Practical Radiotherapy Planning, Fourth Edition

Practical
Radiotherapy
Planning
This page intentionally left blank
Practical
Radiotherapy
Planning
Fourth Edition
Ann Barrett MD, FRCP, FRCR ■

Professor Emeritus of Oncology,
University of East Anglia, Norwich
Jane Dobbs MA, FRCP, FRCR ■
Consultant Emeritus in Clinical Oncology,
Guy’s and St Thomas’ NHS Foundation Trust
Stephen Morris MRCP, FRCR ■
Consultant Clinical Oncologist,
Guy’s and St Thomas’ NHS Foundation Trust
Tom Roques MRCP, FRCR ■
Consultant Clinical Oncologist,
Norfolk and Norwich University Hospital NHS Foundation Trust
Illustrations editor:
Jonathan Harrowven BSc (Hons),
Radiographer, Norfolk and Norwich University Hospital
NHS Foundation Trust
Hodder Arnold, an imprint of Hodder Education, an Hachette UK Company
338 Euston Road, London NW1 3BH
http://www.hoddereducation.com
© 2009 Ann Barrett, Jane Dobbs, Stephen Morris and Tom Roques
All rights reserved. Apart from any use permitted under UK copyright law, this
publication may only be reproduced, stored or transmitted, in any form, or by any means
with prior permission in writing of the publishers or in the case of reprographic
production in accordance with the terms of licences issued by the Copyright Licensing
Agency. In the United Kingdom such licences are issued by the Copyright Licensing
Agency: Saffron House, 6-10 Kirby Street, London EC1N 8TS.
Hachette UK’s policy is to use papers that are natural, renewable and recyclable products
and made from wood grown in sustainable forests. The logging and manufacturing
processes are expected to conform to the environmental regulations of the country
of origin.
Whilst the advice and information in this book are believed to be true and accurate at
the date of going to press, neither the author[s] nor the publisher can accept any legal
responsibility or liability for any errors or omissions that may be made. In particular (but
without limiting the generality of the preceding disclaimer) every effort has been made to
check drug dosages; however it is still possible that errors have been missed. Furthermore,
dosage schedules are constantly being revised and new side-effects recognized. For these
reasons the reader is strongly urged to consult the drug companies' printed instructions
before administering any of the drugs recommended in this book.
British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress
ISBN 978 034 0927731
1 2 3 4 5 6 7 8 9 10
Commissioning Editor: Gavin Jamieson

Project Editors: Eva Senior and Joanna Silman
Production Controller: Joanna Walker
Cover Designer: Helen Townson
Indexer: Jan Ross
Typeset in 10/12 Galliard by Macmillan Publishing Solutions

Printed and bound in Italy
What do you think about this book? Or any other Hodder Arnold title?
Please visit our website: www.hoddereducation.com
Contents
Preface vii
1 What you need to know before planning radiotherapy
treatment 1
2 Principles of radiotherapy planning 9
3 Radiobiology and treatment planning 32
4 Organs at risk and tolerance of normal

tissues 44
5 Principles of brachytherapy 54
6 Emergency and palliative radiotherapy 62
7 Skin 71
8 Head and neck: general considerations 88
9 Lip, ear, nose and treatment of the

neck in skin cancer 111
10 Oral cavity 122
11 Oropharynx cancer and unknown

primary tumour 134
12 Hypopharynx 146
13 Nasopharynx 156
14 Larynx 165
15 Salivary glands 177
16 Sinuses: maxilla, ethmoid and

nasal cavity tumours 185
17 Orbit 196
18 Central nervous system 205
19 Thyroid and thymoma 231
20 Lung 241
21 Mesothelioma 258
22 Breast 265
23 Lymphomas 283
24 Oesophagus and stomach 294
v
CONTENTS
25 Pancreas and liver 303
26 Rectum 311
27 Anus 323
28 Prostate 332
29 Bladder 351
30 Testis 359
31 Penis 365
32 Cervix 370
33 Uterus 384
34 Vagina 394
35 Vulva 401
36 Sarcoma 408
37 Paediatric tumours 415
38 Systemic irradiation 426
Appendix: List of abbreviations 433
Index 435
vi
Preface
Since the last edition of Practical Radiotherapy Planning in 1999, the practice of
radiotherapy has changed radically. Advances in imaging have been integrated with
technological developments in radiotherapy delivery so that 3D planning of
volumes has replaced 2D field arrangements. Major developments in tumour
localisation have included the publication of ICRU reports on target definition
(Report 62 in 1999 and Report 71 in 2004) and the possibility of registration of
different imaging modalities including CT, MRI and PET. In treatment delivery,
multi-leaf collimation has enabled treatments to be shaped to tumours and
intensity-modulated dose plans have provided solutions to previous planning
dilemmas. Accuracy of treatment delivery has been ensured by development of
portal imaging and daily image-guided and adaptive radiotherapy techniques.
This is therefore a completely new book, with new introductory chapters and a
changed structure within each tumour site chapter. We have been able to make
much more use of clinical images to illustrate important planning concepts. The
aim of the book, however, remains unchanged: to provide a guide to radiotherapy
treatment planning that is based on sound pathological and anatomical principles.
Complexity of treatment planning has increased greatly but this new edition
continues to emphasise the underlying principles of treatment, which can be
applied to conventional, conformal, and novel treatments, taking into account
advances in imaging and treatment delivery.
Much treatment is now given according to local, national or international
protocols and these should always be consulted and used when appropriate. Details
of drug and of some radiotherapy regimens are not given as they change frequently.
Again we have largely excluded recommendations for treatment of benign disease.
The ICRU report proposed a colour convention for outlining target volumes,
but with regret we have chosen different colours to ensure optimal reproduction
of plans. We have used the following scheme in all illustrations:
GTV dark blue
CTV cyan (light blue)
CTV2 magenta (purple)
PTV red
PTV2 lime green
OAR yellow, light yellow, light green, dark green
Two authors from the first three editions have been joined by two younger
colleagues from our departments to ensure that we continue to reflect the most
up-to-date approach to treatment planning. It would have been impossible to
vii
PREFACE
produce a book like this one, in the age of site-specialised practice and
multidisciplinary team working, without relying very heavily on expert colleagues
within our respective departments. We are most grateful to them for the time and
effort they have put in to ensuring the accuracy and appropriateness of what we
have written. Special thanks are due to Anna Winship, George Mikhaeel, David
Convery and Kim Ball at Guy’s and St Thomas’ Hospital, Mark Gaze at University
College Hospital, Richard Benson at Addenbrooke’s Hospital, and at the Norfolk
and Norwich University Hospital: Dinos Geropantas, Suat Loo, Adrian Harnett,
Max Dahele, Rob Wade, Jenny Tomes, and our radiological colleagues. We would
like to acknowledge our radiographer colleagues in both departments who have
helped us to find illustrations and plans. The work was supported in part by a
sabbatical leave award to Jane Dobbs from the Clinical Oncology Dean’s Fund of
the Royal College of Radiologists, London, whom we thank.
This small textbook cannot describe all the research which has been undertaken
to develop treatment schedules. We aim always to use evidence-based solutions
where they exist and have suggested, in a short list of information sources at the
end of each chapter, where more detailed data may be found. Some fields of research,
such as the use of gating and adaptive therapy, 3D rotational arc therapy, and
stereotactic radiotherapy are still undergoing evaluation and are therefore beyond the
scope of this book. We give a brief introduction to the principles and practice of
brachytherapy but details must be found elsewhere. Commonly used abbreviations
have been spelled out only at first mention in the book but are included in the
appendix for further reference.
Trainees in radiation oncology and radiographers, working within a
multidisciplinary team, will, we hope, continue to use our book to produce safe
and appropriate plans for common tumours.
Ann Barrett; Jane Dobbs; Stephen Morris; Tom Roques
2008
viii
What you need to 1
know before planning
radiotherapy
treatment
Introduction
Radiotherapy treatment can only produce good effects if it is delivered in an
appropriate clinical context. Attempting radical treatment for a patient with
metastatic disease, or one who is likely to die soon from cardiac or lung disease is
inappropriate. These decisions require a fine balance of judgement between therapeutic
optimism and nihilism and must be firmly based in good clinical history taking and
examination. The clinician must then be able to synthesise all the information about
the patient, tumour, investigations and previous treatment to make a decision about
whether radiotherapy should be given and if so, with radical or palliative intent.
Comorbidities, such as diabetes or vascular disease, which would affect the toxicity
of treatment, must also be considered.
Sometimes the decision to offer radiotherapy may be relatively simple if the
disease is common, the treatment effective and standardised, the histological
features well categorised, and imaging easy to interpret. Some breast cancers fit well
into this category. In contrast, decisions may be very difficult if there is no treatment
of proven benefit, the prognosis is uncertain, the patient’s general performance
status is poor, imaging is of limited usefulness or there is histological uncertainty.
Clinical experience and judgement is then critically important. This expertise is
built on the foundation of good history taking which enables us to set the patient’s
disease in the context of their own ideas, concerns and expectations. Have other
family members had radiotherapy with good or bad outcomes? Are they so
claustrophobic that they will not go into a scanner or treatment room? Do they
have other problems which would affect the feasibility of radiotherapy – arthritis
which limits joint movement, shortness of breath which prevents them lying flat,
heart valves or prostheses which may affect dose delivery?
Clinicians may consider that the new era of cross-sectional and functional imaging
has made examination of the patient irrelevant but it remains the essential foundation
of appropriate clinical judgements; for example, detection of a lymph node in the
axilla, otherwise overlooked in imaging, or the progression of tumour since the last
scan, may change a decision taken earlier in a multidisciplinary team meeting.
Classification systems
Many classification systems have been developed to ensure that clinicians throughout
the world share a common language as they describe patients, tumours, and their
1
PLANNING RADIOTHERAPY TREATMENT
treatment. This is essential to allow effective cancer registration and comparisons

of incidence, prognosis and outcome of treatments. Many protocols for treatment
are also based on such classification systems.
Pathological classification
The International Classification of Disease (ICD) version 10 of the World Health
Organization (WHO) has been in use since 1994. It is the international standard
diagnostic classification for epidemiology and health management. It is used in
hospital records and on death certificates, which in turn are the basis for compiling
mortality and morbidity statistics nationally and internationally. There is a
subclassification, the International Classification of Disease for Oncology version 3
(ICD-O-3) which is used in cancer/tumour registries to code site (topography)
and type (morphology) of neoplasms from the histopathology report.
Information about malignancy (malignant, benign, in situ or uncertain) and
differentiation is also coded. In the UK, this information is abstracted from clinical
notes by trained coders. The introduction of computerised systems suggests that
clinicians will be required to develop greater awareness of this classification, at least
in their own areas of expertise, especially if the information becomes essential data
before income is assured. ICD 10 and ICD-O-3 are available online.
The morphological information for ICD-O-3 coding comes from the pathologist
whose expertise is essential to establish a precise diagnosis and choose appropriate
treatment volumes. A pathology report will contain a description of the macroscopic
appearance of the gross tumour specimen, its size, margins and anatomical
relationships. It will describe the microscopic appearance after appropriate staining of
cut sections of the tumour, including features such as areas of necrosis. Recognition
of the tissue of origin and grading of the tumour will then often be possible.
There has been an explosion of new techniques in pathology, such as
immunocytochemical staining, immunophenotyping and fluorescence in situ
hybridisation, which may help to remove uncertainties about diagnoses following
conventional histopathological examination. Oncologists must be in constant
dialogue with their colleagues in pathology to ensure that they understand the
significance of results of these special investigations and know how to assess the
degree of certainty of the report.
Staging
Tumour stage, histological classification and grading determine prognosis and
treatment decisions. An internationally agreed system of staging is essential to
interpret outcomes of treatment and compare results in different treatment centres.
The behaviour of tumours in different sites is determined by the anatomical situation,
blood supply and lymphatic drainage, as well as the histological classification and
grading. Any staging system must take into account this variability. The most
commonly used systems are the UICC (Union Internationale Contre le Cancer)
TNM, AJCC (American Joint Committee on Cancer) and FIGO (Federation
Internationale de Gynecologie et d’Obstetrique) for gynaecological malignancy. The
TNM system describes the tumour extent (T) nodal involvement (N) and distant
metastases (M). This defines a clinical classification (cTNM) or a pathological one
2
Staging
(pTNM) which incorporates information derived from an excised tumour and any
draining lymph nodes that are also removed or sampled. Details of this system are
given in the TNM atlas which should be available and used wherever patients are
seen or results of investigation are correlated.
T staging includes measurement of the tumour either clinically, by imaging
techniques or by macroscopic examination of an excised specimen. Correct
pathological T staging can only be assured if the pathologist receives a completely
excised tumour with a rim of surrounding tissue. The tumour should not be cut into
or fixed except by the pathologist. Examination of the whole specimen is needed
to determine the highest grade of tumour (as there is frequently inhomogeneity
across the tumour), any vascular or lymphatic invasion or invasion of adjacent
tissues. Spread into a body space such as the pleural or peritoneal cavity affects T
stage as it changes prognosis.
Numbers are added to indicate extent of disease. T0 implies no primary tumour
(as after spontaneous regression of a melanoma). Categories T1–4 indicate tumours
of increasing size and/or involvement of lymphatic vessels or surrounding tissue. If
it is impossible to ascertain size or extent of primary tumour, it is designated Tx.
N classification describes whether there is lymph node involvement and if so,
how many nodes are involved. Pathological staging requires adequate excision of
the relevant lymph node compartment and a minimum number of nodes which
indicates that this has been achieved may be defined for each site. If there are
positive nodes, the ratio of negative to positive is of prognostic significance. For
example, one node positive out of four removed indicates a worse prognosis than
one out of 12. The size of tumour in the nodes must be recorded as well as any
extension through the capsule. Micrometastases are classified differently from
tumour emboli in vessels and this affects the N staging.
Identification and sampling of a sentinel node may give useful prognostic
information about other potential node involvement and help to choose appropriate
treatment strategies. In breast cancer, for example, it appears highly predictive
(90 per cent) for axillary node involvement.
M category indicates presence (M1) or absence (M0) of metastases to distant
sites.
For some sites, other staging systems have proved clinically more useful. These
include the FIGO system for gynaecological malignancy and Dukes’ classification
of colonic tumours. Useful atlases of patterns of lymph node involvement have
been devised for several tumour sites. These are included in subsequent chapters
where relevant. Recommendations for the most appropriate imaging techniques
for staging in different sites have been published.
Residual tumour after surgical excision is an important poor prognostic factor.
Examination of resection margins assigns tumours to categories: R0, no residual
tumour; R1, histologically detectable tumour at margins; and R2, macroscopic
evidence of residual tumour. Where serum markers (S) convey important prognostic
information, as in tumours of the testis, an S category has been introduced to the
TNM system.
TNM categorisation is often used to group tumours subsequently into stages
indicating local and metastatic extent and correlating with likely outcome.
Grading is defined by degree of differentiation as: G1, resemblance to tissue of
origin; G2, moderately well differentiated; G3, poorly differentiated; and G4,
3
undifferentiated tumours; with Gx used when it is not possible to determine

grading, as for example from a damaged specimen. GB signifies a borderline
malignant tumour (for example in the ovary). Grading systems have been agreed
to reduce the subjective element of these assessments.
Performance status
Performance status measures attempt to quantify cancer patients’ general wellbeing.
They are used to help to decide whether a patient is likely to tolerate a particular
treatment such as chemotherapy, whether doses of treatment need to be adjusted
or whether palliative treatment has been effective. The status of all patients should
be recorded using one of these scores at presentation and with any change in
treatment or the disease. They are also used as a measure of quality of life in clinical
trials.
There are various scoring systems, of which the most commonly used are the
Karnovsky and WHO scales for adults, and the Lansky score for children. The SF-36
is a short form (SF) survey of overall health (originally with 36 questions, now 12)
which gives a profile of mental and physical health. It has produced statistically
reliable and valid results in many reported studies. There are modifications of this
questionnaire for specific tumour sites. Another commonly used scale in quality of
life assessment is the HADS (Hospital Anxiety and Depression Scale) which is used
to measure changes in mental and emotional wellbeing during treatment or with
progression of disease.
Prognostic factors
All possible information which may help in predicting prognosis should be
collected in order to advise the patient and help make the most appropriate
treatment decision.
Screen detected cancers may have a better prognosis than tumours presenting
symptomatically because diagnosis is made earlier (breast and oesophagus).
However, in other situations, screening is as yet of unproven benefit (for example
chest X-ray for detecting early lung cancer).
Histological tumour type, grading and staging are most influential in
determining outcome for an individual patient, and new techniques of tumour
examination are yielding more information on gene function and expression which
may affect prognosis. Other factors which must also be considered include
epidemiological ones such as age, sex, lifestyle factors such as smoking, alcohol and
other drug use, obesity, and family history of disease. Other biological factors such
as performance status, which may reflect comorbidities, must be considered.
Biochemical tumour markers may be specific enough to give prognostic
information by their absolute value, as for example β subunit of human chorionic
gonadotrophin (β-hCG) levels in testicular cancer. However, they may be only
relatively poorly correlated with tumour volume, such as carcinoembryonic
antigen (CEA) in bowel cancer, but still be useful to indicate treatment response
or disease progression by their rise or fall.
One of the most important prognostic factors is whether there is effective
treatment for the condition. Because new treatments are being introduced all the
4
Influence of other treatments on radiotherapy
time, prognostic predictions must also be constantly reviewed and validated in
prospective controlled clinical trials.
Predictive tools based on population datasets are available for some tumours
such as ‘Adjuvant! On line’ for breast cancer, and Partin tables and the Memorial
Sloane Kettering nomogram for prostate cancer.
Predictive indicators are variables determined before treatment which give
information on the probability of a response to a specific treatment. Predictive
indices based on multiple indicators give an individual score which may help to
make decisions. These tools can only be developed by painstaking retrospective
analysis and careful prospective studies, but it is likely that their use will increase
steadily. They are important in determining strategies for treating different tumour
subsets in guidelines and protocols.
Increasingly, specific genetic profiles are correlated with natural history of
disease or outcome of treatment. Examples are the predictive value of MYCN
amplification in neuroblastoma, oestrogen/HER2 receptor status and response to
hormone therapy or trastuzumab, and predilection of patients with Li–Fraumeni
syndrome to development of second tumours. Microarray technology will provide
more genetically determined prognostic factors which will have to be taken into
account in planning treatment.
Influence of other treatments on radiotherapy

Surgery is most commonly used as a primary treatment for cancer. If performed
before radiotherapy, it removes the gross tumour volume (GTV) so that a clinical
target volume (CTV) must be used for planning (see Chapter 2).
If it is known from the time of diagnosis that both treatments will be needed,
the best sequencing has to be decided. If the tumour is initially inoperable,
radiotherapy first may produce tumour shrinkage which makes complete excision
possible, thereby improving outcome. Radiotherapy may increase surgical
complications if the interval between the two treatments is not optimal. Surgery
before radiotherapy will alter normal anatomy, causing problems for planning
unless pre- and postsurgical image co-registration is used.
Effective chemotherapy may produce complete resolution of the primary tumour
(no residual GTV), and potentially increased acute normal tissue effects. The treatment
the patient receives will be the best possible only if all these potential interactions are
taken into consideration by all members of the team working together.
A true therapeutic gain from chemotherapy and radiotherapy together requires
either more cell kill for the same level of normal tissue damage (which is useful
in radio-resistant tumours) or the same cell kill with reduced normal tissue
effects (useful for tumours cured with low dose radiotherapy). Concurrent
chemoradiotherapy is now used for many tumours. It may improve outcomes by
promoting spatial cooperation in cell killing, where the chemotherapy kills metastatic
cells, and the radiotherapy those cells in the local tumour, or in-field cooperation
where exploitation of differing molecular, cellular or tissue effects produces more
cell kill than when the two agents are given sequentially.
Patients themselves may be using alternative therapies with possible effects on
efficacy and complications of surgery, chemotherapy and radiotherapy. Antioxidants
may interfere with free radical production which effects cell kill. St John’s wort,
5
used for depression, may affect the metabolism of some drugs. Aspirin and gingko
biloba may increase risk of bleeding, and phyto-oestrogens may affect hormonally
sensitive cancers. Lifestyle factors such as smoking or eating many vegetables
during radiotherapy may influence the severity of acute treatment side effects such
as mucositis or diarrhoea.
Systems for recording outcomes of treatment

In the past, various systems have been used for recording outcomes of treatment,
including the WHO and European Organisation for Research and Treatment of
Cancer (EORTC) criteria. Following advances in imaging techniques and
improvements in treatment, a new set of tumour response criteria have been agreed –
the Response Evaluation Criteria in Solid Tumours (RECIST). Other outcomes are
recorded as the time to a specific event: time to progression for tumours with partial
response (progression-free survival), time to local recurrence where there has been a
complete remission (relapse-free survival), time to distant metastases, time to death
from any cause (overall survival), time to second malignancy. The start and end dates
for these time periods must be clearly defined and stated as, for example, date of first
treatment, date of randomisation, date of imaging of relapse, or date of histological
proof of relapse. There is still no clear convention for defining these dates.
Quality of life measures are also essential for assessing treatment effects.
Standardised scales such as the EORTC, QLQ-C36 and C30 have been validated
and can be used for many cancer sites and in many countries.
Acute side effects can be recorded using the NCI-CTC 3 (Common Toxicity
Criteria version 3). This is a complex and comprehensive system that has only
partially been adopted. Late effects are recorded either using the Radiation Therapy
Oncology Group (RTOG) criteria or the European LENT-SOMA classification or
other simpler schemes for individual body sites such as gynaecological tumours.
Use of the NCI-CTC 3 is recommended until a more concise system gains
international recognition.
Clinical anatomy
Modern radiotherapy planning requires a comprehensive knowledge of cross-
sectional anatomy and the ability to visualise structures in three dimensions.
Formal teaching in anatomy should be part of training, using standard atlases,
various online resources, e.g. the Visible Man, and three-dimensional (3D) images,
which have become more accessible with picture archiving and conservation
systems (PACS). We attempt to give some relevant anatomical details in the
following chapters, but collaboration with a diagnostic radiologist will be essential
for accurate GTV delineation. Oncologists will tend to develop expertise in their
own specialist area, but are unlikely to be familiar with all the possible normal
variants and anomalies which may occur.
Protocols and guidelines

In the UK, there has been a proliferation of guidance, guidelines and protocols for
the management of cancer. Some, such as Improving Outcomes Guidance (IOG)
6
Information sources
and decisions of the National Institute for Health and Clinical Excellence (NICE)
have a mandatory element which ensures their adoption. The firmest base for
clinical decision-making is evidence of effectiveness of treatment from well-
designed prospective randomised clinical trials (RCTs). Where this is lacking,
careful analysis of outcomes data can be very informative and has the advantage
of wider generalisability. National and international trial protocols offer useful
information for treatment planning, as they represent current consensus on best
practice, as for example, how to scan lung cancer for treatment planning, or what
quality assurance programme to use. Departments will have their own written
protocols for treatment at different sites to ensure consistency and quality and to
avoid errors.
Not all patients’ circumstances will fit within standardised guidelines and
protocols, although these should be followed whenever possible for best outcomes.
However, with the rate of change in treatment in oncology, there should be a
constant cycle of writing guidelines and using, auditing, challenging and rewriting
them.
Information sources
Adjuvant! Online. www.adjuvantonline.com (accessed 26 November 2008).
Clark A, Fallowfield LJ (1986) Quality of life measurement in patients with malignant disease. Royal
Soc Med 79:165–9.
Cochran AJ, Roberts AA, Saida T (2003) The place of lymphatic mapping and sentinel node biopsy
in oncology. Int J Clin Oncol 8: 139–50.
Common Toxicity Criteria for Adverse Events v3.0 (CTCAE) (2006)
http://ctep.cancer.gov/forms/CTCAEv3.pdf (accessed 26 November 2008).
Dukes CE (1949) The surgical pathology of rectal cancer. J Clin Path 2: 95–8 and at
www.emedicine.com.
ECOG/WHO Oken MM, Creech RH, Tormey DC et al. (1982) Toxicity and response criteria of the
Eastern Oncology Cooperative Group. Am J Clin Oncol 5: 649–55.
Ernst E, Cassileth BR (1998) The prevalence of complementary/alternative medicine in cancer.
Cancer 83: 777–82.
Gipponi M, Solari N, Di Somma FC et al. (2004) New fields of application of the sentinel node biopsy
in the pathological staging of solid neoplasms: Review of the literature and surgical
perspectives. J Surg Oncol 85: 171–9.
Imaging for Oncology: Collaboration Between Clinical Radiologists and Clinical Oncologists in
Diagnosis, Staging, and Radiotherapy Planning. (2004) Royal College of Radiologists, London.
Karnovsky DA, Burchenal JH (1949) The clinical evaluation of chemotherapeutic agents in cancer.
In: MacLeod CM (ed) Evaluation of Chemotherapeutic Agents. Columbia University Press,
New York, p. 196 and at http://en.wikipedia.org/wiki/Performance_status.
Kattan MW, Eastham JA, Stapleton AM et al. (1998) A prospective nomogram for disease
recurrence following radical prostatectomy for prostate cancer. J Natl Cancer Inst 90: 766–71.
Lansky SB, List MA, Lansky LL et al. (1987) The measurement of performance in childhood cancer
patients. Cancer 60: 1651–6.
NICE guidance. www.nice.org.uk.
Partin AW, Mangold LA, Lamm DM et al. (2001) Contemporary update of prostate cancer staging
nomograms (Partin tables) for the new millennium. Urology 58: 843–8.
7
Pecorelli S, Ngan HYS, Hacker NF (2006) Staging Classifications and Clinical Practice Guidelines for
Gynaecological Cancers. Reprinted from Int J Gynae Obstet 2002; 70: 207–312 at
www.figo.org/docs/staging_booklet.pdf (accessed 2 December 2008).
QLC-30. http://groups.eortc.be/qol/questionnaires_qlqc30.htm (accessed 26 November 2008).
Recommendations for Cross-sectional Imaging in Cancer Management (2006). Royal College of
Radiologists, London.
Seiwert TY, Joseph K, Salama JK et al. (2007) The concurrent chemoradiation paradigm – general
principles. Nat Clin Pract Oncol 4: 86–100.
SF36. www.pdmed.bham.ac.uk/trial/Clinicians/SF36%20Questionnaire.pdf (accessed 2 December
2008).
Sobin LH, Wittekind CH (2002) TNM Classification of Malignant Tumours, 6th edn. John Wiley,
Chichester.
Therasse P, Arbuck SG, Eisenhauer EA et al. (2000) New guidelines to evaluate the response to
treatment in solid tumours. J Natl Cancer Inst 92: 205–16 and at
http://ctep.cancer.gov/forms/TherasseRECISTJNCI.pdf (accessed 26 November 2008).
The Visible Human Project, National Library of Medicine 1989–1995.
www.nlm.nih.gov/research/visible/visible_human.html (accessed 2 December 2008).
Radiation Therapy Oncology Group. www.rtog.org/hnatlas/main.html.
Wittekind CH, Hutter R, Greene FL et al. (2005) TNM Atlas: Illustrated Guide to the TNM
Classification of Malignant Tumours, 5th edn. John Wiley, Chichester.
World Health Organization (2007) International Classification of Disease – Version 10, Chapter 2
Neoplasms. www.who.int/classifications/apps/icd/icd10online (accessed 26 November 2008).
Zigmond AS, Snaith RP (1983) Hospital Anxiety and Depression Scale. Acta Pyschiatr Scand 67:
361–70 and at www.sign.ac.uk/guidelines (Guideline 57).
8
Principles of 2
radiotherapy planning
The practice of radiotherapy requires not only excellent clinical skills but also
appropriate technical expertise. Chapter 1 considered some factors contributing to
making good clinical judgements; Chapter 2 outlines the specialist knowledge
required to plan radiotherapy treatment.
Target volume definition

A common international language for describing target volumes is found in
International Commission on Radiation Units (ICRU) published recommendations
Report 50 (1993), 62 (1999) and 71 (2004). These contain clear definitions (Fig. 2.1)
to enable centres to use the same criteria for delineating tumours for radiation so
that their treatment results can be compared.
IrV
TrV PTV
CTV
GTV
Figure 2.1 ICRU target volume definitions showing

GTV, CTV, PTV, treated and irradiated volume.
Reproduced with permission from ICRU (1993)
Prescribing, Recording and Reporting Photon Beam
Therapy. ICRU report 50.
■ Gross tumour volume

Gross tumour volume (GTV) is the primary tumour or other tumour mass shown
by clinical examination, at examination under anaesthetic (EUA) or by imaging.
GTV is classified by staging systems such as TNM (UICC), AJCC or FIGO.
Tumour size, site and shape may appear to vary depending on the imaging technique
used and an optimal imaging method for each particular tumour site must therefore
also be specified. A GTV may consist of primary tumour (GTV-T) and/or metastatic
lymphadenopathy (GTV-N) or distant metastases (GTV-M). GTV always contains
the highest tumour cell density and is absent after complete surgical resection.
9
PRINCIPLES OF RADIOTHERAPY PLANNING
■ Clinical target volume

Clinical target volume (CTV) contains the GTV when present and/or subclinical
microscopic disease that has to be eradicated to cure the tumour. CTV definition is
based on histological examination of post mortem or surgical specimens assessing
extent of tumour cell spread around the gross GTV, as described by Holland et al.
(1985) for breast cancer. The GTV-CTV margin is also derived from biological
characteristics of the tumour, local recurrence patterns and experience of the
radiation oncologist. A CTV containing a primary tumour may lie in continuity with
a nodal GTV/CTV to create a CTV-TN (e.g. tonsillar tumour and ipsilateral cervical
nodes). When a potentially involved adjacent lymph node which may require elective
irradiation lies at a distance from the primary tumour, separate CTV-T and CTV-N
are used (Fig. 2.2), e.g. an anal tumour and the inguinal nodes.
CTV can be denoted by the dose level prescribed, as for example, CTV-T50 for a
particular CTV given 50 Gy. For treatment of breast cancer, three CTVs may be used
for an individual patient: CTV-T50 (50 Gy is prescribed to the whole breast);
CTV-T66 (66 Gy to the tumour bed); and CTV-N50 (50 Gy to regional lymph nodes).
Variation in CTV delineation by the clinician (‘doctor’s delineation error’) is the
greatest geometrical uncertainty in the whole treatment process. Studies comparing
outlining by radiologists and oncologists have shown a significant inter-observer
variability for both the GTV and/or CTV at a variety of tumour sites. This is
greater than any intra-observer variation. Published results for nasopharynx, brain,
lung, prostate, medulloblastoma and breast all show significant discrepancies in the
volumes outlined by different clinicians. Improvements can be made with training
in radiological anatomy which enables clinicians to distinguish blood vessels from
lymph nodes and to identify structures accurately on computed tomography (CT)
and magnetic resonance imaging (MRI). Joint outlining by radiologists and
oncologists can improve consistency and ensure accurate interpretation of imaging
of the GTV. Consensus guidelines such as those for defining CTV for head and
neck nodes (Gregoire et al. 2000) and pelvic nodes (Taylor et al. 2005) have
improved CTV delineation greatly. Protocols for outlining GTV and CTV at all
tumour sites are needed and suggestions are made in each individual chapter.
■ Planning target volume

When the patient moves or internal organs change in size and shape during a
fraction of treatment or between fractions (intra- or inter-fractionally), the
position of the CTV may also move. Therefore, to ensure a homogeneous dose to
the CTV throughout a fractionated course of irradiation, margins must be added
around the CTV. These allow for physiological organ motion (internal margin)
and variations in patient positioning and alignment of treatment beams (set-up
margin), creating a geometric planning target volume. The planning target volume
(PTV) is used in treatment planning to select appropriate beams to ensure that the
prescribed dose is actually delivered to the CTV.
■ Organ motion/internal margin

Variations in organ motion may be small (e.g. brain), larger and predictable (e.g.
respiration or cardiac pulsation), or unpredictable (e.g. rectal and bladder filling).
10
Figure 2.2 ICRU illustrations to show (a) GTV-T plus GTV-N in continuity (CTV-TN) and
(b) CTV-T and CTV-N at a distance. Reproduced with permission from ICRU (2004)
Prescribing, Recording and Reporting Electron Beam Therapy. ICRU report 71.
When treating lung tumours, the displacement of the CTV caused by respiration can
be dealt with in several ways: by increasing the CTV-PTV margin eccentrically to
include all CTV positions during a respiratory cycle; by using suspended respiration
with a technique such as the active breathing control (ABC) device; or by delivery
of radiation using gating or respiratory correlated CT scanning and treatment.
Protocols for minimising effects on the CTV of variations in bladder and rectal
filling are described in relevant chapters. Uncertainties from organ motion can also
11
be reduced by using fiducial markers, and published results are available for lung,
prostate and breast tumours. Radio-opaque markers are inserted and imaged at
localisation using CT or MRI, and at treatment verification, using portal films,
electronic portal imaging devices (EPIDs) or online cone beam CT image-guided
radiotherapy (IGRT).
The internal margin therefore allows for inter- and intra-fractional variations in
organ position and shape which cannot be eliminated.
■ Set-up variations/set-up margin

During a fractionated course of radiotherapy, variations in patient position and in
alignment of beams will occur both intra- and inter-fractionally, and a margin for
set-up error must be incorporated into the CTV-PTV margin. Errors may be
systematic or random.
Systematic errors may result from incorrect data transfer from planning to dose
delivery, or inaccurate placing of devices such as compensators, shields, etc. Such
systematic errors can be corrected.
Random errors in set-up may be operator dependent, or result from changes
in patient anatomy from day to day which are impossible to correct. Accuracy of
set-up may be improved with better immobilisation, attention to staff training
and/or implanted opaque fiducial markers, such as gold seeds, whose position can
be determined in three dimensions at planning, and checked during treatment
using portal imaging or IGRT. Translational errors can thereby be reduced to
1 mm and rotational errors to 1°.
Each department should measure its own systematic and random errors for each
treatment technique by comparing portal imaging and digitally reconstructed
radiographs (DRRs). These measurements are then incorporated into the CTV-
PTV margin using the formula devised by Van Herk, where the CTV is covered
for 90 per cent of the patients with the 95 per cent isodoses:
PTV margin 2.5Σ 0.74 σ
where Σ total standard deviation (SD) computed as the square root of the sum of
the squared individual SD values of all systematic errors for organ motion and set-up;
and σ total SD of all random errors combined quadratically in a similar way.
This provides a population-derived standard CTV-PTV margin for a particular
technique in a given centre and can be non-isotropic in cranio-caudal, transverse
and anteroposterior (AP) directions. Accurate treatment delivery depends on
reducing or eliminating systematic errors and requires a high level of awareness of
all staff throughout the many different work areas from localisation through to
treatment.
Other theories about how to incorporate organ motion and the uncertainty of
the ‘mean’ position of the CTV on a snapshot CT scan used for localisation have
been proposed. Van Herk suggests a volume large enough to contain the mean
position of the CTV in 90 per cent of cases, called the systematic target volume
(STV) (British Institute of Radiology 2003). Collection of data on the precise CT
location of tumour recurrences in relation to the original target volume is
important to improve margin definition.
12
Immobilisation
■ Treated volume
This is the volume of tissue that is planned to receive a specified dose and is
enclosed by the isodose surface corresponding to that dose level, e.g. 95 per cent.
The shape, size and position of the treated volume in relation to the PTV should
be recorded to evaluate and interpret local recurrences (in field versus marginal)
and complications in normal tissues, which may be outside the PTV but within the
treated volume.
■ Conformity index
This is the ratio of PTV to the treated volume, and indicates how well the PTV is
covered by the treatment while minimising dose to normal tissues.
■ Irradiated volume
This is the volume of tissue that is irradiated to a dose considered significant in
terms of normal tissue tolerance, and is dependent on the treatment technique
used. The size of the irradiated volume relative to the treated volume (and integral
dose) may increase with increasing numbers of beams, but both volumes can be
reduced by beam shaping and conformal therapy.
■ Organs at risk
These are critical normal tissues whose radiation sensitivity may significantly
influence treatment planning and/or prescribed dose. Any movements of the organs
at risk (OAR) or uncertainties of set-up may be accounted for with a margin similar
to the principles for PTV, to create a planning organ at risk volume (PRV). The
size of the margin may vary in different directions. Where a PTV and PRV are close
or overlap, a clinical decision about relative risks of tumour relapse or normal tissue
damage must be made. Shielding of parts of normal organs is possible with the use
of multi-leaf collimation (MLC). Dose–volume histograms (DVHs) are used to
calculate normal tissue dose distributions.
Immobilisation
The patient must be in a position that is comfortable and reproducible (whether
supine or prone), and suitable for acquisition of images for CT scanning and
treatment delivery. Immobilisation systems are widely available for every anatomical
tumour site and are important in reducing systematic set-up errors. Complex
stereotactic or relocatable frames (e.g. Gill–Thomas) are secured to the head by
insertion into the mouth of a dental impression of the upper teeth and an occipital
impression on the head frame, and are used for stereotactic radiotherapy with a
reproducibility of within 1 mm or less. Perspex shells reduce movement in head and
neck treatments to about 2 mm. The technician preparing the shell must have
details of the tumour site to be treated, e.g. position of the patient (prone, supine,
flexion or extension of neck, arm position, etc.). An impression of the relevant area
13
(made using quick setting dental alginate or plaster of Paris) is filled with plaster and
this form is used to make a Perspex shell by vacuum moulding. The shell fits over
the patient and fastens to a device on the couch with Perspex straps and pegs in at
least five places. Alternatively, thermoplastic shells can be made by direct moulding
of heat-softened material on the patient and these have a similar degree of accuracy.
Relocatable whole body fixation systems using vacuum moulded bags of
polystyrene beads on a stereotactic table top restrict movement to 3–4 mm and are
used to immobilise the trunk and limbs with markings on the bag instead of on
the patient’s skin. Where the patient has kyphosis, scoliosis or limitation of joint
movement, extra limb pads or immobilisation devices may be required. Metallic
prostheses, abdominal stomata and the batteries of pacemakers must be located
and excluded from the radiation volume where possible. Details of immobilisation
devices are discussed in each tumour site chapter.
Parameters of limb rests, thoracic or belly boards, foot rests and leg restraints,
Perspex shells and skin tattoos should be clearly recorded to avoid transfer errors
between the planning process and subsequent treatment. Gantry and couch top
flexibility should be measured and couch sag avoided by using rigid radiolucent
carbon fibre tables. Table tops must have fixtures for immobilisation devices and
laser light systems are essential in CT, simulator and treatment units (Fig. 2.3).
Protocols for bladder and rectal filling, respiration and other patient parameters
must be documented at localisation, and reproduced daily during treatment to
minimise uncertainties.
CT scans taken for localisation are only a single snapshot, and the CT scan
should be repeated daily on several days to measure variation in organ motion and
systematic set-up errors for an individual patient. These values can then be used to
inform the CTV-PTV margin on an individual basis rather than using population
derived margin values. This is known as adaptive radiotherapy (ART). Kilovoltage
(kV), cone beam CT, and megavoltage (MV) imaging on treatment machines
make it possible to obtain CT images immediately before treatment. While
Figure 2.3 Patient positioned to show immobilisation on the CT scanner with arms up
and laser lights used to prevent rotational set-up errors.
14
Data acquisition
resolution is not as good as with diagnostic CT, the use of fiducial markers and
image registration protocols enables daily online IGRT. With this technique only
intra-fractional variations and the doctor’s CTV delineation error remain.
Data acquisition
Accurate 3D data about tumour, target volumes and organs at risk are acquired in
relation to external reference points under exactly the same conditions as those used
for subsequent treatment. Optimal diagnostic imaging modalities are chosen for
each tumour site according to protocols developed with diagnostic radiologists.
Multi-slice CT, MR with dynamic scanning, 3D ultrasound, positron emission
tomography (PET) and single photon emission computed tomography (SPECT)
have provided a wealth of anatomical, functional and metabolic information about
the GTV. These MRI and PET images are fused with CT planning scans to optimise
tumour localisation using a CT scanner with virtual simulation. Alternatively, when
these are unavailable, a simulator or simulator CT is used for 2D planning.
■ CT scanning
CT scanning provides detailed cross-sectional anatomy of the normal organs, as well
as 3D tumour information. These images provide density data for radiation dose
calculations by conversion of CT Hounsfield units into relative electron densities
using calibration curves. Compton scattering is the main process of tissue interaction
for megavoltage beams and is directly proportional to electron density. Hence CT
provides ideal density information for dose corrections for tissue inhomogeneity,
such as occurs in lung tissue. Clinical studies have shown that 30–80 per cent of
patients undergoing radiotherapy benefit from the increased accuracy of target
volume delineation with CT scanning compared with conventional simulation. It has
been estimated that the use of CT improves overall 5-year survival rates by around
3.5 per cent, with the greatest impact on small volume treatments.
CT scans taken for radiotherapy treatment planning usually differ from those taken
for diagnostic use. Ideally, planning CT scans are taken on a dedicated radiotherapy
CT scanner by a therapy trained radiographer. The scanner should have the largest
possible aperture to aid positioning of the patient for treatment. The standard
diagnostic CT aperture is 70 cm but 85 cm wide-bore scanners are available, which
are helpful for large patients and those being treated for breast cancer, who lie with
both arms elevated on an inclined plane that can be extended up to around 15°. The
CT couch must be flat topped with accurate couch registration to better than 1 mm.
The patient is positioned using supporting aids and immobilisation devices and
aligned using tattoos and midline and lateral laser lights identical to those used for
subsequent radiotherapy treatment. A tattoo is made on the skin over an immobile
bony landmark nearest to the centre of the target volume (e.g. pubic symphysis). It
is marked with radio-opaque material such as a catheter or barium paste for
visualisation on the CT image. Additional lateral tattoos are used to prevent lateral
rotation of the patient and are aligned using horizontal lasers (see Fig. 2.3).
Oral contrast medium is used in small concentrated dose to outline small bowel
as an organ at risk (e.g. for pelvic intensity-modulated radiotherapy [IMRT]
treatment), but care must be taken to avoid large quantities which may cause
15
diuresis and overfill the bladder. Intravenous contrast is given for patients with
lung and mediastinal tumours to differentiate between mediastinal vascular
structures, tumour and lymph nodes, and in the pelvis to enhance blood vessels for
CTV delineation of lymph nodes. Structures such as the vulva, vaginal introitus,
anal margin, stomata and surgical scars may be marked with radio-opaque material.
Patients with locally advanced tumours should be examined in the treatment
position and tumour margins clearly marked. Protocols to reduce organ motion,
for example by emptying the bladder before bladder radiotherapy and the rectum
before prostate treatment, need to be in place. Patients are given information
leaflets to explain the importance of, and rationale for, these procedures.
Multi-slice CT scanners perform a scan of the entire chest or abdomen in a few
seconds with the patient breathing normally. Scanning for lung tumours can
involve ‘slow’ CT scans, respiratory correlated CT scans, gating, or use of the ABC
device to cope with the effect of respiratory movement.
Protocols for CT scanning are developed with the radiologist to optimise tumour
information, to ensure full body contour in the reconstruction circle and scanning of
relevant whole organs for DVHs. DRRs are produced from CT density information
and are compared with electronic portal images (EPIs). Contiguous thin CT slices
are obtained at 2–5 mm intervals for the head and 3–10 mm for the body.
■ Contouring
CT scans are transferred digitally to the target volume localisation console using
an electronic network system which must be compliant with DICOM 3 and
DICOM RT protocols (Fig. 2.4). The GTV, CTV, PTV, body contour and normal
organs are outlined by a team of radiation oncologist, specialist radiologist and
planning technician, with appropriate training. Where MRI is the optimal imaging
modality for tumours such as the prostate, uterus, brain, head and neck and
sarcomas, it is incorporated into target volume definition by image fusion. Treatment
planning based on MR images alone is reported, but more commonly CT and MR
images are co-registered, ideally scanning using a flat MRI couch top to aid
matching. PET or PET-CT images may give additional information for head and
neck tumours, lymphomas, lung and gynaecological tumours, and SPECT for
brain tumours. Multimodality image fusion of all these images in the treatment
planning process is ideal for accurate delineation of the target.
Departmental protocols for target volume delineation are essential for each
tumour site; these should define optimal window settings, how to construct the
CTV, 3D values for CTV-PTV margins, type of 3D expansion software and
method of outlining for each OAR to be used. ‘Doctor’s delineation errors’
resulting from contouring are said to be the most uncertain part of the whole
planning process and training in, and validation of these procedures is essential.
Contouring starts with definition of the GTV on a central slice of the primary
tumour and then on each axial CT image moving superiorly and then inferiorly.
Involved nodes can then be defined in the same way. Viewing the GTV on coronal
and sagittal DRRs ensures consistency of definition between slices so that no artificial
steps in the volume are created. A volume should not be copied or cut and pasted
onto sequential slices for risk of pasting an error: it is more accurate to redraw the
GTV on each slice. If there are slices where the GTV cannot be defined – for example
16
Data acquisition
Figure 2.4 Network for transferring data between steps in the radiotherapy planning
process.
due to dental artefact – the planning software will usually allow the missing
contours to be interpolated from those either side.
■ CT virtual simulation
Using CT data, software generates images from a beam’s eye perspective, which
are equivalent to conventional simulator images. External landmarks are used to
define an internal isocentre for treatment set-up. The CT simulator provides
maximal tumour information as well as full 3D capabilities (unlike the simulator
CT facility). It is particularly useful for designing palliative treatments such as for
lung and vertebral metastases, as well as for some breast treatments using
tangential beams, which can be virtually simulated and then 3D planned. The
ability to derive CT scans, and provide target volume definition, margin generation,
and simulation all on one workstation, provides a rapid solution.
■ Conventional simulator
For palliative treatments, a simulator may still be used to define field borders following
the 50 per cent isodose line of the beam, rather than a target volume. A simulator is
an isocentrically mounted diagnostic X-ray machine which can reproduce all the
movements of the treatment unit and has an image intensifier for screening. The
patient is prepared in the treatment position exactly as described above for CT
17
scanning. The machine rotates around the patient on an axis centred on a fixed point,
the isocentre, which is 100 cm from the focal spot and is placed at the centre of the
target volume. Digital images or radiographs are used to record the field borders
chosen by reference to bony landmarks. The simulator is commonly used either for
palliative single field treatments of bone metastases or to define opposing anterior and
posterior fields for palliative treatment to locally advanced tumour masses.
■ Simulator-CT
A CT mode attached to the simulator gantry can be used to produce images with
a relatively limited resolution during the simulation process. This provides both
external contouring and some normal anatomical data, such as lung and chest wall
thickness, for simple inhomogeneity corrections. Images do not give detailed
tumour information or accurate CT numbers. These scans are time consuming to
obtain, and are therefore usually limited to the central, superior and inferior levels
of the target volume.
Dose solutions
When the PTV and normal organs have been defined in 3D, the optimal dose
distribution for treating the tumour is sought. Consultation with a dosimetrist is vital
to select the best parameters. For example, a treatment machine must be chosen
according to percentage depth dose characteristics and build up depth which will vary
with energy and beam size as shown in Table 2.1. These can be used to calculate doses
for treatment using single fields and to learn the construction of isodose distributions
using computer modelling. Other factors to be considered in the choice of machine
are the effect of penumbra on beam definition, the availability of independent or
multi-leaf collimators, facilities for beam modification and portal imaging.
Table 2.1 Data from treatment machines in common use, showing variation of Dmax and
percentage depth dose (DD) with energy
Machine Energy (MV) FSD (cm) 10 ⴛ 10 cm field*
Dmax depth (cm) % DD at 10 cm
Cobalt-60 1.25 100 0.5 58.7

Linear accelerator 6 100 1.5 67.5
10 100 2.3 73
16 100 2.8 76.8
*Maximum field size at 100 cm is 40 40 cm.
Three levels of dose planning and treatment are described here:

■ Conventional treatment uses single or opposing beams, with or without 2D
dose distributions, with compensators and simple shielding.
■ Conformal treatment involves target volume delineation of tumour and normal
organs according to ICRU principles with 3D dose calculations using MLC to
shape beams.
18
Dose solutions
■ Complex treatment includes the use of IMRT to shape the fluence of the beam,
dynamic treatments, IGRT and 3D or 4D delivery.
Most dose computation is done using 3D computerised treatment planning
systems which are programmed with beam data from therapy machines. These
systems require careful quality control programmes. Following production of
a satisfactory isodose distribution to a given target volume using 2D or 3D
algorithms, the calculations are checked by a physicist and detailed instructions for
delivery are prepared by radiographers on the therapy unit.
■ Conventional treatment
Single fields may be used with borders defined by the 50 per cent isodose for bone
metastases. Parallel opposing beams are used for speed and ease of set-up for
palliative treatments (e.g. lung), for target volumes of small separation (e.g. larynx)
or tangential volumes (e.g. breast). Isodose distributions show that the 95 per cent
isodose does not conform closely to the target volume, the distribution of dose is
not homogeneous and much normal tissue is irradiated to the same dose as the
tumour. Beam modification with the use of wedges alters the dose distribution to
compensate for missing tissue, obliquity of body contour or a sloping target
volume, and may produce a more homogeneous result.
For many tumours seated at depth, a radical tumour dose can only be achieved
with a combination of several beams if overdose to the skin and other superficial
tissues is to be avoided. When multiple beams are chosen for a plan, variable
wedges can be used to attenuate the beam and thereby avoid a high dose area at
beam intersections. To achieve the same dose at the patient, the number of
monitor units set will have to be increased compared with those for an open field.
Computerised dose planning systems are used to construct an isodose distribution
with beams of appropriate energy, size, weighting, gantry angle and wedge to give
a homogeneous result over the target volume.
■ Inhomogeneity corrections
Attenuation of an X-ray beam is affected by tissue density, being less in lung than
bone. This variation affects both the shape of the dose distribution and the values
of the isodoses. Lung tissue should therefore be localised when planning treatment
for tumours of the thorax (e.g. lung, breast, oesophagus, mediastinum). The
relative electron density of lung compared with water is in the range 0.2–0.3 and
these values are used to correct for inhomogeneity.
When 2D conventional planning is used, correction is only valid at the planned
central slice of the target volume, e.g. breast treatment planned with a simulator
using central lung distance (CLD). Using CT scanning, the whole lung is localised
in 3D and a pixel by pixel correction made for all tissue densities by conversion of
CT numbers into relative electron densities using calibration curves. CT numbers
are affected by contrast agents but dose distributions in the chest and abdomen are
not significantly changed by the quantities used in most CT scanning protocols.
However, large amounts of gas in the rectum can cause organ motion of the
prostate and uterus as well as affecting CT densities.
19
■ Beam junctions
When treatment is given to target volumes that lie adjacent to one another,
consideration must be given to the non-uniformity of dose in the potential overlap
regions caused by divergence of the adjacent beams. If the beams abut on the
skin surface, they will overlap with excess dose at depth. If there is a gap between
beams at the skin, there will be a cold area in the superficial tissues. Clinical examples
of this problem include treatment of (a) adjacent vertebrae with single posterior
fields which may be separated in time, where there is a risk of overlap of dose at the
underlying spinal cord, (b) primary breast cancer and adjacent lymph nodes, where
a single isocentric technique centred at the junction can be used (see Chapter 22),
(c) primary head and neck tumours and their regional nodes which can be treated
with IMRT or matched photon and electron fields to avoid overlap over the spinal
cord (see Chapter 8) and (d) primary central nervous system (CNS) tumours such as
medulloblastoma, where beams are matched at the anterior spinal cord and junctions
between beams shifted during the course of treatment (see Chapter 18).
Various techniques have been developed to minimise dose heterogeneity at beam
junctions in these different clinical situations. Half beam blocking using shielding
or independent collimator jaws can be used to eliminate divergence up to the match
line, but accuracy is then dependent on precise immobilisation and reliability of skin
marks to reproduce the match perfectly. Couch rotation can be used to remove
beam divergence when matching breast and lymph node irradiation. However, for
some sites, it is still common to match beams by using a gap between beams so that
the beam edges converge at a planned depth (Fig. 2.5). The dose in the triangular
gap (x) below the skin surface will be lower than at the point P where the beams
converge because it lies outside the geometric margins of both beams. Doses at (y)
are higher because they include contributions from both beams. The positioning of
point P anatomically will vary according to the aim of treatment. If treatment is for
medulloblastoma, a homogeneous dose is required to potential tumour cells within
the spinal cord which is therefore placed at point P, and point P is moved in a
cranio-caudal direction at regular intervals to prevent any risk of overdose at
Figure 2.5 Technique for matching two

beams I and II (length I1 and l2) at a point P
at depth d by allowing a gap on the skin
surface (x, y, s – see text).
20
Dose solutions
junctions. Where treatment is aimed at metastases in adjacent vertebral bodies, it is
important to avoid overdosage at the spinal cord which is therefore placed in the
superficial cold triangle (x) with point P anterior to it. The gap on the skin is the
sum of the beam divergence of each beam. It is calculated as the distance from
the edge of the beam as defined by the 50 per cent isodose to the point of
convergence P measured perpendicular to the central axis and marked (s) spread
(beam divergence) for each beam. Graphs have been drawn up expressing beam
divergence as a function of the depth below the skin for different field sizes and
focus skin distance (FSD). Once the gap has been calculated, it may be necessary to
increase it slightly to allow for possible movement of patient or skin tattoos, to
ensure that there is no overdosage. Whenever possible, a patient should be treated
in the same position (supine or prone) for matching adjacent fields.
When planning a new treatment for metastatic disease in the spine, previous
treatment fields should be reconstructed from films and records and the patient
placed in the same position to ensure there is no overlap.
■ Electron therapy
Electron therapy may be used to treat superficial tumours overlying cartilage and
bone (for example nose, ear, scalp and dorsum of hand), in preference to superficial
or orthovoltage therapy where there is increased bone absorption due to the
photoelectric effect. There is a sharp fall in dose beyond the 90 per cent isodose
(4–12 MeV) and electron energy is chosen so that the target volume is
encompassed by the 90–95 per cent isodose at the deep margin. Electrons at higher
energies (15–25 MeV) may be used for treating cervical lymph nodes overlying
spinal cord, parotid tumours and in mixed beams with photons.
The effective treatment depth in centimetres is about one-third of the beam energy
in MeV and the total range about half (Fig. 2.6) but this is dependent on field size
(especially at 40 mm). Different tissue densities such as bone and air (as found in
ribs overlying lung and facial bones containing air filled sinuses) cause
inhomogeneous dose distributions. Doses beyond air cavities may be higher than
expected even after density corrections and this limits the usefulness of electrons for
treating in these clinical situations. Electron beam edges do not diverge geometrically
due to lateral scatter, which is greater at low energies with the characteristic shape
shown in Figure 2.7. Wider margins must be added when choosing the beam width
for adequate treatment of tumours at depth. Where there is tumour infiltration of
skin, the skin sparing characteristics of electron beams below 16 MeV should be
removed by adding bolus material, which can also be used to provide tissue equivalent
material for an irregular contour such as the nose or ear. Beam sizes 40 mm should
be avoided because of inadequate depth of penetration and loss of beam flatness.
■ Conformal treatment
3D conformal radiotherapy (CFRT) links 3D CT visualisation of the tumour with
the capability of the linear accelerator to shape the beam both geometrically and
by altering the fluence of the beam (IMRT). This encloses the target volume as
closely as possible while reducing dose to adjacent normal tissues. The radiation
oncologist and dosimetrist agree the final PTV, which has been created using 3D
21
Figure 2.6 Percentage depth dose of varying electron energies for 10 10 cm applicator.
Figure 2.7 9 MeV electron

dose distribution.
growth algorithm software and departmental protocols, taking into consideration

OAR. Discussions between oncologist and dosimetrist include an understanding
of the tumour cell density pattern within the PTV, requirements for homogeneity
of dose distribution, dose constraints to adjacent OAR, avoidance of maximum or
minimum dose spots in 3D and review of a preliminary plan of likely beam
arrangements. Basic CFRT may consist of coplanar and static beams with MLC or
conformal blocks shaping the volume. For coplanar non-standard configuration of
beams, DVHs may aid selection of the best plan, but they do not indicate which
part of the organ is receiving a high or low dose; DVHs of the PTV, CTV and all
PRVs are needed to allow subsequent correlation with clinical outcome. Selection
of the final dose plan is made at the treatment-planning terminal by inspecting
22
Dose solutions
3D physical dose distributions and DVHs. Good communication is important
between radiographers, dosimetrist, physicist and clinician to avoid transfer errors
occurring between CT, treatment planning and treatment machine.
Conformal therapy may involve the use of mixed beams combining photons and
electrons for part of the treatment. Beams can be modified using bolus, wedges,
compensators, MLCs and shielding blocks. Optimisation of skin dose is achieved by
skin sparing using higher megavoltage energies, or by maximising skin dose with
tissue equivalent bolus. Higher beam energies (10–18 MV) are preferred for pelvic
treatments, to increase dose to the centre and reduce dose to skin and subcutaneous
tissues. Lower energies (6–8 MV) are used for breast and head and neck treatments
to avoid excess skin sparing and treat the relatively superficial target volume.
Treatment is delivered using beam shaping with MLCs which have multiple
leaves or shields which can block part of the radiation beam. Typical MLCs have
20–80 leaves arranged in opposing pairs which can be positioned under computer
control to create an irregular beam conforming to the target shape. As well as
eliminating the hazards of production and use of alloy blocks, this system allows
rapid adjustment of shaping to match the beam’s eye view of the PTV.
■ Complex treatment
IMRT
IMRT is created using MLC to define the beam intensity independently in
different regions of each incident beam, to produce the desired uniform distribution
of dose, or a deliberate non-uniform dose distribution, in the target volume. The
position of the leaves of the MLC can be varied in time with a fixed or moving
gantry. IMRT can be delivered using dose compensation, multiple static fields,
step-and-shoot, dynamic MLC or tomotherapy.
A sequence of static MLC fields can be used with the beam being switched off
between changes in position – the step-and-shoot technique. Alternatively, there
may be automatic sequencing of beam segments without stopping treatment –
dynamic MLC. Other methods include tomotherapy and other devices where
there is intensity modulated rotational delivery with a fan beam.
Forward planned or segmental IMRT provides simple tissue compensation with a
beam’s eye view of the PTV and sub-segments which are shaped with different MLC
to create a uniform dose in the PTV. Inverse planning requires specification of dose
prescription to GTV, PTV and PRV in terms of dose–volume constraints, fluence
optimisation and 3D dose planning. Very careful quality assurance must be developed
to assure accuracy of the beam. Verification of an IMRT plan requires either
measurement of the dose distribution in a phantom, or an independent monitor unit
calculation with portal dosimetry. Dose delivery is verified throughout the course of
treatment using radiographic film or adapted EPIDs or transit dosimetry. Accurate
patient positioning, target volume delineation and reduction of organ and patient
motion uncertainties, especially respiration, are critical for safe IMRT.
IMRT techniques modulate the intensity of the beam as well as its geometric
shape, delivering complex dose distributions, using forward or inverse treatment
planning. Plans can be produced with concave shapes, and critical structures at
sites such as head and neck (eye or spinal cord), prostate (rectum) and thyroid
(spinal cord) avoided. Late toxicity can be reduced significantly for tumour sites
23
such as prostate, pelvis, breast and head and neck. Dose escalation studies in
prostate cancer show an improvement in biochemical relapse-free survival using
IMRT with reduced late rectal toxicity. There is proof of sparing of salivary gland
function with IMRT for head and neck cancer with no loss of tumour control.
Late fibrotic changes in the breast can be reduced, and IMRT pelvic treatments,
more conformal to the tumour lymph node drainage, have reduced bone marrow
and acute bowel and bladder toxicity. However, integral dose may be greater with
some IMRT solutions, with increased risk of late malignancies. IMRT dose plans
with steep dose gradients may risk underdosage of tumour if margins are close and
organ motion still present.
It is difficult to produce evidence of benefit from new technology until it is widely
enough available to conduct RCTs, preferably on a multicentre basis. Often,
pioneering groups develop and test the technology from the physical viewpoint and
clinical implementation goes from pilot (phase 1 type) studies to routine use without
a rigorous evidence base of efficacy. A further problem of particular relevance for
radiotherapy is that unwanted late effects of treatment cannot be quantified for many
years, so that determination of therapeutic ratio – the true test of efficacy – is delayed.
IGRT
Organ motion during treatment makes it necessary to consider the fourth
dimension of time. IGRT refers to all techniques in which cross-sectional, X-ray or
ultrasound images obtained during treatment are used to check that the actual
treatment delivered matches that which has been planned. Ideally the moving
tumour outline is imaged during treatment using daily EPIs or real time cone
beam CT. Alternatively, EPIs can be taken on the first few days of treatment and
used to adjust treatment volumes where necessary.
Variations in overall positioning of the body during treatment can be monitored
using optical imaging devices, sometimes in association with markers attached to
the skin. To avoid respiratory motion, treatment may be delivered while the
patient holds his or her breath using the ABC device. This demands cooperation
from the patient and may be difficult in those with lung diseases. Treatment may
be gated to a specific phase of the respiratory cycle, usually expiration, using
optical devices or X-ray fluoroscopic measurements. CT scans obtained by imaging
devices on the linear accelerator can be compared with respiration-correlated spiral
CT planning images. Treatment delivery is triggered when the two images match.
These techniques may reduce the PTV and restrict the dose to normal lung but
are time consuming. Some evidence is accumulating that respiratory gating may be
of limited benefit overall since, although it reduces organ movement effects, it
prolongs treatment time. Composite multi-field conformal plans delivered in
multiple fractions may have a similar overall effect in a more cost effective way.
Further randomised trials are needed to assess 4D treatment delivery.
Adaptive radiotherapy (ART) involves regular changes (weekly or daily) to
treatment delivery for an individual patient based on analysis of images taken before
treatment or ideally during treatment. Linear accelerator mounted imaging devices
are essential for this technique. One integrated approach is to use rotational IMRT
or tomotherapy. The MLC is linked to a machine-mounted CT scanner with
images taken during therapy used to gate treatment to the correct body slice.
24
Dose-fractionation
Special techniques
■ Protons
Some radiation oncologists consider that many of the advantages of IMRT could
be better obtained by using the superior dose distributions of protons. There are
some indications where the benefits seem to be established firmly enough for this
to be the treatment of choice (some paediatric and skull-based tumours, and radio-
resistant tumours in difficult sites, such as vertebral chondrosarcoma). At present
the cost of installing proton facilities precludes widespread use, but guidelines for
referral to specialist centres have been drawn up.
■ Stereotactic radiotherapy
This technique is only available in a limited number of centres but has been used for
many years, mainly for the treatment of small brain tumours and arteriovenous
malformations. Accuracy of patient positioning to approximately 1 mm is maintained
using a stereotactic frame attached to the patient’s skull. Radiotherapy may be delivered
as a single or multiple fractions and may be considered as an alternative to surgery. The
gamma knife device uses multiple cobalt sources arranged around a half circle, which
irradiate a very conformal volume by blocking selected collimator openings with
different collimation helmets for different time intervals. Alternatively, a linear
accelerator with specialised collimators can be used to deliver multiple arc therapy.
This technique requires very careful quality assurance because of steep dose
gradients and problems of electron equilibration with very small beams. It also
requires close collaboration within a team of people with relevant expertise in
imaging, neuroanatomy, tumour management and physics. This approach may
also be beneficial for some small volume lung and liver tumours.
Dose-fractionation
Prescription of radiotherapy treatment is the responsibility of the radiation
oncologist and usually follows agreed guidelines, taking into consideration individual
patient factors, such as the expected risk–benefit ratio of treatment, comorbidities
and consideration of scheduling of other treatment modalities. Radiotherapy
regimens vary internationally. Fractions of 2 Gy or less delivered 5 days a week are
the standard of care in much of North America and Europe. Regimens given in
this book are safe evidence-based schedules but national and international
protocols or trials should be used as appropriate.
Alternative fractionation schedules using fewer larger fractions in a shorter
overall time (hypofractionation) have been developed, especially in the UK and
Canada, driven initially by resource constraints, but now supported by extensive
published clinical data, e.g. for breast and prostate cancer as well as for palliative
treatments. Accelerated fractionation gives the same overall dose in a shorter time,
often using smaller fraction sizes to reduce toxicity, and has been used successfully
in head and neck cancer trials. Hyperfractionation regimens deliver treatment
twice or three times a day using smaller fraction sizes, thereby increasing the total
dose and remaining within tolerance for late toxicity (see Table 3.2, p. 42).
25
The linear quadratic (LQ) model of radiation-induced cell killing is currently the
most useful for comparing different fractionation schedules (see Chapter 3), taking
into account the effect of dose per fraction and repopulation on tumour and normal
tissue late effects. Clinical outcome depends on the total dose, dose per fraction,
overall treatment time, volume of tumour and normal tissues irradiated, dose
specification points and quality control procedures. If treatment has to be stopped
unexpectedly for operational or clinical reasons causing an unscheduled gap in
treatment, the dose-fractionation schedule may need to be altered (see Chapter 3).
Dose specification
All dose distributions are inhomogeneous and so the dose throughout the PTV
varies. The biological effect of a given dose is difficult to predict because of
variations in cell density at the centre and periphery of the CTV, heterogeneity of
tumour cell populations and inadequate knowledge of cellular radio-sensitivity.
Nevertheless, one must attempt to specify a dose which is representative of the
absorbed dose in the target volume as a whole, to assess effectiveness of treatment.
To facilitate understanding and exchange of precise and accurate radiotherapy
treatment data, it is important that all centres report their results using the same
volume concepts and prescribing definitions.
The ICRU reports 50 and 62 recommend that the radiation dose should be
reported, and hence is also best prescribed, at or near the centre of the PTV, and
when possible at the intersection of the beam axes (ICRU reference dose). This
ICRU reference point for prescription is selected because it is clinically relevant,
usually situated where there is maximum tumour cell density, easy to define, often
lies on the central axis of the beam where dose can be accurately determined, and is
not in a region of steep dose gradient. It is essential that this dose specification point
is accompanied by a statement of the homogeneity of the irradiation as defined by at
least the maximum and minimum doses to the PTV. The maximum target dose is
the highest dose in the target volume which is clinically significant (to a volume
greater than 15 mm3 unless in a critical tissue where special considerations apply).
The minimum target dose is an important parameter because it correlates with the
probability of tumour control. Additional information, such as average dose and its
standard deviation, DVHs and an accurate description of dose to OAR, is also very
important (Fig. 2.8).
Superficial treatment machines, with energies ranging from 50 kV to 150 kV and
appropriate filtration which defines percentage depth dose characteristics, are used
to treat superficial skin tumours. The dose prescription point is at Dmax, the
maximum dose, which is at the skin surface. An appropriate energy is selected from
tables for different beam sizes for a given FSD to encompass the target volume
both on the skin and at depth with a 90 per cent isodose.
When a single megavoltage beam is used, dose is prescribed to the ICRU point at
the centre of the target volume rather than to Dmax. For example, for bone metastases,
the prescription point may be at the centre of the vertebra e.g. 40–50 mm depth
for a thoracic vertebra. Alternatively, for palliative treatments, the ICRU point may
be chosen at the dose-limiting structure, such as the spinal cord, and a dose
prescribed to maximum tolerance.
26
(a) (b
)
(c) (d)
Figure 2.8 Dose distribution to treat carcinoma of the prostate showing (a) axial view with isodoses, ICRU point (100 per cent): maximum 101 per
cent; minimum 95 per cent. (b) DVH showing PTV, and doses to bladder, rectum, left (LFH) and right (RFH) femoral head. (c) Coronal view with
dose colourwash. (d) Sagittal view with dose colourwash.
27
For co-axial opposing lateral or AP beams, the dose is specified at the midplane
dose (MPD) on the central axis of the beam, as recommended by ICRU. In
subsequent chapters of this book, these ICRU dose specification conventions have
been followed for all dose distributions and prescriptions.
Verification
Verification is needed of the geometrical set-up of the treatment and the dose being
delivered. Electronic portal imaging is available on most linear accelerators and
images can be compared with DRRs obtained from the treatment planning system.
Image evaluation software tools can be used to match predefined bony landmarks on
both images noting displacements of beam borders, to match the edges of beams
measuring the change in position of bony structures or to match fiducial markers.
Errors may be systematic or random. Repeating the portal image daily for the
first few days will identify any systematic errors, which can then be corrected.
Random errors in set-up may be reduced by better patient immobilisation or staff
education, or the CTV-PTV margin may have to be increased if correction is
impossible.
Image matching of EPIs with DRRs is usually carried out by the treating
radiographers after appropriate training. Verification protocols, both online and
off-line, will define the level of action for any deviation. EPIs or portal films should
be signed by the clinician, verifying and recording any actions taken.
Verification of the dose delivered to the patient is essential using semiconductor
silicon diodes with instant readout, or thermo-luminescent dosimetry (TLD) with
delayed readout, and is performed on the first day of each patient’s treatment.
Transit dosimetry uses a transmission portal image to measure the dose delivered
to the patient, which can be compared with the planned dose distribution. Equally
important are the mechanical checks of MLC that are required to ensure safe
delivery of treatments. These must include examination of the stability of leaf speed,
accuracy of leaf position and transmission through and between leaves.
Patient care during treatment

