Cheng H Et Al - Am Soc Clin Oncol Educ Book - 2018

CHENG ET AL
Practical Methods for Integrating Genetic Testing Into

Clinical Practice for Advanced Prostate Cancer
Heather Cheng, MD, PhD, Jacquelyn Powers, LCGC, Kerry Schaffer, MD, and Oliver Sartor, MD
OVERVIEW
Recent advances clearly demonstrate the potential clinical relevance of germline genetic testing and somatic genomic pro-
filing in identifying possible therapeutic and/or clinical trial options, particularly in advanced prostate cancer. In addition,
if a germline genetic mutation/pathogenic variant is identified, there may be important family implications and possible
life-saving changes to healthcare management. However, there is substantial debate and uncertainty about how best to
offer genetic testing services, which tests to use, which patients to test, what sequence of testing, what timing, by whom,
and with what kind of follow-up. To help address this new area of potential benefit and confusion, we provide a practical
overview of recent advances, discuss options and considerations for both germline and somatic testing, and offer practical
Downloaded from ascopubs.org by 143.58.229.255 on March 13, 2024 from 143.058.229.255
advice on what providers should understand before referring and/or ordering testing, key discussion points for patients
and families, and available genetics resources.
Copyright © 2024 American Society of Clinical Oncology. All rights reserved.
O ver the past several years, it has become abundantly

clear that molecular events not only can be defined in
prostate cancer but also can affect the treatment of patients
Unexpectedly, 8% of the DNA repair mutations were found
in the germline DNA, the first indication that there may be
a higher than previously recognized prevalence of cancer
and can have important implications for family members. predisposition among men with metastatic prostate cancer.
A substantial proportion of aggressive prostate cancers These discoveries represented major strides in the field and
(approximately 25%) have characteristic DNA repair alter- a new era marked by greater interplay between somatic
ations that suggest precision therapy opportunities; ap- genomics and germline genetics.
proximately 10% have alterations found in the germline
and potentially represent inherited cancer predisposition. HOMOLOGOUS RECOMBINATION DNA
These findings have led to rapid and far-reaching changes REPAIR DEFECTS AND NEW TREATMENT
to the field that include exciting new precision treatment STRATEGIES
opportunities and increased responsibility and challenges Specific treatment strategies could quickly be gleaned from
surrounding tumor genomic profiling and genetic counsel- research in other BRCA1/2-associated cancers, particularly
ing around inherited cancer risk. breast cancer and ovarian cancer. The phase II TOPARP-A
study showed that 14 of 16 patients (88%) with heavily pre-
GENOMICS VERSUS GENETICS AND treated metastatic castration-resistant prostate cancer and
IMPLICATIONS FOR PROSTATE CANCER CARE somatic alterations in DNA repair genes achieved objective
In 2015, The Cancer Genome Atlas Research Network re- responses to the PARP inhibitor olaparib.3 The identified
ported findings from 333 primary prostate cancers and the genes involved included BRCA2, ATM, and BRCA1, among
identification of 19% of primary tumors with mutations in others (Table 1). The responses were sustained for a large
DNA repair genes, including 3% in the homologous recom- portion of patients (in many, for more than 6 months). This
bination repair gene, BRCA2.1 That same year, the Inter- trial highlights the strong potential therapeutic role for
national Stand Up to Cancer/Prostate Cancer Foundation/ PARP inhibitor therapy in tumors with DNA repair defects
American Association for Cancer Research Prostate Cancer and has led to a number of subsequent clinical trials testing
Dream Team applied exome sequencing to 150 metastatic PARP inhibitors for biomarker-selected (i.e., DNA repair–
biopsies and found that 23% of metastatic prostate cancers deficient) advanced prostate cancer.
carry alterations in genes critical for DNA repair, again in- Although PARP inhibitors are currently garnering the most
volving homologous recombination repair (BRCA2, ATM, attention in the clinical trial realm, platinum chemotherapy
and BRCA1) as well as mismatch repair (MLH1 and MSH2).2 has also been used in other BRCA1/2-associated cancers
From the University of Washington, Seattle, WA; University of Pennsylvania, Philadelphia, PA; University of Rochester, Rochester, NY; Tulane Cancer Center, New Orleans, LA.
Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook.
Corresponding author: Heather Cheng, MD, PhD, University of Washington, 825 Eastlake Ave E, Seattle, WA 98109-1023; email: [email protected].
© 2018 American Society of Clinical Oncology
372 2018 ASCO EDUCATIONAL BOOK | asco.org/edbook

PRACTICAL INTEGRATION OF GENETIC TESTING FOR ADVANCED PROSTATE CANCER
and has the advantage of being readily available and famil- together can help identify effective “precision” treatments
iar to practicing medical oncologists. A retrospective case for the subset of patients who are likely to respond and can
series evaluated three patients who displayed exceptional minimize ineffective treatments for those who are unlikely to
response (complete or partial response ranging from 6 to benefit. As a result of these data, other studies are ongoing
30 months) to platinum chemotherapy after disease progres- to evaluate the role of PARP inhibitors and platinum-based
sion while receiving prior standard therapy and whose tumors chemotherapy alone or in combination with other therapy
were available for analysis.5 Clinical targeted next-generation for patients with DNA repair defects.
sequencing on tumor DNA demonstrated the presence of
biallelic BRCA2 inactivation: two patients had a single ger- GERMLINE DNA REPAIR DEFECTS ARE
mline mutation with an additional acquired somatic event, ASSOCIATED WITH POOR PROGNOSIS AND
and the third patient had two somatic mutations. ENRICHED IN METASTATIC PROSTATE CANCER
In a retrospective analysis of 141 men with prostate can- Recognition of genomic factors critical to driving a cancer
cer who received at least two doses of carboplatin and may aid in efforts toward characterizing natural history and
docetaxel for metastatic castration-resistant prostate can- thus may serve as prognostic markers. It has been estab-
cer, treatment demonstrated benefit for patients with ger- lished that men who carry relatively rare pathogenic germ-
mline DNA repair deficiency with BRCA2 mutation.6 Eight line variants in BRCA2 have an increased risk of developing
of 141 men (5.7%) had a pathologic germline mutation; six prostate cancer and, if cancer is present, have worse out-
of these patients (75%) experienced a prostate-specific an- comes compared with men with prostate cancer who do not
tigen decline greater than 50% within 12 weeks compared carry BRCA2 pathogenic variants (mutations).8-10
with 23 of 133 noncarriers (17%; absolute difference, 58%;
95% CI, 27% to 88%; p < .001). This retrospective analysis TABLE 1. DNA Repair Genes, Prevalence of
of a prospective study demonstrated increased response to Mutations/Deletions in Metastatic Prostate Cancer,
platinum-based chemotherapy among patients with meta- and Early Evidence of Therapeutic Response to
static castration-resistant prostate cancer who also carried PARP Inhibitors
a BRCA2 germline mutation or pathogenic variant. A pros-
tate-specific antigen response greater than 50% was associ- Prevalence of Germline Preliminary Association
ated with prolonged survival in the entire group of 141 men, Mutations in Metastatic With PARP Inhibitor
Gene Prostate Cancer (%)a Response (X)b
and median survival was 18.9 months for carriers compared
with 9.5 months for noncarriers. ATM 1.59 X
These two studies highlight a biologic subgroup of pros- ATR 0.29
tate cancer with particular sensitivity to platinum therapy. BAP1 0.0
In contrast, other studies of platinum have failed to show BARD1 0.0
clinical benefit among unselected patients with prostate
BRCA1 0.87 X
cancer,7 illustrating the concept that genomics and genetics
BRCA2 5.35 X
BRIP1 0.18
PRACTICAL APPLICATIONS CHEK2 1.87 X

