is also being evaluated in the
Concurrently, COG conducted AALL1621 (NCT02981628), frontline setting in combination with chemotherapy for young
a phase II trial of single-agent inotuzumab ozogamicin in adults with newly diagnosed B-ALL (Alliance A041501,
pediatric patients with multiply R/R B-ALL using the same NCT03150693).
fractionated dosing regimen as the INO-VATE trial.61
Twenty-eight out of 48 patients achieved CR/CRi after cy- Incorporation of Tisagenlecleucel for VHR B-ALL
cle 1 (58.3%; 95% CI, 43.2–72.4%). Two patients with
Subgroups of patients with newly diagnosed B-ALL remain
progressive disease in the central nervous system had
at VHR of relapse despite current intensive chemotherapy
marrow CR (MRD 0.02% and , 0.01% in one case each),
regimens. Many of these patients with VHR disease can be
highlighting the lack of central nervous system penetration
identified early based on a poor response to initial therapy.
of inotuzumab ozogamicin and need for concurrent central
Data from the AALL0232 trial showed extremely poor 5-year
nervous system–directed therapy. MRD was reported for 26
DFS in patients with NCI HR B-ALL who had persistent MRD
out of 28 patients with CR/CRi; of these, 17 (65.4%) had
0.01% or higher after induction and consolidation. Patients
MRD less than 0.01%, and four (15.4%) had MRD between
with EOC MRD 0.01% or higher had a 5-year DFS of 39%
0.01% and 0.099%. The most common AEs in cycle 1 were
compared with 79% for those with end-of-induction MRD
febrile neutropenia (27%) and infection (18.8%), and the
0.1% or higher but EOC MRD less than 0.01% (p ,
most common dose-limiting toxicities were hematologic
.0001).8 These patients are in need of novel therapeutic
(absolute neutrophils , 500/mL or platelets , 20,000/mL
approaches.
beyond 42 days from the start of cycle 1 [7 cases]). Four
(8.3%) patients had grade 3 transaminase elevation, and HSCT is one approach that has been used for patients with
one had grade 3 bilirubin in cycle 1. No dose modifications anticipated poor survival with chemotherapy alone. How-
were required for hepatic toxicity. Overall, six out of 48 ever, relapse risk is higher for those with detectable MRD at
patients (12.5%) developed SOS, and all cases occurred in the time of HSCT.62-64 Further intensification of chemo-
the setting of HSCT after inotuzumab ozogamicin therapy. therapy to decrease or eliminate MRD as well as HSCT itself
Of the 20 patients who received subsequent HCST after carry substantial risk of morbidity and mortality.65-68 Per-
inotuzumab ozogamicin, six (30%) developed SOS. All haps most importantly, persistent disease after two cycles of
cases were grade 3 and treated with defibrotide; five re- multiagent chemotherapy and the HR of relapse with ad-
solved quickly, whereas one was more severe but resolved ditional intensive multiagent chemotherapy suggest that
at the time of death from other HSCT complications. these leukemias are chemorefractory. To address these
considerations, the COG is exploring cellular immuno-
Inotuzumab ozogamicin trials in newly diagnosed B-ALL therapy using CAR-modified T cells, a novel therapy with
The primary aim of the COG AALL1732 trial is to determine demonstrated efficacy and a mechanism of action distinct
in a randomized manner if the addition of two blocks of from cytotoxic chemotherapy. This section will discuss the
inotuzumab ozogamicin to COG-modified BFM chemo- efficacy data for the CD19-directed CAR T-cell product
therapy will improve outcomes in children and young adults tisagenlecleucel, the toxicity profile, and the design of
with NCI HR B-ALL (Fig. 2). A small subgroup of patients AALL1721/Cassiopeia (NCT03876769), a COG and Novartis

collaborative phase II trial of tisagenlecleucel for NCI HR patients with ALL.78-80 In its mildest form, CRS causes flulike
B-ALL with persistent EOC MRD. symptoms, including high persistent fevers, myalgias, head-
ache, fatigue, nausea/vomiting, and anorexia. In its most
Efficacy of CAR T-cell therapy in R/R B-ALL Adoptive transfer
severe form, CRS can lead to life-threatening multiorgan
of T cells engineered to express a CAR targeting CD19 has
system failure. CRS was observed in close to 90% of patients
demonstrated great promise in multiply relapsed and highly
treated with CTL019, with grade 4 CRS reported in 25% of
refractory B-ALL with CR rates as high as 93%.69-75 A CAR
patients in ELIANA.71,77 Across all studies and CD19 CAR
links an antigen recognition domain with T-cell signaling
T-cell products, high bone marrow disease burden is asso-
domains to combine the specificity of a monoclonal anti-
ciated with an increased risk of severe CRS.69,73-75,80 Con-
body with activation domains that trigger T-cell cytotoxic
versely, CRS is mild in the absence of morphologic disease.
machinery. Once engaged by their target, CARs activate
a cytotoxic T-cell response that kills the bound antigen- Neurotoxicity, a second distinct toxicity associated with
expressing cell (Fig. 1). CTL019 (now known as tisa- T-cell–engaging immunotherapies, can occur in the pres-
genlecleucel) became the first CAR T-cell therapy approved ence or absence of CRS symptoms.81,82 Although often
by the FDA in August 2017. This CAR T-cell product, de- termed encephalopathy, this class of AEs can also include
veloped by the University of Pennsylvania, Children’s Hospital confusion, delirium, hallucinations, aphasia, focal neuro-
of Philadelphia, and Novartis, contains an anti-CD19 single- logic deficits, tremor, somnolence, and, less commonly,
chain fragment variable domain linked to the CD3ζ cyto- seizure.71,75,81-83 In clinical trials of CTL019, neurotoxicity
plasmic domain and the 4-1BB costimulatory domain.76 has been reported in 40% to 45% of patients.71,83 Although
neurotoxicity has been observed at low disease burden,
In a phase I/IIa single-institution trial of CTL019 conducted
increased incidence and severity of neurotoxicity has been
at Children’s Hospital of Philadelphia, a CR rate of 93% was
associated with higher grade CRS.71,82,83
observed in 60 patients with multiply R/R B-ALL.75,77 Re-
missions were achieved across a wide range of disease AALL1721/Cassiopeia Study Design In designing the next
burden. At the time of infusion, 73% of patients had de- phase of frontline trials for B-ALL, COG sought a novel
tectable disease, 53% had more than 5% leukemic blasts, mechanistic approach to improve outcomes for patients
and 38% had more than 50% leukemic blasts in the bone with poor early response to therapy. Based on the excellent
marrow. Relapse-free survival was 60% (95% CI, 48–75%) remission rates reported with CTL019 in populations of
and OS was 79% (95% CI, 67–88%) at 12 months with patients who were refractory to standard therapy, combined
a median follow-up of 15 months. CTL019 received Break- with the possibility of durable remission without con-
through Therapy designation by the FDA in 2014. A phase II solidative HSCT, COG chose to study CAR T-cell therapy
single-arm, multicenter, global registration trial (ELIANA) was using the FDA-approved tisagenlecleucel in this VHR pop-
conducted by Novartis across 25 international centers with ulation. As disease burden is associated with CRS severity
results reported in 2018. MRD– remission was achieved and with incidence and severity of neurotoxicity, it is an-
by 3 months in 81% of 75 patients treated with CTL019. ticipated that the incidence of these toxicities will be low in
Relapse-free survival and OS at 12 months were 59% (95% this patient population with persistent MRD during frontline
CI, 41–73%) and 76% (95% CI, 63–86%), respectively, therapy.
with a median follow-up of 13 months.71
Through collaboration between COG and Novartis, AALL1721/
Both trials demonstrated durable remissions without further Cassiopeia was designed as a phase II single-arm trial of
therapy; only 11% proceeded to HSCT in remission after tisagenlecleucel in children and young adults with NCI HR
CTL019. CAR-modified T cells persisted in the circulation B-ALL with persistent MRD at EOC. Participating sites
as long as 39 months at data cutoff.71,77 Based on ELIANA include COG centers in the United States and Canada as
data, with supporting data from the University of Pennsyl- well as pediatric centers in the United Kingdom and
vania/Children’s Hospital of Philadelphia phase I/IIa trial and Europe. Patients age 1 to 25 diagnosed with CD19-
a Novartis U.S. multicenter phase II trial (NCT02228096), expressing NCI HR B-ALL are eligible in first CR after
CTL019 was granted FDA approval as tisagenlecleucel for induction/protocol IA and consolidation/protocol IB che-
children and young adults up to age 25 with B-ALL that is motherapy if MRD is detected by central multiparameter
refractory or in second or greater relapse. flow cytometry at a level of 0.01% or greater. Patients with
Toxicity of CAR T-cell therapy The principle toxicity of CAR induction failure (. 25% marrow blasts), Ph+ ALL, and
T-cell therapy is CRS resulting from elevations in in- hypodiploid ALL are excluded, as are those who have
flammatory cytokines associated with exponential pro- received tyrosine kinase inhibitor therapy.
liferation of CAR T cells. Clinical manifestations are related The trial schema is outlined in Fig. 2. Patients receive in-
to immune activation and inflammation and typically occur duction/protocol IA and consolidation/protocol IB chemo-
within the first 7 to 14 days after CAR T-cell infusion in therapy according to a COG or similar HR B-ALL protocol.

McNeer et al
Once determined to be eligible by positive EOC MRD, pa- following 15-year long-term follow-up trial. Patient-reported
tients are enrolled on AALL1721/Cassiopeia. Leukapheresis outcome measures and cognitive testing will be used to
can occur after induction in patients who have high-end assess the impact of this approach on patients’ quality of life.
induction MRD or after consolidation, once a patient has
a qualifying EOC MRD result. Enrolled patients proceed to CONCLUSIONS
IM, the next phase of standard of care therapy, during the The current COG portfolio of frontline trials for pediatric
period of tisagenlecleucel manufacture. After stopping IM B-ALL incorporates promising immunotherapies with novel
chemotherapy, patients receive lymphodepleting chemo- mechanisms into current regimens with the goal of treat-
therapy followed by a single infusion of tisagenlecleucel. ment intensification without the excess toxicity that has
After infusion, no further cancer-directed chemotherapy hindered recent attempts at chemotherapy intensification.
(including intrathecal chemotherapy) is administered. These targeted immunotherapies have been selected for
Response is assessed at day 28, and patients are followed specific populations based on patient risk stratification and
every 3 months until relapse or 5 years after infusion. The outcomes with chemotherapy alone, efficacy in the R/R set-
primary endpoint is 5-year DFS; secondary endpoints in- ting, and toxicity profiles. Ultimately, if these agents improve
clude OS, safety, cellular kinetics, the percentage of patients outcomes with acceptable toxicity, this portfolio of trials could
in remission without HSCT at 1 year, and time to B-cell profoundly change the landscape of therapy for children and
recovery, a surrogate marker of CD19 CAR T-cell functional young adults with newly diagnosed B-ALL.
persistence. In addition, as the long-term effects of CAR
T-cell therapy and chronic B-cell aplasia have not been well ACKNOWLEDGMENT
studied, treatment-emergent AEs, including infections and J.L. McNeer and R. Rau are co-first authors. The authors
secondary malignancies, will be reported on this trial or the thank Abigail Wung for the design of Fig. 1.

1
University of Chicago Comer Children’s Hospital, Chicago, IL Maureen M. O’Brien, MD, MS, Cancer and Blood Diseases Institute,
2
Baylor College of Medicine, Texas Children’s Hospital, Houston, TX Cincinnati Children’s Hospital Medical Center, 3333 Burnet Ave., MLC
3
The Hospital for Sick Children and University of Toronto, Toronto, 7018, Cincinnati, OH 45229; email: [email protected].
Ontario, Canada
4
Children’s Hospital of Philadelphia, University of Pennsylvania Perelman
School of Medicine, Philadelphia, PA
5
Cincinnati Children’s Hospital Medical Center, University of Cincinnati
College of Medicine, Cincinnati, OH
EDBK_278171.
REFERENCES
1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013. CA Cancer J Clin. 2013;63:11-30.
2. Hunger SP, Mullighan CG. Acute lymphoblastic leukemia in children. N Engl J Med. 2015;373:1541-1552.
3. Schrappe M, Reiter A, Zimmermann M, et al. Long-term results of four consecutive trials in childhood ALL performed by the ALL-BFM study group from 1981 to
1995. Berlin-Frankfurt-Münster. Leukemia. 2000;14:2205-2222.
4. Smith M, Arthur D, Camitta B, et al. Uniform approach to risk classification and treatment assignment for children with acute lymphoblastic leukemia. J Clin
Oncol. 1996;14:18-24.
5. Mrózek K, Heerema NA, Bloomfield CD. Cytogenetics in acute leukemia. Blood Rev. 2004;18:115-136.
6. Harrison CJ. Cytogenetics of paediatric and adolescent acute lymphoblastic leukaemia. Br J Haematol. 2009;144:147-156.
7. Borowitz MJ, Devidas M, Hunger SP, et al; Children’s Oncology Group. Clinical significance of minimal residual disease in childhood acute lymphoblastic
leukemia and its relationship to other prognostic factors: a Children’s Oncology Group study. Blood. 2008;111:5477-5485.
8. Borowitz MJ, Wood BL, Devidas M, et al. Prognostic significance of minimal residual disease in high risk B-ALL: a report from Children’s Oncology Group study
AALL0232. Blood. 2015;126:964-971.
9. Matloub Y, Bostrom BC, Hunger SP, et al. Escalating intravenous methotrexate improves event-free survival in children with standard-risk acute lymphoblastic
leukemia: a report from the Children’s Oncology Group. Blood. 2011;118:243-251.
10. Maloney KW, Devidas M, Wang C, et al. Outcome in children with standard-risk B-cell acute lymphoblastic leukemia: results of Children’s Oncology Group trial
AALL0331. J Clin Oncol. 2020;38:602-612.

11. Angiolillo A, Schore R, Devidas M, et al. Intensification of oral methotrexate is not superior to standard methotrexate dosing during maintenance in children with
National Cancer Institute (NCI) standard-risk B acute lymphoblastic leukemia (SR B-ALL): a report from Children’s Oncology Group (COG) study AALL0932.
Blood. 2017;130(suppl 1):140.
12. Angiolillo A, Schore RJ, Kairalla J, et al. Excellent outcomes with reduced frequency of vincristine and dexamethasone pulses in children with National Cancer
Institute (NCI) standard-risk B acute lymphoblastic leukemia (SR B-ALL): a report from Children’s Oncology Group (COG) study AALL0932. Blood. 2019;
134(suppl 1):824.
13. Seibel NL, Steinherz PG, Sather HN, et al. Early postinduction intensification therapy improves survival for children and adolescents with high-risk acute
lymphoblastic leukemia: a report from the Children’s Oncology Group. Blood. 2008;111:2548-2555.
14. Larsen EC, Devidas M, Chen S, et al. Dexamethasone and high-dose methotrexate improve outcome for children and young adults with high-risk B-acute
lymphoblastic leukemia: a report from Children’s Oncology Group study AALL0232. J Clin Oncol. 2016;34:2380-2388.
15. Salzer WL, Burke MJ, Devidas M, et al. Triple intrathecal therapy (methotrexate/hydrocortisone/cytarabine) does not improve disease-free survival versus
intrathecal methotrexate alone in children with high risk B-lymphoblastic leukemia: results of Children’s Oncology Group Study AALL1131. Blood. 2018;
132(suppl 1):35.
16. Salzer WL, Burke MJ, Devidas M, et al. Toxicity associated with intensive postinduction therapy incorporating clofarabine in the very high-risk stratum of patients
with newly diagnosed high-risk B-lymphoblastic leukemia: a report from the Children’s Oncology Group study AALL1131. Cancer. 2018;124:1150-1159.
17. Burke M, Salzer W, Chen S, et al. Substitution with cyclophosphamide and etoposide does not improve outcome for children and young adults with very high risk
B-lymphoblastic leukemia: Children’s Oncology Group study AALL1131. Pediatr Blood Cancer. 2017;64:S16 (suppl; abstr O-019).
18. Schultz KR, Carroll A, Heerema NA, et al; Children’s Oncology Group. Long-term follow-up of imatinib in pediatric Philadelphia chromosome-positive acute
lymphoblastic leukemia: Children’s Oncology Group study AALL0031. Leukemia. 2014;28:1467-1471.
19. Slayton WB, Kairalla JA, Schultz KR, et al. Outcomes of dasatinib plus intensive chemotherapy or stem cell transplant (SCT) for Philadelphia chromosome-positive
acute lymphoblastic leukemia (Ph+ ALL) on Children’s Oncology Group AALL0622. J Clin Oncol. 2015;33:15s (suppl; abstr 10006).
20. Reshmi SC, Harvey RC, Roberts KG, et al. Targetable kinase gene fusions in high-risk B-ALL: a study from the Children’s Oncology Group. Blood. 2017;
129:3352-3361.
21. Roberts KG, Li Y, Payne-Turner D, et al. Targetable kinase-activating lesions in Ph-like acute lymphoblastic leukemia. N Engl J Med. 2014;371:1005-1015.
22. Roberts KG, Pei D, Campana D, et al. Outcomes of children with BCR-ABL1–like acute lymphoblastic leukemia treated with risk-directed therapy based on the
levels of minimal residual disease. J Clin Oncol. 2014;32:3012-3020.
23. Den Boer ML, van Slegtenhorst M, De Menezes RX, et al. A subtype of childhood acute lymphoblastic leukaemia with poor treatment outcome: a genome-wide
classification study. Lancet Oncol. 2009;10:125-134.
24. Heerema NA, Nachman JB, Sather HN, et al. Hypodiploidy with less than 45 chromosomes confers adverse risk in childhood acute lymphoblastic leukemia:
a report from the children’s cancer group. Blood. 1999;94:4036-4045.
25. Nachman JB, Heerema NA, Sather H, et al. Outcome of treatment in children with hypodiploid acute lymphoblastic leukemia. Blood. 2007;110:1112-1115.
26. Schrappe M, Hunger SP, Pui CH, et al. Outcomes after induction failure in childhood acute lymphoblastic leukemia. N Engl J Med. 2012;366:1371-1381.
27. McNeer JL, Devidas M, Dai Y, et al. Hematopoietic stem-cell transplantation does not improve the poor outcome of children with hypodiploid acute lymphoblastic
leukemia: a report from Children’s Oncology Group. J Clin Oncol. 2019;37:780-789.
28. Pui CH, Rebora P, Schrappe M, et al; Ponte di Legno Childhood ALL Working Group. Outcome of children with hypodiploid acute lymphoblastic leukemia:
a retrospective multinational study. J Clin Oncol. 2019;37:770-779.
29. Mattano LA, Devidas M, Friedmann AM, et al. Outstanding outcome for children with standard risk-low (SR-low) acute lymphoblastic leukemia (ALL) and no
benefit to intensified Peg-asparaginase (PEG-ASNase) therapy: results of Children’s Oncology Group (COG) study AALL0331. Blood. 2014;124:793.
30. Curran E, Stock W. Taking a “BiTE out of ALL”: blinatumomab approval for MRD-positive ALL. Blood. 2019;133:1715-1719.
31. Topp MS, Gökbuget N, Stein AS, et al. Safety and activity of blinatumomab for adult patients with relapsed or refractory B-precursor acute lymphoblastic
leukaemia: a multicentre, single-arm, phase 2 study. Lancet Oncol. 2015;16:57-66.
32. Topp MS, Gökbuget N, Zugmaier G, et al. Phase II trial of the anti-CD19 bispecific T cell-engager blinatumomab shows hematologic and molecular remissions in
patients with relapsed or refractory B-precursor acute lymphoblastic leukemia. J Clin Oncol. 2014;32:4134-4140.
33. Kantarjian H, Stein A, Gökbuget N, et al. Blinatumomab versus chemotherapy for advanced acute lymphoblastic leukemia. N Engl J Med. 2017;376:836-847.
34. von Stackelberg A, Locatelli F, Zugmaier G, et al. Phase I/phase II study of blinatumomab in pediatric patients with relapsed/refractory acute lymphoblastic
leukemia. J Clin Oncol. 2016;34:4381-4389.
35. Parker C, Waters R, Leighton C, et al. Effect of mitoxantrone on outcome of children with first relapse of acute lymphoblastic leukaemia (ALL R3): an open-label
randomised trial. Lancet. 2010;376:2009-2017.
36. Brown PA, Ji L, Xu X, et al. A randomized phase 3 trial of blinatumomab vs. chemotherapy as post-reinduction therapy in high and intermediate risk (HR/IR) first
relapse of B-acute lymphoblastic leukemia (B-ALL) in children and adolescents/young adults (AYAs) demonstrates superior efficacy and tolerability of bli-
natumomab: a report from Children’s Oncology Group Study AALL1331. Blood. 2019;134(suppl 2):LBA-1.
37. Locatelli F, Zugmaier G, Rizzari C, et al. Superior event-free survival with blinatumomab versus chemotherapy in children with high-risk first relapse of B-cell
precursor acute lymphoblastic leukemia: a randomized, controlled phase 3 trial. Presented at the 46th Annual Meeting of the European Society for Blood and
Marrow Transplantation. Madrid, Spain; 2020.

McNeer et al
38. Gökbuget N, Zugmaier G, Klinger M, et al. Long-term relapse-free survival in a phase 2 study of blinatumomab for the treatment of patients with minimal residual
disease in B-lineage acute lymphoblastic leukemia. Haematologica. 2017;102:e132-e135.
39. Goekbuget N, Dombret H, Bonifacio M, et al. BLAST: a confirmatory, single-arm, phase 2 study of blinatumomab, a bispecific T-cell engager (BiTE®) antibody
construct, in patients with minimal residual disease B-precursor acute lymphoblastic leukemia (ALL). Blood. 2014;124:379.
40. Lee DW, Gardner R, Porter DL, et al. Current concepts in the diagnosis and management of cytokine release syndrome [published correction appears in Blood.
2015;126:1048]. Blood. 2014;124:188-195.
41. Maude SL, Teachey DT, Porter DL, et al. CD19-targeted chimeric antigen receptor T-cell therapy for acute lymphoblastic leukemia. Blood. 2015;125:4017-4023.
42. Elitzur S, Arad-Cohen N, Barzilai-Birenboim S, et al. Blinatumomab as a bridge to further therapy in cases of overwhelming toxicity in pediatric B-cell precursor
acute lymphoblastic leukemia: report from the Israeli Study Group of Childhood Leukemia. Pediatr Blood Cancer. 2019;66:e27898.
43. Wood B, Wu D, Crossley B, et al. Measurable residual disease detection by high-throughput sequencing improves risk stratification for pediatric B-ALL. Blood.
2018;131:1350-1359.
44. Buitenkamp TD, Izraeli S, Zimmermann M, et al. Acute lymphoblastic leukemia in children with Down syndrome: a retrospective analysis from the Ponte di Legno
study group. Blood. 2014;123:70-77.
45. Meyr F, Escherich G, Mann G, et al. Outcomes of treatment for relapsed acute lymphoblastic leukaemia in children with Down syndrome. Br J Haematol. 2013;
162:98-106.
46. Hitzler JK, He W, Doyle J, et al; CIBMTR Pediatric Cancer Working Committee. Outcome of transplantation for acute lymphoblastic leukemia in children with
Down syndrome. Pediatr Blood Cancer. 2014;61:1126-1128.
47. Wadhwa A, Kutny MA, Xavier AC. Blinatumomab activity in a patient with Down syndrome B-precursor acute lymphoblastic leukemia. Pediatr Blood Cancer.
2018;65:e26824.
48. Topp MS, Zimmerman Z, Cannell P, et al. Health-related quality of life in adults with relapsed/refractory acute lymphoblastic leukemia treated with blinatumomab.
Blood. 2018;131:2906-2914.
49. Gore L, Locatelli F, Zugmaier G, et al. Survival after blinatumomab treatment in pediatric patients with relapsed/refractory B-cell precursor acute lymphoblastic
leukemia. Blood Cancer J. 2018;8:80.
50. DiJoseph JF, Armellino DC, Boghaert ER, et al. Antibody-targeted chemotherapy with CMC-544: a CD22-targeted immunoconjugate of calicheamicin for the
treatment of B-lymphoid malignancies. Blood. 2004;103:1807-1814.
51. Shor B, Gerber HP, Sapra P. Preclinical and clinical development of inotuzumab-ozogamicin in hematological malignancies. Mol Immunol. 2015;67(2 Pt A):
107-116.
52. Kantarjian H, Thomas D, Jorgensen J, et al. Inotuzumab ozogamicin, an anti-CD22-calecheamicin conjugate, for refractory and relapsed acute lymphocytic
leukaemia: a phase 2 study. Lancet Oncol. 2012;13:403-411.
53. Kantarjian H, Thomas D, Jorgensen J, et al. Results of inotuzumab ozogamicin, a CD22 monoclonal antibody, in refractory and relapsed acute lymphocytic
leukemia. Cancer. 2013;119:2728-2736.
54. DeAngelo DJ, Stock W, Stein AS, et al. Inotuzumab ozogamicin in adults with relapsed or refractory CD22-positive acute lymphoblastic leukemia: a phase 1/2
study. Blood Adv. 2017;1:1167-1180.
375:740-753.
56. Jabbour E, O’Brien S, Huang X, et al. Prognostic factors for outcome in patients with refractory and relapsed acute lymphocytic leukemia treated with inotuzumab
ozogamicin, a CD22 monoclonal antibody. Am J Hematol. 2015;90:193-196.
57. Jabbour E, Advani AS, Stelljes M, et al. Prognostic implications of cytogenetics in adults with acute lymphoblastic leukemia treated with inotuzumab ozogamicin.
Am J Hematol. 2019;94:408-416.
58. Kantarjian HM, DeAngelo DJ, Advani AS, et al. Hepatic adverse event profile of inotuzumab ozogamicin in adult patients with relapsed or refractory acute
lymphoblastic leukaemia: results from the open-label, randomised, phase 3 INO-VATE study. Lancet Haematol. 2017;4:e387-e398.
59. Bhojwani D, Sposto R, Shah NN, et al. Inotuzumab ozogamicin in pediatric patients with relapsed/refractory acute lymphoblastic leukemia [published correction
appears in Leukemia. 2019;33:1061-1062]. Leukemia. 2019;33:884-892.
60. Brivio E, Lopez-Yurda M, Ownes C, et al. A phase I study of single-agent inotuzumab ozogamicin in pediatric CD22-positive relapsed/refractory acute lym-
phoblastic leukemia: preliminary results of the ITCC-059 Study. Blood. 2019;134(suppl 1):2629.
61. O’Brien MM, Ji L, Shah NN, et al. A phase 2 trial of inotuzumab ozogamicin (InO) in children and young adults with relapsed or refractory (R/R) CD22+ B-acute
lymphoblastic leukemia (B-ALL): results from Children’s Oncology Group protocol AALL1621. Blood. 2019;134(suppl 1):741.
62. Pulsipher MA, Carlson C, Langholz B, et al. IgH-V(D)J NGS-MRD measurement pre- and early post-allotransplant defines very low- and very high-risk ALL
patients. Blood. 2015;125:3501-3508.
63. Pulsipher MA, Langholz B, Wall DA, et al. Risk factors and timing of relapse after allogeneic transplantation in pediatric ALL: for whom and when should
interventions be tested? Bone Marrow Transplant. 2015;50:1173-1179.
64. Pulsipher MA, Langholz B, Wall DA, et al. The addition of sirolimus to tacrolimus/methotrexate GVHD prophylaxis in children with ALL: a phase 3 Children’s
Oncology Group/Pediatric Blood and Marrow Transplant Consortium trial. Blood. 2014;123:2017-2025.

65. Bader P, Kreyenberg H, von Stackelberg A, et al. Monitoring of minimal residual disease after allogeneic stem-cell transplantation in relapsed childhood acute
lymphoblastic leukemia allows for the identification of impending relapse: results of the ALL-BFM-SCT 2003 trial. J Clin Oncol. 2015;33:1275-1284.
66. Balduzzi A, Dalle JH, Wachowiak J, et al. Transplantation in children and adolescents with acute lymphoblastic leukemia from a matched donor versus an HLA-
identical sibling: is the outcome comparable? Results from the International BFM ALL SCT 2007 Study. Biol Blood Marrow Transplant. 2019;25:2197-2210.
67. Conter V, Bartram CR, Valsecchi MG, et al. Molecular response to treatment redefines all prognostic factors in children and adolescents with B-cell precursor
acute lymphoblastic leukemia: results in 3184 patients of the AIEOP-BFM ALL 2000 study. Blood. 2010;115:3206-3214.
68. Pieters R, de Groot-Kruseman H, Van der Velden V, et al. Successful therapy reduction and intensification for childhood acute lymphoblastic leukemia based on
minimal residual disease monitoring: Study ALL10 from the Dutch Childhood Oncology Group. J Clin Oncol. 2016;34:2591-2601.
69. Turtle CJ, Hanafi LA, Berger C, et al. CD19 CAR-T cells of defined CD4+:CD8+ composition in adult B cell ALL patients. J Clin Invest. 2016;126:2123-2138.
70. Gardner RA, Finney O, Annesley C, et al. Intent-to-treat leukemia remission by CD19 CAR T cells of defined formulation and dose in children and young adults.
Blood. 2017;129:3322-3331.
73. Davila ML, Riviere I, Wang X, et al. Efficacy and toxicity management of 19-28z CAR T cell therapy in B cell acute lymphoblastic leukemia. Sci Transl Med. 2014;
6:224ra25.
74. Lee DW, Kochenderfer JN, Stetler-Stevenson M, et al. T cells expressing CD19 chimeric antigen receptors for acute lymphoblastic leukaemia in children and
young adults: a phase 1 dose-escalation trial. Lancet. 2015;385:517-528.
75. Maude SL, Frey N, Shaw PA, et al. Chimeric antigen receptor T cells for sustained remissions in leukemia. N Engl J Med. 2014;371:1507-1517.
76. Kalos M, Levine BL, Porter DL, et al. T cells with chimeric antigen receptors have potent antitumor effects and can establish memory in patients with advanced
leukemia. Sci Transl Med. 2011;3:95ra73.
77. Maude SL, Teachey DT, Rheingold SR, et al. Sustained remissions with CD19-specific chimeric antigen receptor (CAR)-modified T cells in children with relapsed/
refractory ALL. J Clin Oncol. 2016;34:15s(suppl; abstr 3011).
78. Maude SL, Barrett D, Teachey DT, et al. Managing cytokine release syndrome associated with novel T cell-engaging therapies. Cancer J. 2014;20:119-122.
79. Teachey DT, Lacey SF, Shaw PA, et al. Identification of predictive biomarkers for cytokine release syndrome after chimeric antigen receptor T-cell therapy for
acute lymphoblastic leukemia. Cancer Discov. 2016;6:664-679.
80. Hay KA, Hanafi LA, Li D, et al. Kinetics and biomarkers of severe cytokine release syndrome after CD19 chimeric antigen receptor-modified T-cell therapy. Blood.
2017;130:2295-2306.
81. Gust J, Hay KA, Hanafi LA, et al. Endothelial activation and blood-brain barrier disruption in neurotoxicity after adoptive immunotherapy with CD19 CAR-T cells.
Cancer Discov. 2017;7:1404-1419.
82. Santomasso BD, Park JH, Salloum D, et al. Clinical and biological correlates of neurotoxicity associated with CAR T-cell therapy in patients with B-cell acute
lymphoblastic leukemia. Cancer Discov. 2018;8:958-971.
83. Gofshteyn JS, Shaw PA, Teachey DT, et al. Neurotoxicity after CTL019 in a pediatric and young adult cohort. Ann Neurol. 2018;84:537-546.

PEDIATRIC ONCOLOGY
Before It’s Too Late: Multistakeholder

Perspectives on Compassionate Access to
Investigational Drugs for Pediatric Patients
With Cancer
Elena Gerasimov, MA, MPH1; Martha Donoghue, MD2; Josh Bilenker, MD3; Tanya Watt, MD4,5; Nancy Goodman, JD1; and
Theodore W. Laetsch, MD4,5,6
Patients and their families, physicians, drug companies, and regulatory agencies have common goals: to find
overview
effective therapies for life-threatening conditions. In oncology, the lines between clinical research and
treatment are often blurred; parents and physicians of patients who have exhausted standard-of-care
treatments and cannot participate in a clinical trial are likely to consider seeking compassionate use access to
investigational drugs; however, knowledge and perspectives about compassionate use may differ among these
groups. There are unique considerations associated with providing compassionate use to children diagnosed
with cancer, including evaluation for potential developmental toxicities, the need for pediatric-specific dosing
and formulations, informed consent, and, when appropriate, patient assent. Positive impacts of providing
access to investigational therapies to children include potential treatment benefits to patients who obtain
investigational agents as well as benefits to future patients if data from expanded access support drug
development for childhood cancer. Challenges for physicians seeking compassionate use access to in-
vestigational drugs for their patients include obtaining the drug sponsor’s agreement to provide the in-
vestigational drug as well as lack of knowledge about the process and regulatory requirements. Clinical trials in
oncology provide the possibility of therapeutic benefit for pediatric patients; when feasible and warranted,
these benefits should also be available to patients on a compassionate use basis outside of trials.
INTRODUCTION trial or who cannot travel to or identify a clinical trial that

Access to investigational therapies outside of a clinical is accepting patients.
trial is known as compassionate use. Some other terms A key difference between compassionate use and
for compassionate use include expanded access, clinical trials is therapeutic intent. Although pediatric
preapproval access, early access, single patient in- clinical trials must confer at least a prospect of
vestigational new drug (IND), named patient program, clinical benefit to participants,3 the main purpose of
temporary authorization for use, and treatment use. clinical trials is to conduct research, whereas the
The U.S. Food and Drug Administration (FDA) uses the primary goal of compassionate use is treatment.
term “expanded access,” which it defines as “a po- However, clinical research4 and compassionate use
tential pathway for a patient with an immediately life- access have the potential to provide clinical benefit to
threatening or serious disease to gain access to an patients. In addition, although the quality of data
Author affiliations
investigational medical product for treatment outside of derived from compassionate use is generally not
and support clinical trials when no comparable or satisfactory al- equivalent to clinical trial data, information gained
information (if ternative therapy options are available.”1 from expanded-access programs can influence drug-
applicable) appear development decisions for sponsors and support
at the end of this Conversations between patients or caregivers and
oncologists about compassionate use typically occur in clinical trial data in the regulatory review of new drug
article.
the setting of a life-threatening cancer that has not applications.
Accepted on March
31, 2020 and responded to standard treatment. Patients with cancer Access to investigational drugs through compassionate
published at are often willing to accept significant risks associated use is especially important for pediatric patients, be-
ascopubs.org on May
12, 2020: DOI https://
with treatments, especially in life-threatening situa- cause studies of new drugs are typically conducted first
doi.org/10.1200/ tions.2 Compassionate use may be considered for in adults, and relatively few cancer drugs are FDA
EDBK_278995 patients who do not satisfy eligibility criteria for a clinical approved specifically for pediatric patients. Additionally,

Multistakeholder Perspectives on Compassionate Access for Pediatric Patients With Cancer
3. Widespread treatment use (or treatment IND or protocol).

This category provides access for large numbers of
patients, often for late-stage products during the pe-
• Compassionate use can be the best option for riod between the completion of pivotal trials and drug
patients who seek treatment and cannot par-
approval.10,11 Treatment protocols and INDs are spon-
ticipate in a clinical trial and who have no
sored by the commercial developer of the investigational
satisfactory therapeutic alternatives.
drug. Widespread treatment use INDs are relatively
• Pediatric oncologists should be knowledgeable uncommon, and single patient INDs are the most
about the compassionate use process and
common method for obtaining compassionate use
prepared to advocate for access on behalf of
access for investigational drugs that are in early-stage
patients.
development.6
• The FDA has streamlined the process of ap-
plying for compassionate use and has created EXPLAINING THE PROCESS: HOW TO APPLY
a call center, Project Facilitate, to help guide
Obtaining a single patient IND is a three-step process that
oncologists through the process.
is different for emergency and nonemergency situations
• The FDA encourages drug sponsors to provide (Fig. 1):
expanded access to investigational drugs to
patients who are unable to enroll in a clinical 1. Contact the drug sponsor. A licensed physician submits
trial, when feasible and warranted. an application to the drug sponsor (often a pharma-
ceutical company) via email or web portal. If it agrees to
• Compassionate use can produce data that
enhance drug development for children. provide the drug, the company issues a Letter of Au-
thorization for the FDA to access its file containing
proprietary manufacturing, nonclinical, clinical phar-
eligibility criteria exclude children and adolescents from macology, and clinical information supporting the safety
many “adult” clinical trials. of the drug. Companies are under no obligation to
provide drugs for compassionate use, and the FDA
CATEGORIES OF EXPANDED ACCESS cannot compel a company to provide an investiga-
The FDA divides expanded use programs into three cate- tional drug.
gories based on the number of patients (Table 1): 2. Obtain FDA authorization. A physician contacts the FDA
and usually fills out Form 392612 with information about
1. Individual patient (single patient IND). Requests for an
the drug and the patient’s condition. The FDA reviews it
individual patient, including emergency use, are the
for an authorization to proceed. The FDA authorizes 99.
most commonly submitted requests. In 2018, the FDA
5% of the expanded-access requests it receives, roughly
received 1,576 new individual patient IND requests,
1,400 annually.5
including 528 for emergency use, 5 and 102 IND
3. Obtain institutional review board (IRB) approval and
protocols.6 Data from 2014 show that 37% of requests
informed consent. After a drug sponsor has agreed to
received were for oncology (20%) and hematology (17%)
provide access, a physician must obtain a sign-off from
products.5,7 Most commonly, the treating physician
the chair or other member of an IRB and informed
agrees to serve as a sponsor/investigator for purposes of
consent from the patient.
administering a compassionate use drug and, thus,
accepts responsibility for the new IND.8 However, it is When a patient requires urgent treatment before a written
also possible for pharmaceutical companies to submit submission can be made, emergency use requests for an
a single patient protocol as an amendment to their individual patient with cancer are usually granted by the
existing commercial IND. FDA verbally the same day, and nonemergency requests are
2. Intermediate-size population. This category of expanded generally reviewed within a few days.13 Certain paperwork,
access is used when a sponsor has or anticipates that such as Form 3926 and the Letter of Authorization, is not
it will receive multiple expanded access requests, but required for the FDA to grant authorization for emergency
a treatment protocol or IND is not warranted. The use; this paperwork can be submitted later. With emergency
FDA recommends that sponsors consider opening an use, it is not necessary to obtain IRB approval prior to patient
intermediate-sized expanded-access protocol or IND9 treatment, but the IRB must be notified of emergency use
when it is expected that more than 10 patients will re- within 5 working days.
ceive compassionate use access to an investigational The investigational drug is shipped directly to the physician
drug. Intermediate expanded access is not used fre- who is responsible for overseeing the investigational treat-
quently; in 2018, the FDA received 20 intermediate-size ment and reporting outcomes to the sponsor, the IRB, and
IND requests and 19 IND protocols.6 the FDA.

Gerasimov et al
TABLE 1. Key Requirements for Expanded Access and Right to Try Requests
Intermediate-Size Widespread
Requirement Single Patient IND Population (Treatment) Use Federal Right to Try*
Required Approvals
Mandates drug sponsor to provide the drug No No No No
FDA authorization required Yes Yes Yes No
Informed consent required Yes Yes Yes Yes
IRB approval required Yes** Yes Yes No
Eligible Investigational Products
The investigational product must be under active No No Yes Yes
commercial development
Providing the investigational product must not Yes Yes Yes No
interfere with clinical trials that could support the
product’s development or marketing approval
Sufficient data on safety and effectiveness exist to Yes† Yes† Yes Completed phase I
support use for the indication
Eligible Patients
Disease is serious or life threatening Yes Yes Yes Yes
Patient is unable to participate in a clinical trial Yes Yes Yes Yes
No comparable or satisfactory alternative therapy is Yes Yes Yes Yes
available
The potential patient benefit justifies the potential Yes Yes Yes No
risks of the treatment use
Reporting Requirements
Adverse event(s) reporting Yes, by IND holder; Yes, by IND holder; Yes, by IND holder; Yes, by drug sponsor;
time requirements time requirements time requirements annual summary
apply apply apply only
Abbreviations: IND, investigational new drug; IRB, institutional review board.

*Individual state Right to Try laws may have different requirements.
**Approval by a chair or a single member of an IRB is sufficient. Approval can be deferred for emergency situations.
†FDA exerts flexibility in this assessment, taking into account the clinical context, scientific rationale, and existing nonclinical and clinical data.
PERSPECTIVES ON COMPASSIONATE USE FOR they believe might save or prolong their child’s life. This
INDIVIDUAL PATIENTS includes pleas to companies, energetic social media cam-
Although patients, physicians, regulators, advocates, and paigns, seeking help from politicians and celebrities, offering
drug manufacturers share common goals of finding effective to pay for an investigational drug, or buying a drug from an
treatments, differences in perspectives can contribute to con- overseas pharmacy.
troversy and misunderstandings about compassionate use. Patients and family members may encounter barriers to
Patients compassionate use, such as a drug company’s decision not
to provide a drug or the inability of their treating physician to
Patients often view compassionate use as their last and best
efficiently navigate the process or undertake the adminis-
hope. Many families of children with refractory cancers
understand that investigational drugs may not provide trative burden of applying for compassionate use.
a benefit and can cause toxicities, but they are willing to Some families have overly optimistic expectations about an
accept the risk in exchange for the possibility of prolonging unapproved drug’s effectiveness and may be unaware of
the life of their child. data held by the sponsor that are critical to the evaluation of
Many patients and their family members or caregivers are the benefit-risk analysis for use of the drug for their child.
aware of the possibility of accessing unapproved drugs and Many may not realize that the FDA cannot require a drug
expect their oncologist to pursue a compassionate use sponsor to provide access to an investigational drug and
request. Some parents of a child with cancer are willing to do that there may be valid reasons why a drug company may
everything in their power to obtain access to something that decline to provide a drug for compassionate use.