Response to treatment, both in terms of tumour and acute side effects, should be
monitored regularly. Weekly clinics are held, often led by a therapy radiographer or
radiotherapy clinical nurse specialist, to check for early acute side effects, and to
answer patients’ questions and give them and their carers psychological and
emotional support. Information sheets listing common and uncommon side effects
for each tumour site are discussed with individual patients when they give informed
consent. Protocols for scoring acute side effects, such as NCI-CTC-3 or RTOG,
should be used to document acute reactions and guide appropriate investigations.
The radiation oncologist states on the radiotherapy prescription sheet any specific
imaging or blood tests that are required during treatment and is responsible for
supervising the review clinics and seeing the patient at appropriate intervals. Patients
receiving concurrent radiochemotherapy may experience increased side effects and
require special interventions. Specific instructions for dealing with different acute
reactions are dealt with in each individual tumour site chapter.
28
Quality assurance
Quality assurance
A comprehensive quality assurance programme is needed to ensure that the best possible
care is delivered to the patient by defining and documenting all procedures involved in
radiotherapy treatments. A quality assurance policy must be formulated by a radiotherapy
manager with overall responsibility for implementation, although large parts of this
responsibility may be delegated to other appropriately trained and qualified members of the
radiotherapy team. Internal quality assurance systems are supplemented by external
accreditation visits. Individual treatment techniques may be checked before trials begin, by
requiring participants to submit sample treatment plans and descriptions of methods of
beam matching, techniques used for tissue compensation and beam data for verification
of output and flatness. Phantom measurements are routinely used to give a composite
check of all factors, including dose homogeneity and beam matching. Any internal or
external system must be periodically reviewed with documentation of all activity according
to a defined protocol. Reference values must be specified with tolerances and action to be
taken if these are exceeded. A departmental procedure for investigation of deviations or
errors must be in place for every part of the system.
■ Practical considerations
Routine checks of the following must be included in the quality assurance protocol:
■ machine checks
■ dosimetry protocols
■ planning checks
■ patient documentation.
Machine checks
During installation and acceptance of new equipment, calibration data are obtained
to provide reference against which subsequent checks are made. Inter-comparisons
between different institutions or with national or international standards are useful
for detecting systematic errors. Quality control should then ensure that a unit
performs according to its specification and is safe for both patients and staff.
It should guarantee accuracy of dose delivered, prevent major errors, minimise
downtime for machines and encourage preventative machine maintenance. There
should be a specific quality control protocol for each unit, which outlines the test
to be performed, the methods to be used to ensure consistency in the performance
of each unit, parameters to be tested, frequency of measurement, staff responsibilities,
reference values, tolerances, action to be taken in case of deviation and rules for
documentation. Daily, weekly and extended testing of dosimetry beam alignment
and safety checks are necessary. Regular checks include tests of optical, mechanical
and computer hardware and software systems. Action levels are defined where
correction is needed before treatment can proceed. Similar checks must be carried
out for all imaging equipment and treatment planning systems. The results of daily
checks must be recorded in the control room of the treatment units and
radiographers must also record any problems in machine functioning. All other
checks, actions and maintenance work are recorded in a separate log book. Good
29
cooperation is needed between all staff groups. A physicist who coordinates all
quality control activity checks that tests are up to standard and reports any major
deviations to the clinician.
Dosimetry protocols
These include dose monitor calibration checks, checks of beam quality and
symmetry and evaluation of beam flatness. In vivo dosimetry systems such as TLDs
and silicon diodes must also be regularly checked and calibrated.
Planning checks
Treatment prescriptions are now mostly electronic, and recording of treatment delivery
parameters is automatic by computer systems attached to treatment machines. Reports of
activity obtained from these systems can be used for audit, and central collection of these
output data may give very useful information about patterns of radiotherapy delivery.
Individual weekly review of patients’ treatment records should verify that treatment is being
delivered as planned.
Patient documentation
Electronic systems are being used increasingly. Radiation treatment records are usually
kept separately from other hospital documents to ensure reliable rapid access. Records
should identify the patient, give clinical history and examination findings, histological
diagnosis, staging of the tumour and proposed treatment plan. There should be
written treatment policies for specific tumour sites and data should be recorded to
enable subsequent evaluation of the outcome of treatment. Written consent for
treatment is required. At the end of treatment, a summary detailing actual treatment
parameters must be prepared and appropriate continuing care of the patient assured.
Staffing
A quality programme as described above is essential for the safe delivery of treatment.
It can only be achieved if each member of the team understands clearly the
boundaries of responsibility and if there is excellent coordination of all quality
control activity by a highly qualified physicist acting with the person responsible for
overall management of the radiotherapy department. Careful training of all staff
members must therefore be an integral part of any effective quality assurance system.
Information sources
Bel A, Bartelink H, Vijbrief RE et al. (1994) Transfer errors of planning CT to simulator: a possible
source of set up inaccuracies? Radiother Oncol 31: 176–80.
Bel A, van Herk M, Bartelink H et al. (1993) A verification procedure to improve patient set-up
accuracy using portal images. Radiother Oncol 29: 253–60.
BIR Working Party (2003) Geometric Uncertainties in Radiotherapy: Defining the Planning Target
Volume. British Institute of Radiology, London.
Chao KSC, Majhail N, Huang C-J et al. (2001) Intensity-modulated radiation therapy reduces late
salivary toxicity without compromising tumor control in patients with oropharyngeal carcinoma:
a comparison with conventional techniques. Radiother Oncol 61: 275–80.
30
Information sources
Development and Implementation of Conformal Radiotherapy in the United Kingdom (2002) Royal
College of Radiologists, London.
Dobbs HJ, Parker RP, Hodson NJ et al. (1983) The use of CT in radiotherapy treatment planning.
Radiother Oncol 1: 133–41.
Gregoire V, Coche E, Cosnard G et al. (2000) Selection and delineation of lymph node target
volumes in head and neck conformal therapy. Proposal for standardising terminology and
procedure based on the surgical experience. Radiother Oncol 56: 135–50.
Guidelines for the Management of Unscheduled Interruption or Prolongation of a Radical Course of
Radiotherapy, 2nd edn. (2002) Royal College of Radiologists, London.
Holland R, Veling S, Mravunac M et al. (1985) Histologic multifocality of Tis, T1–2 breast
carcinomas. Implications for clinical trials of breast-conserving surgery. Cancer 56: 979–90.
Hurkmans CW, Remeijer P, Lebesque JV et al. (2001) Set-up verification using portal imaging:
review of current clinical practice. Radiother Oncol 58: 105–20.
International Commission on Radiation Units and Measurements (1993) Prescribing, Recording and
Reporting Photon Beam Therapy ICRU: Report 50. ICRU, Bethesda, Maryland, USA.
Reporting Photon Beam Therapy (supplement to ICRU Report 50): ICRU Report 62. ICRU,
Bethesda, Maryland, USA.
Reporting Electron Beam Therapy: ICRU Report 71. ICRU, Bethesda, Maryland, USA.
LENT-SOMA Tables (1995). Radiother Oncol 35: 17–60.
McKenzie AL, van Herk M, Mijnheer B (2000) The width of margins in radiotherapy treatment plans.
Phys Med Biol 45: 3331–42.
McKenzie AL, van Herk M, Mijnheer B (2002) Margins for geometric uncertainty around organs at
risk in radiotherapy. Radiother Oncol 63: 299–307.
National Cancer Guidance (2008) Proton Referral-Treatment Abroad. Royal College of Radiologists,
London.
Radiotherapy Dose-Fractionation. (2006) Royal College of Radiologists, London.
Seddon B, Bidmead M, Wilson J et al. (2000) Target volume definition in conformal radiotherapy for
prostate cancer: quality assurance in the MRC RT-01 trial. Radiother Oncol 56: 73–83.
Suter B, Shoulders S, Maclean M et al. (2000) Machine verification radiographs: an opportunity for
role extension. Radiography 6: 245–51.
Taylor A, Rockall AG, Reznek RH et al. (2005) Mapping pelvic lymph nodes: guidelines for
delineation in intensity-modulated radiotherapy. Int J Rad Oncol Biol Phys 63: 1604–12.
van Herk M, Remeijer P, Rasch C et al. (2000) The probability of correct target dosage:
dose-population histograms for deriving treatment margins in radiotherapy. Int J Rad Oncol
Biol Phys 47: 1121–35.
Webb S (2005) Contemporary IMRT: Developing Physics and Clinical Implementation. Institute of
Physics Publishing, London.
Wong JW, Sharpe MB, Jaffray DA et al. (1999) The use of active breathing control (ABC) to reduce
margin for breathing motion. Int J Radiat Oncol Biol Phys 44: 911–19.
World Health Organization (1988) Quality Assurance in Radiotherapy. WHO, Geneva.
Yan D, Ziaja E, Jaffray D et al. (1998) The use of adaptive radiation therapy to reduce setup error: a
prospective clinical study. Int J Radiat Oncol Biol Phy 41: 715–20.
Zelefsky MJ, Fuks Z, Hunt M et al. (2002) High-dose intensity modulated radiation therapy for
prostate cancer: early toxicity and biochemical outcome in 772 patients. Int J Radiat Oncol
Biol Phys 53: B1111–16.
31
3 Radiobiology and
treatment planning
Radiation and cell kill

Ionising radiation causes wide-ranging molecular damage throughout cells by the
production of ionised atoms, which cause breakage of chemical bonds, production
of free radicals and damage to DNA. Most clinically significant effects of
radiotherapy are due to irreparable DNA lesions which result in sterilisation – a loss
of proliferative cells’ ability for sustained cell division. In tumours, loss of
proliferative ability by all the cells of the tumour is a necessary condition for tumour
cure. Partial sterilisation of the tumour cell population results in tumour stasis or
regression, giving a clinical remission, followed by regrowth of the tumour from
those cells which have retained their proliferative ability. In self-renewing normal
tissues, sterilisation of proliferative cells leaves the tissues unable to provide
replacements for cells that are ordinarily being lost at a constant rate from the
tissue, and initiates a rundown of the mature cells of the tissue. Proliferative
sterilisation is often referred to as cell kill, with those cells that retain long-term
proliferative ability being described as survivors.
Cell survival curves

Cell culture techniques have been very important in allowing the proliferative
sterilisation of cells to be investigated quantitatively. For an irradiated cell population,
the proportion of cells that still retain the ability to proliferate (relative to an
unirradiated control population) is called the surviving fraction, and a plot of log
surviving fraction against single radiation dose gives a survival curve for the cells
concerned. Typically, survival curves are continuously bending, with a slope that
steepens as the dose increases. Mathematically a continuously bending curve is
most simply described by a linear quadratic (LQ) equation of the form:
SF Exp (αd βd2) (3.1)
where SF is the surviving fraction, d is the given single dose and α and β are
parameters characteristic of the cells concerned. The ratio α/β gives the relative
importance of the linear dose term and the quadratic dose term for those cells, and
controls the shape of the survival curve (Fig. 3.1). When α/β is large, the linear
term predominates, so a plot of log (SF) against d is relatively straight, while if α/β
is small, the quadratic term is more important, giving a plot with greater curvature.
For cells whose survival curves have a lower α/β ratio, doubling the dose leads to
more than doubling of the effect on log (SF). Such cells will be particularly sensitive
to changes in fraction size when radiation is given as a fractionated schedule.
32
Effects on tissues
0 Dose (Gy)
0
Small α/β
0.2 Large α/β

Log (SF)
0.4
0.6
0.8
Figure 3.1 Two contrasting survival curves for irradiated cells, with log of the surviving
fraction (SF) plotted against single radiation dose. The more steeply curving survival curve
has the lower α/β ratio when fitted to the linear-quadratic equation.
Acute and late effects on tissues

The dose–response relationship for normal tissue injury depends on the survival
curves of the tissue stem cells. The timing of expression of injury depends on the
rate of turnover of mature cells in the tissue. Epithelial and haemopoietic tissues
have rapid turnover, and so manifest acute effects (with a timescale of days or
weeks). Late effects (with a timescale of months or years) occur in tissues and
organs with slow turnover (e.g. endothelial and neuroglial tissues, and parenchymal
tissues of lung, liver and kidney). The risk of major late effects is usually dose-
limiting in radiotherapy. Stem cells of late-responding tissues have been found to
have much more curved survival curves (low α/β ratio) than cells of acute-
responding tissues (higher α/β ratio, less curved survival curve). Radiobiologically,
this difference may be related to slow cell turnover in late-responding tissues,
allowing many stem cells to remain in resting states where they are highly proficient
in the repair of damage caused by small doses. Consequently, late-responding
tissues are particularly sensitive to changes in fraction size, larger fractions being
more damaging to these tissues, but small fractions being well tolerated. Acute and
late-responding tissues are also affected differentially by changes in the overall
treatment time. Because surviving stem cells of acute-responding tissues initiate
repopulation during the course of radiotherapy, the time over which the radiation
is distributed makes a difference to the final level of damage. Late-responding
tissues do not experience repopulation during the course of radiotherapy and are
relatively unaffected by the overall treatment time. Expression of late effects will also
be influenced by the organisation of functional subunits within the organ. Serially
arranged subunits, such as are found in the bowel, mean that damage to some units
will lead to change in the whole organ, whereas in those organs where subunits are
arranged in parallel, such as the kidney, each subunit acts independently and
continuing function is possible if only some of the subunits are damaged.
33
RADIOBIOLOGY
The linear quadratic model for tissues

Differences between acute- and late-responding tissues and typical tumours are
important when we consider the biological consequences of replacing one treatment
schedule with a different one. From experience, radiation oncologists developed
‘rules of thumb’ for altering a treatment schedule, e.g. a substantial dose reduction
might be necessary if fraction size was increased, increased treatment time improved
skin tolerance, etc. It is now recognised that tissues react differently to changes in
schedule structure and there is better understanding of how changes in scheduling
will affect different tissues. The application of the LQ model is based on the idea
that the severity of tissue damage is inversely proportional to stem cell survival, and
that stem cell survival can be calculated by mathematical development of the LQ
survival curve equation to allow it to be applied to fractionated treatments. In
particular, two different schedules will have equal biological effects on a tissue (i.e.
they will be ‘isoeffective’ for the tissue) if each schedule produces the same level of
stem cell survival.
We shall not attempt to derive the mathematics of the LQ model in detail, but
will state some results which are useful in practice. Consider two treatment
schedules, namely total dose D1 or D2 given as fractions sized d1 or d2. We shall
assume for now that both schedules are given in the same overall time period. For
a tissue whose cells have a survival curve described by the LQ parameters α and β,
it can be shown that the two schedules are isoeffective, for that tissue, when:
D1 (α/β d1) D2 (α/β d2) (3.2)
Notice that the survival curve parameters appear as a ratio in this expression, which
means that only this single number for a tissue need be known in order to apply
the equation. In fact, α/β estimates have been made and tabulated for many
tissues. Late-responding tissues are usually found to have low values of α/β (about
3 Gy), while acute-responding tissues have higher values (about 10 Gy). (This
means that the cells of late-responding tissues have more steeply curving survival
curves.) Tumours are more variable; many are like acute-responding tissues, with
α/β values of 10 Gy or more, but recent estimates for breast tumours suggest
lower values. In practice, Equation 3.2 is often used to compare an unfamiliar
schedule with a standard schedule which would have the same effect. Most
usefully, the unfamiliar schedule can be assessed by asking what total dose, given
as 2 Gy fractions, would have the same effect (on that tissue) as the unfamiliar
schedule. This is helpful for determining whether the unfamiliar schedule is ‘hot’
or ‘cold’. For example, consider a schedule which consists of 10 twice-weekly
fractions of 4 Gy to a total dose of 40 Gy in 5 weeks. In Equation 3.2, let d1 be
4 Gy and D1 be 40 Gy. If we now set d2 2 Gy, and calculate D2 (for a particular
choice of tissue α/β), we shall get the total dose given as 2 Gy fractions which
would have the same effect on that tissue as the schedule D1, d1. Notice, however,
that the calculation depends on the assumed value of α/β. For this example, we
shall repeat the calculation for acute-responding and late-responding tissues. For
acute-responding tissues (α/β 10 Gy) we find that D2 47 Gy, while for late-
responding tissues (α/β 3 Gy), we find that D2 56 Gy. Therefore, the new
schedule is expected to be ‘hotter’ in terms of its effects on late-responding than
34
Volume effects
on acute-responding tissues, but it is no ‘hotter’ than a conventional radical
regimen (e.g. 60 Gy in 2 Gy fractions).
It should be noted that the simple LQ model does not allow for differences in
total time between the schedules, which are therefore presumed to be given in the
same overall time (5 weeks in this case), despite their different fractionation
patterns. This restriction is not so important for late-responding tissues (for which
the total time is a minor variable), but the results for acute-responding tissues (for
which the time factor can be significant) need to be interpreted cautiously with this
limitation of the model borne in mind.
In recent years, the LQ model has been developed extensively and applied to
more complex schedules, including brachytherapy. The standard schedule with
which an unfamiliar schedule is to be compared is not necessarily one using 2 Gy
fractions. A rather abstract standard schedule, more appealing to mathematicians
than to clinicians, is a hypothetical regimen in which a very large number of small
fractions are given (mathematically, an infinite number of zero-sized fractions are
given, but to a finite total dose). The equivalent total dose calculated for such a
schedule is called the biological effective dose (BED). It is usually quite a large
dose (e.g. 100 Gy for a typical radical regimen) because it represents the limit of
tissue sparing by fractionation, i.e. the total dose that could be given if the
individual fraction size were vanishingly small. Linear quadratic calculations
performed using BED are mathematically equivalent to those performed with the
standard regimen taken to be one using 2 Gy fractions, although the latter has the
advantage of clinical familiarity. An important feature of the LQ model in its
various forms is the recognition that the cells of different tissues differ in their
survival curve shape and therefore respond differently to changes in fraction size.
Since a target volume may contain several tissue types as well as the tumour, a
change of fractionation regimen will affect these components differently. There is
therefore no such thing as a regimen which is ‘generally equivalent’ to some other
regimen – the regimens can only be matched (by choice of total dose) for
equivalent effects on each specific tissue.
Volume effects
Together with the total dose and fractionation schedule, target volume is a major
variable in radiotherapy. For a given fractionation regimen, higher doses can usually
be given when volumes at the same site are small rather than large. Normal tissues
are required to perform orchestrated functions, which can be impaired in various
ways by irradiation. Most normal tissues also cannot regenerate from a single
surviving cell. However, tissue recovery may be assisted by immigration of
unirradiated neighbouring cells, particularly if the treatment volume is small.
Volume is also an important determinant of normal tissue response to a given dose,
first because larger volumes provide less opportunity for tissues to draw on their
‘functional reserve’ and second because larger irradiated volumes make it more
likely that a critical volume element will exceed some upper dose limit. These factors
differ according to tissue structure, and vary from one treatment to another.
In general, the normal tissue complication probability (NTCP) increases with
dose (for a given fractionation regimen) and with the irradiated volume. It is
important to know, at least approximately, how changes in irradiated volume at a
35
RADIOBIOLOGY
particular site will affect the tolerance dose which can safely be given. The
‘tolerance dose’ may arbitrarily be defined as that dose which gives no more than
5 per cent incidence of significant side effects, based on clinical experience. A body
of data has been amassed which provides some simple ‘rules of thumb’ concerning
the trade-off between treatment volume and tolerance dose, but these need to be
used very cautiously. Tolerance is affected not only by volume, but also by
radiation sensitivity and fraction size, and tolerance to the various new schedules
in use must be carefully confirmed by clinical studies. In some cases, radiation
injury may result from an excessively high dose to a small tissue element within the
treatment volume. The possibility of ensuring better homogeneity of dose
distribution with IMRT may help to ameliorate this problem.
Radiation dose and tumour cure probability

In radical radiotherapy, the objective is complete sterilisation of any tumour cells
present, which must be achieved without incurring an unacceptably high risk of
serious injury to normal tissues.
Radiation kills cells randomly, which means that each tumour cell has the same
probability of surviving irradiation, that probability depending on the given dose.
Suppose that SF2 is the probability of any cell surviving a single dose of 2 Gy, the
most commonly used fraction size. For example, SF2 might be 0.5 for a typical
carcinoma. After the first 2 Gy fraction, 50 per cent of the cells survive; after the
second dose, 50 per cent of those survivors still survive (i.e. 25 per cent of the
original population); after the third dose, 50 per cent of those survivors still survive
(i.e. 12.5 per cent of the original population), and so on. Generally, after F
fractions, the final survival probability will be (SF2)F. Therefore, for a conventional
treatment regimen consisting of 30 treatments of 2 Gy, the final survival
probability (in this example) would be 0.530 or 9 10–10.
These relationships have some interesting clinical implications. First, there is no
dose which gives zero probability of cell survival – even after a large dose there will
be some probability, possibly very small, of survival of each cell. However, a visible
tumour will contain a large number of cells, so even if each cell individually only
has a small chance of surviving, there may be a good chance that at least one cell
will survive and could regenerate the tumour. We therefore need to know the
relationship between the given dose and the probability that a whole cell
population will be sterilised with not a single cell surviving – the basic requirement
for tumour cure. This can be computed using the theory of Poisson statistics. We
can then calculate the number of treatments necessary to achieve some value of
cure probability, such as 90 per cent (remember that 100 per cent cure probability
cannot be achieved by any finite dose).
We can repeat the calculation for tumours of different sizes (different cell
population numbers) for our example of SF2 0.5, and also for other values of
SF2 representing more sensitive or less sensitive tumour types. Figure 3.2 shows
these relationships. Although we have used a rather simple model for these
calculations, we can observe some features which also turn up in more complex
and realistic models.
First, note that the relationship between tumour cell number and number of
treatments is logarithmic, i.e. a large change in cell number corresponds to a rather
36
Radiation dose and tumour cure probability
60
Treatment fractions (2 Gy) for 90% cure
50 SF2 0.6
(resistant)
40
SF2 0.5
(intermediate)
30
SF2 0.4
(sensitive)
7 8 9 10 11 12
Log (cell numb
er)
Figure 3.2 Shows the calculated number of 2 Gy fractions to achieve 90 per cent cure
probability, as a function of tumour cell population number, for differing values of cellular
intrinsic radiosensitivity (expressed as the surviving fraction following a single 2 Gy
treatment, SF2). Moderate variation of SF2, as seen between cells of different tumour
types, leads to large differences in the number of 2 Gy treatment fractions required for
90 per cent cure probability. In some cases, the predicted number of fractions required is
much larger than could be safely given.
modest change in the number of treatments and hence the total dose required. For
example, the number of treatments required for a tumour of 104 cells is just half
the number required for 108 cells. It is because of this logarithmic relationship that
quite high total doses have to be given to regions containing only microscopic
spread (in fact often about half the dose given to the bulk tumour). Within regions
of microscopic spread it is likely that the tumour cell density will gradually
decrease, on average, with increasing distance from the visible edge. This suggests
that the radiotherapy dose should similarly decrease with distance, roughly in
proportion to log cell density. Although the cell density distribution will not be
known in detail, it is possible that a tapering dose distribution, such as occurs in a
conventional plan with a peripheral dose gradient, or achieved with IMRT and
deliberate dose modulation within the volume, could be advantageous. A second
feature of the model is that a small change in SF2 has quite a large effect on the
required total dose (compare the three curves shown in Fig. 3.2). Small variations
in the intrinsic radiosensitivity of tumour cells could result in a tumour being easily
curable, or completely incurable, by a radiotherapy regimen.
Another feature of the dose–cure relationship is the steepness of the increase in
tumour cure probability with total dose, illustrated in Figure 3.3. In this figure,
the solid curves show the dose–cure relationships for a series of tumours with
slightly different parameters, and they can be seen to be increasing steeply with
total dose in all cases. This implies that relatively small differences in total dose
37
RADIOBIOLOGY
Figure 3.3 The solid lines in this figure show the expected relationship between total
radiation dose and cure probability for individual tumours with differing radiosensitivity and
cell number. The curves are sigmoid in shape, located at different positions on the dose
axis, and each is quite steep. The broken line shows the much shallower dose-response
usually seen when proportion cured is plotted against dose for groups of tumours, as in
clinical trial studies. The shallower response is thought to result from the heterogeneity of
the tumours in each dose group.
could make a significant difference to cure probability for each tumour. However,
these steep relationships are not seen in the dose–cure relationships for treatment
of groups of tumours with differing parameters, such as occur in clinical studies
with patient groups. The broken line in Fig. 3.3 shows the much shallower
gradient which is typically observed in such studies. The shallow gradient of the
curve for tumour groups is the resultant of a series of steep curves for individual
tumours with differing radiosensitivities. This has some significance when we
consider the importance of moderate changes in total dose when treating
individual patients. The importance of a dose increment in treating an individual
tumour depends on how close the treatment regimen has come to achieving cure.
For a large or resistant tumour (such as glioblastoma) with cure probability close
to zero, a modest dose increment will make little difference. Conversely, for a small
or highly sensitive tumour (such as seminoma) with cure probability close to unity,
a small dose increment again makes little difference. However, if the treatment
regimen achieves a cure probability close to 50 per cent (such as head and neck
tumours), it is in this situation that the dose–response curve is as steep as that
calculated from the Poisson model. This means that for any individual patient
there is some probability, usually unknown, that a large change in tumour control
probability will result from a small change in delivered dose. Even where the
average dose–response curve for patient groups is known to be shallow, a minority
of patients may benefit substantially from small changes in the given dose. It is
these patients for whom the choice of treatment plan may be especially critical.
38
Tumour radiobiology
Tumour radiobiology
It is believed that the main factors controlling tumour response to fractionated
radiotherapy are the so-called ‘five Rs’ of radiobiology (Table 3.1). Currently, the
most important of these factors are thought to be the intrinsic radiosensitivity of
cells and the kinetics of repopulation of surviving cells.
Intrinsic radiosensitivity varies between different tumour cell lines in culture,
with cell lines derived from clinically resistant tumour types having a statistical
tendency towards higher SF2 values. SF2 measurements on tumours are technically
Table 3.1 Radiobiological factors controlling response to fractionated radiotherapy (the

‘five Rs’) and their clinical relevance
Radiobiological Mechanism of effect on Clinical relevance

factor response
Radiosensitivity Intrinsic radiosensitivity differs Can account for variable

between cells of tumours and response of tumours
normal tissue types, and strongly Curative dose is proportional to the
determines final surviving fraction log of cell number (so subclinical
disease needs smaller dose)
Repair Cells differ in their capacity to Repair is maximal in late-responding
repair DNA damage, particularly tissues given small fractions.
after small doses of radiation Hyperfractionation may be advantageous
Repair is usually more effective Treatments need to be well separated
in non-proliferating cells. The in order to avoid compromising repair
repair process takes at least
6 h to complete
Repopulation Surviving cells in many tumours Shortened treatment times (accelerated
and in acute-responding (but therapy) may be advantageous for some
not in late-responding) normal tumours. Acute (but not late) effects will
tissues proliferate more rapidly be increased. Gaps should be avoided
once treatment is in progress
Reoxygenation Hypoxic cells, which occur Very short treatment times could lead to
especially in tumours, are relatively resistance due to persistence of hypoxic
resistant to radiation. Hypoxic cells. Treatment of anaemia is essential
surviving cells reoxygenate, to optimise tumour response
becoming radiosensitive, as
treatment proceeds
Redistribution Cells in certain phases of the Closely spaced treatment fractions could
proliferative cycle (e.g. late S) are lead to resistance due to persistence
relatively resistant and survive of cells in less sensitive phases
preferentially. With time between
fractions, cells redistribute
themselves over all phases
of the cycle
39
RADIOBIOLOGY
difficult and time-consuming, and techniques are not yet clinically available to help
predict radiosensitivity of tumour and normal tissues in individual patients.
Although tumours are very diverse, the radiobiological properties of most
tumours are similar to those of acute-responding tissues, i.e. a high α/β ratio,
moderate sensitivity to changes in fraction size, and some dependence on total
treatment time. The role of treatment time has been controversial, but it is now
widely believed that many tumours repopulate rapidly during the latter part of a
course of radiotherapy, and that any factor which prolongs time in treatment could
lead to significantly reduced tumour cure probability. Attempts are currently being
made to identify which tumours are most capable of rapid repopulation by
measuring kinetic parameters. Measurements of tumour kinetics to detect those
most capable of rapid repopulation are possible using thymidine analogues, but
these tests are not yet available to use as predictive tools for individual patients.
Avoidance of gaps during treatment

The occurrence of rapid repopulation in irradiated tumours, sometimes with
doubling times as short as 3–4 days, has important implications for interruptions
of treatment. Unscheduled gaps occur not infrequently in radiotherapy schedules
because of machine breakdown or patient intercurrent illness or non-attendance.
(We shall exclude from consideration patients whose treatment is deliberately
stopped because of unusually severe acute reactions.) Gaps are important because
they may lead to prolongation of the total treatment time, allowing opportunities
for rapid repopulation for surviving tumour cells towards the end of the schedule.
Although prolongation will often spare acute normal tissue reactions, the risk of
late effects is not reduced. It has been shown for squamous carcinomas of the head
and neck that reductions in cure probability of 1–2 per cent may result from each
day of prolongation. If gaps occur, the best management strategy is ‘post-gap
acceleration’, i.e. the use of twice daily treatments (separated by more than 6 h) or
weekend treatments to enable ‘catching up’ so that treatment is nevertheless
completed within the originally intended period. This does not require any
changes to fraction size or total dose. However, this approach is not always feasible
in practice. Now that the deleterious effect of prolongation is recognised,
avoidance of gaps is an important consideration for all radiotherapy departments.
Treatment scheduling
Conventionally, radiotherapy schedules have consisted of multiple fractions of 2 Gy
delivered for 5 days a week over several weeks. The total dose is usually limited by
anticipated risk of injury to late-responding tissues, although there are situations
where acute responses (e.g. mucosal reactions) are the main concern. Treatment
schedules with this structure probably take advantage of differences between the
survival curves of cells in late-responding tissues (low α/β ratio and fraction size
sensitivity) and the survival curves of those in typical tumours (with higher α/β
ratios). This means that late-responding tissues are spared to a greater extent than
most tumours by the use of small fractions, giving a favourable therapeutic ratio.
Late responses are not strongly influenced by overall treatment time, and it would
be desirable to make the latter as short as possible in order to minimise the
40
Treatment plan and ‘double trouble’
opportunities for tumour repopulation. However, this must be balanced by the
need to allow time for reoxygenation of hypoxic cells during therapy, and there also
may be an adverse effect of reduced treatment time on acute-responding tissues
(which also have reduced opportunities for repopulation).
There is now considerable experience with different schedules of treatment (for
details, see Table 3.2). All these changes in scheduling may bring important gains
in tumour control. However, tumours are known to be extremely heterogeneous
with regard to cell survival parameters and growth kinetics, as well as other
properties. It is unlikely that any one schedule is ideal for treatment of all tumours,
even those of a single pathological type, and it would be highly desirable to select
treatment schedules for individual patients on the basis of the radiobiological and
kinetic parameters for each tumour. Predictive tests are not yet sufficiently reliable
to be used in this way, but individualised scheduling based on biological assay is a
likely development for the future.
Treatment plan, schedules and

‘double trouble’
The design of physical treatment plans usually proceeds without regard to
treatment schedule. However, there may be interplay between dose distributions
and treatment schedules, which gives an altered biological effect sometimes
described as ‘double trouble’. Consider a treatment plan in which the spread of
dose within the target volume is 10 per cent, so if the intended treatment is
30 fractions of 2 Gy to give 60 Gy in total, the spread of dose will be from 57 Gy to
63 Gy. However, the low-point and high-point doses will not have been delivered
as 2 Gy fractions (the dose variation of 10 per cent affects each dose fraction, so the
fraction size will range from 1.9 Gy to 2.1 Gy). A treatment schedule of
30 2.1 Gy 63 Gy is what is ‘seen’ by cells located close to the high-dose point
in this example, and this 63 Gy dose is more damaging, especially to late-responding
tissues with a low α/β ratio, than 63 Gy given as 2 Gy fractions (which is what is
implied when only the physical total dose is cited). Therefore, the spread of
radiobiological damage within a target volume has two components, namely that
due to variation in the total dose, and that due to variation in the fraction size by
which that total dose is given. The ‘double trouble’ effect is most marked for late-
responding tissues, for large treatment volumes with considerable dose variation,
and for intended treatment schedules where the prescribed fraction size is sometimes
large (radiotherapy of breast cancer may be such a situation). Although not yet
routine, radiobiological considerations need to be included in the analysis of the
dose–response relationship in patients who are experiencing side effects.
Two ways of accounting for ‘double trouble’ can be envisaged. First, IMRT
makes it possible to ensure homogeneity of dose distribution across a treatment
volume and this approach is being used clinically after it has been shown, for
example, to reduce late normal tissue effects of fibrosis after treatment for breast
cancer. Second, radiobiological dose plans may be constructed with the physical
dose mathematically replaced by some radiobiological equivalent which includes
the effect of variation in fraction size. For example, the physical dose at each point
in the dose matrix could be replaced by a dose calculated by the LQ model to have
41
42
Table 3.2 Effects of alterations in scheduling of radiotherapy on tumour and normal tissues
Fractionation Rationale Mode of delivery Effect on tumour Effect on Effect on

scheme acute-responding late-responding
tissues tissues
Hyperfractionation Tumours resemble Multiple daily fractions Increased tumour Increased Increased
acute-reacting tissues of less than 2 Gy using cell kill
with high α/β ratio an increased number of
fractions to give an
increased total dose
Accelerated High rate of repopulation Reduced overall treatment Increased tumour Increased Minimal
in tumours may time. Same fraction size, 6 h cell kill Increase
eliminate proliferation. interval between treatments
Time factor for late if used with hyperfractionation.
effects relatively Six or seven times a week
unimportant for daily fractions
Accelerated Combines the advantages Benefit of this approach either Increased tumour Increased Decreased
hyperfractionation of hyperfractionation with reduced total dose or cell kill or same
(CHART, concomitant and acceleration shorter overall treatment time
boost)
Hypofractionation May help to overcome Decreased number of fractions Same or increased Same or increased Increased if total
radio-resistance. Is of increased fraction size. cell kill dose is unchanged.
possible for some sites Suitable for small volume, Same if dose is
where normal tissue high-precision treatments decreased
reactions are not dose
limiting. Reduces overall
time eliminating proliferation
Information sources
the same effect on late-responding tissues when given as 2 Gy fractions. Isodose
curves can be constructed using the ‘radiobiological dose’ and compared with
more conventional isodose plots using the physical dose. Algorithms have been
developed which can be incorporated in commercial treatment planning systems
and used to compute radiobiological treatment plans. Alternatively, LQ transformed
DVHs (which incorporate the biological effects of changing fraction size within
the volume) can be computed and the biologically equivalent dose to the hottest
and coldest parts of the volume calculated.
Future directions
Careful analysis of data from clinical trials of novel fractionation schemes after
long-term follow up of late effects will permit refinement of understanding of the
LQ model. This will stimulate further investigation, particularly of shortened
courses of radiotherapy, which, if isoeffective with conventional fractionation,
would bring benefits to both patients and healthcare providers. Widespread use of
IMRT with functional imaging makes it possible to improve homogeneity within
a treatment volume or to plan for inhomogeneity according to defined biological
characteristics of the tumour.
Information sources
Ling CC, Humm J, Larson S (2000) Towards multi-dimensional radiotherapy (MD-CRT): biological
imaging and biological conformality. Int J Radiat Oncol Biol Phys 47: 551–60.
Royal College of Radiologists (2008) The Timely Delivery of Radical Radiotherapy: standards and
guidelines for the management of unscheduled treatment interruptions. Royal College of
Radiologists, London.
Steel GG (ed) (2002) Basic Clinical Radiobiology, 3rd edn. Hodder Arnold, London.
43
4 Organs at risk and
tolerance of normal
tissues
Therapeutic ratio
When considering how to achieve the best outcome from radiotherapy in the
treatment of tumours, it is critical to understand the concept of therapeutic ratio.
Cure is always achieved at some cost in terms of normal tissue damage. There must
be a balance between trying to ensure that all tumour cells receive a lethal dose of
radiation and that acute and late effects are tolerable. The total dose, dose per
fraction, treated volume and addition of drugs (radiochemotherapy) will all affect
this balance, as will individual factors for each patient (e.g. age, comorbidity and
intrinsic tissue radiosensitivity).
In some clinical situations, the frequency or severity of late effects drives a
reduction in radiotherapy dose. For example, concern about the high incidence of
second malignancies after radiotherapy for Hodgkin lymphoma has led to
chemotherapy being used in preference. In some paediatric tumours with high cure
rates (Wilms’, germ cell tumours), research has focused on lowering radiotherapy
dose to see if late effects can be reduced without compromising cure.
More commonly, a desire to increase radiotherapy dose to improve cure rates may
be limited by the response of normal tissues to radiation. However, although the
optimal balance of cure and side effects may have been reached for one radiotherapy
technique, further dose escalation may be possible with more precise treatment
delivery or better patient support. If a more conformal technique is used, the dose
to critical structures may be reduced and dose hot-spots or cold-spots in individual
sites or in parts of tumours can be minimised. Improvements in patient support, such
as using gastrostomy tubes for feeding patients undergoing head and neck
radiotherapy, may make acute effects more tolerable. A better understanding of the
mechanisms of late effects is helping to produce a less nihilistic approach to their
management. There are specialist teams now providing a multidisciplinary therapeutic
approach to the management of late effects of pelvic and breast radiation.
Any change to a treatment regimen, such as the use of IMRT, a dose increase
or addition of chemotherapy or biological agents, will change the therapeutic
ratio. The challenge for the radiation oncologist is to ensure that this change
improves the ratio and that an increase in dose is not counteracted by an increase
in unmanageable acute or serious late effects. TCP (tumour control probability)
and NTCP (normal tissue complication probability) are mathematical models used
to predict effects of such changes. However, to know whether a new treatment has
really produced better outcomes overall, and to inform and improve the reliability
of these modelling estimates, good clinical data must be collected, not only for
outcome measures relating to tumour control, but also for acute and late normal
tissue damage.
44
Organs at risk
Normal tissue effects can be specified in relation to the probability of them
occurring, their severity and their timing. Radiation injury may be expressed soon
after treatment (early effects) or after 6 months up to many years later (late effects).
Subsequent treatment, as for example with anthracyclines, may reveal latent,
previously asymptomatic, damage. Expression of damage may depend on genetic
susceptibility. It has been estimated that 20 per cent of the observed variation in
normal tissue sensitivity to radiation is random and 80 per cent deterministic,
including that due to genetic variations. No single gene has been isolated but
several conditions are known to predispose to abnormal radiation sensitivity. These
include ataxia telangiectasia, Fanconi’s anaemia and Bloom’s syndrome.
Underlying all normal tissue damage, there is a mechanism of dysregulated
repair of the radiation injury. Fibroblastic proliferation and extracellular matrix
deposition are influenced by cytokine and growth factor release and may lead to
endothelial proliferation and subsequent fibrosis. This is common in soft tissues
such as skin, breast, bowel, lung, kidney and liver. Alternatively, cell death may
lead to atrophy or necrosis of tissues as may occur with bone, nerves or brain.
Late damage will also depend on the hierarchical organisation of the irradiated
tissue at risk – whether the cells are serial in arrangement (e.g. spinal cord) or
parallel (e.g. liver).
Organs at risk
The ICRU defines OAR as those normal tissues which lie adjacent to tumours and
may therefore be included within treated volumes, with a risk that the radiation may
impair their normal functioning. Preparation of a treatment plan involves outlining
in three dimensions not only tumour and its potential extensions, but also any OAR.
OAR should be outlined according to protocols so that dose can be correlated
with end effect and comparisons made between institutions. For example in lung
cancer, the OAR volume for the normal lung is variously defined as the whole of
both lungs, lungs with GTV subtracted or lungs with PTV subtracted. The volume
of OAR can be expanded in three dimensions to take account of organ movement
and of systematic and random errors in treatment delivery. This will create a PRV
in the same way that a CTV is expanded to form a PTV.
There is a risk that the increasing awareness of organ motion and treatment
delivery errors will lead to larger PTVs and larger PRVs, which may overlap when
a dose solution is chosen. While it may be theoretically correct to generate a PRV
for each organ so that with each fraction the true location of that organ will be
within the PRV, in practice, dose limits for organs at risk are usually applied to the
OAR as defined on the planning images. One exception is the spinal cord where a
more conservative approach is often taken because of the potential severity of late
effects (paralysis). The cord itself can be contoured and a 3–5 mm margin added
isotropically to produce a PRV. Alternatively, the spinal canal is contoured as the
organ at risk which effectively adds a margin to the OAR. This approach also
makes it possible to make comparisons with data derived from 2D planning where
the spinal canal was considered to represent the cord.
For some tumours, the PTV can be treated with a plan where accepted tolerance
doses to normal tissues are not exceeded. But sometimes the clinician may need to
make a value judgement about the relative risks of possible normal tissue damage
45
OAR AND TOLERANCE OF NORMAL TISSUES
and loss of tumour control. It is important to consider the type and severity of the
effect, the possible consequences of a local tumour recurrence and how an
individual patient may tolerate radiotherapy. For example, late damage to the spinal
cord resulting in paralysis may be catastrophic, whereas an oesophageal stricture or
a cataract may be treatable. It may be acceptable to irradiate one kidney to high
dose if the contralateral one is functioning normally, but not in the presence of
hypertensive nephropathy. In postoperative radiotherapy for a tumour close to the
optic nerve, using a lower dose of radiation to try to keep within accepted tolerance
may increase the risk of blindness from a local recurrence. In trying to prevent
blindness, the risk of not giving adequate dose to the PTV may be greater than the
risk of normal tissue damage. In the same patient, the acceptable dose to the
ipsilateral and contralateral optic nerves may therefore be different.
Tolerance doses
There are very few prospective dose escalation studies of radiation, analogous to
phase 1 studies of new drugs, to help determine the maximum tolerated dose of
radiotherapy. The tables of tolerance doses which are used in clinical practice are
often extrapolated from laboratory or animal studies or at best relate to
radiotherapy given many years ago with techniques and technology which have
been superseded by 3D planning and treatment delivery. In addition, the fraction
size, dose rate, volume treated, concomitant therapy and comorbidity will all affect
the probability of late effects occurring in an individual. A tolerance dose needs to
be interpreted in this context. Moreover tolerance doses will change over time as 3D
dose distributions are correlated with late effects in the modern era. Nevertheless
it is possible to give guidance as to the chance that a given dose will produce a
given side effect and to define safe limits to use when devising a plan.
Correlating the risk of side effects with 2D dose distributions led to the production
of TD5/5 tables which are still used clinically today. These provide an estimate of
the dose which gives a 5 per cent probability of a given late effect 5 years after
treatment. Similarly a TD50/5 is the dose giving a 50 per cent risk of a particular
effect at 5 years. From these point estimates, models to predict probability of
normal tissue complications were developed such as the Lyman-probit and the
Kallman-relative seriality models.
While point doses such as TD5/5 may be useful for some serial organs such as
the spinal cord where exceeding a dose threshold at any point can compromise
whole organ function, they are less useful in organs composed of parallel subunits.
The advent of 3D planning has made it possible to describe the dose given to a
volume of tissue and to correlate it with acute and late effects. This description is
usually in the form of a DVH.
A DVH is a plot of dose of radiation on the x-axis and per cent volume of the
structure of interest on the y-axis. The shape and area under the DVH curve are used
to ensure that the target volume is adequately covered with a homogeneous dose and
that dose to critical structures is within acceptable limits. A 3D planning system can
calculate the dose in each pixel of the organ outlined and sum these to produce a
DVH. From this, the percentage of the volume of an organ receiving a given dose
(d) can be read – Vd. From the x-axis, the mean, median doses, etc. can be calculated.
Figure 4.1 shows a DVH for a lung plan where the V20 (volume of lung receiving
46
Tolerance doses
more than 20 Gy) is 7.89 per cent and the mean lung dose (total lung volume – PTV)
is 6 Gy. DVHs can also be used to evaluate dose given to a PTV. A plan should ideally
produce a steep curve showing that the dose within the PTV varies from no more than
95 per cent to 107 per cent of that prescribed in accordance with ICRU50 (Fig. 4.2);
DVHs for multiple volumes can be plotted on the same axes.
Dose (Gy)
10.0
12.0
14.0
16.0
18.0
20.0
22.0
24.0
26.0
28.0
30.0
32.0
34.0
36.0
38.0
40.0
42.0
44.0
46.0
48.0
50.0
52.0
54.0
56.0
58.0
60.0
62.0
64.0
66.0
68.0
70.0
72.0
0.0
2.0
4.0
6.0
8.0
100.0 100.0
90.0 90.0
Ratio of total structure volume (%)
80.0 80.0
Mean lung dose 6 Gy
70.0 70.0
60.0 60.0
50.0 50.0
40.0 40.0
30.0 30.0
Lung-PTV
20.0 Dose 30.3% (19.99 Gy) 20.0
Volume 7.89%
10.0 10.0
0.0 0.0
102.0
105.0
108.0
12.0
15.0
18.0
21.0
24.0
27.0
30.0
33.0
36.0
39.0
42.0
45.0
48.0
51.0
54.0
57.0
60.0
63.0
66.0
69.0
72.0
75.0
78.0
81.0
84.0
87.0
90.0
93.0
96.0
99.0
0.0
3.0
6.0
9.0
Relative dose (%)
Figure 4.1 DVH for a lung plan for a small peripheral tumour receiving 66 Gy in 33
fractions where the V20 (volume of lung receiving more than 20 Gy) is 7.89 per cent and
the mean lung dose (total lung volume – PTV) is 6 Gy.
Figure 4.2 Plan with dose colourwash showing dose within the PTV varying from 95 per cent
to 103.2 per cent of that prescribed, in accordance with ICRU50.
If DVHs are obtained from a series of patients in whom acute or late effects are
recorded, points on the DVH can be correlated with the probability of these effects
occurring. Thus, for parotid sparing IMRT, the chance of producing long-term
xerostomia correlates best with a mean dose to the contralateral parotid gland of
more than 24 Gy. For lung radiotherapy, the chance of symptomatic lung fibrosis
correlates best with a V20 of 32 per cent. These are the usual limits for an
acceptable plan. It must be remembered that these limits essentially still simplify a
3D dose distribution into a single dose, albeit the dose that is best correlated with
effect measured in a series of patients.
It is therefore essential to view and record the whole DVH for a volume rather
than just one dose point. This can be particularly useful when comparing two plans
47
for the same patient. Consider DVHs for the PTV and the lung in two possible
plans for a stage 3 lung cancer. Plan A conforms better to the PTV but with a V20
of 35 per cent. Plan B underdoses more of the PTV but has a V20 of 30 per cent.
The clinician must decide whether it is preferable to risk lung fibrosis to optimise
PTV coverage or whether, for this patient, a higher risk of fibrosis is unacceptable
and PTV coverage must be compromised.
DVH calculations do not take into account all the biological variables which
may determine a treatment outcome and the concept of BED has been developed
by Withers and colleagues (see Wilson 2007), based on the LQ model (see Chapter 3)
to allow comparison of equivalent doses delivered to a particular structure for
other dose/fractionation schedules. The formula for BED is:
BED {n*d [1 d/(α/β)]} – [(0.693*T)/( α*Tp)] (4.1)
where n number of treatments, d dose per fraction, T treatment time and
α/β 3 Gy (for late effects). Tp (potential doubling time), and α (linear
component of cell killing) are taken from published data for each tumour.
Equivalent uniform dose (EUD) is another method of summarising and
reporting inhomogeneous dose distributions which assumes that any two dose
distributions are equivalent if they cause the same radiobiological effect. A project,
known as Quantec, has recently been established to try to summarise 3D dose/
volume/outcome data in a clinically useful manner.
Individual organ tolerances

Suggested dose thresholds and dose–volume constraints for commonly irradiated
organs are given in Table 4.1. They assume radiation is prescribed at 2 Gy per fraction.
■ Spinal cord
Because a possible consequence of late radiation damage is irreversible paralysis,
treatments have been cautious and there is a shortage of clinical data on which to
base estimates of spinal cord tolerance. The constraints employed are therefore
necessarily conservative.
The spinal cord should either be defined with a 5 mm margin to produce a PRV
or the spinal canal contoured as a PRV. Dose to any part of the cord should be less
than 46 Gy. A small part (1 cm3) may receive up to 50 Gy. If more than 15 cm
length of cord is treated, the dose to any part of the cord should be less than
44 Gy. If dose per fraction is increased due to inhomogeneity of the plan (e.g. 103
per cent to spinal cord) and/or hypofractionation schedules are used, total dose to
the spinal cord must be lowered.
When palliative radiotherapy is repeated – for example, for spinal cord compression
or when the prognosis is short – a higher dose may be used if withholding treatment
is more likely to give a poor outcome than exceeding a theoretical dose limit.
■ Brain and peripheral nerves

For primary brain tumours, high doses of radiation are usually employed only
where there is a high risk of local recurrence with consequent brain injury, or when
prognosis is poor. The TD5/5 for brain necrosis varies with the volume irradiated.
For the whole brain TD5/5 45 Gy, for 1/3 of the brain TD5/5 60 Gy.
48
Table 4.1 Estimated tolerance doses for various organs expressed as different
parameters with dose delivered with 2 Gy/fraction. These have wide confidence limits and
vary with age, individual sensitivity, vascular status and with other treatments given
OAR TD5/5 TD50/5 DVH Vx % or Tolerance
(Gy) (Gy) mean dose in Gy dose (Gy)
Spinal cord 5 cm 50 70 45–50

10 cm 50 70 40–44 (15 cm)
20 cm 47 EUD 52.5
Brain Whole 45 50–60

1/3 60
Brainstem 1/3 60 54
2/3 53 65 V60 0.9 mL 1% up to 60
3/3 50
Peripheral nerves 60
Pituitary gland 20–24
(hormone production)
Permanent hair loss 45–55
Optic nerve 50–55
Optic chiasm 50
Lacrimal gland 32–35
Lens 10
Retina Whole 45 45–50
Small
volume 60
Cornea 48
Cochlea 50
Parotid 2/3 32 46 V30 45% 24
3/3 32 46
Epiphyses before 10
fusion in children
Femoral heads V50 50
Heart 1/3 60 70 V40 30 Dmax 60
2/3 45 55 V30 40–45
3/3 40 50 V20 50
Lung 1/3 45 65 V30 10–15
2/3 30 40 V20 25
3/3 17.5 24.5 Mean 10
Kidney 1/3 50
2/3 20–30 40 Mean 17.5 Gy
3/3 23 28
(Continued)
49
Table 4.1 Continued
OAR TD5/5 TD50/5 DVH Vx % or Tolerance

(Gy) (Gy) mean dose in Gy dose (Gy)
Oesophagus V50 32

58 (max) Surface dose
32 Gy
34 (mean) 50 Gy 8–12 cm
Small bowel 1/3 50 45 (250 cm3)
Rectum V74 3 65 Gy
V70 15–25
V60 50
V50 60
Liver 1/3 50 55 V30 60 1/3 40–80
2/3 35 45 Mean 30–35 Gy 2/3 30
3/3 30 40 3/3 25
Testis:
hormone production 25–35
spermatogenesis Transient 0.2 Gy;
count down for
2–3 years 2–3 Gy;
permanent
sterility 6 Gy
Ovary Menses 1.5 Gy
Ovarian failure
6–15 Gy
Bladder Whole 65 V74 5 Whole 50–55
2/3 80 V60 25 Partial 65–75
V50 50
Data taken from multiple sources including personal records, Emani et al. (1991), Milano et al. (2007).
EUD, equivalent uniform dose; TD, total dose.
The use of tolerance doses is more applicable to head and neck cancers where
the PTV is close to neural tissue. If the brain is not part of the target volume, dose
to any part should not exceed 60 Gy.
The brainstem is traditionally regarded as more radiosensitive than the cerebrum.
It should not receive more than 54 Gy (1 per cent up to 60 Gy). For peripheral
nerves such as the brachial plexus, dose should be limited to 60 Gy.
■ Optic nerves and orbital tissues

Tolerance doses to optic structures are usually approached when the PTV is close to
those structures. Underdosing the PTV and increasing the risk of local recurrence
may threaten sight more than potentially overdosing the optic pathway.
■ Optic nerve – dose to any point 50 Gy. Accept 55 Gy if the PTV is very close
to one optic nerve
50
■ Optic chiasm – dose to any point 50 Gy
■ Lacrimal gland – dose to any part 35 Gy
■ Lens – dose to any part 10 Gy. The late normal tissue effect (cataract) can be
treated successfully, so this dose may be exceeded with patient consent in some
tumours
■ Retina – dose to whole retina 45 Gy (TD5/5 for visual damage). Dose to a
small volume 60 Gy
■ Mucosa
Volume irradiated should be minimised to reduce acute mucositis. To prevent
oesophageal stricture, the length of oesophagus in the treated volume is kept as
short as possible and ideally 10 cm.
■ Lung
Late fibrosis is best correlated with V20 with a target of 32 per cent of the lung-
PTV volume receiving more than 20 Gy. Other DVH parameters including mean
lung dose are also sometimes used.
■ Kidney
TD5/5 for the whole kidney is 23 Gy. Dose to two-thirds of one kidney (and
ideally both) should be below 20 Gy.
■ Liver
If the whole liver is irradiated, dose should be 30 Gy to avoid radiation hepatitis.
V30 should be below 60 per cent.
■ Testes and ovaries

For the testis, 0.2 Gy can cause transient brief oligospermia; 2–3 Gy can lower
sperm counts for 2–3 years. Doses of more than 6 Gy cause permanent sterility and
of more than 20 Gy affect hormone production.
Menses are suppressed at 1.5 Gy, and 6–15 Gy causes permanent ovarian
failure. Increasing age lowers the threshold for these effects.
■ Second malignancies
Any radiation dose increases the risk of second malignancy so no safe dose limits
can be given. In principle, the irradiated volume should be kept as small as
possible. Techniques such as IMRT which use multiple beams may increase the
volume of normal tissue irradiated and theoretically increase the risk of second
malignancy. It will take decades before data on the incidence of second malignancies
is obtained so the possible increase can only be estimated from biological modelling.
Clinical experience with radiotherapy for Hodgkin lymphoma, where the absolute
incidence of second malignancies after mantle radiotherapy is 30 per cent at 30 years,
shows the importance of long-term follow-up and data collection in the assessment
of late effects, particularly when a new treatment technique is introduced.
51
Measurement of late effects

Local control and survival are usually relatively easy to assess with clinical
examination, imaging and population databases. In contrast, estimation of the
frequency and severity of late effects is often haphazard and incomplete and will
depend on the sensitivity of the test used for their detection. Over time treatments
change frequently, making it difficult to evaluate the role of each component of
treatment to the outcome.
Adverse outcomes of treatment are often poorly documented by physicians and
data are often only collected retrospectively. Pressurised doctors prioritise care of
the patient before recording outcomes and there is no simple internationally
standardised measure of late effects. Most scales involve grading of late effects from
the physician’s perspective and have been shown to underestimate late effects from
the patient’s point of view.
Existing scales include the RTOG/EORTC, the French/Italian scheme for
gynaecological cancer, LENT-SOMA (late effects of normal tissues – subjective,
objective, measured and analytic) and the NCI-CTC (common toxicity criteria) v3 –
the National Cancer Institute of USA terminology criteria for adverse events
recording with a severity scale for each item. This is used widely for clinical trials
but is rather complex for day to day clinic use. Studies of simpler scales for routine
use by physicians or for self-reporting by patients are in progress.
Most endpoints in these scales are clinical. Modern imaging modalities will
be more sensitive at detecting abnormalities than are ascertained by clinical signs
and symptoms or laboratory markers. Currently, imaging is rarely used for the
documentation of late effects of radiation, unless subclinical damage progresses to a
clinical problem, there is some early preventative intervention possible or the
patient is being treated within a trial of a new modality.
Treatment of late effects

Until recently, little attempt has been made to modify any radiation associated
damage although prevention may be possible in some cases. Restricting aggravating
factors to damage by stopping smoking, maintaining good control of blood pressure
and blood sugar levels, and avoiding use of fibrogenic drugs such as bleomycin may
help. Control of acute inflammation by corticosteroids or anti-inflammatory drugs
may reduce late effects. Vascular-directed therapies such as pentoxifylline, hyperbaric
oxygen and angiotensin-converting enzyme inhibitors are being studied with some
promising results. Anti-oxidant therapies such as superoxide dismutase and vitamin
E have also been tried. Established radionecrosis has been treated with antibiotics,
anti-inflammatory agents, bisphosphonates (bone) and hyperbaric oxygen with
variable outcomes. Further improvement in preventing or treating late effects will
come from collaborative research by specialised multidisciplinary teams.
Information sources
Chan L, Xia P, Gottschalk M et al. (2008) Proposed rectal dose constraints for patients undergoing
definitive whole pelvic radiotherapy for clinically localised prostate cancer. Int J Radiat Oncol
Biol Phys 72: 69–77.
52
Information sources
Chon BH, Loeffler JS (2002) The effect of nonmalignant systemic disease on tolerance to radiation
therapy. Oncologist 7: 136–43.
Drzymala RE, Mohan R, Brewster L et al. (1991) Dose-volume histograms. Int J Radiat Oncol Biol
Phys 21: 71–8.
Emani B, Lyman J, Brown A et al. (1991) Tolerance of normal tissue to irradiation. Int J Radiat Oncol
Biol Phys 21: 109–22.
Franklin JG, Paus MD, Pluetschow A et al. (2006) Second malignancy risk associated with treatment
of Hodgkin’s lymphoma: meta-analysis of the randomised trials. Ann Oncol 17: 1749–60.
Larrier NA, Marks LB (2007) What radiation dose is safe in non-small cell lung cancer? Nat Clin
Pract Oncol 4: 80–1.
Lee SP, Leu MY, Smathers JB et al. (1995) Biologically effective dose distribution based on the
linear quadratic model and its clinical relevance. Int J Radiat Oncol Biol Phys 33: 375–89.
Marks L (2008) MO-D-AUD A-02: A clinician’s view of Quantec. Med Phys 35: 2863–4.
Milano MT, Constine LS, Okunieff P (2007) Normal tissue tolerance doses. Semin Radiat Oncol
17: 131–40.
Niemerko A (1997) Reporting and analyzing dose distributions: a concept of equivalent uniform
dose. Med Phys 24: 103–10.
Wilson G (2007) Cell kinetics. Clin Oncol 19: 370–84.
53
5 Principles of
brachytherapy
Introduction
Brachy is from the Greek word for ‘short’ so brachytherapy (also known as sealed
source radiotherapy) roughly translated means short-distance therapy. A radioactive
material is inserted directly into or next to a tumour and concentrates the dose there.
The dose falls off very rapidly according to the inverse square law, and surrounding
normal tissues receive substantially lower doses than the tumour. When 65 Gy are
delivered at 0.5 cm from the source, the dose at 2 cm is only 4.06 Gy.
As well as its physical advantages, there are also biological advantages. Low dose
rate (LDR) brachytherapy is a type of extreme hyperfractionation and is therefore
relatively sparing to normal tissues. The dose rate may be low but it is delivered
continuously, which shortens overall treatment time and reduces the opportunity
for tumour repopulation during treatment. Conversely, high dose rate (HDR)
brachytherapy must be fractionated to avoid normal tissue morbidity. Three dose
rate bands are defined: LDR (1 Gy/h), medium dose rate (MDR) (1 to
12 Gy/h) and HDR (12 Gy/h). It is important to remember that if the dose
rate is increased, a dose reduction is needed to give a biologically isoeffective dose.
When changing from low to medium dose rate (e.g. changing from LDR
intracavitary brachytherapy to MDR), a dose correction of approximately minus
15 per cent is needed. Other advantages of brachytherapy include the accurate
localisation and immobilisation of the tumour, which removes the problems of
organ movement and set-up errors seen with external beam radiotherapy (EBRT).
The disadvantages of brachytherapy are the operative nature of the procedures
often needed to access the tumour, the requirement for skilled personnel, and
the radiation protection measures needed to protect patient, staff and general
public.
Brachytherapy is considered whenever possible for accessible localised tumours
of relatively small volume. It is contraindicated where tumour infiltrates bone,
where the margins of the tumour or target volume are not clearly identifiable and
where there is active infection in the tissues. Brachytherapy is used as a radical
single modality treatment or in combination with EBRT to deliver a boost dose.
It can be used after surgical excision to irradiate a tumour bed. Isotopes used for
brachytherapy are shown in Table 5.1.
■ Delivery systems
With interstitial brachytherapy, sources are inserted directly into tissue. Iridium-
192 wire is ideal and can be cut to any length and curved as required. It is used as
hair pins to treat cancer of the anterior tongue, or looped to treat base of tongue
54
Clinical use
Table 5.1 Isotopes for brachytherapy
Source Form Dose rate Emissions Half-life
Radium-226 Tubes, needles LDR 2.45 MV gamma 1620 years