FAM175A 0.18
• The presence of both germline and somatic mutations FANCA — X
is increasingly well defined for patients with advanced
GEN1 0.46
prostate cancer; however, who to test and how to test is
the subject of considerable debate. HDAC2 —
• It is now clear that the prevalence of men with MLH1 0.0
metastatic prostate cancer carrying pathologic variants MRE11A 0.14
in DNA repair mutations is approximately 10% and
includes genes such as BRCA2 (most predominant), ATM, MSH2 0.14
CHEK2, BRCA1, RAD51D, and PALB2. MSH6 0.14
• Surprisingly, family histories for these patients are not NBN 0.29 X
always revealing and thus cannot be relied on to guide
PALB2 0.43 X
genetic testing.
• Germline pathogenic alterations may have both familial PMS2 0.29
and therapeutic implications. RAD51C 0.14
• Somatic alterations in various genes are increasingly well RAD51D 0.43
described; PARP inhibitors or platinum-based therapies
XRCC2 0.0
may be considered for patients with BRCA mutations,
and treatment with PD-1 inhibitors is warranted for aData are from Pritchard et al.4
patients with mismatch repair alterations and has been b
Data are from Mateo et al.3
approved by the U.S. Food and Drug Administration. Italicized genes indicate that data with inadequate sequencing for metastatic population were
censored.
asco.org/edbook | 2018 ASCO EDUCATIONAL BOOK 373

CHENG ET AL
A retrospective study of tumor characteristics in BRCA1/2 had a second acquired allele aberrancy, providing support
germline pathogenic variant (PV; mutation) carriers versus that these genes were of consequence rather than passen-
noncarriers identified that 67 participants in the carrier gers. It should be noted that there were also new genes iden-
group had more aggressive tumors, a significantly higher tified for which risk/penetrance estimates for prostate cancer
T score (T1, 9% vs. 26%; T2, 48% vs. 39%; and T3, 34% vs. 28%; have not yet been established.
p < .02), and a higher Gleason score (≤ 6, 33% vs. 49%; 7, 31% In addition to the potential therapeutic benefits and prog-
vs. 34%; and ≥ 8, 34% vs. 15%; p < .0001) compared with nostic implications, identifying germline carriers of single,
1,235 noncarriers.8 N1 disease was present in 6% of carriers high- or moderate-penetrance cancer predisposition genes
and 2% of noncarriers (p = .009). Moreover, BRCA1/2 germ- is an opportunity for cascade genetic testing. Early germ-
line carriers treated with curative intent with conventional line genetic testing of family members will allow better un-
treatment strategies (49 with BRCA2 and 18 with BRCA1) derstanding of personal risk for developing cancers and will
developed metastatic disease earlier and were observed create opportunities for early cancer-specific screening and,
to have shorter survival compared with 1,235 noncarriers.9 in some cases, risk reduction therapies and reproductive
Independent of the more aggressive baseline tumor char- planning options. It should be noted that risk/penetrance for
acteristics, BRCA1/2 mutation carriers had worse outcomes prostate cancer is incompletely characterized for many of the
in terms of metastasis-free survival (90%, 72%, and 50% vs. newly implicated genes but there are established guidelines
97%, 94%, and 84% at 3, 5, and 10 years, respectively) and for screening/management of other cancers for some genes.
cause-specific survival (96%, 76%, and 61% vs. 99%, 97%, With the recent exciting advances, the extent of germline
and 85% at 3, 5, and 10 years, respectively) compared with and somatic testing is likely to increase. Ongoing efforts are
individuals who were noncarriers. underway to understand and optimize delivery models, tim-
The prevalence of germline DNA repair alterations associ- ing, and other considerations around germline and somatic
ated with high- and moderate-penetrance cancer predispo- genetic testing. In the meantime, we offer the following
sition is higher than previously recognized in the population practical considerations.
with metastatic prostate cancer.4 An analysis of germline mu-
tations among men with metastatic prostate cancer was per- PRECISION MEDICINE FOR PROSTATE
formed to determine the frequency of germline DNA repair CANCER IN THE REAL WORLD: SOMATIC AND
mutations in this population. This analysis tested 20 genes GERMLINE TESTING CONSIDERATIONS
associated with autosomal dominant cancer predisposition Obtaining specimens from prostate cancer samples can be
among 692 men with metastatic prostate cancer and identi- challenging for numerous reasons. First, metastatic biopsies
fied 84 pathogenic germline mutations (11.8%) in 16 different are more invasive procedures and the most common site of
genes (Table 1). The study used whole-exome sequencing or metastatic disease is bone, which poses some difficulty with
targeted next-generation sequencing assays specific for the specimen acquisition, and can potentially involve specimen
DNA repair genes. The 11.8% frequency of germline muta- processing steps that may interfere with sequencing assays
tions in genes mediating DNA repair processes in men with (e.g., decalcification). Second, there can be heterogeneity
metastatic prostate cancer showed a significantly higher prev- within the tumor tissue; thus, limitations of sampling may
alence than that for patients with localized prostate cancer lead to results only partially reflective of tumor biology.
(4.6%; per the Cancer Genome Atlas Research Network) and Tumor sampling techniques and circulating tumor cell
for patients without cancer (2%–3%; per the Exome Aggrega- (CTC) sequencing methods are not yet agreed upon in the
tion Consortium).4 Tumor was available for sequencing for 61 experimental realm or as standard of care. However, prog-
men with germline DNA repair mutations, and of these, 59% ress is being made through advances in technology and
SIDEBAR 1. Suggested Criteria for Whom to Offer Genetic Counseling/Testing for Germline Alterations
Man With a Diagnosis of Prostate Cancer and Any One of the Following:
• Known mutation in a cancer susceptibility gene within the family
• Metastatic prostate cancer
• High-risk localized prostate cancer (Gleason score ≥ 8, WHO grade group ≥ 3, or PSA ≥ 20)
• Tumor (somatic) sequencing indicating the presence of mutations in hereditary cancer risk genes (e.g., BRCA2,
BRCA1, ATM, MSH2, MSH6, MLH1, PMS2)
• Family history suggestive of hereditary breast and ovarian cancer syndrome
• Family history suggestive of Lynch syndrome
• Family history suggestive of hereditary prostate cancer syndrome
Abbreviations: WHO, World Health Organization; PSA, prostate-specific antigen.