Physicians Drug Sponsors and Companies

Physicians are central to the process of obtaining com- Every company developing an oncology drug knows that it
passionate use, serving as a liaison between patients and may receive requests for compassionate use, and the
drug sponsors—and often, patients and the FDA. In ad- likelihood of receiving these requests increases if positive
dition to the manufacturer of the investigational drug, only results are reported publicly during clinical development.
licensed physicians can submit an expanded access re- Although drug sponsors generally do not disclose the
quest and obtain an IND. They must identify potential number of requests that they receive and the corresponding
treatments and balance uncertainty about the potential decisions, one company indicated that it had denied 98 of
benefits and toxicities of the drug against the risks of the 160 applications for a single drug in a 6-month period.7
disease itself. Another company disclosed that it receives an average of
65 requests per month and approves 95%.15
Because pediatric oncologists spend so much time with
families of children with cancer, the bonds of trust between Requests are not limited to drugs in late-stage trials. A study
physicians and families are tight. Pediatric oncologists are of expanded-access requests at Memorial Sloan Kettering
intimately involved in the families of their patients. Families Cancer Center showed that, over a 6-year period, almost
look to their pediatric oncologists for direction and emo- 40% of requested drugs were in phase III, 36% were in
tional and medical support, especially when their chil- phase II, and 18.8% were still in phase I.16 Increasingly,
dren’s cancers do not respond to treatment. There are oncology drug developers are recognizing the importance of
rarely conflicts, even when compassionate use is preparing for these requests early in the drug-development
considered. process, including establishment of a clear transparent
policy that communicates whether the drug will be made
Some physicians may be reluctant to engage in discussions
available for compassionate use, the criteria that must be
about compassionate use for several reasons, including the
fulfilled for the drug to be made available, and the process
inability to identify a promising investigational drug, con-
for requesting drug access.
cerns about unproven effectiveness of the drug or potential
toxicities of investigational drugs, lack of clinical evidence to Individual companies adopt very different philosophies with
justify investigational treatment of a specific indication, or regard to compassionate use access. Most companies
safety concerns related to the grave condition of a patient. maintain a publicly available web resource that outlines their
The compassionate use application also requires knowl- compassionate use policy. Some companies have stated
edge of the regulations and administrative process, which that their policy is not to provide access to an investigational
may include a contractual arrangement with the company drug outside of a clinical trial.17 Adoption of a more liberal
providing the drug. Many community hospital physicians compassionate use policy becomes increasingly important
have little or no experience with compassionate use and as the body of evidence accumulates showing that the drug
may not even know how to start the process for applying for has robust and reproducible activity in a defined population.
compassionate use medications.14 It is especially vital for companies to provide patient access,
if possible, when clinical results are so compelling that there
Although the compassionate use process has been sim-
is no longer equipoise regarding whether a drug is more
plified, it takes considerable time for a physician to re-
effective than available treatments. Providing compas-
search potentially beneficial drugs, discuss the benefits
sionate use may also be particularly important for drugs that
and risks of the investigational treatment with the patient or
target ultrarare mutations, because it can be impractical for
the family, obtain agreement from the sponsor to provide
a drug sponsor to open clinical sites in each geographic
the investigational treatment, work with an IRB, obtain
location where a patient may be identified.
informed consent, administer the investigational treat-
ment, and fulfill the responsibilities of an IND sponsor, Nonetheless, companies may choose to restrict or deny
which include reporting serious and unexpected adverse requests to provide drugs for compassionate use for many
events related to the investigational drug and providing valid reasons. These include concerns about patient safety,
a brief summary of patient outcome annually and at the because patients not meeting trial eligibility may have rel-
conclusion of treatment. In addition, compassionate use evant risk factors to the safety of the experimental agent;
access often requires the negotiation of confidential dis- concerns regarding fairness of drug distribution when
closure agreements and treatment use agreements with supply is limited; or concerns that the organizational ex-
the drug manufacturer and regulatory and research co- pense of compassionate use may interfere with securing
ordinator support for physicians; also, physicians outside regulatory approval for the experimental therapy. Drug
academic medical centers might need to locate an IRB. In companies can request authorization from the FDA to re-
most cases, all of the costs associated with this time and cover the direct costs incurred in making a drug available to
effort are uncompensated. patients from the onset of manufacturing to the point when it

Gerasimov et al
arrives at the destination where the patient will be treated; and approves after a drug sponsor agrees to provide the
however, the challenges of so-called “request to charge” drug; the number of requests denied by drug sponsors is not
applications and the difficulty of securing payment from known.19,23 In the first 3 months, Project Facilitate staff
insurers results in the provision of most compassionate processed 161 compassionate use requests for oncology
access drugs free of charge to the patient.18 drugs, which represents a 20% increase over the number of
Risk aversion, especially when paired with limited familiarity applications for oncology drugs that the agency received in
with the associated regulatory complexity, is another reason the same period in the prior year.24
why companies avoid granting compassionate use access. FDA review of expanded access requests is beneficial,
Because the FDA has the authority to put a clinical hold on because the agency has data and information from the
a trial, many drug sponsors are concerned that adverse sponsor about safety, toxicity, or dosage that may not be
events that may occur in compassionate use patients will available to physicians. It also ensures patient protections,
negatively impact a drug-development program. Although, such as IRB review and informed consent.7 In approxi-
in 10 years, the FDA placed only 0.2% of clinical studies on mately 11% of expanded access cases, FDA medical offi-
clinical hold as a result of adverse events in patients re- cers recommend changes to a request, such as dose
ceiving drugs through expanded access, drug companies adjustment, safety monitoring, or informed consent.11
can be risk averse. For comparison, the incidence of clinical
The FDA is working to assuage concerns of drug developers
holds for all commercial investigational drug-development
that reports of adverse events affecting patients who receive
programs is 7.9%.19The FDA points out that expanded
drugs under expanded access are likely to delay or derail
access has never led to a negative regulatory decision re-
marketing applications.9,19 The agency points out that the
garding a drug application20 and has stated that concerns
regulations require expedited reporting of serious adverse
about the impact on approval decisions should not cause
events only if “there is evidence to suggest a causal re-
companies to deny expanded access requests.19
lationship between the drug and the adverse event.”25
In addition to providing investigational treatments to pa- Additionally, safety information submitted to the FDA that
tients, compassionate use can provide meaningful in- is generated through compassionate use of oncology drugs
formation that contributes to the overall development is generally reviewed by trained oncologists who understand
program for drugs. For example, single patient protocols that patients with advanced cancer experience severe se-
may explore doses that are higher than those studied in quelae from their cancer unrelated to any drug treatment.
clinical trials to improve brain delivery or to overcome
The FDA notes that there may be important advantages to
a resistance mutation. They may also generate data in new
having the sponsor, rather than a physician, submit an
populations, including children, that are not yet included in
individual patient expanded access protocol to an existing
clinical protocols, or for novel drug-drug combinations,
IND when possible.9 If sponsors hold INDs they are more
based on tumor genomic profiling, that, if successful, could
likely to identify “ultraresponders” who can guide future
launch stand-alone clinical trials. Companies can benefit
discovery and development activities, and the information
from providing expanded access when data received from
accumulated from compassionate use can be more easily
these programs are used to support a marketing application.
leveraged in the clinical development program, particularly
The FDA has used data acquired through compassionate
for drugs under development for rare cancers, including
access to support its regulatory decision-making, particu-
pediatric cancers.26
larly for rare diseases.20,21
U.S. FDA COMPASSIONATE USE FOR PEDIATRIC PATIENTS
The FDA supports providing compassionate use drugs to An analysis of 398 expanded access programs registered on
patients in need.19 On May 31, 2019, in an effort to help ClinicialTrials.gov shows that, although only 6% of programs
patients with cancer access investigational drugs, FDA were open exclusively to children, 41% of all expanded
launched Project Facilitate,22 a pilot program to help on- access programs were open to at least a subgroup of pe-
cologists submit expanded access requests. In addition to diatric patients younger than age 18.27 In a study conducted
staffing a call center to guide oncologists and their staff at a large cancer center, children represented 34% of ex-
through the expanded access process, Project Facilitate will panded access use, although they represented just 2% of
gather data provided by oncologists on outcomes of ex- the cancer center’s patients.This study suggested that
panded applications, such as the reasons why drug single patient INDs “may provide an important means of
sponsors decide not to provide their drugs for compas- pediatric drug access.”16
sionate use.15 Compassionate use in pediatric patients raises unique
The only data available on the demand for compassionate considerations related to formulation, dosing, pediatric-
use drugs is the number of requests that the FDA reviews specific toxicity, and the ability to provide informed

consent. However, compassionate use of investigational development, must be considered, and clinical monitoring
cancer drugs in children can be considered, even if the drug plans must be instituted to address these concerns, as
has not been previously studied in children, provided that appropriate. An appropriate care team will include physi-
there is a reasonable possibility, based upon the drug’s cians with expertise in treating children and in the use of the
mechanism of action, nonclinical data, or prior clinical class of agents, if possible. A routine part of the FDA’s review
experience in pediatric or adult patients, that the drug may of compassionate access applications for pediatric patients
provide a benefit to the child without posing undue risks and includes evaluation of the toxicology data that were sub-
there are no satisfactory alternative treatments. mitted to the sponsor’s original IND to support human in-
vestigation for potential toxicity signals that are relevant to
For pediatric patients to benefit from treatment with an
pediatric patients. However, it should be noted that sub-
investigational drug through compassionate use, there must
stantial interspecies variability may underestimate or over-
be a feasible way to reliably administer the appropriate
estimate such concerns, and the lack of juvenile animal
dose, ideally through an available age-appropriate formu-
data should not preclude compassionate access in children.
lation. School-aged children can often swallow capsules,
but smaller dosing increments, such as those used early in Finally, IRBs have differing requirements for the age at
a phase I trial, may be needed to achieve the desired dose. which assent is required for children, varying from school-
An oral taste-masked liquid may be needed for patients who aged children to teenagers. As with consent, assent should
are too young or unable to swallow capsules. Sometimes, an include a discussion of the investigational nature of the
extemporaneously compounded formulation may serve as treatment and the potential for known and unknown side
an interim solution while formal liquid-development activi- effects of therapy. Regardless of whether written or verbal
ties are underway. If a liquid formulation is not feasible or assent is required, as with participation in a clinical trial, it is
available, “sprinkling” the contents of an open capsule onto vital that pediatric patients be included in the decision-
a semisolid vehicle, such as applesauce, can be consid- making process for compassionate use, to the extent that is
ered, although this approach can be associated with less developmentally appropriate, as well as that the treatment
precise dose measurement. When considering these op- goals of the parents/guardians and pediatric patient are
tions, the potential for alterations in drug stability and the aligned as much as possible.
ability to achieve a consistent dose deemed safe for the Despite the additional work and potential program risks
patient are important factors. associated with supporting a pediatric compassionate use
Although the dosage that is safe for adolescents is typically request, most sponsors feel a moral imperative to be sup-
the same as the adult recommended phase II dose or portive when an investigational agent has begun to show
maximum tolerated dose, an age-appropriate dose should efficacy. In addition, pediatric compassionate use experi-
ence can provide substantial benefits to a promising in-
usually be determined for younger pediatric patients. For
vestigational agent: real-time pharmacokinetics information,
children older than age 2, proportionally reducing the adult
better trial design for formal pediatric studies, opportunities
dose relative to body surface area is often appropriate.28
for new indications, and goodwill among physician in-
Especially when treating infants, ideally the sponsor will
vestigators. However, supporting pediatric compassionate
have already undertaken pharmacokinetics modeling that
use requests prior to efficacy proof of concept or in un-
takes into account the known ontogeny of the relevant drug
selected populations not yet associated with efficacy
metabolism enzymes and profound changes in glomerular
may have a different risk-reward analysis. In these cases,
filtration rate in the first months of life. In young children and
transparency about the risk-benefit assessment with the
for agents without an established adult dose, a program of
petitioning physician typically leads to a consensus-driven
real-time pharmacokinetics assessment and dose adjust-
decision.
ment, although not necessarily required, can often best
protect against underdosing and overdosing. Because EXPANDED ACCESS IN ACTION
compassionate use is undertaken with therapeutic intent, Many patients and their families believe that a compas-
known underdosing—relative to exposures associated with sionate use drug can save or prolong life, but there are
efficacy in adults—is an unwanted outcome. In cases in limited data available about the outcomes of unapproved
which there are age-specific safety concerns or a relative drugs provided outside of a clinical trial. Studies have re-
paucity of information to inform pediatric dosing, it may be ported an overall response rate of approximately 20% for
appropriate to consider a conservative approach to the initial drugs obtained through compassionate use for a variety of
starting dose, with provisions for intrapatient dose escala- cancers.16
tion, as tolerated.
The impacts of expanded access extend beyond the effects
The possibility of adverse events unique to the pediatric on individual patients. Data from expanded access have
setting, such as effects on skeletal maturity or neurologic helped to characterize the safety and efficacy of drugs29 and

Gerasimov et al
could produce insights for off-label coverage by insurers. 2018.36-38 This law is another way for patients who have
Information from expanded access trials may be particularly been diagnosed with life-threatening diseases or conditions,
valuable for the development of drugs in ultrarare diseases, have tried all approved treatment options, and are unable to
including many pediatric cancers.19 For example, a com- participate in a clinical trial to access certain unapproved
plete or partial response was reported in 34% of patients treatments.
treated with LOXO-195, an investigational drug currently
Currently, patients with terminal illnesses and their physi-
under development for the treatment of patients with
cians may seek investigational treatment through the ex-
cancers harboring an NTRK mutation; these results were
panded access process or an RTT request. Individual states’
described in an abstract summarizing data obtained from
RTT laws may have different requirements.
20 patients enrolled in a clinical trial and 11 patients who
received LOXO-195 in an expanded access program.30 Investigational treatments obtained through the federal RTT
mechanism do not include a requirement for evaluation of the
Data from an expanded access program with more than
safety and overall benefit-risk assessment for compassionate
1,000 patients played a role in the FDA’s approval of
use by IRBs or the FDA.39 RTT also does not require drug
LUTATHERA (lutetium 177Lu dotatate), a radiopharmaceuti-
sponsors to provide an investigational product.40 Compre-
cal for the treatment of gastroenteropancreatic neuroendo-
hensive data on the use of RTT are not available.
crine tumors.31 In an announcement about the approval of
Mepsevii (vestronidase alfa-vjbk) for mucopolysaccharidosis COMPASSIONATE USE BEYOND ONCOLOGY
type 7, the FDA stated that “the most compelling evidence of
Although this review focuses on compassionate use in pe-
a treatment effect” was observed in four patients; one of these
diatric patients with cancer, the perspectives are relevant to
four received the drug under an emergency single patient
the treatment of a broad range of life-threatening diseases.
compassionate use request.32 Other drugs for which approval
The process of requesting compassionate use is identical for
included consideration of data from expanded access are
all diseases. Although Project Facilitate is a resource that is
Voraxaze (glucarpidase), Epidiolex (cannabidiol), and the
specifically intended to guide oncologists through the process
BEXSERO meningococcal vaccine.33
of submitting requests for expanded access to investigational
Studies have demonstrated that between 43% and 76% of cancer drugs, the FDA.gov website provides information,
drugs provided on a compassionate use basis were later including relevant FDA contact information, for physicians
approved.7,27 The FDA considers the expanded access seeking assistance with applying for expanded access for
program to be successful “based on both the volume of investigational drugs for other diseases.41
requests handled each year and the extremely high rate of
The ongoing COVID-19 pandemic caused by severe acute
approval of these requests.”19
respiratory syndrome coronavirus 2 highlights the appli-
Although there is limited information regarding the in- cability of this topic to rapidly emerging infectious diseases
cidence of adverse events reported for patients receiving without FDA-approved therapies.42,43 The same requirements
compassionate use, one study of patients with cancer for compassionate use applications, described above and in
treated on compassionate use protocols reported that Figure 1, apply to investigational treatments for COVID-19.
29.7% of patients experienced one or more treatment- Hematologists and oncologists and their research staff, who
related serious adverse events, including 19.1% of pedi- typically submit a substantial fraction of expanded access
atric patients.16 These adverse event rates are in line with requests,6 may be able to serve as a resource for other phy-
the incidence of adverse events reported in pediatric pa- sicians with less experience in requesting compassionate use.
tients with cancer receiving standard therapies.34 As for oncology products, alignment of the goals of patients
Consideration of safety is especially important, because seeking urgent therapy, physicians treating these patients,
patients requesting compassionate use treatment, who drug manufacturers (who have to balance resources between
typically have advanced refractory cancers that are heavily compassionate access requests and clinical trials), and reg-
pretreated, may be particularly vulnerable to drug toxicities. ulators is crucial.
They are also more vulnerable because they have few
treatment options and may have limited data with which to CONCLUSIONS
make an informed decision because of the earlier stage of Although the perspectives of stakeholders differ, they con-
development of many drugs that are available through verge in the common goal of getting effective cancer
compassionate use.35 treatments to patients as quickly as possible. There is also
a broad agreement that, even in the earliest stages, clinical
RIGHT TO TRY research in oncology is intended to provide the possibility of
Starting in 2014, Right to Try (RTT) laws were passed in therapeutic benefit for patients as well as generalizable
many states, and a federal RTT law was enacted in knowledge.

FIGURE 1. Expanded Access for

an Individual Patient Process
Overview
The blurring of once sharp lines between research and even ethical barriers to providing compassionate use ac-
treatment makes access to investigational drugs a high cess. The entire oncology community, including physi-
priority. When patients, their families, and their physicians cians, patients and patient advocates, regulators, drug
learn that an experimental drug may be helping adults in companies, and payers, must work to overcome these
a clinical trial, children who cannot participate in the trial barriers. Education about compassionate use is the starting
deserve access. There can be logistic, commercial, and point.
AFFILIATIONS ACKNOWLEDGMENT
1
Kids v Cancer, Washington, D.C. This article reflects the views of the authors and should not be construed to
2
U.S. Food and Drug Administration, Silver Spring, MD represent the FDA’s views or policies.
3
Loxo Oncology at Lilly, Stamford, CT
4
Department of Pediatrics, UT Southwestern Medical Center, Dallas, TX
5
Pauline Allen Gill Center for Cancer and Blood Disorders, Children’s AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Health, Dallas, TX AND DATA AVAILABILITY STATEMENT
6
Harold C. Simmons Comprehensive Cancer Center, UT Southwestern Disclosures provided by the authors and data availability statement (if
Medical Center, Dallas, TX applicable) are available with this article at DOI https://doi.org/10.1200/
EDBK_278995.
Theodore W. Laetsch, MD, Department of Pediatrics, G3.240, UT
Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX
75390; Twitter: @LaetschT; email: [email protected].

Gerasimov et al
REFERENCES
1. U.S. Food and Drug Administration. Expanded Access. https://www.fda.gov/news-events/public-health-focus/expanded-access. Accessed January 28, 2020.
2. Hoerger M. Right-to-try laws and individual patient “compassionate use” of experimental oncology medications: A call for improved provider-patient com-
munication. Death Stud. 2016;40:113-120.
3. Burris HA III. Correcting the ASCO position on phase I clinical trials in cancer. Nat Rev Clin Oncol. 2020;17:125.
4. Adashek JJ, LoRusso PM, Hong DS, et al. Phase I trials as valid therapeutic options for patients with cancer. Nat Rev Clin Oncol. 2019;16:773-778.
5. U.S. Food and Drug Administration. Expanded Access (compassionate use) submission data. https://www.fda.gov/news-events/expanded-access/expanded-
access-compassionate-use-submission-data#CBERCDER. Accessed January 28, 2020.
6. U.S. Food and Drug Administration. Expanded Access (compassionate use) submission data. CDER, CBER and CDRH Expanded Access INDs and Protocols
(2014-2018). https://www.fda.gov/news-events/expanded-access/expanded-access-compassionate-use-submission-data#CBERCDER. Accessed March 16,
2020.
7. McKee AE, Markon AO, Chan-Tack KM, et al. How often are drugs made available under the Food and Drug Administration’s expanded access process
approved? J Clin Pharmacol. 2017;57(suppl 10):S136-S142.
8. Van Norman GA. Expanding patient access to investigational drugs: single patient investigational new drug and the “Right to Try”. JACC Basic Transl Sci. 2018;
3:280-293.
9. U.S. Food and Drug Administration. Expanded Access to Investigational Drugs for Treatment Use: Questions and Answers. Guidance for Industry. https://www.
fda.gov/media/85675/download. Accessed January 28, 2020.
10. Puthumana J, Miller JE, Kim J, et al. Availability of investigational medicines through the U.S. Food and Drug Administration’s Expanded Access and
Compassionate Use Programs. JAMA Netw Open. 2018;1:e180283.
11. Jarow JP, Lurie P, Ikenberry SC, et al. Overview of FDA’s Expanded Access Program for investigational drugs. Ther Innov Regul Sci. 2017;51:177-179.
12. U.S. Food and Drug Administration. Individual Patient Expanded Access Applications: Form FDA 3926. https://www.fda.gov/regulatory-information/search-fda-
guidance-documents/individual-patient-expanded-access-applications-form-fda-3926. Accessed January 28, 2020.
13. Gottlieb S. Testimony from Scott Gottlieb, MD, on Examining Patient Access to Investigational Drugs Before Committee on Energy and Commerce, October 3,
2017. https://www.hhs.gov/about/agencies/asl/testimony/2017-10/examining-patient-access-to-investigational-drugs.html. Accessed January 28, 2020.
14. Moerdler S, Zhang L, Gerasimov E, et al. Physician perspectives on compassionate use in pediatric oncology. Pediatr Blood Cancer. 2019;66:e27545.
15. Usdin S. FDA to Facilitate Access to Unapproved Drugs. https://www.biocentury.com/article/299854/how-fda-plans-to-help-patients-get-expanded-access-to-
unapproved-drugs. Accessed January 28, 2020.
16. Feit NZ, Goldman DA, Smith E, et al. Use, safety, and efficacy of single-patient use of the U.S. Food and Drug Administration Expanded Access Program. JAMA
Oncol. 2019;5:570-572.
17. Reagan-Udall Foundation for the Food and Drug Administration. Company Directory. https://navigator.reaganudall.org/company-directory. Accessed January
28, 2020.
18. U.S. Food and Drug Administration. Charging for Investigational Drugs Under an IND: Questions and Answers. https://www.fda.gov/media/85682/download.
Accessed January 28, 2020.
19. Jarow JP, Lemery S, Bugin K, et al. Expanded access of investigational drugs: the experience of the Center of Drug Evaluation and Research over a 10-year period.
Ther Innov Regul Sci. 2016;50:705-709.
20. Jarow JP, Moscicki R. Impact of expanded access on FDA regulatory action and product labeling. Ther Innov Regul Sci. 2017;51:787-789.
21. U.S. Food and Drug Administration. Center for Drug Evaluation and Research: Application Number: 210861Orig1s000, 211710Orig1s000. Multi-Discipline
Review. https://www.accessdata.fda.gov/drugsatfda_docs/nda/2018/210861Orig1s000_211710Orig1s000MultidisciplineR.pdf. Accessed March 16, 2020.
22. U.S. Food and Drug Administration. Project Facilitate. https://www.fda.gov/about-fda/oncology-center-excellence/project-facilitate. Accessed January 28, 2020.
23. U.S. Food and Drug Administration. FDA announces Project Facilitate to assist physicians seeking access to unapproved therapies for patients with cancer. June
3, 2019. https://www.fda.gov/news-events/press-announcements/fda-announces-project-facilitate-assist-physicians-seeking-access-unapproved-therapies-
patients. Accessed January 28, 2020.
24. Usdin S. FDA’s Project Facilitate Fielded 161 Compassionate Use Requests In First Three Months. https://www.biocentury.com/article/303728/fda-rsquo-s-
project-facilitate-fielded-161-compassionate-use-requests-in-first-three-months. Accessed January 28, 2020.
25. U.S. Food and Drug Administration. Investigational New Drug Application. 21 CFR 312.32 (2019). https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/
CFRSearch.cfm?fr=312.32. Accessed March 25, 2020.
26. Donoghue M. FDA and the Challenges of Drug Review in the 21st Century: Immunotherapies, Targeted Therapies, and Companion Diagnostics. https://www.
canceradvocacy.org/wp-content/uploads/2015/04/Dr.-Donoghue-Presentation-March-2015-CPR.pdf. Accessed February 3, 2020.
27. Miller JE, Ross JS, Moch KI, et al. Characterizing expanded access and compassionate use programs for experimental drugs. BMC Res Notes. 2017;10:350.
28. Lee DP, Skolnik JM, Adamson PC. Pediatric phase I trials in oncology: an analysis of study conduct efficiency. J Clin Oncol. 2005;23:8431-8441.
29. Ascierto PA, Simeone E, Sileni VC, et al. Clinical experience with ipilimumab 3 mg/kg: real-world efficacy and safety data from an expanded access programme
cohort. J Transl Med. 2014;12:116.

30. The ASCO Post. AACR 2019: Phase I Trial Evaluates LOXO-195 in Patients With NTRK-Positive Solid Tumors. https://www.ascopost.com/News/59906. Accessed
February 3, 2020.
31. U.S. Food and Drug Administration. FDA approves new treatment for certain digestive tract cancers. https://www.fda.gov/news-events/press-announcements/
fda-approves-new-treatment-certain-digestive-tract-cancers. Accessed February 3, 2020.
32. U.S. Food and Drug Administration. Center for Drug Evaluation and Research: Application Number: 761047Orig1s000. Clinical Review(s). https://www.
accessdata.fda.gov/drugsatfda_docs/nda/2017/761047Orig1s000MedR.pdf. Accessed February 2, 2020.
33. Rawson K. Expanded Access Data Can Support Approval Decisions, U.S. FDA Says. https://pink.pharmaintelligence.informa.com/PS124296/Expanded-Access-
Data-Can-Support-Approval-Decisions-US-FDA-Says. Accessed February 3, 2020.
34. Holdsworth MT, Fichtl RE, Behta M, et al. Incidence and impact of adverse drug events in pediatric inpatients. Arch Pediatr Adolesc Med. 2003;157:60-65.
35. Fountzilas E, Said R, Tsimberidou AM. Expanded access to investigational drugs: balancing patient safety with potential therapeutic benefits [published
correction appears in Expert Opin Investig Drugs. 2018;27:209]. Expert Opin Investig Drugs. 2018;27:155-162.
36. Wendler T, Mongiello F, McLinn J, Bellina M. Right to Try Act of 2017. 115th U.S. Congress; May 30, 2018; p. 132, Stat. 1372; Public Law No. 115-176.
37. Simmons Z. Right-to-Try investigational therapies for incurable disorders. Continuum (Minneap Minn). 2017;23:1451-1457.
38. Parson P. Colorado First State To Pass ‘Right to Try’, or the ‘Dallas Buyer’s Club’ Law. https://www.pbs.org/newshour/health/colorado-first-state-pass-right-try-
dallas-buyers-club-law. Accessed March 25, 2020.
39. Klein R. Right to Try legislation should focus on patients, not politics. P T. 2018;43:147-148.
40. Bateman-House A. “Right To Try” Is Law, Now What?: Part 1. https://www.healthaffairs.org/do/10.1377/hblog20181024.111856/full/. Accessed January 28,
2020.
41. Food and Drug Administration. FDA’s Expanded Access Contact Information. https://www.fda.gov/news-events/expanded-access/fdas-expanded-access-
contact-information. Accessed March 25, 2020.
42. Kalil AC. Treating COVID-19-Off-Label Drug Use, Compassionate Use, and Randomized Clinical Trials During Pandemics. JAMA. Epub 2020 Mar 24.
43. Gilead. Gilead Sciences Statement on Access to Remdesivir Outside of Clinical Trials. https://www.gilead.com/news-and-press/company-statements/gilead-
sciences-statement-on-access-to-remdesivir-outside-of-clinical-trials. Accessed March 25, 2020.

PROFESSIONAL
DEVELOPMENT AND
EDUCATION ADVANCES

On the Shoulders of Giants: The Evolution of Renal

Cell Carcinoma Treatment—Cytokines, Targeted
Therapy, and Immunotherapy
Janice P. Dutcher, MD1; Ronan Flippot, MD, MSc2; Jaleh Fallah, MD3; and Bernard Escudier, MD2
overview
The treatment of advanced renal cell carcinoma (RCC) has evolved dramatically over the past 30 years, as has
a better understanding of the biology of the disease, knowledge of multiple subtypes with distinct molecular
abnormalities, and improved comprehension of the perturbed pathways that lead to the development and
growth of RCC. This is no longer a monolithic disease, although the majority of tumors are of the clear cell
subtype. However, progress is being made in other subtypes as well, as molecular profiles are better un-
derstood and as new agents show activity. Immunotherapies remain a major category of treatment, from
cytokines to checkpoint inhibitors to ex vivo activated cellular therapy. Antiangiogenesis tyrosine kinase
inhibitors are also an important part of the armamentarium. Because these approaches have evolved, we are
now in the era of combination therapy using agents of differing mechanisms to try to achieve synergy to
increase response rates and create durable responses leading to prolonged survival. Renal cell carcinoma as
a tumor is unique in that there has always been a subset of patients who achieve complete responses that last
for many years without subsequent treatment. Thus, the goal of further development is to enlarge this subset
using new therapeutic approaches and to achieve further durable responses and treatment-free survival.
THE BEGINNINGS OF IMMUNOTHERAPY FOR RENAL demonstrated that immune cellular activation was key to
CELL CARCINOMA: CYTOKINES this antitumor effect and suggested a dose-response
Immune Cytokines effect as well.5-8 Preclinical studies also demonstrated
synergy of combinations of IL-2 and IFN-α in murine
Immune cytokines are endogenous substances pro- tumor models.9
duced by a wide range of immune cells (including
lymphocytes, dendritic cells, macrophages, and others) In vitro and early in vivo studies of cytokines as ther-
that signal between immune cells and are mediated by apeutic agents used small amounts of immune cell–
receptors. Cytokines induce or inhibit various cellular derived IFN and IL-2, but it was with the development
regulatory proteins, and some have direct effects on of recombinant DNA technology, allowing produc-
tion of milligram quantities of these agents, that large-
various tissues. Therefore, immune cytokines have been
scale therapeutic human clinical trials could be
studied as potential therapeutic compounds, demon-
investigated.10,11
strating clinical activity against RCC. These include in-
terferons (IFNs), such as IFN alpha (IFN-α), IFN beta Immunotherapy for Renal Cell Carcinoma
(IFN-β), and IFN gamma (IFN-γ), as well as interleukin Several clinical observations on the natural history of
(IL)-2. Interferons were identified as substances pro- RCC, and its lack of sensitivity to standard chemo-
Author affiliations duced during viral infections and were found to affect therapies, have led to the investigation of immuno-
and support the function of both normal and malignant cells.1,2 therapy in RCC. Clinical observations were interpreted
information (if Several biologic effects were observed, including anti- to suggest that the immune system could have
applicable) appear
angiogenic activity, downregulation of certain onco- a therapeutic effect on advanced RCC. First, it was
at the end of this
article. genes, and antiproliferative effects, as well as stimulation often observed that the time from nephrectomy to
Accepted on of effector cells in the immune system.3 IL-2 was initially development of metastatic disease could be many
February 18, 2020 identified as a T-cell growth factor for its direct function in years, suggesting some type of biologic control. Sec-
and published at expanding populations of T lymphocytes, natural killer ond, when either primary or metastatic disease was
ascopubs.org on cells, and lymphokine-activated killer cells.4 The anti- resected, the frequent occurrence of lymphocytic in-
March 31, 2020:
tumor potential of IL-2 was initially shown through lysis filtration was noted.12 Third, the spontaneous re-
10.1200/EDBK_ of autologous tumor cells by human lymphocytes cul- gression of metastatic lesions subsequent to removal of
280817 tured in IL-2.5,6 In vitro and murine studies of IL-2 the primary renal tumor was occasionally documented,

Evolution of Renal Cell Carcinoma Treatment: Cytokines, Targeted Therapy, Immunotherapy
suggesting that potential immune or growth factors may

affect tumor progression.13,14
• Immunotherapy in the treatment of metastatic Early Efforts in Immunotherapy of Renal Cell Carcinoma
renal cell cancer, including high-dose
Initial cytokine clinical trials in advanced/metastatic RCC
interleukin-2 and immune checkpoint inhibitors
(mRCC) were with single-agent IFNs, and responses, pri-
to date, yields complete responses of clinically
important duration and is effective in both first- marily partial, were observed with all three types of IFN (IFN-
and second-line treatment. α, IFN-β, and IFN-γ) and some long-lasting responses were
reported.15-17 There also appeared to be a dose-response
• VEGFR–tyrosine kinase inhibitor therapy, tar-
effect.15-17 Early studies also combined IFN-α with IFN-γ
geting angiogenesis, has demonstrated high
response rates, and these agents remain and with various chemotherapeutic agents, including, in-
a cornerstone of systemic therapy for metastatic dividually, vinblastine or carmustine.18-21
renal cell cancer. Based on animal studies that demonstrated a dose re-
• Multi-targeted VEGFR-TKIs demonstrate activ- sponse, the initial clinical trials of IL-2 used maximal doses
ity in non–clear cell renal cell cancer. Clinical (high dose [HD]) of IL-2 in combination with the infusion of
trials combining CPI and anti-VEGF therapy are ex vivo activated autologous immune cells (lymphokine-
ongoing. activated killer cells).22-26 Subsequently, comparative trials
• Immunotherapy-based combination therapies showed that administration of HD IL-2 alone produced
have shown higher efficacy compared with similar in vivo cellular activation and response rates, in-
VEGFR–tyrosine kinase inhibitor monotherapy cluding complete responses.27,28 Further evaluations in-
in first-line treatment of clear cell metastatic cluded studies of intravenous continuous infusion and
renal cell cancer and demonstrate activity in combinations of IL-2 with subcutaneous injections of IFN-α
non–clear cell renal cell cancer.
in HD and moderate-dose regimens.25,29-31 The combina-
• Patients with metastatic renal cell cancer who tion appeared superior in activity to single-agent IL-2 or IFN-
have limited metastatic disease burden can α as administered in this study but also had more toxicity.31
initially be considered for local therapy or ob- Other researchers evaluated strictly subcutaneous regi-
servation to postpone the potential toxicities of
mens to allow outpatient administration with IL-2 alone32
systemic therapy.
or with moderate-dose combinations of IL-2 and IFN-α.33
However, a large randomized trial that evaluated moderate
TABLE 1. Risk Factor Stratification for Metastatic Renal Cell Cancer: Poor Risk Factors and Median Survival
Favorable Median Intermediate Median Poor Median
Composites of Survival Risk Criteria OS OS OS
MSKCC risk criteria—era of KPS , 80%
cytokines38,44
Time from diagnosis to systemic treatment , 0 risk factors 1–2 risk factors 3 risk factors
1 year
Hgb , LLN 20 months 10 months 4 months
Corrected serum calcium . 10 mg/dL
LDH . ULN
IMDC risk criteria—era of KPS , 80% or ECOG 1
anti-VEGF/VEGFR agents41
Time from diagnosis to systemic treatment , 0 risk factors 1–2 risk factors 3 risk factors
1 year
Hgb , LLN 35.3 months 27 months 8.8 months
Corrected serum calcium . ULN
Neutrophils . ULN
Platelets . ULN
Abbreviations: OS, overall survival; MSKCC, Memorial Sloan Kettering Cancer Center; KPS, Karnofsky Performance Status; Hgb, hemoglobin; LLN, lower level
of normal; LDH, lactate dehydrogenase; ULN, upper limit of normal; IMDC, International Metastatic Renal Cell Carcinoma Database Consortium; ECOG,
Eastern Cooperative Oncology Group.

Dutcher et al
doses of subcutaneous IL-2 and IFN-α, either alone or in to 10% complete responses (CRs) as well as surgically
combination, in intermediate prognosis disease failed to show completed CRs, the majority of which are extremely durable
improved outcome for cytokines compared with medrox- with no further treatment.51-57 Durable partial responses are
yprogesterone acetate, and, in that study, cytokines were as- also reported.51-57 The prospective trial again noted that clear
sociated with greater toxicity than medroxyprogesterone cell histology was associated with response to cytokines,
acetate.34 Subsequently, two large randomized studies have confirming historical data.34,51,58,59 Contemporary data also
demonstrated higher response rates and survival with HD demonstrate that HD IL-2 can safely be administered after
intravenous IL-2 compared with lower-dose intravenous or antiangiogenesis therapy60 and after checkpoint inhibitors,61
subcutaneous IL-2 or compared with the combination of with results similar to de novo–treated patients.60,61
moderate-dose SQ IL-2 plus IFN-α.35,36
The most recent HD IL-2 data reported, derived from a real-
Additional cytokine combinations have not successfully life, largely prospectively acquired registry (PROCLAIMSM)
enhanced the overall response rate (ORR) and response collected from 2011 to 2017, presents response and sur-
duration of HD IL-2 alone, and further studies of combi- vival by International Metastatic RCC Database Consortium
nations of IL-2 with newer immunotherapies are ongoing. (IMDC) risk category and demonstrates again the durable
During the evolution of mRCC therapy with cytokines, survival of patients with mRCC treated with HD IL-2, par-
several research groups evaluated parameters associated ticularly those with favorable and intermediate features, with
with outcomes to cytokine therapy, stratifying patients into a median overall survival (OS) that could exceed 5 years
favorable, intermediate, and poor prognosis risk groups (Table 2; Fig. 1).55 Additional information from 11 centers
(Table 1). At that time, no subsequent therapy was that voluntarily compiled institutional experience on 100
available.37-40 However, this approach continues to provide patients with either melanoma or RCC who responded to IL-
a consistent predictor of long-term outcomes with newer 2 showed patients surviving more than 5 years from the
and sequential therapies for mRCC (Table 1).41-43 initiation of therapy.62 Of the 46 patients with mRCC, 38 had
CR from HD IL-2 alone, 5 had CR after local treatment
Current Status of HD IL-2 Therapy following response to IL-2, and 3 had prolonged partial
Many years of effort and experience have led investigators to response with no subsequent therapy. Median disease-free
establish guidelines for the safe administration of HD IL-2, survival was 10.5 years (range, 5–30 years). Of this group,
including a careful physiologic screening process for po- 27 were alive 10 years or longer after HD IL-2, with no
tential patients and guidance for the development of well- subsequent therapy.62
trained and experienced clinical teams.45,46 Reports after
more than 10 years of follow-up from initial HD IL-2 trials FROM TARGETED MOLECULAR THERAPIES TO IMMUNE
showed that responses were durable and, most importantly, CHECKPOINT INHIBITORS IN mRCC
that relapse was infrequent in patients experiencing com- The Decade of Antiangiogenic Therapies
plete responses.47-50 A major breakthrough in the understanding of the biology
Contemporary reports of HD IL-2 as treatment of patients of clear cell RCC (ccRCC) was the discovery in the early
with mRCC, including a multicenter prospective clinical trial,51 1990s of recurrent alterations of the tumor suppressor
single-institution data,52-54 and current registry data,55-57 de- gene Von Hippel-Lindau (VHL) that led to oncogenesis and
scribe a consistent ORR of 15% to 25%, including at least 5% ccRCC.63 Evidence of biallelic inactivation of VHL by loss of
TABLE 2. Outcome of IL-2–Treated Patients With mRCC by IMDC Criteria: Response and Survival—PROCLAIMSM Registry Database*
All Patients (810 Patients) Favorable (249 Patients) Intermediate (480 Patients) Poor (81 Patients)
CR 9 (3.6%) 34 (7%) 1 (1.2%)
PR 52 (21%) 91 (19%) 13 (16%)
SD 128 (51.4%) 185 (38.5%) 24 (30%)
CR + PR 61 (24.5%) 125 (26%) 14 (17.3%)
CR + PR + SD 189 (75.9%) 310 (64.6%) 38 (46.9%)
Median OS, months (95% CI) 63.3 (51.2–73) 42.4 (35.4–56) 14 (10–24.5)
2-year OS 77.6% 68.2% 40.4%
Abbreviations: mRCC, metastatic renal cell carcinoma; IMDC, International Metastatic Renal Cell Carcinoma Database Consortium; CR, complete response;
PR, partial response; SD, stable disease; OS, overall survival.
*Of 939 patients with metastatic RCC, 810 with data for all six IMDC criteria.
Adapted from Fishman et al.57

FIGURE 1. Overall Survival of All

Patients With mRCC Treated With
HD IL-2 in PROCLAIM Registry
With All Six International Meta-
static RCC Database Consortium
Criteria, by Risk Category
Abbreviations: CL, confidence
limit; RCC, renal cell carcinoma.
Adapted from Fishman et al.56
heterozygosity, mutation, or promoter methylation has since patients, with a median PFS of 11 months versus 5 months
been encountered in up to 74% of ccRCC,64,65 and the and a remarkable response rate of 31%70; and sorafenib
discovery of the hypoxia-inducible factor (HIF) pathway and demonstrated superiority versus placebo in patients whose
the link with VHL inactivation were elucidated.66 VHL is disease did not respond to previous cytokine therapy, with
involved in the degradation of the transcription factor HIF in a median PFS of 5.5 versus 2.8 months.69 The most
normoxic conditions. In the absence of VHL, HIF accu- common adverse events (. 20%) with oral tyrosine kinase
mulates and induces the transcription of several genes inhibitors (TKIs) include diarrhea, stomatitis, fatigue, and
involved in tumor growth and angiogenesis, including the palmar-plantar erythrodysesthesia; these side effects may
VEGF gene.67 These discoveries set up the basis to target be managed with dose adjustments or alternate schedules,
angiogenesis in ccRCC, leading to the development of which have been explored with sunitinib.86
therapies directed against VEGF and its receptors (Table 3).
In the span of a few years (from 2005 to 2007), the standard
Bevacizumab, a humanized monoclonal immunoglobulin of care for patients with metastatic ccRCC switched from
directed against VEGF-A, first proved the relevance of this a purely cytokine-based approach to antiangiogenics in
approach. In a randomized phase II clinical trial, bev- both the first-line (sunitinib and bevacizumab plus IFN-α)
acizumab demonstrated an increase in progression-free and the second-line (sorafenib) setting. Whereas pivotal
survival (PFS) more than twofold compared with placebo, trials included mostly patients with good or intermediate
with some patients experiencing partial responses ex- prognoses,44 VEGFR-directed TKIs have been widely used
ceeding 2 years.84 The tolerance profile was excellent, in- in all comers on the basis of retrospective series.87 Several
cluding manageable hypertension and proteinuria. A phase other VEGFR-directed TKIs have since entered standard of
III clinical trial of bevacizumab in combination with IFN-α in care in advanced ccRCC. Pazopanib was approved by the
the first-line setting confirmed the benefit of bevacizumab, U.S. Food and Drug Administration in the first-line setting in
with a median PFS of 10.2 months, compared with 5.4 2009 after a phase III trial demonstrated its superiority over
months in patients treated with IFN-α alone.68 Response placebo.71 Pazopanib was further shown as noninferior to
rate was also improved to 31% from 13% in the control arm, sunitinib in terms of PFS and OS,73 with slight differences in
with a median duration of response exceeding 1 year. This terms of toxicity because pazopanib was associated with
study was confirmed by a second large randomized study.85 less fatigue and improved quality of life. Pazopanib was
In parallel, oral multikinase inhibitors targeting, among preferred by patients in a double-blind, randomized study
others, the VEGF receptor have entered the clinical arena for aiming at determining patient preference.88 In the second-
treatment of mRCC. Sunitinib demonstrated superiority over line setting, axitinib offered compelling activity with a PFS of
IFN-α in a phase III clinical trial that included untreated 6.7 months, compared with 4.7 months with sorafenib.74

422
TABLE 3. Phase III Trials of Targeted Molecular Therapies in Advanced Clear Cell Renal Cell Carcinoma
Trial Experimental Arm Control Arm Line of Therapy N Risk Groups Response Rate Median PFS (months) Median OS (months)
VEGFR-Directed TKI
AVOREN68 IFN-α + bevacizumab IFN-α First line 649 All (, 10% poor risk) MSKCC* 31% vs. 13% 10.2 vs. 5.4 NR vs. 19.8
69
TARGET Sorafenib Placebo Second line and further 903 Low and intermediate MSKCC 10% vs. 2% 5.5 vs. 2.8 NR vs. 14.7
NCT0008388970 Sunitinib IFN-α First line 750 All (, 10% poor risk) MSKCC 31% vs. 6% 11.0 vs. 5.0 NR vs. NR
71,72
NCT00334282 Pazopanib Placebo First and second-line 435 All (, 10% poor risk) MSKCC 30% vs. 3% 9.2 vs. 4.2 22.9 vs. 20.5

COMPARZ73 Pazopanib Sunitinib First line 1110 All IMDC 31% vs. 24% 8.4 vs. 9.5 28.4 vs. 29.3
AXIS74,75 Axitinib Sorafenib Second line and further 723 All (, 10% poor risk) MSKCC 19% vs. 9% 6.7 vs. 4.7 20.1 vs. 19.2
76,77
METEOR Cabozantinib Everolimus Second line and further 658 All MSKCC 21% vs. 5% 7.4 vs. 3.8 21.4 vs. 16.5
TIVO-178 Tivozanib Sorafenib First line 517 All (, 10% poor risk) MSKCC 33% vs. 23% 11.9 vs. 9.1 28.8 vs. 29.3
79 **
TIVO-3 Tivozanib Sorafenib Third and fourth line 350 All IMDC 18% vs. 8% 11.1 vs. 6.0 16.4 vs. 19.7
mTOR Inhibitors
Dutcher et al
GLOBAL ARCC80 Temsirolimus IFN-α First line 626 Intermediate and poor MSKCC 9% vs. 5% 3.8 vs.1.9 10.9 vs. 7.3
Temsirolimus + IFN-α 8% vs. 5% 3.7 vs. 1.9 8.4 vs. 7.3

81
INTORACT Temsirolimus + IFN-α IFN-α + bevacizumab First line 791 All MSKCC 27% vs. 27% 9.1 vs. 9.3 25.8 vs. 25.5
82
INTORSECT Temsirolimus Sorafenib Second line 512 All MSKCC 8% vs. 8% 4.3 vs. 3.9 12.3 vs. 16.6
RECORD-183 Everolimus Placebo Second line and further 410 All MSKCC 1% vs. 0% 4.0 vs. 1.9 NR vs. 8.8
Abbreviations: PFS, progression-free survival; OS, overall survival; IFN, interferon; MSKCC, Memorial Sloan Kettering Cancer Center risk classification; NR, not reached; IMDC, International
Metastatic Renal Cell Carcinoma Database Consortium classification.

*MSKCC risk groups are defined according to the presence of 0 (favorable risk group), 1–2 (intermediate risk group), or 3 (poor risk group) of the five following risk factors: elevated hemoglobin,
elevated corrected calcium, lactate dehydrogenase . 1.5 N, Karnofsky performance status , 80%, interval from diagnosis to initiation of systemic therapy , 1 year.