Caesium-137 Tubes, needles; LDR 0.662 MV gamma 30 years
afterloading pellets
Cobalt-60 Tubes; afterloading HDR 1.17, 1.33 MV 5 years
pellets gamma
Iridium-192 Wires; afterloading LDR 0.38 MV gamma 74 days
pellets HDR
Iodine-125 Seeds LDR 27.4, 31.4, 35.5 kV 60 days
Palladium-103 Seeds LDR 21 kV 17 days
Ruthenium-106/106 Eye plaques LDR 3.54 MeV 373 days
Rhenium
Strontium-90 Eye plaques HDR 0.546 MeV 28.9 years
tumours. Prostate cancer brachytherapy with iodine-125 or palladium-103 seeds is

also classed as interstitial therapy.
Intracavitary brachytherapy places applicators inside a body cavity such as the
uterine canal or vagina. These applicators can then be afterloaded with radioactive
sources. Caesium-137 is the isotope of choice for low dose rate treatments, and
small iridium-192 sources for HDR afterloading systems. Surface applicator
brachytherapy can be used for very superficial lesions less than 1 mm thick such as
strontium-90 eye plaque therapy after resection of pterygium.
Mould brachytherapy uses sealed sources held in a fixed arrangement and
distance from the surface by a custom-made mould. It is usually used to treat
superficial lesions of skin, mouth or vagina. Intraluminal brachytherapy places
applicators loaded with radioactive sources in a lumen such as the bronchus or
oesophagus.
For tumour sites where it would be difficult to remove the sources, or where
very low dose rate is preferred, a permanent implant can be performed with sources
such as iodine-125 seeds for prostate cancer. High dose rate intracavitary afterloading
systems place sources temporarily using flexible catheters. Temporary interstitial
HDR prostate implants are also possible.
To reduce radiation dose to staff, techniques of afterloading have been developed
which involve the initial implantation of non-radioactive applicators, catheters or
carriers into the patient. The radioactive sources can then be ‘manually afterloaded’
by the operator into the applicators, or ‘remotely afterloaded’ by a machine under
computer control. Remote afterloading reduces the doses to staff, patients and
visitors to a minimum and, with appropriately shielded rooms, can be used for LDR
continuous, and HDR fractionated, treatments.
Clinical use
Table 5.2 shows some of the common sites treated with brachytherapy techniques.
55
PRINCIPLES OF BRACHYTHERAPY
Table 5.2 Clinical uses of brachytherapy
Site Indications Technique Source and

dose rate
Anterior tongue Small T1/T2 Hair pins Iridium-192 LDR

Buccal mucosa Small T1/T2 Plastic tube Iridium-192 LDR
Base of tongue Boost after EBRT Loop Iridium-192 LDR
Lip Small T1/T2;boost Needle Iridium-192 LDR
after EBRT
Nasopharynx Boost after EBRT; Moulds; Iridium-192;
re-treatment NPC applicators LDR or HDR
Recurrent disease Re-treatment Plastic tube Iridium-192;
in neck LDR or HDR
Uterine tumours Postoperative; palliative Vaginal applicator; HDR or LDR
Tube and ovoids
Cervical tumours Boost after EBRT Tube and ovoids HDR or LDR
Vagina Small stage 1; boost Perineal template HDR or LDR
after EBRT; palliative
Vulva Boost after EBRT; palliative Perineal template HDR or LDR
Anal Small T1; boost after Perineal implant HDR or LDR
EBRT; palliative through template
Breast Boost after EBRT; palliative Plastic tubes; Needles; HDR or LDR
Bridge template
Prostate Low risk disease; Permanent seeds; LDR I-125 or
boost after EBRT Perineal template Pd-103; HDR
Dosimetry
The spatial configuration of brachytherapy sources in a target volume is chosen to
achieve as homogeneous a dose as possible. The dose distribution is inherently
inhomogeneous with high doses around each source, which can cause necrosis,
and low doses between sources, which can result in recurrence. An established set
of rules for implantation must be followed to achieve good dose distributions.
Several systems have been used to calculate and describe the dose distributions of
brachytherapy implants. The Manchester system is widely used for gynaecological
implants. However the GEC-ESTRO group has recently published recommendations
on target volume concepts and plan evaluation using DVHs (see Chapter 32). The
Paris system was specifically designed for use with iridium wire afterloading
techniques. Both these systems use traditional dose formalism for manual calculations
with reference to precalculated data such as Paterson–Parker tables for needle implants,
and the cross-line graphs or escargot curves for iridium wire. These systems have
been adapted for computer calculations which follow the TG43 formalism,
published in 1995 by the American Association of Physicists in Medicine (AAPM)
Task group 43. The sources must be distributed according to the particular
dosimetry system used and the method of dose specification and prescription
56
System for iridium wire implants
adhered to. Previously it was always important to plan in advance the number and
distribution of radioactive sources. With modern implant and dosimetry techniques,
it is now possible to perform dynamic intraoperative dosimetry, e.g. for prostate
seed implants, which reduces the amount of preplanning needed and avoids
repositioning errors. An estimate of the volume to be implanted and the number
of sources still needs to be made and the sources must be distributed according to
the system used.
The Manchester system for interstitial implants

This was based on the use of radium sources with dose tables that gave the amount
of radium and time needed (mg h tables) to give 1000 roentgens (1000 cGy) to
the treated surface. The Paterson–Parker rules provide sets of distribution rules for
planar or volume implants. For a simple planar rectangular implant, the sources
must be parallel and the distance between sources should not exceed 10 mm. The
end of rows of parallel needles is crossed by needles at right angles with two-thirds
of the sources at the periphery and one-third in the central area. If an end is not
crossed, 10 per cent is deducted from the area when reading from the mg h tables.
The Paris system for iridium wire implants

and afterloading techniques
The Paris system was developed for iridium-192 wire implants and can also be
used to calculate doses for computer-based HDR systems. The distribution rules
for iridium-192 implants are as follows:
■ Active sources should be parallel and straight.
■ The lines should be equidistant.
■ The line or plane on which the mid-point of the sources lies (central plane) should
be at right angles to the axis of the sources.
■ The linear activity should be uniform along the length of each line, and identical
for all lines.
■ The separation of sources may be varied from one implant to another. A
minimum of 8 mm separation is acceptable for the smallest volumes, rising to
20 mm for the largest.
■ For volume implants, the distribution of sources in cross-section (central plane)
should be either in equilateral triangles or in squares.
■ Because it is not usual to cross the ends of the sources, the average length of
active wire must be longer than the target volume by 25–30 per cent depending
on the number and separation of sources used.
A Paris implant can be a single plane with regularly spaced wires, a circular
arrangement of needles/catheters, or a multiple plane arrangement to treat thicker
tumours. The multiple plane arrangement can be triangular, rectangular or square.
The dose calculation is then based on the distribution of sources in the central
plane, that is, the plane which is at right angles to the axis of the mid-point of the
sources. An example of a rectangular implant is when hairpins are used to treat
anterior tongue tumours. Here the central plane should be half way down the legs
57
of the hairpin ignoring the cross piece. Computer systems can now allow rotation
of the implant in 3D to visualise the implant and central plane.
The calculation then uses the basal dose rate, which is the dose in the middle of
the implanted volume where the dose rate is lowest. The basal dose rate at a point
is the summation of dose rate contributions from each source according to the
distance of the source from the point. In the case of a large implant, there may be
several basal dose rate points, and a mean basal dose rate is taken for the implant
as a whole (Fig. 5.1).
BD
BD
BD
(a) (b)
Figure 5.1 Basal dose rate point for two different volume implants arranged in (a) one
and (b) two equilateral triangles.
Once the basal dose rate at the centre of the implant is known, the reference
dose is taken as 85 per cent of the basal dose rate. This is then used to calculate
the duration of the implant and the 85 per cent isodose defines the treated volume.
The time needed for the implant is derived by dividing the prescribed dose by the
reference dose rate and takes into account the activity of the wire used and
radioactive decay during the implant. An example is shown in Table 5.3.
It is important to know the relationship between volume treated and length and
separation of sources used when performing an implant. The following apply to
implants according to the Paris system:
■ the length of the treated volume is approximately 70 per cent of the length of
the active sources (Fig. 5.2)
■ the thickness of the treated volume in a single-plane implant is approximately
50 per cent of the separation between the sources
■ the treatment margin around a volume implant performed in triangles is
30–40 per cent of the distance between the sides of the triangle
■ the ratio of treated volume to source length or separation increases as more
sources are used.
With the increased use of computer programmes for dose calculation, there is a
tendency to prescribe to computer-derived isodoses. The isodose for prescription
is chosen where the dose gradient is very steep, and there may be wide variation
between the dose at the periphery and that at the centre of the target volume.
A considerable proportion of the implanted volume may therefore receive a higher
58
Dose reporting
Table 5.3 Calculation for breast implant
Two-plane implant to deliver 25 Gy

Superficial plane 5-cm wires 2
Deep plane 7-cm wires 3
Separation between sources 18 mm
Activity of wire (midway through treatment) air kerma rate 0.5 μGy/h/mm at 1 m (0.1193 mG/mm)
1 2 5 cm
A C
3 4 5 7 cm
Wire A B C
Distance Dose rate Distance Dose rate Distance Dose rate
(mm) (Gy/h) (mm) (Gy/h) (mm) (Gy/h)
1 10.4 0.1245 10.4 0.1245 20.8 0.0465

2 20.8 0.0465 10.4 0.1245 10.4 0.1245
3 10.4 0.1340 20.8 0.0555 27.5 0.0365
4 10.4 0.1340 10.4 0.1340 10.4 0.1340
5 27.5 0.0365 20.8 0.0555 10.4 0.1340
Total 0.4755 0.4940 0.4755
0.4755 0.4940 0.4755 Gy/h

Mean basal dose rate
3
0.4817 Gy/h
Reference dose rate 0.4817 0.85
(85%) 0.4095 Gy/h
25.00
Treatment time
0.4095
61.05 h
2 days 13 h
dose than that at the periphery. For safe treatment, it is advised that the central
dose should be no more than 20 per cent higher than the peripheral dose.
Dose reporting
The ICRU report 38 (1985) gives guidance on reporting absorbed doses and
volumes for intracavitary brachytherapy. It recommends that a combination of total
reference air kerma, description of the reference volume and absorbed dose at
reference points be used to specify intracavitary applications for cervix carcinoma.
59
I
L Figure 5.2 Relationship between treated

I/L 0.7 volume (l) and length of active sources (L).
The ICRU report 58 (1997) recommends that the following information should
be reported following an interstitial implant:
■ Description of volumes:
– gross tumour volume
– clinical tumour volume
– treated volume
■ Description of sources and techniques:
– description of time pattern
– total reference air kerma (TRAK) (the sum of the products of the reference
air kerma rate and irradiation time for each source)
■ Description of doses:
– prescribed dose
– mean central dose in the central plane (equivalent to basal dose in the Paris
system)
– peripheral dose (equivalent to the reference dose in the Paris system)
■ Description of high and low dose volumes.
A highly skilled and trained team is essential to perform brachytherapy implants
safely. Everyone in the team should be trained in the principles of radiation safety
so that the patient, staff and public are not at risk of unnecessary irradiation.
Protective measures should be in place to keep dose levels as low as reasonably
practicable (ALARP).
Legislation pertaining to brachytherapy

The following legislation governs the use of sealed brachytherapy sources in
hospitals in the UK:
■ Health and Safety at Work Act 1974
60
Information sources
■ Ionising Radiations Regulations 1999
■ Ionising Radiation (Medical Exposure) Regulations 2000 (IR(ME)R)
■ Medicines (Administration of Radioactive Substances) Regulations 1978
(amendment 1995)
■ Radioactive Material (Road Transport) Regulations 2002
■ Radioactive Substances Act 1993
The IR(ME)R practitioner for brachytherapy must be authorised under the
Medicines (Administration of Radioactive Substances) regulations and hold a valid
ARSAC licence before being able to administer radioactive sources for brachytherapy.
The regulations allow for the safe implementation of brachytherapy in clinical practice.
Information sources
Gerbaulet A, Pötter R, Mazeron J-J et al. (2002) The GEC ESTRO Handbook of Brachytherapy.
ESTRO, Brussels.
Hoskin P, Coyle C (2005) Radiotherapy in Practice. Oxford University Press, Oxford.
International Commission on Radiation Units and Measurements (1985) Dose and Volume
Specification for Reporting Intracavitary Therapy in Gynaecology, ICRU Report 38. Bethesda,
Maryland, USA.
International Commission on Radiation Units and Measurements (1997) Dose and Volume
Specification for Reporting Interstitial Therapy, ICRU Report 58. Bethesda, Maryland, USA.
Report of American Association of Physicists in Medicine Radiation Therapy Committee Task Group
43 (1995). Med Phys 22: 209–35.
www.oxfordjournals.org/jicru/backissues/reports.html. Reports 50, 62 and 71.
61
6 Emergency and
palliative radiotherapy
Indications for radiotherapy

Emergency radiotherapy, which should be given within 24 h of diagnosis, is only
indicated for selected patients with spinal cord compression. Urgent palliative
radiotherapy, used to treat various other symptoms from primary disease or
metastases, should be given as soon as possible. The set-up and planning are kept
as simple as possible for palliative treatments, but they have to be individualised for
each patient. Palliation requires as much skill as radical treatment. Accurate
definition of the tumour causing the symptom is important and side effects of
treatment must be minimised to ensure overall benefit to the patient.
Spinal cord compression

This is a medical emergency. The spinal cord is compressed most commonly by
metastatic tumour involving the vertebrae, or less commonly by a benign cause
such as a vertebral fracture, abscess or ruptured intervertebral disc. The spinal cord
ends at approximately L1 and compression below this level causes the cauda
equina syndrome.
Spinal cord compression (SCC) occurs in approximately 5 per cent of patients
with cancer, most commonly with primary tumours of the lung, prostate, breast,
kidney and thyroid, and lymphoma and multiple myeloma.
The most important determinant of outcome is the severity of neurological
damage at the time treatment is initiated which is why treatment must be
considered as an emergency. Of patients without significant neurological deficit,
80 per cent remain ambulant or regain the ability to walk, whereas only 50 per cent
of those with even a mild transverse myelopathy and 5 per cent of those with
paraplegia, do so. The prognosis is dependent on the type and extent of the
primary malignancy. Untreated patients with SCC often die within a month. With
treatment, the median overall survival ranges from 3 to 16 months.
■ Clinical features
A high index of suspicion is needed to detect cases early while neurological
function is still intact. Common features of SCC in the thoracic area are back pain
(typically radicular), sensory disturbance in the lower limbs, bladder or bowel
dysfunction and leg weakness. The neurological signs are of bilateral upper motor
neurone lesions in the legs and a sensory level. Cervical cord involvement may be
suspected if there are signs and symptoms in the arms. In the lumbar spine area,
compression may be of the cauda equina, causing nerve root pain in the back and legs,
62
urinary disturbance, signs of a lower motor neurone lesion in the legs and patchy
asymmetrical sensory loss. Nerve root irritation may be shown by limitation of
straight leg raising. Onset of symptoms may be insidious, or occasionally paraplegia
may develop rapidly with few preceding symptoms. Any delay in diagnosis and
treatment will impact on functional outcome. Complete sudden paraplegia is
usually associated with vascular damage and is commonly irreversible.
■ Investigations
A patient with suspected SCC needs an emergency MR scan of the whole spine,
which is the most informative and least invasive technique. There are frequently
multiple levels involved and clinical signs can appear to be out of keeping with the
vertebral level involved (Fig. 6.1).
Figure 6.1 MRI of spine showing spinal

cord compression at T2 and T9.
A diagnosis of malignancy may already be known. If not, a good history and

general examination should be undertaken to search for a primary tumour.
Investigations such as a chest X-ray, tumour marker estimations, biopsy or fine
needle aspiration and cytology should be performed.
■ Treatment
At presentation, all patients suspected to have SCC should be given high dose
steroids (e.g. dexamethasone 16 mg daily).
■ Sequencing of multimodality treatment

When planning treatment, the overall picture of the patient’s health status,
neurological function, site and histology of the primary, sites of metastases, prognosis
and further treatment options, performance status and previous treatments, especially
with radiotherapy, need to be taken into account. In patients who are not known
to have malignancy, surgical decompression should be the first consideration; or if
malignancy is highly suspected and the patient is not fit for surgery, an image
guided vertebral biopsy can be performed.
63
EMERGENCY AND PALLIATIVE RADIOTHERAPY
A neurosurgeon should urgently assess all patients who are fit for surgery,
reviewing clinical features and MRI to assess whether there is a place for
multimodality treatment. Immediate consultation is made possible with remote
image viewing. Surgery for SCC involves an anterior decompression and stabilisation
of the spine. The indications for surgery are:
■ unknown primary tumour

■ unstable spine or vertebral displacement
■ relapse following spinal radiotherapy
■ neurological symptoms which progress during radiotherapy
■ relatively radio-resistant tumour
■ paralysis of rapid onset.
There is evidence that some patients have a better functional outcome if treated
with emergency spinal decompressive surgery followed by postoperative
radiotherapy.
Some patients have very chemo-sensitive tumours such as lymphoma or small
cell carcinoma of the lung, and chemotherapy can be started urgently before
radiotherapy.
If surgery or chemotherapy are not appropriate, EBRT is given immediately to
prevent further neurological damage, to improve function and for pain relief.
■ Clinical and radiological anatomy

A full neurological examination should include search for motor impairment, sensory
levels, and local pain and tenderness. These symptoms and signs should be correlated
with MRI appearances in consultation with a radiologist. Metastases may be lytic or
sclerotic, and collapse, compression laterally or posteriorly, and any paravertebral soft
tissue mass should be noted. The sensory level detected in a skin dermatome arises
from compression of the corresponding cord segment, which lies at a higher level
than the vertebral body of the same number, e.g. a sensory level at T10 on the skin
arises from compression of its cord segment at the level of the T8 vertebra.
■ Data acquisition
The patient is planned and treated ideally in the prone position using a direct
posterior beam to avoid increased skin dose from treatment through the couch top.
However, the supine position using an undercouch beam may be easier and more
comfortable for the patient. Treatment should be planned using a CT scanner for
virtual simulation or a simulator. Information from clinical examination and the
MR scan is used to design the target volume.
Three reference tattoos are placed at the isocentre and bilaterally.
■ Target volume definition

The GTV includes vertebral and soft tissue tumour as seen on CT planning scan
and diagnostic MRI. The CTV includes the spinal canal, the width of the vertebra
and one vertebra above and below the SCC if the planning is based on MRI, or
two vertebrae above and below if based on X-ray or CT to allow for uncertainty
about extent of microscopic disease. The CTV to PTV margin is 1 cm.
64
In patients who have had surgery, the CTV will also include any metal that has
been used to stabilise the spine.
■ Dose solutions
3D CT planning or virtual simulation may be used (Fig. 6.2). To treat the PTV
adequately at depth, a direct 6 MV photon beam may be used. For lumbosacral
lesions, a better dose distribution may be obtained with opposing beams. If
treatment is delivered with a cobalt-60 source, an extra margin for the penumbra
should be added according to departmental protocol. The field edge defined at the
simulator to cover the PTV represents the 50 per cent isodose.
(a)
Figure 6.2 Virtual

simulation for treatment of
spinal cord compression
showing (a) posterior beam
arrangement, and (b)
sagittal dose distribution
from 6 MV beam. (Depth to
anterior spinal canal
(b) 4.52 cm.)
The dose prescription point is the depth of the anterior spinal canal. This can be
assessed from axial imaging and usually at 5–7 cm in the cervical and thoracic
region, and at 7–8 cm in the lumbar region.
65
Occasionally SCC is caused by a primary tumour such as a plasmacytoma, and radical

radiotherapy can produce permanent local control or even cure. In this situation, a
planned homogeneous dose distribution with wedged posterior oblique beams with or
without a direct posterior beam (Fig. 6.3), or electron therapy can be used.
Figure 6.3 CT plan (with

dose colourwash) to treat
tumour of vertebral body
using 6 MV posterior
oblique wedged beams.
■ Dose fractionation
Palliation for good prognosis disease where radiotherapy is the
first definitive treatment and postoperatively
20 Gy in 5 daily fractions of 4 Gy given in 1 week.
30 Gy in 10 daily fractions of 3 Gy given in 2 weeks.
A single dose of 8 Gy may be used for palliation of pain in patients with
established paraplegia for 24 h.
Radical doses
Solitary plasmacytoma: 45 Gy in 25 daily fractions of 1.8 Gy given in 5 weeks.
Lymphoma: 30–36 Gy in 15–18 daily fractions given in 3–31⁄2 weeks.
■ Treatment delivery and patient care

An experienced multidisciplinary team should care for a patient during treatment.
A patient with an unstable spine or undergoing surgery needs specialist nursing
and physiotherapy. Those undergoing radiotherapy need specialist input from
experienced radiographers, nurses and physiotherapists to help them to rehabilitate
and regain neurological function. Ongoing oncological management needs to be
planned and the palliative care team involved for symptom control and support. Doses
of dexamethasone should be reduced gradually after completion of radiotherapy.
66
SVCO
Superior vena caval obstruction
■ Indications for radiotherapy
Over 90 per cent of cases of superior vena caval obstruction (SVCO) have a
malignant cause. Although uncommon in patients with lung cancer, the commonest
cause of SVCO is nevertheless small cell and non-small cell lung cancer. Other
malignant causes are lymphoma and metastasis from mesothelioma, thymoma or any
tumour that spreads to mediastinal lymph nodes.
The obstruction arises from compression of the SVC by tumour at the right
main or upper lobe bronchus or by large volume mediastinal lymphadenopathy.
Symptoms may be severe when the obstruction is below the entry of the azygous
vein. The clinical features are neck swelling and distended veins over the chest. There
may be swelling of one or both arms, shortness of breath, hoarse voice and
headaches. The diagnosis is made on contrast-enhanced spiral or multi-slice CT
scans, which can accurately identify the site of occlusion or stenosis and the presence
of intravascular thrombus. Impending SVCO may also be diagnosed on CT scans.
SVCO is no longer considered a radiotherapy emergency, as outcome is not
related to the duration of symptoms. Urgent action may be needed to prevent
SVCO leading to airway obstruction from laryngeal or bronchial oedema, or coma
from cerebral oedema.
Confirmation of histology is important and can be obtained by biopsy of the
primary tumour at bronchoscopy or mediastinoscopy, by percutaneous CT-guided
biopsy, or by biopsy of an involved cervical lymph node. Tumour markers such as
α-fetoprotein (AFP) and β-hCG for germ cell tumours, lactic dehydrogenase for
lymphoma, and prostate-specific antigen (PSA) for prostate cancer may be helpful.

Steroids such as high dose dexamethasone are traditionally given as part of SVCO
management. Their use should be of short duration.
SVC stenting has been shown to be the most effective treatment with rapid relief
of symptoms and should be considered first wherever available or for patients who
fail to respond to chemotherapy or radiotherapy.
Chemotherapy should be considered in chemo-sensitive tumours such as small
cell lung cancer, lymphoma, leukaemia and germ cell tumours. Radiotherapy
should be considered in non-small cell lung cancer and other less chemo-sensitive
tumours and when SVC stenting is not available.
Where patients cannot lie down, they may be treated sitting upright using a direct
anterior beam with margins determined from clinical examination, chest X-ray or
CT appearances mapped on the patient by reference to landmarks such as the
suprasternal notch. Beam sizes of 12 12 cm are usually adequate.
Ideally, patients should be treated supine with 3D conformal CT planning,
virtual CT simulation or simulator planning. The CT scans are taken with 3–5 mm
slices from the lower neck to the diaphragm.
67

The GTV is defined on the contrast-enhanced CT scans, including any mediastinal
mass and the site of SVCO (Fig. 6.4).
Figure 6.4 Axial CT slice showing anterior and posterior opposing beams created with
virtual simulation to treat SVCO (arrowed).
The CTV is chosen according to tumour type and patterns of spread. The CTV-
PTV margin is 1–2 cm. The margin is modified to spare normal lung tissue if possible.
■ Dose solutions
3D conformal planning can be used to treat the PTV and spare as much normal
lung and spinal cord as possible. Conventionally, treatment is given with anterior
and posterior beams with MLC or lead shielding.
■ Dose fractionation
For some chemo-sensitive tumours, a single fraction of 4 Gy in conjunction with
chemotherapy may give adequate immediate palliation.
Bone pain
Pain from bone metastases which persists in spite of analgesia can be successfully
treated by radiotherapy with good relief in 80 per cent of cases. The commonest
tumours to metastasise to bone are prostate, breast and lung cancers, but bone
metastases may occur from any primary tumour site. Assessment of metastatic bone
pain requires a full evaluation of the sites of metastases by isotope bone scanning or
68
Bone pain
MRI to determine whether local or systemic therapy is appropriate. If the cortex of
the bone is eroded and there is risk of fracture, or if the bone has already fractured,
surgical stabilisation should be performed followed by postoperative radiotherapy.
Isotope therapy of diffuse prostatic cancer bone metastases may be considered.
Immobilisation is individualised to the patient and the site of the bone metastases
to be treated. Most sites can be treated with the patient supine, except vertebral
lesions which are ideally treated with the patient prone. This is especially relevant
for cancers with a long natural history where a possible need for re-treatment
makes skin sparing desirable, particularly over the sacrum. Ankle stocks and head
rests can be used to aid immobilisation. Lesions in the upper cervical spine are best
treated with the patient supine, immobilised in a thermoplastic shell so that
opposing lateral beams can be used to avoid irradiating the oral cavity and
pharynx. Patients who are to be treated with electron or orthovoltage applicators
can be immobilised supine, prone or on their side.
The area to be treated is planned using a virtual CT or conventional simulation
with reference to diagnostic X-rays, bone scans, CT, MRI and sites of symptoms.

The origin of the pain must be ascertained to ensure the correct site is treated. For
example knee pain may radiate from the hip, femur or spine, and rib pain may
radiate from the vertebral body. The volume chosen must balance symptom relief
with sparing of normal tissues to minimise side effects (e.g. small bowel with pelvic
treatments).
Where possible, the whole structure should be treated, e.g. a whole vertebra
with consideration for matching adjacent fields that may be required with
subsequent treatments. It is usual to include one or two vertebrae above and below
the site of involvement.
When treating a bone postoperatively, the entire prosthesis or intramedullary
nail should be covered with a margin of normal bone. This is the area most at risk
of residual tumour. In patients with multiple painful bone metastases, wide field
half-body volumes can be treated. Treatment portals are marked on the patient
with reference tattoos as a permanent record. Photographs, DRRs or simulator
films should be taken as a record and for reference for future treatment planning.
When planning the treatment volume, it is important to remember that the
beam edge represents the 50 per cent isodose and a margin must be added to
ensure the target volume is covered by the 90–95 per cent isodose.
■ Dose solutions
The majority of treatments are given with a single direct photon beam, for example
to the spine, or as opposing anterior and posterior beams, e.g. pelvis. Sites such as the
ribs can be treated with direct electron or orthovoltage beams. A single fraction of
8 Gy has been shown overall to be equivalent to higher doses. For large volumes,
situations where long-term survival is expected, or where long segments of spinal cord
are included, a fractionated course of treatment may reduce acute and late morbidity.
69
Anterior and posterior opposing beams are used for half body radiotherapy. A
lower dose is used for upper half body radiotherapy to keep lung dose within
tolerance. The prescription point for a single beam, e.g. spine, should be the depth
of the vertebral body taken from imaging and is usually between 5 cm and 7 cm
in the cervical and thoracic region and 7–8 cm in the lumbar region. The prescription
point for opposing anterior and posterior beams is the MPD. The prescription point
for electron therapy is 100 per cent on the central axis and the energy is chosen to
cover the target volume at depth by the 90 per cent isodose.
Orthovoltage beams are prescribed to Dmax at 100 per cent. A 250–500 kV beam
will give an 80 per cent isodose at a depth of 3–3.5 cm, with a relative increase in
the dose to bone compared with megavoltage and electron beams.
■ Dose-fractionation
8 Gy single fraction.
Half body EBRT

Lower 8 Gy single fraction.
Upper 6 Gy single fraction.
Haemorrhage
Bleeding from advanced tumours of the breast, bladder, bronchus and other sites
can be effectively palliated with radiotherapy. Simple arrangements such as
opposing anterior and posterior beams for treatment of the bronchus or bladder,
or small tangential beams for breast tumours, are used. The size is chosen clinically
as the smallest needed to palliate the bleeding effectively with the fewest side
effects, and may not include the whole tumour.
The following dose fractionations can be used:
Single 8 Gy fraction.
20 Gy in 5 daily fractions given in 1 week.
30 Gy in 5 fractions given in 6 weeks (6 Gy once weekly) can be used for patient
convenience where higher doses are needed.
Information sources
NICE (2008) Metastatic Spinal Cord Compression. Guideline 75. www.nice.org.uk/guidance/index
(accessed 4 December 2008).
Patchell RA, Tibbs PA, Regine WF et al. (2005) Direct decompressive surgical resection in the
treatment of spinal cord compression caused by metastatic cancer. A randomised trial. Lancet
Aug 20–26, 336: 643–8.
Rowell NP, Gleeson FV (2002) Steroids, radiotherapy, chemotherapy and stents for superior vena
caval obstruction in carcinoma of the bronchus: a systematic review. Clin Oncol 14: 338–51.
Sze WM, Shelley MD, Held I et al. (2003) Palliation of metastatic bone pain: single fraction versus
multifraction radiotherapy – a systematic review of randomised trials. Clin Oncol 15: 345–52.
70
Skin 7
Each cell type in the skin can give rise to a different type of cancer. It is convenient
to classify skin tumours into non-melanoma skin cancers (NMSC) and malignant
melanoma (MM). Secondary deposits from other cancers can also present in the
skin. This chapter covers the role of radiotherapy for NMSC (basal and squamous
cell carcinomas), MM and other rare tumours such as cutaneous lymphoma,
Kaposi’s sarcoma, angiosarcoma, and Merkel cell tumours. Radiotherapy is also
used for benign conditions such as keloids.
Non-melanoma skin cancer

Basal cell carcinoma
Radiotherapy gives cure rates for primary or recurrent basal cell carcinoma (BCC)
of 90 per cent. BCCs are defined as low and high risk:
■ Low risk BCCs are generally small (2 cm), well defined in a non-critical site
with non-aggressive histology. Only 5 per cent of well-defined BCCs 2 cm
show subclinical spread beyond 5 mm.
■ High risk BCCs are generally large (2 cm), indistinct or morphoeic, in a critical
site (eyes, ears, lips, nose and nasolabial folds), and show aggressive histology such
as morphoeic, infiltrative, micronodular or perineural spread.
In many cases BCCs can be managed equally effectively by surgery or radiotherapy.
Indications for radiotherapy include:
■ large superficial lesions where a better cosmetic result can be obtained with
radiotherapy
■ large lesions where surgery would cause major loss of function such as paralysis,
numbness, dribbling, or ectropion
■ extensive lesions where surgery may require nasectomy, ear amputation or eye
enucleation
■ older patients in whom long-term skin atrophy caused by radiotherapy may not
be relevant
■ multiple superficial lesions where surgery would be onerous for the patient
■ patients who are unfit for, or refuse, surgery
■ selected tumours of the eyelids and canthi of the eyes
■ selected tumours on the nose, ears and lips; larger lesions overlying cartilage are
best treated with electron rather than superficial radiotherapy
71
SKIN
■ large lesions on the cheek which often respond with minimum scarring
■ recurrent lesions after surgery, or with incomplete excision or perineural invasion.
Relative contraindications are:
■ patients under 45 years: there is potential for deterioration of the cosmetic
outcome over time (5–10 years) and risk of second malignancy
■ large lesions involving cartilage, bone, tendons or joints: the risk of
radionecrosis is high, and cure rates are lower
■ lesions where there is uncertainty over the histology
■ lesions that recur after radiotherapy
■ hair-bearing skin such as scalp, eyebrow and eyelashes: risk of permanent epilation
■ lesions around the upper eyelid: risk of lacrimal gland dryness and upper lid
conjunctival keratinisation
■ inner canthus lesions: risk of nasolacrimal duct stenosis
■ lesions on the lower leg, back and dorsum of the hand: poor healing and radiation
sequelae, particularly telangiectasia, pigmentation, ulceration, and atrophic scarring.
These relative contraindications need to be reviewed in each individual case because
alternative treatments may produce even more problems.
Mohs’ micrographic surgery can be used in selected patients with BCCs in critical
sites such as the eyelids, ears, lips, nose, and nasolabial folds, or morphoeic or
infiltrative histological subtype, as well as in patients with recurrent BCC, especially
after radiotherapy. Incompletely excised high risk BCCs may be treated by
re-excision, or by postoperative radiotherapy.
Not all BCCs require treatment. Aggressive treatment might be inappropriate
for patients of advanced age or poor general health, especially for asymptomatic
low risk lesions that are unlikely to cause significant morbidity. Some elderly or
frail patients with symptomatic or high risk tumours prefer treatments designed to
palliate rather than cure.
■ Assessment of primary disease

A biopsy is essential to obtain a histological diagnosis. The six clinicopathological
subtypes of BCC are: nodular, pigmented, cystic, morphoeic, superficial and linear.
BCC rarely metastasises, and a staging work-up is not necessary. However the
extent of recurrent aggressive and neglected BCCs may be delineated by MRI.
The primary tumour should be examined under a bright light. By palpation and
using a magnifying glass, the edges and depth of the tumour are defined. Edges of
morphoeic lesions are difficult to define due to their wide area of spread. Deep
penetration may occur at the inner canthus where tumour may infiltrate along the
medial border of the orbit and also at the nasolabial fold, ala nasi, tragus and post
auricular areas. MRI may be useful to define extensive lesions.
The majority of skin radiotherapy is based on clinical definition of the treatment
volumes and the use of single superficial X-ray or electron beams. In very advanced
cases with deep infiltration, a CT-planned photon or electron treatment may be
needed.
72
Immobilisation
The patient is positioned supine, prone or semi-prone so that the tumour to be
treated can be accessed by the superficial X-ray machine or electron applicators.
Head rests, pillows, sandbags and other supports are used to aid immobilisation as
necessary. If the patient requires a plan to treat an extensive tumour, immobilisation
will be similar to that for a head and neck cancer using a Perspex shell.

The GTV is defined clinically as described above and marked on the skin. The
margin added to the GTV to create the CTV depends on the clinicopathological
type of BCC, the site and size of the lesion being treated, and organs at risk.
A further margin for set-up error is added to create the PTV. The field size is
chosen to ensure the PTV receives 95 per cent of prescribed dose and will vary
between superficial and electron beam radiotherapy.
■ For a low risk small BCC with a well-defined GTV, a margin of 5–8 mm to
create the treatment field using superficial radiotherapy is appropriate.
■ For high risk, larger or poorly defined lesions such morphoeic BCCs, a margin
of 1–1.5 cm may be necessary to create the treatment field using superficial
radiotherapy.
This margin needs to be increased when electron therapy is used to allow for the shape
of the isodoses. This will depend on the size of the lesion and the energy of the
electron beam. For a 6 MeV electron beam treating a 5 cm circle, an extra 1 cm should
be added to the margins above to define the field size with an electron applicator.
Shielding
Superficial radiotherapy
Lead shielding is used to define the treatment field and protect surrounding
structures. The superficial machine applicators are applied directly to the skin or
standard lead cut-outs may be used. Irregular lesions need individualised lead cut-
outs which on the face can be made into lead masks (Fig. 7.1). The thickness of
lead depends on the energy of the beam used: 1.5 mm is adequate for 90–150 kV.
(a) (b)
Figure 7.1 Lead mask with area cut out for treatment of (a) BCC of nose for superficial
radiotherapy and (b) squamous cell carcinoma of the scalp with wax bolus for electron
treatment.
73
SKIN
When treating lesions of the eyelids a lead shield must be used to protect the eye.
Internal lead contact lenses are available in various sizes and shapes (Fig. 7.2a).
They are inserted after instillation of local anaesthetic eye drops. The eye must be
protected following treatment until the local anaesthetic wears off and the corneal
reflex returns. Alternatively, a spade-shaped eye shield may be used under the lower
or upper eyelid (Fig 7.2b). An intranasal shield is used to protect the mucosa and
cartilage of the nasal septum. The gums can be protected with an internal lead
shield when lesions on the skin above the upper lip are treated, but it is important
to ensure the shield does not alter the shape of the area, causing stand-off.
(a) (b)
(c)
Figure 7.2 (a) Internal lead eye shield. (b) Spade-shaped eye shield. (c) Electron eye
shield.
Electron beam radiotherapy

Electron beams may be defined by an electron endplate cut-out inserted into the
electron applicator or by shaped lead placed on the skin. 4 mm of lead is adequate
for electrons up to 10 MeV. The lead shields for electrons need to be lined with
wax or plastic on the inner surface to absorb secondary electrons. Internal eye
contact lenses have been designed for use with electron beam therapy around the
eye and are made of 3–4 mm lead with a 2–3 mm silicon lining depending on the
electron energy used (Fig. 7.2c).
74
■ Dose solutions
Superficial radiotherapy
Most superficial lesions are treated with 80–150 kV with appropriate filtration
which defines the beam characteristics. Percentage DDs for different field
diameters and energies are obtained from tables drawn up for each therapy unit,
and an appropriate energy is selected to encompass the target volume within the
90 per cent isodose (Table 7.1).
Table 7.1 Sample percentage depth dose data for 80 kV (HVL 2.0 mm aluminium)
15-cm SSD
Applicator (cm) 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Equivalent 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
diameter (cm)*
BSF† 1.06 1.09 1.11 1.12 1.14 1.15 1.16 1.17 1.18
Depth (cm)
0 100 100 100 100 100 100 100 100 100
0.5 76 78 80 81 82 83 84 84 85
1.0 58 61 64 66 68 69 71 71 72
2.0 37 40 42 44 46 48 49 50 51
3.0 24 26 28 30 32 33 34 35 36
4.0 16 18 19 21 22 23 23 24 25
5.0 11 12 13 14 15 16 17 17 18
6.0 8 9 10 10 11 12 12 13 13
7.0 6 6 7 7 8 8 9 9 10
8.0 4 5 5 5 6 6 7 7 7
9.0 3 3 4 4 4 5 5 5 6
10.0 2 2 3 3 3 4 4 4 4
25-cm SSD
Applicator (cm) 6.0 8.0 10.0 12.0 15.0 8 10 10 15
Equivalent 6.0 8.0 10.0 12.0 15.0 9.9 13.3
diameter (cm)*
BSF† 1.20 1.22 1.24 1.25 1.26 1.24 1.26
Depth (cm)
0 100 100 100 100 100 100 100
0.5 87 88 89 90 90 89 90
1.0 76 78 79 80 81 79 80
2.0 57 59 60 61 62 60 61
3.0 42 44 45 46 48 45 47
4.0 30 32 34 35 37 34 36
5.0 22 25 26 28 29 26 28
6.0 17 19 21 22 23 21 22
7.0 13 15 16 17 19 16 18
8.0 10 12 13 14 15 13 14
9.0 8 9 10 11 12 10 11
10.0 6 7 8 8 9 8 9
* Equivalent diameter of cut-out treatment area.

†
BSF, back scatter factor.
75
SKIN
Owing to curving body contours, it may not be possible to appose the applicator
of the machine to the lead mask, which will result in positive or negative stand-off
(Fig. 7.3). An allowance for this stand-off must be made according to the inverse
square law. Tables for superficial X-ray therapy units are available which give
multiplication factors to correct for different amounts of both positive and
negative stand-off (Table 7.2).
Applicator Applicator
Lead mask Skin lesion
0.5 cm Lead mask
0.5 cm
(a) Skin lesion (b)

Figure 7.3 (a) Positive stand-off of 0.5 cm between lesion and applicator. (b) Negative
stand-off of 0.5 cm.
Table 7.2 Multiplication correction factors for stand-off calculation
Stand-off distance (cm) 15-cm SSD applicator 25-cm SSD applicator
1.5 1.23 1.13

1.0 1.15 1.09
0.5 1.07 1.04
0.0 1.00 1.00
0.5 0.94 0.96
1.0 0.88 0.92
1.5 0.83 0.89
2.0 0.78 0.86
2.5 0.73 0.83
3.0 0.69 0.80
3.5 0.66 0.77
4.0 0.62 0.74
4.5 0.59 0.72
5.0 0.56 0.69
Sample calculation for superficial radiotherapy

For applicator 3 cm diameter, lead cut-out 2 cm diameter, positive stand-off
0.5 cm treating at 80 kV (HVL 2.0 mm Al) and 15 cm source skin distance
(SSD) with a daily fraction size of 4.5 Gy.
Using data from tables which incorporate back scatter factor (BSF) for size of
applicator and treatment area and surface dose rates and treatment times, the
treatment time can be calculated.
76
Beam data
Surface dose rate (SDR) 2.81 Gy/minute (Table 7.3)
% DD 100 per cent (Table 7.1)
BSF (3 cm) 1.14 (Table 7.1)
BSF (2 cm) 1.11 (Table 7.1)
Stand-off correction factor (SOF) 0.94 (Table 7.2)
Calculation:
BSF (cut out) % DD

Output (Gy/min) SDR SOF (7.1)
BSF (applicator) 100
1.11 100%
Output (Gy/min) = 2.81 0.94 (7.2)
1.14 100
4.5
Treatment time per fraction 1.75 min 1 min 45 s (7.3)
2.57
When there is no stand-off and no cut-out, the treatment time in minutes can
be read directly from a table (Table 7.3).
Electron beam therapy

Beam sizes of less than 4 cm diameter should not be used because the advantage
of beam flatness is lost. The physical characteristics of electron beams are shown in
Table 7.3 Sample table for calculation of surface dose rate for a given applicator size and
treatment time in min (energy 80 kV, HVL 2.0 mm, aluminium filtration 1.7 mm)
Applicator Surface Treatment time (min) for varying doses (Gy)

size (cm) dose rate
1 4 5 6 6.25 6.3 8 9
(Gy/min)
15-cm SSD applicators

1.0 circle 2.46 0.41 1.62 2.03 2.44 2.54 2.56 3.25 3.66
1.5 circle 2.56 0.39 1.56 1.95 2.34 2.44 2.46 3.13 3.52
2.0 circle 2.64 0.38 1.51 1.89 2.27 2.36 2.38 3.03 3.40
2.5 circle 2.75 0.36 1.46 1.82 2.19 2.28 2.29 2.91 3.28
3.0 circle 2.81 0.36 1.42 1.78 2.14 2.22 2.24 2.85 3.20
3.5 circle 2.84 0.35 1.41 1.76 2.12 2.20 2.22 2.82 3.17
4.0 circle 2.87 0.35 1.39 1.74 2.09 2.18 2.19 2.79 3.13
4.5 circle 2.91 0.34 1.37 1.72 2.06 2.14 2.16 2.75 3.09
5.0 circle 2.93 0.34 1.37 1.71 2.05 2.14 2.15 2.73 3.07
25-cm SSD applicators
6.0 circle 1.03 0.97 3.89 4.87 5.84 6.08 6.13 7.79 8.76
8.0 circle 1.06 0.95 3.78 4.73 5.67 5.91 5.96 7.56 8.51
10.0 circle 1.09 0.92 3.68 4.60 5.51 5.74 5.79 7.35 8.27
12.0 circle 1.10 0.90 3.62 4.52 5.43 5.65 5.70 7.24 8.14
15.0 circle 1.11 0.90 3.59 4.48 5.38 5.61 5.65 7.18 8.07
77
SKIN
Figure 2.6 (p. 22). The energy of the electron beam is chosen so that the deep
surface of the target volume is encompassed by the 90 per cent isodose with a
sharp fall in dose beyond. This spares the underlying tissues. The effective
treatment depth expressed in centimetres is about one-third of the beam energy in
MeV but depends on the beam size and depth dose data for any particular
machine. Bolus is placed on the skin surface to increase dose to 100 per cent and
to compensate for irregular surfaces (see Fig. 7.1b, p. 73). This reduces the depth
of the 90 per cent isodose which must be taken into account when choosing the
electron energy. A correction for any stand-off between the applicator and skin
surface can be made using the inverse square law and effective SSD. Tables for
varying electron energy and applicator size should be used.
The daily dose and fractionation scheme depend on the site and size of the lesion,
and the age and performance status of the patient. The convenience of shorter
regimens must be balanced against risks of normal tissue damage.
For superficial radiotherapy, the dose is specified at Dmax and for electrons at
100 per cent on the central axis, ensuring a minimum dose of 90–95 per cent to
the whole target volume. Many different regimens have been shown to be effective
and are in widespread use.
Lesions 3 cm diameter (superficial radiotherapy [SRT] 80–140 kV)

36 Gy in 8 fractions of 4.5 Gy given in 17 days treating M/W/F.
30–32 Gy in 4 fractions of 7.5–8 Gy given in 2–4 weeks (one or two
fractions/week).
18 Gy in a single fraction.
Lesions 3 cm diameter or nose/pinna/poorly vascularised skin (SRT or

consider EBRT*)
45 Gy in 9 fractions of 5 Gy given in 21 days treating on alternate weekdays.
Lesions 5 cm diameter (EBRT* or megavoltage radiotherapy)

50–54 Gy in 20 daily fractions of 2.5–2.7 Gy given in 4 weeks.
For SCC use 54 Gy (see below).
60 Gy in 30 daily fractions given in 6 weeks.
(*Consider increasing dose by 10 per cent to account for the reduced relative
biological dose of electrons.)

The patient lies on the treatment couch with lead mask, cut-out and other
shielding such as internal eye shields in place. Any bolus required is applied and
the treatment applicator positioned over the target volume. Skin marks, tattoos if
appropriate, and photographs of the planned treatment position are used to ensure
the correct set-up each day.
Scabs over lesions may need to be removed before treatment to ensure
adequate depth dose. Skin should be kept dry, and shaving or the application
78
Squamous cell carcinoma
of make up and chemicals to the area should be avoided. Erythema usually
develops in the first week of treatment followed by an exudative reaction. Acute
erythema is treated with aqueous cream or soft paraffin. If the skin becomes
broken, paraffin gauze or hydrogel is applied with a dry dressing. Vaseline can be
used inside the nostril to help prevent scabbing and nose bleeds. If the dose to the
lacrimal gland is kept below 35 Gy, the risk of the late complication of dry eye may
be minimised.
Long-term side effects include atrophy, hyper- and hypopigmentation,
telangiectasia and alopecia. Non-healing skin ulceration, persistent pain and
secondary skin cancers are more serious late side effects. Patients should be advised
to avoid exposure to cold winds and sun, to use ultraviolet sun barrier cream, and
to wear a hat.
Squamous cell carcinoma

Cutaneous squamous cell carcinoma (SCC) is the second most common skin
cancer after BCC. Treatments are highly effective and can achieve cure rates of
90 per cent. In the head and neck area, 20 per cent of NMSC are cutaneous SCC,
but this rises to 43 per cent in sites such as the pinna. The common premalignant
lesion is actinic keratosis.
Primary cutaneous SCCs may grow slowly or rapidly. They may metastasise
initially to regional lymph nodes, and later to viscera, with an overall mortality of
3 per cent. Cutaneous SCCs of the head and neck can spread haematogenously to
the CNS, or via the perineural space.
Overall, lesions recur locally in 25 per cent. Risk factors for local or nodal
recurrence include site (lip and ear SCCs have a higher recurrence rate and are
discussed in Chapter 9), size (tumours 2 cm diameter), depth of invasion
(4 mm), cellular differentiation, perineural involvement, host immune status and
previous treatment. Tumours arising in non sun-exposed sites, and areas of
previous radiation, thermal injury, scarring or chronic ulceration, have higher risk
of recurrence and metastases. Poorly differentiated and anaplastic SCCs
metastasise more frequently than well-differentiated SCCs. Those on the mid-face
and lip are especially prone to neural involvement. Careful follow-up of patients
with these high-risk features is essential.

Treatment of SCC is similar to that described for BCC. However, more radical
surgery is required because of the greater metastatic potential. Patients with high
risk SCC presenting with involved lymph nodes should be reviewed by a
multidisciplinary oncology team including a dermatologist, pathologist, plastic or
maxillofacial surgeon, oncologist and clinical nurse specialist.
Radiotherapy is generally reserved for patients over 45 years because of the
theoretical risk of inducing further malignancies. It is not suitable for tumours
invading underlying cartilage where the risk of radiochondritis is high and cure
rates are lower. There is a relative contraindication to treating SCCs in cardiac or
renal transplant patients as they may be particularly susceptible to further
79
SKIN
cutaneous malignancies. The 5-year cure rate for NMSC with radiotherapy is as
high as 90 per cent and the cosmetic results are good or acceptable in 84 per cent.
The early and late complication rates are low. Radiotherapy is often used as an
adjuvant modality for high risk SCC, e.g. lesions over 2 cm with perineural
invasion in a high risk site. Radiotherapy may also be used palliatively for patients
with lymph node metastases.
Afterloading brachytherapy can be used for SCCs on the dorsum of the hand,
lower limb or curved surfaces such as the scalp. A mould of the area to be treated
is made and catheters distributed over the area to be treated following the
Manchester or Paris rules for an interstitial implant.
Methods of assessment and target volume definition are similar to those
described for BCC. However, a larger margin of 1–2 cm is added to the GTV to
create the PTV according to risk factors. Dose-fractionation regimens are similar
to those described for BCC, but for larger lesions, longer dose-fractionation
schedules are preferred.
Lesions 5 cm diameter (see also dose-fractionation regimens for BCC)
45 Gy in 9 fractions of 5 Gy given in 21 days treating on alternate weekdays.
54 Gy in 20 fractions of 2.7 Gy given in 4 weeks.
Lesions 5 cm diameter (EBRT* or megavoltage radiotherapy)

54 Gy in 20 daily fractions of 2.7 Gy given in 4 weeks.
66 Gy in 33 daily fractions given in 6 1⁄2 weeks.
Postoperative radiotherapy (EBRT* or megavoltage radiotherapy)
HDR brachytherapy
A typical fractionation is 45 Gy in 10 fractions. A more prolonged fractionation
may be advisable in the lower limb.
Palliative radiotherapy
36 Gy in 6 fractions of 6 Gy once weekly given in 6 weeks.
(*Consider increasing dose by 10 per cent to account for the reduced relative
biological dose of electrons.)
Malignant melanoma
The primary treatment of melanoma is surgery. Radiotherapy has a primary role in
treating in situ lentigo maligna, an adjuvant role for postoperative lymph node
areas and a palliative role for skin, nodal and visceral metastases. In vitro studies
have shown a wide shoulder to the cell survival curve for melanoma cell lines
suggesting a possible advantage for hypofractionation. This has to be balanced
against the risk of increased normal tissue reactions with larger fractions.
80
Malignant melanoma
Lentigo maligna is the superficial in situ phase that progresses to lentigo maligna
melanoma in 30–40 per cent of cases. It is typically a large flat, pigmented area most
commonly on the face. The treatment is surgical excision or radiotherapy.
Radiotherapy is very effective with a recurrence rate of 7 per cent and is a good
option for older patients. The lesions may be treated with superficial or electron
beam radiotherapy as described for NMSC. The margins need to be carefully
defined, as there is often a large area of subclinical disease requiring a margin of up
to 2 cm from GTV to PTV. The dose used is the same as for NMSC described above.
There is a 30–50 per cent risk of recurrence following lymph node dissection for
melanoma where there is extracapsular extension, three or more involved lymph
nodes, lymph node size 3 cm, cervical neck location or recurrence after previous
excision. Adjuvant radiotherapy can reduce the risk of local recurrence by over
50 per cent in these high risk cases, but has not been shown to improve survival.
There is a high risk of lymphoedema when treating the axillae and groins
adjuvantly and radiotherapy is usually avoided in these sites. Radiotherapy to the
nodal regions is planned as described in Figure 23.2 (p. 288).
Radiotherapy may be used palliatively to relieve pain and bleeding from skin,
nodal, bone and visceral metastases. Brain metastases can be palliated in patients
with a good performance status. The recursive partitioning analysis (RPA) classification
is helpful in selecting patients for treatment.
■ RPA class 1: age 65, Karnovsky performance status (PS) 70, controlled
primary, no extracranial metastases
■ RPA class 2: all others
■ RPA class 3: Karnovsky PS 70
Patients in RPA class 1 may benefit from surgical excision and postoperative
whole brain radiotherapy, or focal stereotactic radiotherapy. Patients in RPA class
3 and those with meningeal involvement have such a poor survival that they are
best managed with palliative care only. Patients in RPA class 2 may benefit from
palliative whole brain radiotherapy.
Lentigo maligna
See dose-fractionation regimens for NMSC above. Adjuvant radiotherapy after
lymph node dissection:

60 Gy in 30 fractions given in 6 weeks.
36 Gy in 6 fractions of 6 Gy once weekly given in 6 weeks.
Whole brain radiotherapy

12 Gy in 2 daily fractions given on consecutive days.
81
SKIN
Cutaneous lymphoma
Primary cutaneous lymphomas are rare with an incidence of 0.4 per 100 000, but
most are low grade with long survival and therefore the prevalence is much higher.
Two thirds are T cell in origin, the majority of which are mycosis fungoides (MF).
The WHO/EORTC (Willemze et al. 2005) classification is now used and
specialist multidisciplinary teams should manage these patients.
■ T cell
Radiotherapy is a very effective single agent for the treatment of MF. It is used in
every stage to treat patches and plaques, tumours, and lymph nodes. MF is
extremely sensitive to radiotherapy and low doses can be used, allowing adjacent
areas to be treated, and recurrences can be retreated safely.
In early stage IA–IIB MF, radiotherapy is used with skin directed therapy such
as psoralen ultraviolet A (PUVA) to treat patches and plaques which are planned
for treatment in a similar way to NMSC as described above. The margins from
GTV to field edge depend on the area being treated and are usually 0.5–1.0 cm.
In stage IIB–IVB MF, radiotherapy can be used alone or with systemic therapies
to treat skin patches, plaques and tumours, nodal and visceral metastases. Mucosal
involvement of the nasopharynx or pharynx responds well to radiotherapy, which
is fractionated to reduce normal tissue toxicity.
Total skin electron beam therapy

Total skin electron beam therapy (TSEBT) can be used to treat the whole skin in
any stage of MF. It is used in stage IB disease that is becoming resistant to skin
directed treatment such as PUVA, in stage IIB disease to debulk tumours, in stage
III disease to treat erythroderma, and palliatively in stage IV disease where patients
have good PS. The current standard is the modified Stanford technique (Fig. 7.4)
using a conventional linear accelerator in high dose rate electron mode to deliver
matched dual fields at a distance of 3.5–4.5 m. A Perspex screen is placed close to
the patient to degrade the beam to meet the EORTC consensus of dose maximum
at 1 mm depth, 80 per cent isodose at 9 mm depth and 20 per cent isodose at
20 mm. The patient is treated standing and rotates through the six positions
shown to maximise unfolding of the skin. On days 1 and 5 TLD measurements are
taken from various sites on the skin to map dosimetry accurately. Areas of low dose
and inherently shielded areas such as the scalp and soles of the feet are treated
separately. Using modern techniques, the whole body photon contamination dose
is less than 2.3 per cent (0.7 Gy). Acute adverse effects of TSEBT are usually
minor with attention to care of the patient’s skin. The adverse effects include
fatigue, temporary alopecia, nail loss, leg swelling and blisters, minor nose bleeds
(1 in 30), reduced sweating (1 in 30), minor parotitis (1 in 30), gynaecomastia (1
in 30), and skin infection, which is rare (1 in 100) but must be treated
aggressively. Late effects include skin atrophy, hypothyroidism, nail and finger
changes, sun sensitivity and infertility in men. Combined with PUVA, TSEBT
adds to the patient’s risk of other cutaneous malignancies.
There are other non-MF cutaneous T cell lymphomas that can be effectively
treated with radiotherapy using similar doses to MF.
82
Cutaneous lymphoma
(a)
Normal 50% Nominal treatment plane

beam edge
Polycarbonate Beam central

energy axis, 100% dose
degrader
Electron source
50% beam edges

coincide at
35 Hinge central axis
angle
isocentre
250 cm
(b) Nominal 350 cm treatment distance
Figure 7.4 TSEBT technique used at Guy’s and St Thomas’ Hospital modified from the
Stanford technique: (a) six standing positions and (b) dual field arrangement. Courtesy of
Dr P Rudd.
83
SKIN
Dose-fractionation
Patch and plaque
8 Gy in 2 fractions of 4 Gy given in 2–4 days.
Tumour
12 Gy in 3 fractions of 4 Gy given in 3–5 days.
Mucosal disease
Lymph nodes
TSEBT
■ B cell
Primary cutaneous B cell lymphomas (PCBCL) are a heterogeneous group that
present in the skin without evidence of extracutaneous disease at diagnosis. The
EORTC WHO classification describes two indolent types: primary cutaneous
marginal zone lymphoma (PCMZL), primary cutaneous follicle centre lymphoma
(PCFCL) and two more aggressive types: primary cutaneous diffuse large B cell
lymphoma leg type (PCLBCL LT) and non-leg type. The indolent PCMZL and
PCFCL are treated with radiotherapy alone with an excellent 5-year disease-
specific survival of over 95 per cent. The more aggressive PCLBCL LT has a
5-year disease-specific survival of 50 per cent and is best treated with CHOP-R
(cyclophosphamide, vincristine, doxorubicin and prednisolone, and rituximab)
chemotherapy followed by involved field radiotherapy to the primary skin lesions
and regional lymph nodes.
If radiotherapy planning margins are inadequate, the relapse rate is much higher.
It is currently recommended that margins of 2–3 cm are added to the GTV to form
the treatment field when planning superficial radiotherapy, and margins of 3–5 cm
are added to the GTV when planning electron beam radiotherapy. Lymph node
regions are treated in the same way as nodal lymphoma.
Dose-fractionation
PCMZL and PCFCL
PCLBCL LT
Kaposi’s sarcoma
Radiotherapy can be used for classical Kaposi’s sarcoma and as hyperfractionated
accelerated adjuvant radiotherapy (HAART) with chemotherapy for human
84
Cutaneous angiosarcoma
immunodeficiency virus (HIV)-associated Kaposi’s sarcoma. It is very useful for
treatment of localised lesions and for the palliation of pain, bleeding and oedema.
Nodular localised disease can be treated with superficial or electron therapy with a
0.5–1 cm margin. More widespread lower limb skin involvement can be treated by
covering the limb in bolus (or placing it in a water bath) to deliver opposing
photon beams. Mucosal lesions (such as the mouth and conjunctiva) are best
treated with fractionated courses to avoid the severe mucosal reactions to
radiotherapy seen in patients with HIV.
Skin
15 Gy in 3 fractions of 5 Gy given in 1 week.
Mucosal
Cutaneous angiosarcoma
Cutaneous angiosarcoma of the scalp and face is a rare condition that primarily
affects elderly patients. The tumours have ill-defined margins, are often multifocal
and the scalp and face location makes complete resection difficult. It is biologically
aggressive with a high risk of metastases and the 5-year survival is 15 per cent with
a 50 per cent mortality rate at 15 months.
The primary treatment is surgical excision if possible. In cases with involved or
close margins, postoperative adjuvant radiotherapy is advised. For inoperable
disease, high dose palliative electron therapy with large margins around the
tumour produces local control and palliation. Radiotherapy has been combined
with liposomal daunorubicin and with adjuvant recombinant interleukin 2 (rIL2)
with reported success. Lesions on the scalp need to be treated with wide margins
and often require several electron fields with moving match lines to avoid overdose
when matching fields. Lesions involving the face require large electron fields with
electron eye shielding.
Adjuvant
Palliative
44 Gy in 11 fractions of 4 Gy given in 3 1⁄2 weeks
or
85
SKIN
Merkel cell carcinoma

Merkel cell carcinoma is a rare primary dermal tumour that is known to have a high
local recurrence rate, frequent nodal involvement and high risk of metastases. It most
commonly occurs on the head and neck and extremities. The overall 3-year survival is
reported as 31–62 per cent. A multidisciplinary approach to treatment is advised.
Wide local excision (WLE) of the primary tumour with a 2–3 cm margin is the initial
treatment. When the margins are not clear or the tumour is larger than 2 cm in
diameter, postoperative adjuvant radiotherapy to the tumour bed and scar using
electrons with a 3–5 cm margin should be considered. In patients unfit for surgery or
where the tumour is inoperable, radiotherapy can be used as the primary treatment.
Chemoradiotherapy gives similar local control and survival results to surgery.
The regional lymph nodes should be managed by sentinel node biopsy followed
by a completion lymph node dissection (CLND). For patients with extensive
lymph node disease, adjuvant radiotherapy following the CLND can be
considered. This is planned in the same way as regional lymph nodes for
lymphoma. Locally advanced and metastatic Merkel cell tumours can be treated
with standard palliative radiotherapy doses and chemotherapy.
Radical radiotherapy
Radical chemoradiotherapy
Adjuvant tumour bed
Adjuvant lymph nodes
Keloids
Following surgical excision of keloid scars, radiotherapy can be used to prevent the
reformation of scar tissue. Superficial radiotherapy (50–80 kV) is used to treat the
surgical scar with a narrow margin. The treatment should be planned to minimise
any scatter of radiation to normal tissues and a customised lead cut-out made. The
use of radiotherapy for benign conditions such as keloids requires caution,
especially in children and young adults given the risk of carcinogenesis. This is
particularly true for keloids overlying areas that have been shown to be at increased
risk such as the breast and thyroid.
9 Gy in a single fraction 24–72 hours after surgery.
86
Information sources
Information sources
Bichakjian C K, Lowe L, Lao CD et al. (2007) Merkel cell carcinoma: critical review with guidelines
for multidisciplinary management. Cancer 110: 1–12.
Botwood N, Lewanski C, Lowdell C (1999) The risks of treating keloids with radiotherapy. Br J
Radiol 72: 1222–4.
British Association of Dermatology. Patient information and clinical guidelines. www.bad.org.uk.
Derm IS. Dermatology Information System. www.dermis.net.
Holden CA, Spittle MF, Jones EW (1987) Angiosarcoma of the face and scalp, prognosis and
treatment. Cancer 59: 1046–57.
Motley R, Kersey P, Lawrence C (2002) Multiprofessional guidelines for the management of the
patient with primary cutaneous squamous cell carcinoma. Br J Dermatol 146: 18–25.
Schlienger P, Brunin F, Desjardins L et al. (1996) External radiotherapy for carcinoma of the eyelid.
Report of 850 cases treated. Int J Radiat Oncol Biol Phys 34: 277–87.
Senff NJ, Hoefnagel JJ, Neelis KJ et al. (2007) Results of radiotherapy in 153 primary cutaneous
B cell lymphomas classified according to the WHO-EORTC classification. Arch Dermatol 143:
1520–6.
Telfer NR, Colver G, Bowers PW (1999) Guidelines for the management of basal cell carcinoma. Br
J Dermatol 141: 415–23.
Whittaker SJ, Marsden JR, Spittle M et al. (2003) Joint British Association of Dermatologists and UK
Cutaneous Lymphoma Group guidelines for the management of primary cutaneous T-cell
lymphomas. Br J Dermatol 149: 1095–107.
Willemze R, Jaffe ES, Burg G et al. (2005) WHO-EORTC classification for cutaneous lymphomas.
Blood 105: 3768–85.
87
8 Head and neck:
general
considerations
Conformal volume-based radiotherapy of head and neck cancers requires
knowledge of anatomy and patterns of spread of disease, which are often specific
to each tumour site. This chapter explains the common principles of treatment of
these tumours.
Radiotherapy alone with daily 2 Gy fractions is no longer regarded as standard
for locally advanced head and neck cancer. Altered fractionation or the addition of
chemotherapy or targeted agents improves outcomes in patients able to tolerate a
more intensive approach.
Initial patient assessment