Adapted from the National Comprehensive Cancer Network13-15; Giri et al12; and Gillessen et al.16

increasing correlation between ctDNA and tumor tissue. An prostate guidelines (www.nccn.org) recommend that
ideal test would provide information about somatic muta- genetic counseling and/or testing be offered to patients
tions (and germline mutations if desired), would have low with metastatic prostate cancer and those who qualify for
cost, and would be easy to obtain. Ultimately, if a test were the testing based on family history guidelines (Sidebar 1).
simple and affordable, somatic and germline assessment of Patient priorities should be considered, as should cost of
a tumor at diagnosis could someday help physicians guide testing. Utility of results may be not only for the patient's
up-front management and potentially monitor for resis- own prostate cancer care decisions (in which case life
tance and guide subsequent treatment planning. expectancy and performance status should be factored),
but also to test an informative member of the family
AVAILABILITY OF GERMLINE TESTING where cascade testing is also a potential goal. Potentially
Currently, germline genetic testing may be slightly more at-risk relatives would have the option for more informa-
straightforward, because germline variants and mutations tive, less expensive single-site testing if a pathogenic re-
can be easily and reliably evaluated from peripheral blood sult is found.
and saliva as well as from buccal mucosa or skin biopsy and
there are accepted standards for reporting variants (benign, ASSAYS FOR SOMATIC GENETIC ALTERATIONS
likely benign, variant of uncertain significance, pathogenic, Evaluation of somatic alterations in prostate cancer con-
and likely pathogenic).11 tinues to evolve. A variety of DNA, RNA, or protein-based

Several options for testing are present and the exact test is assays can be either tissue or blood based. Tissue-based as-
dependent on arrangements that can be made individually. says are well discussed in the context of other tumors and
Companies such as Ambry Genetics, Color Genomics, Invitae further discussions here do not need to be elaborated on
Genetics, and Myriad all have different mechanisms for test- other than to say that mutations and copy number varia-
ing and billing. A key determination is the “out-of-pocket” tions are only part of the story. Inversions (e.g., the Boland
cost to the patient and what will be paid by insurance, which inversion), methylation patterns (e.g., those relevant for
can be difficult to discern. Many patients are interested in MLH1), and splice variants are additional noteworthy areas.
genetic testing because of the familial implications but out-of- Tissue-based assays have recently been covered by selected
pocket costs are a consideration for many. insurance companies, including Medicare.
It is also notable that the patients with advanced pros- RNA transcriptomic analyses and the generation of vari-
tate cancer may or may not have family histories that ous signatures are both interesting and potentially clinically
qualify them for genetic testing, and at times the criteria relevant. PAM50 categorization into luminal and basal sub-
seem to vary from one insurance company to another. types is potentially noteworthy for those with localized dis-
Testing guidelines for men are less established than those ease. Tissue- and blood-based assays have been developed
for women. on both androgen receptor (AR) amplification and AR splice
Another issue that has been extremely problematic is variants (especially ARV-7). ARV-7 can be assayed in CTCs or
the frequent lack of availability of genetic counselors. This in whole blood. Although the importance of ARV-7 has been
has become a controversial area particularly for clinics that debated, the stability of RNA is potentially problematic.
treat patients who may be traveling from distant locations. RNA-based ARV-7 assays in CTCs are cleared by the Clinical
It may be helpful to create a standardized set of materials Laboratory Improvement Amendment and can be ordered
and information for patients and a mechanism for test- from selected vendors.
ing to occur directly in an oncology clinic without formal A variety of protein-based assays have been studied, with
genetic counseling if access to counseling is a limited re- ARV-7 being a recent protein of focus. The small aberrant
source. This is a potentially controversial approach, yet the extension of the ARV-7 protein (at the C terminus) can be
limitations of genetic counseling resources are substantial targeted with specific antibodies. Antibody-based assays for
in many institutions and this could facilitate triaging of ARV-7 can be done on CTCs. Such testing can soon be com-
more in-depth genetics services for individuals who have mercially available.
variants of significance. Some companies offer genetic Another blood-based assay is based on cell-free DNA or
counseling services over the telephone. Clearly, this is an ctDNA. There is clear progress on both mutations and am-
area that requires further development and collaborative plifications of AR. Non-AR assays have yet to be verified
optimization. but understanding the mutations in DNA repair genes, mis-
match repair genes, and microsatellite instability/mutation
WHO TO TEST FOR GERMLINE ALTERATIONS? burden is a current area of assessment. Such assays can now
There are multiple questions that arise regarding who be ordered from various commercial vendors but their cov-
should be tested, how, and when. A consensus document erage by insurance is highly variable.
was recently published but must be considered explorato-
ry in terms of making recommendations simply because CLINICAL ACTIONABILITY OF SOMATIC
of the lack of data.12 In addition, relevant National Compre- MUTATIONS
hensive Cancer Network guidelines have been updated. At present, the most clinically “actionable” alteration may
The current National Comprehensive Cancer Network be the presence of mismatch repair mutations or high