**IMDC risk groups are defined according to the presence of 0 (favorable risk group), 1–2 (intermediate risk group), or 3 (poor risk group) of the six following risk factors: elevated hemoglobin,
elevated corrected calcium, elevated neutrophils, elevated platelets, Karnofsky performance status , 80%, interval from diagnosis to initiation of systemic therapy , 1 year.
Tivozanib has recently emerged as another viable thera- HIF Inhibitors

peutic option that improved outcomes over sorafenib in HIF inhibitors are newly developed agents that directly
a refractory setting.79 target HIF upstream of angiogenesis activation.99 In a phase
Cabozantinib more recently joined the field of VEGFR- I clinical trial, the first-in-class HIF-2α inhibitor PT2385 was
directed TKIs in ccRCC, improving outcomes for pre- studied in patients with previously treated clear cell mRCC
treated patients with ccRCC compared with everolimus, with and showed a favorable toxicity profile and ORR of 14%
increases in OS from 16.5 to 21.4 months, in PFS from 3.9 (CR, 2%).100 MK-6482 (PT2977) is a second-generation
to 7.4 months, and in response rate from 3% to 17%.76,77 HIF-2α inhibitor that was studied in a phase I/II clinical trial
Most importantly, cabozantinib demonstrated high activ- in patients with clear cell mRCC and resulted in ORR of 24%
ity in aggressive subsets of tumors, including ccRCC met- and a median PFS of 11.0 months.101 The efficacy of MK-
astatic to the bone.89 Cabozantinib also demonstrated 6482 versus everolimus in patients with previously treated
improved PFS (8.2 vs. 5.6 months) and improved response mRCC is currently being investigated in a phase III clinical
rate (33% vs. 12%) over sunitinib in a phase II trial con- trial (MK-6482-005).102
ducted in untreated patients with intermediate and poor-risk
ccRCC.90,91 In another study, patients with bone dissemi- Targeted Therapies in the Management of Non–clear
nation included in the METEOR trial had a particularly Cell RCC
marked improvement in PFS with cabozantinib, with The activity of targeted therapies in ccRCC led to empirical
a hazard ratio for PFS of 0.33.92 The activity of cabozantinib development of these agents in rarer, non–clear cell sub-
in these subgroups of patients may stem from added in- types. Two randomized phase II trials, ASPEN and ESPN,
hibition of tyrosine kinases AXL and MET, which are involved compared sunitinib and everolimus in this population but
in resistance to sunitinib93 and, for the latter, bone me- failed to clearly demonstrate the superiority of one agent,
tastases onset.94 with response rates that did not exceed 20% and a median
The availability of various multikinase inhibitors has pro- PFS below 9 months in both trials.103,104 Such PFS data were
vided a wide range of therapeutic options for patients with consistent with subgroup analyses of the RECORD-3 trial,98
ccRCC. As demonstrated with cabozantinib, their affinity which evaluated the optimal therapeutic sequence between
toward different tyrosine kinases in addition to VEGFR may everolimus and sunitinib, and the GLOBAL-ARCC trial,
also explain the preserved activity after previous TKI,95 which evaluated temsirolimus versus IFN-α or both.80 More
allowing a potentially durable benefit even in late-stage recent experience with pazopanib and axitinib in small
disease.96 phase II trials, however, reached higher response rates of
28% and 37.5%, respectively,105,106 whereas cabozantinib
A Window of Opportunity for mTOR Inhibitors
provided responses in 27% of patients with non–clear cell
mTOR inhibitors were developed concomitantly to VEGFR- RCC (nccRCC) in a retrospective study.107 Such clinical
directed TKIs in ccRCC. The mTOR pathway has been experience allowed the broader use of VEGFR-directed TKIs
shown to be involved in ccRCC promotion but also promotes and mTOR inhibitors for the treatment of patients with
angiogenesis through cross-talk with the HIF pathway.97 metastatic nccRCC. SWOG 1500 is a randomized phase II
Preclinical models confirmed the antitumor activity of clinical trial that is investigating the efficacy of sunitinib
mTOR inhibitors, leading to clinical development of tem- versus three MET inhibitors (cabozantinib, crizotinib, or
sirolimus and everolimus in advanced ccRCC (Table 3). savolitinib) in patients with papillary mRCC.108
Temsirolimus was evaluated in a predominantly poor-risk
population of patients with untreated mRCC. It was the first Combination of Targeted Therapies
therapy to demonstrate a benefit in OS compared with Based on the efficacy of VEGF and mTOR inhibitors, several
IFN-α as monotherapy, with respective median OS rates of studies have evaluated the potential benefit of combinations
10.9 months and 7.3 months,80 establishing a new standard of these agents. Overall, all of these combinations were more
of care in poor-risk patients in 2007. Class-specific adverse toxic but not more active.81,109 Recently, however, the
events included asthenia, rash, stomatitis, diarrhea, hy- combination of the multikinase inhibitor lenvatinib associ-
perglycemia, and anemia in more than 20% of patients. In ated with the mTOR inhibitor everolimus showed an im-
VEGFR-TKI refractory disease, the oral mTOR inhibitor proved PFS of 14.6 months over 5.5 months with everolimus
everolimus showed improved PFS compared with placebo alone in a phase II clinical trial enrolling patients with
in the RECORD-1 phase III trial at 4.0 versus 1.9 months, metastatic ccRCC.110 Other combinations have proven to
respectively.83 Everolimus has thus been an important provide substantial activity in nccRCC. In a phase II clinical
player in the management of late-stage ccRCC,98 although trial of patients with metastatic nccRCC (encompassing
its role has been recently challenged by the advent of predominantly papillary subtypes), everolimus plus bev-
cabozantinib and immune checkpoint inhibitors. acizumab demonstrated a response rate of 29% and

Dutcher et al
a prolonged PFS of 11.0 months.111 In rare RCC variants of 25% versus 5%, respectively.124 Importantly, the benefit
associated with hereditary leiomyomatosis and renal cell of nivolumab was reported irrespective of prognostic risk
cancer syndrome, characterized by loss of the fumarate groups but also irrespective of PD-L1 expression status.125
hydratase gene, the combination of erlotinib plus bev- By demonstrating benefit in all comers, nivolumab was
acizumab has demonstrated an ORR of more than 50% in approved by the U.S. Food and Drug Administration in
patients with advanced disease.112,113 2015 for patients with advanced RCC after prior anti-
Despite the development of nine targeted agents within angiogenic therapy. This led to further development of PD-
10 years for the treatment of metastatic RCC, there was 1/PD-L1 inhibitors in the first-line setting, with a focus on
a plateau in the efficacy of such approaches in both first-line combination strategies.
and second-line settings.
Combining Immune Checkpoint Inhibitors and
New-Generation Immunotherapies Antiangiogenic Therapies
Standard treatments in RCC evolved from cytokines to Combining treatment modalities has been a staple in on-
targeted molecular therapies in less than a decade, but the cology, aiming at additive or synergistic effects between
potential for immune-based approaches had not yet been compounds to improve antitumor efficacy. In the case of
fulfilled in a disease that had long been described as im- immune checkpoints, alterations of PD-1 and CTLA-4 ex-
munogenic. Indeed, ccRCC harbors a high immune in- pression in mouse models have shown distinct effects on
filtration involving T-effector lymphocytes,114,115 although immunity, suggesting that their concurrent inhibition may
frequent expression of immune checkpoints may be re- be synergistic in human cancers.126 It is indeed suspected
sponsible for immune tolerance in both clear cell and that nivolumab acts within the tumor to promote the effector
non–clear cell subtypes.116,117 Immune checkpoints are phase, whereas ipilimumab may affect T-cell priming in
surface receptors expressed on immune and tumor cells lymph nodes.127 The proof-of-concept of dual PD-1 and
that may induce T-cell anergy, such as PD-1, its ligand PD- CTLA-4 inhibition by nivolumab and ipilimumab has been
L1, and CTLA-4. Their discovery and subsequent inhibition demonstrated in melanoma,128 in which survival was im-
with monoclonal antibodies ushered a new treatment era for proved with the combination compared with single-agent
metastatic RCC. immunotherapy in a phase III clinical trial.129 Considering
The first immune checkpoint inhibitor (CPI) evaluated in the immunogenicity and activity of single-agent immuno-
ccRCC was the CTLA-4 inhibitor ipilimumab. Among 40 therapy in ccRCC, development of nivolumab and ipili-
patients included in a phase II clinical trial, ORR was 12.5%, mumab in this population followed suit.
including among patients who were previously treated with Compelling combinations encompass not only associations
IL-2, and some patients experienced responses lasting more of CPIs but also associations of CPIs with antiangiogenic
than a year.118 Subsequent CPI development in monotherapies. Indeed, both have individually demonstrated
therapy for ccRCC (in parallel to its development in many potent antitumor activity, and a growing body of work in-
tumors) focused on the inhibition of the PD-1/PD-L1 axis. dicates a strong interplay between angiogenesis and im-
The anti–PD-1 agent nivolumab has been evaluated in munity. The tumor endothelium may indeed prevent
early-phase trials at different dose levels and has demon- lymphocyte infiltration through impaired leukocyte adhesion
strated substantial activity, with ORR of more than 20% in to the endothelium130 but also through expression of surface
a setting of refractory disease.119-121 Nivolumab has been receptors inducing effector T-cell death.131 Hypoxic con-
particularly well tolerated, with most adverse events conditions promoted by an abnormal vasculature may also favor
sisting in endocrinopathies, such as hypothyroidism, and the onset of an immunosuppressive microenvironment,132
less than 20% of grade 3/4 toxicities, which may, how- whereas VEGF may directly regulate the expression of im-
ever, encompass severe immune-related adverse events, mune checkpoints on immune cells.133 New combinations
including pneumonitis. Its anti–PD-1 counterpart pem- of CPIs with VEGF or VEGFR-targeted agents have thus
brolizumab similarly demonstrated encouraging activity, this been developed, including pembrolizumab plus axitinib,
time in the first-line setting, with ORR up to 34% in 110 anti–PD-L1 avelumab plus axitinib, and anti–PD-L1 ate-
patients with ccRCC and 26% in 165 patients with zolizumab plus bevacizumab in patients with advanced
nccRCC.122,123 ccRCC.
Nivolumab was the first CPI to enter the standard of care in
advanced ccRCC following the results of the CheckMate- SEQUENCING AND COMBINING TARGETED THERAPY
025 phase III clinical trial. In patients whose disease AND IMMUNOTHERAPY
progressed after one or two VEGFR-directed therapies, The combination of immunotherapy and targeted therapy
nivolumab demonstrated improved outcomes against can achieve higher response rates and better outcomes via
everolimus, with OS of 25.0 versus 19.6 months and ORR additive or synergistic mechanisms. Therefore, various

combinations of immunotherapy and targeted therapies activity. The combination of HD IL-2 and pembrolizumab in
have been studied in mRCC, which are discussed below. 18 patients with mRCC was safe and resulted in an ORR of
Combination of Anti–PD-1 and Anti–CTLA-4 Therapy 69%.147 Bempegaldesleukin (NKTR-214), a pegylated form
of IL-2, was evaluated with nivolumab in 24 patients with
CheckMate-214 was a phase III clinical trial that randomly previously untreated mRCC and resulted in an ORR of 54%
assigned patients with previously untreated metastatic and had acceptable safety.148
ccRCC to receive ipilimumab plus nivolumab versus suni-
tinib. The coprimary endpoints of the study were ORR, PFS, Combination of Immunotherapy and Radiotherapy
and OS, evaluated in patients with intermediate- to poor-risk One of the mechanisms that malignant cells use to evade
disease. The ORR, OS, and quality of life were significantly immune recognition is to reduce the expression of tumor
better in the combination arm compared with sunitinib antigens, which results in resistance to immunotherapy.149,150
(Table 4).134,135 Importantly, more than 10% of those pa- RCC was long considered resistant to standard fractionated
tients achieved CR. Immune-related toxicity must be radiation therapy; however, using the stereotactic body ra-
carefully managed with this regimen because 22% of pa- diation therapy (SBRT) approach to a metastatic lesion may
tients in the trial discontinued therapy after severe adverse lead to tumor necrosis and increased release of tumor an-
events. Of note, treatment-free survival in patients treated in tigens into the blood, which subsequently enhances the
the combination arm was significantly longer than that with effect of the immunotherapy on the metastatic disease be-
sunitinib.136,137 Even patients who discontinued the im- yond the irradiated field.150 In a single-arm phase II clinical
munotherapy due to toxicity continued to benefit from im- trial, patients with mRCC were treated with HD IL-2 and
munotherapy, with outcomes comparable with those in the stereotactic ablative radiotherapy to multiple metastatic sites,
intention-to-treat (ITT) population.138 which resulted in an ORR of 40%.151 However, two recent
Combination of Anti–PD-1/PD-L1 and Anti-VEGF Therapy clinical trials of immunotherapy plus radiation therapy (NIVES
and RADVAX) did not show a significant improvement in
IMmotion 151 was a phase III clinical trial that randomized
response to immunotherapy or a response in nonirradiated
patients with previously untreated mRCC (with a component
lesions.152,153 NIVES is a phase II clinical trial of nivolumab
of clear cell histology or sarcomatoid differentiation) to re-
plus SBRT in patients previously treated for mRCC (up to two
ceive atezolizumab plus bevacizumab versus sunitinib. The
anti-VEGF therapies) in whom treatment resulted in ORR of
primary endpoints were PFS in patients with PD-L1 ex-
17% (27% in irradiated lesions), median PFS 4.1, and OS
pression 1% on immune cells and OS in the ITT pop-
22.0 months. This study did not meet its primary endpoint,
ulation. The median PFS in patients with a PD-L1–positive
which was ORR of 40%.152 In the RADVAX clinical trial, pa-
tumor (as defined in the study) was significantly better in the
tients with clear cell mRCC were treated with ipilimumab plus
combination arm; however, there was no statistically sig-
nivolumab and SBRT, which resulted in an ORR of 56%.153
nificant difference in OS between the treatment arms in the
ITT population (Table 4).143 Combination of Immunotherapy and HIF Inhibitors
Sharing a similar design, JAVELIN Renal 101 was a phase III The combination of nivolumab plus PT2385 was studied in
clinical trial that enrolled patients with previously untreated a phase I expansion cohort of 50 patients with mRCC who
mRCC to receive avelumab plus axitinib versus sunitnib. had received up to three prior therapies. The combination
The primary endpoints were PFS and OS in patients with was well tolerated, and the ORR was 22%. The median PFS
PD-L1 expression 1% on immune cells. Although median among patients with therapeutic exposure to PT2385 was
PFS was significantly better in the combination arm, there 10.0 months.154
was no statistically significant difference in OS between the Several clinical trials are currently investigating the clinical
treatment arms (Table 4).144,145 outcomes of various combinations of immunotherapy with
More promising results came from the KEYNOTE-426 phase novel and targeted therapies (Table 5).
III clinical trial, which randomly assigned patients with pre- Immunotherapy-Based Combinations in Patients With
viously untreated mRCC to receive axitinib plus pem- Metastatic nccRCC
brolizumab versus sunitinib. The primary endpoints were OS
and PFS in the ITT population. The ORR, PFS, and OS were Data on the efficacy of the immunotherapy-based combi-
significantly better in the combination arm compared with nation therapy in patients with metastatic nccRCC are
sunitinib (Table 4), and the clinical benefit was observed limited to relatively small phase II clinical trials and ob-
regardless of the IMDSC risk group or PD-L1 expression.139,146 servational studies. In three retrospective studies, the ORR
of nivolumab combined with ipilimumab or anti-VEGF
Combination of Anti–PD-1 and IL-2 therapy was 31% to 44%.172-174 The efficacy of the com-
Additional strategies in development include CPIs combined bination of atezolizumab and bevacizumab in patients with
with cytokines to jointly activate the immune antitumor nccRCC, including 17% with sarcomatoid differentiation,

Dutcher et al
was studied in a single-arm phase II clinical trial, yielding an (Table 4).134-139,143-146,181,182 HD IL-2 can be considered in
ORR of 26%. Slight differences in ORR were reported select patients.182 In patients with newly diagnosed clear cell
between the histologic subtypes: 25% in papillary RCC, mRCC who have a contraindication to immunotherapy,
10% in chromophobe RCC, 29% in unclassified RCC, and treatment with anti-VEGF monotherapy (sunitinib, pazo-
30% in other subtypes (translocation, collecting duct, and panib, or cabozantinib) is recommended.182
medullary RCC); these results highlight the need to in-
vestigate these combinations in larger cohorts of these rare Sarcomatoid differentiation Approximately 10% to 20% of
nccRCC subtypes.175 patients with mRCC (both clear cell and non–clear cell
histologies) have sarcomatoid features, which are associ-
How to Treat Patients With mRCC at the Time of ated with high PD-L1 expression and immune cell
Initial Diagnosis? infiltration183 and poor response to anti-VEGF therapies.
Local therapy Patients with mRCC who have limited met- Patients with mRCC with sarcomatoid features had signif-
astatic disease should be considered for local therapies icantly better outcomes with immunotherapy-based com-
such as SBRT or metastasectomy, which can result in long- binations compared with sunitinib (Table 4).139-142
term disease-free status and improvement in survival
PD-L1 expression Treatment with CPIs in the frontline
outcomes.176-178 Moreover, postponing the initiation of
setting have shown higher efficacy in patients with PD-
systemic therapy and associated toxicities improves the
L1–positive RCC tumors. However, even in patients with
quality of life of these patients.
PD-L1–negative tumors, there was a substantial clinical
In a meta-analysis of eight cohort studies, the range of benefit with CPIs compared with sunitinib.134,135,143-146
median OS was 36.5–142 months among patients who Considering the inconclusive data on the predictive and
underwent complete metastasectomy and 8.4–27 months prognostic value of PD-L1 expression in RCC, its evalua-
among those with incomplete metastasectomy, and the tion is currently not recommended in routine clinical
pooled adjusted hazard ratio for all-cause mortality was 2.37 practice because it does not affect the treatment
(p , .001).178 In a prospective study of 77 patients with decision.124,134,135,143-146
mRCC who underwent local therapy, CR was achieved in 22
patients (29%), and the median duration of CR was 16.7 Non–clear cell histology Options for treatment of patients
months.179 with newly diagnosed metastatic nccRCC include CPIs,
anti-VEGF therapies, and mTOR inhibitors. Currently,
Observation A subset of patients with mRCC have indolent
there is no head-to-head comparison of CPIs and tar-
disease and limited metastatic burden. Initiation of systemic
geted therapies in patients with nccRCC. Moreover, the
treatment can be postponed in this group of patients to
reported response to immunotherapy or targeted ther-
avoid the treatment-related toxicities. In a phase II clinical
apy in patients with nccRCC varies by the histologic
trial, 52 patients with newly diagnosed asymptomatic mRCC
subtype of the tumor. In view of the efficacy and ac-
were enrolled, and their disease was initially managed with
ceptable toxicity of CPIs in patients with nccRCC,
observation, including radiologic assessments of their dis-
immunotherapy-based regimens could be considered,
ease status. The median duration of observation was 14.9
particularly in patients with papillary or unclassified
months, and the median OS was 44.5 months. A higher
subtypes. 172-175,181
IMDC score and a greater number of organs with metastases
were associated with early need for systemic treatment. A Clinical trials All patients with mRCC, and especially those
subset of patients, notably those with asymptomatic, fa- with non–clear cell histology, should be considered for
vorable risk mRCC with low disease burden and often with enrollment into clinical trials when possible. A list of select
a long time from initial diagnosis to development of me- ongoing clinical trials of combination therapy in patients with
tastases, may be considered for initial observation.180 mRCC is provided in Table 5.
Systemic Treatment of Patients With mRCC at the Time of
Systemic Treatment of Patients With mRCC at the Time of
Initial Diagnosis
Disease Progression
The majority of patients with mRCC need systemic treatment
The treatment options for mRCC in the second line depend
at the time of diagnosis.
on the treatment regimen used in the frontline setting. In
Clear cell histology The preferred treatment options for patients who were initially treated with VEGFR-directed TKI
mRCC with clear cell histology in the frontline setting are the monotherapy, the approved treatment options after disease
combinations of ipilimumab plus nivolumab in patients with progression include nivolumab or cabozantinib, both of
intermediate and poor risk characteristics (CheckMate-214) which demonstrated OS benefit in the second-line set-
or axitinib plus pembrolizumab (KEYNOTE-426) in all ting.182 Other options to be considered include ipilimumab
comers, based on the OS benefit compared with sunitinib plus nivolumab,184 axitinib, and lenvatinib plus everolimus.

In patients who were initially treated with the combination of SUMMARY AND FUTURE DIRECTIONS
ipilimumab plus nivolumab, the treatment options after disease In the past 15 years, the treatments available for mRCC have
progression include axitinib,185 cabozantinib,76,77 sunitinib,186 expanded dramatically. Ongoing research has identified
pazopanib,186 lenvatinib plus everolimus,110 and HD IL-2.61 drivers of RCC that have led to numerous targeted agents,
In patients who were initially treated with the combination of with differing nuances that affect their activity and com-
immunotherapy and VEGFR-targeted therapy, treatment binability. Improvements in the comprehension of RCC
options in the second line include cabozantinib, lenvatinib biology, and its various subtypes, should further drive
plus everolimus, everolimus, and HD IL-2. clinical research into the development of new agents. Most
TABLE 4. Clinical Outcomes With the Combination of Immunotherapy and VEGF-Targeted Therapy in Patients With Clear Cell mRCC in the Frontline Setting
Median
Follow-up Favorable Risk Sarcomatoid
Clinical Trial Treatment Arms (months) ITT Population Intermediate-Poor Risk Group Group Histology139-142
CheckMate- Ipilimumab + 32.4 ORR: 41% vs. ORR: 42% vs. 29% ORR: 39% vs. ORR: 57% vs. 19%
214134,135 nivolumab vs. 34% 50%
sunitinib
CR: 11% vs. 2% CR: 11% vs. 1% CR: 8% vs. 4% CR: 18% vs. 0%
PFS: 9.7 vs. 9.7; PFS: 8.2 vs. 8.3; HR, 0.77 PFS: 13.9 vs. 19. PFS: 8.4 vs. 4.9;
HR, 0.85 (0.65–0.90) 9; HR, 1.23 HR, 0.61
(0.73–0.98)* (0.90–1.69) (0.38–0.97)
OS: NR vs. 37.9; OS: NR vs. 26.6; HR, 0.66 OS: NR in either OS: 31.2 vs. 13.6;
HR, 0.71 (0.54–0.80) arm; HR, 1.22 HR, 0.55 (0.33–
(0.59–0.86) (0.73–2.04) 0.90)
IMmotion151143 Atezolizumab + ORR: 37% vs. PFS in the intermediate risk PFS: HR, 0.71 ORR: 49% vs. 14%
bevacizumab vs. 33% group: HR, 0.81 (0.59–1.13); (0.39–1.29)
sunitinib PFS in the poor risk group: HR,
CR: 5% vs. 2% CR: 10% vs. 3%
0.65 (0.36–1.15)
PFS: 15 PFS: 11.2 vs. 8.4; PFS: 8.3 vs. 5.3;
HR, 0.83 HR, 0.56
(0.70–0.97) (0.38–0.83)
OS: 24 OS: 33.6 vs. 34.9; OS: 21.7 vs. 15.4;
HR, 0.93 HR, 0.64 (0.41–
(0.76–1.14) 1.01)
JAVELIN Renal Avelumab + axitinib 10.7–11.6 ORR: 51% vs. ORR: 51% vs. 25% in the ORR: 68% vs. ORR: 47% vs. 21%
101144,145 vs. sunitinib 26% intermediate risk group; 31% 38%
vs. 11% in poor risk group
CR: 3% vs. 2% CR: 4.3% vs. 0%
PFS: 13.8 vs. 8.4; PFS in the intermediate risk PFS: NR vs. 13.8; PFS: 7.0 vs. 4.0;
HR, 0.69; group: 13.8 vs. 8.4; HR, 0.74 HR, 0.54 HR, 0.57
(0.56–0.84) (0.57–0.95) (0.32–0.90) (0.32–1.00)
OS-12**: 86% vs. PFS in the poor risk group: OS: 83% vs. 67%
83%; HR, 0.78 6.0 vs. 2.9; HR, 0.57 (0.38–
(0.55–1.08) 0.88)
KEYNOTE- Pembrolizumab + 12.8 ORR: 59% vs. ORR: 56% vs. 30% ORR: 67% vs. ORR: 59% vs. 32%
426139,146 axitinib vs. sunitinib 36% 50%
CR: 6% vs. 2% CR: 5% vs. 1% CR: 6% vs. 2% CR: 12% vs. 0%
PFS: 15.1 vs. PFS: 12.6 vs. 8.2; HR, 0.67 PFS: 17.7 vs. PFS: NR vs. 8.4;
11.1; HR, 0.69 (0.53–0.85) 12.7; HR, 0.81 HR, 0.54
(0.57–0.84) (0.53–1.24) (0.29–1.00)
OS-12: 90% vs. OS-12: 87% vs. 71%; HR, 0.52 OS-12: 95% vs. OS: NR in either
78%; HR, 0.53 (0.37–0.74) 94%; HR, arm; HR, 0.58
(0.38–0.74) 0.64 (0.24– (0.21–1.59)
1.68)
Abbreviations: mRCC, metastatic renal cell carcinoma; ITT, intention-to-treat; ORR, overall response rate; CR, complete response; PFS, progression-free
survival; HR, hazard ratio; OS, overall survival; NR, not reached.
*Values in parentheses are 95% CI.
**Overall survival at 12 months.

Dutcher et al
recently, refined understanding of interactions of tumors has improved the outcomes and has shown high response
and the immune system has led to new immunotherapeutic rates even in later lines of therapy for advanced disease.
approaches to treat patients with mRCC.
Although cytokines and targeted molecular therapies did
Cytokines demonstrated the initial clinical activity of therapy not improve survival in the localized disease setting as either
against mRCC, and, more importantly, HD IL-2 treatment adjuvant or neoadjuvant therapy, these settings remain
has produced durable disease-free and treatment-free areas of unmet need. Several of the newer immunologic
survival of up to years and decades in a small proportion agents are being evaluated in phase III trials in the neo-
of patients. The development of VEGFR-directed therapies adjuvant and adjuvant setting.
TABLE 5. Select Ongoing Clinical Trials of Combination Therapy in Patients With mRCC
Study Primary
Trial Phase RCC Subtype Experimental Arms Control Arm Line of Therapy Endpoints
NCT02811861 (CLEAR)155 III Clear cell 1. Lenvatinib + everolimus; 2. Sunitinib First PFS
lenvatinib + pembrolizumab
NCT03141177 III Clear cell Nivolumab + cabozantinib Sunitinib First PFS
(CheckMate 9ER)156
NCT03937219 (COSMIC- III Clear cell Ipilimumab + nivolumab + Ipilimumab + nivolumab + First PFS
313)157 cabozantinib placebo
NCT03793166 III Clear cell Ipilimumab + nivolumab followed Ipilimumab + nivolumab First OS
(PDIGREE)158 by nivolumab + cabozantinib followed by nivolumab
NCT03729245 (PIVOT- III Clear cell Bempegaldesleukin + nivolumab Investigator’s choice of First ORR and OS
09)159 VEGF-TKI (sunitinib/
cabozantinib)
NCT03592472 III Clear cell Abexinostat + pazopanib Pazopanib + placebo First and second (up to PFS
(RENAVIV)160 one prior
immunotherapy)
NCT03501381161 II Clear cell High-dose IL-2 + entinostat High-dose IL-2 First, second, and third ( PFS
1 prior CPI)
NCT02781506162 II Clear cell Nivolumab + SAbR N/A Second and higher ORR
NCT03474497163 I/II Clear cell Pembrolizumab + SBRT + N/A Second (one prior CPI) Abscopal
intralesional IL-2 ORR
NCT04090710 II Clear cell, Ipilimumab + nivolumab + SBRT Ipilimumab + nivolumab First PFS
(CYTOSHRINK)164 non–clear to the kidney mass
cell
NCT02293980165 I Clear cell Part 3: PT2385 + cabozantinib N/A Second and higher ( 1 MTD
VEGF-TKI)
NCT03634540166 II Clear cell MK-6482 + cabozantinib N/A First and higher ORR
NCT03075423 II Non–clear cell Ipilimumab + nivolumab Sunitinib First OS at
(SUNIFORECAST)167 12 months
NCT03635892168 II Non–clear cell Nivolumab + cabozantinib N/A First and second (no prior ORR
CPI)
NCT03170960169 I/II Clear cell, Atezolizumab + cabozantinib N/A Clear cell: first; non–clear MTD and
non–clear cell: first and second ORR
cell
NCT03177239170 II Non–clear cell Nivolumab followed by N/A First and above ORR
(ANZUP1602; UNISoN) ipilimumab + nivolumab
NCT02982954 IV Clear cell, Ipilimumab + nivolumab N/A First IMAE
(CheckMate-920)171 non–clear
cell
Abbreviations: mRCC, metastatic renal cell carcinoma; RCC, renal cell carcinoma; PFS, progression-free survival; OS, overall survival; TKI, tyrosine kinase
inhibitor; ORR, overall response rate; IL, interleukin; CPI, checkpoint inhibitor; SAbR, stereotactic ablative radiation; SBRT, stereotactic body radiation
therapy; N/A, not applicable; MTD, maximum tolerated dose; IMAE, incidence of high grade immune-mediated adverse events

The long-term nature of benefit shown with HD IL-2 and the Many ongoing trials are evaluating multiple novel agents,
numbers of patients benefiting from approaches such as therapeutic approaches, combinations, and sequences in
HD IL-2 and VEGFR-TKI highlight the need to expand our treatment of mRCC (Tables 4 and 5). Future studies are
ongoing research into combinations of immunotherapies needed to identify the predictive biomarkers of response to
and targeted therapies in a multitude of variations. As of treatment and treatment-related toxicity to choose the most
today, combinations of CPIs with VEGFR-targeted therapies effective and least toxic treatment option for each patient with
have shown improvement in response and survival out- mRCC. Additional research will attempt to implement these
comes compared with single-agent targeted therapy in
approaches in the earlier stages of RCC. It will take time and
patients with treatment-naive mRCC. Several clinical trials
effort to sort out the most fitting approach, which may have to
are ongoing to investigate triplet therapy (anti–PD-1,
anti–CTLA-4, anti-VEGF) in the frontline setting or the be further refined according to the different subtypes of RCC
combinations of CPIs with other modalities, such as IL-2, and different sequences of therapy. Nevertheless, the ex-
radiation therapy, and HIF inhibitors. New CPIs are also pansion of available options has brought hope to our patients
undergoing evaluation. and encouragement to their oncologists.

1
Cancer Research Foundation of New York, Chappaqua, NY Janice P. Dutcher, MD, 750 Kappock St., #511, Riverdale, NY 10463;
2
Deptartment of Cancer Medicine, Institute Gustave Roussy, Villejuif, email: [email protected].
France
3
Taussig Cancer Institute, Cleveland Clinic Foundation, Cleveland, OH
EDBK_280817.
REFERENCES
1. Ho M. Identification and “induction” of interferon. Bacteriol Rev. 1964;28:367-381.
2. Paucker K, Cantell K, Henle W. Quantitative studies on viral interference in suspended L cells: III. Effect of interfering viruses and interferon on the growth rate of
cells. Virology. 1962;17:324-334.
3. Gutterman JU, Fine S, Quesada J, et al. Recombinant leukocyte A interferon: pharmacokinetics, single-dose tolerance, and biologic effects in cancer patients.
Ann Intern Med. 1982;96:549-556.
4. Lotze MT, Grimm EA, Mazumder A, et al. Lysis of fresh and cultured autologous tumor by human lymphocytes cultured in T-cell growth factor. Cancer Res.
1981;41:4420-4425.
5. Grimm EA, Mazumder A, Zhang HZ, et al. Lymphokine-activated killer cell phenomenon: lysis of natural killer-resistant fresh solid tumor cells by interleukin 2-
activated autologous human peripheral blood lymphocytes. J Exp Med. 1982;155:1823-1841.
6. Gillis S. Interleukin 2: biology and biochemistry. J Clin Immunol. 1983;3:1-13.

7. Mulé JJ, Yang JC, Afreniere RL, et al. Identification of cellular mechanisms operational in vivo during the regression of established pulmonary metastases by the
systemic administration of high-dose recombinant interleukin 2. J Immunol. 1987;139:285-294.
8. Ettinghausen SE, Rosenberg SA. Immunotherapy of murine sarcomas using lymphokine activated killer cells: optimization of the schedule and route of
administration of recombinant interleukin-2. Cancer Res. 1986;46:2784-2792.
9. Puri RK, Travis WD, Rosenberg SA. In vivo administration of interferon alpha and interleukin 2 induces proliferation of lymphoid cells in the organs of mice.
Cancer Res. 1990;50:5543-5550.
10. Goeddel DV, Yelverton E, Ullrich A, et al. Human leukocyte interferon produced by E. coli is biologically active. Nature. 1980;287:411-416.
11. Devos R, Plaetinck G, Cheroutre H, et al. Molecular cloning of human interleukin 2 cDNA and its expression in E. coli. Nucleic Acids Res. 1983;11:4307-4323.
12. Finke JH, Rayman P, Hart L, et al. Characterization of tumor-infiltrating lymphocyte subsets from human renal cell carcinoma: specific reactivity defined by
cytotoxicity, interferon-gamma secretion, and proliferation. J Immunother Emphasis Tumor Immunol. 1994;15:91-104.
13. deKernion JB. Immunotherapy of renal cell carcinoma. Prog Clin Biol Res. 1984;153:409-421.
14. Neidhart JA. Interferon therapy for the treatment of renal cancer. Cancer. 1986;57 (8, suppl):1696-1699.
15. Quesada JR, Rios A, Swanson D, et al. Antitumor activity of recombinant-derived interferon alpha in metastatic renal cell carcinoma. J Clin Oncol. 1985;
3:1522-1528.
16. Krown SE. Interferon treatment of renal cell carcinoma: current status and future prospects. Cancer. 1987;59 (3, suppl):647-651.

Dutcher et al
17. Trump DL, Elson PJ, Borden EC, et al. High-dose lymphoblastoid interferon in advanced renal cell carcinoma: an Eastern Cooperative Oncology Group Study.
Cancer Treat Rep. 1987;71:165-169.
18. Kurzrock R, Rosenblum MG, Quesada JR, et al. Phase I study of a combination of recombinant interferon-alpha and recombinant interferon-gamma in cancer
patients. J Clin Oncol. 1986;4:1677-1683.
19. Aapro MS, Alberts DS, Salmon SE. Interactions of human leukocyte interferon with vinca alkaloids and other chemotherapeutic agents against human tumors in
clonogenic assay. Cancer Chemother Pharmacol. 1983;10:161-166.
20. Cetto GL, Franceschi T, Turrina G, et al. Recombinant alpha-interferon and vinblastine in metastatic renal cell carcinoma: efficacy of low doses. Semin Surg
Oncol. 1988;4:184-190.
21. Creagan ET, Kovach JS, Long HJ, et al. Phase I study of recombinant leukocyte A human interferon combined with BCNU in selected patients with advanced
22. Rosenberg SA, Lotze MT, Muul LM, et al. A progress report on the treatment of 157 patients with advanced cancer using lymphokine-activated killer cells and
interleukin-2 or high-dose interleukin-2 alone. N Engl J Med. 1987;316:889-897.
23. West WH, Tauer KW, Yannelli JR, et al. Constant-infusion recombinant interleukin-2 in adoptive immunotherapy of advanced cancer. N Engl J Med. 1987;
316:898-905.
24. Fisher RI, Coltman CA Jr., Doroshow JH, et al. Metastatic renal cancer treated with interleukin-2 and lymphokine-activated killer cells. A phase II clinical trial.
Ann Intern Med. 1988;108:518-523.
25. Parkinson DR, Fisher RI, Rayner AA, et al. Therapy of renal cell carcinoma with interleukin-2 and lymphokine-activated killer cells: phase II experience with
a hybrid bolus and continuous infusion interleukin-2 regimen. J Clin Oncol. 1990;8:1630-1636.
26. Négrier S, Mercatello A, Bret M, et al. Intensive regimen of cytokines with interleukin-2 and interferon alfa-2b in selected patients with metastatic renal
carcinoma. J Immunother Emphasis Tumor Immunol. 1995;17:62-68.
27. Negrier S, Philip T, Stoter G, et al. Interleukin-2 with or without LAK cells in metastatic renal cell carcinoma: a report of a European multicentre study. Eur
J Cancer Clin Oncol. 1989;25 (suppl 3):S21-S28.
28. Rosenberg SA, Lotze MT, Yang JC, et al. Prospective randomized trial of high-dose interleukin-2 alone or in conjunction with lymphokine-activated killer cells for
the treatment of patients with advanced cancer. J Natl Cancer Inst. 1993;85:622-632.
29. Weiss GR, Margolin KA, Aronson FR, et al. A randomized phase II trial of continuous infusion interleukin-2 or bolus injection interleukin-2 plus lymphokine-
activated killer cells for advanced renal cell carcinoma. J Clin Oncol. 1992;10:275-281.
30. Sznol M, Mier JW, Sparano J, et al. A phase I study of high-dose interleukin-2 in combination with interferon-alpha 2b. J Biol Response Mod. 1990;9:529-537.
31. Negrier S, Escudier B, Lasset C, et al. Groupe Français d’Immunothérapie. Recombinant human interleukin-2, recombinant human interferon alfa-2a, or both in
metastatic renal-cell carcinoma. N Engl J Med. 1998;338:1272-1278.
32. Sleijfer DT, Janssen RA, Buter J, et al. Phase II study of subcutaneous interleukin-2 in unselected patients with advanced renal cell cancer on an outpatient
basis. J Clin Oncol. 1992;10:1119-1123.
33. Bukowski RM, Olencki T, Wang Q, et al. Phase II trial of interleukin-2 and interferon-alpha in patients with renal cell carcinoma: clinical results and immunologic
correlates of response. J Immunother. 1997;20:301-311.
34. Negrier S, Perol D, Ravaud A, et al; French Immunotherapy Intergroup. Medroxyprogesterone, interferon alfa-2a, interleukin 2, or combination of both cytokines
in patients with metastatic renal carcinoma of intermediate prognosis: results of a randomized controlled trial. Cancer. 2007;110:2468-2477.
35. Yang JC, Sherry RM, Steinberg SM, et al. Randomized study of high-dose and low-dose interleukin-2 in patients with metastatic renal cancer. J Clin Oncol. 2003;
21:3127-3132.
36. McDermott DF, Regan MM, Clark JI, et al. Randomized phase III trial of high-dose interleukin-2 versus subcutaneous interleukin-2 and interferon in patients
with metastatic renal cell carcinoma. J Clin Oncol. 2005;23:133-141.
37. Elson PJ, Witte RS, Trump DL. Prognostic factors for survival in patients with recurrent or metastatic renal cell carcinoma. Cancer Res. 1988;48:7310-7313.
38. Motzer RJ, Mazumdar M, Bacik J, et al. Survival and prognostic stratification of 670 patients with advanced renal cell carcinoma. J Clin Oncol. 1999;
17:2530-2540.
39. Negrier S, Gomez F, Douillard JY, et al; Groupe Français d’Immunothérapie. Prognostic factors of response or failure of treatment in patients with metastatic
renal carcinomas treated by cytokines: a report from the Groupe Français d’Immunothérapie. World J Urol. 2005;23:161-165.
40. Manola J, Royston P, Elson P, et al; International Kidney Cancer Working Group. Prognostic model for survival in patients with metastatic renal cell carcinoma:
results from the international kidney cancer working group. Clin Cancer Res. 2011;17:5443-5450.
41. Heng DY, Xie W, Regan MM, et al. Prognostic factors for overall survival in patients with metastatic renal cell carcinoma treated with vascular endothelial growth
factor-targeted agents: results from a large, multicenter study. J Clin Oncol. 2009;27:5794-5799.
42. Ko JJ, Xie W, Kroeger N, et al. The International Metastatic Renal Cell Carcinoma Database Consortium model as a prognostic tool in patients with metastatic
renal cell carcinoma previously treated with first-line targeted therapy: a population-based study. Lancet Oncol. 2015;16:293-300.
43. Yip SM, Wells C, Moreira R, et al. Checkpoint inhibitors in patients with metastatic renal cell carcinoma: results from the International Metastatic Renal Cell
Carcinoma Database Consortium. Cancer. 2018;124:3677-3683.
44. Motzer RJ, Bacik J, Schwartz LH, et al. Prognostic factors for survival in previously treated patients with metastatic renal cell carcinoma. J Clin Oncol. 2004;
22:454-463.

45. Schwartzentruber DJ. Guidelines for the safe administration of high-dose interleukin-2. J Immunother. 2001;24:287-293.
46. Dutcher JP, Schwartzentruber DJ, Kaufman HL, et al. High dose interleukin-2 (Aldesleukin): expert consensus on best management practices-2014.
J Immunother Cancer. 2014;2:26.
47. Fyfe G, Fisher RI, Rosenberg SA, et al. Results of treatment of 255 patients with metastatic renal cell carcinoma who received high-dose recombinant
interleukin-2 therapy. J Clin Oncol. 1995;13:688-696.
48. Fyfe GA, Fisher RI, Rosenberg SA, et al. Long-term response data for 255 patients with metastatic renal cell carcinoma treated with high-dose recombinant
interleukin-2 therapy. J Clin Oncol. 1996;14:2410-2411.
49. Klapper JA, Downey SG, Smith FO, et al. High-dose interleukin-2 for the treatment of metastatic renal cell carcinoma: a retrospective analysis of response and
survival in patients treated in the surgery branch at the National Cancer Institute between 1986 and 2006. Cancer. 2008;113:293-301.
50. Shablak A, Sikand K, Shanks JH, et al. High-dose interleukin-2 can produce a high rate of response and durable remissions in appropriately selected patients
with metastatic renal cancer. J Immunother. 2011;34:107-112.
51. McDermott DF, Cheng S-C, Signoretti S, et al. The high-dose aldesleukin “select” trial: a trial to prospectively validate predictive models of response to treatment
in patients with metastatic renal cell carcinoma. Clin Cancer Res. 2015;21:561-568.
52. Payne R, Glenn L, Hoen H, et al. Durable responses and reversible toxicity of high-dose interleukin-2 treatment of melanoma and renal cancer in a Community
Hospital Biotherapy Program. J Immunother Cancer. 2014;2:13.
53. Hanzly M, Aboumohamed A, Yarlagadda N, et al. High-dose interleukin-2 therapy for metastatic renal cell carcinoma: a contemporary experience. Urology.
2014;83:1129-1134.
54. Chow S, Galvis V, Pillai M, et al. High-dose interleukin2: a 10-year single-site experience in the treatment of metastatic renal cell carcinoma: careful selection of
patients gives an excellent outcome. J Immunother Cancer. 2016;4:67.
55. Alva A, Daniels GA, Wong MKK, et al. Contemporary experience with high-dose interleukin-2 therapy and impact on survival in patients with metastatic
melanoma and metastatic renal cell carcinoma. Cancer Immunol Immunother. 2016;65:1533-1544.
56. Curti B, Daniels GA, McDermott DF, et al. Improved survival and tumor control with Interleukin-2 is associated with the development of immune-related adverse
events: data from the PROCLAIMSM registry. J Immunother Cancer. 2017;5:102.
57. Fishman M, Dutcher JP, Clark JI, et al. Overall survival by clinical risk category for high dose interleukin-2 (HD IL-2) treated patients with metastatic renal cell
cancer (mRCC): data from the PROCLAIMSM registry. J Immunother Cancer. 2019;7:84.
58. Motzer RJ, Bacik J, Mariani T, et al. Treatment outcome and survival associated with metastatic renal cell carcinoma of non-clear-cell histology. J Clin Oncol.
2002;20:2376-2381.
59. Upton MP, Parker RA, Youmans A, et al. Histologic predictors of renal cell carcinoma response to interleukin-2-based therapy. J Immunother. 2005;
28:488-495.
60. Lam ET, Wong MKK, Agarwal N, et al. Retrospective analysis of the safety and efficacy of high-dose interleukin-2 after prior tyrosine kinase inhibitor therapy in
patients with advanced renal cell carcinoma. J Immunother. 2014;37:360-365.
61. Buchbinder EI, Dutcher JP, Daniels GA, et al. Therapy with high-dose Interleukin-2 (HD IL-2) in metastatic melanoma and renal cell carcinoma following PD1 or
PDL1 inhibition. J Immunother Cancer. 2019;7:49.
62. Clark JI, Curti B, Davis E, et al. Long-term disease-free survival of patients with metastatic melanoma or renal cell cancer following high-dose interleukin-2.
Poster presented at: SITC 2017; November 2017; National Harbor, MD.
63. Gnarra JR, Tory K, Weng Y, et al. Mutations of the VHL tumour suppressor gene in renal carcinoma. Nat Genet. 1994;7:85-90.
64. Cancer Genome Atlas Research Network. Comprehensive molecular characterization of clear cell renal cell carcinoma. Nature. 2013;499:43-49.
65. Banks RE, Tirukonda P, Taylor C, et al. Genetic and epigenetic analysis of von Hippel-Lindau (VHL) gene alterations and relationship with clinical variables in
sporadic renal cancer. Cancer Res. 2006;66:2000-2011.
66. Maxwell PH, Wiesener MS, Chang GW, et al. The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature.
1999;399:271-275.
67. Ivan M, Kondo K, Yang H, et al. HIFalpha targeted for VHL-mediated destruction by proline hydroxylation: implications for O2 sensing. Science. 2001;
292:464-468.
68. Escudier B, Pluzanska A, Koralewski P, et al; AVOREN Trial investigators. Bevacizumab plus interferon alfa-2a for treatment of metastatic renal cell carcinoma:
a randomised, double-blind phase III trial. Lancet. 2007;370:2103-2111.
69. Escudier B, Eisen T, Stadler WM, et al; TARGET Study Group. Sorafenib in advanced clear-cell renal-cell carcinoma. N Engl J Med. 2007;356:125-134.
70. Motzer RJ, Hutson TE, Tomczak P, et al. Sunitinib versus interferon alfa in metastatic renal-cell carcinoma. N Engl J Med. 2007;356:115-124.
71. Sternberg CN, Davis ID, Mardiak J, et al. Pazopanib in locally advanced or metastatic renal cell carcinoma: results of a randomized phase III trial. J Clin Oncol.
2010;28:1061-1068.
72. Sternberg CN, Hawkins RE, Wagstaff J, et al. A randomised, double-blind phase III study of pazopanib in patients with advanced and/or metastatic renal cell
carcinoma: final overall survival results and safety update. Eur J Cancer. 2013;49:1287-1296.
73. Motzer RJ, Hutson TE, Cella D, et al. Pazopanib versus sunitinib in metastatic renal-cell carcinoma. N Engl J Med. 2013;369:722-731.