Head and neck tumours and their treatments can cause complex anatomical and
functional deficits. A thorough initial assessment of tumour and patient factors
including function, comorbidity and personal preference is essential to choose the
optimal treatment pathway. The ideal forum for this assessment is a multidisciplinary
clinic where surgeon and oncologist assess the patient together with input from a
clinical nurse specialist, dietician, speech and language therapist and restorative
dentist.
The extent of the primary tumour should be clearly recorded in the patient
record with the aid of diagrams and photographs. This can be especially useful if
postoperative radiotherapy is later recommended. Both sides of the neck should be
examined and any palpable lymph nodes recorded with a measurement of their size
and position.
Assessments by a dietician and a speech and language therapist are important to
document initial functional problems and to plan support through radiotherapy
and surgery.
Smoking and alcohol abuse are the two principal causes of head and neck tumours,
and their role in cardiovascular and respiratory diseases means patients often have
comorbidity.
Cross-sectional imaging to document local tumour extent and assess nodes is
recommended in all but very early vocal cord tumours. Histological confirmation
should be obtained by fine needle aspiration of lymph nodes or by incisional
biopsy of the primary tumour or nodes. Tissue samples should be reviewed by a
specialist head and neck pathologist.
88
Indications for radiotherapy including
sequencing with surgery
Although the primary site and involved or at-risk cervical lymph nodes are usually
treated with the same modality, it is useful to consider the indications for radiotherapy
for the primary site and nodes separately.
■ Primary tumour – curative treatment

If a tumour is technically resectable with clear margins, local control rates with non-
surgical therapy can never exceed those with surgery. However, it is important to
consider not only tumour control but also long-term function – particularly
swallowing and speech. A radiotherapy-based approach can provide equivalent local
control rates but better long-term function, as long as there is careful follow-up, so
that salvage surgery can be used if tumours recur. Improvements in radiotherapy
with more conformal treatments, altered fractionation and the addition of
chemotherapy or molecular agents mean that radiotherapy is the treatment of
choice for many patients with head and neck cancer. The indications for primary
radiotherapy as opposed to primary surgery are considered in site-specific chapters.
■ Primary tumour – adjuvant treatment

After a curative resection the surgeon, pathologist and oncologist should meet to
discuss the role and extent of adjuvant radiotherapy. For each patient, the most
likely sites of residual disease or recurrence can be specifically targeted with modern
radiotherapy techniques. The clearest indication for adjuvant radiotherapy is where
resection margins are positive and further surgery is not possible. It should also be
considered when factors predicting local recurrence after surgery are present,
including locally advanced tumours (usually T 3/4), close resection margins
(5 mm), high grade and perineural or vascular invasion. A clinicopathological
discussion is especially important when a laser excision has been carried out. The
piecemeal excision of a tumour and frozen section analysis of radial margins will
preclude measurement of margins of excision, which is the most useful indicator
of the need for adjuvant radiotherapy.
■ Cervical nodes – prophylactic treatment

Historical series of neck dissections or observational follow-up provide the best
evidence for estimating the risk of recurrence in clinically negative neck nodes. If
the risk is 20 per cent the relevant nodal levels should be removed surgically or
irradiated prophylactically. Modern imaging techniques are more likely to discover
enlarged nodes and stage patients N so patients may be upstaged when compared
with historical controls. The AJCC TNM (6th edition, 2002) staging for head and
neck nodes – excluding nasopharyngeal and thyroid cancer – is shown in Table 8.1.
Selecting the appropriate nodal levels depends on a thorough knowledge of
lymph node drainage pathways of the head and neck as well as data from previous
series of patients found to have nodal metastases when clinically N0.
The choice of surgery or radiotherapy to treat the N0 neck usually depends on
the treatment of the primary tumour.
89
HEAD AND NECK
Table 8.1 AJCC TNM (6th edn, 2002*) nodal staging for head and neck cancer except
nasopharynx and thyroid
Nx Regional lymph nodes cannot be assessed

N0 No regional lymph node metastasis
N1 Metastasis in a single ipsilateral lymph node,
3 cm in greatest dimension
N2a Metastasis in a single ipsilateral lymph node 3 cm and
N2b Metastasis in multiple ipsilateral lymph nodes, all
N2c Metastasis in bilateral or contralateral lymph nodes, all
N3 Metastasis in a lymph node 6 cm in greatest dimension
*With permission.
■ Cervical nodes – curative treatment

When staging indicates lymph node involvement the neck is treated with a neck
dissection, radiotherapy or a combination of the two. Radical dissection of the
neck, where all nodal levels are resected with removal of the internal jugular vein,
accessory nerve and sternocleidomastoid muscle, is being replaced by more selective
approaches sparing these structures and removing nodal levels at high risk of
containing tumour. Surgical and CT-based descriptions of lymph node levels are
similar though not identical.
If radiotherapy is to be used as treatment of the primary tumour where there is N1
neck disease, radiotherapy alone to selected nodal levels is adequate. For N2 and N3
disease, particularly where nodes are 3 cm in diameter, a combination of surgery
and radiotherapy has been recommended. Surgery followed by radiotherapy has the
advantages of quickly obtaining local control of disease (useful if there is a rapidly
growing mass or skin involvement) and providing definitive staging information. But
radiotherapy volumes are more difficult to define with certainty postoperatively than
in the unoperated neck and there is increasing evidence that (chemo)radiation alone
can control neck disease, particularly when involved nodes are smaller than 3 cm at
diagnosis. We recommend initial radiotherapy to selected nodal levels in the neck
with consideration of a selective neck dissection 3 months after radiation if there is
evidence of residual nodal disease. There is increasing data that PET-CT is most
useful in providing this evidence as pathological data have shown that residual
lymphadenopathy clinically or on CT or MRI may not contain viable tumour.
■ Cervical nodes – adjuvant treatment

If an initial neck dissection is carried out adjuvant radiotherapy is recommended if
there is macroscopic residual disease (e.g. nodes dissected off the carotid artery),
microscopic extracapsular nodal spread or if two or more nodes contain tumour.
Adjuvant radiotherapy should also be considered if a single involved node is 3 cm
in greatest diameter.
■ Palliative treatment
It is often challenging to select patients for a palliative approach when they present
with locally advanced disease that is theoretically curable with surgery and or
radiation. When the chance of cure is low or potential cure would entail significant
morbidity – for example when a total glossectomy or laryngopharyngectomy is
90
Dose-fractionation
proposed or when comorbidity precludes optimal curative approaches – palliative
radiotherapy may provide control of local symptoms such as pain and airways
obstruction. Palliative radiotherapy to the primary site can also be useful when
metastases are present at diagnosis, or in locally recurrent disease to ameliorate
fungating tumour or reduce bleeding.
Dose-fractionation
For many years, daily 2 Gy fractions, 5 days a week, have been regarded as the
standard of care in curative radiotherapy for head and neck cancer. In the UK, 66 Gy
has often been prescribed to the primary and involved nodes but 70 Gy is the standard
dose in most clinical trials and international centres and is recommended here.
In adjuvant radiotherapy 60 Gy has been the standard dose with 66 Gy if there
are positive resection margins or extranodal spread.
The prophylactic dose to uninvolved nodal levels has often been 50 Gy. We
recommend 44 Gy, as supported by the control arm of the CHART trial and other
clinical series, because recurrence rates are equally low and there is no need to
match a posterior electron field.
There is now good evidence that improved local control and cure rates can be
achieved by using altered fractionation regimens, by combining drugs with
radiation, or possibly by a combination of these approaches. The choice depends on
the resources and philosophy of the treating clinician, but for locally advanced
(T3/4 or N) head and neck cancer in patients able to tolerate more aggressive
therapy, one of these methods should be used.
Each of these approaches not only improves local control but also increases
acute and/or late side effects, suggesting that they make dose escalation possible
rather than improving the therapeutic ratio. Careful collection of data on side
effects is as important as local control rates in assessing outcomes.
Refining target volumes in the context of increasingly conformal radiotherapy will,
on the other hand, improve the therapeutic ratio. Better imaging techniques and
clinical expertise lead to more accurate selection and definition of target volumes,
and improvements in treatment delivery such as IMRT lead to more conformal dose
solutions. This allows the same dose to be delivered to the tumour with reduced side
effects or may make dose escalation possible without increasing side effects.
■ Accelerated radiotherapy
Accelerated radiation schedules shorten the overall treatment time to reduce tumour
repopulation during a course of radiotherapy and theoretically increase local control
or cure. Examples of pure accelerated techniques include using 6 fractions per week
(DAHANCA) or a concomitant boost schedule where the smaller second phase
volume is treated at the same time as the larger prophylactic volume rather than after
it (Fig. 8.1). Between fractions given on the same day there should be a gap of at
least 6 hours to allow normal tissues to repair sublethal damage.
Individual trials and meta-analyses have shown an improvement in local control
with accelerated regimens of 7–10 per cent but an absolute survival benefit has
been harder to demonstrate. Acute effects are increased so careful support through
radiotherapy is necessary. There is also evidence that late effects are increased.
91
HEAD AND NECK
fraction of radiotherapy to prophylactic volume and proven disease
fraction of radiotherapy to proven disease alone
x drug dose
(a)
(b)
(c)
(d)
(e)
(f)
x x x
(g)
x x x x x x x
Figure 8.1 Schematic examples of fractionation and radiochemotherapy schedules for
head and neck cancer. (a) Conventional 2 Gy/fraction. (b) Accelerated DAHANCA – 6
fractions/week. (c) Concomitant boost. (d) Hyperfractionation (1.2 Gy/fraction).
(e) GORTEC very accelerated regimen (2 Gy/fraction). (f) Radiochemotherapy with
3 weekly cisplatin. (g) Radiochemotherapy with weekly cisplatin or cetuximab.
The GORTEC group has used a very accelerated regimen of twice daily 2 Gy
fractions throughout a course of treatment and shown that 62–64 Gy over 32–33
days produces better local control and equivalent survival to 70 Gy in 35 daily
fractions but with more acute toxicity.
92
Radiotherapy with chemotherapy
■ Hyperfractionation
Reducing the dose per fraction (usually 1.0–1.2 Gy) and using two fractions per
day should, according to radiobiological principles, reduce the risk of late effects
for the same dose and allow dose escalation with an improved therapeutic ratio.
An example is the RTOG 9003 schedule of 1.2 Gy given twice daily to a total dose
of 81.6 Gy in 68 fractions. A meta-analysis suggests improvement in local control
and overall survival of 8 per cent at 5 years with hyperfractionation. Again, acute
effects are increased with hyperfractionated regimens.
■ Altered fractionation – hyperfractionation

and acceleration
Many altered fractionation regimens are both accelerated and hyperfractionated.
The CHART protocol uses 54 Gy in 36 fractions of 1.5 Gy over 12 days and has
been shown to produce equivalent local control to 66 Gy in 33 daily fractions. The
benefit of altering fractionation needs to outweigh the total dose reduction
necessary to make acute side effects tolerable.
■ Hypofractionation
There is evidence that more hypofractionated regimens can produce local control
and survival rates comparable with the regimens described above but there are few
RCT data to support this. Examples are: 50 Gy in 16 fractions, 55 Gy in 20
fractions and 60 Gy in 25 fractions. A shorter overall treatment time will reduce
the risk of tumour repopulation at the cost of a theoretical increase in late effects
due to the higher dose per fraction. The small volumes in early laryngeal cancer
make this less of a concern and 55 Gy in 20 fractions is used for T1/2 N0 laryngeal
tumours in many centres.
■ Palliative
When the goal of treatment is to improve symptoms and quality of life, a short
course of radiotherapy with minimal side effects is ideal. Several regimens have
been advocated including 20 Gy in 5 daily fractions and 30 Gy in 5 fractions (treating
twice a week). Such regimens improve symptoms in between 50 and 70 per cent
of patients and responses typically last for some months.
Radiotherapy with chemotherapy

■ Concomitant radiotherapy and chemotherapy
Chemotherapy given during a course of conventionally fractionated radiation
improves survival compared with radiation alone. A meta-analysis confirms an
absolute improvement in survival of 8 per cent with various chemotherapy
regimens in the curative setting. Cisplatin-based chemotherapy is particularly effective.
Carboplatin is substituted in renal impairment or if cisplatin is not tolerated.
Concomitant chemotherapy increases the acute side effects of radiation – particularly
the intensity and duration of mucositis. There is less good evidence to quantify late
effects of combined regimens.
93
HEAD AND NECK
Cisplatin-based chemotherapy is most commonly used with a small weekly dose

(30–40 mg/m2) or larger dose every 3 weeks (100 mg/m2). Regimens including
5-fluorouracil may increase mucositis. There are theoretical and practical advantages
to a weekly schedule: the chemotherapy is present for more fractions, there are
fewer side effects, and it is easier to omit chemotherapy if acute toxicity is a
problem. The chemotherapy is usually given early in the week and before that day’s
fraction of radiotherapy, but there is no evidence that the timing of chemotherapy
is critical.
■ Concomitant radiotherapy with cetuximab

Cetuximab is a humanised monoclonal antibody against the epidermal growth
factor receptor (EGFR) which is overexpressed in many head and neck cancers. An
RCT adding cetuximab to conventional radiotherapy showed improvements in
overall survival and local control comparable with that obtained with
radiochemotherapy. Mucositis is not increased but cetuximab does cause an
acneiform rash which can be severe.
■ Concomitant radiotherapy and hypoxic sensitisers

Hypoxic cell sensitisers attempt to mimic the effects of oxygen in fixing radiation-
induced DNA damage in tissues thereby increasing cell kill. The DAHANCA 5-85
trial combined EBRT with nimorazole, and showed an improvement in local
control compared with EBRT alone.
■ Altered fractionation with concomitant

chemotherapy
Altering fractionation or adding cisplatin or cetuximab to conventional EBRT gives
similar improvements in local control and overall survival with increased acute
toxicity and no good evidence of worsening late toxicity. Radiochemotherapy is the
most frequently used of these strategies as it requires less resource, at least for the
delivery of EBRT. The next step is to see if these approaches can be combined to
improve the therapeutic ratio and therefore local control and cure rates.
■ Induction chemotherapy
The excellent (50–70 per cent) response rate of head and neck cancers to cisplatin-
based chemotherapy makes the idea of induction chemotherapy, given before
radiation, attractive but the evidence shows only a non-significant 2 per cent
improvement in overall survival. Chemotherapy can be started quickly, can reduce
the volume of disease and theoretically make curative EBRT more likely to succeed,
and may reduce the rate of distant metastases.
The side effects of induction chemotherapy mean that a proportion of patients will
be less well equipped to deal with the toxicity of subsequent radiochemotherapy.
There is also a concern that increasing the overall duration of anticancer therapy may
lead to tumour repopulation and may negate any benefit of smaller treatment
volumes. Radiochemotherapy as the gold standard for definitive treatment must
not be compromised by induction chemotherapy.
94
3D data acquisition
Brachytherapy
Brachytherapy can be used as definitive treatment for small tumours of the lip, oral
tongue, floor of mouth and buccal mucosa, when expertise is available. It can also be
used to boost dose to primary tumour after EBRT. Treatment with intraluminal
catheters can increase dose to the primary site in nasopharyngeal cancer. Interstitial
brachytherapy can be used to aid local control in the neck where there is extracapsular
spread.
Re-irradiation
Potentially curative re-irradiation can be given in highly selected patients with
unresectable local recurrence after previous radiotherapy or for an adjacent unresectable
second primary tumour. The disease-free interval should be at least 2 years, and the
volume as small as possible. The patient should have excellent performance status and be
fully aware of the increased risk of potentially serious late effects. IMRT can be useful to
conform dose more closely to target volumes and to minimise the volume treated to
a high dose. In some cases, though, it may be preferable to use a simpler solution, as
IMRT will irradiate a greater volume of tissue, albeit to a lower dose, and therefore
overlap more with previously irradiated volumes.
3D data acquisition
■ Immobilisation
The proximity of tumour to critical normal tissues in head and neck cancer with
relatively limited intra-fractional motion of organs means that good immobilisation
will enable smaller treatment margins and reduce side effects.
A Perspex shell constructed from a plaster cast of the patient has been regarded
as the most accurate way of immobilising the patient. This is usually constructed
with the patient supine with their head on a customised head rest and as flat as
possible to maintain the spinal cord parallel to the couch top. Some patients,
particularly those with excessive secretions, need to have their head more elevated.
The shell should be fixed to the couch top in at least five places to reduce
movement (Fig. 8.2). A mouth bite can be used to push the hard palate away from
the treated volume in oral cavity tumours or to depress the tongue when sinonasal
tissues are treated. Grip bars at the side of the couch may help to pull the shoulders
inferiorly. Once anterior and lateral reference marks have been made on the shell,
selected parts can be cut out to reduce skin dose in regions where full dose to
the skin is not required. An anterior midline tattoo below the inferior extent of
the shell is useful to improve shoulder alignment. Modern thermoplastic materials
can immobilise patients to a similar degree of accuracy and have a skin-sparing
effect.
All immobilisation solutions can induce anxiety in patients prone to
claustrophobia. The skill of the radiographers and technicians can help alleviate this
but some patients need benzodiazepines or hypnotherapy to relax them in order
to construct a mask that will be tolerable and practical.
95
HEAD AND NECK
Figure 8.2 Perspex

immobilisation shell fixed to
the couch top in six places.
Each department should assess the random and systematic errors of their
immobilisation system to determine the margin to be added from the CTV to the
PTV. A margin of 3 mm should be achievable with high quality immobilisation.
■ CT scanning
Ideally, CT slices should be 3 mm (but no more than 5 mm) thick to aid accurate
target volume definition and to produce good quality DRRs for verification.
Intravenous contrast will highlight the internal jugular vein and may help to define
involved lymph nodes more accurately but is not essential when diagnostic imaging
is available. Lateral and midline reference marks are drawn on the shell.
Fusing MRI and CT scans can help to define both tumour volumes and critical
normal tissues (e.g. optic chiasm, lacrimal glands) particularly in sinonasal (Fig. 8.3)
and oropharyngeal tumours or those involving the skull base.

General principles for target volume definition which can be applied throughout
the head and neck are defined here.
■ GTV
The GTV is defined as the primary tumour and any lymph nodes over 10 mm in
short axis dimension or smaller nodes with necrotic centres or rounded contours
thought to contain tumour. If induction chemotherapy has been used, the post-
chemotherapy GTV is contoured, but the pre-chemotherapy extent of tumour
should be included in the CTV70.
The diagnostic images and records and photographs of initial evaluation (in
clinic or at examination under anaesthesia [EUA]) are critical to assess the extent
of the primary tumour and the exact site of involved lymph nodes. Discussion with
the surgeon performing the endoscopic assessment and with an experienced
radiologist will enable GTV to be defined as accurately as possible. A planning CT
scan done at an interval after the diagnostic one should be carefully reviewed for
evidence of residual local disease after surgery or unexpected new cervical
96
(a) (b)
Figure 8.3 Adjuvant radiotherapy of an ethmoid tumour. (a) Planning CT. (b) Fused
MRI–CT of the same axial slice to help to define volumes and critical normal structures.
Note ethmoid opacities seen on CT are shown to be postoperative secretions on MRI.
lymphadenopathy. We recommend reviewing the planning system settings with a

radiologist and defining optimal settings for each region of the body, always using
these settings when defining volumes.
Starting on a central slice of the primary tumour, the GTV is defined on each axial
CT image moving superiorly and then inferiorly. Involved nodes can then be defined
in the same way. Viewing the GTV on coronal and sagittal DRRs ensures consistency
of definition between slices so that no artificial steps in the volume are created.
A volume should not be cut and pasted onto sequential slices for risk of pasting an
error: it is more accurate to redraw the GTV on each slice. If there are slices where the
GTV cannot be defined – for example due to dental artefact – the planning software
will usually allow the missing contours to be interpolated from those either side.
■ Primary tumour CTV (CTV-T)

We recommend growing the GTV by an isotropic margin of 10 mm except where
there are natural tissue barriers on the basis that a surgical margin of 10 mm is
considered adequate for local excision. The CTV is then edited slice by slice – again
starting in the centre of the volume – to subtract air and adjacent structures such
as bone which are definitely not involved. The CTV margin can be locally
expanded by more than 10 mm if local structures such as muscle or soft tissue are
at risk of involvement (Fig. 8.4).
■ Postoperative primary tumour CTV

It is more difficult to specify guidance for CTV definition postoperatively as the
anatomy is distorted by the resection and by any reconstructive myocutaneous
flaps. A discussion between surgeon, pathologist and oncologist is important to
97
HEAD AND NECK
(a) (b)
(c)
Figure 8.4 Step-wise CTV definition for a left tonsillar cancer invading the base of tongue.
(a) GTV is grown isotropically by 10 mm to produce a CTV. (b) CTV is edited off
uninvolved bone and air. (c) CTV is expanded to cover possible invasion of the
contralateral base of tongue and to include adjacent level II nodes.
98
define the exact anatomical sites at risk of residual disease. Radio-opaque clips used
to mark close or involved margins can be helpful but should not be confused with
vascular ligation clips which may be distant to the original tumour. It is not known
whether the whole operative field should be included in the CTV but a more
individualised, selective approach seems reasonable. It is sometimes useful to
define the site of the preoperative GTV on the planning CT to orientate the
possible sites of microscopic disease but this is less helpful when the anatomy is
very different following a major resection. If there is bulky residual disease a GTV
can be defined and expanded to form the CTV as above.
■ Cervical lymph node CTV (CTV-N)

In previous decades head and neck radiation fields included all lymph nodes on
both sides of the neck for all but small or well-lateralised tumours. Surgical
evidence that a selective approach to neck dissection is as effective as more radical
operations, and the ability to create and treat relevant smaller nodal volumes, has
changed this approach.
Defining nodal CTVs requires both selection of appropriate nodal levels and
delineation of those levels on the CT dataset. Selection is covered in individual
tumour chapters as it will necessarily differ for each site and stage of disease. Often
two nodal CTVs are selected – a high risk volume containing involved nodes or
those close to the primary site and a lower risk volume containing levels thought to
be at risk of microscopic disease. These are labelled according to dose, e.g. CTV44.
■ Lymph node delineation

There are published guidelines for delineation of nodal CTVs in the node-
negative, node-positive and postoperative neck. It is very useful to have these
guidelines available when defining nodal CTVs. The consensus guidelines for the
node-negative neck are available as an online atlas and specify CT anatomy of
nodal levels in the neck. The retropharyngeal nodes, levels Ia, Ib and II–V, are
defined according to CT-based anatomical criteria which correspond closely to the
surgical definitions of nodal levels (Table 8.2; Fig. 8.5).
In the node-positive neck, two further volumes are defined. When level II nodes
contain tumour, the fatty retrostyloid space around the jugulo-carotid vessels
between the superior portion of level II and the skull base should be outlined.
A nodal space corresponding to the supraclavicular fossa is contoured when level
IV or lower level V nodes contain tumour (Table 8.3).
The nodal volumes in the positive neck are adjusted from the guidelines for the
node-negative neck to take account of the risk of extracapsular spread, usually into
adjacent muscles. Nodal size is related to the risk of extracapsular spread with
25 per cent of 10 mm nodes and 80 per cent of 30 mm nodes having breached the
capsule. Where involved nodes are large or where adjacent structures look involved
on imaging, that structure should be included in the nodal CTV. Muscle fascia
provides a barrier to tumour invasion but once breached, tumour can spread
longitudinally along muscle fibres. It is not known whether part or all of an
involved muscle should be included in the CTV (Fig. 8.6).
It should be remembered that the division into nodal levels is on anatomical and
not functional grounds. Where a low level II node contains tumour it is appropriate
99
100 Table 8.2 CT-based anatomical guidelines for the delineation of nodal volumes in the N0 neck (from Gregoire et al. 2003*)
Level Anatomical boundaries
Cranial Caudal Anterior Posterior Lateral Medial

RP Base of skull Cranial edge of Fascia under the Prevertebral muscles Medial edge of Midline
body of hyoid bone pharyngeal mucosa internal carotid artery
Ia Geniohyoid muscle, Plane tangential to Symphysis menti, Body of hyoid bone Medial edge of anterior N/A
plane tangential to body of hyoid bone platysma muscle belly of digastric muscle
basilar edge of mandible
Ib Mylohyoid muscle, Plane through Symphysis menti, Posterior edge of Basilar edge/inner side Lateral edge of
cranial edge of central part of platysma muscle submandibular gland of mandible, platysma anterior belly of
submandibular gland hyoid bone muscle, skin digastric muscle
II Caudal edge of Caudal edge of Posterior edge of Posterior edge of Medial edge of Medial edge of
lateral process of C1 body of hyoid submandibular gland, sternocleidomastoid sternocleidomastoid internal carotid
bone anterior edge of internal muscle artery, paraspinal
carotid artery, posterior muscles
edge of posterior belly
of digastric muscle
III Caudal edge of Caudal edge of Posterolateral edge of Posterior edge of Medial edge of Medial edge of
body of hyoid bone cricoid cartilage sternohyoid muscle, sternocleidomastoid sternocleidomastoid internal carotid
anterior edge of artery, paraspinal
sternocleidomastoid muscles
muscle
IV Caudal edge of 2 cm cranial to Anteromedial edge of Posterior edge of Medial edge of Medial edge of
cricoid cartilage sternoclavicular sternocleidomastoid sternocleidomastoid sternocleidomastoid internal carotid artery,
joint paraspinal muscles
V Cranial edge of CT slice including Posterior edge of Anterior border of Platysma muscle, skin Paraspinal muscles
body of hyoid the transverse sternocleidomastoid trapezius muscle
bone cervical vessels
VI Caudal edge of Sternal manubrium Platysma muscle, skin Separation between Medial edges of N/A
body of thyroid trachea and thyroid gland, skin and
cartilage oesophagus anteromedial edge of
sternocleidomastoid
*With permission.
Figure 8.5 Lateral DRR and corresponding axial planning CT slices to illustrate cervical lymph node levels Ib and II, levels III and V, and level IV.
101
HEAD AND NECK
Table 8.3 CT-based anatomical guidelines for delineation of the retrostyloid space and
supraclavicular fossa (from Gregoire et al. 2006*)
Level Anatomical boundaries
Cranial Caudal Anterior Posterior Lateral Medial
Retrostyloid Base of Upper limit Parapharyngeal Vertebral Parotid Lateral

skull (jugular of level II space body/base space edge of
foramen) of skull RP nodes
Supraclavicular Lower Sternoclavicular Sternocleidomastoid Anterior edge Lateral Thyroid

fossa border of joint muscle, skin, clavicle of posterior edge of gland/
level IV/V scalenus posterior trachea
muscle scalenus
muscle
*With permission.
Figure 8.6 Level II node with possible

extracapsular spread into
sternocleidomastoid muscle (SM). Nodal
clinical target volume is therefore extended
to cover the whole muscle.
to include both level II and level III nodes within the high dose CTV. When
a matched anterior neck field is used to treat the uninvolved low neck, nodal volumes
need not be defined. If there are involved nodes in the low neck,
nodal GTV and CTV should be contoured to ensure adequate coverage (see below).
For postoperative neck irradiation, the whole operative field is usually defined in
the CTV, particularly when there was extracapsular spread. Where nodes abutted
muscles or other structures not removed at operation, those structures should be
included in the CTV.
■ PTV
Each CTV is grown isotropically to create a PTV. By recording and analysing
systematic and random errors in a series of patients, CTV-PTV margins for each
102
Dose solutions
centre can be defined, with the assumption that there is no intra-fractional organ
motion in the head and neck. Such margins are usually 3–5 mm.
■ OAR and PVR

OAR are outlined in a similar fashion to the GTV – on serial axial CT slices.
Ideally, the tumour volumes should be hidden or turned off so that they do not
compromise OAR definition. Depending on the location of the PTV, OAR
contours will include the spinal cord, parotid glands, optic nerves and chiasm,
lacrimal glands and lenses. With more information on how DVHs relate to side
effects, contouring of other structures such as the pharyngeal constrictor muscles
may be useful in the future.
Owing to the catastrophic effect of late spinal cord damage, the spinal cord PRV
is usually defined either by adding an isotropic 5 mm margin round the cord or by
contouring the bony spinal canal as a surrogate for the cord PRV. Some centres
also recommend adding a 3–5 mm margin to the optic nerves and chiasm to create
PRVs.
Dose solutions
Dose solutions for tumour subtypes will be covered in the relevant chapters but
general principles are covered here.
■ Conventional/conformal
Anterior neck field
The standard conventional beam arrangement of opposing lateral beams or a
unilateral plan to treat the primary site and involved nodes is matched to an
anterior neck beam. With this arrangement recurrences in the low neck are
uncommon but if nodal volumes are delineated as described above, a 6 MV
anterior photon beam will not provide adequate coverage of nodal volumes (Fig. 8.7).
In most patients the risk of recurrence in the low neck is low, and we recommend
continuing with an anterior beam. The lateral border is 1 cm lateral to the
intersection of the first rib and clavicle on a posteroanterior radiograph and the
inferior border is at the inferior head of the clavicle. If the target volume is
unilateral the medial border is 1 cm from midline to avoid the cord, pharynx and
larynx. If it is bilateral, 2 cm midline shielding is added. MLC shielding is used
inferior to the clavicle to spare the apex of the lung (Fig. 8.8).
When the low neck is included in the high risk volume (e.g. involved level IV nodes
or in the high risk postoperative neck) the nodal CTV should be formally delineated.
Either anteroposterior opposing photon beams can be used (which will increase dose
to the posterior neck) or an anterior beam with dose prescribed at 3 cm depth is used
to improve coverage at the expense of a hot-spot underneath the skin surface.
Matching techniques
The common practice of using an anterior beam to treat the low neck nodes
necessitates matching at the junction with superior fields. The problems are the
divergent edges of the photon beams in different planes and the beam penumbra.
103
HEAD AND NECK
(a)
(b)
(c)
Figure 8.7 Possible beam arrangements for prophylactic treatment of low neck (level IV)
nodes. Beam edges are chosen as described in the text. (a) Anterior beam prescribed to
Dmax. (b) Anterior beam prescribed to 3 cm. (c) Anterior and posterior opposing beams.
104
Dose solutions
Figure 8.8 Standard borders of an anterior
neck beam to treat the low neck
prophylactically. The superior border is
determined by the inferior extent of the high
dose planning target volume. Sites of normal
level III (blue) and IV (pink) nodal volumes are
shown for illustration.
The most elegant solution – which deals with both problems – is to use a single
isocentre technique. Both the anterior neck beam and the plan superior to it use
only half the beam (the other half shielded by an asymmetric jaw) so that they
match at the isocentre without penumbra or beam divergence.
Another technique is to calculate angles required so that both beam edges
diverge perpendicular to the match plane. This will result in slight overlap at depth
as the width of the penumbra increases. Simpler solutions are to match the beams
at the 50 per cent (light beam) isodose or to leave a gap of 5–10 mm on the skin
surface. Both will produce a perfect match at one depth but underdose anterior to
this and overdose posteriorly where the divergent beams overlap.
The level at which the anterior beam is matched depends on the PTV. Ideally it
should be below the high dose PTV or in between nodal CTVs as matches
overlying the PTV risk underdosing disease at the match plane. If this is
unavoidable the junction can be moved by 1 cm half way through the course of
treatment to blur this match.
Electron therapy
If IMRT is not available, electron beams of 9–12 MeV may be used to treat posterior
nodes overlying the spinal cord, especially if the prescribed dose would exceed spinal
cord tolerance, i.e. if 50 Gy in 25 fractions is chosen as the prophylactic dose rather
than 44 Gy in 22 fractions. The lateral ‘bowing’ of electron isodoses at depth means
perfect matching with the photon beam is not possible so a compromise of matching
the 50 per cent photon surface isodose with the 50 per cent electron surface isodose
is usually chosen (Fig. 8.9). This is achieved by matching the edges of the two light
beams on the surface of the shell. With CT algorithms for electron beams it is
possible to view electron isodoses on a three-dimensional plan so that underdosing
at areas of high risk (e.g. involved nodes) can be avoided if possible.
Tissue equivalent bolus

When volumes are defined as described above, the nodal PTV often comes close to
the skin surface. Unless there is felt to be a risk of tumour in the epidermis or dermis,
it is preferable to accept a slight underdosing of this part of the volume rather than
use tissue equivalent bolus which will increase skin reactions significantly.
105
HEAD AND NECK
Figure 8.9 Treatment of a

high dose planning target
volume extending posterior
to the spinal cord.
Opposing photon beams
are matched to 12 MeV
electrons. Note the hot-
spot created by the lateral
bowing of electron
isodoses at depth and the
unavoidable matching
close to the GTV. Only the
left electron beam is shown
for illustration purposes.
■ Complex
IMRT is particularly useful for head and neck tumours, given the concave PTVs and
their proximity to critical structures. IMRT offers parotid sparing in oropharyngeal
and some nasopharyngeal cancers to reduce the risk of xerostomia, avoids the risk
of underdosing or overdosing at photon–electron junctions and provides a more
uniform dose distribution in sinonasal and nasopharyngeal tumours.
In the future, careful mapping of recurrent disease and improved knowledge of
the relationship between toxicity and critical organ doses should allow volume
definition to be refined so that increased dose can be given to the tumour with
tolerable side effects – thereby improving cure. This will be enhanced by the ability
to paint dose to areas of highest risk of recurrence as identified by functional
imaging or radiogenomics.
Treatment delivery and verification

■ Gaps in treatment delivery
Squamous cell cancers of the head and neck are particularly sensitive to prolongation
of a course of treatment so gaps in therapy should be avoided whenever possible. If an
unintended gap occurs, hyperfractionation is used, avoiding days when chemotherapy
is also given.
■ Verification
Several off-line and online strategies for verification and correction are described.
A commonly used protocol is for the treatment isocentre on lateral and anterior
DRRs from the CT simulation to be compared with EPIs from the treatment
machine taken on days 1–3 and weekly thereafter. An off-line correction is made
relative to the isocentre position if the mean error in any one plane is 3 mm. If
106
Patient care
there is a 5 mm error on any one day the patient should not be treated until the
error has been corrected. Re-verification in the simulator or a repeat planning CT
scan may be required.
■ Inter-fractional motion
During a course of head and neck radiotherapy, tumour shrinkage or unintentional
weight loss can occur leading to two problems.
First, immobilisation may become less effective introducing error into treatment
delivery. The mould room technician may be able to adjust the immobilisation but
if there is concern that subsequent changes in the set-up are significant (i.e. outside
the margin allowed for systematic and random errors), the planning CT should be
repeated. The original beam arrangement is then overlaid onto the new CT dataset
and corrections made as necessary. If the immobilisation cannot be corrected
either a new plan must be created with a larger CTV-PTV margin to account for
the increased motion or a new shell must be constructed. Either runs the risk of
introducing a gap in the course of treatment.
Second, a reduction in tissue volume (usually manifest as a gap between the shell
and the skin) may mean the beam penetrates more deeply towards critical normal
tissues, increasing the risk of damage. In these cases repeating a planning or cone-
beam CT can provide reassurance or enable the plan to be changed.
Patient care before and during radiotherapy

The severity of acute side effects of head and neck radiotherapy mandates
meticulous assessment and support before, during and after radiotherapy. This is
particularly important where dose escalation with altered fractionation and/or
concomitant drugs are used. Those most at risk of nutritional problems are single
men and this should be taken into account when supporting them through
treatment or even when deciding on dose-escalation approaches.
■ Dental assessment
Before radiotherapy commences, a thorough dental assessment should be carried
out and an orthopantomogram (OPG) performed to look for signs of decay in teeth
likely to be within the treated volume. Any teeth at risk of decay should be removed
to avoid the potential for radionecrosis and chronic infection that can occur when
dental work is carried out in a previously irradiated mouth or in patients with
xerostomia. Patients should be given advice about maintaining good oral hygiene.
■ Smoking
Continued smoking increases the severity of mucositis, and patients still smoking
should be offered formal cessation advice.
■ Haemoglobin
There is evidence that low haemoglobin predicts poor outcomes in head and neck
cancer. There is less good evidence that maintaining haemoglobin above 12 g/dL with
transfusions improves survival. Erythropoietin is ineffective. Anaemia may therefore be
a marker of poor prognosis rather than a correctable influence of survival.
107
HEAD AND NECK
■ Weight loss
The common acute side effects of mucositis, pain, altered taste and altered saliva all
contribute to a reduction in calorie intake during radiotherapy. Patients with pre-
existing swallowing problems (e.g. tongue base or hypopharyngeal tumours or
significant oral or pharyngeal pain) should be assessed by a specialist dietician before
radiotherapy. Losing more than 10 per cent weight during treatment is associated
with delayed recovery and increased complications as well as making immobilisation
more difficult. When a patient is felt to be at risk of losing more than 10 per cent
weight during treatment a prophylactic enteral feeding tube is recommended.
During treatment patients should be seen weekly by a dietician as part of a
multidisciplinary assessment team (with doctor or specialist radiographer and
speech and language therapist) and weighed each week. Soft and high calorie diets
and high energy protein-based supplement drinks can be useful to maintain oral
intake when it is limited by pain or dysphagia.
Speech and language therapy can be valuable during treatment to maintain
effective swallowing, to assess aspiration risk and to advise on vocal care. When
swallowing stops completely, rehabilitation after treatment is more difficult and
therefore only patients with risk of aspiration should be denied oral intake.
Patients in whom treatment includes part of the eyes (e.g. sinonasal tumours)
should be seen regularly throughout treatment by an ophthalmologist for advice
on preventing corneal damage with lubricating drops and to treat corneal abrasions
or infections promptly.
As mucosal reactions peak up to 2 weeks after therapy, particularly with accelerated
fractionation or hypofractionation, careful weekly assessment should continue until
acute effects are subsiding.
■ Pain and mucositis

Mucositis and the consequent pain is the principal side effect for most patients
having head and neck radiotherapy. Benzydamine hydrochloride (Difflam) is the
only substance shown to delay the onset of mucositis in an RCT and should be
used four to eight times a day from the start of treatment, diluted with water if
necessary. No other topical mouthwashes are of proven value and alcohol-based
washes should be avoided. Topical anaesthetic agents such as lidocaine (Xylocaine)
can be helpful if used before eating, but they have a short duration of action.
Candida infections are common and should be treated with fluconazole, or
nystatin if not severe.
Pain should be managed proactively with systemic analgesia to avoid a reduction in
oral intake as much as possible. Many patients having significant portions of their oral
cavity or pharynx irradiated require opioids. Transdermal patches can be especially
effective when oral intake is difficult.
Systemic steroids can occasionally be useful if pharyngeal or laryngeal oedema is
severe.
■ Saliva and taste
The combination of dysphagia and the build up of thick, sticky saliva is a major
problem during treatment. Good oral hygiene is important while a humid
108
Information sources
atmosphere helps keep secretions moist and easier to swallow. Nausea caused by
secretions stimulating the soft palate is best managed with 5-hydroxytryptamine
antagonists. Taste is commonly affected by radiotherapy but the mechanisms of
dysgeusia are complex and poorly understood.
■ Skin
Skin reactions can be minimised by applying aqueous cream topically four times daily
to treated areas. Tight collars should be avoided and men should avoid wet shaving.
key trials
Adelstein DJ, Li Y, Adams GL et al. (2003) 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
21: 92–8.
Bonner JA, Harari PM, Giralt J et al. (2006) Radiotherapy plus cetuximab for
squamous-cell carcinoma of the head and neck. N Engl J Med 354: 567–78.
Bourhis J, Lapeyre M, Tortochaux J et al. (2006) Phase III randomized trial of very
accelerated radiation therapy compared with conventional radiation therapy in
squamous cell head and neck cancer: a GORTEC trial. J Clin Oncol 24: 2873–8.
Bourhis J, Overgaard J, Audry H et al. (2006) Hyperfractionated or accelerated
radiotherapy in head and neck cancer: a meta-analysis. Lancet 368: 843–54.
Fu KK, Pajak TF, Trotti A et al. (2000) A radiation therapy oncology group
(RTOG) phase III randomized study to compare hyperfractionation and two
variants of accelerated fractionation to standard fractionated radiotherapy for
head and neck squamous cell carcinomas: first report of RTOG 9003. Int J
Radiat Oncol Biol Phys 48: 7–16.
Overgaard J, Hansen HS, Sprecht L et al. (2003) Five compared with six fractions
per week of conventional radiotherapy of squamous-cell carcinoma of head and
neck: DAHNACA 6&7 randomised controlled trial. Lancet 362: 933–40.
RTOG 0129: A phase III trial of concurrent radiation and chemotherapy (followed
by surgery for residual primary/N2–3 nodal disease) for advanced head and
neck carcinomas (standard fractionation v concomitant boost). In progress.
RTOG 0421: A phase III trial for locally recurrent, previously irradiated head and
neck cancer: concurrent re-irradiation and chemotherapy versus chemotherapy
alone. In progress.
RTOG 0522: A randomised phase III trial of concurrent accelerated radiation and
cisplatin versus concurrent accelerated radiation, cisplatin, and cetuximab
(C225) (followed by surgery for selected patients) or stage III and IV head and
neck carcinomas. In progress.
Information sources
Bernier J, Bentzen SM (2006) Radiotherapy for head and neck cancer: latest developments and
future perspectives. Curr Opin Oncol 18: 240–6.
109
HEAD AND NECK
Bernier J, Cooper JS, Pajak TF et al. (2005) Defining risk levels in locally advanced head and neck
cancers: a comparative analysis of concurrent postoperative radiation plus chemotherapy –
trials of the EORTC (#22931) and RTOG (# 9501). Head Neck 27: 843–50.
Cancer Care Ontario Head and Neck Cancer Evidence-based Series and Practice Guidelines [online]
(2008). Available at:
www.cancercare.on.ca/english/home/toolbox/qualityguidelines/diseasesite/head-neck-ebs/
(accessed 6 October 2008).
Corvo R (2007) Evidence-based radiation oncology in head and neck squamous cell carcinoma.
Fowler JF (2007) Is there an optimum overall time for head and neck radiotherapy? A review with
new modelling. Clin Oncol 19: 8–22.
Gregoire V, Levendag P, Ang KK et al. (2003) CT-based delineation of lymph node levels and
related CTVs in the node-negative neck: DAHANCA, EORTC, GORTEC, NCIC, RTOG
consensus guidelines. Radiother Oncol 69: 227–36.
Gregoire V, Coche E, Gosnard G et al. (2004) Selection and delineation of lymph node target
volumes in head and neck conformal and intensity-modulated radiation therapy. In: Gregoire V,
Scalliet P, Ang KK (eds) Clinical Target Volumes in Conformal and Intensity-modulated Radiation
Therapy. Springer, Berlin, pp. 69–90.
Gregoire V, Eisbruch A, Hamoir M et al. (2006) Proposal for the delineation of the nodal CTV in the
node-positive and the post-operative neck. Radiother Oncol 79: 15–20.
Gregoire V, Levendag P. DAHANCA, EORTC, GORTEC, NCIC and RTOG endorsed consensus
guidelines for the delineation of the CTV in the N0 neck of patients with head & neck squamous
cell carcinoma. [online] Available at: www.rtog.org/hnatlas/main.html (accessed 14 July 2008).
J Clin Oncol (2006) 24 – series of review articles covering controversies in head and neck cancer.
Lengele B, Hamoir M, Scalliet P et al. (2007) Anatomical bases for the radiological delineation of
lymph node areas. Major collecting trunks, head and neck. Radiother Oncol 85: 146–55.
Pignon JP, Bourhis J, Domenge C et al. (2000) Chemotherapy added to locoregional treatment for
head and neck squamous-cell carcinoma: three meta-analyses of updated individual data.
MACH-NC Collaborative Group. Meta-Analysis of Chemotherapy on Head and Neck Cancer.
Lancet 355: 949–55.
Scottish Intercollegiate Guidelines Network Guideline 90 (2006) Diagnosis and Management of Head
and Neck Cancer. A National Clinical Guideline. Available at: www.sign.ac.uk/guidelines/
fulltext/90/index.html.
110
Lip, ear, nose and 9
treatment of the neck
in skin cancer
This chapter describes radiotherapy for squamous cell cancers of the lip, external
ear canal and middle ear and nasal vestibule, as well as management of the neck in
skin cancers of the head and neck. Superficial skin tumours of the nose and pinna
are considered in Chapter 7.
Lip
Most lip cancers arise on the vermillion border of the lower lip – the junction
between the skin and the lip itself – and are diagnosed at an early stage because
they are visible. They are usually superficial squamous cell tumours linked to long-
term sun exposure and smoking. Surgery, EBRT and brachytherapy all give local
control rates of 90 per cent or higher at 5 years. The choice of treatment depends
on the expertise available, the likely cosmetic and functional outcomes and patient
choice. Radiotherapy is particularly appropriate when the commissure of the lip is
involved (as function may not be as good after surgery), and for larger tumours
where more extensive resection and reconstruction would be required. Commissure
tumours have higher rates of local recurrence than lower lip cancers.
If excision margins are positive or close (3 mm), adjuvant radiotherapy should be
considered if it would result in better function and cosmesis than a further excision.
In elderly patients with significant comorbidity it may be appropriate to observe
clinically rather than treat with adjuvant radiotherapy, even if margins are close.
Involved neck nodes are usually managed with a neck dissection and postoperative
radiotherapy if required (see below).
■ Sequencing of multimodality therapy
When adjuvant radiotherapy is indicated it should ideally commence within
6 weeks of surgery but a longer gap may be required to allow adequate recovery
from a major resection.

Most lip cancers are superficial and have a low risk of lymph node spread. The
relevant nodes should be assessed clinically and with imaging (CT or MRI) of the
neck if there is palpable lymphadenopathy. The lymphatic drainage of the lower lip
is to ipsilateral level Ib and then level II nodes. The midline portion drains to level
Ia and then to bilateral levels Ib and II but there can be direct drainage to level
III. The upper lip drains to level Ib, sometimes via the buccal nodes, which lie
under the superficial muscle layer of the face (Fig. 9.1).
111
LIP, EAR, NOSE, NECK
Partoid
duct
Buccinator
muscle
Facial nodes Intraparotid nodes
Facial
artery Buccinator nodes Mastoid nodes
Masseter
muscle
Parotid
gland
Figure 9.1 Superficial lymph nodes of the head and neck.

Careful clinical examination with a strong light and magnifying lens is essential to
define the extent of the tumour. Palpation is used to assess thickness. An OPG and
cross-sectional imaging are recommended if bony erosion is suspected.

For small (2 cm diameter) well-defined tumours, a 3–5 mm margin can be added
to the GTV to produce a CTV with a further 5 mm CTV to PTV margin. In
practice, the PTV is drawn on the surface of the skin and lip with a fine marker pen
using a strong light, magnifying lens and a ruler to measure the margin from the
visible GTV. Larger tumours or those with indistinct edges need a larger GTV-
CTV margin of 5–10 mm.
When adjuvant radiotherapy is used a 5 mm margin from the resection edge is
used to define the CTV.
■ Dose solutions
Most tumours are treated with electrons or superficial X-rays. The PTV should be
covered by the 90 per cent isodose of the electron or superficial X-ray beam. Once
the PTV has been defined and marked as above, isodose charts are used to select the
required energy to give 90 per cent coverage from the surface to the deep margin.
Tissue equivalent bolus is used as required either to increase the surface dose or to
reduce unwanted deep penetration. The applicator size can then be calculated. If
electrons are used, the 90 per cent isodose in the lateral plane is 3–5 mm inside the edge
of the applicator, which represents the 50 per cent dose. Electrons bow inwards at
depth at higher energies, so if a high energy is chosen the applicator size will need to
be correspondingly larger to avoid underdosing the deep lateral margin.
112
Lip
A 3–4 mm thick lead mask is then constructed with a cut-out area over the
target volume. An intraoral lead shield is used to protect the gums and teeth from
the exit beam. It is lined with wax to absorb secondary electrons.

Mucositis of the outer and inner lip occurs from the third week of treatment.
White soft paraffin is used to keep the lips moisturised and systemic analgesics
should be prescribed if necessary. Sunburn and smoking should be avoided.
■ Verification
Verification is by daily inspection of the set-up compared with photographs taken
at planning.
■ Other points
Brachytherapy for lip squamous cell cancer
Where technical expertise exists, brachytherapy can produce excellent local control
rates, cosmesis and function. The target volume is defined as above. Rigid needles
are implanted horizontally along the axis of the lip, using either a single plane for
superficial lesions, or three or more sources distributed in a equilateral triangle or
square in the cross-sectional plane for deeper tumours. A plastic template can be
used to provide stability. Sources are usually 10 mm apart and 5–8 cm long which
inevitably means treating most of the lip, making this a useful technique in more
extensive tumours or those with indistinct margins (Fig. 9.2).
(a)
(b)
Figure 9.2 Three-source implant for carcinoma of the lip. (a) Anterior view. (b) Sagittal
view showing arrangement of sources in an equilateral triangle.
113
Ear
Cancers of the external auditory canal and middle ear are rare. There is no agreed
staging system but most tumours present when locally advanced with bone erosion
or cranial nerve palsies. Local invasion and the complexity of a temporal bone
resection mean that close or positive resection margins are common. For these
patients, surgery and postoperative radiotherapy are recommended and can
produce 5-year survival rates of 40–60 per cent. Early tumours confined to the
external ear canal without soft tissue or bone involvement can be treated with
either primary radiotherapy or surgery.

When adjuvant radiotherapy is indicated it should ideally commence within 6 weeks
of surgery but a longer gap may be required to allow adequate recovery from a
major resection.

The external ear canal is a 25 mm long tube with an outer cartilaginous portion and
an inner bony segment lined with mucosa. The middle ear contains the ossicles and
semicircular canals and communicates posteriorly with the mastoid air cells.
Ulceration and submucosal spread in the ear canal is best assessed by otoscopy.
Medially, tumour can spread into the temporal bone (causing VII nerve palsy),
around the internal carotid artery and through the Eustachian tube into the
nasopharynx. Anterior extension into the temporomandibular joint, parotid gland
and masticator space can cause trismus. Posterior spread occurs into the mastoid
air cells and thence to the posterior cranial fossa. Superiorly, tumour can invade
into the middle cranial fossa and inferiorly into the jugular foramen, causing IX, X
and XI cranial nerve palsies, or into the cervical vertebrae.
Lymph node spread at presentation is uncommon. Tumour can spread to the
parotid nodes (sometimes divided into the preauricular, subparotid and superficial
and deep intraparotid).The posterior part of the external canal drains to the mastoid
nodes. Further spread to level II can occur. If tumour reaches the nasopharynx, it
can spread to the retropharyngeal nodes.

A combination of clinical examination to look for cranial nerve palsies, MRI for soft
tissue extent and CT to evaluate bone destruction is most useful to assess local spread.
Immobilisation
The patient is immobilised lying supine in a custom-made shell with the neck
extended to move the orbit superiorly out of the treated volume.
114
Ear
CT scanning
CT slices are obtained from the skull base to the hyoid bone. Slices should be no
more than 5 mm thick and ideally 3 mm. Fusion of MR and CT planning images
can be particularly helpful to define skull base anatomy.
Simulator
3D conformal planning is recommended in all patients because of the complex
anatomy and proximity of critical structures.

Discussion with the surgeon and pathologist is critical to establish the sites at
highest risk of recurrence after a temporal bone resection. Preoperative imaging
should be available during planning. If there is macroscopic residual disease or
positive resection margins can be defined, these sites are contoured as a GTV,
ideally with the surgeon. The GTV is expanded isotropically by 10 mm and then
edited depending on natural tumour barriers to form a CTV66.
For adjuvant treatment, the CTV60 is contoured on each axial CT slice with the
aid of the corresponding slices on preoperative imaging and the operation details,
aiming to encompass all resection margins in the CTV60 and to have a 10 mm
margin from the resection edge at high risk sites. This 10 mm margin should be
individualised depending on natural tumour barriers and possible routes of spread.
Though the risk of involvement is low, adjacent parotid, mastoid and high level II
nodes are included in the CTV60 as they tend to be superficial to the resection
margins and can be treated without an increase in morbidity. The CTV is expanded
isotropically to form the PTV by a margin determined for each department by the
observed random and systematic errors – usually 3–5 mm.
The brainstem, spinal cord and adjacent temporal lobe should be contoured as
organs at risk. The CTV and PTV should be defined without reference to the
organs at risk and vice versa. The PTV may end up overlapping a critical structure
but the dosimetrist and oncologist can then evaluate the beam arrangements and
MLC shielding to balance the risks and benefits. If the CTV or PTV is edited away
from an organ at risk to protect that organ, the opportunity to deliver dose to the
whole target by an innovative beam arrangement may be lost and the good coverage
of the PTV will be falsely reassuring. It is better to accept a compromise in dose to
the PTV when the plan is produced, than to guess where this compromise needs to
be made at the volume definition stage.
For early external ear canal tumours confined to the mucosa, the GTV as defined
by clinical examination is outlined on a planning CT. A 10 mm isotropic margin is
added to produce a CTV but this is expanded to include the adjacent temporal bone.
■ Dose solutions
The photon beam arrangement is individualised depending on the PTV, but
usually anterior and posterior oblique wedged beams are used, often with an
additional lateral beam. The angle of the posterior oblique beam is chosen to avoid
the brainstem and spinal cord. Where the PTV comes close to these structures or
to the temporal lobe, the risk of late radiation damage must be weighed against
the risk of tumour recurrence. The oncologist should take this decision but must
115
inform the patient of the possible risks of late effects and possible benefits of giving
more dose to high risk sites.
If a compromise is needed it is better to use two phases of treatment rather than
underdose the PTV throughout the whole course. The whole PTV should be
included in the phase 1 volume (usually for 20–25 fractions), with a smaller phase
2 shaped with MLCs to protect critical structures. A summated plan is created from
the two phases to allow evaluation of DVHs. Numbers of fractions for each phase
can be varied so that the best compromise can be achieved for individual patients.
It may also be necessary to reduce the total dose to protect OAR (Fig. 9.3).
Figure 9.3 Beam arrangement for postoperative external beam radiotherapy of a middle
ear tumour with skull base invasion. The PTV60 is covered by the 95 per cent isodose but
prescribing 60 Gy would exceed tolerance of the brainstem (BS). The prescribed dose
must be reduced or the PTV coverage compromised for at least some of the treatment.
A PTV66 (purple) has also been defined to include known positive resection margins.
60 Gy in 30 daily fractions given in 6 weeks to PTV60.
66 Gy in 33 daily fractions given in 61⁄2 weeks to PTV66.
Higher doses are associated with an increased risk of osteoradionecrosis of the
temporal bone.

After a temporal bone resection patients usually have considerable morbidity with
cranial nerve palsies affecting swallowing and speech, and a feeding gastrostomy
tube is often in place. A dietician and speech and language therapist should assess
116
Nose
patients weekly. Although the local acute effects of radiotherapy are limited to skin
erythema and localised mucositis, lethargy and anorexia can be debilitating when
radiotherapy follows such major surgery. The shell can be cut out over the treated
volume to reduce skin reaction if the PTV is not close to the skin surface.
■ Verification
Portal images are compared with DRRs as for other head and neck tumours with
in vivo dosimetry on the first day of treatment.
Nose
Cancers of the nasal vestibule behave like skin tumours, are usually diagnosed
when small and have a 90 per cent local control rate with radiotherapy or surgery.
Radiation gives better cosmesis except for very small lesions. External beam
radiotherapy can be combined with brachytherapy. Larger tumours with bone
involvement are rare but have a low chance of cure by radiation alone. Resection may
be followed by adjuvant radiotherapy.

When adjuvant radiotherapy is indicated it should ideally commence within
6 weeks of surgery but a longer gap may be required to allow adequate recovery
from a major resection.

The nasal vestibule is the entrance to the nasal cavity, lined by squamous
epithelium. Tumours of the proximal nasal cavity lined by respiratory epithelium
are covered in Chapter 16. The columella is the midline, medial wall of the
vestibule and tumours arising here can spread submucosally onto the upper lip or
posteriorly along the roof of the hard palate. Cross-sectional imaging with MR or
CT is important to define their extent.
Lymph node spread is uncommon at diagnosis but the vestibule drains to levels
Ia and Ib or to the buccinator node overlying the buccinator muscle and then to
level II.

Careful clinical examination with a strong light and magnifying lens is essential to
define the extent of the tumour. Palpation is used to assess thickness.
Immobilisation
The patient is immobilised in a thermoplastic or Perspex shell with wax nostril
plugs to help produce a more homogeneous dose distribution.
117
CT scanning
Slices 3 mm thick are obtained from the inferior orbit to the hyoid bone to include
level I nodes.
Simulator
3D conformal therapy is recommended given the small fields required and possible
dose inhomogeneities created by adjacent air spaces.

The GTV is outlined on the planning CT scan with the help of clinical examination
and diagnostic imaging. To produce the CTV it is easiest to add a 10 mm isotropic
margin to the GTV and edit this to take account of natural tumour barriers, air and
possible submucosal spread into the lower lip. The CTV-PTV margin is determined
by local audit but ideally should be no more than 3 mm.
■ Dose solutions
Two lateral oblique photon beams provide the most conformal coverage plan for
the PTV. The angle of the beams is chosen to ensure adequate dose to any posterior
extension along the roof of the hard palate. Wax is applied on the external surface
of the nose and upper lip as needed to increase the surface dose. The epithelial
surface thus becomes encased in wax internally and externally to allow the hollow
vestibule to be treated as a block, producing a more homogeneous dose distribution
(Fig. 9.4).
Wax
Identical beam
angles
Figure 9.4 Beam arrangement for conformal treatment of nasal columella tumour. Note
beams are angled posteriorly to cover possible tumour extension along the roof of the
hard palate.
Electrons can also be used for superficial lesions but the contour of the nose,
small field sizes and underlying cartilage make conformal photons the preferred
solution.
Longer fractionation is not necessary given the small volumes treated.
118
Postoperative neck irradiation
The inside of the nasal cavity can become especially sore and topical local
anaesthetic or steroid creams as well as systemic analgesics are often needed to
enable the patient to tolerate insertion of the wax plugs throughout treatment.
■ Verification
Postoperative neck irradiation

Squamous cell skin cancers of the head and neck may recur in the regional lymph
nodes, with a median interval to recurrence of 12 months. Treatment at this time
is usually surgical with postoperative radiotherapy indicated unless a neck
dissection only reveals a solitary level II–V node without extracapsular spread.
It is possible to identify head and neck skin tumours at higher risk of having
occult metastases with size, depth of invasion, high grade and perineural or vascular
invasion all increasing the risk. Tumours on or near the external ear are at higher
risk of spread to the intraparotid nodes. There is, however, no consensus on the role
of prophylactic nodal radiotherapy (or neck dissection) in these tumours.

Ideally radiotherapy should commence within 6 weeks of surgery, as long as there
is adequate healing.

The commonest sites for skin cancers (ear, temple, forehead, anterior scalp and
cheek) drain to the intraparotid nodes and thence to level II and on to levels Ib
and III–V. Many patients therefore present with a parotid mass. In 25 per cent of
patients a primary skin tumour is not identified and a mucosal head and neck
tumour should be excluded. The pattern of lymph node spread (intraparotid
nodes), immunohistochemistry and the presence of sun-damaged skin may point
to the skin as the likely primary source.
■ Assessment of nodal disease

The intraparotid nodes and levels Ib–V should be assessed by clinical examination
with MR preferred to CT for imaging the parotid. Surgery usually involves a
superficial parotidectomy if there are intraparotid nodes and a neck dissection of
levels Ib–III in the clinically negative lower neck. If there are enlarged level Ib–V
nodes an appropriate neck dissection is performed.
Discussion with the surgeon and pathologist is important to determine which
nodal levels require radiotherapy because of microscopic residual disease or for
prophylaxis.
119
■ Dose solutions
The radiotherapy technique is a combination of that described in Chapter 15 for
parotid tumours and in Chapter 8 for the neck. Examples of nodal levels to be
covered in the CTV60 and CTV44 are shown in Table 9.1. If the CTV60 includes
the parotid bed and low neck, it will be difficult to provide adequate coverage with
one plan owing to the change in neck contour. A planned volume superiorly
should then be matched to an anterior photon field treating the lower neck with
the match plane at a level at relatively low risk of harbouring disease.
Table 9.1 Examples of CTV60 and CTV44 for the neck nodes in head and neck skin
squamous cell depending on clinical and pathological findings
Clinically involved Surgery and pathology CTV60 CTV44

nodes
Intraparotid only Superficial parotidectomy alone Parotid bed Levels Ib, II, III
Intraparotid only Superficial parotidectomy and Parotid bed Not defined
level Ib–III dissection. Involved
intraparotid nodes only
Intraparotid only Superficial parotidectomy and Parotid bed Levels Ib, III–V
level Ib-III dissection. Involved and level II
intraparotid and level II nodes*
Intraparotid Superficial parotidectomy and Parotid bed and Not defined
and neck level Ib–V dissection. Involved levels Ib–V
intraparotid and level Ib–V nodes*
*Neck irradiation can be omitted if there is only one level Ib–V node involved without extracapsular spread.
66 Gy in 33 daily fractions to sites of positive margins or extracapsular nodal
spread.
44 Gy in 22 daily fractions given in 4 1⁄2 weeks to PTV44.

See Chapters 8 and 15.
■ Verification
See Chapter 8.
key trials
Trans-Tasman Radiation Oncology Group (TROG) trial 05.01. Post-operative

concurrent chemo-radiotherapy versus post-operative radiotherapy in high-risk
cutaneous squamous cell carcinoma of the head and neck. In progress.
120
Information sources
Information sources
De Visscher JGAM, Grond AJK, Botke G et al. (1996) Results of radiotherapy for squamous cell
carcinoma of the vermilion border of the lower lip. A retrospective analysis of 108 patients.
Mendenhall WM, Stringer SP, Cassisi NJ et al. (1999) Squamous cell carcinoma of the nasal
vestibule. Head Neck 21: 385–93.
Ogawa K, Nakamura K, Hatano K et al. (2007) Treatment and prognosis of squamous cell
carcinoma of the external auditory canal and middle ear: a multi-institutional retrospective
review of 87 patients. Int J Radiat Oncol Biol Phys 68: 1326–34.
Veness MJ, Porceddu S, Palme CE et al. (2007) Cutaneous head and neck squamous cell
carcinoma metastatic to parotid and cervical lymph nodes. Head Neck 29: 621–31.
121
10 Oral cavity
The oral cavity comprises the anterior two-thirds of the tongue, the floor of mouth,
buccal mucosa, hard palate and gingivae (Fig. 10.1). Tumours of the lip are discussed
in Chapter 9.
Upper
Anterior alveolus
Buccal mucosa tongue (gingiva)
Mucosa of
the upper lip
Upper bucco-
alveolar sulcus
Retromolar
trigone
Cheek
mucosa
Lower bucco-
alveolar sulcus
Mucosa of
the lower lip
Floor of Lower
Hard palate mouth alveolus
Soft palate (gingiva) Figure 10.1 Subsites of the oral cavity.