CHENG ET AL
microsatellite instability in which PD-1 inhibitors are U.S. Food syndrome may be suspected based on family history but
and Drug Administration–approved in a tumor-tissue agnostic testing tissue in every patient seems unjustified. Given the
manner. Although prostate cancer data remain relatively lim- activity of PD-1 antibodies, testing on accessible tissue and
ited, remissions are documented in patients with castration- more diligent attempts to obtain tissue may be warranted
resistant prostate cancer after treatment with PD-1 antibodies. for patients with either germline alterations or family histo-
Given that today these assays are tissue based and that ries that are suspect for Lynch-type patterns. The frequency
prostate metastatic tissue is often hard to obtain, there is a of microsatellite instability and mismatch repair is unclear
high need for assays that are blood based. but estimates ranged as high as 12% of patients in an au-
Blood-based ctDNA is of considerable interest but it is topsy series.17
early to conclude that these assays can be used reliably.
Considerable differences between assays have been noted PRACTICAL GUIDANCE ON GENETIC
by respected investigators (K. J. Pienta et al, unpublished COUNSELING WITHOUT GENETIC
data). The bottom line is that the ctDNA assays are still to be COUNSELOR SUPPORT
proven in the context of multi-institutional trials and more As next-generation tumor sequencing becomes more em-
research must be done. bedded into routine clinical oncology practice, it remains
DNA repair defects, especially those in BRCA1/2, have critically important that practitioners recognize and better
been linked to PARP inhibitor and platinum sensitivity when understand its role in potentially unearthing germline or
using tissue-based assays.3,5 inherited cancer risk. Although the primary objective of
The AR amplifications and the selected AR mutations next-generation tumor sequencing is to determine thera-
have been linked to resistance to abiraterone and/or en- peutic targets, results may also suggest or identify under-
zalutamide in various assays. Certain mutations such L702H, lying germline disease-associated variants regardless of
H875Y, and T878A are linked to abiraterone/enzalutamide whether they were suspected prior to testing.18 It could
resistance and poor clinical outcomes in some studies. be argued that “germline genetics” is best approached via
W742C/L is associated with agonistic actions of bicalut- thorough risk assessment under the provision of a special-
amide and F877L converts apalutamide and enzalutamide ist such as a genetic counselor. Although this is considered
to agonists. The AR amplifications and mutations are best the “gold standard,” the avenues by which individuals are
studied in blood-based assays using cell-free DNA. assessed for inherited cancer risk, including “incident to”
ARV-7 has been linked to resistance to abiraterone and next-generation tumor sequencing, have altered the mech-
enzalutamide in both RNA- and antibody-based assays in anisms of ascertainment. When specialist support is not
CTCs. AR amplification in CTCs is similarly linked. Taxanes available, the traditional approach of pre- and postgenetic
may be a better choice for those with ARV-7 detected counseling may be untenable.
in CTCs. This section will review practical guidance and tools to
Various morphologic patterns in CTCs including heteroge- help demystify and make “germline versus somatic” more
neity are also linked to abiraterone/enzalutamide resistance navigable.
when analyzing CTCs in the bloodstream. These assays are The main components of clarifying and demystifying
not clinically available at this time. germline versus somatic variants are as follows: (1) know
the test and what is or is not reported, (2) identify the
SOMATIC GENETICS: WHO TO TEST IN laboratory contact(s) and/or laboratory genetic counselor,
CLINICAL PRACTICE? (3) have general familiarity with genes or findings that
The importance of testing escalates when the clinician be- could be relevant to inherited susceptibility, and (4) under-
comes better informed as to appropriate treatment choic- stand how to triage. Through this foundation, one can have
es. Currently, the empirical approaches to management better-informed conversations with patients and their
represent a reasonable alternative to biomarkers for many families.
patients. Current biomarkers are not perfect and the re-
sponse/resistance patterns need further prospective verifi- Germline Versus Somatic
cation in broad groups and with readily repeatable assays. In distinguishing between germline and somatic variants,
The use of PARP inhibition represents an opportunity for germline DNA is inherited material from both the egg and
those with DNA repair alterations, as does the use of plati- sperm that may be passed down to offspring. A germline PV
num. Who to test depends on access to clinical trials and the is constitutional and within every cell of the body. With few
availability of germline results. The majority of individuals exceptions, germline PVs that predispose to known inherit-
with germline mutations in DNA repair genes also have ed cancer syndromes (e.g., BRCA1/2 and hereditary breast
somatic mutations. The use of platinum in an empirical man- and ovarian cancer syndrome) are inherited in an autosomal
ner may be justified. To our knowledge, no data on blood- dominant manner. Put another way, each child (offspring)
based ctDNA assays for use of PARP inhibitors or platinum has a 50/50 chance of inheriting the germline PV and asso-
currently exist. ciated cancer risks. Approximately 5% to 10% of cancers are
The testing of tissue for high microsatellite instability or inherited and are attributable to a single highly penetrant
mismatch repair is a bit of a conundrum. Who to test? Lynch PV in a DNA repair gene.