Dutcher et al
74. Rini BI, Escudier B, Tomczak P, et al. Comparative effectiveness of axitinib versus sorafenib in advanced renal cell carcinoma (AXIS): a randomised phase 3
trial. Lancet. 2011;378:1931-1939.
75. Motzer RJ, Escudier B, Tomczak P, et al. Axitinib versus sorafenib as second-line treatment for advanced renal cell carcinoma: overall survival analysis and
updated results from a randomised phase 3 trial. Lancet Oncol. 2013;14:552-562.
76. Choueiri TK, Escudier B, Powles T, et al; METEOR Investigators. Cabozantinib versus everolimus in advanced renal-cell carcinoma. N Engl J Med. 2015;
373:1814-1823.
77. Choueiri TK, Escudier B, Powles T, et al; METEOR investigators. Cabozantinib versus everolimus in advanced renal cell carcinoma (METEOR): final results from
a randomised, open-label, phase 3 trial. Lancet Oncol. 2016;17:917-927.
78. Motzer RJ, Nosov D, Eisen T, et al. Tivozanib versus sorafenib as initial targeted therapy for patients with metastatic renal cell carcinoma: results from a phase III
trial. J Clin Oncol. 2013;31:3791-3799.
79. Rini BI, Pal SK, Escudier BJ, et al. Tivozanib versus sorafenib in patients with advanced renal cell carcinoma (TIVO-3): a phase 3, multicentre, randomised,
controlled, open-label study. Lancet Oncol. 2020;21:95-104.
80. Hudes G, Carducci M, Tomczak P, et al; Global ARCC Trial. Temsirolimus, interferon alfa, or both for advanced renal-cell carcinoma. N Engl J Med. 2007;
356:2271-2281.
81. Rini BI, Bellmunt J, Clancy J, et al. Randomized phase III trial of temsirolimus and bevacizumab versus interferon alfa and bevacizumab in metastatic renal cell
carcinoma: INTORACT trial. J Clin Oncol. 2014;32:752-759.
82. Hutson TE, Escudier B, Esteban E, et al. Randomized phase III trial of temsirolimus versus sorafenib as second-line therapy after sunitinib in patients with
metastatic renal cell carcinoma. J Clin Oncol. 2014;32:760-767.
83. Motzer RJ, Escudier B, Oudard S, et al; RECORD-1 Study Group. Efficacy of everolimus in advanced renal cell carcinoma: a double-blind, randomised, placebo-
controlled phase III trial. Lancet. 2008;372:449-456.
84. Yang JC, Haworth L, Sherry RM, et al. A randomized trial of bevacizumab, an anti-vascular endothelial growth factor antibody, for metastatic renal cancer. N Engl
J Med. 2003;349:427-434.
85. Rini BI, Halabi S, Rosenberg JE, et al. Phase III trial of bevacizumab plus interferon alfa versus interferon alfa monotherapy in patients with metastatic renal cell
carcinoma: final results of CALGB 90206. J Clin Oncol. 2010;28:2137-2143.
86. Jonasch E, Slack RS, Geynisman DM, et al. Phase II study of two weeks on, one week off sunitinib scheduling in patients with metastatic renal cell carcinoma.
J Clin Oncol. 2018;36:1588-1593.
87. Keizman D, Neiman V, Sella A, et al. Comparison of sunitinib (su) versus temsirolimus (tem) in patients (pts) with poor-risk metastatic renal cell carcinoma
(prmRCC). J Clin Oncol. 2016;34:15s (suppl; abstr 539).
88. Escudier B, Porta C, Bono P, et al. Randomized, controlled, double-blind, cross-over trial assessing treatment preference for pazopanib versus sunitinib in
patients with metastatic renal cell carcinoma: PISCES Study. J Clin Oncol. 2014;32:1412-1418.
89. McKay RR, Lin X, Perkins JJ, et al. Prognostic significance of bone metastases and bisphosphonate therapy in patients with renal cell carcinoma. Eur Urol.
2014;66:502-509.
90. Choueiri TK, Halabi S, Sanford BL, et al. Cabozantinib versus sunitinib as initial targeted therapy for patients with metastatic renal cell carcinoma of poor or
intermediate risk: The Alliance A031203 CABOSUN trial. J Clin Oncol. 2017;35:591-597.
91. Choueiri TK, Hessel C, Halabi S, et al. Cabozantinib versus sunitinib as initial therapy for metastatic renal cell carcinoma of intermediate or poor risk (Alliance
A031203 CABOSUN randomised trial): progression-free survival by independent review and overall survival update. Eur J Cancer. 2018;94:115-125.
92. Escudier B, Powles T, Motzer RJ, et al. Cabozantinib, a new standard of care for patients with advanced renal cell carcinoma and bone metastases? Subgroup
analysis of the METEOR trial. J Clin Oncol. 2018;36:765-772.
93. Zhou L, Liu X-D, Sun M, et al. Targeting MET and AXL overcomes resistance to sunitinib therapy in renal cell carcinoma. Oncogene. 2016;35:2687-2697.
94. Grünwald V, Eberhardt B, Bex A, et al. An interdisciplinary consensus on the management of bone metastases from renal cell carcinoma. Nat Rev Urol. 2018;
15:511-521.
95. Albiges L, Choueiri T, Escudier B, et al. A systematic review of sequencing and combinations of systemic therapy in metastatic renal cancer. Eur Urol. 2015;
67:100-110.
96. Iacovelli R, Farcomeni A, Sternberg CN, et al. Prognostic factors in patients receiving third line targeted therapy for metastatic renal cell carcinoma. J Urol. 2015;
193:1905-1910.
97. Thomas GV, Tran C, Mellinghoff IK, et al. Hypoxia-inducible factor determines sensitivity to inhibitors of mTOR in kidney cancer. Nat Med. 2006;12:122-127.
98. Motzer RJ, Barrios CH, Kim TM, et al. Record-3: Phase II randomized trial comparing sequential first-line everolimus (EVE) and second-line sunitinib (SUN)
versus first-line SUN and second-line EVE in patients with metastatic renal cell carcinoma (mRCC). J Clin Oncol. 2013;31:15s (suppl, abstr 4504).
99. Fallah J, Rini BI. HIF inhibitors: status of current clinical development. Curr Oncol Rep. 2019;21:6.
100. Courtney KD, Infante JR, Lam ET, et al. Phase I dose-escalation trial of PT2385, a first-in-class hypoxia-inducible factor-2α antagonist in patients with previously
treated advanced clear cell renal cell carcinoma. J Clin Oncol. 2018;36:867-874.
101. Choueiri TK, Plimack ER, Bauer TM, et al. Phase I/II study of the oral HIF-2 α inhibitor MK-6482 in patients with advanced clear cell renal cell carcinoma (RCC).
J Clin Oncol. 2020;38:15s (suppl; abstr 611).

102. NCT04195750. A Study of MK-6482 Versus Everolimus in Participants With Advanced Renal Cell Carcinoma (MK-6482-005). https://clinicaltrials.gov/ct2/
show/NCT04195750. Accessed February 20, 2020.
103. Armstrong AJ, Halabi S, Eisen T, et al. Everolimus versus sunitinib for patients with metastatic non-clear cell renal cell carcinoma (ASPEN): a multicentre, open-
label, randomised phase 2 trial. Lancet Oncol. 2016;17:378-388.
104. Tannir NM, Jonasch E, Albiges L, et al. Everolimus versus sunitinib prospective evaluation in metastatic non-clear cell renal cell carcinoma (ESPN): a ran-
domized multicenter phase 2 trial. Eur Urol. 2016;69:866-874.
105. Jung KS, Lee SJ, Park SH, et al. Pazopanib for the treatment of non-clear cell renal cell carcinoma: a single-arm, open-label, multicenter, phase II study. Cancer
Res Treat. 2018;50:488-494.
106. Park I, Lee SH, Lee JL. A multicenter phase II trial of axitinib in patients with recurrent or metastatic non-clear-cell renal cell carcinoma who had failed prior
treatment with temsirolimus. Clin Genitourin Cancer. 2018;16:e997-e1002.
107. Chanzá NM, Xie W, Bilen MA, et al. Cabozantinib in advanced non-clear-cell renal cell carcinoma: a multicentre, retrospective, cohort study. Lancet Oncol.
2019;20:581-590.
108. Pal SK, Tangen CM, Thompson IM, et al. A randomized, phase II efficacy assessment of multiple MET kinase inhibitors in metastatic papillary renal carcinoma
(PRCC): SWOG S1500. J Clin Oncol. 2017;35:15s (suppl; abstr TPS4599).
109. Ravaud A, Barrios CH, Alekseev B, et al. RECORD-2: phase II randomized study of everolimus and bevacizumab versus interferon α-2a and bevacizumab as
first-line therapy in patients with metastatic renal cell carcinoma. Ann Oncol. 2015;26:1378-1384.
110. Motzer RJ, Hutson TE, Glen H, et al. Lenvatinib, everolimus, and the combination in patients with metastatic renal cell carcinoma: a randomised, phase 2, open-
label, multicentre trial. Lancet Oncol. 2015;16:1473-1482.
111. Voss MH, Molina AM, Chen Y-B, et al. Phase II trial and correlative genomic analysis of everolimus plus bevacizumab in advanced non-clear cell renal cell
112. Choi Y, Keam B, Kim M, et al. Bevacizumab plus erlotinib combination therapy for advanced hereditary leiomyomatosis and renal cell carcinoma-associated
renal cell carcinoma: a multicenter retrospective analysis in Korean patients. Cancer Res Treat. 2019;51:1549-1556.
113. Srinivasan R, Su D, Stamatakis L, et al. 5 Mechanism based targeted therapy for hereditary leiomyomatosis and renal cell cancer (HLRCC) and sporadic
papillary renal cell carcinoma: interim results from a phase 2 study of bevacizumab and erlotinib. Eur J Cancer. 2014;50:8.
114. Wang QJ, Hanada K, Robbins PF, et al. Distinctive features of the differentiated phenotype and infiltration of tumor-reactive lymphocytes in clear cell renal cell
carcinoma. Cancer Res. 2012;72:6119-6129.
115. Fridman WH, Pagès F, Sautès-Fridman C, et al. The immune contexture in human tumours: impact on clinical outcome. Nat Rev Cancer. 2012;12:298-306.
116. Giraldo NA, Becht E, Pagès F, et al. Orchestration and prognostic significance of immune checkpoints in the microenvironment of primary and metastatic renal
cell cancer. Clin Cancer Res. 2015;21:3031-3040.
117. Choueiri TK, Fay AP, Gray KP, et al. PD-L1 expression in nonclear-cell renal cell carcinoma. Ann Oncol. 2014;25:2178-2184.
118. Yang JC, Hughes M, Kammula U, et al. Ipilimumab (anti-CTLA4 antibody) causes regression of metastatic renal cell cancer associated with enteritis and
hypophysitis. J Immunother. 2007;30:825-830.
119. Brahmer JR, Drake CG, Wollner I, et al. Phase I study of single-agent anti-programmed death-1 (MDX-1106) in refractory solid tumors: safety, clinical activity,
121. Motzer RJ, Rini BI, McDermott DF, et al. Nivolumab for metastatic renal cell carcinoma: results of a randomized phase II trial. J Clin Oncol. 2015;33:1430-1437.
122. McDermott DF, Lee J-L, Szczylik C, et al. Pembrolizumab monotherapy as first-line therapy in advanced clear cell renal cell carcinoma (accRCC): results from
cohort A of KEYNOTE-427. J Clin Oncol. 2018;36:15s (suppl. abstr 4500).
123. Lee J-L, Ziobro M, Gafanov R, et al. KEYNOTE-427 cohort B: First-line pembrolizumab (pembro) monotherapy for advanced non‒clear cell renal cell carcinoma
(NCC-RCC). J Clin Oncol. 2019;37:15s (suppl; abstr 4569).
124. Motzer RJ, Escudier B, McDermott DF, et al; CheckMate 025 Investigators. Nivolumab versus everolimus in advanced renal-cell carcinoma. N Engl J Med.
2015;373:1803-1813.
125. Escudier B, Sharma P, McDermott DF, et al; CheckMate 025 investigators. CheckMate 025 randomized phase 3 study: outcomes by key baseline factors and
prior therapy for nivolumab versus everolimus in advanced renal cell carcinoma. Eur Urol. 2017;72:962-971.
126. Curran MA, Montalvo W, Yagita H, et al. PD-1 and CTLA-4 combination blockade expands infiltrating T cells and reduces regulatory T and myeloid cells within
B16 melanoma tumors. Proc Natl Acad Sci USA. 2010;107:4275-4280.
127. Korman AJ, Peggs KS, Allison JP. Checkpoint blockade in cancer immunotherapy. Adv Immunol. 2006;90:297-339.
128. Wolchok JD, Kluger H, Callahan MK, et al. Nivolumab plus ipilimumab in advanced melanoma. N Engl J Med. 2013;369:122-133.
129. Larkin J, Chiarion-Sileni V, Gonzalez R, et al. Combined nivolumab and ipilimumab or monotherapy in untreated melanoma. N Engl J Med. 2015;373:23-34.
130. Griffioen AW, Damen CA, Martinotti S, et al. Endothelial intercellular adhesion molecule-1 expression is suppressed in human malignancies: the role of
angiogenic factors. Cancer Res. 1996;56:1111-1117.
131. Motz GT, Santoro SP, Wang L-P, et al. Tumor endothelium FasL establishes a selective immune barrier promoting tolerance in tumors. Nat Med. 2014;
20:607-615.

Dutcher et al
132. Palazón A, Aragonés J, Morales-Kastresana A, et al. Molecular pathways: hypoxia response in immune cells fighting or promoting cancer. Clin Cancer Res.
2012;18:1207-1213.
133. Voron T, Colussi O, Marcheteau E, et al. VEGF-A modulates expression of inhibitory checkpoints on CD8+ T cells in tumors. J Exp Med. 2015;212:139-148.
134. Motzer RJ, Tannir NM, McDermott DF, et al; CheckMate 214 Investigators. Nivolumab plus ipilimumab versus sunitinib in advanced renal-cell carcinoma.
N Engl J Med. 2018;378:1277-1290.
135. Motzer RJ, Rini BI, McDermott DF, et al; CheckMate 214 investigators. Nivolumab plus ipilimumab versus sunitinib in first-line treatment for advanced renal cell
carcinoma: extended follow-up of efficacy and safety results from a randomised, controlled, phase 3 trial. Lancet Oncol. 2019;20:1370-1385.
136. McDermott DF, Rini BI, Motzer RJ, et al. Treatment-free interval (TFI) following discontinuation of first-line nivolumab plus ipilimumab (N+I) or sunitinib (S) in
patints (pts) with advanced renal cell Carcinoma (ARCC): CheckMate 214. Ann Oncol. 2018;29 (suppl, 8):viii303-viii331.
137. McDermott DF, Rini BI, Motzer RJ, et al. Treatment-free survival (TFS) after discontinuation of first-line nivolumab (NIVO) plus ipilimumab (IPI) or sunitinib
(SUN) in intention-to-treat (ITT) and IMDC favorable-risk patients (pts) with advanced renal cell carcinoma (aRCC) from CheckMate 214. J Clin Oncol. 2019;37:
15s (suppl; abstr 564).
138. Tannir NM, Motzer RJ, Plimack ER, et al. Outcomes in patients (pts) with advanced renal cell carcinoma (aRCC) who discontinued (DC) first-line nivolumab +
ipilimumab (N+I) or sunitinib (S) due to treatment-related adverse events (TRAEs) in CheckMate 214. J Clin Oncol. 2019;37:15s (suppl; abstr 581).
139. Rini BI, Plimack ER, Stus V, et al. Pembrolizumab (pembro) plus axitinib (axi) versus sunitinib as first-line therapy for metastatic renal cell carcinoma (mRCC):
Outcomes in the combined IMDC intermediate/poor risk and sarcomatoid subgroups of the phase 3 KEYNOTE-426 study. J Clin Oncol. 2019;37:15s (suppl;
abstr 4500).
140. McDermott DF, Choueiri TK, Motzer RJ, et al. CheckMate 214 post-hoc analyses of nivolumab plus ipilimumab or sunitinib in IMDC intermediate/poor-risk
patients with previously untreated advanced renal cell carcinoma with sarcomatoid features. J Clin Oncol. 2019;37:15s (suppl; abstr 4513).
141. Rini BI, Motzer RJ, Powles T, et al. Atezolizumab (atezo) + bevacizumab (bev) versus sunitinib (sun) in pts with untreated metastatic renal cell carcinoma
(mRCC) and sarcomatoid (sarc) histology: IMmotion151 subgroup analysis. J Clin Oncol. 2019;37:15s (suppl; abstr 4512).
142. Choueiri TK, Larkin JMG, Pal SK, et al. Efficacy and biomarker analysis of patients (pts) with advanced renal cell carcinoma (aRCC) with sarcomatoid histology
(sRCC): subgroup analysis from the phase III JAVELIN renal 101 trial of first-line avelumab plus axitinib (A 1 Ax) vs sunitinib (S). Ann Oncol. 2019;30 (suppl, 5):
v356-v402.
143. Rini BI, Powles T, Atkins MB, et al; IMmotion151 Study Group. Atezolizumab plus bevacizumab versus sunitinib in patients with previously untreated metastatic
renal cell carcinoma (IMmotion151): a multicentre, open-label, phase 3, randomised controlled trial. Lancet. 2019;393:2404-2415.
145. Choueiri TK, Motzer RJ, Campbell MT, et al. Subgroup analysis from JAVELIN Renal 101: outcomes for avelumab plus axitinib (A + Ax) versus sunitinib (S) in
advanced renal cell carcinoma (aRCC). J Clin Oncol. 2019;37:15s (suppl; abstr 544).
2019;380:1116-1127.
147. Chatzkel JA, Swank J, Ludlow S, et al. Overall responses with coordinated pembrolizumab and high dose IL-2 (5-in-a-row schedule) for therapy of metastatic
clear cell renal cancer: a single center, single arm trial. J Clin Oncol. 2019;37:15s (suppl; abstr 657).
148. Diab A, Hurwitz ME, Cho DC, et al. NKTR-214 (CD122-biased agonist) plus nivolumab in patients with advanced solid tumors: preliminary phase 1/2 results of
PIVOT. J Clin Oncol. 2018;36 (suppl; abstr 3006).
149. Sharabi AB, Nirschl CJ, Kochel CM, et al. Stereotactic radiation therapy augments antigen-specific PD-1-mediated anti-tumor immune responses via cross-
presentation of tumor antigen. Cancer Immunol Res. 2015;3:345-355.
150. Singh AK, Winslow TB, Kermany MH, et al. A pilot study of stereotactic body radiation therapy combined with cytoreductive nephrectomy for metastatic renal cell
carcinoma. Clin Cancer Res. 2017;23:5055-5065.
151. Hannan R, Ishihara D, Louder K, et al. Phase II trial of high-dose interleukin-2 (IL-2) and stereotactic radiation therapy (SABR) for metastatic clear cell renal cell
carcinoma (ccRCC): interim analysis. J Clin Oncol. 2016;34:15s (suppl; abstr 532).
152. Masini C, Iotti C, De Giorgi U, et al. Nivolumab (NIVO) in combination with stereotactic body radiotherapy (SBRT) in pretreated patients (pts) with metastatic
renal cell carcinoma (mRCC): first results of phase II NIVES study. J Clin Oncol. 2020;38:15s (suppl; abstr 613).
153. Hammers HJ, Vonmerveldt D, Ahn C, et al. Combination of dual immune checkpoint inhibition (ICI) with stereotactic radiation (SBRT) in metastatic renal cell
carcinoma (mRCC) (RADVAX RCC). J Clin Oncol. 2020;38:15s (suppl; abstr 614).
154. Rini BI, Appleman LJ, Figlin RA, et al. Results from a phase I expansion cohort of the first-in-class oral HIF-2α inhibitor PT2385 in combination with nivolumab in
patients with previously treated advanced RCC. J Clin Oncol. 2019;37:15s (suppl; abstr 558).
155. Motzer RJ, Grünwald V, Hutson TE, et al. A phase III trial to compare efficacy and safety of lenvatinib in combination with everolimus or pembrolizumab vs
sunitinib alone in first-line treatment of patients (Pts) with metastatic renal cell carcinoma (RCC). J Clin Oncol. 2017;35:15s (suppl; abstr TPS4595).
156. Choueiri TK, Apolo AB, Powles T, et al. A phase 3, randomized, open-label study of nivolumab combined with cabozantinib vs sunitinib in patients with
previously untreated advanced or metastatic renal cell carcinoma (RCC; CheckMate 9ER). J Clin Oncol. 2018;36:15s (suppl; abstr TPS4598).
157. NCT03937219. Study of Cabozantinib in Combination with Nivolumab and Ipilimumab in Patients With Previously Untreated Advanced or Metastatic Renal Cell
Carcinoma (COSMIC-313). https://clinicaltrials.gov/ct2/show/NCT03937219. Accessed February 20, 2020.
158. Zhang T, Ballman KV, Choudhury AD, et al. PDIGREE: An adaptive phase 3 trial of PD-inhibitor nivolumab and ipilimumab (IPI-NIVO) with VEGF TKI
cabozantinib (CABO) in metastatic untreated renal cell cancer (Alliance A031704). J Clin Oncol. 2019;37:15s (suppl; abstr TPS4596).

159. Tannir NM, Agarwal N, Pal SK, et al. A phase III randomized open label study comparing bempegaldesleukin (NKTR-214) plus nivolumab to sunitinib or
cabozantinib (investigator’s choice) in patients with previously untreated advanced renal cell carcinoma. J Clin Oncol. 2019;37:15s (suppl; abstr TPS4595).
160. Aggarwal RR, Thomas S, Hauke RJ, et al. RENAVIV: A randomized phase III, double-blind, placebo-controlled study of pazopanib with or without abexinostat in
patients with locally advanced or metastatic renal cell carcinoma. J Clin Oncol. 2019;37:15s (suppl; abstr TPS681).
161. NCT03501381. Prometheus Laboratories, Syndax Pharaceuticals: High Dose IL 2 and Entinostat in RCC. https://clinicaltrials.gov/ct2/show/NCT03501381.
162. NCT0278150. Nivolumab and Stereotactic Ablative Radiation Therapy (SAbR) for Metastatic Clear Cell Renal Cell Carcinoma. https://clinicaltrials.gov/ct2/show/
NCT02781506. Accessed February 20, 2020.
163. NCT03474497. UCDCC#272: IL-2, Radiotherapy, and Pembrolizumab in Patients Refractory to Checkpoint Blockade. https://clinicaltrials.gov/ct2/show/
NCT03474497. Accessed February 20, 2020.
164. Lalani AA, Swaminath A, Pond GR, et al. Phase II trial of cytoreductive stereotactic hypofractionated radiotherapy with combination ipilimumab/nivolumab for
metastatic kidney cancer (CYTOSHRINK). J Clin Oncol. 2020;38:15s (suppl; abstr TPS761).
165. NCT02293980. A Phase 1, Dose-Escalation Trial of PT2385 Tablets in Patients With Advanced Clear Cell Renal Cell Carcinoma. https://clinicaltrials.gov/ct2/
166. NCT03634540. A Trial of PT2977 in Combination With Cabozantinib in Patients With Clear Cell Renal Cell Carcinoma (ccRCC). https://clinicaltrials.gov/ct2/
167. NCT03075423. Randomized Phase-II Study of Nivolumab Plus Ipilimumab vs. Standard of Care in Untreated and Advanced Non-Clear Cell RCC (SUNI-
FORECAST). https://clinicaltrials.gov/ct2/show/NCT03075423. Accessed February 20, 2020.
168. NCT03635892. A Study of Nivolumab in Combination With Cabozantinib in Patients With Non-Clear Cell Renal Cell Carcinoma. https://clinicaltrials.gov/ct2/
169. NCT03170960. Study of Cabozantinib in Combination With Atezolizumab to Subjects With Locally Advanced or Metastatic Solid Tumors. https://clinicaltrials.
170. NCT03177239. Phase II Sequential Treatment Trial Of Single Agent Nivolumab, Then Combination Ipilimumab + Nivolumab in Metastatic or Unresectable Non-
Clear Cell Renal Cell Carcinoma (ANZUP1602) (UNISoN). https://clinicaltrials.gov/ct2/show/NCT03177239. Accessed February 20, 2020.
171. NCT02982954. A Study To Evaluate the Safety of Nivolumab and Ipilimumab in Subjects With Previously Untreated Advanced or Metastatic Renal Cell Cancer
(CHECKMATE 920). https://clinicaltrials.gov/ct2/show/NCT02982954. Accessed February 20, 2020.
172. McKay RR, Bossé D, Xie W, et al. The clinical activity of PD-1/PD-L1 inhibitors in metastatic non-clear cell renal cell carcinoma. Cancer Immunol Res. 2018;
6:758-765.
173. Chahoud J, Msaouel P, Campbell MT, et al. Nivolumab for the treatment of patients with metastatic non-clear cell renal cell carcinoma (nccRCC): a single-
institutional experience and literature meta-analysis. Oncologist. 2019;24:1-7.
174. Gupta R, Ornstein MC, Gul A, et al. Clinical activity of ipilimumab plus nivolumab (ipi/nivo) in patients (pts) with metastatic non-clear cell renal cell carcinoma
(nccRCC). J Clin Oncol. 2019;37 (suppl; abstr e16084).
175. McKay RR, McGregor BA, Gray K, et al. Results of a phase II study of atezolizumab and bevacizumab in non-clear cell renal cell carcinoma (nccRCC) and clear
cell renal cell carcinoma with sarcomatoid differentiation (sccRCC). J Clin Oncol. 2019;37 (suppl; abstr 548).
176. van der Poel HG, Roukema JA, Horenblas S, et al. Metastasectomy in renal cell carcinoma: a multicenter retrospective analysis. Eur Urol. 1999;35:197-203.
177. Vogl UM, Zehetgruber H, Dominkus M, et al. Prognostic factors in metastatic renal cell carcinoma: metastasectomy as independent prognostic variable. Br
J Cancer. 2006;95:691-698.
178. Zaid HB, Parker WP, Safdar NS, et al. Outcomes following complete surgical metastasectomy for patients with metastatic renal cell carcinoma: a systematic
review and meta-analysis. J Urol. 2017;197:44-49.
179. Brecheteau F, Carrouget J, Lebdai S, et al. How did we obtain complete remission in patients who had metastatic renal cancer using targeted therapies? Eur
Urol. 2016;15 (3, suppl):e531.
181. Rini BI, Battle D, Figlin RA, et al. The society for immunotherapy of cancer consensus statement on immunotherapy for the treatment of advanced renal cell
carcinoma (RCC). J Immunother Cancer. 2019;7:354.
182. Motzer RJ, Jonasch E, Michaelson MD, et al; CGC. NCCN guidelines insights: kidney cancer, version 2.2020. J Natl Compr Canc Netw. 2019;17:1278-1285.
183. Kawakami F, Sircar K, Rodriguez-Canales J, et al. Programmed cell death ligand 1 and tumor-infiltrating lymphocyte status in patients with renal cell carcinoma
and sarcomatoid dedifferentiation. Cancer. 2017;123:4823-4831.
184. Hammers HJ, Plimack ER, Infante JR, et al. Safety and efficacy of nivolumab in combination with ipilimumab in metastatic renal cell carcinoma: The CheckMate
016 Study. J Clin Oncol. 2017;35:3851-3858.
185. Ornstein MC, Pal SK, Wood LS, et al. Individualised axitinib regimen for patients with metastatic renal cell carcinoma after treatment with checkpoint inhibitors:
a multicentre, single-arm, phase 2 study. Lancet Oncol. 2019;20:1386-1394.
186. Shah AY, Kotecha RR, Lemke EA, et al. Outcomes of patients with metastatic clear-cell renal cell carcinoma treated with second-line VEGFR-TKI after first-line
immune checkpoint inhibitors. Eur J Cancer. 2019;114:67-75.

The Next Step in Your Career: Getting the Right

Opportunity and the Right Contract
Claire F. Verschraegen, MD1; Yousif Abubakr, MD2; and Mark Lee, MD, PhD3
overview
A new job is always a difficult step to take in one’s career. Negotiating the optimal contract is paramount. The
contract should be a reasonable agreement that satisfies both parties: the enterprise and the prospective
candidate. Some contract clauses are usually non-negotiable because these items are required for the job.
Examples include board certification, insurance for liabilities, faculty standing for a specific position, and so
forth. Other clauses are negotiable and should be negotiated, including salary, resources to perform the job,
work hours, and so on. This article details what to expect in a contract, how to think about it, and how to act on
it. Possible careers for an oncologist are reviewed and include academia, private practice, and industry at the
largest sense. Recommendations are to be certain that the choice of career feels comfortable, to read and fully
understand the contract, to obtain a few contracts from different job offers to draw comparisons, to ask
questions until all items are clarified, and to not be shy in asking what is not reflected in the contract but would
assuage the unknown of assuming the duties and deliverables of a new job.
INTRODUCTION or a private hospital could “employ” university faculty

Taking the next step in your career will require find- under various terms of agreements.
ing the right opportunity and effectively negotiating “Industry” is the broad setting of for-profit companies
a contract that sets you up for success. This article in health care and the life sciences. Examples can
incorporates practical advice from experts with dif- range from traditional entities like pharmaceutical and
ferent backgrounds and discusses negotiation tips diagnostics companies to insurance companies and
helpful for every oncologist looking for a professional investment firms and to less traditional opportunities
position in academia, private practice, or industry. in high-technology companies and digital health (in-
The vocabulary used in this article is defined in the cluding businesses focused on data and advanced
following manner. “Academia” means an institution of analytics). Additional opportunities can include non-
higher learning, recognized as such legally, and whose profit and public sector entities such as patient ad-
main purposes are (1) to train the next generation of vocacy groups, research foundations, and regulatory
health care workers, usually by managing a medical agencies.
school; (2) to advance research to improve standards Finally, “enterprise” will reflect the employer side, and
of care; and (3) to treat patients through their asso- “prospective candidate” will reflect the individual
ciated hospitals. Faculty in academic settings usually seeking employment.
bear the “professor” label.
UNDERSTANDING OPPORTUNITIES
“Private practice” means that the providers are either
independent professionals with or without privileges to For many oncologists, especially those who are just
Author affiliations one or a few specific hospitals or salaried employees finishing fellowship, exposure to career opportunities
and support of a private for-profit or not-for-profit hospital who do outside of academia or private practice may be limited.
information (if
not work in an academic setting. Independent pro- Even for careers in academia and private practice,
applicable) appear
fessionals can be solo practitioners or have employ- there is tremendous variation by region and institu-
at the end of this tion, with different business models, access to re-
article. ment or partnership in an incorporated physician
group. Private practitioners might have an “adjunct or sources, pay structures, and benefits, and there are
Accepted on
February 21, 2020 associated professor” title. To note, the boundary often differences in culture and career development
and published at between the academia and private practice settings opportunities.
ascopubs.org on
can be difficult to clarify, as many private hospitals Beyond academia and private practice, the last few
March 17, 2020:
contract with academic centers for naming and other decades have seen an explosion in the number and
10.1200/EDBK_ business purposes. In some settings, a hospital can be type of roles that are available to oncologists. What
279635 a university hospital with privileged private practitioners, follows is an illustration of this great breadth of

Negotiating for Success
With the rise of “big data” and advanced analytics in health

care, opportunities arise to participate in businesses aimed
at leveraging this new knowledge to accelerate research and
• Understanding opportunities is the first step to development or to optimize cancer care in everyday prac-
choosing the right career. Opportunities exist in
tice. These new roles are emerging in the pharmaceutical
academia (e.g., a scientist or clinician), private
and diagnostics industry as well as in companies dedicated
practice (e.g., partner or employee), industry,
and other settings ranging from the traditional to digitalization of health care, including health insurance
pharmaceutical company and health care and large high-technology companies, which have ambi-
insurance or analytics organization to not-for- tions in branching out into this space. Finally, there are also
profit philanthropic societies or other private roles working with payors, investors (e.g., venture capital
enterprises. firms assessing investments in new companies targeting
• Understanding of the fair market values and oncology), and nonprofit and public sector entities such as
market opportunities is paramount and varies patient advocacy groups, research foundations and asso-
by area of interest and region. Employers have ciations (e.g., ASCO), and regulatory agencies (e.g., U.S.
a duty to maintain equity among personnel, Food and Drug Administration, Centers for Disease Control
especially as gender differences in salary are and Prevention, and National Institutes of Health).
a well-known societal fact; women should not
For the oncologist starting a professional life, it is critical to
be afraid to negotiate equivalent salary and
positions. educate oneself about opportunities. There are multiple
avenues to pursue in seeking out this self-education, such
• Contract basics include the exact position, the
as tapping into the network at your institution (including
key expectations of deliverables required for the
graduates of the program and faculty), seeking out in-
enterprise’s success, salary, benefits, liability,
other ancillary rewards, and termination formational interviews in your network with people who have
provisions. jobs of potential interest to you, considering industry fel-
lowships (some academic institutions have arrangements
• Do your homework to understand the contract
with companies), engaging recruiters to learn more about
in detail with the use of legal help (optional),
comparison of job offers, discussion with peers opportunities, and visiting with representatives from the full
or your current professional network, and range of potential employers at conferences like the ASCO
so forth. Annual Meeting. It is important to gain clarity on opportu-
nities that could match your professional goals and to un-
• Transparency fosters trust between parties, and
one should not hesitate to negotiate the best derstand elements of the job that would best fit your current
deal possible. The best relationship is a long- interests, aspirations, and life situation.
term and successful one for both parties. For the areas that interest you most, there is tremendous
value in going through the application and interview process
itself.
CONTRACTING BASICS
opportunity and is not meant to be exhaustive. In the
pharmaceutical and diagnostics industry, oncologists have Contracts
abundant opportunities for traditional roles in clinical sci- Successful contracting is imperative for any prospective
ence (e.g., development of clinical strategies for drug/ candidate who wants to embrace a professional career.
diagnostic development, study design, conduct, and analy- Good contracts help ensure equitable pay, good benefits,
sis), translational science (e.g., biomarker discovery and and enough resources to complete the goals that must be
development), medical affairs (e.g., development of strat- achieved for the enterprise’s success and for the pro-
egies for integration of drugs/diagnostics into the clinical spective candidate’s personal success. A contract should
care paradigm; interaction with academic researchers and include both basic and high-level issues, including the
clinicians, medical societies, and guideline committees), exact position, the key expectations of deliverables required
and other stakeholders in the health care ecosystem, infor the enterprise’s success, salary, benefits, liability, other
cluding medical education and health communications. ancillary rewards, and termination provisions. It is important
Beyond these areas, oncologists seeking employment with that the prospective candidate understands the contract
industry might also consider roles in commercial affairs, reg- in detail. Many enterprises have standard contracts, which
ulatory compliance, drug safety, or business development. have been vetted by their legal teams, especially for the
Importantly, these roles can be found in well-established, common items by which every employee abides. The most
publicly traded companies and in a vast number of small negotiable contract areas are generally salary and the re-
and medium-sized start-up companies. sources needed for job success.

Verschraegen, Abubakr, and Lee
Legal Help spirit of a contract should be to ensure the success of any

Because of the perceived complexity of contracts, some tasks that must happen on the job. See Table 1 for
prospective candidates might want to hire a lawyer to help examples.
them understand the contract. This is usually not neces- Multiple Options
sary, unless there is an impasse in obtaining an item that is Finally, it is always a good idea to apply to a few places to be
indispensable to accomplish the job properly. Although able to compare offers.
legal counsel can determine whether a contract contains
appropriate legal language, only the prospective candi- STANDARD TERMINOLOGY
date can determine whether the workload is achievable at A standard language is often used for key areas im-
the site. portant to the employer. For academic enterprises, the
Fair Market Value most commonly addressed items in a physician employment
contract include position, reporting relationships, duties and
The physician profession is usually easily benchmarked for responsibilities, compensation, term and termination, mis-
each specialty. There are two major benchmarking asso- cellaneous, and execution (Fig. 1A). Clinical expectations are
ciations: the Medical Group Management Association, usually measured in relative value units and are becoming
which establishes the clinical workload for private practi- a standard clinical work outcome.3
tioners, and Vizient Inc. and the Association of American
Medical Colleges,1 which launched the Clinical Practice Duties and responsibilities have the most variations, as they
Solutions Center for academic and community practice are tailored to you as an individual with a specific job to
organizations. Salaries also vary by geographic area and accomplish. You should make sure you understand what
whether the practice is urban or rural. objectives you have to reach and what resources are at your
disposal to do so. There are two main kinds of resources:
Understanding of the Context budget and personnel. For clinical care, you must un-
It is also important to understand the various types of derstand what team will support you for the various tasks
contracts and know whether the employer permits negoti- required. These tasks include scheduling, test and treat-
ation. For example, some federal jobs, such as those offered ment preauthorizations, billing, compliance audits, phar-
through the National Cancer Institute, generally offer little macy support, treatment administration, patient education,
room for negotiation. With limited ability to negotiate, the patient consent and insurance help, patient communica-
prospective candidate must compare both positive and tions, nursing and advanced practice provider support,
negative elements in the contract and verify that the de- social worker support, nutritional support, psychological
liverables are doable under the allotted resources. Large support, and other possible items needed for your practice.
organizations usually provide contract templates to pro- Most clinical enterprises have a solid organization to support
spective candidates. Using a standard form might expedite you, but this is not always clear-cut. These organizational
the process for both parties. On the contrary, contracts for items do not have to be listed in a contract, but you should
private employment, such as those sponsored by phar- inquire and visit the clinic site to make sure that what you
maceutical companies or private enterprise, are usually need exists. The contract needs to stipulate that you will be
open for negotiation. Elements deemed undesirable by fully supported for your clinical duties. For your research
the prospective candidate may be negotiated with the en- activities, you must make sure that you have enough time
terprise in an attempt to reach a mutually agreeable con- allocated (full-time equivalent or 100% of time) and that you
clusion. Prospective private practice candidates might understand what research you will embark on (laboratory
generally be more involved in the initial preparation of the bench research, public health research, clinical trials, or
contract to ensure (1) that there is clarity about potential other research areas), how the outcomes will be measured,
partnership in the near future and (2) that the bound- and what resources you will be given initially until you can
ary between professional duties and private life is well obtain a full-time equivalent (or 100% of time) of grants or
balanced.2 contracts. In addition, ask whether there is support for
protocol/grant writing and submission, as well as how the
Negotiations system allocates this resource to you. Sometimes, group
Negotiations should be conducted in a timely manner. The incentives are available. For example, in a division that
first step is for the prospective candidate to ask questions opens many clinical trials, often the research work becomes
each time an item is not clear. Often the point is agreeable a team effort. Resources include money for salary, per-
and small edits are made to the contract language for sonnel you must hire, shared resource expertise, equip-
clarification. The second step is to make a list of the items ment, and other items as appropriate. To facilitate the
the prospective candidate wants to perform in his or her negotiations, try to be as realistic as possible with your
future job. Long detailed lists are discouraged, and the proposed research budget. For teaching and administrative

TABLE 1. How to Read Deliverables and Other Terms in a Contract

Employment
Setting Deliverable Comments*
Academia RVUs Two main standards: MGMA and AAMC/CPSC
Scientific achievements This is what you must pay attention to
Teaching Teaching is usually done for free, unless you teach an official
university course
Administrative service Administrative service on committees is usually done for free;
administrative responsibilities (e.g., medical directorships of
clinics or operating rooms), outreach, and others should be paid
for
Private practice RVUs
Charges
Patient satisfaction
Marketing participation Not required by all employers
Tumor board participation Marketing and tumor board normally without compensation
Timely medical record completion
Compliance with value-based contracts Some practices may bonus efforts
Industry Deliverables are highly specific for each type of position Understand from the hiring manager what the initial scope of work
will be and obtain examples of quarterly/yearly goals for someone
Key terms (general)
coming into that position.
Title and job level (including individual contributor vs. Clarify management responsibilities, reporting structure, access to
management track) resources and collaborations needed to do the work, and the kinds
Base salary of decisions that are in the scope of the role
Bonus structure (annual target as a percentage of base salary)

Stock plans (restricted stock units and options, including
vesting schedule)
Sign-on bonus (if applicable)
Additional considerations may include special features of the
benefits plan such as 401(k) matching, deferred compensation
plans, licensure, board certification and society membership
fees, and so on
Abbreviations: RVU, relative value unit; MGMA, Medical Group Management Association; AAMC, Association of American Medical Colleges; CPSC, Clinical
Practice Solutions Center.
*See the text for discussion.
activities, the offers and deliverables are usually self- alternative retirement 401(k) matching plans and 403(b)
explanatory. Administrative duties tend to be better paid and 457(b) supplemental retirement accounts.
than teaching duties, which are often pro bono. You should
not accept a greater than 5% effort unpaid. READING YOUR CONTRACT
You must read the contract from top to bottom. Circle
For industry positions, there will be substantial variation in
what you do not understand, and write down any
terms of the deliverables, depending on the role. Terms of
questions for clarification. These questions will be the
the contract will generally center on a few standard areas as
start of the negotiations. You must be comfortable with
follows (Table 1): the job title and level (including where that
the contract clauses before you sign the contract. An-
level sits within the job family and the peers in the group),
other item to consider is the noncompete clause, which
base annual salary and bonus structure (which is cus-
could apply to further employment you might want to
tomarily a targeted percentage of the base salary, subject to
seek in the future. 4
adjustment based on company financial performance),
long-term incentive plans (usually provided as stock or stock If there are any links to supplemental information, make
options that vest over a 3- to 4-year window, as long as you sure to click on these links. Sidebar 1 lists an example of
are still with the company), and additional benefits such as terms in a faculty group practice contract clause, which

FIGURE 1. Suggested Templates to Assist With Nego-

tiating a Contract
(A) Academic contract elements. (B) Private practice
contract elements.
Abbreviations: CART, clinical, administration, research,
teaching; CME, continuing medical education; FTE, full-
time equivalent; RVU, relative value unit.
contains a lot of legal language that might be more difficult types of benefits you need for you and your family. You
to negotiate. Sidebar 2 lists benefits that are usually of- must also understand your benefits and those of your
fered by employers. You must make decisions on the dependents.

candidates must consider worst-case scenarios and be

sure the contract guarantees the protection of each party.
Using a checklist that identifies all major elements of
a contract is a useful way to confirm that all required el-
ements are included. This checklist is particularly useful
for comparing offers from different employers. Each pro-
spective candidate should also use a checklist that lists all
factors to be addressed in the contract to reassure the
applicant.
NEGOTIATING
Negotiations are always best when conducted in a pro-
fessional and timely manner, but they must lead to a realistic
outcome. Remember that both parties are working toward
the same goal: to guarantee the success of the enterprise.
Employers want employees who meet goals and beyond,
and it is the responsibility of the employer to ensure that
these goals will be met under the terms of the contract. This
cannot happen if the candidate is not an integral part of this
success.5
After the prospective candidate receives a draft contract
or letter of offer, the next step is negotiating. To ensure
a smooth negotiation process, the prospective candidate
must prepare to engage the enterprise in a fruitful discussion
(Sidebar 3). In addition to reading and understanding the
contract, initial preparation should involve understanding of
fair market value, duties and deliverables, benefits, and
termination conditions. Understanding the perspective of the
enterprise is always advantageous to help figure out whether
the contract offers a match. The best match will occur when
visions and expectations coincide between employer and
candidate. It is important to determine what elements, if any,
require negotiation. If several elements are identified, they
should be ranked in order of importance, from items that
must be modified versus the ones that can remain and still
ensure long-term success. Be prepared to provide docu-
mentation that supports items that are negotiated. An ex-
ample would be cost of living in one area versus another area.
Employers understand that there are regional and geographic
differences.
In some cases, the candidate may have special consider-
ations that could make the position substantially more at-
tractive if met. For industry roles, examples could include
the ability to serve part-time as (adjunct) faculty at an ac-
ademic medical center, including some limited amount of
clinic time per week or month; limitations on travel ex-
pectations; or options for flexible working arrangements,
such as working from home a certain number of days each
week. Although there is no guarantee that an employer can
FIGURE 1. Continued.
meet such considerations, the candidate should not hold
Although a contract is always important, it becomes es- back from sharing these considerations, especially if they
pecially vital when things are not going as planned. During could affect the candidate’s decision on whether to accept
the initial contracting process, employers and prospective the role.