■ Curative radiotherapy for local disease
For T1 and early T2 tumours, surgery and radiotherapy give equivalent local control
rates but only if some or the entire radiation dose is given by brachytherapy. Five-
year local control rates for T1 and T2 tumours are 75–95 per cent and
50–85 per cent, respectively. However, brachytherapy is often technically
impossible (tumour is close to bone) or the necessary expertise unavailable.
Therefore most T1 and T2 tumours of the oral cavity are treated with surgical
excision as long as clear margins and good functional outcome can reasonably be
expected. Where surgical excision is unlikely to achieve these outcomes, external
beam radiation alone is used, most often in the case of a second primary tumour
122
close to a previous resection site. For T1 and T2 cancers of the retromolar trigone,
radiotherapy alone can be used if a good functional outcome after surgery is
unlikely. There is no proven role for radiotherapy for carcinoma in situ of the oral
cavity. For tumours of the hard palate, brachytherapy is not technically possible
and relative hypoxia in bone may reduce cure rates.
■ Adjuvant radiotherapy for local disease

For larger T2 (3 cm), T3 and T4 tumours, local control is best achieved by
surgery and adjuvant radiotherapy. Adjuvant local radiotherapy is also indicated
where a smaller primary tumour is excised with positive margins and the preferred
option of further excision is not possible. Where a small primary tumour has been
excised with close (5 mm) margins, a discussion with the surgeon is important to
assess the risk of microscopic residual disease. The measured margin must be
considered in the context of adjacent structures (whether the margin is adjacent to
bone), possible specimen shrinkage, and how easy it may be to detect local
recurrence early. Single modality treatment should be the goal for early oral cavity
cancer as the best way to maintain function and reduce complications such as
osteoradionecrosis. In addition, second primary oral cancers are relatively common
and once part of the oral cavity has been irradiated, further surgery or radiation may
be difficult. Overall survival rates in locally advanced disease are 40 per cent at best.
■ Radiotherapy to the N0 neck

The clinically and radiologically node negative neck should be treated where the
risk of nodal metastases is thought to be 20 per cent. Tumours of the floor of
mouth or midline tongue or hard palate are at risk of bilateral neck metastases and
both sides of the neck should be treated. Several factors increase the risk of occult
nodal metastases including depth of primary tumour invasion ( 4 mm), larger
primary tumour and site (tongue and floor of mouth tumours are at higher risk
owing to richer lymphatic drainage). At risk nodal levels can be treated by a selective
neck dissection or radiotherapy.
When surgery is the initial treatment for the primary cancer, a selective neck
dissection is usually performed if access to the neck is required for reconstruction,
or if the preoperative risk of occult neck metastases is 20 per cent. Radiotherapy
to the neck is used if histological features of the primary tumour indicate a higher
risk of occult metastases than was thought preoperatively, or if radiation is used to
treat the primary site. If brachytherapy alone is used to treat the primary it is
usually followed by radiation or surgery to the N0 neck (if indicated).
■ Radiotherapy to the Nⴙ neck

When tumour is found in lymph nodes from a neck dissection, adjuvant
radiotherapy is recommended unless there is only a solitary positive node
3 cm
in diameter without extracapsular spread.
If radiation is used to treat the primary disease in a patient with clinically
involved lymph nodes, the neck should also be irradiated. Tumours of the floor of
mouth or midline tongue or hard palate are at risk of bilateral neck metastases and
both sides of the neck should be treated.
123
ORAL CAVITY
■ Palliative radiotherapy
T3 and T4 tumours are best managed by a combination of surgery and adjuvant
(chemo) radiation. However, surgery may be inappropriate for locally advanced
cancer where clear resection margins are not possible, if the expected functional
outcome after surgery is not acceptable to the patient, or if comorbidity precludes
surgery. In these cases, radiotherapy alone can be the best method of palliation and
may provide a small chance of long-term cure, especially if combined with
concomitant chemotherapy.
Sequencing of multimodality therapy

Where EBRT is followed by brachytherapy the overall treatment time should be kept
as short as possible to avoid tumour repopulation. A 1-week gap is recommended.
If adjuvant radiotherapy is indicated it should ideally commence 4–6 weeks after
surgery but after adequate wound healing has occurred. If radiotherapy is delayed
to more than 3 months after surgery because of surgical complications, the potential
risks of EBRT may outweigh the benefits.
Concomitant chemotherapy with adjuvant radiotherapy improves local control
and disease free survival when there is a high risk of local recurrence, especially if
excision margins are positive or there is extracapsular nodal spread.
Clinical and radiological anatomy

Anterior tongue tumours usually present as ulcers, which can spread radially or
invade deeply into the tongue muscles or as an exophytic mass. The lateral tongue
drains to ipsilateral level Ib nodes. Whilst the level Ib nodes have efferent
lymphatic connections to level II and thence to levels III and IV, there are also
direct lymphatic connections from the tongue and floor of mouth to level II and
level III nodes and occasionally to level IV. This explains the relatively high
incidence of skip metastases in oral cavity tumours (10–15 per cent) (Fig. 10.2).
Lymphatics from the floor of mouth and the tip of the tongue drain to level Ia
nodes and thence to bilateral Ib nodes, but there can again be direct spread to level
III nodes. Floor of mouth tumours are often infiltrative and invade the mandible
anteriorly, tongue posteriorly and deep muscles of the floor inferiorly. They
commonly present late so surgery and radiotherapy are often both required for
local control.
Primary tumours of buccal mucosa can invade deep structures including the
mandible and cheek and spread initially to ipsilateral level Ib nodes. Retromolar
trigone cancers can spread inferiorly to the mandible and posteriorly to invade the
pterygoid muscles causing trismus.
The mucosa of the upper and lower alveolus and hard palate is fixed to the
underlying periosteum so invasion of the adjacent bone occurs relatively early
making these tumours less suitable for primary radiotherapy. Hard palate tumours
may originate from minor salivary glands and spread via the nerve roots towards
the skull base. They spread first to ipsilateral level Ib nodes. Midline tumours of
the hard palate can spread bilaterally to level Ib. Hard palate tumours invading the
124
Assessment of primary disease
(a) (b)
Figure 10.2 Primary floor of mouth tumour and lymphadenopathy. (a) Coronal
T1-weighted contrast-enhanced MRI with tumour (T) and level Ib node (N) arrowed.
(b) Axial T1-weighted contrast-enhanced MRI with tumour (T) and level II node (N) arrowed.
soft palate posteriorly can also spread to level II or retropharyngeal nodes which
should be specifically evaluated by cross-sectional imaging.
T staging of oral cavity tumours is shown in Table 10.1.
Table 10.1 UICC TNM (6th edn, 2002*) tumour staging for oral cavity cancer
T1 Tumour
T2 Tumour 2 cm but
T3 Tumour 4 cm in greatest dimension
T4a Tumour invades through cortical bone, into deep (extrinsic) muscles of the tongue
(genioglossus, hyoglossus, palatoglossus and styloglossus), maxillary sinus or skin
of face (i.e. considered resectable)
T4b Tumour invades masticator space, pterygoid plates or skull base and/or encases
internal carotid artery (i.e. considered unresectable)
*With permission.

The oncologist and maxillofacial surgeon should assess primary disease by careful
examination using bimanual palpation to assess tumour thickness when possible as
this predicts for occult metastases. EUA should be considered for more extensive
tumours to assess local invasion before planning curative treatment.
MRI gives the most accurate information to assess local invasion but a CT scan can
also be useful. Superficial mucosal tumours can be difficult to see on cross-sectional
imaging. An OPG can help assess invasion of the mandible. Either MRI or CT can be
used to stage neck nodes.
125
ORAL CAVITY
Data acquisition
■ Immobilisation
Patients should lie supine with a straight spine, immobilised in a Perspex or
thermoplastic shell. A custom-made mouth bite may help to push the tongue
inferiorly when irradiating the hard palate or upper alveolus or to separate the roof
of the mouth from the inferior oral cavity when irradiating the tongue. Mouth
bites can distort the anatomy and make volumes on CT more difficult to define
accurately. Some patients find them difficult to tolerate and they may precipitate
swallowing and thus cause movement of critical structures.
■ CT scanning
CT slices are obtained from the base of skull to the arch of the aorta with the patient
immobilised in the treatment position. Ideally slices should be 3 mm thick to
improve volume definition and quality of the DRR but 5 mm thick slices may suffice.
■ Simulator
Where low dose palliative radiotherapy with opposed lateral fields is planned, a
lateral simulator radiograph is used to define volumes and field edges after marking
any palpable tumour or lymphadenopathy with radio-opaque wires. A midplane
separation is taken to calculate the required monitor units.

When defining volumes, it is important to have all relevant information available
including clinical assessments, EUA reports, operation notes, histology results and
diagnostic imaging.
■ Curative radiotherapy
The GTV is outlined on a planning computer with window settings adjusted to
show soft tissues. There are no studies correlating primary tumour size clinically or
on imaging with microscopic spread histologically on which to base GTV-CTV
margins. A 10 mm margin from macroscopic tumour edge to surgical resection
margin would be regarded as likely to produce local control. We therefore
recommend growing GTV isotropically by 10 mm to produce a CTV70.
The CTV70 is edited to take account of local patterns of tumour spread
and natural tissue barriers, for example bone can be spared if not clinically involved.
The medial pterygoid muscle should be included in the CTV70 if there is local
invasion of part of the muscle, for example from a retromolar trigone tumour.
The CTV70 is then further edited to include adjacent positive or high risk nodal
volumes. These will depend on the site of the primary tumour but include all levels
where there is any lymphadenopathy and levels between the primary site and skip
metastases (usually at least levels Ib and II).
The CTV70 is then copied to form the CTV44 which is expanded to include
other nodal levels at lower risk of occult nodal metastases. Recommended CTV44
126
levels depend on the nodal status and whether both sides of the neck are
considered to be at risk (Table 10.2).
Table 10.2 Recommendations for the selection of prophylactic nodal levels (CTV44) for
oral cavity tumours – anterior two-thirds of the tongue, floor of mouth, buccal mucosa,
hard palate and gingivae
Stage Ipsilateral nodal levels Contralateral nodal levels
N0/N1 with Ib, II, III; include Ia for tumours Not treated
well-lateralised involving tip of tongue, floor of
primary mouth or anterior mandible; include
IV for tumours of the anterior
tongue or involving the oropharynx
N0/N1 primary Ib, II, III; include Ia for tumours Ib, II, III; include Ia for tumours
not well involving tip of tongue, floor of involving tip of tongue, floor of
lateralised mouth or anterior mandible; mouth or anterior mandible;
include IV for tumours of the include IV for tumours of the
anterior tongue or involving anterior tongue or involving the
the oropharynx oropharynx
N2a/N2b with Ib, II, III, IV; include Ia for tumours Ib, II, III; consider not treating
well-lateralised involving tip of tongue, floor of contralateral neck
primary mouth or anterior mandible;
include IV for tumours of the
anterior tongue or involving the
oropharynx; include V if level IV
involved
N2a/N2b Ib, II, III, IV; include Ia for tumours Ib, II, III; include IV for tumours of
primary not involving tip of tongue, floor of mouth the anterior tongue or involving
well lateralised or anterior mandible; include IV for the oropharynx
tumours of the anterior tongue
or involving the oropharynx;
include V if level IV involved
N2c Ib, II, III, IV; include Ia for tumours Ib, II, III; include IV if multiple
involving tip of tongue, floor of mouth contralateral nodes or for tumours
or anterior mandible; include IV for of the anterior tongue or involving
tumours of the anterior tongue or the oropharynx; include V if level
involving the oropharynx; include V IV involved
if level IV involved
N3 Ib, II, III, IV; include Ia for tumours As for N2a/b (no contralateral
involving tip of tongue, floor of mouth nodes) or N2c (contralateral
or anterior mandible; include IV for nodes) respectively
tumours of the anterior tongue or
involving the oropharynx; include V
if level IV involved
127
ORAL CAVITY
■ Adjuvant radiotherapy – primary

The CTV60 is defined using the planning CT scan after discussion with the surgeon
and pathologist to define the sites at risk of microscopic residual disease. Postoperative
oral cavity CTVs can be difficult to define because oral mucosa is not well defined
on CT, dental artefact can obscure the oral anatomy and reconstructive soft tissue flaps
can be confused with normal anatomical structures. If there has been reconstructive
surgery, the whole flap does not need to be included in the CTV60 but the mucosa
adjacent to the primary site should be included with a 10 mm isotropic margin to
account for microscopic spread. A CTV66 (which may be equivalent to the CTV60
or smaller) is defined in the case of positive excision margins and includes the sites
of incomplete excision.
The CTV-PTV margin should be individualised in each department according
to measured random and systematic set-up errors. There should be minimal organ
or tumour movement in a head and neck shell with the exception of small tumours
of the oral tongue. For these, a separate internal margin may need to be added.
Overall CTV-PTV margins should be 3–5 mm in each direction.
■ Adjuvant radiotherapy – neck

If there is an indication to irradiate the neck after a neck dissection (N2/3 disease
or extracapsular spread), a CTV60 is defined to include nodal levels where there
was tumour at surgery or adjacent levels at high risk of microscopic disease (for
example level IV if a selective neck dissection of levels I–III revealed disease in level
III). A CTV66 is defined in the case of extracapsular nodal extension.
A CTV44 is defined to include lymph nodes at risk of containing micrometastases
in levels not removed at surgery. In the case of a lateral tongue tumour with a
heavily node positive ipsilateral neck at neck dissection, levels Ib–III in the
contralateral neck could be included in the CTV44.
■ Prophylactic radiotherapy – N0 neck

At risk nodal levels are defined according to patterns of lymphatic drainage from
the primary site and outlined as CTV44.
GTV, CTV and PTV are defined as for curative treatment either on a planning CT
scan or lateral simulator radiograph. To minimise normal tissue toxicity
(particularly mucositis), smaller margins can be used than for curative treatment –
e.g. 5 mm from GTV to CTV.
Dose solutions
■ Conformal
Conformal solutions are preferred because of the complex three-dimensional shape
of target volumes in oral cavity tumours and the advantage for the patient of sparing
128
Dose solutions
some of the mucosa of the oral cavity and pharynx. The exact multiple beam
arrangement will depend on the tumour doses required, the shape of the volumes
and consideration of doses to OAR.
In evaluating the plan a careful assessment of mandible dose is essential to
reduce the risk of late osteoradionecrosis. Hot-spots of 107 per cent within the
mandible should be avoided.
A midline PTV may still need to be treated with lateral opposed beams but
MLC shielding will help to spare normal tissues, and wedges can be used to
compensate for the change in neck contour and produce a more even dose
distribution (Fig. 10.3). When only one side of the neck is treated, an arrangement
of three coplanar beams can usually provide good tumour coverage while sparing
the contralateral mucosa and parotid gland. At least one of the beams must have
no exit dose through the spinal cord for this organ to remain within tolerance.
In practice this means the angle of the posterior oblique beam is chosen
to provide best coverage of the PTV while avoiding the cord (Fig. 10.4). A
matched anterior neck beam is often required to treat low neck nodes (see
Chapter 8).
■ Complex
IMRT has the potential to deliver the most conformal radiotherapy to complex
target volumes with greater sparing of normal tissues. There are relatively few instances
in oral cavity tumours where this is likely to give a therapeutic advantage as PTVs are
usually convex and as the contralateral parotid can already be spared if the treated
volume is unilateral. Moreover, the extra dose to the oral mucosa and contralateral
mandible from IMRT may increase acute and late effects.
IMRT may offer an advantage when both sides of the neck are included in the
PTV, both to spare one parotid gland and reduce the risk of xerostomia, and to
avoid the uncertainties inherent in a plan with matched posterior electron and
anterior neck photon fields. Beam arrangements similar to those chosen for
oropharynx IMRT are used.
■ Conventional
Where resources are not available for conformal planning, conventional techniques
may be unavoidable. For tumours involving the midline or where bilateral
neck nodes require radiotherapy, opposed lateral fields can be used but this does
not allow sparing of any adjacent mucosa in the treated volume. If the high dose
volume extends posteriorly to the spinal cord, a two phase technique is
used with large lateral fields for phase 1 and smaller lateral fields matched to
posterior electron fields for phase 2. A matched anterior neck field treats
lower neck nodes with midline shielding to reduce dose to the larynx, pharynx and
spinal cord.
Where treatment is unilateral, anterior and posterior oblique wedged fields are
chosen, with a lateral field sometimes used to improve homogeneity medially in
the target volume. An ipsilateral anterior neck field is matched to treat inferior
neck nodes if required.
129
ORAL CAVITY
(a) (b)
(c)
Figure 10.3 Adjuvant radiotherapy for a floor of mouth tumour following excision with
close margins and bilateral negative neck dissections (pT4N0). (a) Axial T2-weighted
fat-saturated MRI showing primary tumour (T). (b) Lateral DRR showing beam borders
and PTV. (c) Axial planning CT slice showing dose distribution.
130
Dose-fractionation
Figure 10.4 Adjuvant radiotherapy for a pT4aN0 buccal mucosa tumour invading the
mandible, which has been resected with close margins but no involved lymph nodes.
Dose-fractionation
■ Curative treatment
EBRT with brachytherapy
EBRT
LDR brachytherapy boost

30 Gy to the reference isodose over 3 days.
LDR brachytherapy alone

65 Gy to the reference isodose over 7 days.
EBRT alone
70 Gy in 35 daily fractions given in 7 weeks concomitant cisplatin or alternative
fractionation, e.g. with concomitant boost.
44 Gy in 22 daily fractions to PTV44 given in 4 1⁄2 weeks.
■ Adjuvant treatment
60 Gy in 30 daily fractions to PTV60 given in 6 weeks concomitant cisplatin.
66 Gy in 33 daily fractions given in 6 1⁄2 weeks to high risk sites.
30 Gy in 5 fractions of 6 Gy given in 3 weeks.
131
ORAL CAVITY
Treatment delivery and patient care

See Chapter 8. Excellent oral hygiene is particularly important when part of the oral
cavity is being irradiated. Mucositis of the lip can be treated with steroid based
paste preparations if there is no concurrent infection. Fungal infections are common
and can recur during a course of radiotherapy. The oral cavity should be carefully
inspected at least once a week and antifungal agents used as necessary.
Verification
See Chapter 8.
Brachytherapy
Brachytherapy offers excellent conformal radiotherapy to small tumours of the
tongue and buccal mucosa that are well demarcated, not close to bone and accessible
for implantation. These tumours are usually treated by surgery and the decision to
treat with brachytherapy instead is often based on the expertise available. The
disadvantages of brachytherapy include the risk of bleeding and infection, the
radiation risk to staff and the patient isolation required.
Anterior tongue tumours or small floor of mouth cancers not too close to the
mandible are usually treated with iridium-192 hairpins after metal gutters have been
inserted. With the patient anaesthetised, the GTV is marked with ink. The depth
of tumour infiltration on MRI corresponds accurately with surgical invasion and
can be used to assess depth. The CTV is the GTV with a 10 mm margin. There is
(a) (b)
Figure 10.5 Source arrangements for brachytherapy of oral cavity tumours. (a) Iridium-
192 hairpins for anterior tongue or floor of mouth. (b) Iridium-192 single plane implant for
buccal mucosa.
132
Information sources
no margin added to form a PTV as there is no daily set-up error or organ motion
relative to the radiation. The gutters are inserted 12–15 mm apart according to
Paris rules to ensure adequate CTV coverage and their position verified by
orthogonal radiographs before the iridium is inserted (Fig. 10.5).
Small buccal mucosa tumours can be treated with a single plane implant using
iridium-192 placed in flexible plastic tubes. After the CTV has been defined, hollow
needles are placed to provide adequate CTV dose and a guidewire is introduced
over which the plastic tubes are placed. Steroids may be required if trauma during
the procedure causes enough swelling to compromise the airway. The skin should
be cleaned regularly with an antiseptic agent. Analgesia is provided with opiates and
non-steroidal anti-inflammatory drugs. The implant is reviewed daily.
key trials
There are no phase III trials evaluating the role of radiotherapy in oral cavity
cancers alone, although these patients are often included in studies of head and
neck cancer in general (see key trials in Chapter 8).
Information sources
See Chapter 8.
Nag S, Cano ER, Demanes J et al. (2001) The American Brachytherapy Society recommendations
for high-dose-rate brachytherapy for head-and-neck carcinoma. Int J Radiat Oncol Biol Phys
50: 1190–8.
Parsons JT, Mendenhall WM, Stringer SP et al. (1997) An analysis of factors influencing the
outcome of postoperative irradiation for squamous cell carcinoma of the oral cavity. Int J Radiat
Oncol Biol Phys 39: 137–48.
Shah JP, Candela FC, Poddar AK (1990) The patterns of cervical lymph node metastases from
squamous cell carcinoma of the oral cavity. Cancer 66: 109–13.
133
11 Oropharynx cancer
and unknown primary
tumour
Two approaches can be recommended for curative treatment of oropharyngeal
squamous cell cancers: primary radiotherapy chemotherapy, or surgery
postoperative (chemo)radiation. There are no randomised controlled trials comparing
these two approaches, although there is indirect evidence suggesting cure rates are
similar. The choice depends on likely functional outcomes, particularly with respect to
swallowing and speech, local medical expertise, patients’ ability to tolerate treatment
and individual patient choice. Treatment of the primary site and the neck can be
considered separately though one modality is usually used as the initial treatment to all
sites of disease. Brachytherapy may have a role in combination with EBRT and is
considered in Chapter 5.
Oropharyngeal tumours usually have lymphadenopathy at presentation and thus
stage III or IV disease but metastases are uncommon. Patients have often smoked
heavily and abused alcohol so assessing functional physiological reserve is
important before deciding on a treatment for each patient.
■ Radiation to the primary tumour

Radiation to the primary site is the preferred treatment in oropharyngeal cancer
because of the difficulty in obtaining adequate surgical margins while maintaining
good swallowing and speech. Target volumes can be defined with greater accuracy
if there has been no surgery prior to radiation. The use of concurrent
chemotherapy or altered fractionation regimens may improve local control and
survival compared with standard radiotherapy in stage III and IV disease. Cure
rates vary from 80–90 per cent for early tonsil cancer to 40–50 per cent for larger
primary tumours with nodal involvement.
If surgery is used to treat the primary site, adjuvant (chemo)radiation is
recommended if surgical margins are positive and should be considered if the
margins are close or if there is vascular or perineural invasion.
■ Radiation to the neck N0/N1

The N0 or N1 neck can be managed with either radiation or surgery alone
depending on the modality used to treat the primary. If there is no clinical
evidence of nodal involvement (N0) prophylactic treatment of the neck is
recommended if the risk of occult metastases is over 20 per cent. Because the
oropharynx has a rich lymphatic network this is usually the case. In well-lateralised
tumours of the soft palate or tonsil, the ipsilateral nodes alone need treatment. The
risk of contralateral nodal involvement is higher in T 3/4 tumours or any tumour
134
involving the tongue base so prophylactic radiation to the contralateral neck is
recommended in these circumstances.
■ Radiation to the neck N2/3
Curative radiotherapy to involved neck nodes is recommended when treatment of the
primary site is with radiotherapy. If the risk of nodal metastases at uninvolved nodal levels
in the ipsilateral or contralateral neck is greater than 20 per cent, a prophylactic dose is
given to these sites. Residual neck disease after radiation requires a neck dissection.
If surgery is used as initial therapy to the primary, a neck dissection is usually
performed. Adjuvant radiotherapy is then recommended in the case of N2b/N2c/N3
disease or if there is extracapsular nodal spread. It should be considered in N2a disease
particularly if there are poorer prognostic features (see Table 11.1).
Table 11.1 Poor prognostic features after potentially curative surgery
Primary Nodes
*Positive margins *Extracapsular spread

Close margins (0–5 mm) More than one involved node
Perineural invasion One node 3 cm
Lymphovascular invasion
Poorly differentiated histology
T4 tumour (invading local structures)
*Positive excision margins and extracapsular nodal spread are definite indications for adjuvant
radiotherapy.
In the presence of metastases or if tumour size or comorbidity preclude a curative
approach, palliative radiotherapy may improve pain, dysphagia and speech.
■ Unknown primary tumour sites in the
head and neck
Patients presenting with a level II/III neck node without a clear primary site often
have a primary site in the tonsil, base of tongue, pyriform fossa or nasopharynx.
Pathology of the involved nodes obtained by biopsy or excision can suggest a likely
primary site as can careful examination of the pharynx. If no primary can be found,
radiation is given either to the neck alone or to the neck and possible primary sites,
usually encompassing the ipsilateral tonsil, tongue base and pyriform fossa.
■ Oropharyngeal lymphoma
Lymphomas of the oropharynx (usually tonsil) are treated with radiotherapy alone
or sequential chemotherapy and radiation to residual disease or high risk sites.

The oropharynx is the posterior continuation of the oral cavity. Superiorly it
joins the nasopharynx at the level of the hard palate and inferiorly it meets
135
OROPHARYNX
the hypopharynx at the level of the hyoid bone. An appreciation of the

complex anatomy and physiology of this area is vital in order to understand the
mechanisms for local invasion and lymph node spread. The oropharynx has four
walls (Fig. 11.1).
Soft palate
Palato- Tonsil Larynx

pharyngeal Epiglottis
arch
(posterior
tonsillar pillar)
Palato-
glossal arch
(anterior
tonsillar pillar)
Posterior wall
Hard palate
Base of Vallecula
Tonsil Uvula Soft palate tongue
(a) (b)
Figure 11.1 Anatomy of the oropharynx. (a) Anterior view. (b) Superior view looking down
from the nasopharynx. The valleculae (glossoepiglottic fossae) lie between the base of
tongue and the epiglottis.
The commonest site of oropharyngeal tumours is the lateral wall, which is made
up of the tonsils, tonsillar fossae and palatoglossal and palatopharyngeal arches.
Tonsil cancers tend to be infiltrative and can spread inferiorly to the tongue base
or superiorly to the soft palate. The rich lymphatic drainage of the tonsils means
spread to ipsilateral lymph nodes is common at presentation. The tonsil is a
frequent site for occult primaries to be detected in patients presenting with
enlarged neck nodes. Tonsillar tumours extending onto the tongue base have a
greater risk of bilateral nodal spread.
The base of tongue and valleculae (glossoepiglottic fossae) comprise the anterior
wall and have a rich lymphatic drainage to both sides of the neck. Tumours
involving the tongue base are therefore treated with bilateral neck irradiation. The
infiltrative nature of most base of tongue tumours in a site critical for swallowing
means that patients should be assessed before therapy for evidence of aspiration
and a prophylactic feeding tube is advised.
Tumours of the superior wall (soft palate and uvula) and posterior wall are less
common. Lateral soft palate tumours spread to ipsilateral lymph nodes while more
medial tumours or those arising in the posterior wall have a higher chance of
bilateral nodal involvement. The lack of submucosal tissue on the posterior
pharyngeal wall means radiation to these tumours often results in a shallow ulcer,
which takes many months to heal, can be very painful and can be confused with
residual tumour.
136
Level II nodes are the usual first site of spread. Disease then spreads inferiorly to
levels III and IV or superiorly to upper level II (and retrostyloid) or retropharyngeal
nodes. Level V involvement is relatively uncommon. Pattern of nodal spread may
be less predictable in patients who have had a previous neck dissection.

The oropharynx, oral cavity and larynx should be carefully examined by the
radiation oncologist and ENT surgeon together to assess the mucosal sites of
disease. Tonsil and soft palate tumours can be seen by inspection through the oral
cavity. The tongue base is best assessed by flexible nasendoscopy or indirect
(mirror) examination combined with careful bimanual palpation. EUA may be
helpful to define the extent of tumour and to obtain histology.
Investigation of a squamous cell cancer in level II/III nodes without an obvious
primary site on examination should include EUA with bilateral tonsillectomies,
biopsies of the nasopharynx and pyriform fossae and deep biopsies of the base
of tongue.
PET-CT is the most sensitive modality to detect a primary site and can help to
direct the biopsies (Fig. 11.2). False positive results may occur if the PET-CT follows
the EUA.
Figure 11.2 PET-CT for an unknown primary head and neck cancer presenting with right
level II nodes. Uptake in the right tonsil strongly suggests this is the primary site.
Cross-sectional imaging is essential to define extent of the primary tumour and

MRI is preferable to CT in oropharyngeal tumours (Fig. 11.3); however, MRI and
CT are equally accurate at assessing neck disease. Cross-sectional imaging should
include the whole neck from the skull base to the superior mediastinum to assess
retropharyngeal nodes and levels I–V. A node 10 mm is considered pathological
but nodes larger than 5 mm should be regarded as suspicious if in a location typical
137
OROPHARYNX
for nodal metastases. In this case ultrasound and fine needle aspiration cytology
may be useful to confirm metastases.
(a) (b)
Figure 11.3 T1-weighted contrast-enhanced MRI of a T2N2b left tonsil cancer. (a) Axial
slice showing tumour (T) and level II node (N). (b) Coronal slice showing tumour (T) and
level Ib node (N) adjacent to submandibular gland (S).
Data acquisition
■ Immobilisation
The patient lies supine with the spine as straight as possible and no mouth bite,
but any dentures should be left in place. A shell with at least five fixation points is
constructed to ensure immobilisation (see Chapter 8).
■ CT scanning
CT images – ideally with intravenous contrast – are acquired with 3 mm thick slices
from the skull base superiorly to the top of the aortic arch inferiorly. Lateral and
midline reference marks are drawn on the mask. The CT dataset is imported into
the treatment planning system. Many oropharyngeal tumours are more easily
defined with MRI than CT so co-registered images may improve GTV definition.
■ Simulator
If 3D CT-based planning is not available, beam borders can be defined on
orthogonal radiographs in the simulator. Lateral and anteroposterior (AP) films
are taken when the immobilisation device is fitted. The GTV is drawn on the films
with the help of diagnostic clinical information and imaging. The CTVs and PTVs
are then defined as for CT planning and beam arrangements chosen.
138
The diagnostic images and records and photographs of initial evaluation (in clinic
or at EUA) are critical to assess the extent of the primary tumour and the exact site
of involved lymph nodes. The GTV is defined as the primary tumour and any
lymph nodes over 10 mm in short axis dimension or smaller nodes with necrotic
centres or rounded contours thought to contain tumour.
A high dose CTV70 is created to include sites of local and nodal spread and a
lower dose CTV44 to include uninvolved nodal levels to receive a prophylactic dose.
The GTV-CTV70 expansion at the primary site is individualised and non-
uniform reflecting local barriers to spread and the possible involvement of local
structures. For example the CTV70 for a left base of tongue tumour should include
the whole tongue base and left tonsillar fossa; the CTV70 for a T1 tonsil cancer is
the tonsillar fossa but not tongue base or the nearby medial pterygoid muscle.
For N0 tumours, the CTV70 will include level II nodes adjacent to the primary
either ipsilaterally (lateral tonsil or soft palate primary) or bilaterally. In node-
positive disease the CTV70 includes all levels with involved lymph nodes and those
adjacent to the primary site. If there is involvement of an adjacent muscle due to
extracapsular spread, this should be included in the CTV70.
There is no evidence correlating CT data with pathological specimens from
which to derive a GTV-CTV margin. In practice, the CTV70 is best achieved by
automatically growing the GTV by 10 mm (the pathological margin that a surgeon
aims for), and editing it, to include sites of local spread and possible nodal
involvement and to exclude soft tissues where there are natural barriers to tumour
spread, air spaces and uninvolved bone.
A low dose CTV44 is created from the CTV70 by also including other lymph
node sites thought to be at risk of microscopic nodal disease. If this volume is to
be treated with a single anterior neck beam, a 3D volume need not be defined.
If the posterior pharyngeal wall is involved the bilateral retropharyngeal nodes
should be included in the CTV44. Examples of nodal levels to be included in the
CTV44 for different tumours are given in Table 11.2.
Table 11.2 Suggested nodal CTVs for oropharyngeal tumours
Disease site CTV44
Lateral tonsil or soft palate T1/T2 N0/N1 Ipsilateral levels II–IV (and Ib, RP, retrostyloid if N1)
Lateral tonsil or soft palate T1/T2 N2a,b Ipsilateral levels Ib–IV, retrostyloid, RP and consider
ipsilateral level V
Base of tongue, midline soft palate, Bilateral levels II–IV (and RP if posterior pharyngeal
posterior pharyngeal wall T1/T2 N0 wall involved)
Tonsil or soft palate T 3/4 and/or Consider each hemi-neck separately: N neck – levels
N2c, N3 Ib–IV, retrostyloid and RP nodes and consider level V
Base of tongue, midline soft palate N0 neck – levels II–IV (and RP if posterior pharyngeal
Any T 3/4 and/or N wall involved)
RP, retropharyngeal.
139
OROPHARYNX
The CTVs are grown to PTVs by applying a margin which can be determined
by the local assessment of random and systematic errors, assuming that tumour
movement in head and neck cancer is negligible. This is usually 3–5 mm.
The spinal cord is outlined and grown by 5 mm to produce a PRV. Each parotid
gland is contoured where parotid sparing is feasible.
■ CTV for adjuvant radiotherapy

In addition to imaging and clinical information, discussion with the surgeon and
pathologist is important in determining the sites at greatest risk of recurrent disease.
There is no GTV. If radiation to the primary site is indicated, the CTV60 is the
operative tumour bed paying particular attention to sites where excision margins are
positive or close. If radiation to the nodes is indicated, the CTV60 includes the nodal
levels where there were positive lymph nodes and adjacent structures at sites of
extracapsular spread. A CTV44 is only defined if there are nodal levels that were not
operated on but have a 20 per cent chance of containing occult metastases (see
Table 11.2).
■ CTV for unknown primary

Any macroscopically enlarged nodes are contoured as GTV. The CTV70 is the
nodal levels containing tumour with at least a 10 mm margin superiorly and
inferiorly and is usually the ipsilateral level II and upper level III nodes. There are
two approaches to contouring the CTV.
One is to assume the primary site is in the ipsilateral base of tongue, tonsil or
pyriform fossa (and nasopharynx in some parts of the world) and to include these sites
in the CTV44 along with their draining lymph nodes. The CTV44 will include levels
(a) (b)
Figure 11.4 Suggested CTVs for an unknown primary head and neck cancer presenting with
a level II node. (a) Axial slice showing the left tonsil (t) and base of tongue (B) in the CTV44
(magenta), CTV70 (cyan). (b) Axial slice showing the left pyriform fossa (P) in the CTV44.
140
Dose solutions
Ib–IV, retropharyngeal and retrostyloid nodes on the involved side and levels II–IV
contralaterally (Fig. 11.4). Another approach is to not treat the possible primary sites
and only include the ipsilateral levels Ib–IV, retrostyloid and retropharyngeal nodes.
This is less toxic as much of the oropharyngeal mucosa can be spared but, if the
primary site later becomes apparent it will be very difficult to treat with radiation so
close to the original CTV. We recommend the first approach unless comorbidity and
performance status suggest such extensive treatment volumes will be poorly tolerated.
Dose solutions
■ Conformal
Phase 1 aims to treat the PTV44 to a dose thought sufficient to eradicate microscopic
disease in lymph nodes. Unilateral PTVs are treated with a plan using two or three
beams unilaterally to provide dose to the PTV according to ICRU62 (Fig. 11.5). The
angle of the posterior oblique beam is chosen so as to avoid the spinal cord PRV
which can sometimes mean accepting reduced dose to the most posterior part of the
nodal CTV. This is matched to an ipsilateral low neck beam.
Figure 11.5 Beam arrangement and dose colourwash for treatment of a tonsillar cancer
and ipsilateral neck nodes.
Those PTVs extending to the contralateral side are usually treated with opposing
lateral beams with AP wedges and sometimes superior–inferior wedges to
compensate for changes in contour. A three-beam arrangement can sometimes be
used to try to spare the contralateral parotid gland. Usually the planned superior
volume is matched isocentrically to a lower anterior neck beam, with midline
shielding to treat more inferior nodes without irradiating midline structures.
141
OROPHARYNX
Phase 2 aims to treat the PTV70 to a curative dose and the beam arrangements are
similar to those in phase 1 but with more shielding if possible. If the PTV70 extends
posteriorly to the posterior level II or level V nodes, a matched posterior electron
beam (9 MeV or 12 MeV) may be needed. The plans for phase 1 and phase 2 should
be reviewed together to ensure PTVs are adequately covered according to ICRU62
and critical structures are not overdosed. Acute side effects will be reduced if the dose
to the oral cavity and pharynx outside the PTV is minimised.
■ Complex
IMRT is the ideal treatment for oropharyngeal cancer requiring bilateral neck
irradiation as it allows parotid sparing to reduce the risk of xerostomia and obviates the
need for matched electron fields (Fig. 11.6). Different PTVs can be treated to different
doses within the same plan. If bilateral level II nodes contain tumour, the need to
include the retrostyloid nodes in the treated volume may preclude parotid sparing.
Figure 11.6 Volumes, five beam arrangement, and dose distribution for
intensity-modulated radiotherapy of a right base of tongue tumour. Colourwash scale is
set from 79 per cent (i.e. 54 Gy) to illustrate the conformality of the plan.
Target volumes are defined as above. Nodal PTVs often come close to the skin
surface. To artificially create a skin sparing effect with IMRT the nodal volumes are
edited so as to be a minimum of 5 mm from the body contour. The only exception
is when an involved node is within 5 mm of the skin surface.
The spinal cord is outlined with a 5 mm isocentric margin as a PRV. Both parotid
glands are contoured: they can be difficult to define on a planning CT, particularly
when there is dental artefact, though the diagnostic MRI may be helpful. The
brainstem is also contoured as an OAR. There is increasing evidence that the dose
to the pharyngeal constrictor muscles may relate to the risk of long-term dysphagia.
They can be spared inferior to the midline PTV and can either be contoured or can
142
Dose solutions
be included as part of an avoidance volume, including the larynx, hypopharynx and
trachea, which can help to reduce dose to these midline structures.
When the high dose PTV extends inferiorly below the inferior cricoid cartilage the
whole neck is treated in one plan, usually with a seven-beam arrangement of
equispaced IMRT beams. If the more inferior nodes only require a prophylactic dose
they can be treated with a matched anterior neck beam as for conformal treatment, in
which case the superior part of the volume can usually be treated with five equispaced
beams.
In practice, the five IMRT beam angles are selected with beam sizes chosen to cover
the PTV. The computer software uses an iterative process to try to meet
the constraints applied to the target volumes as illustrated in Table 11.3. The operator
guides the computer by defining the constraints and weighting them according to
their importance. The target constraints and weights are gradually changed until
Dose [Gy]
10.0
12.0
14.0
16.0
18.0
20.0
22.0
24.0
26.0
28.0
30.0
32.0
34.0
36.0
38.0
40.0
42.0
44.0
46.0
48.0
50.0
52.0
54.0
56.0
58.0
60.0
62.0
64.0
66.0
68.0
70.0
0.0
2.0
4.0
6.0
8.0
100.0 100.0
90.0 90.0
Ratio of total structure volume [%]
80.0 PTV65 80.0

70.0 PTV54 70.0
60.0 60.0
50.0 50.0
40.0 40.0
30.0 30.0
20.0 20.0
10.0 10.0
0.0 0.0
0.0
3.0
6.0
9.0
12.0
15.0
18.0
21.0
24.0
27.0
30.0
33.0
36.0
39.0
42.0
45.0
48.0
51.0
54.0
57.0
60.0
63.0
65.0
68.0
72.0
75.0
78.0
81.0
84.0
87.0
90.0
93.0
96.0
99.0
102.0
105.0
108.0
(a) Relative dose [%]
Dose [Gy]
10.0
12.0
14.0
16.0
18.0
20.0
22.0
24.0
26.0
28.0
30.0
32.0
34.0
36.0
38.0
40.0
42.0
44.0
46.0
48.0
50.0
52.0
54.0
56.0
58.0
0.0
2.0
4.0
6.0
8.0
100.0 100.0
90.0 90.0
Ratio of total structure volume [%]
80.0
LP 80.0
70.0 70.0
60.0 60.0
50.0 50.0
40.0 40.0
30.0 30.0
BS
20.0 20.0
10.0 10.0
SC 5mm
0.0 0.0
0.0
3.0
6.0
9.0
12.0
15.0
18.0
21.0
24.0
27.0
30.0
33.0
36.0
39.0
42.0
45.0
48.0
51.0
54.0
57.0
60.0
63.0
66.0
69.0
72.0
75.0
78.0
81.0
84.0
87.0
90.0
(b) Relative dose [%]
Figure 11.7 DVHs for the volumes and plan shown in Figure 11.6, meeting the targets in
Table 11.3. (a) PTV65 (red) and PTV54 (green). (b) Spinal cord PRV (yellow, max dose
45.1 Gy), brainstem (blue, max dose 48.2 Gy), left parotid gland (pink, mean dose 22.6 Gy).
143
OROPHARYNX
Table 11.3 Suggested targets for PTV and OAR coverage for oropharyngeal IMRT
Structure Volume Target dose
PTV65 99 per cent 58.5 Gy (90 per cent)

95 per cent 61.75 Gy (95 per cent)
50 per cent 65 Gy (100 per cent)

5 per cent 68.25 Gy (105 per cent)
PTV54 99 per cent 48.6 Gy
95 per cent 51.3 Gy
50 per cent 54 Gy
Cord 5 mm Any 46 Gy (1 cm3 can receive 48 Gy)
Brainstem Any 54 Gy
Contralateral parotid Mean dose 24 Gy
Body – PTVs 5 per cent 71.5 Gy (110 per cent)
spinal cord and brainstem are within tolerance, the PTVs are adequately covered and,
if possible, the mean contralateral parotid dose is less than 24 Gy (Fig. 11.7).
Dose-fractionation
T1T2N0
70 Gy in 35 daily fractions given in 7 weeks
or
55 Gy in 20 daily fractions of 2.75 Gy given in 41⁄2 weeks.
T 3/4 any N or any T N

70 Gy in 35 daily fractions given in 7 weeks concomitant cisplatin
or
alternative fractionation, e.g. with concomitant boost.
44 Gy in 22 daily fractions to PTV44 given in 41⁄2 weeks.
■ Simultaneous boost IMRT

65 Gy to high dose PTV and 54 Gy to low dose PTV in 30 fractions given in
6 weeks.
60 Gy in 30 daily fractions to PTV60 given in 6 weeks concomitant
cisplatin.
66 Gy in 33 daily fractions given in 61⁄2 weeks to high risk sites.
44 Gy in 22 daily fractions to PTV44 given in 41⁄2 weeks.
144
Information sources
30 Gy in 5 fractions of 6 Gy twice weekly given in 21⁄2 weeks.

Patients are treated daily. Any gap in treatment must be corrected – ideally by
treating twice in one day. There must be at least a 6-hour gap between fractions
on any day when a patient receives two doses.
Weekly assessment by doctor or radiographer and dietician is important. Regular
advice by a speech and language therapist can help maintain safe, functional
swallowing. Most patients are at high risk of weight loss and should be considered
for a gastric feeding tube inserted for prophylaxis before radiotherapy. Further
details are given in Chapter 8.
Verification
See Chapter 8.
key trials
See Chapter 8 – most phase III trials in head and neck cancer contain a majority
of patients with oropharyngeal cancers.
Calais G, Alfonsi M, Bardet E et al. (1999) Randomized trial of radiation therapy
versus concomitant chemotherapy and radiation therapy for advanced-stage
oropharynx carcinoma. J Natl Cancer Inst 9: 2081–6.
Denis F, Garaud P, Bardet E et al. (2004) Final results of the 94–01 French Head
and Neck Oncology and Radiotherapy Group randomised trial comparing
radiotherapy alone with concomitant radiochemotherapy in advanced-stage
oropharynx carcinoma. J Clin Oncol 22: 69 –76.
Information sources
Chao KS, Majhail N, Huang CJ et al. (2001) Intensity-modulated radiation therapy reduces late
salivary toxicity without compromising tumour control in patients with oropharyngeal carcinoma:
a comparison with conventional techniques. Radiother Oncol 61: 275–80.
Eisbruch A, Ten Haken RK, Kim HM et al. (1999) Dose, volume, and function relationships in parotid
salivary glands following conformal and intensity modulated irradiation of head and neck cancer.
Int J Radiat Oncol Biol Phys 45: 577–87.
Jackson SM, Hay JH, Flores AD et al. (1999) Cancer of the tonsil: the results of ipsilateral radiation
treatment. Radiother Oncol 51: 123–8.
Maes A, Weltens C, Flamen P et al. (2002) Preservation of parotid function with uncomplicated
conformal radiotherapy. Radiother Oncol 63: 203–11.
145
12 Hypopharynx

■ Curative treatment for T1T2 N0 tumours
Table 12.1 gives tumour staging for hypopharyngeal cancer. Early stage
hypopharyngeal tumours are rare but can be treated with curative intent by surgery
or radiotherapy. Choice depends on local expertise and patient preference but as
clear margins and good pharyngeo-laryngeal function are difficult to achieve
surgically, radiotherapy is recommended. There are no RCTs comparing these
approaches.
Table 12.1 UICC TNM (6th edn, 2002*) tumour staging for hypopharyngeal cancer
T1 Tumour limited to one subsite (postcricoid, pyriform sinus or posterior pharyngeal wall) and

T2 Tumour invades more than one subsite of hypopharynx or an adjacent site, or measures
2 cm but
4 cm in greatest dimension, without fixation of the hemilarynx
T3 Tumour 4 cm in greatest dimension, or with fixation of the hemilarynx
T4a Tumour invades any of the following: thyroid/cricoid cartilage, hyoid bone, thyroid gland,
oesophagus, central compartment soft tissue (prelaryngeal strap muscles and
subcutaneous fat)
T4b Tumour invades prevertebral fascia, encases carotid artery, or invades mediastinal structures
*With permission.
Pyriform fossa tumours have the best prognosis – radiotherapy can achieve local
control in 90 per cent of T1 tumours and 80 per cent of T2 tumours with a 5-year
overall survival of 50–65 per cent. Involvement of the apex of the pyriform fossa
(which may imply occult invasion of cartilage) predicts a lower chance of cure.
Postcricoid carcinomas have a 20–50 per cent cure rate with radiotherapy which is
equivalent to that in surgical series. There are fewer data available for posterior wall
tumours but cure rates are similar to those for postcricoid cancer.
■ Curative treatment for T3–4 or Nⴙ tumours

Most hypopharyngeal cancers are locally advanced at presentation and many
patients have significant smoking and alcohol exposure and cardiac, respiratory and
other comorbidities. Treatment is therefore tailored and options range from
pharyngolaryngectomy with adjuvant radiochemotherapy, radiochemotherapy alone
to no attempt at cure with palliative care only. Treatment choice depends on the
146
site and stage of disease and patient preference acknowledging the balance between
possible cure, treatment morbidity and likely functional outcomes.
There are phase III data to support an organ preserving approach for advanced
pyriform fossa cancer. Large volume primary disease, particularly with cartilage
destruction, is best treated with surgery and postoperative radiotherapy.
■ Adjuvant radiotherapy
Adjuvant radiotherapy to the site of primary tumour is recommended for all
resected T3/T4 tumours and for T1/T2 tumours with positive or close resection
margins (usually 5 mm or 10 mm where there is surgical concern). Adjuvant
radiotherapy to the involved neck is recommended in N2/3 disease or N1 disease
with extracapsular spread.
Locally advanced hypopharyngeal cancers – particularly those originating in the
postcricoid and posterior pharyngeal wall – have a low chance of cure with any
approach. Patients’ wishes, advanced age and comorbidity mean that palliative
treatment should be discussed with all patients with locally advanced disease.
Palliative radiotherapy may improve pharyngeal pain and obstructive symptoms
and help to control neck disease but hardly ever improves complete dysphagia.
Supportive care alone is an option for patients with incurable disease, complete
dysphagia and few other symptoms.

If primary surgery is used and adjuvant radiotherapy is indicated, radiation should
begin 4–6 weeks postoperatively or when adequate healing has taken place.
Concomitant cisplatin chemotherapy should be considered, especially when resection
margins are positive or there is extracapsular nodal spread.
Radiotherapy following induction chemotherapy in patients with a response
should ideally begin 3 weeks after the start of the last chemotherapy cycle. Addition
of concomitant chemotherapy to this regimen is recommended if the patient can
tolerate such intensive treatment.

The pyriform fossae are mucosa-lined spaces on either side of the larynx (Fig. 12.1).
Each is shaped like an inverted pyramid with the apex between the cricoid cartilage
medially and the thyroid cartilage laterally at the level of the cervical oesophagus. The
posterior pharyngeal wall is the posterior mucosa in between the pyriform fossae
extending from the level of the hyoid bone to the inferior border of the cricoid
cartilage. The postcricoid mucosa is the anterior surface of the hypopharynx from the
arytenoid cartilages to the inferior border of the cricoid. It is continuous
posterolaterally with the inferior portion of the posterior pharyngeal wall (Fig. 12.2).
Tumours can spread mucosally to involve adjacent mucosal sites – other subsites
of the hypopharynx, the supraglottic larynx or tongue base. If other sites are
invaded, lymph node target volumes should include levels draining these sites.
Because the hypopharyngeal mucosa is thin, tumours readily spread into the adjacent
147
HYPOPHARYNX
Figure 12.1 Transnasal oesophagoscopic

view of the hypopharynx. Note the
pyriform fossa (P) and larynx (L). There is a
tumour of the posterior pharyngeal wall (T).
C4 cervical
vertebra
Epiglottis
Hyoid bone
Thyroid cartilage
Arytenoid cartilage
Vocal cords
Cricoid cartilage
Posterior pharyngeal wall Cervical Figure 12.2 Midline sagittal

Posterior cricoid mucosa oesophagus
anatomy of the hypopharynx.
sites – medially from the pyriform fossae to the paraglottic space (and then into the
larynx), laterally into the soft tissues of the neck, or posteriorly into the prevertebral
fascia. They can also invade the arytenoid, thyroid and cricoid cartilages. Submucosal
spread is common and should influence GTV-CTV margins. Postcricoid tumours
can spread into the cervical oesophagus.
Lymph node spread is common at presentation and occult disease is found in up
to 60 per cent of patients with no lymphadenopathy. The pyriform fossae drain to
ipsilateral levels II and III nodes initially, and other sites to levels II, III and IV,
usually bilaterally. Tumours involving the posterior pharyngeal wall can also spread
directly to the retropharyngeal nodes. If levels II–IV are involved, tumour can
spread to level Ib or V or to retropharyngeal nodes. Postcricoid cancers can also
spread inferiorly to the paratracheal and paraoesophageal nodes.

Most pyriform fossa cancers are visible at nasendoscopy during which vocal cord
function is assessed. Vocal cord paralysis is a poor prognostic sign. Tumours are
more visible if the fossae are distended by the patient performing a Valsalva
148
Data acquisition
manoeuvre with the nostrils pinched. Transnasal oesophagoscopy allows other
hypopharyngeal sites to be assessed and biopsies to be taken if necessary. Palpable
lymphadenopathy should be recorded and fine needle aspiration of lymphadenopathy
is often an easier way to confirm the diagnosis histologically.
EUA allows assessment of mucosal extent and deep fixation of the tumour and
should be performed if surgery is considered. The EUA should include laryngoscopy,
oesophagoscopy and bronchoscopy to assess involvement of adjacent organs.
Cross-sectional imaging with either CT or MRI is performed (Fig. 12.3). Both
are equally good at demonstrating lymphadenopathy though MRI may provide
slightly better imaging of the primary tumour. CT is preferable for more inferior
tumours that may have spread to the superior mediastinal nodes as it allows easy
imaging of the head, neck and thorax contiguously. A barium swallow can be
helpful to document the superior and inferior extent of the primary tumour as well
as to assess aspiration and fixation to the prevertebral muscles.
Figure 12.3 Diagnostic CT scan of a right

pyriform fossa tumour (T) growing into the
paraglottic space (PS), through the thyroid
cartilage (TC) and invading onto the
posterior pharyngeal wall (PW). (Note the
adjacent involved lymph node (N).)
Data acquisition
■ Immobilisation
Patients lie supine on a headrest to keep the spine straight, with a custom-made
shell fixed to the couch top in at least five places to reduce movement. The treated
volume will usually extend inferior to the level of the shoulders which should be
as low as possible to facilitate beam entry. No mouth bite is required.
■ CT scanning
With the patient immobilised, CT images are obtained from the skull base to the
carina. The inferior border is at the carina to facilitate planning with lateral beams
angled inferiorly if required. Slices should ideally be 3 mm thick. Intravenous
contrast may help in the definition of lymph node volumes but is unlikely to add
more information than contrast-enhanced diagnostic imaging which must be
available at the time of target volume definition. Contrast can be particularly helpful
when there has been tumour shrinkage after induction chemotherapy as the residual
GTV will differ from that demonstrated on the original diagnostic scan.
149
HYPOPHARYNX
There is as yet no proven role for image fusion with MRI or PET in defining
tumour volumes. MRI is unlikely to add much to CT-based planning as
hypopharyngeal tumours are seen almost as well on CT. The role of PET for target
volume definition is evolving and may facilitate dose escalation by identifying a
smaller target volume within the GTV which could receive a higher dose.
■ Simulator
Ideally all patients with hypopharyngeal cancer are planned three-dimensionally
using CT images in view of the complex volume shape and position. If opposing
lateral beams are to be used, the borders can be defined in the simulator on a
lateral radiograph.

■ CTV for curative radiotherapy T1/T2 N0
The GTV is defined using endoscopy, EUA reports and diagrams, and diagnostic
imaging. There is no evidence on which to base GTV-CTV margins at the primary
site but submucosal spread is common and can extend at least 10 mm from the
GTV. We recommend the GTV is grown by the planning computer by a 10 mm
axial margin and a 15 mm longitudinal margin to create a CTV70. This is then
edited to take account of patterns of spread and natural barriers to tumour
progression (e.g. bone). The CTV70 is also edited to include adjacent lymph
nodes (usually level III but occasionally part of levels II and IV) on the same axial
slices as the CTV70. For lateral pyriform fossa tumours these high risk nodes will
be ipsilateral; for other hypopharyngeal tumours they will be bilateral.
The CTV70 is then copied to become the CTV44. The CTV44 is edited to
include other lymph node groups at risk of micrometastases. For lateral pyriform
fossa tumours this will include ipsilateral levels II–IV. For other tumours with
bilateral lymph node drainage, bilateral nodes are included. Retropharyngeal
nodes are included for tumours involving the posterior pharyngeal wall.
■ CTV for curative radiotherapy T 3/4 or Nⴙ

Some patients in this group will have been treated with and responded to induction
chemotherapy. CTVs should therefore be based on the initial pattern of disease as
well as on the residual tumour seen on the planning CT. The GTV should be
defined on the planning CT from the residual volume but should include any nodes
that were involved at diagnosis even if they are not now enlarged.
To allow for submucosal spread, the GTV is enlarged by 10 mm axially and 15 mm
longitudinally to form the CTV70. The CTV70 is edited to take account of natural
barriers to tumour progression (e.g. bone, air and uninvolved muscle) and to include
all sites of primary disease at presentation. For example, a pyriform fossa cancer
invading the tongue base at diagnosis may respond to induction chemotherapy to
leave residual tumour in the pyriform fossa alone. The tongue base should be
included in the CTV70. The CTV is further edited to include sites of high risk nodal
disease. For N disease the CTV70 includes level II–IV nodes adjacent to the
primary GTV, any involved nodes at other levels and any nodes in between.
150
When the CTV70 has been defined it is copied to become the CTV44. This
volume is then edited to include nodal sites at lower risk of containing microscopic
disease (Table 12.2; Fig. 12.4).
Table 12.2 Suggested CTVs for hypopharyngeal tumours
Disease site CTV44
T 1/2 N0 pyriform fossa (confined to Ipsilateral II–IV

lateral wall)
T 1/2 N0 post cricoid /posterior Bilateral II–IV. Include bilateral RP nodes if posterior
pharyngeal wall pharyngeal wall involved
T 3/4 N0 pyriform fossa Bilateral level II–IV
T 3/4 N0 postcricoid/posterior Bilateral levels II–IV. Include bilateral RP nodes if posterior
pharyngeal wall pharyngeal wall involved. Include level VI nodes if tumour
extends into cervical oesophagus
Pyriform fossa N Bilateral level II–IV. Level Ib and V nodes on the side of
any lymphadenopathy. Bilateral RP nodes if N2b/3 or if
posterior pharyngeal wall involved
Postcricoid/posterior pharyngeal Bilateral level II–IV, RP nodes. Level Ib and V nodes on the
wall N side of any lymphadenopathy. Level VI nodes if tumour
extends into the cervical oesophagus
RP, retropharyngeal.
Figure 12.4 GTV, CTV70 and CTV44 for a

T1N1 right pyriform fossa tumour.
■ CTV for adjuvant radiotherapy

After careful discussion with the surgeon and pathologist, the CTV60 is defined as
sites of possible residual microscopic disease. If a large resection with
reconstruction and flaps has been performed, the anatomy visible on the planning
CT will be very different from that on the initial diagnostic images.
151
HYPOPHARYNX
The CTV60 should include the margins of resection and sites of any dissected
nodal levels where there was tumour. Sites of positive resection margins or where
there was extracapsular nodal spread should be further defined as CTV66. In
practice it can be impossible to accurately define these sites separately from the
CTV60, in which case only the CTV66 should be defined. When the patient has
had a laryngectomy, the stoma should be included in the CTV60 if subglottic
extension was present or if the surgeon is concerned about the risk of parastomal
recurrence.
If a clinically node negative tumour has been excised with an elective neck
dissection and there is unexpected tumour in the neck nodes, a CTV44 can also be
defined. This should ensure that the nodal levels defined in Table 12.2 have been
treated either surgically or with radiotherapy. For example if surgery for a T2 pyriform
fossa tumour included an ipsilateral level II–IV neck dissection at which multiple
involved nodes were identified (pN2b), the CTV44 should include ipsilateral levels
Ib and V, contralateral levels II–IV and bilateral retropharyngeal nodes.
■ CTV for palliative radiotherapy

The goal of palliative radiotherapy is to treat all symptomatic disease with minimal
toxicity. The GTV is defined from clinical information and imaging. If treatment
is planned on the simulator, a 15 mm margin from GTV to beam edge (50 per cent)
is recommended. If CT planning is used, the GTV should be expanded by 5 mm
to produce the CTV.
■ Planning target volume

CTVs are expanded isotropically to form PTVs by a margin depending on the
measured random and systematic errors in the department – usually 3–5 mm.
Dose solutions
■ Conformal
Both PTV44 and high dose volumes often extend bilaterally into the neck. The usual
conformal plan for many hypopharyngeal tumours uses opposing lateral photon
beams shaped to the PTV with MLCs. Wedges in both the AP and superoinferior
planes may be needed to compensate for changes in the contour of the neck.
If the high dose PTV extends posterior to the plane of the spinal cord a two-
phase technique will be needed. Opposing lateral beams in phase 1 will include the
spinal cord. A second phase using smaller lateral photon beams with posterior
border anterior to the spinal cord is then matched to electron beams to the
posterior neck. The electron energy is chosen to keep within spinal cord tolerance
but cover PTV if possible. Alternatively, the opposing photon beams can be angled
to treat the PTV while avoiding the cord (Fig. 12.5).
The PTV usually extends inferiorly to below the level of the shoulders. The
inferior part of the volume can be treated with a matched anterior photon beam
with midline shielding. The match plane should ideally be inferior to the lower
border of the high dose volume to avoid matching through sites of macroscopic
or high risk microscopic disease. An alternative is to angle the opposing lateral
152
Dose solutions
(a)
(b)
Figure 12.5 Treatment of a

T4aN1 posterior pharyngeal
wall tumour with
radiotherapy. In this patient
with significant comorbidity,
a decision was made not to
treat the level II nodes to
reduce acute effects.
(a) Axial slice to show phase
1 beams angled posteriorly.
(b) Coronal slice to show
phase 1 beams angled
inferiorly. (c) Axial slice to
show phase 2 beams angled
(c) to avoid the spinal cord.
153
HYPOPHARYNX
beams caudally by 10–30°. Superoinferior wedges compensate for the change in

contour of the neck but it can still be difficult to produce a homogeneous dose
distribution for adequate dose to the inferior part of the volume. An anterior neck
beam can help to achieve this.
If the volume extends both posterior to the cord and inferior to the shoulders
there is no conformal option other than to use opposing photon and posterior
electron beams and a matched anterior neck beam. If the match plane is through
a high risk site, it should be moved once during the course of treatment to reduce
the risk of under- or overdosing at the junction.
If the volume extends unilaterally (e.g. lateral T1/2 N0 pyriform fossa),
a unilateral beam arrangement will provide a more conformal dose distribution
whilst sparing contralateral pharyngeal mucosa and salivary glands. A plan with
three beams can be used with beam angles chosen to avoid the spinal cord with
the posterolateral beam and to minimise dose to the uninvolved mucosa.
■ Complex
IMRT has several advantages. No anterior neck beam matching is needed and
therefore the risk of under or overdosing at the match plane is reduced. Matched
electron beams to treat posteriorly behind the cord will not be needed. It can
provide a more even dose distribution if the volume extends inferiorly into the low
neck or superior mediastinum. It facilitates dose escalation by allowing smaller,
high dose volumes to be treated in one phase with a simultaneous boost technique.
It can reduce the risk of xerostomia by sparing salivary glands.
A seven-beam coplanar technique is usually needed to achieve these aims
though five beams are sometimes adequate. Spinal cord (with a 5 mm isotropic
margin), parotid glands and brainstem are contoured as organs at risk. Oral or
pharyngeal mucosa outside the PTV should be contoured to allow these sites to
be spared as much as possible.
■ Conventional
Patients having palliative radiotherapy can be treated conventionally with lateral
opposing beams chosen in the simulator. If the volume extends inferior to the
shoulders or for any curative treatment, a conformal CT based plan is recommended.
Dose-fractionation
■ Curative treatment T1/T2 N0
■ Curative treatment T3/4 or Nⴙ

70 Gy in 35 daily fractions given in 7 weeks concomitant cisplatin weekly.
or
alternative fractionation, e.g. with concomitant boost.
44 Gy in 22 daily fractions to PTV44 in 4 1⁄2 weeks.
154
Information sources
60–66 Gy in 30–33 daily fractions given in 6–61⁄2 weeks concomitant
chemotherapy.

65 Gy to high dose PTV and 54 Gy to low dose PTV in 30 fractions given in 6 weeks.
or 20 Gy in 5 fractions of 4 Gy given in 1 week.