TABLE 3. Suggested Reference Checklist of Considerations When Ordering Genetic/Genomic Testing
Checklist Provider Action Plan (Summarized in 5–10 Minutes)21,22

Pretest checklist
Know your test (1) Brief assessment of pretest clinical suspicion of a high-risk cancer syndrome (Sidebar 1) and temper
2(a) to 2(e) according to suspicion23
Q: Tumor only?a (2) Brief dialogue with patient/family:
Q: Filter out/computationally mask (a) Outline the primary goal of tNGS
presumed germline?
A: Use traditional risk assessment to (b) Real but small chance test result may suggest an inherited risk with implication for relatives
inform suspicion of inherited cancer
risk
Q: Who is my laboratory contact? (c) If suggestive a priori, “indispensable” information directly relevant to the family is as follows:
Informed consent: no precedent, verbal Inherited mutations in cancer risk genes are rare. Most mutations found within tumors are
acknowledgment acquired changes, are not present at birth, and cannot be passed down to children
Implications of inherited risk vary depending on the gene and result (associated cancer risks and
our depth of knowledge are gene specific)
If determined to be inherited, the family could take steps to clarify and reduce risk through medical
care/intervention
Potential for various uncertainties

(d) Testing is voluntary. In context of tNGS, primary benefit of therapeutics typically outweighs
potential risks associated with incidental germline
(e) E lements of disclosure: possibility of new sample (blood/saliva) for germline confirmation,
possible cascade testing in family members if mutation is confirmed
Post-test check list to help determine whether germline analysis appears warranteda
Q: Has the laboratory provided variant Pathogenic: potentially actionable
interpretation (e.g., pathogenic, VUS,
VUS: not actionable
benign)?
Variant interpretation discrepancies between somatic and germline: refer to ClinVar (www.ncbi.nlm.nih.
gov/clinvar/)
Benign: not actionable
Unknown: contact the laboratory for confirmation
Q: Have mutations in the gene of interest BRCA1 or BRCA2: warrants germline confirmation per NCCN v1.2018
been reported as associated with an
Consistent clinical phenotype: If patient’s personal/family history is consistent with disease, germline
inherited cancer predisposition
confirmation warranted
syndrome and/or an increased cancer
risk (Table 4)? Example 1: APC mutation found in any tumor type in patient reporting a history of greater than 20
colon polyps
Action plan: APC germline confirmation appears warranted
Example 2: VHL mutation found in any tumor type in patient that does not exhibit features of Von
Hippel Lindau (Table 1 and www.omim.org)
Action plan: VHL germline confirmation does not appear warranted
Moderate penetrance: RAD51C, RAD51D, BRIP1, PALB2, etc.
Consideration of germline confirmation
Q: Is this gene variant a common “founder Strongly consider germline confirmation
mutation” (Table 5)?
Q: Is this gene commonly mutated in Example: The TP53 gene is the most commonly mutated gene in human cancers. Inherited TP53
disease? mutations cause Li-Fraumeni syndrome, a high-risk cancer syndrome with well-established clinical
diagnostic criteria (Sidebar 1 and www.omim.org). Less than 1% of TP53 mutations identified in tumors
are present in the germline. Rarely do TP53 mutations warrant germline confirmation, unless cancer
type alone supports a possible diagnosis
If germline variants are reported in parallel with tumor results, ensure that the laboratory is validated in germline sequencing and interpretation (per American College of Medical Genetics and Genomics–
a
Association for Molecular Pathology guidelines; skip to the last section titled “Key Discussion Points for Patients and Their Families.”).
Abbreviations: tNGS, next-generation tumor sequencing; VUS, variant of uncertain significance; NCCN, National Comprehensive Cancer Network.
Somatic variants are alterations in genes that devel- this mutation through a blood or saliva specimen, the
op over one’s lifetime, and they are not present in the PV would be absent. The majority of genetic alterations
egg or sperm and cannot be passed down to off- detected in tumors are somatic (“It’s not you, it’s your
spring. In this context, should one pursue analysis of tumor.”).

CHENG ET AL
TABLE 4. Genes Associated With Increased Risk of Cancer
Gene Syndrome Association Predominant Tumor Type(s)

High penetrance Consistent phenotype: available clinical diagnostic criteria,
early onset, multigenerational, specific constellation of
clinical features in individual +/− family
APC FAP Colorectal, polyposis (adenomatous)
BAP1 Renal, uveal melanoma
BRCA1 a
HBOC Breast, ovarian, prostate, pancreas, melanoma
BRCA2a HBOC Breast, ovarian, prostate, pancreas, melanoma
BMPR1A Colorectal, large and small bowel polyposis (juvenile type)
CDH1 Hereditary diffuse gastric cancer Diffuse gastric, breast
CDK4 Melanoma cancer syndrome (FAMMM) Malignant melanoma
CDKN2A Melanoma cancer syndrome (FAMMM) Malignant melanoma
FH Hereditary leiomyomatosis and renal cell cancer Renal, cutaneous, and uterine leiomyomas
FLCN Birt Hogg-Dubé Renal, fibrofolliculomas, pneumothorax, lung cysts
MEN1 Multiple endocrine neoplasia type 1 Endocrineb

MLH1 HNPCC, Lynch syndrome Tier 1: colorectal, endometrial, ovarian
Tier 2: gastric, small bowel, brain, pancreas, prostate
MSH2 HNPCC, Lynch syndrome Tier 1: colorectal, endometrial, ovarian
MSH6 HNPCC, Lynch syndrome Tier 1: colorectal, endometrial, ovarian
RB1 Retinoblastoma (pediatric), melanoma, sarcoma
RET Medullary thyroid
PALB2 Breast, pancreas
PMS2 HNPCC, Lynch syndrome Tier 1: colorectal, endometrial, ovarian
PTEN c
Cowden syndrome b
SDHA Hereditary pheochromocytoma-paraganglioma Pheochromocytoma, paraganglioma

SDHB Hereditary pheochromocytoma-paraganglioma Pheochromocytoma, paraganglioma
SDHC Hereditary pheochromocytoma-paraganglioma Pheochromocytoma, paraganglioma
SDHD Hereditary pheochromocytoma-paraganglioma Pheochromocytoma, paraganglioma
STK11 Peutz-Jegher syndrome Tier 1: polyposis (Peutz-Jegher type), distinct lip frecklingb
Tier 2: breast, colorectal, pancreas, lung, small bowel
TP53d Li-Fraumeni syndrome Pediatric adrenocortical carcinoma, choroid plexus tumor,
breast cancer (younger than 35 years), soft tissue
sarcoma
TSC1 Tuberous sclerosis type 1 b
TSC2 Tuberous sclerosis type 2 b
VHL Von Hippel Lindau b
WT1 Wilms tumor, including WAGR b
NF2 Neurofibromatosis type 2 b
Continued