SIDEBAR 1. EXAMPLE OF GROUP PRACTICE ITEMS THAT YOU MUST UNDERSTAND PRIOR TO SIGNING THE CONTRACT
• Introduction
• Board of trustees approval
• Physician duties and qualifications
• Clinical, teaching, and research duties
• University integrity program
• Ethics law
• Work authorization
• Self-disclosure of criminal convictions and background check policy
• Performance review
• Medical staff appointment and clinical privileges
• Licensure
• Exclusive practice
• Scheduling and patient care responsibilities
• Absences
• Essential position
• Nondiscrimination
• Support of the university medical center
• Other benefits
• Compliance with Section 409A of the Internal Revenue Code
• Space, services, and equipment
• Professional liability insurance
• Billing and collection
• Third-party payor agreements
• Compliance with billing compliance program
• Authorization
• Eligibility to participate in health care programs
• Termination
• Confidentiality, access to records, and restrictive covenants
• Incorporated group physician practice
• Compliance with laws
• Amendments
• Applicable law
• No waiver
• Assignment
• Accountable care organization participation
SIDEBAR 2. BENEFIT EXAMPLES

• Medical, dental, and vision benefits
• Employee assistance program
• Flexible spending accounts
• Healthy life support
• Retirement programs
• Life insurance
• Disability insurance
• Tuition assistance
• Adoption assistance
• Stock options
• Additional benefits

SIDEBAR 3. FURTHER SUGGESTED READING

• American Academy of Family Physicians. Employment Contracting: Five Key Elements of a Physician Employment
Agreement. https://www.aafp.org/practice-management/payment/contracts.html.
• Felberg RA. “Anatomy of a complex physician contract negotiation.” The White Coat Investor, September 6, 2017. https://
www.whitecoatinvestor.com/anatomy-of-a-complex-physician-contract-negotiation/.
• Karass CL (ed). In Business, As in Life, You Don’t Get What You Deserve, You Get What You Negotiate. Beverly Hills, CA:
Stanford St. Press; 1996.
When negotiating by electronic means, be sure to use the CONCLUSION

track-change feature for ease of interpretation. This helps Taking the next step in your career may seem daunting, but
employers quickly address the proposed modifications. If it is important to recognize the breadth of opportunities
electronic communication does not resolve issues, call the available to oncologists. Although engaging in a contract may
other party and start negotiating by phone and then provide also seem daunting for a prospective candidate, both parties
a written follow-up to confirm agreement. Often, the employer have a vested interest in finding a good match. If a job leads to
might call the prospective candidate to clarify remaining ques- burnout or unsatisfactory outcomes, everyone loses. Negotiating
tions or to establish a personal relationship that will become long could be seen more as a game than an art, in which one seeks
lasting. Transparency fosters trust between parties and one to obtain the optimal conditions to perform defined tasks. In the
should not hesitate to negotiate the best deal possible. Employers, end, you must ask what you think you need and justify why. If
on the other hand, have to make sure they maintain equity you receive a rebuttal, the best way to understand the situation is
between the faculty/personnel. This is quite important, as gender to compare various offers side by side using a checklist and to
differences in salary are a well-known societal fact. Women ask specific questions to obtain what is needed for you to
should not be afraid to negotiate equivalent salary and positions. perform your job in the best possible conditions.

1
Ohio State University Comprehensive Cancer Center, Columbus, OH AND DATA AVAILABILITY STATEMENT
2
Cancer Specialists of North Florida, Jacksonville, FL Disclosures provided by the authors and data availability statement (if
3
Genentech Inc., South San Francisco, CA applicable) are available with this article at DOI https://doi.org/10.1200/
EDBK_279635.
Claire F. Verschraegen, MD, Ohio State University Comprehensive Cancer
Center, A455 Starling Loving Hall, 320 W. 10th Ave., Columbus, OH
REFERENCES
1. Association of American Medical Colleges. Vizient and AAMC launch Clinical Practice Solutions Center to drive efficiency for physician practices [press release].
Washington, DC: Association of American Medical Colleges; June 12, 2019. www.aamc.org/news-insights/press-releases/vizient-and-aamc-launch-clinical-
practice-solutions-center-drive-efficiency-physician-practices. Accessed February 2, 2020.
2. Eisemann BS, Wagner RD, Reece EM. Practical negotiation for medical professionals. Semin Plast Surg. 2018;32:166-171.
3. Barr TR, Towle EL. Oncology practice trends from the national practice benchmark. J Oncol Pract. 2012;8:292-297.
4. LaFoe L. Negotiating noncompetition agreements. Mo Med. 2019;116:30-31.
5. Leverence R, Nuttall R, Palmer R, et al. Using organizational philosophy to create a self-sustaining compensation plan without harming academic missions. Acad
Med. 2017;92:1133-1137.

SARCOMA

SARCOMA
How Technology Is Improving the

Multidisciplinary Care of Sarcoma
Inga-Marie Schaefer, MD1; Kelvin Hong, MD2; and Anusha Kalbasi, MD3
overview
Sarcomas are rare tumors but comprise a wide histologic spectrum. Advances in technology have emerged to
address the biologic complexity and challenging diagnosis and treatment of this disease. The diagnostic
approach to sarcomas has historically been based on morphologic features, but technologic advances in
immunohistochemistry and cytogenetic/molecular testing have transformed the interdisciplinary work-up of
mesenchymal neoplasms in recent years. On the therapeutic side, technologic advances in the delivery of
radiation have made it a linchpin in the treatment of localized and oligometastatic sarcoma. In this review, we
discuss recent advances in the pathologic diagnosis of sarcomas and discuss select sarcoma types that il-
lustrate how newly discovered diagnostic, prognostic, and predictive biomarkers have refined existing
classification schemes and substantially shaped our diagnostic approach. Such examples include conven-
tional and epithelioid malignant peripheral nerve sheath tumors (MPNSTs), emerging entities in the group of
round cell sarcomas, and other mesenchymal neoplasms with distinct cytogenetic aberrations. Recent ad-
vances in radiation oncology, including intensity-modulated, stereotactic, MRI-guided, and proton radio-
therapy (RT), will be reviewed in the context of neoadjuvant or adjuvant localized soft-tissue sarcoma and
oligometastatic or oligoprogressive disease. Innovations in translational research are expected to be in-
troduced into clinical practice over the next few years and will likely continue to affect the rapidly evolving field
of sarcoma diagnostics and therapy.
INTRODUCTION Here, we will first cover select recent technologic ad-

Sarcomas are rare tumors, accounting for approxi- vances and their implications for sarcoma diagnostics.
mately 1% of cancers, and comprise a diverse spec- We will then explore in detail the impact of technology
trum of mesenchymal neoplasms with varied prognosis, on treatment, with an emphasis on therapeutic ap-
including approximately 70 with intermediate or ma- proaches in radiation oncology.
lignant biologic potential.1,2 The number of tumors of HOW THE RAPIDLY EVOLVING FIELD OF PATHOLOGY IS
soft tissue and bone included in the World Health Or- INTERFACING WITH THE COMPLEXITY OF SARCOMA
ganization classification has been increasing over the
Brief Overview of Established Diagnostic Techniques
past decades. This is attributed largely to continuously
refined classification schemes and technical advances Given the complexity of the sarcoma histologic spec-
in the immunohistochemical and cytogenetic/molecular trum, thoughtful integration of clinical presentation,
genetic work-up that enabled the discovery of bi- patient (and family) history, radiologic imaging (when
ologically distinct entities with specific prognostic and/or available), gross and histomorphologic clues, immu-
predictive implications. nohistochemical staining pattern, and (when appro-
priate) cytogenetic and molecular genetic testing are
Technology has also affected the care of patients with
crucial to establish a correct diagnosis in most cases.
Author affiliations sarcoma beyond diagnosis, with an imprint on the
and support
As determined by rigorous classification schemes, the
efficacy and safety of treatment approaches. RT is
information (if histopathologic diagnosis of sarcomas is largely based
a technology-based therapeutic modality that has been
applicable) appear on morphology and distinguishes tumors with spindle
at the end of this a critical adjuvant to surgery since the emergence of
cell, epithelioid/epithelial-like, round cell, myxoid, and
article. limb-preservation approaches for soft-tissue sarcoma.
pleomorphic appearances (Table 1). In fact, many
Accepted on Continued technologic advances in the areas of image entities with distinct features can be diagnosed on
February 23, 2020 guidance (including MRI guidance, intensity modu- hematoxylin and eosin–stained sections alone.
and published at lation, stereotactic RT, and proton-based RT) have
ascopubs.org on May
18, 2020: DOI https://
improved the efficacy of RT while reducing its toxicity Added value of immunohistochemistry The imple-
doi.org/10.1200/ and expanding the role of RT for patients with oligo- mentation of immunohistochemistry into the histo-
EDBK_280729 metastatic or oligoprogressive disease. pathologic work-up of soft-tissue tumors in the past

Schaefer, Hong, and Kalbasi
sarcoma and help inform and refine current classification

schemes.
• Although rare, sarcomas comprise a wide his- Recent Advances in Sarcoma Diagnostics
tologic spectrum of diseases characterized by Recent advances in the discovery of distinct recurrent cyto-/
unique prognostic and therapeutic genetic aberrations and development of associated im-
implications. munohistochemical markers have had a substantial impact
• Initial sarcoma diagnostic work-up may require on the diagnostic approach to certain sarcomas and will be
ancillary immunohistochemical, cytogenetic, reviewed in detail.
and/or molecular genetic testing as appropriate
to confirm or further refine a diagnosis based on Conventional and epithelioid malignant peripheral nerve
morphology alone. sheath tumors MPNSTs are aggressive sarcomas associ-
• For localized extremity or trunk soft-tissue ated with high rates of distant metastases and 5-year sur-
sarcoma, advanced radiation techniques, in- vival rates of 35% to 50%.4 Their histopathologic diagnosis
cluding CT or MRI guidance and IMRT, should has been challenging because diagnostic features of (1)
be considered standard. identifiable origin from a peripheral nerve or neurofibroma,
• Selective use of stereotactic RT for (2) immunohistochemical/ultrastructural evidence of
oligometastatic and oligoprogressive disease is Schwann cell differentiation, and/or (3) association with
a well-tolerated and effective approach to neurofibromatosis type I are often absent. Histologically,
bridge patients between systemic therapies. MPNST appears as spindle cell sarcoma with alternation of
hyper- and hypocellular areas and accentuation of tumors
cells around blood vessels (Fig. 1A–C). However, these
decades enabled the detection of lineage-specific markers
features are relatively nonspecific in isolation. Expression of
(Table 1). For instance, myogenic markers such as desmin,
nerve sheath markers (S-100 protein, SOX10, and GFAP) is
smooth muscle actin (SMA), and caldesmon are generally
detectable in only up to 40% of MPNSTs, and, if positive,
expressed in tumors exhibiting smooth muscle differentiation,
usually limited in extent (Table 1). The discovery of char-
such as leiomyoma and leiomyosarcoma.1 Another example
acteristic loss-of-function mutations in SUZ12 or EED,
are the neural markers S-100 protein and SOX10, expressed
encoding components of the polycomb repressive complex 2
in benign and MPNSTs.1 However, none of these immuno-
(PRC2),5,6 and subsequent loss of trimethylation at lysine 27
histochemical stains is entirely specific or sensitive for a di-
of histone 3 (H3K27me3) led to the introduction of
agnosis per se.
H3K27me3 immunohistochemistry as a useful diagnostic
Many sarcomas exhibit characteristic cytogenetic and/or marker.7-9 H3K27me3 loss can be detected in up to 80% of
molecular genetic aberrations (Table 2), with an expo- MPNSTs, in particular in high-grade tumors, and has been
nentially growing number of newly discovered alterations, shown to be quite specific in the distinction of MPNST from
owing to increased sensitivity and more frequent application other spindle cell neoplasms.8 H3K27me3 loss by immu-
of molecular testing in routine diagnostics. Several immunohistochemistry represents a biomarker that demonstrates
nohistochemical markers have been discovered that either the epigenetic consequences of genomic PRC2 inactivation
directly or indirectly correspond to distinct genetic or epi- characteristic of conventional MPNST.
genetic alterations, as will be discussed more in detail.
In contrast, epithelioid MPNST, which is considered a rare
Ancillary cytogenetic and molecular genetic analysis histologic variant of MPNST, follows a less aggressive
Cytogenetic and molecular genetic testing has substantially clinical course, is generally not associated with neurofi-
advanced the routine diagnostic work-up of sarcomas. This bromatosis type I, and lacks PRC2 genomic inactivation
includes conventional karyotyping and fluorescence in situ (Fig. 1D–F).5,8,10 Instead, these tumors arise sporadically or
hybridization (FISH) for detection of known genetic rear- in association with conventional or epithelioid benign nerve
rangements as well as targeted next-generation sequencing sheath tumors and harbor loss of the tumor suppressor
(NGS), anchored multiplex polymerase chain reaction, SMARCB1 (i.e., INI-1) by immunohistochemistry in 70% of
targeted or whole transcriptome RNA sequencing for de- cases (Table 1).10 As demonstrated recently, SMARCB1
tection of unknown gene fusions, and/or reverse tran- loss in epithelioid MPNST results from inactivating muta-
scription polymerase chain reaction.3 These technologies tions of SMARCB1 on chromosome 22.11 SMARCB1 en-
are expected to further decrease in cost and effectively codes a key subunit of the SWI/SNF1 chromatin remodeling
complement the morphology and immunohistochemistry- complex that orchestrates chromatin organization and ac-
based diagnosis of sarcomas. In addition, they allow for the cessibility and has, in part, biologic functions that oppose
detection of hitherto unknown gene fusions, which may add PRC2. SMARCB1 inactivation and SMARCB1 protein loss
to our current understanding of the genetic drivers of can be found in a wide range of benign and malignant

FIGURE 1. Genetic and Immu-

nohistochemical Characteristics
in Conventional and Epithelioid
Malignant Peripheral Nerve Sheath
Tumors
A high-grade malignant periph-
eral nerve sheath tumor (MPNST)
of the psoas region (A), with loss
of trimethylation at lysine 27 of
histone 3 (H3K27me3) in tumor
cells (B; vascular endothelial cells
and inflammatory cells serve as
positive internal control). This
tumor shows a homozygous
c.274+1G.A splice site muta-
tion (allele fraction 80%) in
SUZ12 (C, arrow), which leads
to inactivation of the polycomb
repressive complex 2 (PRC2)
and loss of H3K27me3. This
MPNST also had biallelic in-
activation of NF1 (not shown).
Epithelioid MPNST of the pop-
liteal fossa with epithelioid morphology of tumor cells showing nuclear atypia, prominent nucleoli, and frequent mitoses (D) showing loss of
SMARCB1 expression in tumor cells (E; inflammatory cells serve as positive internal control, arrow) resulting from a homozygous deletion affecting
the entire coding region of SMARCB1 at 22q11.23 (F, arrow).
mesenchymal neoplasms,12 including 40% of epithelioid consist of moderately pleomorphic round to ovoid tumor
schwannomas, which rarely give rise to epithelioid cells with frequent mitoses, apoptoses, and necrosis, which
MPNST.13 SMARCB1 loss is therefore not specific for epi- aides in the distinction from Ewing sarcoma, which usually
thelioid MPNST, but when interpreted in the context of comprises a monomorphic cell population and infrequent
cytomorphology, the presence of necrosis and frequent mitoses, apoptoses, or necrosis (Fig. 2D–F).
mitoses, and strong and diffuse expression of S-100 protein Because access to cytogenetic analyses such as conven-
and SOX10 support a diagnosis of epithelioid MPNST. In tional karyotyping or FISH for detection of characteristic CIC-
addition, SMARCB1 loss helps rule out other malignancies rearrangement and turnaround time can be limitations in
with overlapping histologic and immunohistochemical some institutions, immunohistochemical expression of WT1
features, such as (metastatic) malignant melanoma, in (. 90% of cases) and ETV4 (90% of cases)19,20 can be
which SMARCB1 expression is generally retained.14 sufficient to support a diagnosis of CIC-rearranged sarcoma
Emerging entities in the group of round cell sarcomas in the presence of unequivocal histologic features; CD99
Added value of immunohistochemical markers has also staining is usually limited in CIC-rearranged sarcoma
been demonstrated in the differential diagnosis of recently (Table 1). In contrast, diffuse membranous expression of
discovered entities in the group of round cell sarcomas. First CD99 and nuclear expression of the transcription factor
described in 2012 as a type of round cell sarcoma lacking NKX2.2 and the presence of EWSR1 rearrangement by
EWSR1 rearrangement,15 CIC-rearranged sarcoma shows FISH would favor a diagnosis of Ewing sarcoma (Fig. 2D–F;
predilection for the soft tissue of trunk and extremities of Tables 1 and 2).21-23
younger male adults, follows a more aggressive clinical Another distinct type of round cell sarcoma lacking EWSR1
course compared with Ewing sarcoma with overall survival rearrangement initially reported in 201224 with predilection
rates of 43% vs. 76%, and generally does not respond well for bone and soft tissue of male children24,25 is characterized
to systemic therapies established for Ewing sarcoma16 (Fig. by BCOR-CCNB3 rearrangement resulting from inv(X)(p11)
2A–C). Distinction of CIC-rearranged sarcoma from Ewing (i.e., X-chromosomal paracentric inversion; Tables 1 and 2).
sarcoma therefore has prognostic and predictive value. Rare cases harbor an alternate rearrangement of BCOR with
These tumors harbor characteristic t(4;19)(q35;q13) or MAML3 or ZC3H7B.26 BCOR-rearranged sarcomas follow
t(10;19)(q26;q13), resulting mostly in CIC-DUX4 fusion an aggressive clinical course with 5-year overall survival
(Table 2); rare cases with alternate CIC-FOXO4 fusion have rates of approximately 75%, similar to Ewing sarcoma but
been reported.17,18 Histologically, CIC-rearranged sarcomas less aggressive than CIC-rearranged sarcoma.25,27 BCOR

FIGURE 2. Useful Diagnostic

Markers for Round Cell Sarcomas
CIC-rearranged sarcoma of the
perineal region is characterized
by a morphologically heteroge-
neous population of primitive
round to ovoid or spindled tumor
cells (A) with frequent mitoses
(A, inset) and nuclear expres-
sion of WT1 (B). Rearrangement
of the CIC locus at 19q13.2 with
several break-apart signals was
detected by fluorescence in
situ hybridization (FISH, C). In
contrast, Ewing sarcoma com-
prises sheets of uniform tumor
cells exhibiting rounded nuclei
and inconspicuous nucleoli (D)
with diffuse membranous ex-
pression of CD99 (D, inset) and
diffuse nuclear expression of the
transcription factor NKX2.2 (E).
Rearrangement of EWSR1 at
22q12 can be detected by FISH (F; break-apart signal indicated by arrow).
immunohistochemistry can aid in the diagnosis of BCOR- rearrangement. ALK staining has been reported in various
rearranged sarcomas, in particular when genetic testing for ALK-rearranged mesenchymal neoplasms, such as epi-
BCOR rearrangement is not available.25,28 thelioid fibrous histiocytoma,36 Spitz nevus,37 so-called mela-
nocytic myxoid spindle cell tumor with ALK rearrangement,38
Other mesenchymal neoplasms with distinct cytogenetic and single cases of leiomyosarcoma,39 with potential im-
aberrations Inflammatory myofibroblastic tumor (IMT), plications for targeted therapies using ALK inhibitors.
epithelioid vascular neoplasms, and postradiation angio- However, ALK expression is also found, for instance, in
sarcoma represent examples of mesenchymal neoplasms subsets of epithelioid and spindle cell rhabdomyosarcomas
with characteristic genetic aberrations that have been with TFCP2 fusion lacking ALK rearrangement,40 suggesting
ttranslated into useful diagnostic immunohistochemical that ALK staining is not always associated with ALK
markers. rearrangement.
IMT shows predilection for the visceral soft tissues of With the discovery of recurrent fusions in the group of
children and young adults, with a tendency for local re- vascular neoplasms, highly specific and sensitive immu-
currence but a small risk of distant metastasis.29 IMT nohistochemical markers have been introduced into the
comprises spindled tumor cells with fasciitis-like, compact routine diagnostic setting over the past few years. Although
spindle cells and hypocellular fibrous patterns, with minimal they are classified as low-grade malignant endothelial
cytologic atypia and a scattered inflammatory infiltrate (Fig. neoplasm, some cases of epithelioid hemangioendothe-
3A–C). Rearrangements of ALK at 2p23 are identified in lioma (EHE) behave in a frankly malignant fashion, and their
about 50% of cases, particularly when arising in younger distinction from other mesenchymal and nonmesenchymal
patients. TPM3-ALK fusion30 represents the most frequent neoplasms with overlapping histologic appearances, such
aberration in IMT, but ALK fusions with various other as metastatic carcinoma, is important for patient manage-
partners have been reported (Table 2). ALK rearrangement ment. EHEs occur over a wide age range and show pre-
results in upregulation of ALK expression, which is de- dilection for the soft tissue of extremities and trunk, often
tectable by immunohistochemistry (Fig. 3A–C; Table 1).29 In arising in association with a large vein. Occasionally, these
the distinction from true smooth muscle tumors, this marker tumors present as multifocal disease in lung, liver, and
can be very useful; however, in older patients, ALK staining bone. Local recurrence occurs in 15%, distant metastasis in
is often negative and does not rule out a diagnosis of IMT. In 30% (in soft-tissue sites), and mortality ranges from 15% for
addition to ALK-positive anaplastic large cell lymphoma soft-tissue sites to 50% for tumors arising in liver and lung.
and ALK-positive large B-cell lymphoma,31,32 certain car- Histologically, EHE comprises cords and strands of round to
cinomas of lung,33 thyroid,34 and kidney35 harbor ALK epithelioid endothelial cells with characteristic glassy to

TABLE 1. Select Sarcomas With Spindle Cell, Epithelioid, Myxoid, Pleomorphic, and Round Cell Morphologies and Useful Diagnostic Features and/or
Immunohistochemical Markers
Immunohistochemistry
Morphologic
Pattern Sarcoma Type Diagnostic Clues Positive Markers Negative Markers
Spindle cell Leiomyosarcoma Fascicular growth, brightly eosinophilic Desmin, SMA,
cytoplasm, cigar-shaped nuclei caldesmon
MPNST Perivascular accentuation, alternation of hyper- S100, SOX10, GFAP H3K27me3 (up to 80% of cases)
and hypocellular areas, wavy nuclei (, 50% each)
Monophasic synovial Overlapping nuclei, monomorphic TLE1, EMA/keratins
sarcoma cytomorphology, hemangiopericytoma-like
blood vessels
Dedifferentiated Adjacent well-differentiated liposarcoma MDM2, CDK4,
liposarcoma HMGA2
GIST, spindle cell type Location (GI tract); uniform cytomorphology, DOG1, KIT
fibrillary/syncytial cytoplasm
Spindle cell/sclerosing Presence of rhabdomyoblasts Desmin, Myf4,
rhabdomyosarcoma MYOD1
Solitary fibrous tumor “Patternless” architecture, STAT6, CD34
hemangiopericytoma-like blood vessels
Inflammatory Fascicles of uniform plump spindle cells with ALK (50%), ROS1
myofibroblastic tumor ovoid to tapering nuclei, myxoid stroma, (, 10%), smooth
inflammatory infiltrate muscle markers
(subset)
DFSP Uniform spindle cells with storiform or whorled CD34
growth pattern, infiltration of adipose tissue
(honeycomb appearance)
Epithelioid/ Epithelioid sarcoma Diffuse infiltration along fascial planes, EMA/keratins, CD34 SMARCB1 (90%)
epithelial- pseudogranulomatous appearance (55%)
like
Alveolar soft part sarcoma Pseudo-alveolar/nested growth, large tumors TFE3
cells with abundant granular eosinophilic
cytoplasm
Epithelioid MPNST Multinodular architecture, large nucleoli S100, SOX10, GFAP SMARCB1 (70%)
(60%)
GIST, epithelioid type Location (GI tract); multinodular growth in DOG1, KIT SDHB in SDH-deficient GIST,
gastric tumors suggestive of SDH deficiency additional SDHA loss in SDHA-
mutant GIST; KIT often negative
or limited in PDGFRA-mutant
GIST
Epithelioid Cords and strands of tumor cells embedded in ERG, CD31,
hemangioendothelioma (chondro-)myxoid stroma CAMTA1 (90%) or
TFE3 (5%)
Pseudomyogenic Sheets and loose fascicles of plump spindle CD31, ERG, keratin,
hemangioendothelioma cells with abundant eosinophilic cytoplasm, FOSB (. 95%)
sometimes mimicking rhabdomyoblasts

TABLE 1. Select Sarcomas With Spindle Cell, Epithelioid, Myxoid, Pleomorphic, and Round Cell Morphologies and Useful Diagnostic Features and/or
Immunohistochemical Markers (Continued)
Immunohistochemistry
Morphologic
Pattern Sarcoma Type Diagnostic Clues Positive Markers Negative Markers
Round cell Embryonal Monomorphic round to spindled cell, loose Desmin, Myf4
rhabdomyosarcoma myxoid stroma (limited to
a subset of cells),
MYOD1
Alveolar Monomorphic large cells with even chromatin Desmin (diffuse),
rhabdomyosarcoma and prominent nucleoli, pseudo-alveolar Myf4 (diffuse),
architecture MYOD1
Desmoplastic small round Intra-abdominal location, young males WT1 (C-terminal),
cell tumor keratin, EMA,
desmin
Ewing sarcoma Small amounts of cytoplasm, rare mitoses, CD99 (100%), ETV4, WT1, BCOR, CCNB3
monomorphic cytomorphology NKX2.2 (. 90%)
CIC-rearranged sarcoma Primitive round to ovoid and sometimes CD99 (20%), ETV4 NKX2.2
spindled cytomorphology, irregularly shaped (. 90%), WT1 (.
vesicular nuclei, mitoses, necrosis, 90%)
morphologic heterogeneity
BCOR-rearranged sarcoma Monomorphic or primitive appearing round to CD99 (80%), BCOR NKX2.2
ovoid and occasionally spindled tumor cells (. 90%), CCNB3
arranged in intersecting fascicles or (90%)
a patternless fashion
Myxoid Myxofibrosarcoma Curvilinear blood vessels, pleomorphic cells None (can express
focal MDM2)
Low-grade fibromyxoid Abrupt transition between myxoid and MUC4
sarcoma nonmyxoid areas, bland and uniform spindle
cell morphology
Extraskeletal myxoid Lobulated growth, reticular architecture, S100 (, 50%), EMA SMARCB1 (17%)
chondrosarcoma uniform nuclear morphology (30%)
Myxoid liposarcoma Delicate plexiform vasculature (“crow’s feet” None
vessels), uni- or bivacuolated lipoblasts
palely eosinophilic cytoplasm and intracytoplasmic vacu- resulting from t(X;11)(p11;q22) (Tables 1 and 2).45 This
oles, surrounded by myxohyaline or collagenous stroma subset of EHE has recently been shown to follow a less
(Fig. 3D–F; Table 1). The discovery of recurrent t(1;3)(p36.3; aggressive clinical course compared with those with ca-
q25) translocation in 201141 in approximately 90% of nonical WWTR1-CAMTA1 fusion, with 5-year overall survival
cases resulting in WWTR1-CAMTA1 fusion,42,43 led to the rates of 86% versus 59%,46 and exhibits distinct morpho-
development of a highly specific and sensitive CAMTA1 logic features, such as tumor cells with prominent, volu-
immunohistochemical stain for the diagnosis of EHE in minous eosinophilic cytoplasm and focally well-formed
distinction from histologic mimics (Tables 1 and 2).44 A vascular channels. This subset of EHE is negative for
small subset (, 10% of cases) lack WWTR1-CAMTA1 CAMTA1 immunohistochemistry and instead shows nuclear
fusion and instead harbor alternate YAP1-TFE3 fusion expression of TFE3.

TABLE 2. Select Sarcomas With Diagnostically Relevant Distinct Cytogenetic and Molecular Alterations
Sarcoma Type Cytogenetic Alteration Molecular Alteration
Alveolar rhabdomyosarcoma t(2;13)(q35;q14) PAX3-FOXO1 fusion
t(1;13)(p36;q14), double PAX7-FOXO1 fusion
minutes
t(2;2)(q35;p23) PAX3-NCOA1 fusion
t(X;2)(q35;q13) PAX3-AFX fusion
Alveolar soft part sarcoma t(X;17)(p11.2;q25) TFE3-ASPSCR1 fusion
BCOR-rearranged sarcoma Inv(X)(p11.4p11.22) BCOR-CCNB3 fusion
t(X;4)(p11;q31) BCOR-MAML3 fusion
t(X;22)(p11;q13) ZC3H7B-BCOR fusion
CIC-rearranged sarcoma t(4;19)(q35;q13) or t(10; CIC-DUX4 fusion
19)(q26;q13)
t(X;19)(q13;q13.3) CIC-FOXO4 fusion
Clear cell sarcoma t(12;22)(q13;q12) EWSR1-ATF1 fusion
t(2;22)(q32.3;q12) CREB1-EWSR1 fusion
Dedifferentiated liposarcoma Ring and giant marker Amplification of 12q13-15: MDM2, CDK4 (HMGA2)
chromosomes
Desmoplastic small round cell tumor t(11;22)(p13;q12) EWSR1-WT1 fusion
DFSP Ring form of chromosomes 17 COL1A1-PDGFB fusion
and 22
Endometrial stromal sarcoma, low t(7;17)(p15;q21) JAZF1-SUZ12 fusion
grade
t(6;7)(p21;7p15) PHF1-JAZF1 fusion
t(6;10)(p21;p11) EPC1-PHF1 fusion
t(1;6)(p34;p21) MEAF6-PHF1 fusion
t(X;17)(p11;q21) MBTD1-CXorf67 fusion
Endometrial stromal sarcoma, high t(10;17)(q22;p13) YWHAE-NUT2 fusion
grade
t(X;22)(p11;q13) ZC3H7B-BCOR fusion
Epithelioid hemangioendothelioma t(1;3)(p36;q25) WWTR1-CAMTA1 fusion
t(X;11)(p11;q22) YAP1-TFE3 fusion
Epithelioid sarcoma Deletion 22q SMARCB1 inactivation
t(8;22)(q22;q11)
t(10;22)
Ewing sarcoma t(11;22)(q24;q12) EWSR1-FLI1 fusion
t(21;22)(q12;q12) EWSR1-ERG fusion
Others Others
EWSR1/FUS-NFATC2 fusion sarcoma t(20;22)(q13.2;q12.2) EWSR1-NFATC2 fusion
t(16;20)(p11;q13.2) FUS-NFATC2 fusion
Extraskeletal myxoid chondrosarcoma t(9;22)(q22;q12) EWSR1-NR4A3 fusion
t(9;17)(q22;q11) TAF2N-NR4A3 fusion
t(9;15)(q22;q21) TCF12-NR4A3 fusion
t(3;9)(q11;q22) TFG-NR4A3 fusion
t(9;17)(q22;q11) RBP56-NR4A3 fusion
GIST Deletion 14q, 22q, 1p, 15q KIT or PDGFRA mutation (85% of cases); (NF1, SDH, BRAF mutation, SDHC
hypermethylation, other)

TABLE 2. Select Sarcomas With Diagnostically Relevant Distinct Cytogenetic and Molecular Alterations (Continued)
Sarcoma Type Cytogenetic Alteration Molecular Alteration
Infantile fibrosarcoma t(12;15)(p13;q25) ETV6-NTRK3 fusion
t(2;15)(p21;q25) EML4-NTRK3 fusions
Inflammatory myofibroblastic tumor t(1;2)(q22;p23) TPM3-ALK fusion
t(2;19)(p23;p13) TPM4-ALK fusion
t(2;17)(p23;q23) CLTC-ALK fusion
t(2;2)(p23;q13) RANBP2-ALK fusion
t(2;2)(p23;q35) ATIC-ALK fusion
t(2;11)(p23;p15) CARS-ALK fusion
t(2;4)(p23;q21) SEC31L1-ALK fusion
t(2;12)(p23;p12) PPFIBP1-ALK fusion
t(3;6)(q12;q22) TFG-ROS1 fusion
t(6;17)(q22;p13)? YWHAE-ROS1 fusion
RRBP1-ALK fusion
Low-grade fibromyxoid sarcoma t(7;16)(q33;p11) FUS-CREB3L2 fusion
t(11;16)(p11;p11) FUS-CREB3L1 fusion
Mesenchymal chondrosarcoma t(8;8)(q13;q21) HEY1-NCOA2 fusion
MPNST Complex changes SUZ12 or EED mutation (80% of cases), NF1 inactivation
Myoepithelial tumor/carcinoma of soft t(6;22)(p21;q12) EWSR1-POU5F1 fusion
tissue
t(19;22)(q13;q12) EWSR1-ZNF444 fusion
t(1;22)(q23;q12) EWSR1-PBX1 fusion
Myxoid liposarcoma t(12;16)(q13;p11) FUS-DDIT3 fusion
t(12;122)(q13;q12) EWSR1-DDIT3 fusion
Postradiation angiosarcoma Gain 8q24 MYC amplification
Sclerosing epithelioid fibrosarcoma t(7;16)(q33;p11) FUS-CREB3L2 fusion
Solitary fibrous tumor Inv(12)(q13q13) NAB2-STAT6 fusion
Synovial sarcoma t(X;18)(p11;q11) SS18-SSX1/SSX2 fusion
Abbreviations: DFSP, dermatofibrosarcoma protuberans; GIST, gastrointestinal stromal tumor; MPNST, malignant peripheral nerve sheath tumor.
Pseudomyogenic hemangioendothelioma (PHE), another cutaneous in location.52-54 Presenting as small erythem-

vascular neoplasm of intermediate biologic potential, was atous to violaceous papules, nodules, or large plaques with
recently found to harbor recurrent SERPINE1-FOSB fusion skin discoloration, postradiation angiosarcoma infiltrates
resulting from t(7;19)(q22;q13)47; rare cases of PHE with the reticular dermis, often with subtle radial extension of
alternate ACTB-FOSB48 fusion have been described.FOSB individual neoplastic vessels at a distance from the primary
rearrangement leading to FOSB overexpression can be lesion and deeper infiltration into subcutis. In contrast to
detected by recently introduced FOSB immunohisto- conventional mammary angiosarcoma, which arises in
chemistry, which is positive in 96% of cases of PHE breast parenchyma, postradiation angiosarcoma of the
(Table 1).49 However, subsets of other vascular neoplasms skin is characterized by high-level MYC amplification55,56
may harbor FOSB rearrangement, such as the “cellular (Fig. 3G–I), which can be detected by immunohisto-
variant” of epithelioid hemangioma (i.e., ZFP36-FOSB or chemical staining for MYC and by genomic evidence of
WWTR1-FOSB),50,51 with positive FOSB staining demon- high-level copy number gain at 8q24.21. Diffuse nu-
strated in approximately 50% of cases.49 clear MYC expression in postradiation angiosarcoma
Secondary postradiation angiosarcoma of the breast de- enables distinction from atypical postradiation vascular
velops with a median latent interval of 5 to 6 years after proliferation and can be particularly useful in cases of
radiation following breast-conserving surgery with in- postradiation angiosarcoma with deceptively bland
creasing incidence over the past years and is mostly cytomorphology.

FIGURE 3. Mesenchymal Neoplasms With Distinct Cytogenetic Aberrations and Associated Immunohistochemical Markers
Inflammatory myofibroblastic tumor is characterized by a population of spindled tumor cells arranged in fascicles with scattered inflammatory cells (A) and
expression of ALK in tumor cells (B). Rearrangement of the ALK locus at 2p23 can be detected by fluorescence in situ hybridization (C; break-apart signal
indicated by arrow). Malignant epithelioid hemangioendothelioma (EHE) consisting of strands of tumor cells with epithelioid morphology, nuclear atypia,
and glassy amphophilic cytoplasm embedded in a myxohyaline to collagenous stroma (D). Most cases of EHE have diffuse nuclear expression of CAMTA1
(E) resulting from WWTR1-CAMTA1 fusion. In this case, next-generation sequencing detected rearrangement involving the WWTR1 coding as evidenced
by split reads (F, arrow) that match to CAMTA1 (not shown). Postradiation angiosarcoma of the breast consists of atypical endothelial cells growing in
strands and sheets and diffusely infiltrating preexisting fat (G) is characterized by nuclear expression of MYC (H) resulting from high-level MYC am-
plification (I, arrow).

Newly Emerging Entities and Predictive Biomarkers recognized (Fig. 4E and F); however, their biologic potential,
With increased use of molecular analyses in the diagnostic clinical course, and potential for response to inhibitors of the
work-up of soft-tissue tumors, newly discovered genomic TRK family of kinases, such as larotrectinib and entrectinib,
aberrations have led to the identification of novel entities which have recently been U.S Food and Drug Administration-
defined by distinct gene fusions: these include NTRK- approved for the treatment of NTRK-rearranged solid
rearranged spindle cell neoplasms57,58 and EWSR1- tumors,66-70 remain to be defined.
SMAD3–positive fibroblastic tumors, 59,60 which are be-
RADIATION ONCOLOGY: ADVANCEMENTS AND NEW
ing included as “emerging entities” in the upcoming
APPROACHES FOR THE TREATMENT OF SARCOMA
World Health Organization Classification of Tumors.
However, biologic potential and clinical implications of RT in the Neoadjuvant or Adjuvant Setting for Extremity
any newly reported entity defined by a recurrent genomic and Trunk Soft-Tissue Sarcoma
event remain to be defined. As exemplified by the no- Here, we focus on the use of RT in the treatment of extremity
torious “promiscuity” of EWSR1 and ALK as common and trunk soft-tissue sarcoma, a common scenario for the
fusion partners, many genetic alterations initially con- use of RT in sarcoma care. However, the concepts dis-
sidered “tumor-specific” or even “disease-defining” evolve to cussed here apply to RT for the treatment of sarcoma in
be quite nonspecific over time and must be carefully eval- other scenarios, such as the neoadjuvant treatment of
uated before they can add diagnostic, predictive, or prog- retroperitoneal sarcoma, treatment of oligometastatic dis-
nostic value to existing classification schemes. ease, and in palliative settings.
As an example, various solid tumors (including rare benign Despite early proclamations that sarcoma was a “radio-
and malignant mesenchymal neoplasms) harbor fusions of resistant” tumor type, RT has been a part of the care of
NTRK1, NTRK2, or NTRK3 encoding neurotrophic tyrosine patients with sarcoma for nearly a century.71 RT began to
kinases NTRK1-3 (i.e., TRKA-C), which are generally be- take hold as a critical adjuvant therapy for primary soft-
lieved to be mutually exclusive with other genomic aber- tissue sarcoma in the second half of the 20th century along
rations. A classic example of a mesenchymal neoplasm with with conservative surgery.72 Ultimately, randomized evi-
canonical NTRK rearrangement is infantile fibrosarcoma, dence emerged to illustrate that limb salvage surgery
a pediatric spindle cell sarcoma with characteristic ETV6- combined with RT could provide a less morbid alternative to
NTRK3 fusion in most cases and rare alternate EML4- amputation for extremity soft-tissue sarcoma.73
NTRK3 fusion (Fig. 4A and B; Table 2).61,62 A recently
At that time, RT was delivered to patients in what today
introduced “pan-TRK” immunohistochemical stain has
would be considered medieval fashion. The target area for
been shown to be highly sensitive63 but not entirely specific
RT was delineated clinically or using two-dimensional ra-
for tumors with NTRK fusions and can be expressed, for
diographs, without any refined capacity to control the dis-
instance, in ALK-rearranged tumors.
tribution of the radiation dose, avoid or reduce radiation
The testing approach for detection of NTRK fusions in dose to nearby organs at risk, maximize radiation dose to the
sarcoma remains to be determined. A recent retrospective target structures, or account for daily variability in treatment
analysis evaluated the performance of immunohistochem- setup. Two major advances in technology in the late 20th
istry and DNA-based NGS to detect NTRK fusions relative to and early 21st century revolutionized RT for all disease sites
RNA-based NGS.64 Among a total of 33,997 patients, the including sarcoma.
authors identified 87 patients with oncogenic NTRK1-3
The major advances in RT technology have been the use of
fusions in solid tumors.64 The reported sensitivity and
more refined imaging for treatment planning and image
specificity for detection of NTRK fusions were 81.1% and
guidance prior to daily treatments. Radiation treatment
99.9% for DNA-based sequencing and 87.9% and 81.1%
plans are now based on three-dimensional imaging of tumor
for immunohistochemistry, respectively.64 Specifically, im-
and normal anatomy. In some cases, four-dimensional
munohistochemistry showed 96% to 100% sensitivity for
imaging is used to account for changes in anatomy over
fusions of NTRK1 and NTRK2 but only 79% sensitivity for
time (e.g., during the respiratory cycle). With this imaging,
NTRK3. Both sensitivity and specificity were found to be
radiation treatment plans can account for differences in the
poor in sarcomas.64
densities of patients’ tissues (air, soft tissue, bone) that
Lipofibromatosis-like neural tumor, which harbors mor- directly affect the distribution of radiation dose. Further-
phologic resemblance to lipofibromatosis but exhibits locally more, radiation target volumes can be defined more ac-
aggressive behavior in children and young adults, has re- curately, minimizing unnecessary radiation dose to normal
current NTRK1 rearrangement65 and expresses pan-TRK tissue. Furthermore, just prior (or even during) each radi-
(Fig. 4C and D).63 In addition, rare cases of unclassified ation treatment, two-dimensional and/or three-dimensional
sarcomas with positive pan-TRK staining are increasingly imaging of a patient can be acquired to confirm patient and

primary tumor. Predictably, important late toxicities related

to radiation of the extremity (joint stiffness, fibrosis, and
edema) occurred in 5% or less of the patients, compared
with rates between 15% and 30% in patients using older
techniques without image guidance.75
The second key advance in the technology of RT was the
introduction of intensity-modulated radiation therapy (IMRT).
With conventional three-dimensional conformal RT, beams
are arranged to maximize coverage of the target structure
while minimizing radiation dose to surrounding organs at risk.
The radiation dose from each beam can be modulated to
a certain extent, but this process is limited by human capacity
for calculation. With IMRT, the modulation of each radiation
beam can be performed by computer-based calculations,
which are derived from constraints to normal structures and
organs at risk determined and prioritized a priori by the ra-
diation oncologist. Thus, radiation oncologists have greater
ability to control the dose of radiation from a particular ra-
diation beam over time, resulting in improved accuracy and
homogeneity of the radiation dose with respect to target
volumes and improved avoidance of normal tissues.
In a phase II study of preoperative image-guided RT for soft-
tissue sarcoma exclusively using IMRT, reconstructive tis-
sue flaps were less frequently required to manage wound
complications, and late toxicities (joint stiffness, fibrosis,
edema) were improved compared with historical rates using
conventional three-dimensional conformal RT tech-
niques. 76 Subsequent retrospective data have further
supported low late toxicity rates with the use of IMRT as well
as potentially lower local recurrence rates.77,78 According to
FIGURE 4. Examples of Mesenchymal Neoplasms Showing Pan-TRK the National Cancer Database, the use of IMRT in the
Expression by Immunohistochemistry neoadjuvant or adjuvant treatment of soft-tissue sarcoma
Infantile fibrosarcoma comprises monotonous population of spindle cells has risen sharply over the past 10 years (V. Reddy, et al,
(A) with ETV6-NTRK3 rearrangement, detected by ETV6 break-apart unpublished data, 2019).
fluorescence in situ hybridization (A, inset, arrows), and shows diffuse
staining by pan-TRK immunohistochemistry (B). Lipofibromatosis-like To maximize the benefit of image-guided RT and IMRT,
neural tumor with spindled tumor cells lacking nuclear atypia containing there has been increased use of preoperative compared
wavy neural-like nuclei and eosinophilic cytoplasm with diffuse in- with postoperative RT for extremity and trunk soft-tissue
filtration of adjacent fat (C) shows diffuse pan-TRK staining (D). Un- sarcoma. Although randomized data provided evidence for
classified sarcoma of the endocervix comprising ovoid to spindled tumor reduced late toxicities with preoperative RT, there has also
cells with nuclear atypia scattered in a collagenous stroma with prom- been a competing concern for an increase in perioperative
inent hyalinized blood vessels (E) showing positive pan-TRK staining (F) wound complications in patients receiving preoperative RT.
suggestive of underlying NTRK rearrangement.
Improvements in image guidance and IMRT have provided
further impetus to shift practice patterns toward the use of
target positioning and ensure treatment accuracy within as preoperative RT, where the presence of a tumor targe-
few as 2 to 3 mm of error. t allows for more discrete target delineation to maximize the
benefit of these technologies.79 Still, there are advantages to
RTOG 0630 was an early example of the potential for image- both preoperative and postoperative approaches that must
guided RT to improve the care of patients with sarcoma.74 be considered on a case-by-case basis (Table 3).
This multi-institutional phase II study was designed to as-
sess the frequency and severity of late toxicities in patients Stereotactic RT for the Treatment of Oligometastatic or
receiving preoperative RT for extremity soft-tissue sarcoma. Oligoprogressive Sarcoma
With the availability of image guidance, RT was delivered to Another common scenario for the use of RT in sarcoma that
a reduced volume using smaller margins around the has benefited from advances in technology has been the