Weekly assessment by doctor or radiographer and dietician is important. Regular
speech and language therapist advice can help maintain safe, functional
swallowing. Most patients are at high risk of weight loss and should be considered
for a gastric feeding tube inserted for prophylaxis before radiotherapy. Further
details are given in Chapter 8.
Verification
See Chapter 8.
key trials
See chapter 8 – many phase III trials in head and neck cancer contain patients
with hypopharyngeal cancers.
Lefebvre JL, Chevalier D, Lunoinski B et al. (1996) Larynx preservation in
pyriform sinus cancer: preliminary results of a European Organisation for
Research and Treatment of Cancer phase III trial. EORTC Head and Neck
Cancer Cooperative Group. J Natl Cancer Inst 88: 890–9.
Information sources
Mendenhall WM, Parsons JT, Stringer SP et al. (1993) Radiotherapy alone or combined with neck
dissection for T1–T2 carcinoma of the pyriform sinus: an alternative to conservation surgery. Int
J Radiat Oncol Biol Phys 27: 1017–27.
Montgomery PQ, Rhys Evans PH, Henk JM (2003) Tumours of the hypopharynx. In: Rhys Evans
PH, Montgomery PQ, Gullane PJ (eds) Principles and Practice of Head and Neck Oncology.
Martin Dunitz, London, pp. 253–78.
155
13 Nasopharynx

■ Curative
Radiotherapy is the principal treatment modality in nasopharyngeal cancer (NPC)
and is often combined with chemotherapy in advanced disease (stage IIb, III and
IV). The role of surgery is confined to neck dissections for persistent or recurrent
lymphadenopathy or, rarely, to salvage recurrent nasopharyngeal disease.
NPC is endemic in southern China, relatively common around the Mediterranean,
and uncommon in northern Europe and the USA. The WHO classification identifies
three histological subtypes. Type 1 (keratinising squamous cell carcinoma) is more
common in the West but less common in endemic areas, and local control is harder
to achieve. Because there is a relatively low incidence of occult cervical lymph node
involvement in WHO1 disease with no palpable lymphadenopathy, elective neck
irradiation can be less extensive. Types 2 (non-keratinising differentiated carcinoma)
and 3 (undifferentiated) make up most of endemic cases. While lymphadenopathy at
presentation is common, these tumours are relatively sensitive to chemotherapy and
radiotherapy, so local control rates are good. Failure, particularly in WHO3 tumours,
is often with metastases.
Genetic, environmental and dietary factors may all play a role in the aetiology of
NPC. High levels of Epstein–Barr virus antibodies are found in the WHO3
subtype and can correlate with stage and response to treatment.
The most frequently used staging systems are the UICC TNM and Ho’s
staging. T staging reflects local invasion as described below. N staging criteria in
the UICC system are different from those for other head and neck cancers to
reflect the high incidence of lymphadenopathy and relative sensitivity to radiation
(Table 13.1).
Overall 5-year survival rates are about 50 per cent, but different staging systems
and case mix make comparisons between series difficult. T1/2 N0/1 disease
(stage I and II in most systems; Table 13.2) has a 5-year local control rate of
70–90 per cent with radiotherapy alone. Chemotherapy is recommended in
addition to radiotherapy for stage IIb–IV disease, to achieve 5-year local control
rates of 50–70 per cent.
■ Recurrent disease
Local recurrence in the nasopharynx is difficult to treat surgically. Brachytherapy
with iodine-125 or gold-198 seeds can be used to treat small volume persistent or
recurrent disease, or can be combined with surgery. However, most local recurrences
are detected when they are too large to be cured with brachytherapy.
156
Table 13.1 Comparison of UICC TNM (6th edn, 2002*) and Ho’s (1978) staging for
nasopharyngeal cancer
Stage UICC TNM (6th edn, 2002) Ho’s (1978)
T1 Tumour confined to nasopharynx Tumour confined to nasopharynx

T2 Tumour extends to soft tissues. Extension to the nasal fossa, oropharynx, or
T2a – extends to oropharynx and/or nasal adjacent muscles or nerves below the
cavity without parapharyngeal extension. base of the skull
T2b – with parapharyngeal extension
T3 Tumour invades bony structures and/or Beyond T2 limits. T3a Bone involvement of
paranasal sinuses the base of the skull – includes floor of the
sphenoid sinus. T3b Involvement of the base
of the skull. T3c Involvement of the cranial
nerve(s). T3d Involvement of the orbits,
laryngopharynx, or infratemporal fossa
T4 Tumour with intracranial extension Not defined
and/or involvement of cranial nerves,
infratemporal fossa, hypopharynx,
orbit or masticator space
N1 Unilateral metastasis in nodes Node(s) wholly in the upper cervical level

6 cm in greatest dimension, bounded below by the skin crease extending
above the supraclavicular fossa laterally and backward from or just below
the thyroid notch (laryngeal eminence)
N2 Bilateral metastasis in nodes
6 cm in Node(s) palpable between the crease and
greatest dimension, above the the supraclavicular fossa
supraclavicular fossa
N3 Metastasis in nodes 6 cm in dimension Supraclavicular fossa nodes and/or skin
or in the supraclavicular fossa. N3a – involvement
6 cm in dimension. N3b – in the
supraclavicular fossa
*With permission.
Table 13.2 Stage groupings for UICC TNM
(6th edn, 2002*) classification
Stage I T1 N0 M0
Stage IIA T2a N0 M0
Stage IIB T1 N1 M0
T2a N1 M0
T2b N0–1 M0
Stage III T1 N2 M0
T2a–2b N2 M0
T3 N0–2 M0
Stage IVA T4 N0–2 M0
Stage IVB Any T N3 M0
Stage IVC Any T any N M1
*With permission.
157
NASOPHARYNX
Re-irradiation to a curative dose can be considered. It provides cure in 35 per cent

but must be balanced against the high rate (at least 30 per cent) of serious late
effects, including bone necrosis and temporal lobe damage. An interval of more than
2 years from first treatment, and highly conformal planning are desirable.

The relative chemo-sensitivity of nasopharyngeal tumours has led to the use of
chemoradiation to try to improve local control and overall survival. Neoadjuvant
or adjuvant chemotherapy, concomitant (usually cisplatin) chemotherapy, and a
combination of both have all been reported to improve local control and overall
survival rates. The relative chemo-sensitivity of NPC and high rate of distant
metastases makes neoadjuvant treatment theoretically attractive, but there is so far
no proof of benefit from clinical trials. For stage IIb–IV disease, three cycles of
cisplatin-based induction chemotherapy are given initially (except in patients with
significant comorbidity), followed by radiation with concomitant cisplatin either
weekly or 3-weekly for all patients.

The nasopharynx is a mucosa-lined space behind the nasal cavities and above the
oropharynx. Understanding the anatomy of this region helps correlate presenting
symptoms with local invasion and is vital in order to define the CTV accurately
(Fig. 13.1). Tumours most commonly arise in the roof or lateral wall – often the
fossa of Rosenmüller behind the Eustachian tube orifice. Tumour can spread via
the mucosa or submucosa, to invade the nasal cavity anteriorly or the oropharynx
inferiorly. The lateral wall is formed by the pharyngeal fascia which offers
relatively little resistance to tumour spread. Deep to the fascia is the
parapharyngeal space containing the lateral retropharyngeal lymph nodes (nodes
of Rouvière), cranial nerves IX–XII, carotid artery and internal jugular vein.
Direct extension or nodal involvement can lead to IX–XII cranial nerve palsies.
Tumour can grow out of the parapharyngeal space superiorly into the middle
cranial fossa, and anteriorly into the pterygopalatine fossa and inferior orbital
fissure towards the orbit.
The roof slopes downwards to become the posterior wall. Medially it is formed
by the sphenoid sinus and laterally by the foramen lacerum at the skull base. This
provides relatively little barrier to local invasion into the cavernous sinus, which
contains cranial nerves III–VI and the internal carotid artery. Skull base invasion
occurs in 30 per cent of cases.
The nasopharynx drains directly to both the lateral retropharyngeal nodes in
the parapharyngeal space and to level II and upper level V nodes. Ipsilateral
lymphadenopathy is detected in 60–90 per cent of patients at diagnosis and
50 per cent have involved contralateral nodes. The lateral retropharyngeal nodes
lie medial to the internal carotid artery and anterior to the spinal column from
the occiput to C3. These nodes are not palpable and should be assessed on MRI
or CT, with a node 5 mm considered pathological. Involved cervical nodes are
usually palpable but the whole neck and supraclavicular fossae should be imaged.
158
(a) (b)
(c)
Figure 13.1 Diagnostic MR scans to illustrate nasopharyngeal anatomy and routes of
tumour spread. (a) Axial T1-weighted contrast-enhanced MRI showing tumour (T) invading
parapharyngeal space (P). The normal contralateral fossa of Rosenmüller (R) and normal
Eustachian tube orifice (E) are also shown. (b) Coronal T1-weighted contrast-enhanced
MRI showing cavernous sinus invasion (C), the sphenoid sinus (S) and internal carotid
artery (IC). (c) Axial T1-weighted contrast-enhanced MRI showing bilateral retropharyngeal
nodes (arrowed).
Distant metastases are uncommon at presentation. All patients should have a

chest radiograph performed but chest CT, liver imaging and bone scans are only
required in the presence of advanced disease (N3) or symptoms and signs suggestive
of metastases.
159
NASOPHARYNX

Careful clinical examination and cross-sectional imaging are required to assess local
invasion and lymph node spread. Particular attention should be paid to sites where
local invasion is suspected on the basis of clinical symptoms and signs. Clinical
examination should include a nasendoscopy or EUA to assess mucosa, submucosa
and extent of spread in the nasopharynx. Documentation of cranial neuropathies
can direct imaging assessment. MRI is preferred to CT with contrast as it provides
better imaging of soft tissue invasion, particularly the parapharyngeal space and
retropharyngeal lymph nodes. CT with bone windows is better for detecting bone
involvement and may also be required if skull base invasion is suspected.
All patients should have a dental assessment and OPG as xerostomia after
treatment is common. A baseline audiogram is useful as tumour, radiotherapy and
chemotherapy can all contribute to hearing loss.
Data acquisition
■ Immobilisation
Patients are treated supine with head and shoulders immobilised in a Perspex shell
or thermoplastic mask with at least five fixation points. The chin is elevated to
spare the oral cavity and orbit, but the spine should be kept as straight as possible
if posterior neck nodes are present, to facilitate matching of an electron boost.
A mouth bite may be used to depress the tongue away from the treated volume. If
IMRT is used, the patient can be immobilised with the chin in a neutral position.
■ CT scanning
CT scan slices measuring 3–5 mm are obtained from 2 cm above the superior orbital
ridge (to include the skull base) to the arch of the aorta inferiorly. Intravenous
contrast may help definition of cervical nodes. Reference marks are placed on the
shell at the CT visit to aid verification.
■ Simulator
3D conformal radiotherapy or an IMRT technique is preferred because local extent
of disease can be better determined. However, Ho’s technique (modified here
slightly) relates likely patterns of spread to bony anatomy, so 2D simulator-based
planning with opposing lateral fields can be used.

The GTV is first contoured on the planning CT using diagnostic images and clinical
information. Particular attention should be given to the parapharyngeal space as
described above, and to the lateral pharyngeal lymph nodes. Retropharyngeal nodes
5 mm and cervical nodes 10 mm in short axis diameter are contoured as GTV.
If induction chemotherapy has been used, the GTV should reflect the initial sites
of disease.
Three CTVs are defined: a high dose CTV70 reflecting the clinically apparent
disease; a high risk CTV60 reflecting the high risk of local spread in and adjacent
160
to the nasopharynx; and a prophylactic CTV50 to treat at risk but clinically
uninvolved nodes.
The GTV is expanded isotropically by 5 mm to form the CTV70 which is then
edited to reflect natural tumour barriers. In particular, the posterior margin can
be reduced if the vertebral column is not involved, and this will help minimise
brainstem dose.
The CTV70 is copied to form the CTV60 which is expanded to reflect possible
local spread in the nasopharynx. It should include the whole nasopharynx, adjacent
retropharyngeal lymph nodal regions, parapharyngeal space, pterygoid plates,
pterygomaxillary fissures, floor of the sphenoid sinus, foramen lacerum and the
posterior part of the nasal cavity (5 mm anterior to the GTV) (Fig. 13.2). If a T1
primary tumour is small and well defined on imaging, some of these sites can be
excluded from the CTV60 (e.g. contralateral parapharyngeal space).
(a) (b)
Figure 13.2 GTV, CTV70 (cyan) and CTV60 (purple) for a T2b nasopharyngeal cancer
shown on two axial levels: (a) and (b).
The CTV60 is copied to form the CTV50. This is expanded to include bilateral
level Ib and II–V nodes. In N0 disease, level Ib can be omitted bilaterally. In N1
disease, the contralateral level Ib can be omitted. In the rare situation of WHO1
disease that is clinically N0, level IV and inferior level V nodes can also be omitted.
If level IV or low level V nodes are involved the supraclavicular nodes on that side
should be included in the CTV50. A CTV-PTV margin is applied (usually
3–5 mm) based on measured set-up errors assuming no tumour motion.
The brainstem and spinal cord should always be defined. The parotid glands,
pituitary gland, temporomandibular joints (TMJs), optic apparatus, temporal lobes
and middle ear apparatus (adjacent to the fossa of Rosenmüller) should also be
contoured as critical structures if IMRT is used, as dose sparing may be possible.
161
NASOPHARYNX
Dose solutions
■ Conformal
As the PTV50 includes both a midline tumour and bilateral cervical nodes, it is
difficult to obtain adequate coverage conformally with anything other than opposing
lateral photon beams. Conformal planning is still preferable as it will allow MLC
shielding to shape to PTVs rather than to bony anatomy. Treatment of the PTV60
and PTV70 with a conformal plan will reduce the dose to adjacent critical structures
such as the brainstem and temporal lobes.
PTV50 is initially treated to 40 Gy in 20 fractions with opposing lateral photon
beams as shown (Fig. 13.3). MLC shielding is applied to oral cavity, orbit and
brain. This is matched to a bilateral anterior neck beam with the match plane at
least 10 mm below the nodal GTV.
Figure 13.3 BEV of lateral beam for

phase 1 treatment of nasopharyngeal
cancer.
The second phase treats the nasopharynx (PTV60) with a further 20 Gy in 10

fractions with a three-beam plan using an anterior infraorbital in addition to small
lateral beams to reduce temporal lobe and TMJ dose. This arrangement is also
used for the final 10 Gy in 5 fractions with additional shielding to reduce the
treated volume if possible (Fig. 13.4).
The second phase also includes a matched anterior neck beam covering the
elective nodal PTV50 bilaterally. This can be reduced to cover sites of involved
nodes only (nodal PTV70) from 50 Gy. If the nodal PTV70 is small, it can be
treated with electrons from 50 Gy to 70 Gy.
An anterior neck beam prescribed to Dmax will not cover all the cervical node
PTV with the 95 per cent isodose, but this arrangement has been used for decades
without a high incidence of isolated nodal recurrence, so the dose delivered seems
to be adequate for tumour control. If a higher dose to the nodes is required, for
example when level V nodes are present, either the prescription point can be
changed to 3 cm (which will produce a hot-spot of 115 per cent) or opposing
anterior and posterior fields weighted anteriorly can be used (which will increase
the volume of normal tissue treated to a high dose).
162
Dose-fractionation
Figure 13.4 Conformal treatment of the PTV70 for phase 3 in nasopharyngeal cancer.
There is evidence that toxicity (especially xerostomia) is reduced using a more

complex arrangement of five to six coplanar beams, sometimes with an additional
non-coplanar vertex beam.
■ Complex
The complex PTVs and adjacent critical normal structures make IMRT of the
nasopharynx a good solution in theory. Five or seven equally spaced coplanar
IMRT beams can provide good target volume coverage, keep neural tissues within
tolerance and spare parotid function to prevent long-term xerostomia. Several
dose levels can be stipulated in the plan so that the three PTVs can be treated to
different doses in a single phase. The low neck can be included in the IMRT
volume or treated with a matched anterior neck beam as above.
■ Conventional
may be unavoidable. Ho’s technique describes a similar solution to the conformal
one described above with bony landmarks used to define beam edges. For example
beam borders for the first phase of treatment are: anterior – bisecting antrum;
posterior – 2 cm posterior to nodes; superior – 5 mm above anterior clinoid.
Dose-fractionation
65 Gy to PTV70, 60 Gy to PTV60 and 50.4 Gy to PTV50 in 30 fractions given in
6 weeks.
■ Conformal/conventional
70 Gy in 35 daily fractions given in 7 weeks to PTV70 concomitant cisplatin
(60 Gy/30fractions and 50 Gy/25 fractions to PTV60 and PTV50, respectively).
163
NASOPHARYNX

A dental assessment is carried out before radiotherapy because of the risk of long-
term xerostomia. We recommend that a prophylactic feeding tube is inserted
before radiotherapy in all patients with N2/3 disease and is considered for others
thought to be at increased risk of weight loss during treatment.
See Chapter 8 for details.
Verification
See Chapter 8.
key trials
Al-Sarraf M, LeBlanc M, Giri PG et al. (1998) Chemoradiotherapy versus

radiotherapy in patients with advanced nasopharyngeal cancer: Phase III
randomised Intergroup study 0099. J Clin Oncol 16: 1310–17.
Information sources
Chau RMC, Teo PML, Choi PHK et al. (2001) Three-dimensional dosimetric evaluation of a
conventional radiotherapy technique for treatment of nasopharyngeal carcinoma. Radiother
Oncol 58: 143–53.
Jen Y-M, Shih R, Lin YS et al. (2005) Parotid gland-sparing 3-dimensional conformal radiotherapy
results in less severe dry mouth in nasopharyngeal cancer patients: a dosimetric and clinical
comparison with conventional radiotherapy. Radiother Oncol 75: 204–9.
Lee AWM, Poon YF, Foo W et al. (1992) A retrospective analysis of 5037 patients with
nasopharyngeal carcinoma treated during 1976–1985, overall survival and patterns of failure.
Int J Radiat Oncol Biol Phys 23: 261–70.
Liu L-Z, Zhang GY, Xie CM et al. (2006) Magnetic resonance imaging of retropharyngeal lymph node
metastasis in nasopharyngeal carcinoma: patterns of spread. Int J Radiat Oncol Biol Phys 66:
721–30.
Wolden SL, Chen WC, Pfister DH et al. (2006) Intensity-modulated radiation therapy (IMRT) for
nasopharyngeal cancer: update of the Memorial Sloan-Kettering experience. Int J Radiat Oncol
Biol Phys 64: 57–62.
164
Larynx 14

■ Early stage squamous cell cancer of glottic
larynx (T1–2, N0)
Squamous cell carcinomas confined to the glottic larynx can be treated with either
surgery (often laser excision) or radiotherapy. Large retrospective series suggest
equivalent cure rates as long as radiotherapy is followed by close surveillance to
detect and treat recurrences. Radiation alone gives 5-year local control rates of
75–90 per cent in T1 tumours. Local surgical and radiotherapeutic expertise,
patient choice and likely voice quality all influence the treatment decision. If the
anterior commissure is involved, voice quality with surgery may be worse as it can
be more difficult to oppose the vocal cords after resection.
■ Early stage squamous cell cancer of supraglottic

larynx (T1–2, N0)
The supraglottic larynx has a richer lymphatic drainage than the glottic larynx.
Although surgery and radiotherapy have equal cure rates, the ability to preserve
organ function and to treat the adjacent neck nodes means radiotherapy is
preferred.
■ Advanced laryngeal cancer (T3–4, Nⴙ)

The preferred treatment for many years for advanced laryngeal cancer has been
surgery (total laryngectomy and neck dissection) with adjuvant radiotherapy in
selected cases. In practice, adjuvant radiation is recommended to the primary site
in T4 cancer or where resection margins are close or involved and to the neck in
N2–3 disease or in N1 disease with extracapsular nodal spread. Adjuvant
radiotherapy is therefore recommended for the majority of patients who have a
laryngectomy.
An alternative approach is organ preservation – initial radiotherapy, usually
combined with chemotherapy with laryngectomy reserved for recurrence. There
are phase III data to support this approach, provided there is careful follow-up to
detect recurrence and allow salvage surgery. Without this, (chemo)radiation will
produce inferior cure rates compared with laryngectomy.
If the primary tumour invades through the laryngeal cartilage, laryngectomy is
preferred. In other cases we recommend initial radiochemotherapy with close
follow-up and salvage laryngectomy in case of residual or recurrent disease.
165
LARYNX
■ Palliative EBRT
If distant metastases are present initially or if the patient is not suitable for curative
treatment, palliative EBRT may improve pain and reduce the chance of laryngeal
obstruction or tumour ulceration in the neck.

Locally advanced laryngeal disease is often treated with surgery, radiotherapy and
chemotherapy. Initial laryngectomy may be followed by adjuvant radiotherapy
with concomitant chemotherapy for selected patients (see Chapter 8). Concomitant
radiochemotherapy can be used as initial treatment with surgery (laryngectomy
and/or neck dissection) for residual or recurrent disease.
Induction (neoadjuvant) chemotherapy has been used to predict response to
radiation and to select patients for organ preservation. Two to three cycles of
cisplatin and 5-fluorouracil are given with assessment of disease response at each
cycle. Most patients respond and then have organ preservation with radiotherapy,
but if response is poor, a laryngectomy is recommended. The role of concomitant
radiochemotherapy after induction chemotherapy is not established.
Exophytic glottic and subglottic tumours may present with stridor. Even if
organ preservation is the preferred treatment, surgical debulking may be required
initially to preserve a clear airway.

The larynx is divided into three subsites: the supraglottic larynx (epiglottis, false
cords, ventricles, aryepiglottic folds and arytenoids), glottic larynx (true cords) and
subglottic larynx (from the under surface of the cords to the inferior border of the
cricoid cartilage) (Table 14.1 and Fig. 14.1). Primary tumours can spread
mucosally or submucosally between these subsites or to the adjacent oropharynx
or hypopharynx. They can invade deep structures including the laryngeal cartilages,
Table 14.1 UICC TNM (6th edn, 2002*) tumour staging for supraglottic and glottic cancer
Supraglottis (subsites – suprahyoid epiglottis, aryepiglottic fold, arytenoids, infrahyoid

epiglottis, false cords)
T1 Tumour limited to one subsite with normal vocal cord mobility

T2 Tumour invades mucosa of more than one adjacent subsite of supraglottis or glottis or region
outside the supraglottis (e.g. mucosa of base of tongue) without fixation of the larynx
T3 Tumour limited to the larynx with vocal cord fixation and/or invades postcricoid areas,
pre-epiglottic tissues, paraglottic space, and/or with minor thyroid cartilage erosion (e.g.
inner cortex)
T4a Tumour invades through the thyroid cartilage and/or invades tissues beyond the larynx
e.g. trachea, soft tissues of the neck including deep/extrinsic muscle of tongue, strap
muscles, thyroid, oesophagus
T4b Tumour invades prevertebral space, mediastinal structures or encases carotid artery
(i.e. unresectable)
(Continued)
166
Table 14.1 Continued
Glottis (subsites – vocal cords, anterior commissure, posterior commissure)
T1 Tumour limited to vocal cords with normal mobility

T1a – limited to one vocal cord
T1b – involves both vocal cords
T2 Tumour extends to supraglottis and/or subglottis and/or with impaired vocal cord mobility
T3 Tumour limited to the larynx with vocal cord fixation and/or invades paraglottic space, and/or
with minor thyroid cartilage erosion (e.g. inner cortex)
T4a Tumour invades through the thyroid cartilage and/or invades tissues beyond the larynx,
e.g. trachea, soft tissues of the neck including deep/extrinsic muscle of tongue, strap
muscles, thyroid, oesophagus
T4b Tumour invades prevertebral space, mediastinal structures or encases carotid artery
(i.e. unresectable)
*With permission.
Vallecula
Free portion
of epiglottis
Hyoid bone
Pre-epiglottic fat Laryngeal wall
of epiglottis
Aryepiglottic fold
False cord
Laryngeal ventricle
Arytenoid
Thyroid cartilage
True cord
Subglottic area
Prelaryngeal Sections of
node cricoid cartilage
Figure 14.1 Midline sagittal view of the larynx.
pre-epiglottic and paraglottic fat, or into nearby structures such as the carotid
sheath or prevertebral muscles.
The glottic larynx has very few lymphatics so early tumours confined to this site
are usually N0 and adjacent nodes do not require treatment. The prelaryngeal
(Delphian) node lies on the cricothyroid membrane and is usually included in
radiotherapy treated volumes incidentally. Tumours originating in the supraglottic
larynx can spread to adjacent level II or III nodes unilaterally or bilaterally if the
primary tumour crosses the midline. Locally advanced cancers spread to adjacent
level II–IV nodes. If there is subglottic involvement, tumours can spread to level
IV or midline level VI nodes.
167
LARYNX

For early disease, clinical assessment of local invasion and vocal cord mobility is
more important than cross-sectional imaging. Most patients have an EUA to
obtain biopsies and to evaluate supraglottic and subglottic spread, although a
transnasal oesophagoscopy can be used instead. Both oncologist and surgeon
should examine the larynx with a nasendoscope or indirect laryngoscopy to
visualise the primary tumour and to assess vocal cord mobility (Fig. 14.2).
(a) (b)
Figure 14.2 Nasendoscopy of the larynx. (a) T1a tumour (T) of the anterior half of the right
vocal cord. (b) T3 tumour (T) of the left vocal cord extending into the supraglottic larynx.
T1N0 squamous cell cancer confined to one vocal cord does not require cross-
sectional imaging as the risk of deep invasion or nodal spread is very small. All other
laryngeal cancer requires cross-sectional imaging with CT or MRI. Invasion of
paraglottic fat and laryngeal cartilages may be assessed by CT or MRI (Fig. 14.3).
Ultrasound can be useful particularly to help assess cartilage sclerosis where cartilage
invasion is suspected but not proven with MRI or CT.
Data acquisition
■ Immobilisation
All patients having radiotherapy should be immobilised in a Perspex or thermoplastic
shell fixed to the couch in at least five places. The spine should be straight. The
shoulders are immobilised in the shell as inferiorly as possible so that the shoulder
tips are inferior to the lower border of the cricoid cartilage thus permitting lateral
radiation beams to treat the larynx without the need to angle them inferiorly. Grip
bars on the side of the couch may help to achieve this.
■ Simulator
Radiotherapy for early (T1) glottic carcinomas with no extra-laryngeal disease
can be planned on a simulator because the position of the primary tumour
can be determined – the vocal cords extend horizontally from 10 mm below the
thyroid promontory – and lymph nodes do not require treatment. The whole
larynx can therefore be treated with opposing lateral beams with the following
168
Data acquisition
(a) (b)
(c)
Figure 14.3 Cartilage invasion in laryngeal tumours. (a) Diagnostic CT scan showing
tumour (arrowed) with possible invasion of the thyroid cartilage. (b) Same slice with
window and level optimised for bone/cartilage to show cartilage invasion. (c) T4a tumour
(T) invading through the cricoid cartilage (CC) into the soft tissues of the neck.
borders: superior – mid-body of the hyoid, inferior – inferior margin of cricoid

cartilage, anterior – to cover skin, posterior – anterior vertebral column. A 1 cm
thick tissue equivalent bolus should be applied anteriorly if tumour extends to the
anterior commissure. As the larynx moves superiorly on swallowing, fluoroscopy is
used in the simulator to ensure that the glottis remains within the treated volume
when the patient swallows.
■ CT scanning
CT slices 3–5 mm thick are obtained from the base of skull to the top of the aortic
arch with the patient immobilised in the treatment position. As treatment of
locally advanced glottic cancer or adjuvant radiation after a laryngectomy may
require lateral beams angled inferiorly, the CT scan in these patients should be
extended inferiorly to the carina.
169
LARYNX

■ T1 N0 glottic larynx
As conventional treatment with small lateral beams is usually successful and well
tolerated, target volumes should be defined with these borders in mind. Records and
pictures from nasendoscopy and EUA should be available as small primary vocal cord
tumours are difficult to see on CT scan. GTV is defined as the site of primary
tumour. CTV is the glottic larynx with a border at least 10 mm superiorly and
inferiorly from the GTV. Axially the bilateral mucosal laryngeal surface is included as
field change is common. PTV is CTV with a 3–5 mm isotropic margin depending on
local assessment of random and systematic errors assuming tumour movement is
minimal. Motion can be assessed by watching laryngeal movement with fluoroscopy
in the simulator to ensure that the vocal cords remain within the PTV on swallowing.
In practice, it can be easier to define the superior (mid-body of hyoid) and
inferior (inferior margin of cricoid) beam borders on the lateral DRR from the
planning CT scan and then check on the axial slices that the whole glottic larynx
is included within the CTV, assuming a 5 mm penumbra from PTV to beam edge.
Using smaller beams defined solely on the basis of GTV-CTV-PTV definition has
little therapeutic advantage and increases the possibility of a geographical tumour
miss due to intra-fractional motion from swallowing.
■ Other curative radiotherapy

The GTV is defined on the planning CT scan as the primary tumour and any
involved lymph nodes. There are no clinicopathological studies to help determine
the GTV-CTV70 margin, but in practice the GTV is grown isotropically by 10 mm
and then edited to take into account likely patterns of spread and barriers to
tumour growth. The high dose CTV70 also includes lymph nodes of normal size
that are at high risk of having microscopic tumour involvement – in practice this
includes the level III nodes immediately adjacent to the primary tumour.
Table 14.2 Suggested nodal CTVs for laryngeal cancer
Site and stage CTV70 nodal levels CTV44 nodal levels
T1 N0 glottis None None

T2 N0 glottis Level III nodes immediately adjacent to tumour None
T1/2 N0 supraglottis Level II, III nodes immediately adjacent to tumour Bilateral level II, III
T3/4 N0/1 glottis Involved nodal levels level III nodes immediately Bilateral low level II, levels
adjacent to tumour III–IV. Level VI for subglottic
tumours
T3/4 N2/3 glottis Involved nodal levels level III nodes immediately Bilateral level II–IV. Level V
adjacent to tumour if 1 node involved in that
side of the neck. Level VI
for subglottic tumours
N supraglottis Involved nodal levels level III nodes immediately Bilateral levels II–IV
adjacent to tumour Consider Ib, V depending
on site of primary and of
involved nodes
170
Dose solutions
The CTV44 in the patient with clinically negative nodes includes all nodes at
risk for microscopic metastases as shown in Table 14.2. If adjacent structures are
involved, the CTV44 must reflect their lymphatic drainage, e.g. level II nodes if
there is tongue base invasion.
The CTV after a laryngectomy is determined by the initial site of the tumour and
which local structures were invaded, and by the pattern of lymph node spread found
on initial imaging and at neck dissection. A CTV66 can be defined as sites where
surgical margins were involved or where there was extracapsular nodal spread but
in practice it is often difficult to delineate this volume separately from the CTV60.
Preoperative imaging, clinicopathological correlation, a clear operation note and
discussion with the surgeon all help to determine the CTV60, which includes sites
at risk of microscopic residual disease. Particular thought should be given to
establishing whether there was subglottic invasion as this predicts for recurrence in
the tracheostomy site and mandates inclusion of the stoma in the CTV60.
For a pT4N0 tumour that has been completely excised with a bilateral negative
selective neck dissection, the CTV60 is the tumour bed alone. For node positive
disease, the CTV60 should include all nodal levels which contained disease in the
dissected neck. A separate CTV44 can be defined to treat contralateral nodes that are
at lower risk of metastases if there has not been a selective neck dissection on that side.
Dose solutions
■ T1 N0 glottic larynx
Conventional
The field borders defined above will provide adequate coverage for early glottic
cancers. Opposing lateral beams of approximately 5 cm 5 cm require 15–45°
wedges to compensate for missing tissue anteriorly. A 1 cm tissue equivalent bolus
is needed over the apex of the larynx if the anterior commissure is involved.
Conformal
A more conformal dose distribution can be obtained by either using anterior
oblique rather than lateral beams or by the addition of an anterior beam which can
obviate the need for bolus anteriorly. MLCs are used to conform each beam to the
shape of the PTV. In this way dose to the lateral neck is reduced which may reduce
the risk of damage to the carotid vessels. However, the adjacent lymph nodes will
not receive a therapeutic dose as they would with an opposing beam arrangement
(Fig. 14.4). If there is any question of supraglottic involvement the adjacent nodes
must be included in the CTV and the volume treated accordingly.
■ Other tumours
Conformal
For a PTV that extends bilaterally and involves midline structures, opposing or
slightly angled lateral photon beams may be the only way to provide adequate
coverage. The beams are shaped to the PTV contour with MLC leaves to improve
171
LARYNX
(a)
(b)
Figure 14.4 Conformal radiotherapy for T1a cancer of the left glottis with three beams.
(a) Axial slice showing beams and dose distribution. (b) Sagittal slice. The superior beam
border is the mid-body of the hyoid bone and the inferior border is the inferior margin of
the cricoid cartilage.
the therapeutic ratio. When the PTV extends inferiorly below the shoulders, the
opposing lateral beams are angled inferiorly by 10–30° to provide dose to the low
neck. Because the contour of the neck changes both from anterior to posterior and
also from superior to inferior, the beams are wedged in two planes to provide a
conformal distribution. This is impossible with MLCs, so it is necessary to use two
identical sized fields for each lateral beam – each with a wedge in a different plane.
An anterior beam can also improve the dose distribution in some plans. Even with
these modifications it can be difficult to get a uniform dose distribution and
particularly to get adequate dose to the inferior part of the PTV.
172
Dose solutions
Where a single PTV anterior to the cord is defined one beam arrangement is
used throughout (Fig. 14.5). If a PTV70 and PTV44 are defined two phases are
used, usually with similar beam angles but different MLC positions (Fig. 14.6). If
the high dose PTV extends posterior to the spinal cord on one side, the lateral
beams can sometimes be angled by 20–30° which may also allow some sparing of
the contralateral parotid gland if the PTV extends superiorly. For some volumes,
matched electron beams may be required to treat the posterior extent of high dose
PTVs in phase 2. The posterior border of the opposing photon beams is set just
anterior to the spinal cord and the electron energy is chosen to provide as good
coverage of the PTV as possible without exceeding spinal cord tolerance.
(a)
Figure 14.5 Adjuvant radiotherapy for a

pT4N0 laryngeal tumour. (a) Axial view.
Tissue planes are difficult to define in the
postoperative setting. (b) Sagittal view
showing the stoma (arrowed) included
(b) within the treated volume.
Complex
IMRT has the potential to produce a more conformal dose distribution to the
PTVs particularly if the PTV extends inferiorly below the shoulders or posteriorly
behind the spinal cord. It also means matching within the PTV to an anterior neck
or posterior electron beam is unnecessary. Dose escalation may then be possible.
Parotid sparing can be accomplished if the PTV extends superiorly into level II. A
seven beam coplanar arrangement is likely to produce the best plan though five
beams may be adequate.
173
LARYNX
(a)
(b)
Figure 14.6 Radiotherapy

for T3 N0 carcinoma of the
right glottis. Phase 1 lateral
beams angled (a) posteriorly
and (b) inferiorly. (c) Phase
2 lateral beams to treat the
larynx and adjacent level III
(c) nodes.
174
Verification
Conventional
may be unavoidable with beam arrangements as described above.
Dose-fractionation
■ T1–2 N0 glottic larynx
50 Gy in 16 daily fractions of 3.125 Gy given in 22 days.
■ Other curative treatments

70 Gy in 35 daily fractions given in 7 weeks concomitant cisplatin or alternative
fractionation, e.g. with concomitant boost.
60 Gy in 30 daily fractions given in 6 weeks concomitant cisplatin.

See Chapter 8. Particular care of the tracheostomy site is needed if the stoma is
included in the treated volume. Tracheitis will occur and may make it painful to
replace a stoma button. Desquamation anteriorly can make fixing stoma devices
difficult. These patients should be assessed weekly throughout treatment by a speech
and language therapist. Patients having an intact larynx irradiated will develop varying
degrees of laryngitis and need advice to rest their voice until acute effects subside.
Verification
See Chapter 8.
EaStER – Early Stage glottic cancer: Endoscopic excision or Radiotherapy.

Key trials
Feasibility Study. A randomised, controlled, feasibility study developed by the

EaStER trial management group on behalf of the NCRI Head and Neck Cancer
Clinical Studies Group (in progress).
Forastiere AA, Goepfert H, Maor M et al. (2003) Concurrent chemotherapy and
radiotherapy for organ preservation in advanced laryngeal cancer. N Engl J Med
349: 2091–8.
The Department of Veteran Affairs Laryngeal Cancer Study Group (1991)
Induction chemotherapy plus radiation compared with surgery plus radiation in
patients with advanced laryngeal cancer. N Engl J Med 324: 1685–90.
175
LARYNX
Information sources
Cellai E, Frata P, Magrini SM et al. (2005) Radical radiotherapy for early glottic cancer: Results in a
series of 1087 patients from two Italian radiation oncology centers. I. The case of T1N0
disease. Int J Radiat Oncol Biol Phys 63: 1378–86.
Clark CH, Bidmead AM, Mubata CD et al. (2004) Intensity-modulated radiotherapy improves target
coverage, spinal cord sparing and allows dose escalation in patients with locally advanced
cancer of the larynx. Radiother Oncol 70: 189–98.
Pfister DG, Laurie SA, Weinstein GS et al. (2006) American Society of Clinical Oncology clinical
practice guideline for the use of larynx-preservation strategies in the treatment of laryngeal
cancer. J Clin Oncol 24: 3693–704.
176
Salivary glands 15

Most salivary gland tumours arise in the parotid glands, but tumours of the
submandibular or other minor salivary glands can also occur. Most tumours are
treated with surgery, followed by postoperative radiotherapy when risk of local
recurrence is high. Retrospective series suggest the addition of radiotherapy can
reduce local recurrence rates from 30 per cent to 10 per cent but there is no effect
on overall survival.
Patterns of local and metastatic spread vary with histological subtype. Careful
pathological assessment is important to help predict risk of local recurrence and
the need for adjuvant radiotherapy. Pre- and postoperative discussions with the
surgeon are useful to define extent of surgery and likely sites of macroscopic
or microscopic residual disease though some tumours will only be found to be
malignant at operation. Tumours close to the facial nerve within the parotid gland
may often be excised with positive or very close margins in order to preserve the
nerve, with the expectation that adjuvant radiotherapy will be used. Primary skin
cancers of the head and neck can metastasise to intraparotid lymph nodes, but
radiotherapy for these cancers is considered separately in Chapter 9.
■ High grade tumours (high grade mucoepidermoid,

high grade adenocarcinoma, carcinoma arising from
pleomorphic adenoma)
Adjuvant radiotherapy to the tumour bed is recommended for all high grade
salivary tumours except for T1 tumours completely excised with clear margins.
Ipsilateral neck levels Ib, II and III should be treated prophylactically in view of
the high risk of occult neck node metastases, unless a negative selective neck
dissection has been performed. In node-positive patients, adjuvant radiotherapy is
recommended for N2/3 disease or in the presence of extracapsular spread.
Tumours of intermediate grade should be managed as high grade cancers.
■ Low grade tumours (low grade mucoepidermoid,

low grade adenocarcinoma, acinic cell carcinoma)
Adjuvant radiotherapy is recommended where excision margins are positive or
close (5 mm) after discussion with the surgeon and pathologist. The deep
excision margin close to the facial nerve is usually most critical. The risk of occult
neck metastases is smaller than for high grade tumours, so prophylactic treatment
of the N0 neck is not recommended.
177
SALIVARY GLANDS
■ Adenoid cystic carcinoma

These tumours have a relatively high local recurrence rate and a propensity for
perineural spread, therefore adjuvant radiotherapy is recommended for all adenoid
cystic cancers except in rare T1 tumours without pathological evidence of
perineural invasion.
■ Pleomorphic adenoma
Pleomorphic adenomas, though histologically benign, can be difficult to control
locally with surgery alone. Radiotherapy is indicated if excision margins are
positive and no further surgery is possible (e.g. tumour close to the facial nerve).
Radiotherapy should also be considered to prevent further recurrences in patients
who have had a pleomorphic adenoma excised on more than one occasion
previously, particularly if there is a short interval between recurrences relative to
the life expectancy of the patient, or if further surgery would compromise cosmesis
or function.

Adjuvant radiotherapy should ideally commence 4–6 weeks after surgery as long as
adequate wound healing has occurred. There is no proven role for concomitant
chemotherapy.

Parotid tumours usually arise in the portion of the gland lateral to the plane of the
facial nerve – the superficial lobe – though there is no anatomical distinction
between the superficial and deep lobes. They can invade locally throughout the
gland, compromising facial nerve function if trunks of the nerve are invaded.
Extraparotid extension can occur laterally into skin or medially into the ptery-
gopalatine fossa and lateral parapharyngeal space, resulting in trismus or invasion
of the carotid sheath. Bone invasion is uncommon. Adenoid cystic carcinomas in
particular can invade nerve fibres spreading up the facial nerve towards the
stylomastoid foramen.
The parotid gland contains several intraparotid lymph nodes. The superficial
intraparotid nodes are on the external surface of the gland, and the deep nodes are
found within the gland, mainly adjacent to the external carotid artery and external
jugular vein. Parotid tumours can spread via the intraparotid nodes to the
subparotid nodes in the retrostyloid space and thence to the retropharyngeal
nodes, or directly to level II nodes (Fig. 15.1).
Tumours of the submandibular salivary gland can invade locally or perineurally
in the marginal branch of the facial nerve, the lingual nerve, nerve to mylohyoid
and hypoglossal nerve. Lymphatic drainage is to level Ib nodes lying adjacent to
(but rarely within) the salivary gland and then to ipsilateral level II nodes.
There are minor salivary glands submucosally throughout the upper
aerodigestive tract and malignant salivary tumours can occur in any site. The hard
palate is the most common location for such tumours which spread to the same
lymph nodes as squamous cell carcinomas at those sites.
178
Data acquisition
Figure 15.1 Axial T1-weighted MRI
showing intraparotid lymph node (N), and
the location of the subparotid nodes in the
retrostyloid space (R).

Clinical examination can reveal invasion of local structures such as the skin, facial
nerve (palsy) or pterygoid muscles (trismus) or spread to draining lymph nodes.
Fine needle aspiration either in clinic, or under ultrasound guidance, usually
provides confirmation of malignancy.
Cross-sectional imaging is performed to assess extent of the primary tumour
particularly at the deep margin adjacent to the parapharyngeal space and to assess
local lymph nodes. MRI is preferred to CT as primary tumours are better defined
and nerve enhancement can be assessed. Scans should include imaging of the skull
base. Cross-sectional preoperative imaging should be obtained on all patients with
malignant tumours to enable more accurate postoperative volumes to be defined
for radiotherapy.
Data acquisition
■ Immobilisation
Patients should be immobilised lying supine with the neck slightly extended to
move the orbits superiorly and reduce the chance of beams exiting through the
eye. A Perspex or thermoplastic shell with at least five fixation points should ideally
be used even if the neck is not included in the treatment volume, as systematic and
random errors will be smaller and CTV-PTV margins can be tighter.
■ CT scanning
CT slices are obtained from the skull base to the hyoid in patients not requiring
neck radiotherapy, or from the skull base to the arch of the aorta if the neck is to
be irradiated. Slices should be no more than 5 mm thick and ideally 3 mm.
179
SALIVARY GLANDS
Fusion of planning MRI and CT images can be particularly helpful where there
is extensive perineural invasion which necessitates inclusion of the skull base in the
target volume.
■ Simulator
CT scanning is preferred to simulator-based planning to allow individualised
ipsilateral beam arrangements to conform better to the PTV and avoid critical
structures and mucosa.

■ Parotid
The planning CT (and MRI if performed) should be carefully evaluated to detect
macroscopic residual disease or lymphadenopathy. Preoperative imaging and
discussions with the surgeon and pathologist are important. As radiotherapy is
usually given adjuvantly after surgery, no GTV is defined unless there is
macroscopic residual disease. The CTV60 is contoured as the sites of possible
microscopic disease. Particular attention is given to the deep excision margin
which is likely to be close or involved if the facial nerve has been preserved. As a
minimum, the medial extent of the CTV60 should be to the lateral surface of the
internal jugular vein, but if the deep lobe of the parotid is thought to contain
tumour, the parapharyngeal space should be included (Figs 15.2 and 15.3). The
lateral extent of the CTV60 will be close to the surface of the skin. The position
of the contralateral parotid on the planning CT can be a useful guide to the
superior and inferior limits of the CTV60.
In adenoid cystic carcinomas, the CTV60 should include the course of the facial
nerve up to the stylomastoid foramen at the skull base. In recurrent pleomorphic
adenomas, the CTV should be individualised depending on the location of
recurrent disease as defined by preoperative imaging and surgical discussions.
Figure 15.2 CTV for a small high grade

tumour completely excised by a superficial
parotidectomy with a close deep margin
adjacent to the facial nerve. Arrow
indicates internal jugular vein.
180
(a) (b)
(c)
Figure 15.3 Target volume definition for an adenoid cystic tumour of the deep lobe of the
parotid gland. (a) Axial T1-weighted contrast-enhanced MRI showing primary tumour (T).
(b) Corresponding planning CT scan showing CTV including the parapharyngeal space
(arrowed). (c) Axial planning CT slice close to skull base. The course of the facial nerve up
to the stylomastoid foramen (arrowed) is included in the CTV.
If there are indications for neck radiotherapy adjuvantly after a neck dissection,
the levels to be treated are included in the CTV60. If there are indications (high
grade tumour) for prophylactic neck radiotherapy, the ipsilateral level Ib, II and
III nodes should be included in the treated volume. While a separate CTV44 can
be defined to give these sites a prophylactic dose, the proximity of the nodes to
the parotid bed means that including them in the CTV60 and treating the whole
volume in one phase is a more pragmatic approach.
181
SALIVARY GLANDS
Sites where resection margins are involved, or where there was extracapsular nodal
extension, should be defined in a CTV66 – though again it may be more appropriate
to treat the whole CTV to this dose rather than to define separate dose levels.
The CTV is expanded isotropically to form the PTV by a margin determined for
each department by the observed random and systematic errors – usually 3–5 mm.
The contralateral parotid gland does not usually receive sufficient dose to cause
xerostomia but it should be contoured as an organ at risk if the mean dose to the
gland is expected to be 24 Gy. The inner ear should be defined as an OAR, as
reducing dose to the cochlear apparatus may reduce the risk of deafness.
■ Other sites
Similar principles can be applied for volume definition for tumours of the
submandibular or minor salivary glands. In adenoid cystic carcinomas the nerve
innervating the primary tumour site should be included up to the skull base. In
adenoid cystic carcinomas of the submandibular gland this should include the
lingual nerve (a branch of the mandibular nerve, V3) back to the foramen ovale
and the marginal mandibular branch of the facial nerve to the stylomastoid
foramen. For tumours arising in or close to midline (e.g. hard palate), prophylactic
lymph node volumes should be outlined bilaterally if lymph nodes are to be
included in the CTV.
Dose solutions
■ Conformal
Two or three ipsilateral photon beams will usually provide homogeneous dose
distribution to the CTV without exceeding the tolerance of adjacent critical
structures. The anterior oblique beam angle is chosen according to the shape of
the anteromedial edge of the PTV while trying to minimise dose to the mucosa
of the oral cavity and oropharynx. The posterior oblique angle is chosen according
to the contour of the posterolateral edge of the PTV and should be lateral to the
spinal cord and brainstem (Fig. 15.4). The exit dose from this beam should be
inferior to the contralateral eye. This is usually achieved by immobilising the patient
with the neck slightly extended but half beam blocking may be needed if the PTV
extends more superiorly. An additional lateral photon beam may provide a more
homogeneous distribution but will increase dose to the contralateral parotid gland
and possibly to the spinal cord. Alternatively, a lateral electron beam can be used
but current algorithms make the calculation of mixed photon and electron beams
less reliable. All beams should be shaped to the PTV contour with MLCs.
The PTV may come close to the skin surface, in which case it can be difficult to
cover the lateral surface of the PTV unless tissue equivalent bolus is used.
However, bolus is only recommended if there is a risk of microscopic residual
disease in the skin. This is an uncommon situation as involved skin is usually
resected and a myocutaneous flap used to fill the defect. This new skin will not
contain microscopic disease. However, if a Perspex shell is used it should not be
cut out over the treated volume.
Hotspots in the mandible of 107 per cent should be avoided to reduce the risk
of osteoradionecrosis. Excessive dose in the temporomandibular joint (TMJ)
182
Dose-fractionation
Figure 15.4 Axial planning
CT slice showing three-
beam plan for adjuvant
radiotherapy of a parotid
tumour.
should also be avoided to reduce the risk of long-term TMJ dysfunction and
trismus. The cochlear dose should be kept below 50 Gy if possible to reduce the
risk of long-term sensorineural hearing damage.
If level III and IV nodes are to be treated as an adjuvant to neck dissection,
a matched anterior neck beam can be used. The match plane should be inferior to
any preoperative lymphadenopathy to avoid a junction through microscopic
residual disease.
■ Complex
IMRT planning studies have reported reduced dose to the cochlea. An equispaced
nine-beam coplanar technique has been described, but this risks increasing dose to
the contralateral parotid. An ipsilateral four-beam IMRT planning solution has
also been used but may not be better than a 3D-conformal beam arrangement.
Neutron therapy
The one small RCT of fast neutron therapy versus conventional radiotherapy for
unresectable salivary gland cancers reports a better local control rate with neutrons
at the risk of increased late toxicity.
■ Conventional
The irregularly shaped PTV will be treated best with conformal radiotherapy to
reduce dose to organs at risk, but an ipsilateral anterior and posterior oblique
wedged beam arrangement can be planned conventionally. Care should be taken
to avoid organs at risk, especially exit dose to the contralateral eye.
Dose-fractionation
■ Adjuvant
66 Gy in 33 fractions for positive margins or extracapsular nodal spread.
183
SALIVARY GLANDS
■ Recurrent pleomorphic adenoma

There are no data to support the use of altered fractionation regimens for
salivary gland tumours.

The amount of oral cavity and oropharynx included in the treatment volume may
predict the degree of swallowing problems seen during treatment. Treatment of
mucositis should be given within a multidisciplinary team, which reviews the patient
weekly. Advice on jaw exercises can reduce the risk of trismus and TMJ dysfunction.
Conductive hearing loss due to middle ear effusions can occur during radiotherapy
and take several months to improve after treatment has finished. If subjective hearing
loss persists 2 months after treatment, an audiogram should be performed. If there
is evidence of conductive hearing loss, a grommet may be indicated.
Verification
In addition to the procedures outlined in Chapter 8, if the posterior oblique beam
is shown on BEV to exit close to the contralateral eye, lens doses should be
measured by TLD on the first day of treatment to ensure tolerance is not exceeded.
key trials
COchlear Sparing Therapy And conventional Radiation COSTAR – A multicentre

randomised study of cochlear sparing intensity modulated radiotherapy versus
conventional radiotherapy in patients with parotid tumours. Available at
www.ncrn.org.uk (accessed 4 December 2008).
Laramore GE, Krall JM, Griffin TW et al. (1993) Neutron versus photon irradiation
for unresectable salivary gland tumors: final report of an RTOG-MRC
randomized trial. Int J Radiat Oncol Biol Phys 27: 235–240.
Information sources
Chen AM, Garcia J, Lee NY et al. (2007) Patterns of nodal relapse after surgery and postoperative
radiation therapy for carcinomas of the major and minor salivary glands: what is the role of
elective neck irradiation? Int J Radiat Oncol Biol Phys 67: 988–94.
Terhaard CHJ, Lubsen H, Rasch CRN et al. (2005) The role of radiotherapy in the treatment of
malignant salivary gland tumours. Int J Radiat Oncol Biol Phys 61: 103–11.
184
Sinuses: maxilla, 16
ethmoid and nasal
cavity tumours
Tumours arising in the sinonasal area usually present with symptoms of local invasion
and the importance of local extent is reflected in the T stage (see Table 16.1).
Lymphatic spread and distant metastases are unusual so surgery and
radiotherapy to the primary site are the main treatments. Fifty per cent of tumours
appear to arise in the maxilla with 25 per cent each in the nasal cavity and
ethmoids. Primary tumours of the frontal or sphenoid sinus are very rare.
Table 16.1 UICC TNM (6th edn, 2002*) tumour staging for nasal cavity and paranasal
sinuses
Maxillary sinus Nasal cavity or ethmoid sinus
T1 Tumour limited to the mucosa with Tumour restricted to one subsite of nasal cavity
no erosion or destruction of bone or ethmoid sinus,† with or without bony invasion
T2 Tumour causing bone erosion or Tumour involves two subsites in a single site or
destruction, including extension extends to involve an adjacent site within the
into hard palate and/or middle nasoethmoidal complex, with or without bony
nasal meatus, except extension invasion
to posterior wall of maxillary
sinus and pterygoid plates
T3 Tumour invades any of the following: Tumour extends to invade the medial wall
bone of posterior wall of maxillary or floor of the orbit, maxillary sinus, palate, or
sinus, subcutaneous tissues, cribriform plate
floor or medial wall of orbit,
pterygoid fossa, ethmoid sinuses
T4a Tumour invades any of the following: Tumour invades any of the following: anterior
anterior orbital contents, skin of orbital contents, skin of nose or cheek, minimal
cheek, pterygoid plates, masticator extension to anterior cranial fossa, pterygoid
space, cribriform plate, sphenoid plates, sphenoid or frontal sinuses
or frontal sinuses
T4b Tumour invades any of the following: Tumour invades any of the following: orbital
orbital apex, dura, brain, middle apex, dura, brain, middle cranial fossa, cranial
cranial fossa, cranial nerves other nerves other than V2, nasopharynx, clivus
than maxillary division of trigeminal
nerve V2, nasopharynx, clivus
*With permission.
†
Subsites are: septum, floor, lateral wall and vestibule of nasal cavity and left and right ethmoid sinuses.
185
SINUSES
Squamous cell cancers are the commonest histological subtype (50 per cent) and
radiotherapy is often combined with chemotherapy in advanced disease on the
basis that there is additional benefit from combined treatment in other head and
neck squamous cell cancers.
There are many other tumour types that all have slightly different clinico-
pathological characteristics. Adenoid cystic cancers have a propensity for perineural
spread so radiotherapy volumes need to include the course of the relevant nerve to
the skull base. Olfactory neuroblastomas (aesthesioneuroblastoma) arise in the
olfactory epithelium of the superior nasal cavity and can invade the cribriform plate
and anterior cranial fossa. This is important when planning both surgery (craniofacial
resection) and radiotherapy.
Malignant melanomas can arise from the nasal cavity mucosa, especially the lateral
wall. Depth of invasion does not correlate well with prognosis. Sinonasal melanomas
behave unpredictably but almost inevitably recur at some point, often locally. This
means a more palliative approach should be considered in the elderly or those with
poor performance status. Adenocarcinomas usually start in the middle meatus or
ethmoid sinus, and are related to exposure to hardwood dust. Other rarer tumours
include chondrosarcomas and sinonasal undifferentiated carcinomas (SNUC).
Whilst surgery is the treatment of choice for almost all sinonasal malignancies,
adjuvant radiotherapy is recommended in most cases as it is difficult to resect these
tumours en bloc with clear margins. There is non-randomised trial evidence that
radiotherapy improves local recurrence rates for all tumour types. The exception is
completely resected T1 disease where recurrence rates are likely to be low. It may
also be appropriate not to irradiate after surgery for sinonasal melanoma, reserving
radiotherapy for the almost inevitable recurrence.
Combined surgery and postoperative radiotherapy lead to optimal 5-year
survival rates of 50 per cent in maxillary sinus squamous cell cancers, 60 per cent in
ethmoid adenocarcinoma, 75 per cent in olfactory neuroblastoma and 30 per cent
in sinonasal melanoma.
■ Primary radiotherapy
If complete resection is considered impossible because of invasion of local
structures (e.g. cranial fossae, masticator space) or if the patient declines surgery,
primary radiotherapy is used to obtain local control and, occasionally, cure.
Mucosal melanomas and other locally advanced sinonasal tumours in patients with
poor performance status are treated with palliative radiotherapy as cure is unlikely.
This approach should be discussed with individual patients.

Adjuvant radiotherapy should begin after adequate surgical healing – usually
within six weeks of operation. Concomitant chemotherapy with cisplatin may be
given in stage III and IV squamous cell carcinomas, particularly if disease is
unresectable or if excision margins are positive.
186
If primary radiotherapy is given, disease should be reassessed 4–6 weeks later by
a surgeon to see if resection is then possible.

The nasal cavity, ethmoid sinuses and maxillary sinuses are interconnected mucosa-
lined spaces in close proximity to the orbit and anterior cranial fossa (Fig. 16.1).
Most tumours present with symptoms from spread outside the sinuses. An
understanding of the 3D anatomy is important to assess disease and to determine
target volumes for radiotherapy.
Ethmoid sinuses
Cribriform plate
Lamina papyracea
Floor of orbit
Nasal cavity
Maxillary antrum
Hard palate
Nasal septum
Alveolar process
of maxilla
Figure 16.1 Anterior view of the paranasal sinuses.
Maxillary tumours can extend through the anterior wall to invade the cheek or
posteriorly into the pterygopalatine fossa and masticator space (infratemporal
fossa) causing trismus, and from there to the middle cranial fossa. Inferior
extension into and through the floor of the maxilla may result in loose teeth or an
oroantral fistula.
It is relatively easy for tumour to grow into the orbit superiorly through the
inferior orbital fissure or for ethmoid tumour to grow into the orbit through the thin
lamina papyracea. Ethmoid tumours can also grow superiorly through the cribriform
plate and into the anterior cranial fossa and anteriorly into the nasal cavity.
MRI with gadolinium enhancement is the imaging modality of choice as it can
assess local extent and differentiate tumour from retained secretions. CT may
provide additional information if cribriform plate erosion or early orbital involve-
ment is suspected.
Lymph node involvement is seen in less than 20 per cent of tumours but the
neck should be examined clinically and radiologically. Tumours invading the
anterior nose and cheek have a higher risk of lymphatic spread than those
contained within the sinuses, with level Ib and II nodes most likely to be involved.
Lymphatic spread is more common in tumours invading adjacent mucosal surfaces
such as the nasopharynx.
Metastases at presentation are uncommon but all patients should have a chest
radiograph and appropriate investigation of symptoms suggestive of metastases.
187
SINUSES

Owing to the late presentation, it can be difficult to determine the exact primary site
so possible spread to all the above sites should be assessed clinically and radiologically.
Clinical assessment includes nasendoscopy to assess the nasal cavity and examination
of the oral cavity to check for inferior extension. Pterygopalatine fossa extension may
lead to trismus, infraorbital canal involvement to facial pain and paraesthesia and
orbital cavity spread to proptosis and diplopia, all of which should be sought.
Several surgical approaches are possible and need to be understood to define
target volumes in the adjuvant setting.
A lateral rhinotomy allows access to the medial maxilla, the nasal cavity and the
ethmoid, sphenoid and frontal sinuses. It has been superseded for many tumours
by a midfacial degloving approach. In this operation, a sublabial incision allows the
soft tissues of the face to be elevated to provide greater bilateral access though
access to the frontal sinus is more limited.
A craniofacial approach enables assessment and resection of the anterior skull
base at the cribriform plate with en-bloc resection of tumour involving the
ethmoids. If there is tumour invading the medial wall of the orbit but not the
periosteum (i.e. no tumour within the orbital cavity), part of the orbital
periosteum can be resected. This enables the eye to be spared but presents a
challenge in planning radiotherapy. Frozen sections are used to define margins of
excision. Surgery for selected patients is performed endoscopically. Tumours are
removed piecemeal but with frozen sections to assess margins. The technique is
not suitable when the orbit and skull base are thought to be at risk.
Data acquisition
■ Immobilisation
Patients should be immobilised supine in a Perspex or thermoplastic shell. If the
neck is not irradiated, the shoulders do not need to be immobilised. If the low
neck nodes are to be treated (level III–V) the neck should be extended to allow
treatment of most of the neck nodes with an anterior beam, avoiding the oral
cavity and pharynx where possible.
A mouth bite is used to depress the tongue and oral cavity away from the treated
volume and reduce acute morbidity. Patients should be asked to look straight
ahead to avoid rotating the lens or retina, particularly if the orbital cavity is
included in the treated volume. Wax plugs in the nostrils are used if the tumour
extended inferiorly in the nasal cavity to enable a more uniform dose distribution.
■ CT scanning
A CT scan is performed with 3 mm slices from 2 cm superior to the superior orbital
ridge to the hyoid bone (but extended to include the low neck if neck nodes are
to be treated). The whole head should be imaged if non-coplanar beams are to be
used in the treatment plan (see below).
Fused CT-MRI images can be useful in the definition of the optic pathways and
skull base. MRI also allows retained secretions to be differentiated from tumour
where resection has been incomplete.
188
■ Simulator
3D CT-planned conformal radiotherapy is recommended for all patients, given the
complex anatomy of this region. If this is not available, orthogonal simulator films
are taken and volumes defined on these films using knowledge of tumour extent
and patterns of spread.

Where resection is not possible or has been incomplete, the GTV is outlined but
defining the CTV is the most important step for most patients.
The proximity of these tumours to critical structures such as the optic nerves and
chiasm, brainstem, and lacrimal glands mandates meticulous CTV definition so
that radiotherapy can be targeted to sites at highest risk of relapse in individual
tumours and reduce the risk of long-term radiation damage. The importance of
discussion between the surgeon, pathologist and radiation oncologist in defining
sites at greatest risk of recurrence after surgery cannot be overemphasised.
Preoperative imaging should be viewed beside the CT planning dataset to ensure
that initial sites of disease are covered.
The CTV should encompass all initial sites of disease (presurgery GTV), the
mucosa of adjacent compartments of the sinonasal complex and a 10 mm margin
at least from initial sites of GTV where no good bony barrier to invasion exists
(e.g. masticator space, cribriform plate and infraorbital fissure) (Fig. 16.2).
(a) (b)
Figure 16.2 Definition of CTV for a pT4a carcinoma of the maxilla resected with clear
margins. (a) Preoperative T1-weighted contrast-enhanced MRI showing primary tumour
invading the cheek (C), masticator space (M) and lateral pterygoid muscle (P).
(b) Corresponding planning CT slice showing CTV.
189
SINUSES
For most tumours, the CTV will include the ipsilateral maxillary sinus and nasal
cavity and the ethmoid sinuses bilaterally. The superior extent can be modified for
a very inferior tumour. Where the primary involved the maxilla, consideration
should be given to including the pterygopalatine and masticator space. When a
maxillary tumour has invaded inferiorly, the hard palate should be included in the
CTV so as to allow a 10 mm margin around original disease.
The CTV for tumours involving the ethmoid sinuses should include the
sphenoid sinus. Where initial disease came close to the orbit or invaded the lamina
papyracea, the CTV should include that portion of the medial and inferior orbital
wall (Fig. 16.3). The orbital cavity should be included in the CTV if the orbital
wall has been breached by tumour or if tumour has grown superiorly through the
inferior orbital fissure. Where a craniofacial excision has been carried out, the CTV
should extend 10 mm superior to the cribriform plate or 10 mm superior to initial
sites of disease, whichever is greater. Olfactory neuroblastomas arise from the
cribriform plate and particular attention should be paid to the CTV at this site.
(a) (b)
Figure 16.3 CTV for an olfactory neuroblastoma invading the lamina papyracea but not
into the orbit. (a) Preoperative T1-weighted contrast-enhanced MRI showing primary
tumour (T) close to the left orbit. (b) Corresponding planning CT slice showing CTV
including the medial portion of the orbital wall.
If the primary was close to or invading the nasopharynx, the adjacent ipsilateral
retropharyngeal nodes should be included in the CTV. When cervical node
radiotherapy is indicated, the intraparotid nodes, level Ib and superior level II
nodes can be included in the CTV but including these nodes will make a complex
volume harder to treat adequately. As local relapse in the primary site is usually the
greatest risk, the nodes are often not treated in sinonasal tumours.
The CTV is expanded isotropically (usually by 3–5 mm but determined by local
audit) to form the PTV.
Organs at risk to be outlined include the lenses, lacrimal glands (in the
superolateral orbit and upper eyelid), optic nerves and chiasm, spinal cord, brainstem
190
Dose solutions
and pituitary gland. Some centres also expand these structures (in particular the optic
nerves) by 2–3 mm to create a PRV to account for systematic and random errors.
The direction of such errors is likely to be in the same direction as changes in the
PTV so adding a 3 mm margin to both the CTV and the OAR may be unnecessary.
Dose solutions
■ Conformal
The commonest beam arrangement for sinonasal tumours uses an anterior beam
to provide most of the dose with an ipsilateral or bilateral wedged lateral beams
added to provide extra dose to the posterior part of the PTV (Fig. 16.4). This
arrangement cannot provide a uniform dose distribution. Hot-spots of 110 per
cent are usually found anteriorly and inadequate posterior coverage occurs in spite
of the lateral beams. MLCs are used to shape each beam to the PTV. The lateral
fields have their anterior border behind the lens and can be angled 5° posteriorly
to avoid exiting through the contralateral lens. As a result, not all the PTV will be
within the lateral beams.
(a) (b)
Figure 16.4 Conformal radiotherapy for sinonasal tumours. (a) Beam arrangement.
Colour wash scale is set from 95 per cent to demonstrate the hot spot anteriorly and the
underdosing posteriorly. (b) BEV of left lateral beam. This beam is to add dose to the
posterior part of the PTV only. Left eyeball green; lens yellow.
The course of the optic nerves becomes more medial at the posterior part of the
orbital cavity as they exit through the optic canal. At this point, the nerves
commonly overlap the PTV. The authors recommend a dose limit of 50 Gy (in
2 Gy fractions) for the optic nerve and chiasm, but where there is particularly high
risk of local recurrence (which could itself cause blindness) 55 Gy can be accepted
to one optic nerve. In practice, this necessitates a two-phase technique. The whole
PTV is treated to 50 Gy. Then the MLCs are moved to ensure the optic nerve
191
SINUSES
doses remain acceptable, although inevitably the posterolateral part of the PTV
will not be in the treated volume.
Where the orbital contents are included in the CTV (or after an orbital
exenteration) equally weighted anterior and lateral wedged beams are used.
■ Complex
IMRT provides a more conformal dose distribution to the unusual PTVs in sinonasal
cancer. Five- or seven-field coplanar beams have been used but these arrangements
will increase dose to the orbital contents. A non-coplanar arrangement of three to
five sagittal midline beams with right and left lateral beams avoids entry or exit of
beams through the eyes and provides a uniform dose distribution (Figs 16.5–16.7).
Where the PTV overlaps the optic nerve, there must still be either an acceptance of
increased risk of blindness or a reduction in PTV coverage.
Figure 16.5 Beam arrangement for IMRT of sinonasal tumours.
■ Conventional
The shape of the PTV means that simple conventional planning will not be able to
produce adequate dose to the PTV and spare critical normal structures.
Dose-fractionation
■ Adjuvant
66 Gy in 33 daily fractions if possible where there is residual disease.
■ Palliative
36 Gy in 6 fractions of 6 Gy treating once weekly.
192
(a)
(b)
Figure 16.6 IMRT dose solution for a sinonasal tumour. (a) Note the more homogeneous
coverage of the PTV than the conformal solution (Fig. 16.4) provides. Colourwash scale is
set from 95 per cent. (b) Dose colourwash to illustrate relative sparing of the left lacrimal
gland (pink) and lens (blue).