TABLE 4. Gennes Associated With Increased Risk of Cancer (Cont'd)
Gene Syndrome Association Predominant Tumor Type(s)

Moderate penetrance Syndrome Association Predominant Tumor Type(s)
(monozygote)
ATM Breast, pancreas
BRIP1 Ovarian
CHEK2 Breast, prostate, colon
MITF Melanoma
NBN Breast
No established syndrome, inconsistent phenotype
RAD51C Ovarian
RAD51D Ovarian
Low penetrance
APC I1307K Colorectal
MUTYH monozygote Colorectal

CHEK2 I157T Breast
a
Per National Comprehensive Cancer Network13 guidelines (v1.2018), all patients with somatic BRCA1 and BRCA2 pathogenic findings regardless of tumor type and/or family history should be offered germline
confirmation.
b
Clinical diagnosis typically made based on noncancerous features. Please refer to www.omim.org for explanation and expansion.
c
Per the Cancer Genome Atlas Research Network, approximately 5% of primary prostate cancers have a PTEN mutation, with 0.1% germline frequency. Please refer to OMIM.org or ACMG-NSGC practice
guidelines for appropriate consideration of germline confirmation and/or referral.
d
Most frequently mutated gene in human cancers. Please refer to Online Mendelian Inheritance in Man (OMIM) or ACMG-NSGC practice guidelines20 for appropriate consideration of germline confirmation
and/or referral.
Abbreviations: FAP, familial adenomatous polyposis; HBOC, heritable breast and ovarian cancer; FAMMM, familial atypical multiple mole–melanoma; HNPCC, hereditary nonpolyposis colorectal cancer;
WAGR, Wilms tumor/aniridia/genitourinary malformation/mental retardation; ACMG, American College of Medical Genetics and Genomics; NSGC, National Society of Genetic Counselors.
Deciphering What Is Somatic and What Could Be always apparent whether germline variants are reported,
Germline: Know Your Test and Know Your Laboratory and if they are, by what standard of interpretation and rigor.
More frequently seen in research-based next-generation tu- This section focuses on laboratories that sequence only
mor sequencing, although there will likely be an uptick in the the tumor DNA, because this is the more likely encountered
pursuant of commercial entities, laboratories may sequence scenario in the clinical realm. Tumor-only sequencing without
a normal, matched control DNA sample in parallel with the matched normal DNA introduces the possibility of a germline
tumor DNA. When using “paired tumor-normal,” germline mutation but does not confirm or rule it out. It is at the prac-
variants may be evident.19 In such cases, however, it is not titioner’s discretion when to suspect a reported mutation as
TABLE 5. Ashkenazi Jewish and European Founder Mutations
Mutation Comment
BRCA1 c.68_69delAG (p.Glu23Valfs*17)
BRCA1 c.5266dupC (p.Gln1756Profs*74)
BRCA2 c.5946delT (p.Ser1982Argfs*22)
BRCA2 c.771_775delTCAAA
MSH2 c.1906G>C (p.Ala636Pro)
MSH6 c.3959_3962delCAAG (p. Ala1320Glufs*6)
CHEK2 c.1100delC (p.Thr367Metfs*15)
CHEK2 c.1283C>T (p.Ser428Phe)
APC c.3920T>A (p.Ile1307Lys) Low penetrance, unclear clinical utility; does not cause polyposis
MUTYH c.1187G>A (p.Gly396Asp) Biallelic MUTYH carriers have polyposis condition known as MUTYH polyposis
MUTYH c.536A>G (p.Tyr179Cys) Biallelic MUTYH carriers have polyposis condition known as MUTYH polyposis
NBN c.657del5
FANCC c.456 + 4A>T
PALB2 c.1592delT
CHEK2 (p.I157T) and (p.S428F) Low penetrance, unclear clinical utility