TABLE 3. Comparison of Preoperative and Postoperative RT for Soft- after 24 and 43 months of follow-up, respectively.80,81 In
Tissue Sarcoma a separate study of metastatic sarcoma to the spine, the
Advantage of Preoperative RT Advantage of Postoperative RT actuarial rate of local control among surviving patients
Reduced late toxicities (e.g., Pathologic evaluation of an treated with SRS or SBRT was over 85%.82 Another
fibrosis, lymphedema, joint untreated specimen in cases of consideration is that these are noninvasive procedures
stiffness) for extremity tumors diagnostic uncertainty performed on an outpatient basis, are typically well tol-
Smaller RT volumes Reduced wound-complication erated (depending on the anatomic location of the target
rate lesion), and can be safely combined with some systemic
Shorter course of RT (5 vs. 6 Pathologic evaluation of tumor therapies. However, radiation oncologists should always
weeks) extent to define RT volume discuss less common but serious toxicities, including ra-
Maximize benefits of image- Earlier removal of tumor can diation neuritis or plexopathy (nerve injury resulting in pain
guided RT due to presence of reduce patient stress/anxiety syndrome, weakness, and/or muscle atrophy), radiation
tumor target
pneumonitis, or radiation-induced bowel injury (ulcera-
Abbreviation: RT, radiotherapy. tion, bleeding, perforation). Ultimately, the appropriate
tumor-directed therapy (e.g., metastasectomy, interven-
setting of oligometastatic or oligoprogressive disease, in tional ablation) should be considered in a multidisciplinary
which a patient may have a small number of progressing or setting and on a case-by-case basis.
metastatic lesions that can be addressed with tumor- MRI-Guided RT
directed therapies. The lung and spine are two common
areas of metastases arising in patients with primary soft- Most modern radiation planning is based on CT scans of the
tissue and bone sarcoma. Although systemic therapy (e.g., target area. Although some soft-tissue sarcomas are well
chemotherapy, targeted therapy, or immunotherapy) is the visualized on CT, others can be difficult to localize without
mainstay of overall disease control in the metastatic setting, the improved soft-tissue resolution of MRI.83 One approach
individual tumor lesions may be poorly controlled by sys- to circumvent this issue has been to obtain diagnostic MRI
temic therapy and may require additional tumor-directed that can be fused with CT-based treatment planning scans.
therapy (oligoprogressive disease). As improvements in This has limited utility because of the difficulty of obtaining
systemic therapies allow patients to live longer with disease, diagnostic imaging with the patient in the precise setup used
tumor-directed therapy can allow for a much-needed hiatus for treatment delivery, thus making the fusion of MRI and CT
from the toxicities of systemic therapies. In other cases, scans imprecise. The adoption of MRI-based simulation, in
where patients develop oligometastatic disease after a long which patients undergo MRI-based imaging in the treat-
disease-free interval, local therapies can obviate the need ment position, can facilitate fusion with CT-based planning
for immediate systemic therapy. scans and improve accuracy of target delineation.84 MRI-
only treatment planning may also be a future alternative,
Several options are available for tumor-directed therapy. circumventing CT planning entirely.85
Among these, stereotactic body RT (SBRT) and stereotactic
surgery (SRS) are two modern RT technologies that have Identification of the tumor on daily image guidance prior to
been widely adopted and refined for the treatment of pa- treatment delivery is also essential for accurate treatment; in
tients with metastatic disease of all histologies over the past some cases, the tumor is not visible using the on-board
20 years. SRS and SBRT refer to cases where a high dose of cone-beam CT scans used for setup verification. In these
RT is delivered per fraction (treatment) to a tumor target in cases, MRI-guided RT allows for imaging of patients prior to
one (SRS) or up to five (SBRT) total fractions (treatments). and during delivery of RT.86 Imaging just prior to treatment
SRS and SBRT are predicated on the use of the advanced delivery ensures accurate treatment delivery for fixed le-
immobilization devices to ensure accuracy of patient setup sions. In cases of mobile tumors that are susceptible to
reproducibility, high-resolution imaging for target de- respiratory motion, such as those in the abdomen or thorax
lineation, optimal image guidance for treatment delivery, near the diaphragm or even cardiac tumors susceptible to
and advanced treatment planning techniques (e.g., inverse cardiac motion, cine imaging of tumors during RT can allow
planning). for target tracking and better accuracy throughout the
treatment delivery. As MRI-based linear accelerators be-
Because of the unique biology of delivering high radiation come more widely adopted, their value in selected cases of
doses per fraction, the antitumor effect of SRS or SBRT is soft-tissue sarcoma treatment will be difficult to ignore (Fig.
potent and can result in long-term disease control or cure of 5).
the irradiated tumor. In two separate studies of patients with
metastatic high-grade sarcoma in the lung treated with Proton Therapy
SBRT at two separate institutions, over 86% and 94% of Proton-based RT is an alternative approach to delivery ra-
irradiated tumors (25 patients and 39 patients) were controlled diation. Protons are heavy, positively charged particles with

distinct physical characteristics compared with photons, The use of proton therapy is particularly encouraged in
which are particles more commonly used to deliver RT and pediatric tumors, including pediatric sarcoma, to limit pu-
have negligible mass and no charge. Although proton tative long-term effects of low- and medium-dose exposure
therapy has been available regionally at select institutions to nearby normal structures. Specific use cases include
for several decades, it has emerged as an alternative to rhabdomyosarcoma, especially those arising in the head
conventional photon-based RT over the past 15 years. This and neck, including parameningeal and orbital rhabdo-
is the result of an advance in proton therapy that allowed its myosarcoma, where protons may allow sparing of the optic
delivery on advanced isocentric gantry systems, resulting in apparatus, central nervous system structures, and salivary
more flexibility for its use. Proton therapy provides an ad- glands.89,90 Other pediatric sarcoma cases where proton
vantage to photon therapy in that the dose of RT has a finite therapy may provide an advantage include osteosarcoma of
penetration in tissue and delivers the majority of its energy at the spine or skull base91 and pelvic Ewing sarcoma.92
a specific depth window. As a result, in certain scenarios,
proton therapy may be able to reduce the exposure of low to In some classically radioresistant tumors, including chor-
medium doses of RT in the vicinity of the tumor target. doma and chondrosarcoma, the putative higher biologic
Furthermore, pound-for-pound, proton therapy may also potency of proton therapy may result in improved tumor
deliver a slightly higher biologic effect compared with control.93 Given the location of these tumors in the spine and
photons.87 skull base, the radiation dose characteristics of proton beam
therapy are important in sparing the optic apparatus and
This advantage is particularly relevant when the tumor target other central nervous system structures. It should be noted
is located on one side of a critical normal structure. For
that all proton therapy is not created equal; more advanced
example, tumors located posterior to the spinal cord and
delivery using pencil beam scanning and intensity-
anterior to the mediastinum may maximally benefit from the
modulated proton beam therapy provide the most con-
use of proton therapy, which could significantly reduce
formal therapy and the greatest potential therapeutic
radiation dose to the structures immediately beyond the
advantage.
tumor target (e.g., spinal cord, mediastinum). As another
example, proton therapy has been proposed as an alter- Thus, although there is a paucity of clinical data at this point
native approach to deliver dose-escalated RT to the high- to support the superiority of proton therapy in terms of
risk margin of retroperitoneal sarcoma due to its ability to reduced late toxicities or improved disease outcomes, the
spare tissue immediately adjacent small bowel.88 accumulation of clinical case series supports the selected
FIGURE 5. MRI-Guided Radio-

therapy for Sarcoma Liver
Metastasis
Radiation dose distribution (top
panel) for a patient receiving
magnetic resonance (MR)-guided
stereotactic body radiotherapy for
oligometastatic spindle cell sar-
coma of the liver. A 172s balance
steady-state free precession MR
sequence was used to acquire
three-dimensional anatomic MRI
with 1.5 mm isotropic resolution
using MRIdian’s 0.35T on-board
MRI. There are distinct advan-
tages of MR-guided radiotherapy
for sarcoma at simulation, treat-
ment planning, and patient setup
prior to treatment delivery. In this
case, the target lesion was not
identifiable on CT imaging. The
ability to visualize the tumor just
prior to treatment allowed safe
reduction of the planning margin from 1 cm to 5 mm, reducing radiation dose to surrounding normal tissue. Given the challenging treatment setup for
sarcoma extremity lesions, the added imaging allowed more reproducible patient positioning. From left to right in both top and bottom panels: axial, sagittal,
and coronal MR images. Isodose colors (top panel): red, 99%; green, 50%, light blue, 33%.

use of proton therapy on a case-by-case scenario for the radiation can reportedly exceed 95%.96 A multi-institutional
treatment of patients with sarcoma. study evaluating radiation dose reduction in the pre-
operative treatment of myxoid liposarcoma is ongoing
Active Investigations to Improve RT as a Therapeutic (Table 4).
Modality for Sarcoma
How Technology Can Fill the Gap
Condensed radiation treatment regimens (SBRT and SRS)
for metastatic sarcoma and other cancers are common- Technologic advances will be necessary to continue to
place. However, the standard neoadjuvant or adjuvant improve the impact of RT on patients with sarcoma over the
treatment of primary soft-tissue sarcoma involves a 5- to 6- next few decades. In particular, better understanding of the
week course of daily radiation that takes place Monday biology of radiation responses in tumor and normal tissue
through Friday. This is a burdensome therapy that is lo- will permit the personalization of RT with respect to toxic-
gistically and socioeconomically burdensome for patients, ities, secondary malignancies, and disease control. Bio-
especially those who wish to be treated at tertiary sarcoma markers to identify a priori patients most at risk for
centers (where there is an association with improved out- RT-associated toxicities, such as wound complications or
comes), but these centers are often inconveniently located late extremity complications, or even RT-associated ma-
at a greater distance than community oncology practices. lignancies will improve the therapeutic window. Tailoring
radiation dose and volumes to tumor subtypes—as defined
Shorter treatment regimens may soon challenge this
not only by histology but also by biology—will further amplify
treatment paradigm. For decades, a neoadjuvant ap-
the potential impact of RT on patients with sarcoma.
proach using an eight-treatment neoadjuvant radiation
regimen in combination with neoadjuvant chemotherapy CONCLUSIONS
has resulted in acceptable (but not ideal) local control and
The pathologic work-up of sarcomas relies mainly on the
toxicity.94 More recently, we evaluated a condensed 5-day
thoughtful integration of clinical information (i.e., patient
radiation regimen for primary high-risk soft-tissue sarcoma
age, sex), anatomic location, radiologic features, gross and
of the trunk and extremity in a single-institution phase II
histomorphologic appearances, as well as immunohisto-
study.95 Early toxicity and disease control outcomes were
chemical staining and ancillary cytogenetic and molecular
encouraging, and the 5-day approach also improved uti-
genetic analyses. With the development of targeted ther-
lization and access of neoadjuvant RT at our high-volume
apies directed against distinct oncogenic aberrations and
sarcoma center. Other studies using shorter RT regimens
tumor-specific signatures that predict sensitivity to im-
(5–15 fractions) for soft-tissue sarcoma are ongoing
munotherapy, the accurate and timely identification of
(Table 4).
subsets of patients most likely to benefit from systemic
Traditionally, radiation for soft-tissue sarcoma has been therapies gains importance. Whereas some cytogenetic/
agnostic of histologic subtype, but this is also poised to molecular genetic aberrations or immunohistochemical
change. Myxoid liposarcoma has long been known to be markers are considered tumor specific, others can be
a well-defined histologic subtype with a unique chromo- present in various benign and malignant mesenchymal
somal aberration and a clearly radiosensitive phenotype. neoplasms and require careful evaluation in the diagnostic
Local control for myxoid liposarcoma after surgery and context.
TABLE 4. Active Clinical Trials Using Modified Radiation Regimens in Preoperative or Postoperative Treatment of STS
NCT No. Study Title Patients Radiotherapy Regimen
NCT02106312 Dose Reduction of Preoperative Radiotherapy in Myxoid Liposarcomas Adults with myxoid 2 Gy 18 fractions = 36 Gy total (3.5
(DOREMY) liposarcoma weeks)
NCT02701153 Phase II Study of 5-Day Hypofractionated Preoperative Radiation Adults with extremity 5-6 Gy 5 fractions = 30 Gy (1 week)
Therapy for Soft Tissue Sarcomas: Expansion Cohort and trunk STS
NCT02634710 Hypofractionated Pre-operative Radiation Therapy for Soft Tissue Adults with extremity 7 Gy 5 fractions = 35 Gy (2 weeks)
Sarcomas of the Extremity and Chest-wall and chest-wall STS
NCT03819985 Shorter Course, Hypofractionated Pre-Surgery Radiation Therapy in Adults with extremity 2.85 Gy 15 fractions = 42.75 Gy (3
Treating Patients With Localized, Resectable Soft Tissue Sarcoma of and trunk STS weeks)
the Extremity of Superficial Trunk
NCT02565498 Preoperative vs. Postoperative IMRT for Extremity/Truncal STS Adults with extremity 2 Gy 25 fractions = 50 Gy
and trunk STS preoperatively or postoperatively (5
weeks)
Abbreviations: IMRT, intensity-modulated radiation therapy; STS, soft-tissue sarcoma.

Despite improvements in these diagnostic tools and the discoveries will lead to refined classification systems and
expanding armamentarium of systemic therapies, ef- novel therapeutic approaches that, in the context of
fective local therapy involving surgery and radiation a multidisciplinary approach, will continue to raise the bar
therapy remains the primary curative approach. Ad- for the care of patients with sarcoma.
vances in image guidance, intensity modulation, and
proton beam technology have expanded the therapeutic ACKNOWLEDGMENT
window of radiation therapy for local disease. Further- Dr. Schaefer and Dr. Kalbasi contributed equally to this
more, for limited-burden metastatic disease refractory to article. The authors thank Dr. Christopher D. M. Fletcher,
systemic therapy, stereotactic approaches offer a reli- MD, FRCPath, Department of Pathology, Brigham and
able, noninvasive and well-tolerated tool for local disease Women’s Hospital, for sharing representative cases; and the
control. Center for Advanced Molecular Diagnostics and Division of
We await the outcomes of ongoing efforts in the translational Cytogenetics, Department of Pathology, Brigham and
research setting because we anticipate these will further Women’s Hospital, and Dr. Yingli Yang, PhD, Department of
promote discoveries that expand our understanding of Radiation Oncology, University of California Los Angeles for
sarcoma biology, development, and progression. These providing select images.

1
Department of Pathology, Brigham and Women’s Hospital/Harvard Anusha Kalbasi, MD, Department of Radiation Oncology, UCLA, B-262
Medical School, Boston, MA Factor Building, 700 Tiverton Avenue, Los Angeles, CA 90095; Twitter:
2
Division of Vascular & Interventional Radiology, Johns Hopkins @AnushaKalbasiMD; email: [email protected].
University, School of Medicine, Baltimore, MD
3
Division of Molecular and Cellular Oncology, Department of Radiation
Oncology, Jonsson Comprehensive Cancer Center Sarcoma Program,
University of California Los Angeles, Los Angeles, CA
EDBK_280729.
REFERENCES
1. Fletcher C, Bridge JA, Hogendoorn PCW, et al. In Fletcher CDM, (ed). WHO Classification of Tumours of Soft Tissue and Bone. Lyon: IARC Press; 2013.
3. Dickson BC, Swanson D. Targeted RNA sequencing: a routine ancillary technique in the diagnosis of bone and soft tissue neoplasms. Genes Chromosomes
Cancer. 2019;58:75-87.
4. Zou C, Smith KD, Liu J, et al. Clinical, pathological, and molecular variables predictive of malignant peripheral nerve sheath tumor outcome. Ann Surg. 2009;
249:1014-1022.
5. Lee W, Teckie S, Wiesner T, et al. PRC2 is recurrently inactivated through EED or SUZ12 loss in malignant peripheral nerve sheath tumors. Nat Genet. 2014;
46:1227-1232.
6. De Raedt T, Beert E, Pasmant E, et al. PRC2 loss amplifies Ras-driven transcription and confers sensitivity to BRD4-based therapies. Nature. 2014;514:247-251.
7. Prieto-Granada CN, Wiesner T, Messina JL, et al. Loss of H3K27me3 expression is a highly sensitive marker for sporadic and radiation-induced MPNST. Am
J Surg Pathol. 2016;40:479-489.
8. Schaefer IM, Fletcher CD, Hornick JL. Loss of H3K27 trimethylation distinguishes malignant peripheral nerve sheath tumors from histologic mimics. Mod Pathol.
2016;29:4-13.
9. Cleven AH, Sannaa GA, Briaire-de Bruijn I, et al. Loss of H3K27 tri-methylation is a diagnostic marker for malignant peripheral nerve sheath tumors and an
indicator for an inferior survival. Mod Pathol. 2016;29:582-590.
10. Jo VY, Fletcher CD. Epithelioid malignant peripheral nerve sheath tumor: clinicopathologic analysis of 63 cases. Am J Surg Pathol. 2015;39:673-682.
11. Schaefer IM, Dong F, Garcia EP, et al. Recurrent SMARCB1 inactivation in epithelioid malignant peripheral nerve sheath tumors. Am J Surg Pathol. 2019;
43:835-843.
12. Hollmann TJ, Hornick JL. INI1-deficient tumors: diagnostic features and molecular genetics. Am J Surg Pathol. 2011;35:e47-e63.
13. Jo VY, Fletcher CDM. SMARCB1/INI1 loss in epithelioid schwannoma: a clinicopathologic and immunohistochemical study of 65 cases. Am J Surg Pathol. 2017;
41:1013-1022.
14. Hornick JL, Dal Cin P, Fletcher CD. Loss of INI1 expression is characteristic of both conventional and proximal-type epithelioid sarcoma. Am J Surg Pathol. 2009;
33:542-550.

15. Italiano A, Sung YS, Zhang L, et al. High prevalence of CIC fusion with double-homeobox (DUX4) transcription factors in EWSR1-negative undifferentiated small
blue round cell sarcomas. Genes Chromosomes Cancer. 2012;51:207-218.
16. Antonescu CR, Owosho AA, Zhang L, et al. Sarcomas with CIC-rearrangements are a distinct pathologic entity with aggressive outcome: a clinicopathologic and
molecular study of 115 cases. Am J Surg Pathol. 2017;41:941-949.
17. Sugita S, Arai Y, Tonooka A, et al. A novel CIC-FOXO4 gene fusion in undifferentiated small round cell sarcoma: a genetically distinct variant of Ewing-like
sarcoma. Am J Surg Pathol. 2014;38:1571-1576.
18. Solomon DA, Brohl AS, Khan J, et al. Clinicopathologic features of a second patient with Ewing-like sarcoma harboring CIC-FOXO4 gene fusion. Am J Surg Pathol.
2014;38:1724-1725.
19. Specht K, Sung YS, Zhang L, et al. Distinct transcriptional signature and immunoprofile of CIC-DUX4 fusion-positive round cell tumors compared to EWSR1-
rearranged Ewing sarcomas: further evidence toward distinct pathologic entities. Genes Chromosomes Cancer. 2014;53:622-633.
20. Hung YP, Fletcher CD, Hornick JL. Evaluation of ETV4 and WT1 expression in CIC-rearranged sarcomas and histologic mimics. Mod Pathol. 2016;
29:1324-1334.
21. Yoshida A, Sekine S, Tsuta K, et al. NKX2.2 is a useful immunohistochemical marker for Ewing sarcoma. Am J Surg Pathol. 2012;36:993-999.
22. Hung YP, Fletcher CD, Hornick JL. Evaluation of NKX2-2 expression in round cell sarcomas and other tumors with EWSR1 rearrangement: imperfect specificity
for Ewing sarcoma. Mod Pathol. 2016;29:370-380.
23. Shibuya R, Matsuyama A, Nakamoto M, et al. The combination of CD99 and NKX2.2, a transcriptional target of EWSR1-FLI1, is highly specific for the diagnosis of
Ewing sarcoma. Virchows Arch. 2014;465:599-605.
24. Pierron G, Tirode F, Lucchesi C, et al. A new subtype of bone sarcoma defined by BCOR-CCNB3 gene fusion. Nat Genet. 2012;44:461-466.
25. Kao YC, Owosho AA, Sung YS, et al. BCOR-CCNB3 fusion positive sarcomas: a clinicopathologic and molecular analysis of 36 cases with comparison to
morphologic spectrum and clinical behavior of other round cell sarcomas. Am J Surg Pathol. 2018;42:604-615.
26. Specht K, Zhang L, Sung YS, et al. Novel BCOR-MAML3 and ZC3H7B-BCOR gene fusions in undifferentiated small blue round cell sarcomas. Am J Surg Pathol.
2016;40:433-442.
27. Cohen-Gogo S, Cellier C, Coindre JM, et al. Ewing-like sarcomas with BCOR-CCNB3 fusion transcript: a clinical, radiological and pathological retrospective study
from the Société Française des Cancers de L’Enfant. Pediatr Blood Cancer. 2014;61:2191-2198.
28. Kao YC, Sung YS, Zhang L, et al. BCOR overexpression is a highly sensitive marker in round cell sarcomas with BCOR genetic abnormalities. Am J Surg Pathol.
2016;40:1670-1678.
29. Coffin CM, Hornick JL, Fletcher CD. Inflammatory myofibroblastic tumor: comparison of clinicopathologic, histologic, and immunohistochemical features
including ALK expression in atypical and aggressive cases. Am J Surg Pathol. 2007;31:509-520.
30. Lawrence B, Perez-Atayde A, Hibbard MK, et al. TPM3-ALK and TPM4-ALK oncogenes in inflammatory myofibroblastic tumors. Am J Pathol. 2000;
157:377-384.
31. Morris SW, Kirstein MN, Valentine MB, et al. Fusion of a kinase gene, ALK, to a nucleolar protein gene, NPM, in non-Hodgkin’s lymphoma. Science. 1994;
263:1281-1284.
32. Pan Z, Hu S, Li M, et al. ALK-positive large B-cell lymphoma: a clinicopathologic study of 26 cases with review of additional 108 cases in the literature. Am J Surg
Pathol. 2017;41:25-38.
33. Mano H. Non-solid oncogenes in solid tumors: EML4-ALK fusion genes in lung cancer. Cancer Sci. 2008;99:2349-2355.
34. Chou A, Fraser S, Toon CW, et al. A detailed clinicopathologic study of ALK-translocated papillary thyroid carcinoma. Am J Surg Pathol. 2015;39:652-659.
35. Mariño-Enrı́quez A, Ou WB, Weldon CB, et al. ALK rearrangement in sickle cell trait-associated renal medullary carcinoma. Genes Chromosomes Cancer. 2011;
50:146-153.
36. Dickson BC, Swanson D, Charames GS, et al. Epithelioid fibrous histiocytoma: molecular characterization of ALK fusion partners in 23 cases. Mod Pathol. 2018;
31:753-762.
37. Kiuru M, Jungbluth A, Kutzner H, et al. Spitz tumors: comparison of histological features in relationship to immunohistochemical staining for ALK and NTRK1. Int
J Surg Pathol. 2016;24:200-206.
38. Perron E, Pissaloux D, Charon Barra C, et al. Melanocytic myxoid spindle cell tumor with ALK rearrangement (MMySTAR): report of 4 cases of a nevus variant with
potential diagnostic challenge. Am J Surg Pathol. 2018;42:595-603.
39. Davis LE, Nusser KD, Przybyl J, et al. Discovery and characterization of recurrent, targetable ALK fusions in leiomyosarcoma. Mol Cancer Res. 2019;17:676-685.
40. Le Loarer F, Cleven AHG, Bouvier C, et al. A subset of epithelioid and spindle cell rhabdomyosarcomas is associated with TFCP2 fusions and common ALK
upregulation. Mod Pathol. 2020;33:404-419.
41. Mendlick MR, Nelson M, Pickering D, et al. Translocation t(1;3)(p36.3;q25) is a nonrandom aberration in epithelioid hemangioendothelioma. Am J Surg Pathol.
2001;25:684-687.
42. Errani C, Zhang L, Sung YS, et al. A novel WWTR1-CAMTA1 gene fusion is a consistent abnormality in epithelioid hemangioendothelioma of different anatomic
sites. Genes Chromosomes Cancer. 2011;50:644-653.
43. Tanas MR, Sboner A, Oliveira AM, et al. Identification of a disease-defining gene fusion in epithelioid hemangioendothelioma. Sci Transl Med. 2011;3:98ra82.

44. Doyle LA, Fletcher CD, Hornick JL. Nuclear expression of CAMTA1 distinguishes epithelioid hemangioendothelioma from histologic mimics. Am J Surg Pathol.
2016;40:94-102.
45. Antonescu CR, Le Loarer F, Mosquera JM, et al. Novel YAP1-TFE3 fusion defines a distinct subset of epithelioid hemangioendothelioma. Genes Chromosomes
Cancer. 2013;52:775-784.
46. Rosenbaum E, Jadeja B, Xu B, et al. Prognostic stratification of clinical and molecular epithelioid hemangioendothelioma subsets. Mod Pathol. Epub 2019 Sep
19.
47. Walther C, Tayebwa J, Lilljebjörn H, et al. A novel SERPINE1-FOSB fusion gene results in transcriptional up-regulation of FOSB in pseudomyogenic hae-
mangioendothelioma. J Pathol. 2014;232:534-540.
48. Zhu G, Benayed R, Ho C, et al. Diagnosis of known sarcoma fusions and novel fusion partners by targeted RNA sequencing with identification of a recurrent
ACTB-FOSB fusion in pseudomyogenic hemangioendothelioma. Mod Pathol. 2019;32:609-620.
49. Hung YP, Fletcher CD, Hornick JL. FOSB is a useful diagnostic marker for pseudomyogenic hemangioendothelioma. Am J Surg Pathol. 2016;41:596-606.
50. Antonescu CR, Chen HW, Zhang L, et al. ZFP36-FOSB fusion defines a subset of epithelioid hemangioma with atypical features. Genes Chromosomes Cancer.
2014;53:951-959.
51. Huang SC, Zhang L, Sung YS, et al. Frequent FOS gene rearrangements in epithelioid hemangioma: a molecular study of 58 cases with morphologic reappraisal.
Am J Surg Pathol. 2015;39:1313-1321.
52. Brenn T, Fletcher CD. Radiation-associated cutaneous atypical vascular lesions and angiosarcoma: clinicopathologic analysis of 42 cases. Am J Surg Pathol.
2005;29:983-996.
53. Strobbe LJ, Peterse HL, van Tinteren H, et al. Angiosarcoma of the breast after conservation therapy for invasive cancer, the incidence and outcome: an
unforseen sequela. Breast Cancer Res Treat. 1998;47:101-109.
54. Billings SD, McKenney JK, Folpe AL, et al. Cutaneous angiosarcoma following breast-conserving surgery and radiation: an analysis of 27 cases. Am J Surg Pathol.
2004;28:781-788.
55. Manner J, Radlwimmer B, Hohenberger P, et al. MYC high level gene amplification is a distinctive feature of angiosarcomas after irradiation or chronic
lymphedema. Am J Pathol. 2010;176:34-39.
56. Guo T, Zhang L, Chang NE, et al. Consistent MYC and FLT4 gene amplification in radiation-induced angiosarcoma but not in other radiation-associated atypical
vascular lesions. Genes Chromosomes Cancer. 2011;50:25-33.
57. Suurmeijer AJ, Dickson BC, Swanson D, et al. The histologic spectrum of soft tissue spindle cell tumors with NTRK3 gene rearrangements. Genes Chromosomes
Cancer. 2019;58:739-746.
58. Miettinen M, Felisiak-Golabek A, Luiña Contreras A, et al. New fusion sarcomas: histopathology and clinical significance of selected entities. Hum Pathol. 2019;
86:57-65.
59. Kao YC, Flucke U, Eijkelenboom A, et al. Novel EWSR1-SMAD3 gene fusions in a group of acral fibroblastic spindle cell neoplasms. Am J Surg Pathol. 2018;
42:522-528.
60. Michal M, Berry RS, Rubin BP, et al. EWSR1-SMAD3-rearranged fibroblastic tumor: an emerging entity in an increasingly more complex group of fibroblastic/
myofibroblastic neoplasms. Am J Surg Pathol. 2018;42:1325-1333.
61. Knezevich SR, McFadden DE, Tao W, et al. A novel ETV6-NTRK3 gene fusion in congenital fibrosarcoma. Nat Genet. 1998;18:184-187.
62. Church AJ, Calicchio ML, Nardi V, et al. Recurrent EML4-NTRK3 fusions in infantile fibrosarcoma and congenital mesoblastic nephroma suggest a revised testing
strategy. Mod Pathol. 2018;31:463-473.
63. Hung YP, Fletcher CDM, Hornick JL. Evaluation of pan-TRK immunohistochemistry in infantile fibrosarcoma, lipofibromatosis-like neural tumour and histological
mimics. Histopathology. 2018;73:634-644.
64. Solomon JP, Linkov I, Rosado A, et al. NTRK fusion detection across multiple assays and 33,997 cases: diagnostic implications and pitfalls. Mod Pathol. 2020;
33:38-46.
65. Agaram NP, Zhang L, Sung YS, et al. Recurrent NTRK1 gene fusions define a novel subset of locally aggressive lipofibromatosis-like neural tumors. Am J Surg
Pathol. 2016;40:1407-1416.
66. Doebele RC, Drilon A, Paz-Ares L, et al. Entrectinib in patients with advanced or metastatic NTRK fusion-positive solid tumours: integrated analysis of three phase
1-2 trials. Lancet Oncol. 2020;21:271-282.
67. Laetsch TW, DuBois SG, Mascarenhas L, et al. Larotrectinib for paediatric solid tumours harbouring NTRK gene fusions: phase 1 results from a multicentre, open-
label, phase 1/2 study. Lancet Oncol. 2018;19:705-714.
68. Drilon A, Laetsch TW, Kummar S, et al. Efficacy of larotrectinib in TRK fusion-positive cancers in adults and children. N Engl J Med. 2018;378:731-739.
69. Farago AF, Le LP, Zheng Z, et al. Durable clinical response to entrectinib in NTRK1-rearranged non-small cell lung cancer. J Thorac Oncol. 2015;10:1670-1674.
70. Drilon A. TRK inhibitors in TRK fusion-positive cancers. Ann Oncol. 2019;30 (suppl 8):viii23-viii30.
71. Sarcoma and radiotherapy. Edinb Med J. 1930;37:707-708.
72. Lindberg RD, Martin RG, Romsdahl MM, et al. Conservative surgery and postoperative radiotherapy in 300 adults with soft-tissue sarcomas. Cancer. 1981;
47:2391-2397.

73. Rosenberg SA, Tepper J, Glatstein E, et al. The treatment of soft-tissue sarcomas of the extremities: prospective randomized evaluations of (1) limb-sparing
surgery plus radiation therapy compared with amputation and (2) the role of adjuvant chemotherapy. Ann Surg. 1982;196:305-315.
74. Wang D, Zhang Q, Eisenberg BL, et al. Significant reduction of late toxicities in patients with extremity sarcoma treated with image-guided radiation therapy to
a reduced target volume: results of Radiation Therapy Oncology Group RTOG-0630 trial. J Clin Oncol. 2015;33:2231-2238.
75. O’Sullivan B, Davis AM, Turcotte R, et al. Preoperative versus postoperative radiotherapy in soft-tissue sarcoma of the limbs: a randomised trial. Lancet. 2002;
359:2235-2241.
76. O’Sullivan B, Griffin AM, Dickie CI, et al. Phase 2 study of preoperative image-guided intensity-modulated radiation therapy to reduce wound and combined
modality morbidities in lower extremity soft tissue sarcoma. Cancer. 2013;119:1878-1884.
77. Folkert MR, Casey DL, Berry SL, et al. Femoral fracture in primary soft-tissue sarcoma of the thigh and groin treated with intensity-modulated radiation therapy:
observed versus expected risk. Ann Surg Oncol. 2019;26:1326-1331.
78. Folkert MR, Singer S, Brennan MF, et al. Comparison of local recurrence with conventional and intensity-modulated radiation therapy for primary soft-tissue
sarcomas of the extremity. J Clin Oncol. 2014;32:3236-3241.
79. Haas RLM, Miah AB, LePechoux C, et al. Preoperative radiotherapy for extremity soft tissue sarcoma; past, present and future perspectives on dose fractionation
regimens and combined modality strategies. Radiother Oncol. 2016;119:14-21.
80. Baumann BC, Nagda SN, Kolker JD, et al. Efficacy and safety of stereotactic body radiation therapy for the treatment of pulmonary metastases from sarcoma:
a potential alternative to resection. J Surg Oncol. 2016;114:65-69.
81. Mehta N, Selch M, Wang PC, et al. Safety and efficacy of stereotactic body radiation therapy in the treatment of pulmonary metastases from high grade sarcoma.
Sarcoma. 2013;2013:360214.
82. Leeman JE, Bilsky M, Laufer I, et al. Stereotactic body radiotherapy for metastatic spinal sarcoma: a detailed patterns-of-failure study. J Neurosurg Spine. 2016;
25:52-58.
83. Wu JS, Hochman MG. Soft-tissue tumors and tumorlike lesions: a systematic imaging approach. Radiology. 2009;253:297-316.
84. Paulson ES, Erickson B, Schultz C, et al. Comprehensive MRI simulation methodology using a dedicated MRI scanner in radiation oncology for external beam
radiation treatment planning. Med Phys. 2015;42:28-39.
85. Owrangi AM, Greer PB, Glide-Hurst CK. MRI-only treatment planning: benefits and challenges. Phys Med Biol. 2018;63:05tr1.
86. Massaccesi M, Cusumano D, Boldrini L, et al. A new frontier of image guidance: organs at risk avoidance with MRI-guided respiratory-gated intensity modulated
radiotherapy: technical note and report of a case. J Appl Clin Med Phys. 2019;20:194-198.
87. Ilicic K, Combs SE, Schmid TE. New insights in the relative radiobiological effectiveness of proton irradiation. Radiat Oncol. 2018;13:6.
88. DeLaney TF, Chen YL, Baldini EH, et al. Phase 1 trial of preoperative image guided intensity modulated proton radiation therapy with simultaneously integrated
boost to the high risk margin for retroperitoneal sarcomas. Adv Radiat Oncol. 2017;2:85-93.
89. Leiser D, Calaminus G, Malyapa R, et al. Tumour control and quality of life in children with rhabdomyosarcoma treated with pencil beam scanning proton therapy.
90. Doyen J, Jazmati D, Geismar D, et al. Outcome and patterns of relapse in childhood parameningeal rhabdomyosarcoma treated with proton beam therapy. Int
91. DeLaney TF, Park L, Goldberg SI, et al. Radiotherapy for local control of osteosarcoma. Int J Radiat Oncol Biol Phys. 2005;61:492-498.
92. Rombi B, DeLaney TF, MacDonald SM, et al. Proton radiotherapy for pediatric Ewing’s sarcoma: initial clinical outcomes. Int J Radiat Oncol Biol Phys. 2012;
82:1142-1148.
93. Weber DC, Malyapa R, Albertini F, et al. Long term outcomes of patients with skull-base low-grade chondrosarcoma and chordoma patients treated with pencil
beam scanning proton therapy. Radiother Oncol. 2016;120:169-174.
94. Pennington JD, Eilber FC, Eilber FR, et al. Long-term outcomes with ifosfamide-based hypofractionated preoperative chemoradiotherapy for extremity soft tissue
sarcomas. Am J Clin Oncol. 2018;41:1154-1161.
95. Kalbasi A, Kamrava M, Chu FI, et al. A phase II trial of 5-day neoadjuvant radiation therapy for patients with high-risk primary soft tissue sarcoma. Clin Cancer Res.
Epub 2020 Feb 13.
96. Guadagnolo BA, Zagars GK, Ballo MT, et al. Excellent local control rates and distinctive patterns of failure in myxoid liposarcoma treated with conservation surgery
and radiotherapy. Int J Radiat Oncol Biol Phys. 2008;70:760-765.

SARCOMA
Secondary Sarcomas: Biology, Presentation, and

Clinical Care
Vanessa Eulo, MD1,2; Harry Lesmana, MD3; Leona A. Doyle, MD4; Kim E. Nichols, MD3; and Angela C. Hirbe, MD, PhD1,2
overview
Secondary sarcomas are a subset of sarcomas that occur in patients with prior cancer diagnoses and are
associated with environmental or genetic factors. Although secondary sarcomas are rare in general, there are
predisposing factors that can substantially increase this risk in certain populations. Herein, we review the
environmental factors with the strongest association of sarcoma risk, including chemical exposure, certain
viruses, cytotoxic and immunosuppressive agents, chronic edema, and radiation exposure. Additionally, the
most common genetic disorders that carry a predisposition for sarcoma development will be discussed,
including hereditary retinoblastoma (RB), Li-Fraumeni syndrome (LFS), neurofibromatosis type 1 (NF1), and
DICER1 syndrome. Although treatment does not generally differ for sporadic versus secondary sarcomas,
awareness of the risk factors can alter therapeutic strategies to minimize risk, aid prompt diagnosis by in-
creasing clinical suspicion, and allow for appropriate surveillance and genetic counseling for those patients
with cancer predisposition syndromes.
INTRODUCTION soft tissue and bone sarcomas. Other environmental

Sarcomas are a heterogeneous group of rare tumors of risk factors include industrial chemicals, viruses such
mesenchymal origin.1,2 There are more than 70 sub- as human herpesvirus-8 and Epstein-Barr virus (EBV),
types, and they make up less than 1% of all new cancer and chronic lymphedema, particularly in the setting of
diagnoses annually but up to 15% to 20% of pediatric breast cancer treatment. Medications, including im-
cancers.3 The two primary groups of sarcomas are munosuppressants and other cancer treatments, are
soft tissue sarcomas (STSs) and bone sarcomas. The also associated with an increased risk of developing
estimated new cases of STSs and bone sarcomas in sarcoma. Survivors of childhood cancer are at a par-
the United States in 2020 are 13,130 and 3,600, ticularly increased risk of secondary malignant neo-
plasms, including sarcomas.7 Finally, there are several
respectively.4 Approximately 1,600 of these cases
genetic syndromes with links to soft tissue and bone
would be expected to occur in children.5 Sarcomas
sarcomas, including RB, LFS, NF, and DICER1 syn-
are also varied at the genomic level. Some have rel-
drome, among others. This review will focus on the
atively simple genetics based on the presence of
causes of secondary sarcomas.
somatic point mutations affecting key cancer-
associated genes (usually KIT or PDGFRA for gas- ENVIRONMENTAL EXPOSURES
trointestinal stromal tumor [GIST] and CTNNB1 or
A variety of environmental exposures increase the
APC for desmoid tumor) or specific translocations
risk of sarcomas. These include specific chemicals, vi-
(EWS-FLI1 in Ewing’s, SS18-SSX in synovial sarcoma,
ruses, medication, inflammation/edema, and radiation.
FUS-DDIT3 in myxoid liposarcoma), whereas others
have very complex genetics or highly abnormal kar- Chemicals
Author affiliations
and support
yotypes (leiomyosarcoma, angiosarcoma, undifferentiated
Several chemicals have been associated with the de-
information (if pleomorphic sarcoma, malignant peripheral nerve sheath
velopment of sarcoma. However, given small patient
applicable) appear tumors [MPNST]).6
at the end of this
numbers in cohorts, there is difficulty in establishing
article. Most sarcomas are idiopathic and account for a pa- clear causal roles. Vinyl chloride is a chemical used in
Accepted on March tient’s only primary cancer. However, a subset are the plastics industry in the 1970s and has been as-
4, 2020 and secondary sarcomas, or second cancers, in a patient sociated with the development of hepatic angiosarcoma.8,9
published at who previously had a different primary cancer. These A study of a of cohort of 9,951 men employed be-
ascopubs.org on secondary sarcomas are associated with environ- tween 1942 and 1972 at vinyl chloride or polyvinyl
March 26, 2020:
mental and/or genetic factors. The International chloride plants found the median latency for angio-
10.1200/EDBK_ Agency for Research on Cancer has deemed ionizing sarcoma of the liver was 36 years (range, 13.5–55.9
280985 radiation as the strongest environmental cause for both years).8 Mortality substantially increased for cancers

Eulo et al
adaptive immune responses, including antigen pre-

sentation.14 There are four subtypes of KS: classic, African
endemic, immunosuppression related, and AIDS related.
• Although only a small proportion of primary Classic KS is seen in older men of Mediterranean, Eastern
sarcomas can be attributed to environmental or
European Jewish, and South American origin. It typically
underlying genetic causes, this proportion is
presents as cutaneous lower-extremity lesions and has
increased for secondary sarcomas.
a more chronic course.15 Endemic KS is seen in young men
• Ionizing radiation is the most common envi- in sub-Saharan Africa, where rates of Podoconiosis are high
ronmental exposure leading to a secondary
because of chronic lymphedema.16 This form has a varied
sarcoma, particularly when it develops within or
clinical course, from indolent to aggressive, with the most
just adjacent to the field of prior radiation.
aggressive form presenting as adenopathy in younger
• Host genetic factors, independently and in children.15 Immunosuppression-related or iatrogenic KS is
conjunction with environmental exposures,
seen in patients who are immunosuppressed because of
contribute to the development of secondary
solid organ transplantation and subsequent immunosup-
sarcomas.
pression, which is described later in this review. AIDS-
• Identifying an underlying cancer predisposition related KS is the most common cancer in patients living
enables tailored treatment of primary cancers to
with HIV/AIDS and is an AIDS-defining malignancy.14 It can
minimize the risk for secondary sarcomas and
present more diffusely and involve viscera. The clinical
informs approaches to surveillance to detect
and treat second malignant neoplasms, in- course is varied, and treatment includes starting highly
cluding sarcomas, as early as possible. active antiretroviral therapy and/or chemotherapy.
• Treatment is similar for primary and secondary EBV is carried asymptomatically in most of the population.
sarcomas. However, there are EBV-associated smooth muscle tumors,
which are found in patients with impaired T cell–mediated
immunity, such as in HIV or after solid organ transplantation.17-19
of connective and soft tissues (standardized mortality ratio
These are rare, soft tissue, spindle cell neoplasms. These
[SMR], 2.43; 95% CI, 1.48–3.75) and cancers of the liver,
differ from primary leiomyosarcomas, because they often
biliary tract, and gall bladder (SMR, 2.87; 95% CI, 2.40–
occur at atypical sites, such as the central nervous system,
3.40).8 Hepatic angiosarcoma risk was increased and re-
and are EBV-encoded small nuclear RNA positive. They
lated to cumulative exposure, with a hazard ratio of 38.1
are designated as leiomyosarcomas based on nuclear
(95% CI, 15.1–96.0) for those with exposures of 5,686 to
atypia and mitotic activity but are usually low grade.18,19 A
less than 10,551 ppm-years and a hazard ratio of 73.6
single-institution study of more than 5,000 solid organ
(95% CI, 28.8–188.4) in the patients with 10,551 or more
recipients older than age 30 noted three cases of post-
ppm-years of exposure.8 There have also been epidemi-
transplant EBV-associated smooth muscle tumors, all of
ologic studies about pentachlorophenol, a chemical used
which were seronegative pediatric heart transplant re-
in wood preservatives and insecticides. One Swedish study
cipients, suggesting that this group may be at higher
found an excess risk of STS in those exposed to chlor-
risk.17
ophenols, although other studies have not confirmed that
association.6,10 A meta-analysis including 31 studies with Iatrogenic
29,605 cancer cases were included, and it found that
Several medications, including some chemotherapeutic
higher levels of 2,3,7,8-tetrachlorodibenzo-p-dioxin were
agents, are associated with the development of sarcoma,
associated with development of STS (SMR, 1.60; 95% CI,
likely because of suppression of immune responses. Com-
1.15–2.23).11 There have also been cohort studies sug-
mon drugs that are associated include cyclosporine, aza-
gesting increase risk of STSs in patients with exposure to
thioprine, corticosteroids, and rituximab. These sarcomas are
phenoxy herbicides, particularly with higher chlorinated
generally cutaneous and will regress when immunosup-
dioxins and 2,3,7,8-tetrachlorodibenzo-p-dioxin, but the
pression is decreased.15
data remain inconclusive.11-13
There also appears to be an increased incidence of uterine
Viruses sarcoma and carcinosarcoma among women treated with
There are several virally mediated sarcomas. Kaposi sar- tamoxifen for breast cancer.20-22 Tamoxifen selectively
coma (KS) is a well-established complication of immuno- stimulates the endometrium and increases the risk for
suppression and immunodeficiency associated with human uterine malignancies. Most commonly, tamoxifen is asso-
herpesvirus-8/KS-related herpesvirus infection. KS-related ciated with low-grade endometroid carcinoma, but uterine
herpesvirus infection encodes viral proteins that inhibit sarcomas are seen in approximately 10% of uterine ma-
innate immune pathways and expresses proteins that impair lignancies in long-term tamoxifen users.20 Treatment is