Patients are seen weekly during treatment in a multidisciplinary clinic. Exercises
to reduce trismus (e.g. Therabite™) are recommended if there is pre-existing
trismus or where the masticator muscles are in the treated volume. Prophylactic
feeding tubes should be considered when the patient has trismus and the treated
volume includes a significant portion of the oral cavity (e.g. in maxillary sinus
tumours). Careful oral hygiene is essential with prompt treatment of oral Candida
infections.
193
SINUSES
180.0
170.0
160.0
150.0
140.0
130.0
120.0
3D Conformal
110.0
Volume (cm )
3
100.0
IMRT
90.0
80.0
70.0
60.0
50.0
40.0
30.0
20.0
10.0
0.0
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
20.0
22.0
24.0
26.0
28.0
30.0
32.0
34.0
36.0
38.0
40.0
42.0
44.0
46.0
48.0
50.0
52.0
54.0
54.0
58.0
60.0
62.0
64.0
66.0
68.0
70.0
(a) Dose (Gy)
0.6
0.5
0.4 3D Conformal
Volume (cm )
3
IMRT
0.3
0.2
0.1
0.0
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
20.0
22.0
24.0
26.0
28.0
30.0
32.0
34.0
36.0
38.0
40.0
42.0
44.0
46.0
48.0
50.0
52.0
54.0
56.0
58.0
60.0
(b) Dose (Gy)
Figure 16.7 DVHs comparing plans shown in Figure 16.4 (three-dimensional conformal)
and 16.6 (IMRT). Prescribed dose 60 Gy in 30 fractions. (a) PTV. (b) Ipsilateral (left) optic
nerve. Note that in the conformal plan, dose to the nerve exceeds tolerance (55 Gy) and
the prescribed dose would have to be reduced.
Where the cornea is within the treated volume, ophthalmic review should be
carried out both during and after radiotherapy. Lubricating eye ointments can be
applied during the day and at night. If there is a pre-existing facial nerve palsy, the
eyelid should be taped shut at night to avoid a dry eye.
Pituitary function tests should be carried out annually during follow-up to
evaluate late radiotherapy effects to the pituitary gland.
Verification
194
Information sources
key trials
■ None in this rare cancer.
Information sources
Claus F, De Gersem W, De Wagter C et al. (2001) An implementation strategy for IMRT of ethmoid
sinus cancer with bilateral sparing of the optic pathways. Int J Radiat Oncol Biol Phys 51:
318–31.
Dirix P, Nuyts S, Vanstraelen B et al. (2007) Malignancies of the nasal cavity and paranasal sinuses:
long-term outcome with conventional or three-dimensional conformal radiotherapy. Int J Radiat
Lund VJ (2003) Tumours of the upper jaw and anterior skull base. In: Rhys Evans PH, Montgomery
PQ, Gullane PJ (eds) Principles and Practice of Head and Neck Oncology. Martin Dunitz:
London, 337–54.
Mendenhall WM, Amdur RJ, Hinerman RW et al. (2005) Head and neck mucosal melanoma. Am J
Clin Oncol 28: 626–30.
Mendenhall WM, Mendenhall CM, Riggs CE Jr et al. (2006) Sinonasal undifferentiated carcinoma.
Am J Clin Oncol 29: 27–31.
195
17 Orbit

■ Intraocular tumours
Metastases to the vascular choroidal layer (or uvea) – particularly from breast, lung
or gastrointestinal tumours – are the commonest intraocular tumours. They usually
present with reduced visual acuity and are treated with palliative radiotherapy.
The most frequent primary intraocular tumour is choroidal melanoma. Tumours
arising in the choroid have a 30 per cent chance of metastasis at 5 years while rarer
iris or conjunctival melanomas have a much lower risk. Tumour size is the most
important prognostic factor. Small melanomas are usually asymptomatic and can
be difficult to differentiate from various benign conditions. Small tumours
(
5 mm diameter and
3 mm thick) can therefore be observed or treated with
iodine-125 or ruthenium-106 plaque brachytherapy, protons or other local
treatments such as laser photocoagulation or surgical excision. Large tumours
(16 mm diameter or 10 mm thick) are treated with enucleation. Preoperative
radiation does not improve cure rates. Where there is extrascleral extension,
prognosis is poor but enucleation and postoperative radiotherapy can be
considered to optimise local control.
A large RCT has shown equivalent cure rates for enucleation and iodine-125
plaque brachytherapy for tumours up to medium size. The plaque is temporarily
sutured to the sclera underlying the melanoma and left in place for 3–7 days.
Doses of 400 Gy to the base of the tumour and 80–100 Gy to the apex can be
achieved. Fractionated proton therapy, available only in specialised centres, gives
similar local control rates.
Radiotherapy can be used for cure of retinoblastoma but because there is a high
incidence of second malignancies, chemotherapy is often used to allow radiation
dose to be reduced. IMRT and proton therapy can produce more conformal dose
distributions than standard EBRT.
Conjunctival lymphomas present as salmon-coloured patches on the sclera.
They are confined to the orbit in 70 per cent of patients and are treated with
radiotherapy. If there is systemic disease chemotherapy is also used. Mucosal
associated lymphoid tissue (MALT) lymphomas of the conjunctiva can also be
treated with topical chemotherapy or intralesional interferon or can be observed.
■ Extraocular tumours
Orbital lymphomas respond well to radiotherapy either given as sole treatment in
stage I disease or following chemotherapy. Primary lacrimal gland carcinomas and
orbital sarcomas are treated with excision and postoperative radiotherapy.
196
Radiotherapy is useful for basal and squamous cell carcinomas arising from the
skin and lower eyelid. Superficial photons or electrons are used for early disease
whereas EBRT can be used as an alternative to exenteration in more advanced
disease or can be considered following exenteration if excision margins are close or
involved.
■ Benign disease
Graves’ disease is an autoimmune lymphocyte activation causing infiltration and
oedema of the extraocular muscles. Radiation is an effective way to reduce local
inflammation and compressive symptoms, especially when combined with systemic
steroids. It can similarly be used for orbital pseudotumour – an idiopathic orbital
inflammation.

The eye is composed of three layers. An outer fibrous layer is formed by the sclera
posteriorly and the cornea anteriorly. The inner layer is the sensory retina with
vision concentrated at the fovea which is lateral to the optic nerve and directly
posterior to the lens. In between these is the vascular layer – the uvea or choroid –
which supplies the retina. The iris is the outer continuation of the vascular layer
and the lens sits just behind it, suspended from the ciliary body. The eye has no
lymphatic drainage (Fig. 17.1).
Cornea
Iris Pupil
Suspensory
ligament Ciliary body
Lens
Optic disc
Retina
Uvea or
choroid
Sclera
Optic nerve Fovea
Artery
Figure 17.1 Anatomy of the eye.
The extraocular structures often need to be considered and contoured as critical

structures in radiotherapy. The optic nerve can clearly be seen on cross-sectional
imaging and leaves the orbit via the optic canal to form the optic chiasm. The
lacrimal gland sits mainly in the lacrimal fossa – the under-surface of the orbital
plate of the frontal bone – but also has a palpebral portion within the upper eyelid.
It can be difficult to see on imaging (see Fig. 16.6b, p. 193). Tears from the eye
197
ORBIT
drain through the nasolacrimal duct within the nasolacrimal canal which runs from
the lacrimal sac to the inferior nasal meatus between the inferior concha and nasal
cavity floor. Tumour invasion, surgery or radiation to the duct can cause epiphora.
The conjunctiva is a clear membrane of lymphoid tissue that covers the sclera
and the inner surface of the eyelids producing protective mucus and tears.
The orbit is conical and formed by the fusion of several bones, the thinnest of
which is the lamina papyracea of the ethmoid bone which forms part of the medial
wall. Tumour can grow into the middle cranial fossa through the superior orbital
fissure and optic canal or into the pterygopalatine fossa and masticator space via
the inferior orbital fissure.
Only the eyelids, conjunctiva and lacrimal glands have lymphatic drainage and
they drain to the preauricular node.

The oncologist and ophthalmologist should jointly assess any intraocular tumour
after the pupil has been dilated. Extraocular tumour is best assessed by cross-
sectional imaging with CT or MR scans with coronal and sagittal reconstructions.
Clinical examination can also help define extraocular spread into the superior
orbital fissure (which causes cranial nerve palsies of VI and then of III, IV and V1)
or into the pterygopalatine fossa and masticator space via the inferior orbital fissure
(causing trismus). Careful examination of eye movements can help distinguish
invasion of the external ocular muscles from cranial neuropathies. Invasion of the
ethmoid sinuses or the nasopharynx can be assessed at nasendoscopy.
Data acquisition
■ Immobilisation
The proximity of target volumes to several critical normal structures means
excellent immobilisation is vital. A custom-made thermoplastic or Perspex shell is
created with the patient supine and the chin in a neutral position.
■ Simulator
The dimensions of the eye vary by only 1–2 mm between subjects, with a vertical
diameter of 23 mm and a slightly larger anteroposterior diameter. Palliative EBRT
beams for choroidal metastases can therefore be defined in the simulator. A
4 4 cm beam is marked onto the shell with the anterior margin at the outer
canthus and the centre of the field in line with the pupil. The beam is angled
posteriorly by 5° to avoid exit through the contralateral lens. Dose is either
prescribed to Dmax or to the depth of the metastasis if known from imaging (Fig.
17.2). An alternative is to use a direct anterior 4 4 cm beam centred on the pupil
and to treat with the eye open to minimise lens dose. For treatment of bilateral
choroidal metastases, opposing beams are used, again angled 5° posteriorly.
The direct anterior beam used to treat conjunctival lymphomas can also be
defined in the simulator. If the lymphoma is confined to the conjunctiva, electrons
or orthovoltage photons can be used. If diagnostic imaging reveals tumour
tracking posterior to the globe, MV photons are preferred (Fig. 17.3).
198
Data acquisition
(a) (b)
Figure 17.2 Palliative radiotherapy for choroidal metastasis. (a) Photograph of a choroidal
metastasis (arrowed); optic nerve (ON). (b) Dose colourwash illustrating the dose
distribution created by a 4 4 cm lateral beam angled 5° posteriorly.
(a) (b)
Figure 17.3 A 6 MV photon beam to treat conjunctival lymphoma. (a) Anterior view.
(b) Sagittal view. Note tissue equivalent bolus used to increase the surface dose.
In Graves’ disease small opposed lateral beams can be defined in the simulator
to treat the retro-orbital tissues. No target volumes are defined in benign disease.
Beam borders are specified by the 50 per cent isodose at the front of the pituitary
fossa and the 10 per cent isodose at the posterior lens edge. Beams are angled 5°
posteriorly to avoid the contralateral lens (Fig. 17.4).
■ CT scanning
As non-coplanar beams are sometimes required, the volume imaged should extend
from the vertex of the skull to the hyoid bone. Slices should be no more than
199
ORBIT
Figure 17.4 Beam arrangement

and dose distribution for treatment
of the retro-orbital tissues in
Graves’ disease.
3 mm thick. Intravenous contrast is not necessary. Fused CT and MR images can

be particularly helpful to define the optic chiasm and any skull base involvement.

For choroidal metastases, conjunctival lymphoma, melanoma or retinoblastoma,
disease can be adequately defined in the simulator as described above.
The variety of rare extraocular tumours makes it difficult to give general
recommendations for target volume definition following surgery or chemotherapy.
Preoperative or pre-chemotherapy imaging should be available at the time of
volume definition so that all original sites of disease are included in the CTV.
Discussion with the surgeon and pathologist may help to define locations at
greatest risk of recurrence. The CTV should take account of likely patterns of
spread: for example for tumours invading behind the eye in the orbital cavity, the
whole orbit should be included in the CTV. A 3 mm isotropic margin is added to
produce a PTV.
It is tempting to edit the target volumes away from critical structures with the
assumption that tolerance of these structures will be exceeded unless the PTV size
is reduced. The complexity of possible planning solutions – conformal or with
IMRT – means that the compensation needed is hard to predict. We strongly
recommend that tumour target volumes are not influenced by critical structures
and vice versa. Any compromise needed can thus be assessed using dose
distributions and DVHs to choose the plan providing the best solution. The optic
nerves, chiasm, lenses, retinas, lacrimal glands and pituitary should all be
contoured as OAR.
200
Dose solutions
Dose solutions
For treatment to the choroid or retina, planning scan data can be used to create
virtual simulation images to define the beam size and angle to minimise lens dose.
A 3D planning scan with 3D conformal planning is the technique of choice for all
curative extraocular tumour treatments because of the inevitable proximity of
critical optic structures to the PTV and the need to identify dose to organs at risk.
Postoperative radiotherapy for carcinomas and sarcomas ideally requires doses of
at least 60 Gy (66 Gy to sites of residual macroscopic disease) so the location and
tolerance of critical structures needs to be considered from the start of the
planning process.
Fractionated doses of more than 10 Gy have a high risk of inducing a cataract but
this late effect is treatable surgically. Lacrimal gland damage can occur with doses over
32 Gy and as a dry eye causes serious morbidity, great care should be taken to avoid
this. Corneal ulceration is uncommon at doses below 48 Gy and can be minimised by
patients being treated with the eye open to avoid any build-up effect of the overlying
eyelid and by good eye care. Retinopathy can occur with doses over 50 Gy and is more
common in diabetic patients. Particular care should be taken to avoid excess dose to
the fovea. Optic nerve and chiasm doses should be kept below 55 Gy.
A variety of beam arrangements can be considered depending on the PTV.
Anterior and lateral wedged beams will usually produce adequate PTV coverage
but the lacrimal gland, retina and optic nerve will usually lie within the treated
volume and the lens dose will exceed tolerance in spite of the lateral beam position
being chosen to be posterior to the lens. It is the preferred beam arrangement after
an enucleation (Fig. 17.5).
Figure 17.5 Adjuvant radiotherapy

to the orbit after an enucleation. The
lateral beam is angled to avoid the
contralateral lens.
Another solution is to use superior and inferior wedged beams in the sagittal
plane (Fig. 17.6). The borders can be chosen to spare the cornea anteriorly but
the lacrimal gland is again difficult to avoid. If the PTV is entirely behind the
201
ORBIT
(a)
(b)
(c)
Figure 17.6 Treatment of residual orbital lymphoma after chemotherapy. (a) Diagnostic
pre-chemotherapy CT scan showing tumour (T). (b) Corresponding slice on planning CT
scan after chemotherapy showing the CTV, PTV and dose distribution. (c) Sagittal view to
show beam angles and dose distribution.
202
Dose-fractionation
globe, as in some extraorbital lymphomas, either a single lateral beam is used
or superior and inferior wedged beams in the coronal plane are angled behind the
globe.
It is usual for the oncologist to have to accept some compromise in PTV dose
to keep all critical structures within tolerance or to accept an increased risk of
late effects from delivering an adequate dose. This decision will be influenced by
the risk and likely site of recurrence and should be discussed with the radiotherapy
team and the patient. It may be necessary to use two phases of treatment with the
whole PTV treated in phase 1 and some compromise in phase 2 to keep within
tolerance. Both phases should be planned at the same time.
IMRT may offer some advantage to conformal planning in these complex
volumes with a combination of coronal and lateral beams as for ethmoid tumours.
Particular care must be taken not to exceed tolerance of the contralateral orbital
structures when the number of beams is increased.
■ Benign disease
As both eyes are usually treated and a relatively low dose required, lateral beams
angled posteriorly to avoid the lenses can be defined using the CT dataset to help
to choose beam angles as for choroidal metastases.
Dose-fractionation
■ Choroidal metastases
■ Conjunctival lymphoma
30 Gy in 15 daily fractions given over 3 weeks.
■ Orbital lymphoma
40 Gy in 20 daily fractions (no chemotherapy).
30–36 Gy in 15–18 daily fractions after chemotherapy depending on the bulk of
any residual disease.
■ Postoperative radiotherapy (sarcoma, carcinoma,

melanoma)
Consider
66 Gy in 33 daily fractions given in 612⁄ weeks to sites of residual disease.
■ Thyroid eye disease

20 Gy in 12 daily fractions given in 212⁄ weeks (1 Gy for first fraction, 1.73 Gy
thereafter).
203
ORBIT

Patients should be advised on meticulous eye care and use of lubricating drops
throughout treatment. There should be prompt assessment and treatment of any
conjunctivitis or corneal damage. Close liaison with the ophthalmic team during
treatment is therefore essential. A full ophthalmic assessment including plotting of
visual fields should be carried out before radiotherapy so that any late effects can
be detected.
Patients should be instructed to keep their eyes open and look directly at the
beam to spare the front of the eye as much as possible.
Verification
TLD is used to confirm dose to the contralateral anterior eye.
Other points
Proton and plaque therapy are available in national specialised centres and referral
protocols should be readily available for consultation by any oncologist responsible
for an ophthalmological practice.
key trials
Collaborative Ocular Melanoma Study Group (2006) The COMS randomized trial
of iodine 125 brachytherapy for choroidal melanoma: V. Twelve-year mortality
rates and prognostic factors: COMS report No. 28. Arch Ophthalmol 124:
1684–93.
Hawkins BS, Collaborative Ocular Melanoma Study Group (2004) The
Collaborative Ocular Melanoma Study (COMS) randomized trial of pre-
enucleation radiation of large choroidal melanoma: IV. Ten-year mortality findings
and prognostic factors. COMS report number 24. Am J Ophthalmol 138:
936–51.
Information sources
Wei RL, Cheng JW, Cai JP (2008) The use of orbital radiotherapy for Graves’ ophthalmopathy:
quantitative review of the evidence. Ophthalmologica 222: 27–31.
204
Central nervous 18
system

This chapter will discuss the following tumours from the WHO classification
(Box 18.1).
Box 18.1 WHO classification of brain tumours*

■ Tumours of neuroepithelial tissue (of which astrocytomas, ependymomas and
oligodendrogliomas are commonest). These are graded I–IV
■ Tumours of cranial or paraspinal nerves
■ Tumours of the meninges
■ Lymphomas and haemopoietic malignancies
■ Tumours of the sellar region
■ Germ cell tumours
■ Metastatic tumours
*With permission.
Primary brain tumours are uncommon and comprise only 1.6 per cent of cancers.
Metastatic spread to the brain from primary cancers elsewhere in the body is much
more common. The variable behaviour of intracranial tumours depends on site
and histology. Some cause problems by their intracranial extension alone (gliomas,
meningiomas, pituitary tumours, metastases) while others (such as lymphomas,
germ cell tumours and primitive neuroendocrine tumours [PNET]) have a
predilection for leptomeningeal spread.
Gliomas
Gliomas constitute 40 per cent of all primary CNS tumours. They arise from glial cells
and are classified by cell type into astrocytomas, ependymoma and oligodendroglioma.
Tumours of mixed cell type can also occur. Grading (I–IV) is determined by
pathological examination with grade I–II classified as low grade astrocytomas, grade III
as anaplastic gliomas and grade IV as glioblastoma multiforme.
Low grade gliomas

Low grade gliomas (LGG) comprise a heterogeneous group of tumours which
often occur at a younger age and present with more insidious symptoms than high
grade tumours (HGG). Prognosis is variable and is adversely affected by age 40,
size 6 cm, corpus callosum involvement and histology of astrocytoma rather than
oligodendroglioma or mixed tumour.
205
CENTRAL NERVOUS SYSTEM
Survival may be prolonged for many years so that treatment related sequelae are
of more significance than for HGG. Low grade gliomas should be excised
completely wherever possible. Surgery alone gives 5-year progression free survival
rates of 65–80 per cent. If resection is incomplete, studies have shown that EBRT
given immediately after surgery improves progression-free survival and control of
seizures when compared with no EBRT, but there is no difference in overall
survival. Trials of early versus delayed radiotherapy have shown similar outcomes,
and the neuro-oncology team and the patient must weigh up the risks and benefits
of further treatment in each individual situation. However, taking into account
age, performance status, site of tumour, and patient choice, immediate treatment
may be indicated because of the risk of transformation to higher grade. Larger
volume of residual tumour is associated with a shorter time to tumour progression.
In young patients (2–3 years) radiotherapy may be deferred until recurrence
because of the increased risk of toxicity.
Comparison of radiotherapy doses of 50.4 Gy with dose escalation to 64.8 Gy
has shown increased toxicity with no improvement in disease control. There is no
proven benefit for chemotherapy, but studies are underway comparing EBRT as
sole treatment, or temozolomide alone or in combination with EBRT.
High grade gliomas

These tumours behave very aggressively with a median survival without treatment
of only 3–6 months. Because of their infiltrative nature and proximity to critical
structures, complete surgical resection is very rarely possible, but prognosis is
improved in proportion to the degree of completeness of excision. Surgery can be
followed by radiotherapy within 4–6 weeks of uncomplicated recovery. Radiotherapy
following surgery extends median survival to 9–12 months. Dose escalation above
60 Gy does not improve outcomes. Chemotherapy with low dose temozolomide
(75 mg/m2 daily) during radiotherapy and for six courses afterwards (200 mg/m2
given for 5 days every 28 days) gives an increase in median and progression-free
survival of 2.5 months and survival at 2 years of 26.5 per cent compared with 10.5
per cent with EBRT alone.
Glioblastoma multiforme can be divided into two prognostic groups on the basis
of performance status, age of the patient and treatment. Elderly patients with good
performance status may be treated as above but patients over the age of 70 with
poor performance status should be considered for biopsy and short course EBRT
for palliation of symptoms only, because of poor results of aggressive treatment.
Studies have shown no benefit from dose escalation, hyperfractionation, addition
of radiosensitiser, or wide compared with local irradiation; 90 per cent of recurrences
occur within 1–3 cm of the original site. For high grade tumours, survival after
treatment is of the order of 3 years for grade III, and 12 months for grade IV.

Gliomas can arise throughout the central nervous system including the spine and
optic nerve. In adults, most are supratentorial in the cerebral cortex but in
children, an infratentorial site is more common. Tumours expand and infiltrate
within the brain and may cross the corpus callosum, but extracranial spread does
not occur. They are best evaluated with MRI but the infiltrative nature of HGG
makes it difficult to determine tumour margins accurately.
206
Gliomas
Gliomas may present with symptoms of raised intracranial pressure, including
headache, nausea and vomiting, cognitive or behavioural problems, focal
neurological deficits or epilepsy. Spinal cord gliomas can cause pain, weakness, or
numbness in the extremities, and glioma of the optic nerve may present with visual
loss. A full general and neurological examination is needed to detect extent of
impairment. PS should be recorded. Early discussion is needed between neurologist,
radiologist, neurosurgeon and oncologist to agree an appropriate plan for each
individual. Since neurosurgical services are often located in specialised centres,
videoconferencing can be very helpful.
Whole brain or spinal CT will reveal the site of the tumour and show areas of
low density (necrosis) or calcification. However, MRI with gadolinium enhancement
is the investigation of choice.
T1-weighted sequences show low signal density and T2-weighted a high signal
density in comparison with the rest of the brain. For grade III tumours, a contrast-
enhanced CT will show a low density tumour with ring enhancement. There is
heterogeneity within the tumour and associated oedema. For grade IV tumours,
oedema will be seen outside the ring enhancement.
Grade cannot be predicted accurately by imaging and 40 per cent of tumours
diagnosed as LGG will be HGG after biopsy. PET and various functional MRI
approaches may give additional information.
Postoperative MRI is used to assess completeness of resection. If EBRT is given
in a different hospital, good liaison between surgeon and oncologist following
resection is important to maintain continuity of care for the patient.
Immobilisation
The patient lies supine with the head immobilised in an individual Perspex or
thermoplastic shell. More rigorous immobilisation with a stereotactic frame and
mouth bite is possible.
CT scanning
MRI is more sensitive than CT scanning for demonstrating tumour extent.
Tumours are non-enhancing with low signal intensity on T1-weighted and high
signal on T2. Active tumour lies mainly within areas of T2 hyperintensity but can
extend up to 2 cm from it. Since MRI cannot be used for planning treatment
alone, CT planning scans using intravenous contrast are taken with 1–3 mm slices
from the vault to the base of the skull. Pre-and postoperative MR images are then
co-registered with the CT planning scans and the target volumes delineated.

GTV
Grade I–II
The initial preoperative GTV seen on T2-weighted MRI is outlined on fused
MR/CT planning images and includes areas of peritumoural oedema shown as
low density on CT scan (Fig. 18.1).
207
(a) (b)
Figure 18.1 Comparison of (a) CT and (b) fused T2-weighted MR/CT images for low
grade glioma (G) to illustrate value of fusion. Note fluid in postoperative cavity anteriorly.
Grade III–IV
The GTV is delineated at the contrast-enhancing edge of the tumour (not
oedema) on postsurgical gadolinium enhanced T1-weighted MRI scans fused with
planning CT (Fig. 18.2).
(a) (b)
Figure 18.2 Comparison of (a) CT and (b) fused T1-weighted contrast-enhanced MR/CT
images for high grade glioma (G) to illustrate value of fusion.
208
Gliomas
For palliative treatment, the GTV includes gross visible tumour as seen on a CT
planning scan.
CTV
Two CTVs are defined according to dose to be delivered and reflect degree of
infiltration dependent on tumour grade:
Grade II
CTV54 GTV 15 mm
Grade III
CTV45 GTV 25 mm
CTV54 GTV 15 mm
Grade IV
CTV50 GTV 25 mm
CTV 60 GTV 15 mm
For palliative treatments a single phase CTV margin of 15 mm is added.
PTV
A 5 mm margin is added to the CTV taking into account departmental
measurements of set-up accuracy.
Volumes must be tailored to minimise dose to OAR, such as optic chiasm, and
take account of natural barriers to spread such as bone and falx.
OAR
These will vary according to the site of the primary tumour. They should be
outlined and a PRV added. A clinical decision about relative risks and benefits is
needed if PTVs and PRVs volumes overlap.
■ Dose solutions
Conventional
Simple coplanar beam arrangements or opposing beams defined on the simulator
using 6 MV photons may be appropriate for palliative treatments, but CT scanning
is needed to define the GTV.
Conformal
Using CT scanning and MLCs, volumes are tailored to avoid as much normal
tissue as possible. Three beam arrangements are often used which may be non-
coplanar and should be wedged as appropriate to obtain a satisfactory dose
distribution (Figs 18.3 and 18.4).
Complex
Better dose homogeneity across the tumour may be achieved using forward
planning IMRT with segmentation or ‘field in field’ arrangements. Full IMRT
may produce optimal plans to meet normal tissue dose constraints if these would
limit effective doses to tumour when long-term survival is expected (such as
treatment of optic glioma in children).
209
(a) (b)
Figure 18.3 Treatment plan for low grade gliomas as in Figure 18.1 showing (a) axial and
(b) off axis sagittal non-coplanar beams to avoid the eyes.
Figure 18.4 Treatment plan for

high grade glioma as in Figure 18.2
showing axial dose distribution.
Grade II/III
CTV 45
CTV 54
9 Gy in 5 daily fractions of 1.8 Gy given in 1 week.
Grade IV
CTV 50
CTV 60
210
Medulloblastoma and infratentorial PNET
Adjustments to this treatment approach may be made in the light of known
prognostic factors:
Grade IV, PS 0–1 age 70

As above (60 Gy) with temozolomide 75 mg/m2 daily throughout treatment.
Grade IV, PS 0–1 Age 65

40 Gy in 15 daily fractions gives equivalent control rates to higher dose radiotherapy
without temozolomide and may be preferred.
Grade IV, PS 2 or age 70 or any palliative treatment

30 Gy in 6 fractions over 2 weeks.

During the first days of treatment, there may be an increase in peritumoral
oedema, which may require adjustment or introduction of steroid dosage to
prevent headache and vomiting. Antiemetics may also be required. Consideration
to cutting out parts of shells to reduce skin dose, for example over the ears, may
help to prevent skin erythema and irritation, but adequate immobilisation must be
maintained. Shells may need to be adjusted with changes in steroid dose which
affect the amount of swelling of the patient’s face. Hair loss from the irradiated
area including sites of exit of the beam will start after about 2 weeks of treatment
and will be permanent in high dose volumes.
Regular weekly review is essential to monitor response and check medication.
This should include assessment of seizure control. Rapid deterioration during
treatment may lead to discontinuation of radiotherapy. Patients and relatives often
require considerable psychological support from the treatment team during this
period. If temozolomide is used, blood counts should be checked weekly.
■ Verification
EPIs are taken on the first 3 days of treatment and then once weekly to ensure
accuracy of set-up. Diodes or TLDs are used to check delivered dose.
Other glial tumours

Oligodendrogliomas are treated as described above, according to prognostic
factors. Choroid plexus carcinomas may be treated palliatively with short-term
improved control. With a dose of 54 Gy, symptoms of gliomatosis cerebri may be
improved for about 6 months.
Medulloblastoma and infratentorial primitive

neuroepithelial tumour
These tumours are discussed together because the radiotherapy technique of
treatment is similar. Craniospinal radiotherapy (CSRT) is also used in the
treatment of some non-germinomatous germ cell tumours and lymphomas. These
211
tumours are rare, optimal treatment is still under investigation and patients should
be cared for by multidisciplinary teams in specialist centres and entered into clinical
trials whenever possible.

Medulloblastoma and infratentorial PNET are treated with a multimodality approach.
Surgery should be as complete as possible without causing disability. Because of the
risk of subarachnoid seeding of tumour, surgery is followed by CSRT with a boost to
the posterior fossa or primary tumour site, and by combination chemotherapy.
Although radiotherapy can be deferred in very young children, most cures are only
achieved with the addition of radiotherapy, and reduced doses, hyperfractionated
regimens and reduced volumes have been associated with poorer control rates.
Control rates with concurrent radiotherapy and chemotherapy followed by multiagent
chemotherapy for 1 year are higher than with surgery and radiotherapy alone.
For medulloblastoma or PNET in adults, the role of chemotherapy is less well
established.

Surgery is considered as essential first treatment and is followed as soon as recovery
permits (usually 2–3 weeks) by radiotherapy and, for medulloblastoma, with
modified concurrent and consequent chemotherapy. Local failure remains the
commonest problem.

Medulloblastoma arises from the roof of the fourth ventricle in the posterior fossa in
the midline. CT scanning commonly shows a well-circumscribed hyperdense tumour
projecting into the fourth ventricle. There may be enlargement of the other
ventricles because of compression. Cysts and necrosis may show as areas of low
density. MRI shows more clearly the point of origin of medulloblastoma from the
roof of the fourth ventricle. On T1-weighted images the tumour is iso- or
hypodense. Leptomeningeal disease is seen as an irregular thin layer of enhancement.
■ Assessment of disease
Full neurological, endocrinological and general examination should be performed
as a baseline to facilitate documentation of response and late effects. Various tests
such as audiometry and echocardiography are indicated depending on which
chemotherapy regimen is used. Initial lumbar puncture may reveal positive
cytology. Bone scan is indicated to rule out metastases if there are suggestive
symptoms or neuraxis involvement.
Within 12–48 hours postoperatively, the patient may develop the posterior fossa
syndrome, characterised by mutism, cerebellar dysfunction, supranuclear cranial
nerve palsy and hemiparesis. Resolution may take several weeks and if the patient’s
condition is poor, EBRT may have to be deferred.
Preoperative MRI of brain and whole neuraxis is performed to assess tumour
extent and operability, and to detect any metastatic disease within the spine.
Postoperative MRI should be performed within 48–72 hours after surgery to
212
determine extent of residual disease, as after this time disease is difficult to
distinguish from postsurgical artefacts.
Immobilisation
Conventional
Formerly the patient was treated prone with an individual facial support and a shell
down over the shoulders to immobilise the head, neck and shoulders.
Conformal
Improved technology has now made it possible to treat the patient in the supine
position and this is preferred as it is more comfortable and reproducible and is safer
if general anaesthesia is required. The patient lies on a carbon-fibre couch top with
neck extended with a vacuum moulded bag to support the head and shoulders. An
individually made Perspex or thermoplastic shell covers the face and shoulders and
is attached onto the couch top. Indexed knee rests are used to ensure that the
spine is straightened and hips are also fixed in a foam form. The sides of these hip
rests act as arm rests to lift the arm above the spine. Anterior and lateral tattoos are
placed at the point of hip fixation.
CT scanning
With the patient in the supine treatment position, whole body images are obtained
with 5 mm slices from the vault of the skull to the bottom of the sacrum, with
3 mm slices through the primary tumour.
Simulator
The initial volume includes the whole brain and extends to the inferior border of
the third or fourth cervical vertebra to allow an adequate margin below the
primary tumour in the posterior fossa, to facilitate the matching of the spinal beam
and to avoid the spinal beam exiting through the mouth.
Spine
The spine is treated from the fourth or fifth cervical vertebra to the fourth sacral
foramina to include the theca and sacral nerve roots.
Primary tumour
The volume is reduced to cover the primary tumour.

CT scanning
Using CT scanning with co-registration of MR images, the GTV-T (preoperative
extent of primary tumour and any residual disease after surgery) is outlined.
Two CTVs are defined:
CTV35 whole brain and spine.
CTV54 posterior fossa, or GTV-T 1 cm margin.
PTV is determined according to departmental protocols, usually:
PTV CTV 3–5 mm.
213
All OAR for both CTV35 and 54 such as the ear, optic chiasm, pituitary,
thyroid, lungs, kidneys, ovaries or testes are outlined for DVH evaluation.
Conventional
For conventional treatment, opposing lateral beams with the lower border at C3–4
are applied to cover the whole brain, with a collimator rotation of 7–10° to match
the divergence of the posterior spinal beam. A template is made to facilitate lead
shielding of extracranial structures (such as eyes, teeth, etc.) or MLC shielding is
designed. It is important to check that cribriform plate, anterior and temporal
lobes are adequately treated. The lower border of the cranial field is tattooed.
Lateral and posteroanterior simulator films of the vertebral column are then taken.
The position of the spinal cord is marked on the lateral film and the dose at its
central axis calculated over its entire length, which extends from the junction with
the cranial field to the fourth sacral foramina. A wax compensator may be required
to improve homogeneity over this length. The width of the spinal beam ranges
from 4 cm in small children to 6 cm in adults (to cover the lateral spinal roots).
For the second phase of treatment, the anterior border passes behind the
posterior clinoid process avoiding the pituitary gland. The inferior border lies at
the bottom of the first cervical vertebra, and the superior and posterior borders are
set to cover the contents of the posterior fossa (Fig. 18.5).
Figure 18.5 Lateral BEV for posterior

fossa treatment for medulloblastoma.
■ Dose solutions
Complex
Beams are designed to cover first the whole brain and spine (CTV 35) (Fig. 18.6) and
then the posterior fossa (CTV54) defined on axial CT scans and are angled posteriorly
to avoid the external auditory meatus and cochlea. MLC is used to shield the face.
Multiple segmented beams are used to ensure a homogeneous dose throughout the
length of the spine and to prevent overdose at sites of beam junction between skull
and spine. Using beam segments, forward planned IMRT, asymmetrical jaws and
dynamic wedges, several boost fields can be added to the posterior spine to top up
areas of underdose. Tomotherapy™ may improve sparing of critical structures but
there is concern about whole body dose, especially in children.
Conformal
Whole brain irradiation is delivered using opposing lateral beams. By using MLC
with 5 mm leaves if available, the face is shielded from the lateral beams and the
214
(a) (b)
(c) (d)
(e)
Figure 18.6 Whole neuraxis for medulloblastoma treatment. (a) Posterior DRRs showing
spinal beam, (b) whole brain, (c) and (d) segment beams to create homogeneous dose
along whole spine. (e) Isodose distribution. Courtesy of Helen Taylor and Dr F Saran,
Royal Marsden Hospital.
kidneys from the posterior spinal beam. Use of posterior oblique beams, rather than
opposing laterals, for treatment to the posterior fossa, makes it possible to avoid the
ears to reduce the likelihood of deafness. Compensation may be applied at the skull
vault and neck. If the uncompensated distribution of dose over the length of the
spine exceeds the ICRU recommended limits of 7 per cent/5 per cent, a
physical compensator of aluminium, wax or Perspex may be used to even out the
overdose superiorly and the underdose inferiorly.
215
Conventional
Whole brain and posterior fossa
Treatment is given isocentrically using a linear accelerator and opposing lateral beams
as defined in the simulator. The position of the lower cranial border is shifted by 1 cm
every seven treatments to change the level of the junction with the spinal field.
Spinal field
Despite the use of an FSD extended up to 140 cm, two adjacent fields are
commonly required to cover the spinal cord in adults and older children. The
technique for calculating the gap at beam junctions is described in Chapter 2. Both
this and the craniospinal beam junction are moved caudally every seven treatments.
Using a 6 MV linear accelerator it is not possible to spare anterior abdominal
structures but shielding can be used to cover the kidneys.
The doses in both phases of the cranial treatment are prescribed to the midplane of
the posterior fossa volume. Doses received at the midplane of the whole brain volume
are also documented. If there is a variation of more than 5 per cent between the two
central doses, compensators must be used with the conventional technique. The
spinal dose is prescribed to the central spinal axis (the middle of the spinal cord).
35 Gy to whole brain and spine in 21 daily fractions of 1.66 Gy given in 41⁄2 weeks.
19 Gy boost to posterior fossa in 12 daily fractions of 1.6 Gy given in 21⁄2 weeks.

The patient is aligned using lasers and shell markings with palpation of the spine
and visualisation of the gaps between head and spine (in the prone set-up). With
a supine treatment position, gaps are more difficult to visualise and checking spine
alignment requires a suitable insert in the couch so that spines can be palpated.
A radio-opaque marker can be placed at the inferior margin of the lateral beams in
the anterior midline for visualisation with on board imaging or EPIs.
Hair loss begins with doses of 10–20 Gy and 80 per cent of those receiving
50 Gy will have permanent hair loss over the occiput. Raised intracranial pressure
should be relieved before radiotherapy starts, but steroids may nevertheless be
required to control vomiting during treatment. The mutism syndrome may make
it difficult for the patient to cooperate. Antiemetics may be required. A general
anaesthetic may be needed to treat very young patients but use of the supine
position and appropriate preparation with play therapy reduces the number for
whom it is necessary. If anaesthesia is used, care must be taken to maintain
adequate nutrition. Skin reactions, especially over the ears, can be reduced by
cutting out the shell. Deafness is a late effect of radiotherapy especially given in
conjunction with platinum-based chemotherapy, but may be reduced by the use
of conformal techniques which exclude the ear from the full posterior fossa dose.
Blood counts should be monitored at least weekly. Other side effects may include
a period of somnolence after radiotherapy, which is self-limiting.
Long-term follow-up should be ensured to diagnose and remediate any late
effects such as hypothyroidism or pituitary insufficiency.
216
Ependymoma
■ Verification
Beams and shielding are verified with portal films. TLDs or diodes are used on the
first day of treatment to measure doses including to the lens of the eye. MV or kV
imaging online is used where available before treatment to check set-up.
Ependymoma
Ependymomas may arise at any site in the neuraxis. In children, 90 per cent are
intracranial and are commonest in the posterior fossa. Spinal lesions are more
common in adults. Extent of resection is the most important prognostic factor.
Supratentorial lesions with complete gross resection in patients older than 3 years
and with benign histology have the best prognosis and require only surgery. Local
failure is the commonest problem and CSRT is not necessary for non-metastatic
disease. Radiotherapy is considered if resection has been incomplete and the
patient is over 3 years old, if the tumour has been grossly resected but there are
adverse histological features, or if the lesion is infratentorial. Children who have
not received radiotherapy at the time of surgery may be treated for tumour relapse.

Ependymomas arise from the floor of the fourth ventricle and extend laterally.
They spread by seeding in the leptomeninges. Tumours may contain calcification
and are best defined as hypodense lesions on T1-weighted MR images.
■ Assessment of disease, data acquisition

See section on gliomas.

The GTV is defined as the tumour and any residual tumour shown on pre- and
postoperative MRI fused with CT planning scans.
CTV GTV 10–15 mm.
PTV CTV 5 mm.
The optic chiasm is the main OAR.
■ Dose solutions
Treatment may be delivered conformally with three to four beams or a full IMRT
solution.
Primary brain
Primary spinal tumours
50.4 Gy in 28 daily fractions of 1.8 Gy given in 51⁄2 weeks.
For metastatic disease, CSRT is given.
217
Craniospinal
35 Gy in 21 daily fractions of 1.66 Gy given in 4 weeks, with a boost to
individual sites of spinal disease up to 50.4 Gy.
Primary tumour
As above.
Meningioma
These tumours arise from the meninges, most commonly adjacent to the falx,
along the sphenoid ridge, in the olfactory grooves, the sylvian region,
cerebellopontine angle, and the spinal cord. Most are benign (grade I) but atypical
(grade II) and malignant (grade III) types also occur. Surgery is the treatment of
choice. Radiotherapy is indicated when surgery is impossible, if there is infiltration
of adjacent tissue, biopsy shows grade II or III histology or the patient is suffering
a second or subsequent relapse after surgery. When radiotherapy is indicated, there
is evidence that it is best given early rather than deferred.

Tumours are usually well defined and localised although they may spread along the
dura. They may lie flat (en plaque) along the meninges or be lobulated and indent
the adjacent brain. Resectability is in part determined by the relationship to
adjacent structures, particularly major blood vessels, such as the superior sagittal
sinus, anterior and middle cerebral arteries and cavernous sinus, and must be
evaluated for each site. Spread through bone and foramina into the orbit, temporal
fossa and other sites may make radical excision very difficult.
CT scans usually show a well-defined hyperdense mass which enhances with
intravenous contrast. Calcification and lower density is seen in about 25 per cent
of tumours. Evidence of bone involvement may be seen. Oedema may be a
prominent feature. MRI is useful to demonstrate the attachment of the tumour to
the dura and relationship to vascular structures.

Full clinical and radiological examination as described above is carried out to assess
operability or need for radiotherapy. Tumours may be asymptomatic or may
present with focal seizures, neurological impairment or more rarely with signs of
raised intracranial pressure.
Immobilisation
See glioma. Stereotactic frames may be used when treatment with protons or
stereotactic techniques is used.
218
Meningioma
CT scanning
With the patient immobilised, CT scans are obtained from the skull vault to the
base of brain or the first cervical vertebra depending on site of origin of the
tumour, with 1–3 mm slice thickness for fusion with MR images.

If surgery is not performed, the whole tumour with any spread along the meninges
or through bone must be encompassed within the GTV. Gadolinium-enhanced
T2-weighted MR images are co-registered with CT planning scans and GTV
delineated by contouring areas of enhancement. CTV is created by adding a 5 mm
margin. Normal barriers to spread (such as bone) may in fact be invaded and this
must be taken into account if any editing of volumes is done.
After macroscopic surgical removal, information from surgeon and pathologist
must be taken into account in designing volumes. CTV is defined using presurgical
MR images fused with CT scans to identify areas at greatest risk of recurrence,
which are the point of attachment to the dura and any meningeal extensions, and
intravascular or bony involvement. Invasion into the brain is rare and therefore
volume of brain tissue included in the CTV should be minimal. A variable margin
(from 1 mm to 5 mm) which will increase with grade of tumour should be added
around these areas to allow for microscopic spread. A PTV margin is added
according to departmental protocols and measurements and is usually 5 mm.
OAR are defined according to the primary site and a PRV created and edited as
appropriate.
■ Dose solutions
Conventional
Conventional planning and treatment with opposing beams only has a place in the
palliative treatment of recurrent tumours where long-term control is not expected.
Conformal
Conformal planning and treatment delivery are essential because of the proximity
of critical normal organs. An arrangement of three 6 MV beams is commonly used,
chosen to avoid normal structures as much as possible with MLC shielding and
wedges to improve dose distribution.
Complex
Small meningiomas may be most appropriately treated by proton therapy or
stereotactic techniques for which referral to a specialised treatment centre may be
necessary. For other tumours, a non-coplanar beam arrangement with appropriate
MLC shielding and wedges should be used.
Reduced doses of 51 Gy in 30 daily fractions of 1.7 Gy over 6 weeks may be used
for tumours adjacent to optic nerves, chiasm or spinal cord.
219

See general recommendations above. Most patients are in good general condition
and few complications arise during treatment other than skin reactions.
■ Verification
See above.
Primary CNS lymphoma

Ninety per cent of intracranial lymphomas are diffuse large B cell tumours. They
are associated with immune-suppression or compromise, or Epstein–Barr virus
infection. They invade the perivascular space and in 30 per cent there is spread into
the cerebrospinal fluid (CSF) and meninges.
High dose methotrexate, which crosses the blood–brain barrier, in conjunction
with systemic chemotherapy is the most effective treatment. The role of
radiotherapy is still unclear although it is known to be effective treatment. CHOP
(cyclophosphamide, vincristine, doxorubicin and prednisolone) chemotherapy
given after cerebral radiotherapy added no further benefit. A study by the EORTC
used high dose methotrexate, teniposide, carmustine, and intrathecal therapy
before 40 Gy of whole brain radiotherapy. The overall response rate at the end of
treatment was 81 per cent, but these results were associated with significant acute
chemotherapy-related toxicity (10 per cent death rate). It is difficult to use these
results to guide treatment since the commonest age group with this condition
(patients 65) were excluded from the study. Most trials reported to date have
noted the deleterious effect of increasing age on outcome.
In patients in whom complete remission is obtained with chemotherapy, there
appears to be no significant difference in overall or progression-free survival when
radiotherapy is added. Chemotherapy, with radiotherapy deferred until further
relapse, is now recommended, except where there is residual disease after
chemotherapy when whole brain radiotherapy is given.

Lymphoma infiltrates widely throughout the brain and shows on contrast-enhanced
MRI or CT scanning as a homogeneous poorly defined low density mass.
Patients present with symptoms from tumour mass effect, or from discrete deposits
on cranial or spinal nerves or with ocular involvement in the vitreous. Complete
examination should exclude lymphoma outside the brain and full neurological
testing and slit lamp eye examination should be performed.
Data acquisition, target volume definition, dose solutions and verification are
described in the section on whole brain irradiation for intracranial metastases
below.
220
CNS irradiation for acute leukaemia
Whole brain
40 Gy in 20 daily fractions of 2 Gy given in 4 weeks
or
CNS irradiation for acute leukaemia

Patients with high risk disease and those with CNS disease at presentation, who are
assigned by a current protocol to receive cranial irradiation, have treatment to the
cranial meninges with extensions retro-orbitally and into the foramen magnum, after
complete remission has been achieved by systemic and intrathecal chemotherapy.
Palliative treatment is rarely given but may be beneficial for facial nerve
involvement or spinal cord compression when the volume is determined by the
extent of disease.
■ Radiotherapy
A thermoplastic or other type of shell is used to immobilise the patient in the
supine position and the outer bony canthi of both orbits are marked with wire or
lead pellets before simulation or virtual simulation. Two opposing lateral beams
are used to cover the whole cranium extending to the lower border of the second
cervical vertebra (Fig. 18.7). The posterior part of the orbit is included because of
the reflection of the meninges along the optic nerve. The anterior orbit and
nasopharynx are shielded with MLCs or a custom-made block. The exit dose to
the contralateral eye may be reduced by angling beams posteriorly by 5° or,
preferably, by placing the central axis of the beam at the anterior orbital margin
with appropriate shielding. Lasers are used to check alignment and portal images
are taken on the first day of treatment to ensure correct positioning of the beams
and shielding. Lithium fluoride TLDs or diodes are used to measure the dose to
the eyes, which should not exceed 10 per cent of the MPD.
Figure 18.7 Opposing lateral beams on

virtual simulation for cranial treatment for
lymphoma/leukaemia, with whole brain
covered including cribriform plate and
cervical vertebrae.
221
24 Gy in 15 daily fractions of 1.6 Gy given in 3 weeks (designed to attain the
highest antileukaemic dose and minimise effect on developing brain).

Most patients will already have alopecia from chemotherapy. Acute skin reactions
are mild. Patients may have headaches or nausea during treatment because of
concomitant chemotherapy. Most experience somnolence at 6 weeks after
treatment and should be warned that lethargy, nausea and anorexia are self-
limiting and do not require special treatment. Blood counts should be checked
regularly according to protocol.
Tumours of the sellar region

Pituitary adenoma (in adults) and craniopharyngioma (commonly in children or
young adults) are benign tumours for which surgery is the primary treatment of
choice. For pituitary tumours, radiotherapy may be indicated for residual or
progressive tumour after surgery or for recurrence, but individual assessment
should be made by the multidisciplinary team. For good risk craniopharyngiomas,
surgery is followed by careful surveillance. If resection is incomplete, radiotherapy
is indicated for poor risk tumours larger than 2–4 cm, where there has been a
breach of the floor of the third ventricle, or where the patient has hydrocephalus
or the hypothalamic syndrome. Radiotherapy is also indicated after cyst aspiration,
partial trans-sphenoidal resection, shunting or debulking.

Pituitary adenomas arise in the anterior pituitary gland within the sella, whereas
craniopharyngiomas are primarily suprasellar although 50 per cent have an intrasellar
component. Clinical presentation is related either to endocrine or hypothalamic
dysfunction or to the effects of extension outside the site of origin with compression
of surrounding brain tissue.
Contrast-enhanced MRI is the investigation of choice to demonstrate
relationship to the optic chiasm superiorly (Fig. 18.8a) and the sphenoid sinus
inferiorly. Pituitary macroadenomas will show low signal on T1-weighted images
but there may be high signal in areas of haemorrhage. In craniopharyngiomas, CT
will show calcification in a majority of tumours and 70–75 per cent have a cystic
component, which will show high signal intensity on T2-weighted MR images.
There will be contrast enhancement of solid areas.

Patients are seen initially by endocrinologist and neurosurgeon and discussed
when appropriate in a multidisciplinary meeting with the oncologist. Before
radiotherapy, there must be a full ophthalmological and endocrinological
assessment and neurological examination.
222
Tumours of the sellar region
(a) (b)
Figure 18.8 (a) MRI showing pituitary tumour (t) and relation to optic chiasm (oc).
(b) Sagittal dose distribution showing optic chiasm (yellow).
Immobilisation
The patient is immobilised in a Perspex shell or relocatable frame with the head in
a neutral position.
CT planning
CT scans are obtained with 3 mm slice thickness from the skull vault to the first
cervical vertebra with intravenous contrast, and co-registered with postoperative
gadolinium-enhanced T1-weighted MR images or planning MR scan.
Any abnormality showing enhancement on T1-weighted MR images postoperatively
is outlined as GTV for pituitary tumours. For craniopharyngioma GTV is defined
as the cystic and solid elements of the tumour defined by pre- and postoperative
MRI. Decompression of large tumour masses may lead to alterations in normal
anatomy and post-decompression changes must be taken into account. For
pituitary adenomas, no GTV to CTV margin is needed as there is no microscopic
spread, but for craniopharyngioma a GTV-CTV margin of 3–5 mm is allowed
because of the tendency of the tumour to adhere to surrounding structures.
A CTV to PTV margin is added according to departmental protocols, but is
usually 5 mm.
■ Dose solutions
Conformal
Using CT planning, MLCs are used to shape a superior oblique and opposing
lateral beams (Fig. 18.8b), or other beam arrangement to the target volume.
Complex
Three to six non-coplanar conformal beams are used to try to reduce dose to optic
structures and temporal lobes. Higher energies than 6 MV may be appropriate to
223
reduce dose to temporal lobes. Stereotactic techniques may be used for small
intrasellar lesions.
Conventional
An anterior and two opposing lateral beams are chosen to cover the target volume,
angled to avoid optic structures.
Craniopharyngioma or large pituitary adenomas
50–54 Gy in 30 daily fractions of 1.67–1.8 Gy given in 4–5 weeks.
Small pituitary adenomas
MRI at weeks 3 and 5 may make it possible to reduce the volumes if tumour
shrinkage is confirmed.
Cystic recurrences of craniopharyngioma may be treated with aspiration and
intracystic instillation of radioactive colloids such as Y-90 (half-life 2.67 days,
range 11 mm) to give a planned dose of 200–300 Gy.
Steroid and other endocrine replacement therapy is given as necessary. Visual function
should be carefully monitored each week as reaccumulation of fluid in a cystic
component of a craniopharyngioma may require urgent neurosurgical decompression.
■ Verification
See above.
Intracranial germ cell tumours

These tumours, which arise from primitive germ cell precursors, comprise 2–5 per
cent of primary CNS tumours; 75 per cent occur between the ages of 10 and 20 years.
All histological types of germ cell tumours may be seen, but 57 per cent are
germinomas (analogous to seminomas).
Introduction of effective chemotherapy regimens has led to reduction in
radiotherapy volumes and dose. A series of international trials has indicated that
cure rates of nearly 100 per cent can be achieved in germinomas with chemotherapy
and radiotherapy. Outcomes in non-germinomatous germ cell tumours are poorer
(65–70 per cent 5-year survival) and are improved when chemotherapy and surgical
excision are followed by radiotherapy if complete remission is not obtained. CSRT
is reserved for patients with disseminated disease. Children are treated within agreed
international protocols which must be consulted.

Primary treatment is with cisplatin-based chemotherapy. If complete remission is
achieved, focal radiotherapy to the primary tumour site and ventricles is given
224
Intracranial germ cell tumours
starting within 3–6 weeks of the end of chemotherapy. If there is residual disease,
surgical excision is performed before radiotherapy. For disseminated disease,
chemotherapy is given before CSRT.

Germ cell tumours may be unifocal in the pineal or suprasellar region or bifocal in
both areas (Fig. 18.9). There may be CSF involvement (M1) or macroscopic
metastases (M2/3). Tumours may be secreting or non-secreting.
Figure 18.9 MRI to show sites of origin of

germ cell tumours in the suprasellar region
(SR) and pineal gland (P).
MRI will demonstrate primary tumour extent, and special techniques including
magnetic resonance spectroscopy, diffusion-weighted or diffusion tensor imaging
may make it possible to distinguish germ cell from non-germ cell tumours.
Spread from the primary site is first to the ventricular system and thence to the
leptomeninges. Metastases to other sites occur late with uncontrolled CNS disease.
Symptoms and signs at presentation depend on site and size of tumour. There may
be raised intracranial pressure, lethargy, visual loss, hormone abnormalities or
hypothalamic disturbance.
Factors predicting unfavourable outcome include older age, metastases at
presentation, elevated markers (β-hCG or α-fetoprotein [AFP]), histological
subtype and residual disease after initial chemotherapy.
Diagnosis may be possible without surgery if the characteristic position of the
tumour is associated with elevations in β-hCG or AFP in blood or CSF. If biopsy
is needed to confirm the diagnosis, it may be achieved endoscopically with or
without a procedure designed to restore CSF flow. MRI of the spine and CSF
cytology are required to exclude disseminated disease.
Full endocrine and visual assessment should be carried out as well as baseline
measurement of blood counts and renal and liver function before chemotherapy.
Immobilisation
The patient lies supine and is immobilised in a Perspex or thermoplastic shell,
with the shoulders as low as possible and the neck in a neutral position. Foot
225
restraints and a vacuum moulded bag to immobilise the body may also be
helpful.
CT scanning
CT scans are taken of the whole brain from the vault of the skull to the first cervical
vertebra and are fused with MR images taken pre- and postoperatively.

The pre-chemotherapy/surgery GTV is defined by contrast-enhanced T1- and T2-
weighted MR scans. CT may demonstrate fatty or calcified areas better and fused
images may be used. The patient should be in remission at the time of radiotherapy
so that there is no residual GTV. Two CTVs are designed, the first corresponding
to the pre-chemotherapy tumour extent and the second to the potential sites of
spread within the ventricular system (whole ventricular radiotherapy; WVRT). The
WVRT CTV should be grown by 5–10 mm to create the PTV (Fig. 18.10). OAR
should be outlined (eyes, parotid, pituitary gland, supra- and infratentorial brain)
and PRVs designed with 3 mm margins to reduce the incidence of late effects.
Figure 18.10 Lateral DRR showing

outlining of CTV and PTV for whole
ventricular radiotherapy.
CSRT
See medulloblastoma.
■ DOSE SOLUTIONS
Conventional
Opposing lateral beams may be used to cover the PTV with shielding to other
areas of the brain.
Conformal
Conformal solutions are preferred, using either two opposing, or three or four
beam non-coplanar arrangements. DVHs are created and OAR doses are
minimised with MLC shielding.
Complex
IMRT solutions may permit sparing of supratentorial brain and pituitary gland and
make it possible to give a simultaneous integrated boost to the primary tumour site
226
Cerebral metastases
during WVRT. Care must be taken that the use of five to seven non-coplanar
beams does not result in unwanted radiation to normal areas of brain.
Germinomas
If CR after chemotherapy – WVRT (CTV 24)
If PR after chemotherapy, an additional boost – primary tumour bed (CTV40)
16 Gy in 10 daily fractions of 1.6 Gy given in 2 weeks to give a total dose of 40 Gy
in 25 fractions of 1.6 Gy given in 5 weeks.
Non-germinomatous GCT
WVRT
Primary tumour bed
An additional dose of 25.2 Gy in 14 daily fractions of 1.8 Gy given in 2.5–3 weeks.
Metastatic
CSRT
Boost to primary tumour
25.2 Gy in 14 daily fractions of 1.8 Gy.
These doses are still being investigated and current protocols should be checked.
■ Treatment delivery and patient care, verification

See medulloblastoma.
Cerebral metastases
Forty per cent of intracranial neoplasms are metastatic, arising in order of frequency
from lung, breast, melanoma, renal, and colon cancers. They are increasing in
frequency with improved treatment of the primary tumour leading to longer survival.
If CT scan or MRI confirms a single metastasis, surgery is the treatment of choice.
Postoperative whole brain radiotherapy with or without a local boost improves
progression-free survival. More commonly, MRI will reveal multiple metastases and
radiotherapy is then indicated. Median survival after whole brain irradiation for
multiple metastases is 3–6 months depending on the number of lesions and the tumour
type as well as performance status. The side effects of radiotherapy, which include hair
loss, and possibly headache, nausea, or other symptoms, are unfortunate with such a
short survival period but quality of life is nevertheless often much improved.
Stereotactic radiosurgery may be appropriate to treat small solitary or few
metastases and gives a median survival of 11 months. Highly collimated beams are
focused on the tumour to give a high dose treatment (17–18 Gy) in a single
227
session. It may be used after surgical excision for focal treatment to the area of
metastasis. Details of this technique are beyond the scope of this book.

If chemotherapy is required for systemic disease in a sensitive tumour, radiotherapy
can be deferred unless symptoms of intracranial disease worsen. Following surgical
treatment of a single metastasis, radiotherapy, either focally or to the whole brain,
may be added to improve cure rates, especially if there is any concern about
whether clear margins have been achieved.
The need for treatment of secondary tumours in the brain must be assessed in the
light of the extent of the primary tumour and other metastatic disease, and the general
condition of the patient. MRI is more sensitive in detecting intracranial metastases
and should be used if surgery is envisaged for single metastases. Routinely, contrast-
enhanced CT scanning may be more readily available and is adequate.
Immobilisation
The patient lies supine in a thermoplastic shell with the neck in a comfortable
neutral position.
Simulator
Unless focal EBRT is being given after surgical resection of a single metastasis,
planning can be adequately done with the simulator or, if available, virtual
simulation to define borders of treatment volumes (Fig. 18.11). Focal higher dose
treatment can be planned using the general principles outlined above.
Figure 18.11 Lateral DRR with BEV for

whole brain irradiation.