CHENG ET AL
Basic Genetic Education and Implication for the Family

Sidebar 2. Logistics and Resources Most mutations in tumors are not inherited; rather, they de-
velop only within cancerous cells. Germline mutations are rare
• National Society of Genetic Counselors frequently but can be passed down to children and can be associated
asked questions (aboutgeneticcounselors.com/FAQs- with an increased risk for cancer. In most cases, a parent who
resources) has a germline mutation has a 50/50 chance of passing down
• How to find a genetic counselor, including tele- this mutation to each child. We all have an expected 7 to
medicine (aboutgeneticcounselors.com) 10 genes (of 20,000) that do not function due to a mutation.
• National Cancer Institute fact sheet on genetic test-
ing for hereditary cancer syndromes (www.cancer. Implications of Results Vary Depending on the Gene,
gov/about-cancer/causes-prevention/genetics/ Result, and Age and Sex of the Patient
genetic-testing-fact-sheet#q7) There are high-risk genes, moderate-risk genes, and genes
for which there may be a cancer risk association but it re-
mains unclear (Tables 4 and 5). A positive test result for
possibly germline, which would warrant confirmation on a new a mutation does not guarantee that patients will develop
sample type (blood, saliva) by a Clinical Laboratory Improvement cancer in their lifetime; however, this test may show a need
Amendments–College of American Pathologists laboratory for additional medical care to protect health. Some gene
proficient in germline analysis, reporting, and interpretation. mutations cause higher risks for cancer (BRCA1/2), whereas
Medical oncologists must also appreciate that the goals others may confer modest increased risk (CHEK2 for breast
of somatic testing and variant interpretation are different cancer and RAD51C/D for ovarian cancer). It is important to
than for germline variants. It is possible that certain tumor recognize that not all gene mutations confer the same risk,
testing laboratories filter or computationally mask what is and many are not yet characterized for either cancer risk or
presumed germline. for disease biology or treatment implications. Many newly
implicated rare variants have limited to no data around the
Tumor Testing Should Not Be Used as a Substitute relative risk of developing cancer or a therapeutic response
When Genetic Cancer Risk Is Suspected profile. However, there are medical care guidelines and in-
Germline versus somatic laboratories have different report- terventions for some key genes, some of which are proven
ing mechanisms (e.g., are missenses and/or variants of un- to be life-saving and others more controversial. There are
certain significance reported?) and different institutional evidence-based recommendations in some cases, consen-
knowledge. Furthermore, absence of a gene finding on a so- sus guidelines in others, and no clear consensus or guide-
matic testing does not equate to a negative result in the ger- lines others.24 With few exceptions (TP53), most testing of
mline. As with all patients, if clinical suspicion is high for an this nature is recommended for individuals older than 18
inherited cancer predisposition syndrome, genetic evalua- years. Reproductive issues may influence decisions, espe-
tion and/or referral to a genetic specialist is recommended.20 cially in younger women.
There is recognition that it is not feasible for busy oncol-
ogy clinics to implement detailed family risk screening Psychosocial Assessment
for every patient; furthermore, relying solely on family his- It is important to establish a mutual agenda and evaluate
tory has its shortcomings (e.g., inaccurate reporting, the family dynamics. Although inherited health risk information
reduced penetrance of moderate risk mutations making may not directly benefit the patient’s care, the patient is the
family history difficult to use, and stringent diagnostic and most informative individual to evaluate to see whether this
testing criteria).18 finding could be inherited or passed down. Identification
If possible, it benefits the provider to be anticipatory; the of an inherited gene mutation does not mean an individual
likelihood of a somatic finding present in the germline is has or will develop cancer; however, the risks may be higher.
low but real. A brief discussion up front at the time of next- Conversely, testing negative for one or more inherited gene
generation sequencing ordering can help explain the possibil- mutations does not eliminate all likelihood of developing
ity of identifying inherited risk, reassurance that information cancer. In addition, relatives should be followed based on
will be communicated to the patient/family if determined their own personal and family history.
to be relevant, and highlighting that incidentals are at times It is also recommended that physicians assess the pa-
“unavoidable.” Table 3 proposes a pre- and postchecklist tient’s willingness to share information with relatives and
that could be referenced quickly to mitigate (although not to potentially establish a point of care (next of kin), as in-
eliminate) some of the current issues. herited cancer genetics is evolving rapidly and there could
be updates in the future, especially if results are uncertain.
KEY DISCUSSION POINTS FOR PATIENTS AND Individuals are adaptive and most can effectively assim-
THEIR FAMILIES ilate risk information. Sometimes patients do not want to
If a germline mutation is suspected and/or confirmed, here “burden” their families with genetic risk information and
are some main discussion points to consider, recognizing they feel guilty or ashamed. It is important to share with
that discussions are tailored and nuanced for each patient. patients that most individuals, including offspring, would

feel that the benefits of having predictive health risk infor- risk or disease (intellectualism), insert humor, or take the
mation outweigh the risks. Most individuals would and “fighting spirit” approach (e.g., those motivated to avoid a
do prefer to take charge of their medical care if there similar fate). Individuals may pull from multiple strategies.
are interventions for early detection and prevention. Fi- By evaluating the coping strategy, providers can normalize
nally, coping strategies differ between individuals and un- emotions and correct misinformation and thus make navi-
derstanding a patient’s coping mechanism may be used to gation of health risk information and care easier.
potentially clarify misinformation. Mutation carriers may The goal of this article is to provide a basic framework
experience fatalistic thoughts (e.g., that cancer is not a mat- for considering somatic and germline testing, as well as the
ter of “if” but “when” and that there is nothing to mitigate downstream familial impact of an inherited finding, it cannot
such a high risk for disease). Others may avoid addressing fully address all of the current issues nor should this be used
their risk out of fear. In “high-risk” families, individuals’ as a substitute when genetic specialist support is available. It
perceptions of risk and mortality are oftentimes shaped by is hoped that this may serve as a supplement to the knowl-
their strong family history of disease (their family legacy). edge gleaned from the experts positioned at both the testing
Others may choose to learn as much as possible about their laboratories and within one’s own community (Sidebar 2).
References
1. Cancer Genome Atlas Research Network. The molecular taxonomy of 14. National Comprehensive Cancer Network. Genetic/Familial High-Risk
primary prostate cancer. Cell. 2015;163:1011-1025. Assessment: Colorectal (v1.2018). https://www.nccn.org/professionals/