Secondary Sarcomas
typically multimodal, with hysterectomy with or without individuals with specific cancer predisposition syndromes.
radiation therapy or chemotherapy. Radiation-associated sarcomas must fulfill the following
Cytotoxic chemotherapy exposure has also been linked to diagnostic criteria: (1) the tumor must arise within, or ad-
the development of sarcomas. The Dutch Childhood Cancer jacent to, a previously irradiated field; (2) the tumor should
Oncology Group–Long-Term Effects After Childhood Cancer arise at least 6 months after the cessation of radiation
cohort is a study that evaluated the long-term risks of therapy; and 3) the tumor must be a histologically confirmed
secondary malignancies in survivors of childhood cancer. It sarcoma, distinct from the patient’s prior malignancy.32-34
included 6,165 survivors of childhood cancer from the Ionizing radiation damages DNA by causing double-strand
Netherlands who were diagnosed from 1963 to 2001. They breaks and generating reactive oxygen species that result in
found that the incidence of subsequent bone sarcoma was a variety of insults, including single-strand breaks and DNA
increased in children who had received cyclophosphamide cross-linking. This damage results in various types of mo-
in a dose-dependent manner.23 Additionally, a nested case- lecular alterations, from single base substitutions to cata-
control study from the Childhood Cancer Survivor Study strophic genomic events. Similar to other environmental
evaluated 105 secondary sarcomas with 422 matched carcinogens, such as tobacco and ultraviolet radiation,
controls in a cohort of 14,372 survivors of childhood cancer a mutational signature of radiation is emerging. A small
and found that anthracycline exposure was associated with number of radiation-associated neoplasms (carcinomas
sarcoma risk (odds ratio, 3.5, 95% CI, 1.6–7.7) at a median and sarcomas) analyzed by whole-genome sequencing
of 11.8 years (range, 5.3–31.3 years) after initial di- showed two features: (1) a greater number of small deletions
agnosis.24 This analysis was adjusted for radiation and other (with proportionately more deletions than insertions) than
therapies. radiation-naı̈ve tumors, a feature also described in BRCA1-
Chronic Edema and BRCA2-associated breast carcinoma, and (2) an en-
richment of balanced inversions, thought to result from
Angiosarcoma/lymphangiosarcoma, or Stewart-Treves syn- chromothripsis and other forms of structural damage.35
drome, is a rare malignant cutaneous lesion that arises from Transcriptome analysis has shown a signature of differen-
chronic lymphedema. Chronic lymphedema can be caused tially expressed genes, suggesting a background of chronic
by congenital lymphedema or secondary causes, including oxidative stress in radiation-induced sarcomas compared
chronic bacterial infections, radical mastectomy, vein har- with sporadic sarcomas.36 Although there is much interest
vesting, radiation, trauma, lymphoma, or filiarasis.16,25 Most in mutational burden and mismatch repair deficiency in
cases are seen in the setting of prior radical mastectomy, solid tumors, early data suggest that the mutational burden
and the reported incidence of developing angiosarcoma in radiation-associated sarcomas is frequently low, but
in patients surviving at least 5 years after radical mastectomy larger studies are needed.37,38 Overall, however, the ge-
is 0.07% to 0.45%, with an onset of 5 to 25 years.16,26,27 nomic landscape of radiation-associated sarcomas is similar
The pathophysiology is thought to be caused by local im- to sporadic sarcomas, with a few exceptions. The mutation
munodeficiency and inflammation from chronic lymphedema.16,25 pattern of TP53 in radiation-associated sarcomas differs
Preclinical and clinical data have shown aberrant ex- from sporadic sarcomas, in that mutations are usually caused
pression of mTORC1 and angiogenic regulators, including by small deletions, with recurrent mutations at codons 135
VEGF. Lymphangiosarcomas often have increased am- and 237.39,40 TP53 mutations occur in one-third of radiation-
plification of MYC, which is not seen in benign atypical associated sarcomas unassociated with hereditary pre-
vascular lesions.28-30 This tumor often appears as reddish disposition syndromes, and inactivation of one allele is an
blue macules or nodules that may coalesce over time. early event in tumor formation.40 Loss of CDKN2A and
Treatment is often multimodal, with surgical resection with CDKN2B occurs with a higher frequency in radiation-
wide margins being the mainstay of treatment. Recurrence associated sarcoma compared with sporadic sarcoma.41
rates remain high despite aggressive surgical treatment, MYC amplification is common in radiation-associated
and outcomes are poor, with a survival of range of 19 to 31 angiosarcoma but rare in sporadic angiosarcoma.42 Co-
months.27,31 amplification of FLT4, a gene involved in lymphatic devel-
Radiation Exposure opment, occurs in up to 25% of secondary angiosarcomas
Although radiation-associated sarcomas in general are rare, and may be associated with response to tyrosine kinase
occurring in less than 1% of patients who receive radiation inhibitors, such as sorafenib.43
therapy, the incidence is expected to increase with the The likelihood of developing radiation-associated sarcoma
increasing use of radiation therapy for certain tumor types is influenced by radiation dose, radiation field, and patient-
alongside the overall increasing survival of patients with specific factors. The primary malignancy type and site, and
cancer. Furthermore, and as outlined below, the develop- the risk of local recurrence, determine radiation dose. Most
ment of radiation-associated sarcomas is increased in patients receive conventionally fractionated radiation

Eulo et al
administered at 1.8 to 2 Gy/day over the course of 5 to 8 the few large outcome studies examining patients with
weeks; patients receiving radiation without plans for sur- radiation-associated sarcomas, the following clinical and
gical resection typically receive higher doses of radiation histologic features have been associated with worse overall
than patients receiving preoperative or postoperative ra- survival: positive resection margins, deep tumor location
diation therapy. The biology of adjacent organs also in- (intrathoracic or intra-abdominal), and high grade.33,34,38,45
fluences radiation dose and field. For example, the bowel is There are conflicting data on whether histologic subtype
only able to tolerate radiation doses of up to approximately influences grade, with some studies suggesting a worse
50 Gy before enteritis occurs, so this tends to be the prognosis for MPNST.34,46
maximum dose for the management of gastrointestinal
Although radiation-associated sarcomas have generally
malignancies. Patients with inherited germline variants in
been considered to have a poor prognosis among sarcoma
genes such as ATM have deficiencies in the double-strand
subtypes, with older literature reporting 5-year survival rates
break repair pathway causing a high risk of secondary
of 14% to 33%, two recent studies have suggested that
malignancy; therefore, radiation therapy is rarely used in
outcomes are improving, with reported 5-year survival rates
this group.44
of 58% to 68%. 34,38,45,47 The strongest driver for better
Most patients with radiation-associated sarcomas are fe- outcome is complete surgical excision, so amenability to
male (approximately 70%), reflecting the frequency of ra- margin-negative excision is associated with better out-
diation therapy for the management of breast cancer. The comes. This improvement in outcome is seen best in
latency period from diagnosis of primary malignancy to radiation-associated angiosarcoma arising after treatment of
radiation-associated sarcoma is very broad, ranging from breast cancer, likely reflecting contemporary surgical ap-
months to more than 50 years, with a median duration of proaches involving excision of not only the neoplasm but
11 years. There is a notably shorter median latency for also the entire radiation field.
patients with radiation-associated angiosarcoma of the
Although an increase in the incidence of radiation-associated
breast compared with patients with other types of radiation-
sarcomas is anticipated, modern radiation therapy ap-
associated sarcomas: 8 versus 15 years, respectively.38 The
proaches that allow higher-dose conformality to the targeted
age range of patients with radiation-associated sarcomas is
area and reduce radiation of adjacent normal tissues have
also wide because of both the variable latency between
improved radiation-associated morbidity, and it seems
initial and secondary malignancy and the diverse patient
possible that they may also reduce the risk of developing
population. Although pediatric patients may develop
radiation-associated sarcoma. In theory, the risk of de-
radiation-associated sarcomas as young adults, older adults
veloping radiation-induced secondary sarcomas may be
who receive radiation therapy may develop radiation-
higher in patients with cancer predisposition syndromes.
associated tumors in their ninth decade and beyond.
Further research is needed in this regard.
Among patients with radiation-associated sarcomas, the
most common primary malignancy for which radiation CANCER PREDISPOSITION SYNDROMES
therapy was administered is breast carcinoma, followed by
Since the first systematic description of NF1 by von
lymphoma, and a smaller number of primary gynecologic or
Recklinghausen in 1882, a growing number of Mendelian
prostatic primaries, unrelated sarcoma types, head and
disorders have been associated with increased pre-
neck cancers, colorectal cancer, or germ cell tumors.33,38
disposition to sarcomas (Table 1). Thus, although most
The incidence in patients with lymphoma is diminishing as
sarcomas are believed to occur sporadically, recent geno-
treatment has changed from high-dose, large-volume ra-
mic investigations have revealed that 5% to 25% of children
diation therapy to chemotherapy only.
and adults with STSs or bone sarcomas harbor an un-
Radiation-associated sarcomas arise within the initial ra- derlying genetic predisposition.48-50 This is likely an un-
diation field; therefore, superficial locations on the chest wall derestimation, because there is growing evidence pointing
or an extremity are most common, followed by intrathoracic, to the role of multiple genetic alterations (such as polygenic
intra-abdominal/pelvic locations, and head and neck. The variants) in carcinogenesis.51 Most of the sarcoma-predisposing
most common histologic types of radiation-associated sar- genes are involved in critical cellular processes, such as dif-
comas are undifferentiated pleomorphic/spindle cell sar- ferentiation, proliferation, apoptosis, repair of DNA damage, and
coma and angiosarcoma, which together account for intracellular signaling. Here, we discuss several of the classic
approximately 80% of all cases. Less common subtypes Mendelian single-gene disorders associated with an increased
include MPNST, leiomyosarcoma, and osteosarcoma (ei- predisposition to sarcoma.
ther arising in soft tissue or bone). High tumor grade (by the
French Fédération Nationale des Centres de Lutte Contre le Hereditary RB
Cancer grading system) in unclassified sarcomas is asso- RB is the most common intraocular malignancy in child-
ciated with a worse outcome on multivariable analysis.38 In hood, affecting 1 in 17,000 live births, of which 40% of

Secondary Sarcomas
TABLE 1. Selected Cancer Predisposition Syndromes Associated With Sarcomas

Inheritance Sarcoma Surveillance
Inherited Syndrome Gene/Locus (OMIM#) Associated Sarcomas Other Associated Cancers Recommendation
Neurofibromatosis NF1 AD (162200) MPNST, GIST, RMS Leukemias (JMML, MDS), OPG Assess with history and
type 1 physical exam annually for
typical signs of MPNST: any
nondermal neurofibroma
with rapid growth, loss of
neurologic function, or
increasing pain or change in
consistency52
Li-Fraumeni TP53 AD (151623) Embryonal RMS, OS, Breast cancer, ACC, leukemias, Annual WBMRI53
syndrome leiomyosarcoma choroid plexus carcinoma
DICER1 syndrome DICER1 AD (601200) Embryonal RMS, anaplastic Sertoli-Leydig cell tumor, Abdominal ultrasound every
sarcoma of the kidney, differentiated thyroid 6 months until 8 years of age,
leiomyosarcoma, carcinoma, pineoblastoma then every 12 months until
carcinosarcoma, 12 years of age
undifferentiated sarcoma
For females beginning at
8–10 years of age: pelvic and
abdominal ultrasound every
6–12 months at least until
age 4054
Hereditary RB1 AD (180200) Osteosarcoma, embryonal Retinoblastoma Education regarding second
retinoblastoma RMS primary tumor risks and
close attention to any new
signs/symptoms
Some consider WBMRI
annually after age 8, but no
consensus55
Familial adenomatous APC AD (175100) Desmoid tumor Colorectal carcinomas, thyroid Annual physical examination: a
polyposis carcinoma, carcinoma of the bdominopelvic MRI (for
ampulla of Vater individuals with positive
family history of desmoids),
every 1–3 years following
colectomy or other surgery
and then lengthen time
frame to 5–10 years56
Constitutional EPCAM, AR (276300) Embryonal RMS Leukemias, lymphomas, Annual WBMRI from age
mismatch repair MLH1, glioblastoma multiforme, 6 years57
syndrome MSH2, ependymoma,
MSH6, medulloblastoma,
PSM2 astrocytoma, neuroblastoma,
basal cell carcinoma,
colorectal cancers
Familial rhabdoid SMARCB1, AD (60932) Malignant rhabdoid tumor Pineoblastoma, For carrier of germline
tumor SMARCA4 medulloblastoma, choroid truncating SMARCB1
predisposition plexus carcinoma mutations, consider WBMRI
syndrome to age 5 years, undetermined
frequency, and ultrasound
every 3 months58
No specific recommendation
for SMARCA4 carriers
Carney-Stratakis SDHB, SDHC, AD (60674) GIST Paraganglioma No sarcoma specific
syndrome SDHD consensus guideline
Familial GIST KIT, PDGFRA AD (60674) GIST None No sarcoma specific
consensus guideline

Eulo et al
TABLE 1. Selected Cancer Predisposition Syndromes Associated With Sarcomas (Continued)

Inheritance Sarcoma Surveillance
Inherited Syndrome Gene/Locus (OMIM#) Associated Sarcomas Other Associated Cancers Recommendation
Hereditary FH AD (150800) Leiomyosarcoma Renal cell carcinoma Abdominopelvic MRI annually
leiomyomatosis and starting at age 8 years59
renal cell
carcinoma
Beckwith-Wiedemann CDKN1C, AD (130650) Embryonal ACC, hepatoblastoma, Wilms No sarcoma-specific
syndrome H19, ICR1, rhabdomyosarcoma tumor consensus guideline
KCNQ1OT1
Werner syndrome RECQL2 AR (277700) Osteosarcoma, embryonal Leukemias, melanoma, thyroid No sarcoma-specific
rhabdomyosarcoma carcinoma consensus guideline
Bloom syndrome RECQL3 AR (210900) Osteosarcoma, embryonal Leukemias, lymphomas, No sarcoma-specific
rhabdomyosarcoma adenocarcinoma, squamous consensus guideline
cell carcinoma
Rothmund-Thompson RECQL4 AR (268400) Osteosarcoma Basal cell carcinoma, No sarcoma-specific
syndrome squamous cell carcinoma consensus guideline
Gorlin syndrome/basal PTCH1, AD (109400) Embryonal RMS Basal cell carcinoma, No sarcoma-specific
cell nevus PTCH2, medulloblastoma consensus guideline
syndrome SUFU
Rasopathies (Noonan, HRAS, AD (218040, Embryonal RMS Leukemias (JMML), malignant From 0 to 8–10 years: physical
Costello syndrome) PTPN11, 163950, schwannoma, bladder examination and
SOS1 610733) carcinoma abdominopelvic ultrasound
6 CXR every 3–4 months
From 10 years: annual
urinalysis60
Nijmegen breakage NBN AR (251260) Embryonal RMS Lymphomas, medulloblastoma No sarcoma-specific
syndrome consensus guideline
Rubinstein-Taybi CREBBP AD (180849) Embryonal RMS Leukemias No sarcoma-specific
syndrome consensus guideline
Abbreviations: OMIM, Online Mendelian Inheritance in Man; AD, autosomal dominant; MPNST, malignant peripheral nerve sheath tumor; GIST,
gastrointestinal stromal tumor; RMS, rhabdomyosarcoma; JMML, juvenile myelomonocytic myeloid leukemia; MDS, myelodysplastic syndrome; OPG, optic
pathway glioma; OS, osteosarcoma; ACC, adrenocortical carcinoma; WBMRI, whole-body MRI; AR, autosomal recessive; CXR, chest x-ray.
cases are hereditary.61 Notably, the study of RB formed the and epithelial cancers (lung, bladder, and breast).54,65 Al-
basis for the seminal “two-hit hypothesis” of tumor for- though most sarcomas developing in individuals with he-
mation, in which Alfred Knudson postulated that individuals reditary RB are secondary to prior radiation therapy, in up to
with hereditary RB are at increased risk to develop bilateral 40% of cases, these sarcomas occur outside the radiation
or multifocal tumors because they carry a germline mutation field. Because of the increased risk for radiation-induced
in a critical growth regulatory gene.62 Knudson speculated sarcomas, treatment has moved away from radiotherapy and
that acquisition of a mutation affecting the second copy of toward alternative approaches such as chemotherapy and/or
this gene in one or more retinal cells would then lead to focal intraocular therapies.
tumor formation. In 1986, Knudson’s prediction was con-
firmed when Stephen Friend isolated RB1, the first cancer LFS
predisposition gene to be identified.63 RB1 encodes a nu- LFS is an autosomal dominant cancer predisposition syn-
clear phosphoprotein that is essential for regulating cell drome caused by germline loss-of-function mutations af-
cycle progression and expression of genes involved in ap- fecting TP53, the gene encoding the p53 tumor suppressor.
optosis, autophagy, and chromosome stability.64 Historically considered a rare condition, a recent large,
Although individuals with hereditary RB are most prone to population-based study estimated the prevalence of path-
developing eye tumors, they are also at a substantially in- ogenic TP53 variants as high as 0.2%.66 The classic defi-
creased risk of developing other cancers—up to 20-fold nition of LFS mandates the presence of a proband diagnosed
compared with the general population. These second can- with sarcoma before age 45, a first-degree relative younger
cers primarily include bone sarcomas and STSs (rhabdo- than age 45 with any cancer, and a first- or second-degree
myosarcoma [RMS] and leiomyosarcomas), melanomas, relative younger than age 45 with any cancer or a sarcoma at

Secondary Sarcomas
any age.67 Subsequent modifications of LFS diagnostic of STSs is significantly higher in patients with NF1 compared
criteria have incorporated additional tumor types falling into with the general population (proportionate mortality ratio,
the LFS spectrum, including premenopausal breast cancer, 34.3; 30.8–38.0), and they account for significant mortality
brain tumors (especially choroid plexus carcinomas), ad- (SMR, 2.02; 1.6-2.6; p , 10–4).74,75 MPNSTs are the most
renocortical carcinomas, and leukemias.68 Individuals with frequent sarcoma seen in NF1, with a lifetime risk of 8% to
LFS have an increased lifetime risk for cancer development, 13%.52 MPNSTs arise from existing plexiform neurofi-
with nearly 100% of affected individuals developing a cancer bromas that undergo malignant transformation, usually in
by age 70 and 50% developing a second primary cancer young adults. Additional somatic loss in the tumor sup-
within 10 years of an initial cancer diagnosis.69 The risk of pressor genes CDKN2A/B is a critical step in pathogenesis
developing second cancers is highest among survivors of of atypical neurofibromas, a premalignant lesion to MPNST,
childhood cancer, likely because of the untoward effects of which then subsequently acquires additional somatic mu-
treatment of the first cancer with chemotherapy and/or ra- tations in genes such as TP53 or polycomb repressive
diation therapy. complex components (EED, SUZ12).53,55 Based on data
Sarcomas are the most common cancer after breast cancer from the U.K. NF1 registry, the median age of diagnosis of
in TP53 mutation carriers.69 A survey from the International MPNST is age 26 compared with age 62 for sporadic
Agency for Research on Cancer TP53 database revealed MPNST. A recent population-based cohort study in Finland
that sarcomas represent 17.4% of all cancers recorded, and investigating the pediatric NF1 population younger than age
36.8% of cancers in TP53 mutation carriers are in patients 20 identified nine MPNSTs out of 53 tumors (17%), with
younger than age 20.70 A similar study from the National a median age of diagnosis of 15 years.56 These data
Cancer Institute LFS cohort identified sarcomas as the most highlight the importance of surveillance in this high-risk age
common first cancers, which were present in more than group, as early diagnosis and complete surgical resection
50% of carriers of the TP53 mutation younger than age improve overall survival.57
17.69 Osteosarcomas were the most common subtype of In addition to MPNST, other STSs, such as RMSs, are
sarcoma at any age (40.4%), followed by sarcoma not enriched in the NF1 population. Retrospective analyses
otherwise specified (17%), RMS (16.5%), leiomyosarcoma from several pediatric RMS studies identified the presence
(9.1%), and liposarcoma (4.9%). It is estimated that germline of germline NF1 mutations in 0.5% to 1% of patients,58,59
TP53 mutations can be identified in 2.5% to 10% of all a prevalence 20 times greater than in the general pop-
sarcomas presenting in children and adults younger than age ulation. Most RMSs are diagnosed before age 3, have an
20.70 Specifically, the diagnosis of anaplastic embryonal RMS embryonal subtype, and arise from the genitourinary tract or
in children is highly suggestive of the presence of a patho- pelvis. Outcomes are similar to those in children and young
genic germline TP53 mutation. Thus, this histology should adults without NF1. NF1 has also been associated with the
prompt genetic counseling and TP53 genetic testing of af- development of GISTs. Data from the Swedish cancer
fected children.71 registry reveal that 7% of individuals with NF1 develop
GISTs, primarily in the small intestine. NF1-associated
NF1
GISTs comprised approximately 1.5% of all GIST cases.60
NF1 is an autosomal dominant disorder affecting 1 in every A recent Finnish, retrospective, case-control study vali-
3,000 births, making it the most common cancer pre- dated this finding, showing an overall hazard ratio of 2.6 for
disposition syndrome in humans.72 It is caused by loss-of- intestinal tumors in NF1 compared with the general
function mutations in NF1, the gene encoding neurofibromin, population.76
a GTPase-activating protein that negatively regulates the Ras
signaling pathway. Loss of GTPase-activating protein function DICER1 Syndrome
leads to hyperactivation of Ras, which promotes the tran- DICER1 syndrome is an autosomal dominant disorder
scription of a number of genes required for cell growth and caused by germline loss-of-function mutations in DICER1,
proliferation, ultimately culminating in tumor formation. The which encodes an essential ribonuclease III endonuclease
diagnosis of NF1 is clinically established based on the that participates in the generation of microRNAs.77 Since
presence of two classic features, such as café-au-lait mac- the identification of heterozygous germline DICER1 muta-
ules, neurofibromas, axillary or inguinal freckling, optic gli- tions in familial pulmonary pleuroblastoma in 2009,78
oma, Lisch nodules, distinctive bony lesions such as a variety of other benign or malignant neoplasms have
sphenoid dysplasia, thickening of long bone cortex, or been associated with germline DICER1 mutations, including
pseudarthrosis, and the presence of a first-degree relative cystic nephroma, Sertoli-Leydig cell tumors, gynandro-
with NF. Approximately one-half of the cases are familial. blastoma, differentiated thyroid carcinoma, nasal chon-
Patients with NF1 are at increased risk of developing both dromesenchymal hamartoma, ciliary body medulloepithelioma,
benign and malignant tumors throughout life.73 The incidence primary brain tumors (especially pituitary blastoma and

Eulo et al
pineoblastoma), and embryonal RMS of the cervix. Although predisposition syndromes affecting children, including
DICER1-associated neoplasms are rare individually, the those discussed herein. Based on positive results using the
prevalence of germline DICER1 mutations in the population “Toronto protocol” that incorporates physical examinations,
has been estimated to be approximately 1:10,600.79 Unlike blood tests, and whole-body MRI (WBMRI), the American
most cancer predisposition syndromes, for which a variety of Association for Cancer Research suggested adopting
second somatic hits results in carcinogenesis, DICER1- a modified version of this protocol to monitor individuals with
associated neoplasms evolve after mutations affecting LFS90 for the development of primary or secondary
only five hotspot codons.80 These second hits are missense neoplasms.91 Although WBMRI has excellent sensitivity,
mutations in the DICER1 RNase IIIb domain, leading to specificity is less than perfect. Toward this end, an in-
neomorphic DICER1 function in microRNA processing. It vestigation from the National Cancer Institute in which
results in an imbalanced ratio between 5p and 3p mature a baseline WBMRI was performed for 116 individuals with
microRNAs, altering the expression of numerous down- LFS identified cancers in 7%. Nevertheless, one-third of
stream target mRNAs across the exome, including those participants required additional screening, with most
critical for embryogenesis or tumor suppression.81 This identified lesions ultimately deemed incidental, resulting in
phenomenon suggests that retention of some microRNA a false-positive rate of 29.6%.92 Remarkably, a second
processing function is required for tumor survival. Germline study from the Li Fraumeni Exploration Research Consor-
mosaicism for these hotspot mutations (instead of the tium that included 578 individuals with LFS revealed es-
typical loss-of-function mutations) has also been reported in sentially identical findings, with 7% identified by WBMRI to
individuals with DICER1 syndrome characterized by earlier have a primary malignant neoplasm and a false-positive rate
onset with multisite disease.82 of 42.5%.93
Since the recognition of DICER1 syndrome, at least 86 The feasibility and clinical value of WBMRI have also been
DICER1-associated sarcomas have been reported.83 The investigated in hereditary RB and NF1. Although there are
spectrum of DICER1-associated sarcomas includes RMS and no established or standard protocols for individuals with
non-RMS STSs primarily affecting the genitourinary region. hereditary RB, some experts recommend annual WBMRI
The median age of sarcoma diagnosis is 12 years, with after ages 8 to 10, when a child is able to lie still for the scan
embryonal RMS as the most common subtype.83,84 A recent without requiring anesthesia.94 A retrospective study of
study incorporating data from three large DICER1 registries WBMRI from investigators at Memorial Sloan Kettering
identified 207 individuals with germline pathogenic DICER1 Cancer Center that included 25 individuals with hereditary
variants. The most common neoplasms identified were RB revealed new osseous abnormalities suspicious for
pulmonary pleuroblastoma, Sertoli-Leydig cell tumors, RMS, malignancy in five patients; two were diagnosed with lo-
and cystic nephroma.85 All RMSs in this study were em- calized high-grade osteosarcoma, and three were found to
bryonal and arose exclusively from the gynecologic tract. have benign osseous abnormalities after dedicated imaging
Non-RMS STSs in DICER1 syndrome include a range of rare and/or biopsy.95 In this study, the sensitivity and specificity
sarcomas, such as leiomyosarcoma, carcinosarcoma, un- of WBMRI in detecting secondary malignancy in survivors of
differentiated sarcoma, and anaplastic sarcoma. Anaplastic hereditary RB were 66.7% and 92.1%, respectively. The
sarcoma of the kidney is a rare tumor known to be enriched in routine use of WBMRI for individuals with NF1 remains an
the DICER1 syndrome.86 In a single retrospective study, out of area of debate, with some investigators not recommending
nine cases of anaplastic sarcomas of the kidney, eight re- WBMRI as a screening tool for asymptomatic individuals.
portedly harbored at least one somatic mutation in DICER1, Rather, they suggest that WBMRI be used to assess whole-
four of which had mutations in the adjacent normal tissue, body tumor burden in individuals with signs and symptoms
suggesting mosaic or germline status. Additional acquired concerning for MPNST.96-98 In contrast, there are some
somatic TP53 mutations also commonly co-occurred in this investigators who suggest completing a baseline WBMRI
tumor. Outside the genitourinary region, rare primary sar- before transition from the pediatric to the adult health care
comas have been reported in the central nervous system, setting to assist in determining the optimal approaches for
such as primitive intracranial sarcoma.87-89 long-term follow-up.97,99 Clearly, additional studies must be
done to further assess the role and to optimize the use of
Cancer Surveillance in Patients at Increased Genetic Risk WBMRI as a means to screen for primary or secondary
for Sarcomas sarcomas in at-risk individuals.
Early tumor detection to reduce cancer- and treatment-
related morbidity and mortality is the primary goal of sur- Genetic Predisposition to Sarcoma: Common and Other
veillance in individuals at increased genetic risk for cancer, Novel Variants Acting As Risk Factors
including sarcoma. In 2016, the American Association for The advent of massively parallel sequencing of paired tumor
Cancer Research published a series of guidelines for several and normal samples is providing new insights into the host

Secondary Sarcomas
and somatic genetic landscape of sarcomas. A large genetic finding, although in a smaller proportion, was also replicated
study including four international sarcoma cohorts identified in an adult cohort.100 Finally, BRCA2 germline mutations are
an excess of pathogenic germline variants in more than half also enriched in survivors of cancer who developed sec-
of the patients, with some of the most frequently altered ondary sarcomas.49
genes including not only TP53 but also genes implicated in
the DNA damage response (ATM, ATR, ERCC2) and ho- CONCLUSION
mologous recombination (BRCA2).51 This study also iden- In summary, sarcomas encompass a heterogeneous group
tified a statistically significant correlation between the number of aggressive cancers of mesenchymal origin. Most sarco-
of genes or alleles with pathogenic variants and the age of mas are sporadic in nature with no known cause; however,
sarcoma onset (odds ratio, 2.21; CI, 1.57–3.13, p , .0001). a subset is associated with environmental and/or genetic
Specifically, individuals with biallelic variants or monoallelic factors. Overall, treatment does not generally differ between
variants in two or more genes had a median age of first cancer primary and secondary sarcomas. Importantly, awareness
diagnosis similar to carriers of pathogenic germline TP53 of the risk factors can alter therapeutic strategies to mini-
variants. Germline BRCA2 pathogenic variants have also mize the possibility of secondary sarcoma development, aid
been identified independently in multiple cancer genomic prompt diagnosis by increasing clinical suspicion, and allow
studies. In a pediatric study, BRCA2 pathogenic variants for appropriate surveillance and genetic counseling for
were identified in three of 43 patients with sarcoma.50 This patients with cancer predisposition syndromes.

1
Division of Medical Oncology, Department of Medicine, Washington Angela C. Hirbe, MD, PhD, Washington University School of Medicine,
University School of Medicine, St. Louis, MO Department of Medicine, Campus Box 8076, 660 S. Euclid Ave., St.
2
Siteman Cancer Center, Washington University School of Medicine, St. Louis MO 63110; email: [email protected].
Louis, MO
3
Department of Oncology, St. Jude Children’s Research Hospital,
Memphis, TN
4
Department of Pathology, Brigham and Women’s Hospital, Boston, MA
EDBK_280985.
REFERENCES
1. Ratan R, Patel SR. Chemotherapy for soft tissue sarcoma. Cancer. 2016;122:2952-2960.
2. Brownstein JM, DeLaney TF. Malignant soft-tissue sarcomas. Hematol Oncol Clin North Am. 2020;34:161-175.
3. Surveillance, Epidemiology, and End Results Program (SEER). SEER*Stat database: incidence: SEER 9 regs research data. www.seer.cancer.gov. Accessed
January 2, 2020.
5. Sarcoma Alliance. How rare is sarcoma? https://sarcomaalliance.org/resources/how-rare-is-sarcoma/. Accessed January 2, 2020.
6. Lacourt A, Amadéo B, Gramond C, et al; Etiosarc study group. ETIOSARC study: environmental aetiology of sarcomas from a French prospective multicentric
population-based case-control study-study protocol. BMJ Open. 2019;9:e030013.
7. Lee JS, DuBois SG, Coccia PF, et al. Increased risk of second malignant neoplasms in adolescents and young adults with cancer. Cancer. 2016;122:116-123.
8. Mundt KA, Dell LD, Crawford L, et al. Quantitative estimated exposure to vinyl chloride and risk of angiosarcoma of the liver and hepatocellular cancer in the US
industry-wide vinyl chloride cohort: mortality update through 2013. Occup Environ Med. 2017;74:709-716.
9. Collins JJ, Jammer B, Sladeczek FM, et al. Surveillance for angiosarcoma of the liver among vinyl chloride workers. J Occup Environ Med. 2014;56:1207-1209.
10. Pentachlorophenol and Some Related Compounds. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. Lyon, France: World Health
Organization; 2019.
11. Xu J, Ye Y, Huang F, et al. Association between dioxin and cancer incidence and mortality: a meta-analysis. Sci Rep. 2016;6:38012.
12. McBride DI, Collins JJ, Humphry NF, et al. Mortality in workers exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin at a trichlorophenol plant in New Zealand.
J Occup Environ Med. 2009;51:1049-1056.
13. Collins JJ, Bodner KM, Aylward LL, et al. Mortality risk among workers with exposure to dioxins. Occup Med (Lond). 2016;66:706-712.
14. Dittmer DP, Damania B. Kaposi sarcoma-associated herpesvirus: immunobiology, oncogenesis, and therapy. J Clin Invest. 2016;126:3165-3175.
15. Vangipuram R, Tyring SK. AIDS-Associated Malignancies. Cancer Treat Res. 2019;177:1-21.

Eulo et al
16. Lee R, Saardi KM, Schwartz RA. Lymphedema-related angiogenic tumors and other malignancies. Clin Dermatol. 2014;32:616-620.
17. Stubbins RJ, Alami Laroussi N, Peters AC, et al. Epstein-Barr virus associated smooth muscle tumors in solid organ transplant recipients: incidence over
31 years at a single institution and review of the literature. Transpl Infect Dis. 2019;21:e13010.
18. Purgina B, Rao UN, Miettinen M, et al. AIDS-related EBV-associated smooth muscle tumors: a review of 64 published cases. Pathol Res Int. 2011;
2011:561548.
19. Dekate J, Chetty R. Epstein-Barr virus-associated smooth muscle tumor. Arch Pathol Lab Med. 2016;140:718-722.
20. Lavie O, Barnett-Griness O, Narod SA, et al. The risk of developing uterine sarcoma after tamoxifen use. Int J Gynecol Cancer. 2008;18:352-356.
21. Matsuo K, Ross MS, Bush SH, et al. Tumor characteristics and survival outcomes of women with tamoxifen-related uterine carcinosarcoma. Gynecol Oncol.
2017;144:329-335.
22. Wickerham DL, Fisher B, Wolmark N, et al. Association of tamoxifen and uterine sarcoma. J Clin Oncol. 2002;20:2758-2760.
23. Teepen JC, van Leeuwen FE, Tissing WJ, et al; DCOG LATER Study Group. Long-term risk of subsequent malignant neoplasms after treatment of childhood
cancer in the DCOG LATER study cohort: role of chemotherapy. J Clin Oncol. 2017;35:2288-2298.
24. Henderson TO, Rajaraman P, Stovall M, et al. Risk factors associated with secondary sarcomas in childhood cancer survivors: a report from the childhood
cancer survivor study. Int J Radiat Oncol Biol Phys. 2012;84:224-230.
25. Farzaliyev F, Hamacher R, Steinau Professor HU, et al. Secondary angiosarcoma: A fatal complication of chronic lymphedema. J Surg Oncol. 2020;121:85-90.
26. Wang LL, Cui LF, Gao Y, et al. Clinicopathologic features of Stewart-Treves syndrome. Int J Clin Exp Pathol. 2019;12:680-688.
27. Murgia RD, Gross GP. Stewart-Treves Syndrome. Treasure Island, FL: StatPearls; 2019.
28. Udager AM, Ishikawa MK, Lucas DR, et al. MYC immunohistochemistry in angiosarcoma and atypical vascular lesions: practical considerations based on
a single institutional experience. Pathology. 2016;48:697-704.
29. Sun S, Chen S, Liu F, et al. Constitutive activation of mTORC1 in endothelial cells leads to the development and progression of lymphangiosarcoma through
VEGF autocrine signaling. Cancer Cell. 2015;28:758-772.
30. Kurisetty V, Bryan BA. Aberrations in angiogenic signaling and MYC amplifications are distinguishing features of angiosarcoma. Angiol Open Access. 2013;
1:17309.
31. Gottlieb R, Serang R, Chi D, et al. Stewart-Treves syndrome. Radiol Case Rep. 2015;7:693.
32. Cahan WG, Woodard HQ, Higinbotham NL, et al. Sarcoma arising in irradiated bone: report of 11 cases. Cancer. 1948;1:3-29.
33. Cha C, Antonescu CR, Quan ML, et al. Long-term results with resection of radiation-induced soft tissue sarcomas. Ann Surg. 2004;239:903-909.
34. Gladdy RA, Qin LX, Moraco N, et al. Do radiation-associated soft tissue sarcomas have the same prognosis as sporadic soft tissue sarcomas? J Clin Oncol. 2010;
28:2064-2069.
35. Behjati S, Gundem G, Wedge DC, et al; ICGC Prostate Group. Mutational signatures of ionizing radiation in second malignancies. Nat Commun. 2016;7:12605.
36. Hadj-Hamou NS, Ugolin N, Ory C, et al. A transcriptome signature distinguished sporadic from postradiotherapy radiation-induced sarcomas. Carcinogenesis.
2011;32:929-934.
37. Doyle LA, Nowak JA, Nathenson MJ, et al. Characteristics of mismatch repair deficiency in sarcomas. Mod Pathol. 2019;32:977-987.
38. Mito JK, Mitra D, Barysauskas CM, et al. a comparison of outcomes and prognostic features for radiation-associated angiosarcoma of the breast and other
radiation-associated sarcomas. Int J Radiat Oncol Biol Phys. 2019;104:425-435.
39. Gonin-Laurent N, Gibaud A, Huygue M, et al. Specific TP53 mutation pattern in radiation-induced sarcomas. Carcinogenesis. 2006;27:1266-1272.
40. Gonin-Laurent N, Hadj-Hamou NS, Vogt N, et al. RB1 and TP53 pathways in radiation-induced sarcomas. Oncogene. 2007;26:6106-6112.
41. Lesluyes T, Baud J, Pérot G, et al. Genomic and transcriptomic comparison of post-radiation versus sporadic sarcomas. Mod Pathol. 2019;32:1786-1794.
42. Mentzel T, Schildhaus HU, Palmedo G, et al. Postradiation cutaneous angiosarcoma after treatment of breast carcinoma is characterized by MYC amplification
in contrast to atypical vascular lesions after radiotherapy and control cases: clinicopathological, immunohistochemical and molecular analysis of 66 cases. Mod
Pathol. 2012;25:75-85.
43. Guo T, Zhang L, Chang NE, et al. Consistent MYC and FLT4 gene amplification in radiation-induced angiosarcoma but not in other radiation-associated atypical
vascular lesions. Genes Chromosomes Cancer. 2011;50:25-33.
44. Bernstein JL, Haile RW, Stovall M, et al; WECARE Study Collaborative Group. Radiation exposure, the ATM Gene, and contralateral breast cancer in the women’s
environmental cancer and radiation epidemiology study. J Natl Cancer Inst. 2010;102:475-483.
45. Bjerkehagen B, Smeland S, Walberg L, et al. Radiation-induced sarcoma: 25-year experience from the Norwegian Radium Hospital. Acta Oncol. 2008;
47:1475-1482.
46. Riad S, Biau D, Holt GE, et al. The clinical and functional outcome for patients with radiation-induced soft tissue sarcoma. Cancer. 2012;118:2682-2692.
47. Davidson T, Westbury G, Harmer CL. Radiation-induced soft-tissue sarcoma. Br J Surg. 1986;73:308-309.
48. Ballinger ML, Pinese M, Thomas DM. Translating genomic risk into an early detection strategy for sarcoma. Genes Chromosomes Cancer. 2019;58:130-136.
49. Wang Z, Wilson CL, Easton J, et al. genetic risk for subsequent neoplasms among long-term survivors of childhood cancer. J Clin Oncol. 2018;36:2078-2087.
50. Zhang J, Walsh MF, Wu G, et al. Germline mutations in predisposition genes in pediatric cancer. N Engl J Med. 2015;373:2336-2346.

Secondary Sarcomas
51. Ballinger ML, Goode DL, Ray-Coquard I, et al; International Sarcoma Kindred Study. Monogenic and polygenic determinants of sarcoma risk: an international
genetic study. Lancet Oncol. 2016;17:1261-1271.
52. Evans DG, Baser ME, McGaughran J, et al. Malignant peripheral nerve sheath tumours in neurofibromatosis 1. J Med Genet. 2002;39:311-314.
53. Widemann BC, Italiano A. Biology and management of undifferentiated pleomorphic sarcoma, myxofibrosarcoma, and malignant peripheral nerve sheath
tumors: state of the art and perspectives. J Clin Oncol. 2018;36:160-167.
54. Marees T, Moll AC, Imhof SM, et al. Risk of second malignancies in survivors of retinoblastoma: more than 40 years of follow-up. J Natl Cancer Inst. 2008;
100:1771-1779.
55. Kim A, Stewart DR, Reilly KM, et al. malignant peripheral nerve sheath tumors state of the science: leveraging clinical and biological insights into effective
therapies. Sarcoma. 2017;2017:7429697.
56. Peltonen S, Kallionpää RA, Rantanen M, et al. Pediatric malignancies in neurofibromatosis type 1: a population-based cohort study. Int J Cancer. 2019;
145:2926-2932.
57. Bernthal NM, Putnam A, Jones KB, et al. The effect of surgical margins on outcomes for low grade MPNSTs and atypical neurofibroma. J Surg Oncol. 2014;
110:813-816.
58. Ferrari A, Bisogno G, Macaluso A, et al. Soft-tissue sarcomas in children and adolescents with neurofibromatosis type 1. Cancer. 2007;109:1406-1412.
59. Sung L, Anderson JR, Arndt C, et al. Neurofibromatosis in children with Rhabdomyosarcoma: a report from the Intergroup Rhabdomyosarcoma study IV.
J Pediatr. 2004;144:666-668.
60. Zöller ME, Rembeck B, Odén A, et al. Malignant and benign tumors in patients with neurofibromatosis type 1 in a defined Swedish population. Cancer. 1997;
79:2125-2131.
61. Broaddus E, Topham A, Singh AD. Incidence of retinoblastoma in the USA: 1975-2004. Br J Ophthalmol. 2009;93:21-23.
62. Knudson AG Jr. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci USA. 1971;68:820-823.
63. Friend SH, Bernards R, Rogelj S, et al. A human DNA segment with properties of the gene that predisposes to retinoblastoma and osteosarcoma. Nature. 1986;
323:643-646.
64. Dick FA, Rubin SM. Molecular mechanisms underlying RB protein function. Nat Rev Mol Cell Biol. 2013;14:297-306.
65. MacCarthy A, Bayne AM, Brownbill PA, et al. Second and subsequent tumours among 1927 retinoblastoma patients diagnosed in Britain 1951-2004. Br
J Cancer. 2013;108:2455-2463.
66. de Andrade KC, Mirabello L, Stewart DR, et al. Higher-than-expected population prevalence of potentially pathogenic germline TP53 variants in individuals
unselected for cancer history. Hum Mutat. 2017;38:1723-1730.
67. Li FP, Fraumeni JF Jr., Mulvihill JJ, et al. A cancer family syndrome in twenty-four kindreds. Cancer Res. 1988;48:5358-5362.
68. Schneider K, Zelley K, Nichols KE, et al. Li-Fraumeni Syndrome. In Adam MP, Ardinger HH, Pagon RA, (eds). GeneReviews. Seattle, WA: University of
Washington; 1993;1-21.
69. Mai PL, Best AF, Peters JA, et al. Risks of first and subsequent cancers among TP53 mutation carriers in the National Cancer Institute Li-Fraumeni syndrome
cohort. Cancer. 2016;122:3673-3681.
70. Ognjanovic S, Olivier M, Bergemann TL, et al. Sarcomas in TP53 germline mutation carriers: a review of the IARC TP53 database. Cancer. 2012;
118:1387-1396.
71. Hettmer S, Archer NM, Somers GR, et al. Anaplastic rhabdomyosarcoma in TP53 germline mutation carriers. Cancer. 2014;120:1068-1075.
72. Evans DG, Howard E, Giblin C, et al. Birth incidence and prevalence of tumor-prone syndromes: estimates from a UK family genetic register service. Am J Med
Genet A. 2010;152A:327-332.
73. Seminog OO, Goldacre MJ. Risk of benign tumours of nervous system, and of malignant neoplasms, in people with neurofibromatosis: population-based record-
linkage study. Br J Cancer. 2013;108:193-198.
74. Duong TA, Sbidian E, Valeyrie-Allanore L, et al. Mortality associated with neurofibromatosis 1: a cohort study of 1895 patients in 1980-2006 in France. Orphanet
J Rare Dis. 2011;6:18.
75. Rasmussen SA, Yang Q, Friedman JM. Mortality in neurofibromatosis 1: an analysis using U.S. death certificates. Am J Hum Genet. 2001;68:1110-1118.
76. Ylä-Outinen H, Loponen N, Kallionpää RA, et al. Intestinal tumors in neurofibromatosis 1 with special reference to fatal gastrointestinal stromal tumors (GIST).
Mol Genet Genomic Med. 2019;7:e927.
77. Doros L, Schultz KA, Stewart DR, et al. DICER1-Related Disorders. In Adam MP, Ardinger HH, Pagon RA, (eds). GeneReviews. Seattle, WA: University of
Washington; 1993;1-29.
78. Hill DA, Ivanovich J, Priest JR, et al. DICER1 mutations in familial pleuropulmonary blastoma. Science. 2009;325:965.
79. Kim J, Field A, Schultz KAP, et al. The prevalence of DICER1 pathogenic variation in population databases. Int J Cancer. 2017;141:2030-2036.
80. Schultz KAP, Williams GM, Kamihara J, et al. DICER1 and associated conditions: identification of at-risk individuals and recommended surveillance strategies.
Clin Cancer Res. 2018;24:2251-2261.
81. Brenneman M, Field A, Yang J, et al. Temporal order of RNase IIIb and loss-of-function mutations during development determines phenotype in pleuro-
pulmonary blastoma / DICER1 syndrome: a unique variant of the two-hit tumor suppression model. F1000 Res. 2015;4:214.

Eulo et al
82. Klein S, Lee H, Ghahremani S, et al. Expanding the phenotype of mutations in DICER1: mosaic missense mutations in the RNase IIIb domain of DICER1 cause
GLOW syndrome. J Med Genet. 2014;51:294-302.
83. Warren M, Hiemenz MC, Schmidt R, et al. Expanding the spectrum of dicer1-associated sarcomas. Mod Pathol. 2020;33:164-174.
84. de Kock L, Foulkes WD. Sarcoma and germ-line DICER1 mutations. Lancet Oncol. 2016;17:e470.
85. Stewart DR, Best AF, Williams GM, et al. Neoplasm risk among individuals with a pathogenic germline variant in DICER1. J Clin Oncol. 2019;37:668-676.
86. Wu MK, Vujanic GM, Fahiminiya S, et al. Anaplastic sarcomas of the kidney are characterized by DICER1 mutations. Mod Pathol. 2018;31:169-178.
87. Das A, Roy P, Modi SK, et al. Germline DICER1-mutant intracranial sarcoma with dual chondroid and spindle cell morphology and pulmonary metastases
treated with multimodal therapy. Pediatr Blood Cancer. 2019;66:e27744.
88. Lee JC, Villanueva-Meyer JE, Ferris SP, et al. Primary intracranial sarcomas with DICER1 mutation often contain prominent eosinophilic cytoplasmic globules
and can occur in the setting of neurofibromatosis type 1. Acta Neuropathol. 2019;137:521-525.
89. Sakaguchi M, Nakano Y, Honda-Kitahara M, et al. Two cases of primary supratentorial intracranial rhabdomyosarcoma with DICER1 mutation which may belong
to a “spindle cell sarcoma with rhabdomyosarcoma-like feature, DICER1 mutant. Brain Tumor Pathol. 2019;36:174-182.
90. Villani A, Shore A, Wasserman JD, et al. Biochemical and imaging surveillance in germline TP53 mutation carriers with Li-Fraumeni syndrome: 11 year follow-up
of a prospective observational study. Lancet Oncol. 2016;17:1295-1305.
91. Kratz CP, Achatz MI, Brugières L, et al. Cancer screening recommendations for individuals with Li-Fraumeni syndrome. Clin Cancer Res. 2017;23:e38-e45.
92. Mai PL, Khincha PP, Loud JT, et al. Prevalence of cancer at baseline screening in the National Cancer Institute Li-Fraumeni Syndrome cohort. JAMA Oncol.
2017;3:1640-1645.
93. Ballinger ML, Best A, Mai PL, et al. Baseline surveillance in Li-Fraumeni syndrome using whole-body magnetic resonance imaging: a meta-analysis. JAMA
Oncol. 2017;3:1634-1639.
94. Kamihara J, Bourdeaut F, Foulkes WD, et al. Retinoblastoma and Neuroblastoma Predisposition and Surveillance. Clin Cancer Res. 2017;23:e98-e106.
95. Friedman DN, Lis E, Sklar CA, et al. Whole-body magnetic resonance imaging (WB-MRI) as surveillance for subsequent malignancies in survivors of hereditary
retinoblastoma: a pilot study. Pediatr Blood Cancer. 2014;61:1440-1444.
96. Ahlawat S, Fayad LM, Khan MS, et al; REiNS International Collaboration Members 2016. Current whole-body MRI applications in the neurofibromatoses: NF1,
NF2, and schwannomatosis. Neurology. 2016;87 (suppl 1):S31-S39.
97. Evans DGR, Salvador H, Chang VY, et al. Cancer and central nervous system tumor surveillance in pediatric neurofibromatosis 1. Clin Cancer Res. 2017;
23:e46-e53.
98. Zhang L, Dessouky R, Xi Y, et al. Clinical value of multiparametric whole-body magnetic resonance imaging over whole-spine magnetic resonance imaging in
patients with neurofibromatosis type I. World Neurosurg. 2017;108:729-737.
99. Evans DGR, Salvador H, Chang VY, et al. Cancer and central nervous system tumor surveillance in pediatric neurofibromatosis 2 and related disorders. Clin
Cancer Res. 2017;23:e54-e61.
100. Schrader KA, Cheng DT, Joseph V, et al. Germline variants in targeted tumor sequencing using matched normal DNA.[published correction appears in JAMA
Oncol. 2016;2:279]. JAMA Oncol. 2016;2:104-111.

SYMPTOMS AND
SURVIVORSHIP

SYMPTOMS AND SURVIVORSHIP
Bringing Life to Death: The Need for Honest,

Compassionate, and Effective
End-of-Life Conversations
Amy R. MacKenzie, MD, FACP1 and Michelle Lasota, BSN, RN1
overview
Conversations about death and dying are a crucial part of all medical care and are particularly relevant in the
field of oncology. Patients express a desire to have discussions about goals of care, and many patients have
thought about their end-of-life (EOL) wishes but have not had an opportunity to openly talk with care providers
about this. Deficiencies in medical training, lack of confidence, limited time, and cultural barriers all
contribute to the paucity of these important discussions. Although physicians are often expected to lead these
conversations, nurses and nurse practitioners also play a vital role in the identification of opportunities to
address EOL goals and should be a resource for the care team in facilitating EOL conversations at all points on
the care continuum. Public engagement is paramount in normalizing conversations about death and dying,
and the health care system needs to partner with public health agencies and private groups to open dialogues
about EOL. Providers at all levels need improved education in having these difficult but essential
conversations.
INTRODUCTION these conversations are necessary and that they will

“Death isn’t a medical failure, it’s a biological certainty. change over time, yet implementation is often lacking.5
But poorly managed death IS a medical failure.” Research demonstrates that patients prefer early ini-
—Dr. Kathryn Mannix1 tiation of advance directive discussions; however,
many patients do not have these discussions with ei-
ther primary care physicians or specialists.5-7 Avoid-
“Dying is a gritty, difficult, and vulnerable human ex-
perience. But it isn’t simply to be suffered. Talking ance of clinician-led EOL conversations can deny the
about dying offers us opportunities to reconnect with patient and family the opportunity for meaningful plans
loved ones. Maybe these assumptions about dying are for the present. In addition, avoidance can contribute
not all true: that it’s the worst thing possible, that it’s to false hope and can rob patients of time to complete
always associated with suffering. Maybe it’s more in- things before death. A recent study conducted by
teresting than that. Perhaps we can see dying is the American Society for Clinical Oncology of 5,000
a developmental phase, like toddler-hood or adoles- American adults examined patients’ thoughts about
cence, a phase in life that offers opportunity for growth.” EOL care. Approximately one-half of these patients
—Dr. Ira Byock2
either had a personal diagnosis of cancer or were
There is a crucial need for effective and compassionate “touched by cancer” (immediate family member or
Author affiliations EOL conversations. In all of medicine, but particularly caregiver). Respondents overwhelmingly agreed that it
and support in oncology, developing a level of comfort around is important to have a plan in place and to commu-
information (if discussing death is essential. As providers, we must nicate with one’s doctor about EOL care wishes, with
applicable) appear
shepherd our patients through every aspect of their even greater agreement among those who have or had
at the end of this
article. illness and, perhaps most important, the transition cancer. Despite this, only 26% of respondents who
Accepted on from ill, to dying, to death. Goals-of-care conversations have/had cancer reported having discussed EOL care
February 21, 2020 can help clarify patients’ priorities and values and, with their doctor. One-half of cancer patients surveyed
and published at ideally, should be started early on. Unfortunately, these had a plan for EOL care, and the vast majority with
ascopubs.org on discussions often happen at emergent decision points a plan agreed that it was at least somewhat helpful.
March 24, 2020:
(e.g., need for life-sustaining treatment) or at a time Seventy percent of caregivers agreed with the state-
10.1200/EDBK_ when patients are within a few days to weeks of ment “I wish I had more information to help prepare me
279767 dying.3,4 Both oncologists and patients recognize that and my loved one for end-of-life care.”8

Improving End-of-Life Conversations
underprepared to engage in these conversations and to

manage their own feelings about death and dying.14 Resi-
dents value opportunities to learn from dying patients but
• Patients are interested in discussing their goals also experience feelings of guilt and failure after a patient’s
of care and end-of-life wishes.
death.15
• Education is generally lacking in end-of-life
communication but can be improved upon. A single-institution survey of U.S. medical residents dem-
onstrated that 88% of those surveyed had little to no
• Care providers experience barriers to serious
classroom training on EOL care during residency. Although
conversations at many levels.
these conversations were frequent during postgraduate
• Communication skills can be honed with guided training, they were largely unsupervised. By contrast,
courses online and in person. conversations were less frequent in medical school but
• Social media is a possible platform for dis- highly supervised. Sadly, slightly more than one-half of these
semination of information about end-of- residents reported minimal training in medical school. It is
life care. not surprising that physicians who had received education
and participated in frequent conversations had significantly
higher levels of ease with EOL conversations (p = .017 and
p = .003, respectively). Residents also reported a higher
Patient preference for EOL care communication is varied comfort level when accompanied by a history of adequate
and is dependent upon a number of factors. A systemic training.16 A similar, albeit small, survey of residents’ ex-
review of preferences showed wide variability in patients’ periences in training in New York revealed that only 16%
preference for high levels of information depending on the surveyed reported having received very good or outstanding
geographical location. However, relatively uniformly, pa- palliative care training. In addition, they cared for few pa-
tients and caregivers stated the importance of exploring tients at EOL and felt that clinical supervision was lacking.17
patients’ goals and expectations and a perceived willingness For physicians to be able to begin to navigate these con-
of a health care provider to initiate and engage in conver- versations, it is clear that adequate training is a must.
sations about death and to use the words “death” and
“dying.” Also universal was the opinion that interactions In the mid-2000s, a multidimensional, week-long, palliative
were considered negative when they were “too speedy” or if care curriculum was piloted with Boston University second-
the doctor was too tentative.9 year medical residents. This EOL curriculum improved at-
titudes, knowledge, and comfort level in caring for dying
It should not go unrecognized that approaching conver- patients, as measured by pre- and post-curriculum ques-
sations about death is quite complex and is rooted in pa-
tionnaires. In addition, this training provided an in situ
tients’ and families’ personal histories of loss, experiences
experience not had by most trainees before. This hands-on
of the dying process, and the family dynamic itself.10 The
training is vital to improving trainees’ comfort and readiness
unpredictability of living with and navigating a cancer di-
to participate in difficult conversations.18
agnosis is accompanied by grief, loss, and death long before
actual physical death occurs. Examples of such loss and Experiential learning can be very effective in education for
death can include but are not limited to daily routines, EOL care. Stilted role-play aside, drama-based education
personal image, relationships, intimacy, travel plans, family has been found to be effective and well received. The
planning, careers, and role of provider. It is evident that “Confessions of a Reluctant Caregiver” Palliative Educa-
these conversations are desired but are challenging. Anna tional Program is based on a play of the same name that
Beck of the Huntsman Cancer Institute describes our incorporates a staged reading of the play followed by
current efforts at EOL conversations as frequently “too little, a guided discussion with the audience. The play chronicles
too late, and not great.”11 Training medical professionals to the experience of a daughter caring for her father at the EOL
be better prepared in facilitating difficult conversations is after already having been her mother’s caregiver. Fifty
crucial.12 physicians who participated in this program were surveyed
as to the perceived usefulness of the program and relevance
TRAINING IN EOL CONVERSATIONS to everyday practice. More than 80% of physicians found
Historically, medical schools have offered courses in palli- the program to either “very relevant” or “extremely relevant”
ative care and, as of the mid-1990s, 90% of medical schools to their practice. Additionally, respondents found this
offer lectures on death and dying. Palliative care has grown method of learning about palliative care far more useful
as a subspecialty and, currently, 72% of hospitals in the than didactic lectures or journal articles. This could be
United States with 50 beds or more have a palliative care a very effective tool for increasing comfort with EOL
program.13 However, medical trainees still report feeling conversations. 19

MacKenzie and Lasota
TABLE 1. Training Programs in Communication and Palliative Care

Program Format Main Skills Taught Target Audience Website
Center to Advance Online training modules Palliative care Clinicians, social workers, case https://www.capc.org
Palliative Care managers, licensed
Toolkits Pain management
professional counselors
Communication skills
Symptom management
End-of-Life Nursing In-person training Nine-module training in Nurses, nurse practitioners https://www.aacnnursing.org/
Education various aspects of palliative ELNEC
“Train-the-trainer” course
Consortium (ELNEC) care
Education on Palliative Classroom presentation Communication skills and Clinicians, social workers, case https://www.bioethics.
and End-of-Life Care clinical competencies in managers, interdisciplinary northwestern.edu/programs/
palliative care team members epec/index.html
Respecting Choices Online training Communication skills Individual clinicians, larger https://respectingchoices.org
health systems
Onsite courses Design and implementation
strategies
Continuing education Advance care planning
Serious Illness Care Online forum Use of serious illness Clinicians, social workers, https://www.ariadnelabs.org/
Program at Ariadne conversation guide to interdisciplinary team areas-of-work/serious-
Webinar
Labs facilitate conversations members illness-care/
VitalTalk Online courses Communication skills for Physicians, advanced practice http://vitaltalk.org
serious illness and end-of- providers, intensive care unit
Workshops
life conversations nurses
Smartphone app
Confessions of Drama-based program End-of-life communication Physicians [email protected] for
a Reluctant involving staged reading information
Breaking bad news
Caregiver and guided discussion
Other interactive programs exist to improve clinician comfort Findings from a 2007 qualitative study of Norwegian pa-
with difficult conversations and are listed in Table 1. tients with cancer showed that patients were receptive to the
offer for communication but not always able to engage as
NURSING INTERVENTIONS an inpatient. Despite this, the article reminds us that “the
Inevitably, nurses in every specialty will be required to care nurses must not be afraid to talk about death. They must be
for dying patients at some time throughout their career. prepared for and able to handle patients’ difficult feelings;
These care experiences are far more frequent within the that they are afraid to die. The nurses need knowledge about
specialty of oncology nursing. It is true that, with the con- these things.”21 It is this fortitude to have the serious dis-
tinuing advances of modern medicine and the complex cussion that can position nurses to advocate for patients at
evolution of cancer care, people are living longer with a vital time in their cancer journey.
cancer and, as a result, people may be dying more slowly
Over the past 40 years, the change in nursing education has
but with equal level of complexities in care requirements.
been important. In the 1970s and 1980s, few nursing
There is an important role that nurses can play in recog- programs offered required or elective content in the care of
nizing opportunities to begin EOL conversations.19 Histori- the dying. Textbooks, too, had either minimal or no EOL
cally, the responsibility of introducing and navigating such content.22 One of the most important contributions to
crucial conversations with patients was a physician-driven nursing education in EOL care is the End-of-Life Nursing
initiative, often diminishing the role of the nurse in goals of Education Consortium, developed with the American As-
care and EOL conversations. Although physicians often sociation of Colleges of Nurses and researchers at City of
deliver technical news or offer treatment options, nurses are Hope in California. This nine-module program was de-
in a unique position to engage with families beyond the veloped in February 2000 in response to deficiencies
confines of technical medical information. Unfortunately, identified locally at City of Hope and nationally by the As-
nurses are not often trained in initiating or leading these sociation of Colleges of Nurses in the report “Peaceful
conversations.20 Proper training and education help in- Death: Recommended Competencies and Curriculum
crease nurses’ ease and competency in the delivery of Guidelines for End-of-Life Care.”23 The modules include
quality EOL care. Nursing Care at the EOL, Pain Management, Symptom

Management, Ethical/Legal Issues, Grief/Loss/Bereave- another, physicians were generally found to avoid emotions
ment, Achieving Quality Care, and Care at the Time of by explaining things medically. (2) Difficult or sensitive
Death. Nurses not only learn skills through lectures, role- discussions are perceived to lengthen medical encounters
play, small-group work, and case discussion but also ex- and often do. Visit times were increased by 6 to 7 minutes on
perience a “train-the-trainer” module to disseminate the average when serious palliative care discussions occurred.
educational content.24 As of 2019, the End-of-Life Nursing (3) Physicians spend more time speaking in these con-
Education Consortium has trained more than 24,000 pro- versations. Five of the 20 studies reviewed in this article
fessionals in all 50 states and 100 countries. Those who reported that physicians used up to 70% of the time
received training have gone on to train 720,000 professionals speaking in these conversations. In addition, physicians
in their own communities. were more likely to reframe than validate the heightened
BARRIERS TO CONVERSATIONS emotions. (4) Patient/family satisfaction is associated
with supportive physician behaviors. Both frequent re-
Several barriers to successful EOL conversations have been assurances about not abandoning the patient at the EOL
identified. As previously mentioned, the lack of training in and increased time for families to speak during conferences
such endeavors leads many young attending physicians to were found to increase family and patient satisfaction.6
feel inadequately prepared. The Royal College of Physicians
conducted five workshops with 100 medical professionals Nurses experience barriers to EOL conversations as well. In
from student level to physicians with advanced training to a 2017 cross-sectional survey conducted with oncology
better understand the barriers to EOL conversations.25 nurses in Lithuania, nurses were asked about obstacles to
Through these workshops, the Royal College of Physicians providing EOL care. The major obstacles identified by the
identified three main barriers to having EOL conversations: 239 oncology nurses were the following: first, the nurse’s
culture, confidence, and practicalities. opinion on patient care was not welcome, valued, or dis-
cussed. Second, nurses felt that families did not have access
Physicians in this study believed that there is a culture in
to emotional or psychological support after a conversation
health care to see death as a failure and a need among
about the diagnosis. A third obstacle was working with
medical providers to “always do something.” There was also
physicians who avoided conversations about diagnosis and
a belief that the public may see modern medicine as able to
prognosis.29
“fix anything.” Doctors at all levels of practice expressed
anxiety about initiating conversations that may involve EOL An earlier study of critical care nurses’ perceptions dem-
care. Prognostic uncertainty in complex patients was onstrated similar findings, although, in this study, nurses
a frequent factor in feelings of discomfort and has been cited behavior of patients’ families that removed them from
reported elsewhere as a key reason to avoid initiating caring directly for patients in addition to families’ lack of
conversations about dying.26 Confidence, or lack thereof, understanding and physician disagreement with the plan of
has a marked impact on the ability to initiate a sensitive care.30 Additional barriers to delivering EOL care as iden-
conversation. tified by oncology nurses across multiple studies included
but were not limited to physician avoidance, dialectic ten-
Practical barriers to conversations are numerous. In-
sion, lack of knowledge and training on treating grieving
creasing complexity of patient illness, with the resultant
family members, lack of skills for providing empathy, and
long list of specialists, can make it difficult to identify which
perceived institutional barriers.31
physician is truly “in charge” of the discussion. 25 In-
terestingly, in a multicenter study in Canada, physicians The American Nurses Association released a position
identified patient- and family-related issues as the greatest statement in 2016 to identify the roles, responsibilities, and
barriers. These included difficulty on the part of the patient opportunities for nurses providing EOL care as well as to
and the family to accept a poor prognosis, family members’ provide direction in carrying out these responsibilities.
difficulty in understanding the limitations and complications Standards for practice, education, research, and adminis-
of life-sustaining treatment, and lack of agreement among tration are delineated. The mandate is clear that nurses
family members.27,28 must provide this care, but is the health system allowing
space for this to happen?
Directly observed interactions can provide us with a unique
perspective on how physicians and families behave in these Media can present a barrier to realistic EOL conversations
circumstances. In a systemic review of observed patient/ but may also offer an opportunity to educate patients and
family-physician discussions surrounding EOL care, several families. Depictions of death, dying, and miraculous re-
themes emerged regarding these interactions: (1) Physi- covery are often delivered through a frosted Hollywood lens,
cians avoid emotional issues and focus on medical or thus blurring the public’s interpretation of the limits of
technical ones. The impact of palliative chemotherapy on modern medicine. A 2015 study evaluated two well-known
social aspects of life was rarely discussed in one study; in medical television dramas for their depiction of cardiopulmonary

resuscitation. In more than 91 episodes in which cardio- Prize–winning writer Ellen Goodman started this project
pulmonary resuscitation was performed 46 times in the after spending several years caring for her mother at the end
hospital, approximately 70% survived and most survived to of her life. As she states, “I realized only after her death how
discharge (72%).32 In 2016, according to the American much easier it would have all been if I heard her voice in my
Heart Association, only about 25% of adults who were re- ear as these decisions had to be made.”37 The Conversation
suscitated survived to discharge. The number is as low as Project offers “starter kits” to help facilitate conversations
10% for those who received cardiopulmonary resuscitation about EOL wishes as well as tools for health care pro-
out of the hospital. More than 11,000 participants in fessionals, faith organizations, and the general community.
a Howard University School of Communications study were Death Over Dinner is a second initiative that was started to
surveyed about television watching habits and how this open the public to feeling more comfortable discussing EOL
might influence beliefs about illness. The study found that issues. “Let’s Have Dinner and Talk About Death,” started
heavy viewers of medical dramas underestimated the se- by Michael Hebb and Scott Macklin as a graduate course
riousness of chronic illness.33 For physicians who are at the University of Washington, has grown into an in-
working toward honest conversations about serious illness, ternational project with an Australian edition, Brazilian
this can be an important barrier. On the other hand, ref- edition, Jewish edition, and health care edition. The un-
erencing well-known television shows or movies to illustrate derlying notion is that the dinner table is a comfortable place
the opportunity for “a good death” can be just as powerful. to start difficult conversations. Thus the inviting phrase,
PUBLIC ENGAGEMENT “Let’s have dinner…,” can perhaps assuage some fears
about embarking on tough topics. The website provides
It is imperative as clinicians that we offer extended guidance
tools to organize and lead a dinner, including an invitation
and support to the public by way of informative resources
email, numerous articles, other media to review, and a guide
about death, the dying process, and the power of conver-
with tips to hosting.38
sation. Ideally, a proactive approach would introduce the
public to a pathway of information gathering in a safe, Other projects and websites that encourage awareness can
nonthreatening manner via structured forums, far in ad- be found in Sidebar 1.
vance of serious illness or crisis. However, in situations that INTERNATIONAL EFFORTS
don’t allow the option of processing this information at
In the United Kingdom in 2018, the Royal College of
leisure, the value of the information obtained does not di-
Physicians undertook a year-long effort to improve public
minish amid crisis. Providing a road map to reliable EOL
health awareness and to drive further research. One of the
resources can help reduce anxiety and the fear that can be
products of this campaign was “Talking About Dying: How
caused by potentially conflicting information found online.
to Begin Honest Conversations About What Lies Ahead.”
Instead, we can guide the public to welcoming resources,
This project engaged the public to bring recommendations
deliver reliable information, and extend invitation for vol-
to the health care system that better reflect the goals and
untary participation in tangible gatherings that serve to
desires of patients.25 A second U.K. initiative, The Departure
normalize their individual and collective experiences sur-
Lounge, ran from May 2019 to December 2019 as a “pop-
rounding death, dying, and the conversation.
up” space meant to mimic a departure lounge in an airport.
In 1997, the Institute of Medicine (IOM) published a report This space was created to allow the public to access re-
entitled “Approaching Death,” which suggested the im- search and have conversations about death and dying.39
portance of “a continuing public discussion…to develop Although less recent, the Project on Death in America (part
a better understanding of the modern experience of of the Open Society Initiative) led to the formation of an
dying….”34 This notion begins with assessing the public International Palliative Care Initiative to enhance palliative
understanding of vocabulary and concepts about EOL. In care in Central Europe, Eastern Europe, the former Soviet
a 2011 survey by the Center to Advance Palliative Care, 78% Union, and areas of South Africa, which are underserved.40
of Americans did not know what palliative care was.35 In
a 2012 California Health Care Foundation survey, 83% of Projects in Australia include the Healthy EOL Project,
respondents had not heard of palliative care.36 In addition, a public health approach to create community models for
many medical professionals are not clear about the differ- EOL care,41 and the GroundSwell Project, a not-for-profit
ence between palliative care and EOL care, which can organization designed to help people better understand
include hospice. Engagement with the public on several death and grieving.42
levels is important to improve understanding. FUTURE DIRECTIONS
An American initiative that was developed to allow patients’ The IOM produced a report in 2015 entitled “Dying in
wishes to be expressed and respected at the EOL is a col- America: Improving Quality and Honoring Individual Pref-
laborative effort known as The Conversation Project. Pulitzer erences Near the End of Life.” 43 In 2009, while the

SIDEBAR 1. ON END OF LIFE, DEATH, AND DYING

• Aging With Dignity (http://agingwithdignity.org)
• The Art of Dying Well (https://www.artofdyingwell.org)
• Before I Die (http://beforeidie.cc)
• Community Conversations on Compassionate Care (Compassion and Support; https://www.compasionandsupport.org)
• Death Café (http://www.deathcafe.com)
• Death Clock (http://www.deathclock.com)
• Death Over Dinner (http://www.deathoverdinner.org)
• Death With Dignity (https://www.deathwithdignity.org)
• DeathWise (https://www.deathwise.org)
• Dying Matters (https://www.dyingmatters.org)
• Engage With Grace (http://www.engagewithgrace.org)
• Life Before Death: The Lien Foundation (http://www.lifebeforedeath.com/index.shtml)
• National Healthcare Decisions Day (http://www.nhdd.org)
• Project Compassion (http://project-compassion.org)
• The Conversation Project (http://theconversationproject.org)
• The Order of the Good Death (www.orderof thegooddeath.com)
Affordable Care Act was taking shape, a bipartisan group • All individuals should be encouraged to participate ac-
of representatives sponsored a provision (Section 1233 tively in their health care, including decision-making as
of HR 3200, 11th Congress) that would allow Medicare they approach death.
reimbursement for physicians discussing living wills, • Clinicians should begin and continue conversations to
advance directives, and EOL care. Unfortunately, this integrate goals into care plans of patients as they evolve.
was politicized and manipulated into the term “death • Palliative care education should be offered at all stages of
panel,”44 which led to widespread fear that the govern- professional education, and accrediting bodies should
ment would take over decisions about EOL care and require knowledge and competency in palliative care
would advocate for euthanasia. When faced with this kind domains.
of jargon, it is not surprising that there are barriers to good • National policy must reflect the need to expand financing
conversations about EOL care.45 With the goals of im- of quality, equitable care at the EOL.
proving the national discourse on death and dying, the • Public and private payers should improve the portability
IOM undertook a large study of the gaps in care sur- of advance care planning documents through improved
rounding this topic. electronic health care records.
This report highlights a number of key issues: (1) the de- • Public health, governmental, community-based, faith-
livery of person-centered, family-oriented EOL care; (2) based, and consumer organizations should provide
clinician-patient communication and advance care plan- fact-based information about care options, encourage
ning; (3) professional education and development; (4) dialogue, and discourage misinformation.
policies and payment systems to support high-quality EOL Education is the cornerstone of beginning to implement the
care; and (5) public education and engagement. Although recommendations made by the IOM. A recent white paper in
the scope of this manuscript does not allow an in-depth the Journal of Clinical Oncology addressing patient-clinician
review of the almost 700-page report, the summary points communication issues in palliative care identifies de-
are well worth noting: ficiencies and pitfalls in communication. Some of these
• To provide person-centered, comprehensive care, en- include introducing “palliative care” as “end-of-life” care,
suring bodies should cover comprehensive care at the a lack of understanding about how to keep hope while
EOL with access to skilled palliative care and should take discussing death, using medical jargon, missing emotional
into account individual values and goals. cues, and avoiding discussions about death and dying.

Opportunities to improve on these deficiencies include media platform, has been host to a number of #lend-
specific communication skills, using a team approach, and yourinstagram posts on behalf of Singapore Hospice
partnering with families and caregivers. Clinician education, Council. Users of Instagram are posting clips of patients in
promoting a culture of reflection and support, and aligning hospice as part of the Live Well, Leave Well campaign,
system level changes with clinician changes are all sug- meant to encourage conversation about death and debunk
gestions to improve the complicated communication that myths about hospice.48
occurs at the EOL.28
The dissemination and implementation of palliative care CONCLUSION
services needs to adapt to the increasing complexity of Addressing death and dying is not easy. As medical pro-
patients, the growing population of aging patients, and fi- fessionals, we spend our lives aiming toward success. The
nancial pressures to keep patients out of acute care settings. death of a patient can feel like a failure, and that pushes us
Current models of care can, and will have to, adapt to in- away from the experience rather than toward it. Talking
clude palliative care more readily. Best practices have been about death can result in improved clinical relationships.
suggested, including maximizing the role of the nurse and Grief advocate Megan Devine explains, “The issue is, we
the advanced practice provider, reducing the demarcation don’t know what to do with other peoples’ pain so we try to
that exists between treatment and supportive care and in- be optimistic or look on the bright side.” In this way, we fail to
corporation of more palliative care in clinical trials.46 acknowledge grief. “Grief is not something to be cured; it is
Social media is one of the defining changes of the 21st a state of being.”49
century and has a definite place in the landscape of Our duty to our patients is to be present for them. We must
medicine and public engagement. Twitter has both ad- educate ourselves, provide resources, and listen.
vantages and disadvantages in the world of medicine. Fa-
“The point, I believe, is not that we should just be talking
cilitating knowledge and building support networks are pros
about death or tweeting about it [for its own sake], but that
of engaging online; an overflow of information, inaccurate a fuller awareness of one’s death makes life more mean-
information, and the threat of one’s professional standing ingful. The best use of the technology is to share stories and
are conceivable cons.47 That being said, Twitter can be to reach out to other people in real time. Death is one of
a platform that reaches members of the public who may the few universals that we have. It brings us together.”
otherwise not engage. Instagram, another popular social —Dr. Lawrence R. Samuel50

1
Thomas Jefferson University, Sidney Kimmel Cancer Center, AND DATA AVAILABILITY STATEMENT
Philadelphia, PA Disclosures provided by the authors and data availability statement (if
EDBK_279767.
Amy R. MacKenzie, MD, FACP, Thomas Jefferson University, Sidney
Kimmel Cancer Center, 925 Chestnut St., Suite 420, Philadelphia, PA
REFERENCES
1. Pilgrims Hospices. In Conversation With Dr. Kathryn Mannix. www.pilgrimshospices.org/news/kathryn-mannix-interview/. Accessed February 17, 2020.
2. Byock I. Reframing Our Conversation and Perspective on Death and Dying With Dr. Ira Byock #future of health. www.thestatenislandfamily.com/reframing-our-
conversation-and-perspective-on-death-and-dying-with-dr-ira-byock/. Accessed February 17, 2020.
3. Mack JW, Cronin A, Taback N, et al. End-of-life care discussions among patients with advanced cancer: a cohort study. Ann Intern Med. 2012;156(3):204-210.
4. Borreani C, Miccinesi G. End of life care preferences. Curr Opin Support Palliat Care. 2008;2:54-59.
5. Schulman-Green D, Smith CB, Lin JJ, et al. Oncologists’ and patients’ perceptions of initial, intermediate, and final goals of care conversations. J Pain Symptom
Manage. 2018;55:890-896.
6. Fine E, Reid MC, Shengelia R, et al. Directly observed patient-physician discussions in palliative and end-of-life care: a systematic review of the literature. J Palliat
Med. 2010;13:595-603.
7. Childers JW, Arnold RM. Expanding goals of care conversations across a health system: the Mapping the Future Program. J Pain Symptom Manage. 2018;56(4):
637-644.

8. ASCO. ASCO National Cancer Opinion Survey, 2019. www.asco.org/research-guidelines/reports-studies/national-cancer-opinion-survey. Accessed December
11, 2019.
9. Parker SM, Clayton JM, Hancock K, et al. A systematic review of prognostic/end-of-life communication with adults in the advanced stages of a life-limiting illness:
patient/caregiver preferences for the content, style, and timing of information. J Pain Symptom Manage. 2007;34:81-93.
10. Waldrop DP. At the eleventh hour: psychosocial dynamics in short hospice stays. Gerontologist. 2006;46:106-114.
11. Beck AC. Using the serious illness conversation guide. End of life care: are you ready? April 4, 2016. https://healthinsight.org/tools-and-resources/send/50-in-
person-events/298-eol-using-the-serious-illness-conversation-guide.
12. Sutherland R. Dying well-informed: the need for better clinical education surrounding facilitating end-of-life conversations. Yale J Biol Med. 2019;92:757-764.
13. Hospice News. Home page. www.hospicenews.com. Accessed January 20, 2020.
14. Sullivan A, Lakoma M, Block S. The status of medical education in end of life care. J Gen Intern Med. 2003;18:685-695.
15. Sattar S, Alibhai SMH, Wildiers H, et al. How to implement a geriatric assessment in your clinical practice. Oncologist. 2014;19:1056-1068.
16. Schmit JM, Meyer LE, Duff JM, et al. Perspectives on death and dying: a study of resident comfort with end-of-life care. BMC Med Educ. 2016;16(1):297.
17. Schwartz CE, Goulet JL, Gorski V, et al. Medical residents’ perceptions of end-of-life care training in a large urban teaching hospital. J Palliat Med. 2003;6:37-44.
18. Yacht AC, Suglia SF, Orlander JD. Evaluating an end-of-life curriculum in a medical residency program. Am J Hosp Palliat Care. 2007;23:439-446.
19. Kerr AM, Kachmar U, Palocko B, et al. “Confessions of a Reluctant Caregiver” Palliative Educational Program: the results of a survey assessing physicians’
perceptions of drama-based education for end-of-life care. J Cancer Educ. Epub 2019 Nov 20.
20. Putman-Casdorph H, Drenning C, Richards S, et al. Advance directives: evaluation of nurses’ knowledge, attitude, confidence, and experience. J Nurs Care Qual.
2009;24:250-256.
21. Kvåle K. Do cancer patients always want to talk about difficult emotions? A qualitative study of cancer inpatients communication needs. Eur J Oncol Nurs. 2007;
11:320-327.
22. DʼAntonio J. End-of-life nursing care and education---end-of-life nursing education: past and present. J Christ Nurs. 2017;34:34-38.
23. American Association of Colleges of Nursing, Robert Wood Johnson Foundation. Peaceful Death: Recommended Competencies and Curricular Guidelines for
End-of-Life Nursing Care. Washington, DC: American Association of Colleges of Nursing; 1998.
24. Ferrell BR, Virani R, Grant M, et al. Evaluation of the End-of-Life Nursing Education Consortium undergraduate faculty training program. J Palliat Med. 2005;
8:107-114.
25. Cogle K, Bailey S, Royal College of Physicians. Talking About Dying: How to Begin Honest Conversations About What Lies Ahead. www.rcplondon.ac.uk/projects/
outputs/talking-about-dying-how-begin-honest-conversations-about-what-lies-ahead. Accessed January 31, 2020.
26. Krawczyk M, Gallagher R. Communicating prognostic uncertainty in potential end-of-life contexts: experiences of family members. BMC Palliat Care. 2016;15(1):
59.
27. You JJ, Downar J, Fowler RA, et al. Barriers to goals of care discussions with seriously ill hospitalized patients and their families: a multicenter survey of clinicians.
JAMA Intern Med. 2015;175:549-556.
28. Back AL. Patient-clinician communication issues in palliative care for patients with advanced cancer. J Clin Oncol. Epub 2020 Feb 5.
29. Blaževičienė A, Newland JA, Čivinskienė V, et al. Oncology nurses’ perceptions of obstacles and role at the end-of-life care: cross sectional survey. BMC Palliat
Care. 2017;16:74.
30. Beckstrand RL, Kirchhoff KT. Providing end-of-life care to patients: critical care nurses’ perceived obstacles and supportive behaviors. Am J Crit Care. 2005;
14:395-403.
31. Banerjee SC, Manna R, Coyle N, et al. Oncology nurses’ communication challenges with patients and families: a qualitative study. Nurse Educ Pract. 2016;
16:193-201.
32. Portanova J, Irvine K, Yi JY, et al. It isn’t like this on TV: revisiting CPR survival rates depicted on popular TV shows. Resuscitation. 2015;96:148-150.
33. Chung JE. Medical dramas and viewer perception of health: testing cultivation effects. Hum Commun Res. 2014;40:333-349.
34. Institute of Medicine. Approaching death: improving care at the end of life. Health Serv Res. 1998;33:1-3.
35. Center to Advance Palliative Care. 2011 Public Opinion Research on Palliative Care: A Report Based on Research by Public Opinion Strategies. https://media.
capc.org/filer_public/18/ab/18ab708c-f835-4380-921d-fbf729702e36/2011-public-opinion-research-on-palliative-care.pdf. Accessed February 1, 2020.
36. California Health Care Foundation. Final chapter: Californians’ attitudes and experiences with death and dying. www.chcf.org/publication/final-chapter-
californians-attitudes-and-experiences-with-death-and-dying/. Accessed February 1, 2020.
37. The Conversation Project. Home page. theconversationproject.org. Accessed February 15, 2020.
38. Death Over Dinner. Home page. deathoverdinner.org. Accessed February 15, 2020.
39. The Health Foundation. The Departure Lounge: A Project to Enable People to Talk More Openly About Death, Dying, and Our Aging Population. www.health.org.
uk/funding-and-partnerships/programmes/the-departure-lounge. Accessed February 15, 2020.
40. Open Society Foundations. Transforming the Culture of Dying: The Project on Death in America 1994-2003. www.opensocietyfoundations.org/publications/
transforming-culture-dying-project-death-america-1994-2003. Accessed February 15, 2020.

41. Grindrod A, Rumbold B. Healthy End of Life Project (HELP): a progress report on implementing community guidance on public health palliative care initiatives in
Australia. Ann Palliat Med. 2018;7 (suppl 2):S73-S83.
42. The GroundSwell Project. Manifesto. www.thegroundswellproject.com/manifesto. Accessed February 16, 2020.
43. Committee on Approaching Death: Addressing Key End of Life Issues, Institute of Medicine. Dying in America: Improving Quality and Honoring Individual
Preferences Near the End of Life. Washington, DC: National Academies Press; 2015.
44. Palin S. Statement on the Current Health Care Debate. www.facebook.com/note.php?note_id=113851103434. Accessed January 15, 2020.
45. Frankford DM. The remarkable staying power of death panels. J Health Polit Policy Law. 2015;40:1087-1101.
46. Ferrell BR, Chung V, Koczywas M, et al. Dissemination and implementation of palliative care in oncology. J Clin Oncol. Epub 2020 Feb 5.
47. Choo EK, Ranney ML, Chan TM, et al. Twitter as a tool for communication and knowledge exchange in academic medicine: a guide for skeptics and novices. Med
Teach. 2015;37:411-416.
48. Asiaone. #Lendyourinstagram’s Got Singaporeans Talking About Hospice Care. www.asiaone.com/singapore/lendyourinstagrams-got-singaporeans-talking-
about-hospice-care. Accessed February 16, 2020.
49. Kottke.org. How Do You Help a Grieving Friend [online video]. https://kottke.org/18/10/how-do-you-help-a-grieving-friend. Accessed February 17, 2020.
50. Bisceglio P. "How social media is changing the way we approach death." The Atlantic, August 20, 2013. www.theatlantic.com/health/archive/2013/08/how-
social-media-is-changing-the-way-we-approach-death/278836/. Accessed February 16, 2020.

ASCO Educational Book 2020

Uploaded by

Copyright:

Available Formats

ASCO Educational Book 2020

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ASCO Educational Book 2020

Uploaded by

Copyright:

Available Formats

American Society of Clinical Oncology

UNITE AND CONQUER:

A PEER-REVIEWED, MEDLINE-INDEXED PUBLICATION

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

© 2020 by American Society of Clinical Oncology Alexandria, VA

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

CANCER PREVENTION, RISK REDUCTION, AND GENETICS

CARE DELIVERY AND REGULATORY POLICY

ii 2020 ASCO EDUCATIONAL BOOK | asco.org/edbook

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

CENTRAL NERVOUS SYSTEM TUMORS

DEVELOPMENTAL THERAPEUTICS—MOLECULARLY TARGETED AGENTS AND TUMOR BIOLOGY

GASTROINTESTINAL CANCER—COLORECTAL AND ANAL

GASTROINTESTINAL CANCER—GASTROESOPHAGEAL, PANCREATIC, AND HEPATOBILIARY

asco.org/edbook | 2020 ASCO EDUCATIONAL BOOK iii

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

GENITOURINARY CANCER—KIDNEY AND BLADDER

GENITOURINARY CANCER—PROSTATE, TESTICULAR, AND PENILE

iv 2020 ASCO EDUCATIONAL BOOK | asco.org/edbook

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

HEAD AND NECK CANCER

HEALTH SERVICES RESEARCH AND QUALITY IMPROVEMENT

asco.org/edbook | 2020 ASCO EDUCATIONAL BOOK v

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

PROFESSIONAL DEVELOPMENT AND EDUCATION ADVANCES

SYMPTOMS AND SURVIVORSHIP

vi 2020 ASCO EDUCATIONAL BOOK | asco.org/edbook

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

asco.org/edbook | 2020 ASCO EDUCATIONAL BOOK vii

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

2020 ASCO MEETINGS AND MEMBER PUBLICATIONS

MANAGING EDITOR EDITORIAL OFFICE

viii 2020 ASCO EDUCATIONAL BOOK | asco.org/edbook

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

COPYRIGHT 2020 INDIVIDUAL PRICES

2020 ASCO EDUCATIONAL BOOK | asco.org/edbook ix

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

Please email [email protected] with speciﬁc questions or concerns.

x ASCO EDUCATIONAL BOOK | asco.org/edbook

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

Tatiana Prowell, MD – Chair DEVELOPMENTAL THERAPEUTICS — MOLECULARLY

DEVELOPMENTAL THERAPEUTICS — GERIATRIC ONCOLOGY

2020 ASCO EDUCATIONAL BOOK | asco.org/edbook xi

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

GYNECOLOGIC CANCER PROFESSIONAL DEVELOPMENT AND EDUCATION

xii 2020 ASCO EDUCATIONAL BOOK |asco.org/edbook

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

2020 ASCO EDUCATIONAL BOOK | asco.org/edbook xiii

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

xiv 2020 ASCO EDUCATIONAL BOOK | asco.org/edbook

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

2020 ASCO EDUCATIONAL BOOK | asco.org/edbook xv

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

Don S. Dizon, MD, FACP, FASCO

2020 ASCO EDUCATIONAL BOOK | asco.org/edbook 1

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

Escalating and De-escalating Therapy for Early-Stage HER2-Positive Breast Cancer

Downloaded from ascopubs.org by 113.193.220.68 on June 22, 2020 from 113.193.220.068

invited article overview Escalating and De-escalating Therapy for

INTRODUCTION binds to the HER2 extracellular domain, causing HER2