2. Robinson D, Van Allen EM, Wu YM, et al. Integrative clinical genomics physician_gls/pdf/genetics_colon.pdf. Accessed March 13, 2018.
of advanced prostate cancer. Cell. 2015;161:1215-1228. 15. National Comprehensive Cancer Network. Prostate Cancer (v1.2018).
https://www.nccn.org/professionals/physician_gls/pdf/prostate.pdf.
3. Mateo J, Carreira S, Sandhu S, et al. DNA-repair defects and olaparib in
Accessed March 13, 2018.
metastatic prostate cancer. N Engl J Med. 2015;373:1697-1708.
16. Gillessen S, Attard G, Beer TM, et al. Management of patients with
4. Pritchard CC, Mateo J, Walsh MF, et al. Inherited DNA-repair gene
advanced prostate cancer: the report of the Advanced Prostate Cancer
mutations in men with metastatic prostate cancer. N Engl J Med.
Consensus Conference APCCC 2017. Eur Urol. 2018;73:178-211.
2016;375:443-453.
17. Pritchard CC, Morrissey C, Kumar A, et al. Complex MSH2 and MSH6
5. Cheng HH, Pritchard CC, Boyd T, et al. Biallelic inactivation of BRCA2 in
mutations in hypermutated microsatellite unstable advanced prostate
platinum-sensitive metastatic castration-resistant prostate cancer. Eur
cancer. Nat Commun. 2014;5:4988.
Urol. 2016;69:992-995.
18. Catenacci DV, Amico AL, Nielsen SM, et al. Tumor genome analysis
6. Pomerantz MM, Spisák S, Jia L, et al. The association between germline
includes germline genome: are we ready for surprises? Int J Cancer.
BRCA2 variants and sensitivity to platinum-based chemotherapy among
2015;136:1559-1567.
men with metastatic prostate cancer. Cancer. 2017;123:3532-3539.
19. Li MM, Datto M, Duncavage EJ, et al. Standards and guidelines for
7. Hager S, Ackermann CJ, Joerger M, et al. Anti-tumour activity of
the interpretation and reporting of sequence variants in cancer: a
platinum compounds in advanced prostate cancer-a systematic
joint consensus recommendation of the Association for Molecular
literature review. Ann Oncol. 2016;27:975-984.
Pathology, American Society of Clinical Oncology, and College of
8. Castro E, Goh C, Olmos D, et al. Germline BRCA mutations are associated American Pathologists. J Mol Diagn. 2017;19:4-23.
with higher risk of nodal involvement, distant metastasis, and poor
20. Robson ME, Bradbury AR, Arun B, et al. American Society of Clinical
survival outcomes in prostate cancer. J Clin Oncol. 2013;31:1748-1757.
Oncology policy statement update: genetic and genomic testing for
9. Castro E, Goh C, Leongamornlert D, et al. Effect of BRCA mutations on cancer susceptibility. J Clin Oncol. 2015;33:3660-3667.
metastatic relapse and cause-specific survival after radical treatment
21. Riley BD, Culver JO, Skrzynia C, et al. Essential elements of
for localised prostate cancer. Eur Urol. 2015;68:186-193.
genetic cancer risk assessment, counseling, and testing: updated
10. Mitra A, Fisher C, Foster CS, et al; IMPACT and EMBRACE Collaborators. recommendations of the National Society of Genetic Counselors. J
Prostate cancer in male BRCA1 and BRCA2 mutation carriers has a Genet Couns. 2012;21:151-161.
more aggressive phenotype. Br J Cancer. 2008;98:502-507.
22. Bradbury AR, Patrick-Miller L, Long J, et al. Development of a tiered and
11. Richards S, Aziz N, Bale S, et al; ACMG Laboratory Quality Assurance binned genetic counseling model for informed consent in the era of
Committee. Standards and guidelines for the interpretation of multiplex testing for cancer susceptibility. Genet Med. 2015;17:485-492.
sequence variants: a joint consensus recommendation of the American
23. Hampel H, Bennett RL, Buchanan A, et al; Guideline Development
College of Medical Genetics and Genomics and the Association for
Group, American College of Medical Genetics and Genomics
Molecular Pathology. Genet Med. 2015;17:405-424.
Professional Practice and Guidelines Committee and National Society of
12. Giri VN, Knudsen KE, Kelly WK, et al. Role of genetic testing for Genetic Counselors Practice Guidelines Committee. A practice guideline
inherited prostate cancer risk: Philadelphia Prostate Cancer Consensus from the American College of Medical Genetics and Genomics and the
Conference 2017. J Clin Oncol. 2018;36:414-424. National Society of Genetic Counselors: referral indications for cancer
13. National Comprehensive Cancer Network. Genetic/Familial High-Risk predisposition assessment. Genet Med. 2015;17:70-87.
Assessment: Breast and Ovarian (v1.2018). https://www.nccn.org/ 24. Bradbury AR, Patrick-Miller LJ, Egleston BL, et al. Patient feedback and
professionals/physician_gls/pdf/genetics_screening.pdf. Accessed early outcome data with a novel tiered-binned model for multiplex
March 13, 2018. breast cancer susceptibility testing. Genet Med. 2016;18:25-33.

Cheng H Et Al - Am Soc Clin Oncol Educ Book - 2018

Uploaded by

Copyright:

Available Formats

Cheng H Et Al - Am Soc Clin Oncol Educ Book - 2018

Uploaded by

Document Information

Original Title

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

Cheng H Et Al - Am Soc Clin Oncol Educ Book - 2018

Uploaded by

Copyright:

Available Formats

CHENG ET AL

Practical Methods for Integrating Genetic Testing Into

O ver the past several years, it has become abundantly

© 2018 American Society of Clinical Oncology

372 2018 ASCO EDUCATIONAL BOOK | asco.org/edbook

PRACTICAL APPLICATIONS CHEK2 1.87 X

asco.org/edbook | 2018 ASCO EDUCATIONAL BOOK 373

Abbreviations: WHO, World Health Organization; PSA, prostate-specific antigen.

374 2018 ASCO EDUCATIONAL BOOK | asco.org/edbook

and likely pathogenic).11 tinues to evolve. A variety of DNA, RNA, or protein-based

asco.org/edbook | 2018 ASCO EDUCATIONAL BOOK 375

376 2018 ASCO EDUCATIONAL BOOK | asco.org/edbook

TABLE 3. Suggested Reference Checklist of Considerations When Ordering Genetic/Genomic Testing

Checklist Provider Action Plan (Summarized in 5–10 Minutes)21,22

Potential for various uncertainties

asco.org/edbook | 2018 ASCO EDUCATIONAL BOOK 377

TABLE 4. Genes Associated With Increased Risk of Cancer

Gene Syndrome Association Predominant Tumor Type(s)

MEN1 Multiple endocrine neoplasia type 1 Endocrineb

SDHA Hereditary pheochromocytoma-paraganglioma Pheochromocytoma, paraganglioma

TSC2 Tuberous sclerosis type 2 b

VHL Von Hippel Lindau b

WT1 Wilms tumor, including WAGR b

NF2 Neurofibromatosis type 2 b

378 2018 ASCO EDUCATIONAL BOOK | asco.org/edbook

TABLE 4. Gennes Associated With Increased Risk of Cancer (Cont'd)

Gene Syndrome Association Predominant Tumor Type(s)

MUTYH monozygote Colorectal

TABLE 5. Ashkenazi Jewish and European Founder Mutations

asco.org/edbook | 2018 ASCO EDUCATIONAL BOOK 379

Basic Genetic Education and Implication for the Family

380 2018 ASCO EDUCATIONAL BOOK | asco.org/edbook

primary prostate cancer. Cell. 2015;163:1011-1025. Assessment: Colorectal (v1.2018). https://www.nccn.org/professionals/

asco.org/edbook | 2018 ASCO EDUCATIONAL BOOK 381

You might also like