Diagnostic images can be fused with CT planning scans to define GTV accurately
if focal EBRT is planned. If whole brain irradiation is given, beams are defined in
the simulator to cover the whole skull from vertex to a line joining the external
228
Information sources
auditory meatus with the outer canthus of the orbit (Reid’s base line). If there is
any evidence of involvement of the skull base this line will need to be 1–2 cm lower
to ensure tumour is included in the treatment volume.
■ Dose solutions
Simple opposing lateral beam arrangements are used for whole brain irradiation
with wedging if necessary to improve dose distribution. For focal treatment, plans
are drawn up after CT scanning for optimal dose to tumour and minimal dose to
normal tissues, often using non-coplanar arrangements.
Whole brain irradiation
12 Gy in 2 daily fractions given on consecutive days.
Focal irradiation alone

40 Gy in 20 daily fractions given in 4 weeks
or
17 Gy by stereotactic radiotherapy.
key trials
Borgelt B, Gelber R, Kramer S et al. (1980) RTOG 7361 1973–1976; The

palliation of brain metastases: final results of the first two studies by the
Radiation Therapy Oncology Group. Int J Radiat Oncol Biol Phys 6:1–9.
Cairncross G–RTOG 9402, Phase III intergroup randomised comparison of
radiation alone vs pre-radiation chemotherapy for pure and mixed anaplastic
oligodendrogliomas.
EORTC Brain Tumor Group Phase III trial on concurrent and adjuvant
temozolomide chemotherapy in non-1p/19q deleted anaplastic glioma. The
CATNON intergroup trial. www.eortc.be
Mead GM, Bleehen NM, Gregor A et al. (2008) Medical Research Council
randomised trial in patients with primary cerebral non-Hodgkin lymphoma:
cerebral radiotherapy with and without cyclophosphamide, doxorubicin,
vincristine, and prednisone chemotherapy. Neurology 29: 401–2A.
Mirimanoff RO, Mason W, van den Bent M et al. (2007) Is long term survival in
glioblastoma possible? Updated results of the EORTC/NCIC phase III
randomised trial on radiotherapy (RT) and concomitant and adjuvant
temozolomide versus RT alone. Int J Radiat Oncol Biol Phys 69(3S): 2.
Information sources
Brandes AA, Franceschi E, Tosoni A et al. (2000) Long-term results of a prospective study on the
treatment of medulloblastoma in adults. Cancer 15: 1359–70.
Chang CH, Housepian EM, Herbert C Jr (1969) An operative staging system and a megavoltage
radiotherapeutic technic for cerebellar medulloblastomas. Radiology 93: 1351–9.
229
Cochrane Database of Systematic Reviews (2002) Chemotherapy in adult high-grade glioma: a
systematic review and meta-analysis of individual patient data from 12 randomised trials. Issue
4. Art. No.: CD003913. Available at: www.cochrane.org/reviews/en/ab003913.html (accessed 1
December 2008).
Gerstner ER, Carson KA, Grossman SA, Batchelor TT (2003) Long-term outcome in PCNSL patients
treated with high-dose methotrexate and deferred radiation. J Clin Oncol 21: 4483–8.
Kleihues P, Burger PC, Scheithauer BW (1993) The new WHO classification of brain tumours, Brain
Pathol 3: 255–68.
Laperriere N, Zuraw L, Cairncross G (2002) The Cancer Care Ontario Practice Guidelines Initiative
Neuro-Oncology Disease Site Group. Radiotherapy for newly diagnosed malignant glioma in
adults: a systematic review. Radiother Oncol 64: 259–73.
Louis DN, Ohgaki H, Wiestler OD et al. (2007) The 2007 WHO Classification of Tumours of the
Central Nervous System. Acta Neuropathol 114: 97–109. (See also BrainLife.org: WHO
Classification 2007 WHO Classification.)
Nieder C, Astner ST, Grosu A-L (2007) The role of postoperative radiotherapy after resection of a
single brain metastasis. Strahlenther Onkol 183: 576–80.
Ogawa K, Shikama N, Toita T et al. (2004) Longterm results of radiotherapy for intracranial
germinoma: A multi-institutional retrospective review of 126 patients. Int J Radiat Oncol Biol
Phys 58: 705–13.
Poortmans PMP, Kluin-Nelemans H C, Haaxma-Reiche H et al. (2003) High-dose methotrexate-
based chemotherapy followed by consolidating radiotherapy in non–AIDS-related primary
central nervous system lymphoma: European Organization for Research and Treatment of
Cancer Lymphoma Group Phase II Trial 20962. J Clin Oncol 44: 1210–16.
Shawl EG, Seiferheld W, Scott C et al. (2003) Re-examining the radiation therapy oncology group
(RTOG) recursive partitioning analysis (RPA) for glioblastoma multiforme (GBM) patients. Int J
Radiat Oncol Biol Phys 57: S135–6.
Stupp R, Hegi ME, Gilbert MR, Chakravarti A (2007) Chemotherapy in malignant glioma: standard of
care and future directions. J Clin Oncol 25: 4127–36.
Wolden SL, Wara WM, Larson DA et al. (1995) Radiation therapy for primary intracranial germ cell
tumors. Int J Radiat Oncol Biol Phys 32: 943–9.
230
Thyroid and 19
thymoma
Thyroid
Well-differentiated thyroid cancer (WDTC – papillary, follicular)
Total thyroidectomy and thyroid remnant ablation with iodine-131 is the
recommended initial treatment for most follicular and papillary tumours. As
patients are usually young and have an excellent prognosis, the possible late effects
of radiotherapy should be carefully considered before advocating treatment.
Adjuvant EBRT is indicated for incompletely excised or inoperable WDTC
(usually T4a or T4b) that does not concentrate iodine-131 sufficiently. The area
of incomplete excision is usually in the thyroid bed but occasionally in the neck
where there is extracapsular lymph node spread or invasion of local structures. Less
well-differentiated follicular or papillary cancers (in particular Hürthle cell
neoplasms) take up iodine poorly.
Occasionally recurrences in the neck or thyroid bed will be inoperable due to
invasion of local structures and will not concentrate iodine. EBRT is then indicated
to improve local control unless the patient has uncontrollable distant metastases.
Radiotherapy can effectively palliate metastases, for example in the bones or
mediastinum.
Medullary thyroid cancer

Surgery is the principal treatment for medullary cancer. There is no role for iodine-
131. Radiotherapy is indicated if there is inoperable or macroscopic residual local
disease unless there are uncontrollable distant metastases. It should also be
considered in patients with microscopic residual disease though careful follow-up
with calcitonin and carcinoembryonic antigen measurements, and neck imaging,
may be appropriate.
Anaplastic thyroid cancer

Many patients with anaplastic cancer are elderly, have poor performance status and
significant local symptoms. Good supportive and palliative care is paramount but
radiotherapy can be used for local control and to relieve or prevent compression
of critical structures such as the trachea and oesophagus.
Occasionally anaplastic carcinoma is operable. Adjuvant radiotherapy should
then be considered to improve local control as there are rare long-term survivors
with this disease.
231
THYROID AND THYMOMA

In follicular and papillary cancer, total thyroidectomy is followed by radio-ablation
of the thyroid remnant by iodine-131 where possible or by EBRT for residual
disease if there is no iodine-131 uptake.
In anaplastic cancer where surgery is possible, postoperative radiotherapy may be
followed by adjuvant doxorubicin-based chemotherapy to reduce the risk of distant
metastases. In the palliative setting, concurrent radiochemotherapy has been
advocated in patients with good performance status and no distant metastases but
this approach increases acute toxicity without a proven survival advantage. If
palliation of local symptoms is paramount, radiotherapy is indicated and palliative
chemotherapy can be considered subsequently for distant metastases.

Most WDTC is confined to the thyroid gland and is therefore completely excised
with a total thyroidectomy. Tumours can invade beyond the capsule into the
soft tissues of the neck, trachea, oesophagus, larynx, neck vessels or recurrent
laryngeal nerve and in these tumours a microscopically complete excision is
unusual. With anaplastic tumours, local structures have usually been invaded at
presentation.
Lymphatic drainage is initially to the midline level VI nodes in the anterior neck
which extend from the hyoid to the suprasternal notch. Subsequent drainage is to
levels III–V and then to the supraclavicular nodes, or inferiorly to nodes in the
tracheo-oesophageal groove or superior mediastinum. This region is sometimes
termed level VII and extends from the suprasternal notch to the brachiocephalic
veins. The term superior mediastinal nodes is used to avoid confusion due to the
different numbering of mediastinal node levels and those in the neck (Fig. 19.1).
Papillary cancer has a relatively high risk of lymph node metastasis though the
prognostic significance of this is debated. Extracapsular lymph node spread is a risk
factor for local recurrence.

Routine imaging is not required before surgery for WDTC unless there is clinical
suspicion of local invasion or lymphadenopathy. A CT scan without contrast
(iodinated contrast may inhibit uptake of therapeutic iodine-131) or MRI may
then be of value to document local extent and help plan surgery and possible
radiotherapy. If the patient is hoarse, nasendoscopy is used to assess vocal cord
movement and hence involvement of the recurrent laryngeal nerve.
Medullary cancer should be treated in a centre with specialist expertise.
Preoperative assessment should include genetic and biochemical tests to look for
multiple endocrine neoplasia.
If treatment is contemplated for anaplastic cancer, a contrast-enhanced CT or
MR scan of the neck should be performed as well as a CT scan of the chest and
mediastinum because of the high incidence of distant metastases.
All patients in whom radiotherapy is being considered should be discussed within
a specialist thyroid multidisciplinary team. A full discussion with the surgeon is vital
to be able to define sites at highest risk of local recurrence.
232
Thyroid
Internal jugular vein
Thyroid cartilage
Thyroid gland
Trachea
Brachiocephalic
vein
Inferior vena cava
Supraclavicular Level III

Superior mediastinal Level IV
Level VI
Figure 19.1 Lymphatic drainage of the thyroid gland.
Immobilisation
Patients should be immobilised in a custom-made Perspex or thermoplastic shell
fixed to the couch top in at least five places. The neck is extended to move the oral
cavity superiorly and to reduce dose to the mandible and salivary glands. The
shoulders should be as low as possible.
CT scanning
Slices 3–5 mm thick are obtained from the base of skull to the carina to allow
assessment of lymphadenopathy. If the target volume is smaller (for example the
thyroid bed only) the extent of the scan can be reduced. Intravenous contrast
should be avoided if iodine-131 is being considered but otherwise can be helpful
to delineate vascular structures.
Simulator
Rapidly growing anaplastic carcinomas may require urgent radiotherapy if there is
tracheal compression and this is often achieved most quickly using a simulator for
planning. Palpable disease is marked with lead wire and parallel opposed
anterior–posterior fields defined to cover the tumour and adjacent lymph nodes if
required (see target volumes). If a dose which exceeds spinal cord tolerance is
prescribed, a CT-planned phase 2 treatment is needed.
233
THYROID AND THYMOMA

Papillary, follicular and medullary cancer
A careful discussion with the surgeon and pathologist is essential to define the sites
at highest risk of local recurrence. Any macroscopic residual disease is contoured
as GTV. Any sites where tumour was excised with a known positive margin
(e.g. from the trachea or vessels) should also be contoured as GTV to facilitate
CTV definition.
The CTVs will depend on the sites of residual disease, excision margins and the
extent and risk of lymph node involvement. If tumour is inoperable or EBRT is
given because of local invasion, the GTV is expanded by 10 mm isotropically to
form the CTV66 and then edited to account for natural tumour barriers,
e.g. bone. The CTV66 is copied to form the CTV60 and edited to include the
sites of preoperative tumour. If a level VI node dissection revealed no tumour and
there are no lateral cervical nodes, there is no indication for lymph nodes to be
included in the CTV60. If there were positive level VI nodes but the lateral
cervical nodes were not dissected or nodes were not assessed surgically then level
VI, the superior mediastinal nodes, and bilateral level IV and lower level III nodes
adjacent to the thyroid bed should be included in the CTV60.
If EBRT is indicated because of extracapsular nodal spread, the CTV60 should
be defined individually depending on the extent of disease and surgery. It will
usually include the level VI, bilateral level III–V, and supraclavicular and superior
mediastinal nodes. If level III nodes were involved, then level II nodes may also
be included but this will increase both acute mucosal toxicity and the risk of late
effects, particularly xerostomia.
The CTV is expanded by 3–5 mm to form the PTV, depending on local assessment
of systematic and random errors and likely organ motion.
Anaplastic thyroid cancer

If radiotherapy is adjuvant to surgery the CTV is defined as for WDTC above.
Most radiotherapy is palliative and there needs to be a careful balance between
trying to cover all sites of possible disease and maintaining quality of life by
minimising acute toxicity. The GTV is defined as the primary tumour and involved
lymph nodes (10 mm short axis diameter). The GTV is expanded by 10 mm
isotropically to form the high dose CTV55. This CTV is edited to reflect local
patterns of tumour spread (e.g. including adjacent muscles if invaded but edited
off bone and air).
A low dose CTV35 may also be defined by expanding the CTV55 to include
uninvolved nodes in levels III, IV, V and VI and the inferior part of level II and in
the supraclavicular fossae and superior mediastinum.
There may be significant shrinkage of tumour during a long course of
radiotherapy so it may be necessary to redefine the CTV on a new planning CT
scan during treatment. The CTV is expanded by 3–5 mm to form the PTV
depending on local assessment of systematic and random errors and likely organ
motion.
When a short palliative course (20 Gy in 5 fractions or 30 Gy in 6 fractions) is
used, a 10 mm margin is added to the GTV to produce the PTV (Fig. 19.2).
234
Thyroid
Figure 19.2 Palliative radiotherapy for
anaplastic thyroid cancer. Anterior and
posterior opposing beams with MLC
shielding to reduce acute toxicity.
■ Dose solutions
Conformal
The proximity of the PTV to critical structures such as spinal cord and oesophagus
makes a conformal approach desirable in all patients with WDTC or medullary
cancer and in anaplastic tumours where a high dose is used.
Adjuvant treatment of WDTC, medullary and anaplastic cancer

The PTV is best covered by an antero-oblique wedged pair arrangement with the
beam angles chosen to avoid exit through the spinal cord. An additional anterior
beam may improve coverage but at the expense of increasing spinal cord dose.
Beams may need to be angled caudally if coverage of the superior mediastinal
nodes is needed. A wedge in the superoinferior dimension can reduce
inhomogeneities in this plane caused by the PTV being more superficial at the
superior end (Fig. 19.3).
Figure 19.3 Adjuvant radiotherapy for a small left-sided anaplastic thyroid cancer
following a total thyroidectomy without involved lymph nodes. The tumour came
close to the skin surface so bolus was used.
235
THYROID AND THYMOMA
Anaplastic cancer
The PTV35 is treated with a pair of anterior and posterior opposed beams with
MLC shielding. The PTV55 is treated as for well-differentiated tumours above.
Both phases should be planned together so as not to exceed spinal cord tolerance.
If the high dose PTV includes both the thyroid bed and level II or III nodes, it
is very difficult to treat the whole PTV in continuity without using IMRT. A pair
of lateral opposing beams angled caudally will not be able to treat posterior nodes
(cord tolerance would be exceeded) and will risk underdosing the PTV in the low
neck. The thyroid bed may be treated with two antero-oblique wedged beams
matched to electrons superiorly to treat the nodes.
Complex
IMRT can provide better target volume coverage than conformal radiotherapy for
thyroid cancer when the concave PTV surrounds the spinal cord. It is of particular
value when lateral cervical nodes are included in the PTV. It can also allow
simultaneous treatment of different volumes to different doses depending on the
risk to each. A five- or seven-beam inverse-planned technique is used.
Conventional
Where a palliative dose is being given (usually for anaplastic carcinoma), opposing
anterior and posterior beams can be defined on the simulator to cover the target
volume with an appropriate margin to the beam edge. MLC shielding can reduce
dose to adjacent normal structures.
WDTC or medullary cancer
66 Gy in 33 daily fractions given in 61⁄2 weeks to PTV66 if defined.
Anaplastic cancer
20 Gy in 5 daily fractions of 4 Gy given in 7 days.
55 Gy in 20 daily fractions of 2.75 Gy given in 4 weeks if high dose treatment
appropriate (35.75 Gy in 13 fractions to PTV35, 19.25 Gy in 7 fractions to PTV55).

Patients should be treated daily without hyperfractionation for missed treatments.
Weekly review is critical to manage acute effects proactively. As mucositis of the
pharynx and oesophagus are inevitable, dietetic advice should be provided to maintain
weight and calorie intake throughout treatment. Enteral feeding with a nasogastric or
gastrostomy tube may be necessary if weight falls more than 10 per cent from baseline.
Pain should be managed with paracetamol, non-steroidal anti-inflammatory drugs
and oral or transdermal opiates as needed. Steroids may be necessary at the start of
treatment in anaplastic cancer to reduce local oedema if there is airway compromise.
■ Verification
Portal images should be taken on days 1–3 and weekly thereafter to ensure any day-
to-day set-up variation is within tolerance. If there is significant shrinkage of tumour,
236
Thymus
a repeat CT scan may be required to evaluate the 3D dose distribution of the original
beam arrangement within the new contour. A new plan should be considered if there
is an increased dose to critical structures or the PTV dose is reduced.
Thymus
Thymic tumours are usually clinically staged according to the Masaoka system
(Table 19.1) which reflects local invasion. Pathological classification (WHO)
separates thymic carcinomas (WHO C) from thymic epithelial tumours (WHO B –
true thymoma), but clinical staging is more useful to predict prognosis and guide
treatment decisions. Surgery is the treatment of choice for thymic tumours. There
are no phase III studies investigating adjuvant treatment or to guide therapy for
unresectable disease.
Table 19.1 The Masaoka clinical staging system* for thymoma
Stage I Macroscopically and microscopically completely encapsulated (includes tumour

invading into but not through the capsule)
Stage II A: microscopic transcapsular invasion
B: macroscopic invasion into the surrounding fatty tissue or grossly adherent to
but not through mediastinal pleura or pericardium
Stage III Macroscopic invasion into neighbouring organs (e.g. pericardium, great vessels, lung)
A: without invasion of great vessels
B: with invasion of great vessels
Stage IVA Pleural or pericardial dissemination
Stage IVB Lymphogenous or haematogenous metastases
*With permission.
Adjuvant radiotherapy is recommended following incomplete excision of stage II

and III tumours but an observational approach is appropriate if excision has been
complete. In unresectable stage III or IV disease, chemotherapy and radiotherapy
are indicated followed by surgery where feasible.

Combined modality treatment with surgery, radiotherapy and chemotherapy
(cisplatin, doxorubicin and cyclophosphamide) is recommended to provide
optimal local control and survival in unresectable thymic tumours but the best
sequence of treatment is not known. Preoperative chemotherapy may shrink a
tumour to allow resection and postoperative radiotherapy. Preoperative radiation
with or without chemotherapy has also been advocated to try to allow subsequent
resection or as sole treatment. If resection is known to be incomplete, radio-
opaque clips placed at sites of residual disease during surgery can be helpful for
volume definition.
237
THYROID AND THYMOMA

The thymus sits in the anterior mediastinum behind the sternum. Fifty per cent of
thymic tumours invade through the capsule and thence into local structures
including the mediastinal pleura, pericardium or great vessels. Lymph node spread
is uncommon. Tumours can metastasise to the pleura or to distant sites.

CT or MR imaging of the mediastinum is used to define local invasion.
Mediastinoscopy can also be useful to assess resectability.
Immobilisation
Patients lie supine with their arms above the head immobilised with a T bar or
similar device. A custom-shaped vacuum bag can be used to aid immobilisation.
CT scanning
Slices 5 mm thick are obtained from the inferior cricoid cartilage to the bottom of
the thoracic cavity. Intravenous contrast can be helpful to delineate vascular
structures in the mediastinum.

When adjuvant radiotherapy is used, discussion with the surgeon and pathologist
is very important to define sites at risk of local recurrence. A GTV is defined if
there is residual macroscopic disease. The GTV is expanded isotropically by 10 mm
to form the CTV60 encompassing sites of residual disease. The CTV50 is formed
by copying the CTV60 and enlarging it to include the sites of preoperative disease,
i.e. the operative bed.
If there is no GTV, it can be helpful to contour any clips marking positive
microscopic margins. The CTV50 is then contoured as the sites of preoperative
disease with a 10 mm isotropic margin with particular attention paid to locations
where the surgeon has marked residual disease with clips, or where there is felt to
be an incomplete resection. After resection of a large tumour the postoperative
anatomy may be very different from that on preoperative imaging (Fig. 19.4).
If primary radiotherapy is the sole treatment for unresectable disease, the initial
GTV is defined on the planning CT. If chemotherapy has been given, the CTV
should incorporate all initial sites of disease and possible sites of microscopic invasion
which may include the mediastinum and adjacent mediastinal pleura and pericardium.
The CTV is expanded by 5 mm to form the PTV though this depends on local
assessment of systematic and random errors and likely organ motion.
■ Dose solutions
Three coplanar beams usually provide adequate PTV coverage without exceeding
tolerance of the lungs, heart or spinal cord. Examples include two antero-oblique
beams with an additional anterior or postero-oblique beam depending on the PTV.
Inverse planned IMRT may provide useful solutions for these rare tumours.
238
Thymus
(a)
(b)
Figure 19.4 Adjuvant radiotherapy for a Masaoka IIA (WHO B) thymoma. (a) Preoperative
diagnostic CT showing thymoma (T). (b) Corresponding slice on planning CT showing
CTV, PTV and beam arrangement. Note the radio-opaque clips used to mark sites at high
risk of microscopic residual disease (arrowed).
Adjuvant radiotherapy in stage II/III disease
60 Gy in 30 daily fractions given in 6 weeks to PTV60 if defined.
Primary radiotherapy for unresectable disease
Pre-operative radiotherapy
239
THYROID AND THYMOMA

Treatment is usually well tolerated without severe acute effects but oesophagitis can
occur and should be managed with analgesia and dietary advice (see Chapter 24).
■ Verification
Portal images should be taken on days 1–3 and weekly thereafter to ensure any
day-to-day set-up variation is within tolerance.
key trials
■ There are no phase III trials of radiotherapy in these tumours.
Information sources
■ Thyroid
British Thyroid Association, Royal College of Physicians (2007) In: Perros P (ed.) Guidelines for the
Management of Thyroid Cancer, 2nd edn. Report of the Thyroid Cancer Guidelines Update
Group, Royal College of Physicians, London. Available at: www.british-thyroid-
association.org/news/Docs/Thyroid_cancer_guidelines_2007.pdf (accessed 8 December 2008).
Cooper DS, Doherty GM, Haugen BR et al. (2006) The American Thyroid Association Guidelines
Taskforce. American Thyroid Association management guidelines for patients with thyroid
nodules and differentiated thyroid cancer. Thyroid 16: 109–42.
Meadows KM, Amdur RJ, Morris CG et al. (2006) External beam radiotherapy for differentiated
thyroid cancer. Am J Otolaryngol 27: 24–8.
Pacini F, Schlumberger M, Dralle D et al. (2006) European consensus for the management of
patients with differentiated carcinoma of the follicular epithelium. Eur J Endocrinol 154:
787–803.
Pasieka J (2003) Anaplastic thyroid cancer. Curr Opin Oncol 15: 78–83.
Schwartz DL, Rana V, Shaw S et al. (2008) Postoperative radiotherapy for advanced medullary
thyroid cancer – local disease control in the modern era. Head Neck 30: 883–8.
■ Thymoma
Giaccone G (2005) Treatment of malignant thymoma. Curr Opin Oncol 17: 140–6.
Masaoka A, Monden Y, Nakahara K et al. (1981) Follow-up study of thymomas with special
reference to their clinical stages. Cancer 48: 2485–92.
Zhu G, He S, Fu X et al. (2004) Radiotherapy and prognostic factors for thymoma: a retrospective
study of 175 patients. Int J Radiat Oncol Biol Phys 60: 1113–19.
240
Lung 20

■ Non-small cell lung cancer
Curative treatment – stage I/II
Although surgery offers the best cure rates for early stage non-small cell lung cancer
(NSCLC), some patients may be medically unfit for surgery due to comorbidity
(e.g. insufficient functional lung reserve for a lung resection) or may elect to have
radiotherapy rather than surgery after an informed discussion. In these patients
curative radiotherapy alone gives a 5-year survival of 20 per cent although studies
from single institutions report cure rates of up to 50 per cent. Cure is more likely
for early stage or smaller tumours. WHO PS 1 and significant weight loss
(10 per cent) predict for poorer outcome, presumably because they indicate
occult distant metastases. There are no phase III trials comparing surgery with
radiotherapy for these patients and indirect comparisons are difficult because
patients having radiotherapy are likely to have significant comorbidity and poorer
PS. There are also no phase III trials of curative radiotherapy compared with lower
palliative dose radiation or observation. In patients with a short life expectancy for
other reasons, both these alternatives should be considered.
In patients with poor lung function (forced expiratory volume in one second
[FEV1] 1.0 L) a balance between attempting cure and maintaining lung function
must be sought. Although there is no agreed lower cut-off below which radiation
is contraindicated, a curative dose of radiotherapy is likely to significantly impair
lung function in patients with an FEV1 of 0.6 L or grade 4 dyspnoea on the
Medical Research Council (MRC) scale.
With the increased use of CT and PET imaging and screening programmes in
lung cancer, the incidence of small primary lung cancers is likely to rise. Dose
escalation approaches such as stereotactic radiotherapy should be considered for these
patients.
Curative treatment – stage IIIA/IIIB

Surgery is only possible for a minority of stage III patients. Curative radiotherapy
is indicated in patients with PS 0–1 and 10 per cent weight loss in the preceding
3 months where disease can be safely encompassed in a radical radiotherapy
volume. It produces 2-year survival rates of 10–25 per cent. Contralateral mediastinal
(IIIB) disease increases both the risk of occult distant metastases and the morbidity
of radiotherapy. Curative radiotherapy should only be offered to patients in this
category if they have good performance status, little comorbidity and relatively
241
LUNG
small volume disease. There are no firm pretreatment volume or functional

indicators to predict which tumours can be encompassed but radiotherapy should
be attempted with caution in patients with stage III disease and FEV11.0 L or
MRC dyspnoea score 3 or 4.
The addition of either sequential or concomitant chemotherapy to radiation
improves cure rates and should be considered in patients able to tolerate combined
treatment. Hyperfractionated regimens such as the CHART protocol of 54 Gy in
36 fractions delivered with three fractions a day on 12 consecutive days offer a
small advantage over conventional fractionation but have been difficult to
implement.
The brain is a common site of relapse in patients treated for IIIA/B disease, but
prophylactic cranial radiotherapy is not widely used. There is evidence that it
reduces the incidence of brain metastases but no evidence that survival is improved.
Results of a further randomised trial (RTOG 0214) are awaited.
The majority of patients treated with palliative radiotherapy for lung cancer obtain
symptomatic benefit particularly for cough, chest pain or haemoptysis. The optimal
fractionation schedule is unclear but longer regimens produce a more durable
response and a small survival advantage compared to shorter regimens. They should
be considered for patients with PS 0–1 who are not suitable for curative radiotherapy.
Other patients should be treated with one or two fractions. There is no evidence to
support the use of palliative radiotherapy in asymptomatic patients.
Intravascular stents relieve the symptoms of SVCO faster than radiation or
chemotherapy and should be considered first. Nonetheless SVCO is effectively
palliated by EBRT in 60 per cent of patients with NSCLC and 80 per cent with
SCLC. Palliative radiotherapy should also be considered for symptomatic distant
metastases in sites such as bone, lung or skin.
Adjuvant radiotherapy
Although for many cancers radiotherapy is used as an adjunct to surgery in patients
at high risk of local recurrence, it is not recommended routinely in NSCLC. Meta-
analyses have shown an improvement in local control but an adverse effect on
overall survival, particularly in stage I/II disease. Most studies used non-conformal
techniques and lower doses of radiotherapy than would be considered today,
so the benefits of modern conformal radiotherapy as an adjunct to surgery are
unclear.
Adjuvant radiotherapy should be considered after surgery where there is
microscopic residual disease (positive resection margins) or unexpected mediastinal
lymphadenopathy at operation.
Neoadjuvant radiotherapy
Some centres recommend radiotherapy in combination with chemotherapy for
tumours that are borderline operable. However, there are only two phase III trials
(both from the 1970s) comparing the addition of preoperative radiation to surgery
and neither showed a benefit. There is no role for neoadjuvant radiotherapy outside
a clinical trial.
242
■ Small cell lung cancer
Curative thoracic radiotherapy
Radiotherapy should be considered for patients with SCLC who have tumour that
can be encompassed within a curative radiotherapy volume. Four to six cycles of
chemotherapy combined with radiotherapy are recommended because of the high
incidence of systemic metastases and the relative chemo-sensitivity of SCLC. If
patients are unable to tolerate chemotherapy, radiation alone is given if all sites of
disease can be encompassed.
Thoracic radiotherapy has proven benefit (2-year survival improved from
13 per cent to 19 per cent) if given after chemotherapy to patients with a partial or
complete response with the aim of treating all sites of initial disease. In practice, this
is mainly patients with limited stage disease, excluding those with a malignant pleural
effusion. There is also an RCT supporting the use of thoracic radiotherapy in extensive
disease if there is a complete response outside the chest and a complete or partial
response in the chest. Radiotherapy should therefore be considered in this situation.
There is also evidence to support starting thoracic radiotherapy early in the course of
chemotherapy (concurrently with cycle 1 or 2) in patients with limited stage disease.
Smoking during thoracic radiotherapy for SCLC has been shown to affect
outcomes adversely so cessation advice and support should be provided.
Palliative thoracic radiotherapy

There are no trials of palliative radiotherapy in SCLC, but the relative radiosensitivity
suggests benefits should be at least equivalent to those in NSCLC. As most patients
with good PS will receive chemotherapy, short fractionation schedules (one or two
fractions) are recommended for patients with thoracic symptoms. There are no data
comparing palliative chemotherapy with palliative radiotherapy – treatment should
be chosen on the basis of which is likely to provide better palliation for the
individual patient.
Prophylactic cranial radiotherapy in patients with SCLC

In patients with limited stage disease, prophylactic cranial radiotherapy (PCI)
reduces the incidence of brain metastases and has a small benefit in overall survival.
The additional acute toxicity of PCI and possible effects on long-term cognitive
function need to be balanced against this small benefit for individual patients. It is
given after curative chemotherapy and radiotherapy as long as there has been a
complete or near-complete response. If thoracic radiation follows chemotherapy,
the brain and chest can be irradiated at the same time.
PCI has also been shown to improve 1-year survival from 13 per cent to 27 per cent
in patients with extensive disease who respond to chemotherapy and should be
considered in this group.

■ NSCLC stage I/II treated with radiotherapy
Patients in this group undergoing radiotherapy will usually have single modality
treatment. Adjuvant chemotherapy after surgery confers a survival benefit in resected
243
LUNG
stage IB–IIIA disease. A similar benefit could be expected for the few patients with
stage IB–II disease having potentially curative radiotherapy who are fit enough to
undergo chemotherapy, but there is no clinical trial evidence to support this.
■ NSCLC stage IIIA/B treated with radiotherapy and

chemotherapy
Chemotherapy and radiation can be combined sequentially or concomitantly.
Sequential regimens use three to four cycles of platinum based chemotherapy
followed 3–4 weeks later by radiotherapy. Concomitant regimens use cisplatin-based
chemotherapy in various schedules concurrent with radiotherapy and usually have
further courses of chemotherapy either before or after concurrent treatment.
No one regimen has been shown to be conclusively superior although there are
trials showing improved survival with concomitant compared with sequential
regimens at the expense of increased acute toxicity (mainly oesophagitis and
myelosuppression).
■ NSCLC adjuvant radiotherapy and chemotherapy

after surgery
Where radiotherapy is indicated as an adjunct to surgery it should ideally begin 4–6
weeks postoperatively or when wound healing is adequate. There are phase III trial
data to support the use of adjuvant chemotherapy in resected stage IB–IIIA NSCLC
and radiotherapy was given to some patients in these trials on a non-randomised
basis. For some patients adjuvant chemotherapy and adjuvant radiotherapy may both
be indicated but there are no data on which to base sequencing of these
treatments. The risks of local recurrence and systemic metastases should be
assessed on an individual basis and radiotherapy or chemotherapy started first as
appropriate. As the majority of patients having adjuvant radiotherapy will have
positive excision margins we recommend adjuvant radiotherapy should be given
first, followed 3–6 weeks later by consideration of adjuvant chemotherapy if
indicated.
■ NSCLC palliative therapy

Patients may have indications for both palliative chemotherapy and radiotherapy.
Sequencing depends on the balance of local and systemic symptoms and the likely
benefits of each treatment. If significant local symptoms exist there is no good
reason to delay radiotherapy as the response is likely to be good.
■ Superior sulcus (Pancoast) tumours

Tumours in the superior sulcus are often considered separately from other lung
cancers and there is support for initial radiotherapy or chemoradiotherapy followed
by surgery from several phase II studies and retrospective series, with 5-year
survival rates of 15–40 per cent. The optimal management of these tumours is
unclear but there is no evidence that surgery improves cure rates when added to
(chemo) radiotherapy in NSCLC overall. We therefore recommend managing
patients with superior sulcus tumours in the same way as other NSCLC patients.
244
■ Small cell lung cancer
The optimal timing of chemotherapy and radiotherapy for SCLC is still debated.
Starting radiotherapy after chemotherapy will result in smaller treatment volumes
and therefore reduced toxicity. It will also mean that patients who respond
unusually poorly to chemotherapy will be spared the toxicity of radiation. There is,
however, increasing evidence that starting radiotherapy early (concurrent with cycle
1 or 2) improves survival, suggesting that the overall treatment time may be
important. We recommend radiation concurrent with cycle 2 in patients of excellent
performance status with relatively small volume disease. The advantage of a shorter
overall treatment time is only seen if optimal chemotherapy can still be given after
the increased toxicity of early radiation.

All pathological subtypes of lung cancer have a high rate of regional lymph node
and distant metastases, so histology alone is not a good predictor of spread. SCLC
often presents with large volume mediastinal lymphadenopathy and distant metastases
are common. Squamous cell carcinomas are more frequently confined to the thorax
than other subtypes and may cavitate.
Primary lung tumours spread within lung parenchyma where they are relatively
asymptomatic. Satellite tumour nodules may be visible on imaging. Spread within
major airways can cause obstruction, and distal collapse or atelectasis which can be
difficult to differentiate from tumour (Fig. 20.1). Tumour may invade the chest
wall, mediastinal structures, major vessels or the heart. Although invasion does not
preclude curative radiotherapy if the treated volume is small enough, it is a predictor
for distant metastasis.
(a) (b)
Figure 20.1 Tumour can be difficult to differentiate from distal collapse and consolidation
on a contrast-enhanced CT scan with either (a) lung windowing or (b) soft tissue windowing.
Lung tumours first spread to lymph nodes within the lung – the subsegmental,
segmental, lobar and interlobar nodes, which lie close to the division of bronchi,
arteries or veins. These are rarely visible on imaging but are often removed at surgery
when involvement may help define adjuvant radiotherapy volumes. Hilar nodes are
245
LUNG
situated at each hilum outside the reflection of the pleura. Adjacent tracheobronchial
nodes are inside the reflection but cannot be differentiated from hilar nodes on CT.
The lymph node stations of the mediastinum are best divided into
tracheobronchial, paratracheal, aorto-pulmonary, anterior mediastinal, subcarinal
and para-oesophageal nodes (Fig. 20.2). There are numerical classification systems
for mediastinal nodes based on surgical series.
(a) (b)
(c) (d)
Figure 20.2 Enlarged mediastinal lymph nodes on diagnostic CT scans:
(a) tracheobronchial (T) and subcarinal (S), (b) paratracheal (P) and anterior mediastinal (A),
(c) aorto-pulmonary window (AP), and (d) para-oesophageal (PO).
N staging in the TNM system reflects involvement of ipsilateral hilar (N1),

ipsilateral mediastinal (N2) or contralateral mediastinal or supraclavicular (N3)
nodes (Tables 20.1 and 20.2).
The right lung drains systematically to hilar and adjacent mediastinal nodes
(e.g. right upper paratracheal for upper lobe tumours, right lower paratracheal and
subcarinal for middle and lower lobes). Left lower lobe tumours drain to the para-
oesophageal, subcarinal, left paratracheal and aorto-pulmonary nodes. In contrast the
left upper lobe can also drain to the right paratracheal nodes. Skip metastases to the
mediastinal nodes when hilar nodes are negative occur in up to 15 per cent of tumours.
The criterion for nodal involvement on CT is a short axis diameter of 10 mm,
but 15 per cent of nodes smaller than 10 mm contain tumour if removed and over
30 per cent of 2 cm nodes are pathologically negative. The likely pattern of nodal
spread, number of enlarged nodes, node shape and presence of obstructive
246
pneumonitis or other lung pathology (e.g. sarcoidosis) all need to be considered
when assessing nodes on cross-sectional imaging as well as size. Primary tumour
size is also correlated with N stage.
Table 20.1 UICC TNM (6th edn, 2002) staging of lung cancer*
T1 Tumour 3 cm or less in greatest dimension, surrounded by lung or visceral pleura,
without bronchoscopic evidence of invasion more proximal than the lobar bronchus
(i.e. not in the main bronchus)
T2 Tumour with any of the following features of size or extent:
■ More than 3 cm in greatest dimension
■ Involves main bronchus, 2 cm or more distal to the carina
■ Invades the visceral pleura
■ Associated with atelectasis or obstructive pneumonitis that extends to the hilar
region but does not involve the entire lung
T3 Tumour of any size that directly invades any of the following: chest wall (including superior
sulcus tumours), diaphragm, mediastinal pleura, parietal pericardium; or tumour in the
main bronchus less than 2 cm distal to the carina, but without involvement of the carina;
or associated atelectasis or obstructive pneumonitis of the entire lung
T4 Tumour of any size that invades any of the following: mediastinum, heart, great vessels,
trachea, oesophagus, vertebral body, carina; or separate tumour nodules in the same
lobe; or tumour with malignant pleural effusion
N1 Metastasis to ipsilateral peribronchial and/or ipsilateral hilar lymph nodes, and
intrapulmonary nodes including involvement by direct extension of the primary tumour
N2 Metastasis to ipsilateral mediastinal and/or subcarinal lymph nodes
N3 Metastasis to contralateral mediastinal, contralateral hilar, ipsilateral or contralateral
scalene, or supraclavicular lymph node(s)
* With permission.
Table 20.2 UICC TNM (6th edn, 2002) stage groupings for lung cancer*
IA T1 N0 M0
IB T2 N0 M0
IIA T1 N1 M0
IIB T2 N1 M0
T3 N0 M0
IIIA T1–3 N2 M0
T3 N1 M0
IIIB T1–4 N3 M0
T4 N0–3 M0
IV Any T Any N M1
* With permission.

Initial assessment and imaging aims to identify disease that is either operable or
treatable within a curative radiotherapy volume. The primary tumour is imaged with
247
LUNG
chest radiograph and contrast-enhanced CT with particular attention to differentiating

tumour from distal collapsed lung and to assessing mediastinal invasion. CT imaging
for staging should include the low neck nodes, liver and adrenal glands. MRI may be
useful to look for local invasion of the brachial plexus in superior sulcus tumours.
A flexible or rigid bronchoscopy is used to determine endobronchial tumour extent
as well as to obtain histology. Cytological confirmation of a possibly malignant
effusion may be necessary if it would alter management.
CT will allow assessment of mediastinal lymph nodes on size criteria but this has
a relatively high false negative (40 per cent) and false positive (20 per cent) rate.
The number of visible nodes, their shape and location can provide further
information. If available, a PET-CT can be used to assess enlarged mediastinal nodes
as well as distant metastases, and should be considered in all patients eligible for
curative treatment for NSCLC (Fig. 20.3). Enlarged mediastinal nodes that are
visible on PET-CT should ideally be confirmed on biopsy at mediastinoscopy as
there is still a significant false positive rate though the false negative rate is small.
If a NSCLC is N0 on CT, a staging PET will still detect occult mediastinal nodes
in some patients. Overall PET-CT will change the treatment in patients eligible for
curative radiotherapy in up to 40 per cent of cases.
(a) (b)
Figure 20.3 Diagnostic PET-CT can change treatment volumes by identifying involved
mediastinal lymph nodes. (a) Contrast-enhanced CT showing a right paratracheal node
9 mm in short axis diameter (arrowed). (b) Corresponding slice on PET-CT showing
increased uptake. This node should be contoured as GTV.
Distant metastases in the liver, adrenals or bones are assessed using the staging
CT scan and PET-CT if performed. Bone scans and brain imaging are useful in
the presence of unexplained bone pain or neurological symptoms but are not
performed routinely.
Data acquisition
■ Immobilisation
Patients should be positioned supine with arms immobilised above the head in a
comfortable, reproducible position to allow a greater choice of beam angle. The
patient holds on to a T-bar device with their elbows supported laterally (Fig. 20.4).
A knee support provides a more comfortable and therefore reproducible set-up.
248
Data acquisition
Figure 20.4 Immobilisation
for thoracic radiotherapy
with a T-bar, lateral elbow
supports and a knee rest.
If treatment delivery is prolonged (e.g. with respiratory gating or stereotactic

radiotherapy) a vacuum bag should be used to reduce movement.
■ Simulator
If AP beams are to be used for palliation, the borders can be defined in the simulator.
The beam centre is marked with a reference tattoo and a photograph of the borders,
drawn on the skin, is taken to aid set-up for treatment. Fluoroscopy can be used
to view tumour movement but the accuracy of assessment in two dimensions is not
enough to predict margins required.
■ CT scanning
CT scans are obtained from the cricoid cartilage to the superior aspect of the L2
vertebra to allow the lung DVH to be calculated. Ideally 3–5 mm slices are used
both to aid volume definition and to create high quality DRRs to aid verification.
A separate simulator verification visit can then be omitted. An isocentre is tattooed
in the CT scanner, as are lateral reference points. Intravenous contrast may help
define mediastinal extent of disease but is not essential if a contrast-enhanced
diagnostic CT scan is available.
A co-registered PET scan can be used to aid volume definition. Defining volumes
on a PET-CT image may help distinguish tumour from collapsed lung. Relatively
poor spatial resolution, movement artefact and difficulty in defining the edge of a
PET positive mass make this an experimental technique at present.
■ Tumour motion
A free breathing ‘slow’ CT scan with a single slice scanner is the simplest method of
acquiring 3D data for planning. Depending on the speed of image acquisition, some
motion will be accounted for with a single CT dataset. Target volumes are then
defined taking into account possible organ motion based on population studies.
Whilst some tumours move several centimetres in one or more planes in a
respiratory cycle, others are relatively stationary. Moreover, it is difficult to predict
from the location of a tumour how much movement there will be. There are several
ways to account for individual tumour motion: imaging the tumour at extremes
of motion thus incorporating motion into the CT data, limiting mobility by
249
LUNG
breath-holding or diaphragmatic compression, tracking respiration with

respiratory gating and tracking the tumour itself with IGRT.
A very slow CT dataset (4 s per slice) taken during normal breathing will effectively
provide a composite image of the tumour as it moves so that volumes defined on
the CT scan already take motion into account. Image quality may be reduced
making accurate GTV definition more difficult. Fast images taken at maximum
inspiration and expiration can be co-registered with a free breathing scan, enabling
margins of tumour movement in each plane to be determined so that extremes of
motion are encompassed. These techniques allow individualisation of margins
while the patient is treated breathing normally.
Several techniques to limit respiratory movement have been researched including
breath-holding and abdominal compression but they are difficult to reproduce
reliably for each fraction of treatment as they are poorly tolerated by patients. In
addition, this approach is time consuming and requires patient education with audio
and/or visual cues.
Respiratory gating uses a fiducial marker on the chest wall to switch the CT
scanner (and linear accelerator) on in a certain phase of respiration. This approach
assumes that an external marker of respiration correlates with internal tumour
movement which is not always the case. The baseline thoracic volume can also vary
from day to day so a relative measure of tidal volume may not correlate with an
absolute position of a tumour.
A 4D CT scan uses a fast multi-slice scanner correlated to the respiratory cycle.
Several CT datasets are therefore obtained, each at a different point in the respiratory
cycle. A composite GTV including all positions of the tumour in the respiratory
cycle can thus be created allowing individual tumour motion to be accounted for
in volume definition.

■ Curative
To define the GTV accurately it is important to have all diagnostic imaging
available – ideally on a separate workstation adjacent to the planning computer.
Clinical information and bronchoscopy or mediastinoscopy reports should be
available. In the case of adjuvant radiotherapy, a discussion with the surgeon and
pathologist is essential to define sites at highest risk of relapse. Clips placed at
surgery at sites of incomplete excision are valuable, but must not be confused with
clips used to ligate vessels. Uncertainty in GTV definition can be reduced by a
radiologist and oncologist collaborating to define the GTV.
The parenchymal extent of the GTV should be defined with CT images viewed
on a standardised lung window setting (Fig. 20.5). The spiculated edge of the
tumour is included within the GTV. Anatomical knowledge, contrast CT scans
and radiologist input can help differentiate tumour from normal structures such as
pulmonary vasculature and the azygous vein. Tumour can be very difficult to
differentiate from adjacent lung collapse or atelectasis.
CT images should be viewed on a mediastinal window setting to define mediastinal
extent of disease and any involved lymph nodes. Contrast-enhanced images (planning
or diagnostic scan) and PET-CT if available may help define local mediastinal extent.
250
(a) (b)
Figure 20.5 GTV definition with the appropriate window setting on a planning CT scan.
(a) Lung window setting to define the primary tumour (T) within the lung. (b) Same slice on
mediastinal window setting to define any invasion into the mediastinum and involved
nodes (N). Note primary tumour appears smaller.
If chemotherapy is used before radiation the GTV should include all sites of
disease at presentation, e.g. any enlarged nodes that have shrunk to less than
10 mm with treatment.
The GTV is grown isotropically to produce a CTV. There is evidence to suggest
that an 8 mm margin from GTV to CTV is adequate to cover microscopic disease in
lung tissue. The CTV is edited to take account of natural barriers to tumour spread
(e.g. uninvolved bone or great vessels). It is extended to encompass likely patterns of
spread and to encompass the original GTV if neoadjuvant chemotherapy has been
used. In practice this usually produces a variable GTV-CTV margin throughout the
volume but adding an isotropic margin which is then edited will be less prone to
systematic errors than manually defining the CTV (Fig. 20.6). Elective nodal
irradiation is not recommended as most recurrences after radiotherapy are within the
primary tumour or as distant metastases rather than as isolated nodal recurrences.
Figure 20.6 Defining the

CTV in lung cancer. The
GTV is grown isotropically
by 8 mm (dashed line) and
then edited to take account
of natural tumour barriers
and likely patterns of
spread to create the final
CTV.
CTV to PTV margins depend on tumour motion and day-to-day set-up errors.
The latter should be measured in each department and should be approximately
5 mm in each direction. The margin added for tumour motion will vary from zero
in the case of perfect IGRT using implanted fiducial markers, to a standard
solution if individual motion is not accounted for. In practice, margins of 7 mm
251
LUNG
axially and 12 mm longitudinally are added to account for tumour motion and set-
up errors. While this may not reflect extremes of movement for all tumours, larger
margins may make it difficult to keep within dose constraints for critical normal
structures. Some centres advocate margins as small as 5 mm for GTV-CTV and
CTV-PTV, on the basis that local control in lung cancer is primarily determined
by the dose that can be delivered to the GTV rather than to the edge of the PTV.
The spinal cord is contoured on axial slices throughout the PTV and an
isotropic 5 mm margin applied to produce a PRV. The oesophagus is contoured
throughout the PTV if the tumour is close to or involves the mediastinum.
Automatic contouring tools can be used to contour the lungs. A lung minus PTV
structure is then constructed by subtraction.
■ Palliative
If CT planning is used, the GTV is defined as above and a 10 mm margin applied
to produce a PTV on the basis that larger margins would reduce the therapeutic
ratio in the palliative setting. If beams are defined on the simulator, the diagnostic
CT images can still be used to define a virtual GTV which can be superimposed
onto the simulator radiograph. A 15 mm margin from this virtual GTV to the
beam edge will give the same effect.
Dose solutions
■ Conventional
Palliative therapy given in one or two fractions can be defined in the simulator as
above. Anterior and posterior photon beams are used with dose prescribed to the
midplane. MLC shielding may reduce the dose to normal lung tissue; 6 MV
photons are adequate unless the separation at the centre is more than 28 cm, in
which case a higher energy (e.g. 10 MV) is needed.
■ Conformal
There are several challenges to covering the PTV within ICRU targets while
maintaining toxicity at acceptable levels. The location and size of the PTV and its
proximity to critical structures, particularly the oesophagus and spinal cord, often
necessitate a compromise in choosing the most acceptable plan for an individual
patient.
When curative radiotherapy is used for stage I or II disease a three-field conformal
plan is commonly used. Many tumours are closer to the chest wall than to the
mediastinum and ipsilateral beams will minimise the dose to contralateral normal
lung tissue and normally provide a homogeneous dose distribution (Fig. 20.7).
Beam angles are chosen to reduce lung dose with anterior oblique, posterior
oblique and lateral beams often used. Wedges compensate for the obliquity of the
beams in relation to the chest wall, and MLC shielding is used to conform each
beam shape to the PTV.
In stage IIIA disease a similar three-field conformal plan is often used. The beam
angles are chosen using the BEV tool while viewing the PTV, spinal cord PRV and
oesophagus contour so as to reduce dose to the spinal cord and oesophagus as
much as possible and with the aim of minimising lung dose.
252
Dose solutions
Figure 20.7 Beam arrangement and dose
colourwash for a T2N0 lung cancer
situated posteriorly in the right lower lobe.
Lung windowing is used to demonstrate
the primary tumour more clearly.
For larger tumours, in particular those crossing the midline in the mediastinum,
it is more difficult to cover the medial extent of the tumour with ipsilateral beams.
Adding a contralateral beam will significantly increase the dose to normal lung.
Such tumours can be treated in two phases. In phase 1, an arrangement of
opposing MLC-shaped photon beams ensures all the mediastinal extent is treated
and also reduces lung dose, at the expense of a poorly conformal plan with
oesophagus and cord within the treated volume. Phase 2 uses a conformal three-
or four-beam plan which will give a higher lung dose. By varying the number of
fractions in each phase it can be possible to deliver adequate dose to the PTV and
remain within cord, lung and oesophageal tolerance (Fig. 20.8).
Primary tumours close to the spinal cord may have a PTV that is very close to,
or even overlaps the spinal cord PRV. Rather than miss some of the PTV
throughout a course, it is preferable to use two phases for treatment accepting full
dose to the cord for phase 1 with the addition of MLCs to shield the cord from at
least two beams in phase 2. Again, the number of fractions in each phase is varied
until the sum of the two plans is within to cord tolerance.
In SCLCs or NSCLCs with N2 disease, the PTV will often be very close to the
oesophagus. The risk of grade 3 acute toxicity or long-term oesophageal stricture
is related to the length of oesophagus in the treated volume and ideally this is kept
below 8 cm by choosing beam angles carefully. For some patients it may be
appropriate to accept some underdosing of the PTV to achieve this. Concomitant
chemotherapy in either non-small cell or small cell disease increases the risk of
oesophagitis.
The plan should be assessed carefully by the oncologist using DVHs. We recommend
aiming to keep the V20 (volume of lung minus PTV receiving 20 Gy) below
32 per cent. In vivo, the risk of radiation pneumonitis depends not only on the dose
absorbed but also on the use of concomitant chemotherapy, pre-existing lung disease,
the lobe being treated and vascular perfusion. A 3D dose distribution cannot take
these points into consideration and is then reduced to a 2D graphic illustration – the
DVH. To then base an assessment of pneumonitis solely on whether this curve is
above or below one point on the graph is perhaps an oversimplification. Other targets
such as minimising V5 and having a mean lung dose lower than 20 Gy have also been
shown to correlate with pneumonitis. A V20 of 32 per cent is a useful target but for
patients with poor lung function, a lower target may be needed, whereas for those
253
LUNG
(a)
(b)
(c)
Dose (Gy)
20.0
48.0
38.0
40.0
66.0
68.0
44.0
26.0
58.0
50.0
78.0
22.0
24.0
30.0
36.0
42.0
56.0
62.0
64.0
14.0
60.0
18.0
46.0
72.0
28.0
74.0
76.0
32.0
52.0
54.0
10.0
12.0
16.0
34.0
70.0
0.0
2.0
4.0
6.0
8.0
100.0 100.0
V20
90.0 90.0
80.0 80.0
70.0 70.0
A-P Beams
60.0 3 Beams 60.0
50.0 50.0
40.0 40.0
30.0 30.0
20.0 20.0
10.0 10.0
0.0 0.0
0.0
3.0
6.0
9.0
12.0
15.0
18.0
21.0
24.0
27.0
30.0
33.0
36.0
39.0
42.0
45.0
48.0
51.0
54.0
57.0
60.0
63.0
66.0
69.0
72.0
75.0
78.0
81.0
84.0
87.0
90.0
93.0
96.0
99.0
102.0
105.0
108.0
111.0
114.0
117.0
120.0
(d) Relative dose (%)
Figure 20.8 Treatment of a large T3N2 NSCLC with a two-phase technique. (a) AP beams
to minimise lung dose. (b) Three-beam plan to reduce cord dose. (c) BEV of posterior
oblique beam from the three-beam plan. The beam angle is chosen to maximise the gap
between the spinal cord PRV and the PTV. (d) DVH for lung-PTV for the AP beams and the
three-beam plan. The V20 would be 46 per cent if the three-beam plan were used for the
whole treatment (66 Gy in 33 fractions) so a two-phase technique is necessary.
with good lung function a value closer to 40 per cent may be acceptable. It is
therefore difficult to use one value in assessing all plans.
Similarly, there are several parameters of the oesophageal DVH identified that
correlate with the risk of grade 3 or 4 acute oesophagitis or the risk of strictures.
254
Dose-fractionation
We recommend aiming to keep the length of oesophagus within the treated
volume to less than 8 cm where possible and certainly to less than 12 cm. Other
variables including the V40 or V50 have also been shown to correlate with toxicity
in some series.
Beam energies above 10 MV should be avoided because secondary electrons
have a greater range in lung tissue so higher energy photon beams have a wider
penumbra. Modern dose calculation engines such as voxel Monte Carlo will
produce more accurate dose calculations within the lung than the older pencil
beam convolution algorithms implemented in most planning systems.
■ Complex
The principal challenges in lung cancer radiotherapy are volume definition and
accounting for motion. Complex treatment dose solutions should only be used
when tumour motion is also accounted for. In such circumstances IMRT may
produce more conformal dose distributions, particularly for tumours close to the
spinal cord.
Dose-fractionation
■ Curative NSCLC
66 Gy in 33 daily fractions given in 61⁄2 weeks.
■ Adjuvant NSCLC
■ Curative SCLC
Consider
66 Gy in 33 daily fractions given in 61⁄2 weeks
or
45 Gy in 30 fractions of 1.5 Gy treating twice daily given in 3 weeks.
■ Palliative
10 Gy in 1 fraction.
16 Gy in 2 fractions of 8 Gy a week apart (the dose has been reduced from
17 Gy in 2 fractions to minimise the risk of spinal cord damage if survival is
prolonged).
■ If PS 0–1 and life expectancy 6 months, consider

20 Gy in 5 fractions of 4 Gy given in 1 week
or
36 Gy in 12 fractions of 3 Gy given in 4 weeks (39 Gy in 13 fractions if spinal cord
not in treated volume).
255
LUNG
■ Prophylactic cranial irradiation


Patients should be reviewed weekly by a trained radiographer or physician so that
acute side effects are treated proactively. A mild increase in dyspnoea or cough is
common but rarely needs treatment though intercurrent infections should be
excluded. Advice on skin care is given. When the oesophagus is within the treated
volume, pain on swallowing and dysphagia usually begin in the third week of
treatment. Systemic analgesia, topical local anaesthetic agents and advice on soft
and high calorie diets from a dietician should be available.
Verification
Ideally the treatment isocentre on a DRR from the CT simulation is compared
with portal images of the isocentre on the treatment machines using electronic
portal imaging. Off-line correction protocols are used for standard conformal
treatment. Images are taken on days 1–3 and weekly thereafter with a correction
made if the mean error in any one plane is 5 mm. If stereotactic radiation or IGRT
is used, online correction protocols are necessary.
Other points
A single fraction of HDR brachytherapy can provide useful palliation in NSCLC.
Other local treatments such as stent insertion, laser debulking and cryotherapy can
produce similar results and the choice of treatment depends largely on local expertise.
EBRT is likely to be more effective in patients who have not had radiation before,
but brachytherapy can offer good palliation after curative radiotherapy.
key trials
Saunders M, Dische S, Barrett A et al. (1999) Continuous, hyperfractionated,

accelerated radiotherapy (CHART) versus conventional radiotherapy in non-small
cell lung cancer: mature data from the randomised multicentre trial. CHART
Steering Committee. Radiother Oncol 52: 137–48.
Slotman B, Faivre-Finn C, Kramer G et al. (2007) Prophylactic cranial irradiation
in extensive small-cell lung cancer. N Eng J Med 357: 644–72.
Turrisi AT 3rd, Kim K, Blum R et al. (1999) Twice-daily compared with once daily
thoracic radiotherapy in limited small-cell lung cancer treated concurrently with
cisplatin and etoposide. N Engl J Med 340: 265–71.
Information sources
Alberts WM (2007) Diagnosis and management of lung cancer executive summary: ACCP
evidence-based clinical practice guidelines, 2nd edition. Chest 132(3 suppl): S1–19.
256
Information sources
Auperin A, Arriagada R, Pignon JP et al. (1999) Prophylactic cranial irradiation for patients with
small-cell lung cancer in complete remission. Prophylactic Cranial Irradiation Overview
Collaborative Group. N Engl J Med 341: 476–84.
Cancer Care Ontario Lung Cancer Evidence-based Series and Practice Guidelines (2005)
Postoperative Adjuvant Radiation Therapy in Stage II or IIIA Completely Resected Non-Small
Cell Lung Cancer: Cancer Care Ontario Practice Guidelines Initiative. Available at:
www.cancercare.on.ca/english/home/toolbox/qualityguidelines/diseasesite/lung-ebs/ [cited
2008 October 11] (accessed 26 September 2008).
Chapet O, Kong FM, Quint LE et al. (2005) CT-based definition of thoracic lymph node stations:
an atlas from the University of Michigan. Int J Radiat Oncol Biol Phys 63: 170–8.
De Ruysscher D, Pijls-Johannesma M, Bentzen SM et al. (2006) Time between the first day of
chemotherapy and the last day of chest radiation is the most important predictor of survival in
limited-disease small-cell lung cancer. J Clin Oncol 24: 1057–63.
Lester JF, Macbeth FR, Toy E et al. (2006) Palliative radiotherapy regimens for non-small cell lung
cancer. Cochrane Database Syst Rev (4): CD002143.
National Collaborating Centre for Acute Care, London. Available at:
www.nice.org.uk/Guidance/Topic/Cancer/Lung (accessed 8 December 2008).
National Institute for Health and Clinical Excellence (2005) Lung Cancer: Diagnosis and Treatment.
Clinical guideline 24. Available at: www.nice.org.uk/guidance/CG24 (accessed 5 December
2008).
Pignon JP, Arriagada R, Ihde DC et al. (1992) A meta-analysis of thoracic radiotherapy for
small-cell lung cancer. N Engl J Med 327: 1618–24.
Rowell NP, Williams CJ (2001) Radical radiotherapy for stage I/II non-small cell lung cancer in
patients not sufficiently fit for or declining surgery (medically inoperable). Cochrane Database
Syst Rev (2): CD002935.
Rowell NP, O’Rourke NP (2004) Concurrent chemoradiotherapy in non-small cell lung cancer.
Cochrane Database Syst Rev (4): CD002140.
Scottish Intercollegiate Guidelines Network Guidance 80 (2005) Management of Patients with Lung
Cancer. National Clinical Guideline 80. Available at: www.sign.ac.uk/guidelines/published/
index.html (accessed 5 December 2008).
Senan S, De Ruysscher D, Giraud P et al. (2004) Literature-based recommendations for treatment
planning and execution in high-dose radiotherapy for lung cancer. Radiother Oncol 71:
139–46.
257
21 Mesothelioma

For many years mesothelioma has been regarded as relatively resistant to
radiotherapy. The increasing incidence of this disease and interest in new
therapeutic strategies with curative intent has led to more patients with
mesothelioma receiving radiotherapy.
A few patients with good PS (WHO 0 –1), low comorbidity and early stage disease
confined to the thoracic pleural surface are treated with combined modality
therapy with curative intent. This comprises chemotherapy, surgery (extrapleural
pneumonectomy [EPP] – en bloc removal of the parietal pleura, lung, diaphragm
and pericardium) and adjuvant radiotherapy. Published single centre series in
highly selected patients support this intensive approach with median survival rates
of 20–30 months, but there is no confirmatory phase III evidence. Eligible patients
should be included in phase III studies where possible or at least treated at centres
with experience of this programme. To irradiate one hemithorax while minimising
dose to critical structures (heart, spinal cord, liver, kidneys and contralateral lung)
presents a major radiotherapeutic challenge. The lowest local recurrence rates with
acceptable toxicity have been achieved by treating the hemithorax to 54 Gy.
Comparative series suggest that lower doses are less effective.
■ Prophylactic radiotherapy
Radiotherapy has been given to prevent painful subcutaneous masses forming at
sites of trans-thoracic interventions such as pleural biopsies or chest drains. Seeding
of malignant cells along such needle tracts is reported to occur in 15–50 per cent
of sites, but there is no good evidence that such masses are symptomatic in most
patients. There are a number of phase III studies comparing prophylactic
radiotherapy with no radiation but all are small and underpowered. Interventions
such as CT-guided fine needle biopsies may pose less risk of seeding than large-
bore chest drains used for treatment of pleural effusions. In the absence of evidence
either in favour of or against this practice, we recommend considering prophylactic
radiotherapy for patients of PS 0–1 who have had a chest drain or thoracotomy
(Fig. 21.1). Ideally it is performed within 2 weeks of the intervention.
Palliative radiotherapy can help alleviate chest wall pain in mesothelioma. Several
small single centre series suggest response rates of 20–70 per cent but the doses
258
Figure 21.1 Prophylactic radiotherapy
with 12 MeV electrons to the site of a
chest drain. The beam edge is marked in
pen – 2 cm from the drain site.
used and methods of response evaluation were very variable. There is no good
evidence of a dose–response relationship. Mesothelioma often presents as diffuse
pleural disease and it can be difficult to decide which part of the pleural disease
should be targeted with palliative radiotherapy. Radiotherapy is best considered as
one method of pain control alongside opiates, antineuropathic agents and
cordotomy.

If radiotherapy is used as part of multimodality therapy, it should follow
neoadjuvant chemotherapy and EPP, ideally commencing 8–10 weeks after
surgery. There is no evidence that high dose hemi-thoracic radiation has any
benefit outside this protocol.

Mesothelioma usually spreads locally within the pleural space on the parietal and
mediastinal pleural surfaces and can invade the chest wall and mediastinum.
A contrast-enhanced CT scan of the thorax and upper abdomen is used to assess
the degree of pleural space involvement and mediastinal lymphadenopathy.

If multimodality therapy is to be attempted careful patient selection is very
important. All such patients should have a mediastinoscopy to evaluate mediastinal
lymph nodes, and a staging PET scan to exclude patients with inoperable disease.
Patients should be assessed by a multidisciplinary team including medical
oncologist, surgeon and radiation oncologist before commencing the treatment
programme to ensure they are eligible for all modalities involved.
If palliative radiotherapy is to be of value, it needs to be targeted to a site likely
to be causing pain and this can be difficult to assess given the diffuse nature of
mesothelioma. A careful history and clinical examination are most useful but chest
radiographs or CT scans can show rib or vertebral destruction, which may
correlate with sites of pain.
259
MESOTHELIOMA
Data acquisition
■ Immobilisation
For adjuvant radiotherapy patients should be supine with arms above the head in
a comfortable and reproducible position. A head rest, knee pillow and arm support
may be helpful, and a vacuum bag for the thorax should be considered.
Patients for prophylactic or palliative electron treatment should be placed in a
comfortable position on the simulator couch which allows skin apposition of the
electron applicator.
■ CT scanning
For adjuvant radiotherapy, drain sites and all chest incisions other than the median
sternotomy scar are marked with radio-opaque wire. CT slices no more than 5 mm
thick are obtained from the cricoid cartilage to the iliac crests. A treatment isocentre
is tattooed at the time of CT scanning in order to reduce systematic set-up errors.
■ Simulator
Palliative radiotherapy can be planned on a simulator if a single electron beam or
parallel opposing photon beams are used. Painful chest wall masses or sites of
origin of pain are marked with radio-opaque wire before screening.

The entire pleural cavity should be marked with radio-opaque clips at the time of
surgery to facilitate CTV definition. Clips at the insertion of the diaphragm and on
the pleural reflections are especially useful. Volumes are defined after consultation
with the surgeon and pathologist to identify high risk sites in the hemithorax such
as residual macroscopic or microscopic disease or locations where there was
tumour on the mediastinal pleural surface.
The GTV is residual macroscopic disease if present. The CTV1 (54) comprises
the GTV with a minimum 10 mm margin in each plane and any sites of
microscopic residual disease (positive resection margins). The CTV2 is defined as
the entire ipsilateral thoracic cavity from lung apex to insertion of the diaphragm,
ipsilateral mediastinal pleura, mediastinal tissues at sites where there was evidence
for tumour invasion, the ipsilateral pericardial surface, and full thickness of the
thorax at the sites of thoracotomy and chest tube incisions. It is very challenging
to treat all the CTV2 to 54 Gy so some dose compromise to this large volume may
be unavoidable. It can be very helpful to review the CTVs with the thoracic
surgeon once they have been defined.
The CTVs are grown by at least 10 mm to produce the PTVs, but this margin
should be adjusted according to measured departmental errors for hemithoracic
radiation.
The spinal cord with a 5 mm margin, contralateral lung, heart, liver, oesophagus
and kidneys are contoured as OAR volumes.
260
Dose solutions
■ Prophylactic radiotherapy
As electrons or 300 kV photons are used and the beam border marked onto the
patient’s skin, GTV, CTV and PTV are not defined. A 2 cm margin from the edge
of the scar being treated to the 50 per cent electron beam edge is recommended
to account for the shape of isodoses at depth. This will usually mean a 6 cm circle-
shaped field for simple drain sites with larger fields for patients who have had a
thoracotomy for a video-assisted thoracoscopic pleurodesis or a decortication.
As beams are usually defined on the simulator, GTV, CTV and PTV are not defined.
A 1 cm margin from the edge of any masses to the beam edge is recommended.
Dose solutions
■ Conformal – adjuvant radiotherapy
It is a major challenge to produce a conformal plan to treat the PTV uniformly to
54 Gy while maintaining critical organ tolerance (Fig. 21.2). Table 21.1 gives
suggested OAR tolerance doses when the dose prescribed to the PTV is given in
1.8 Gy fractions.
Figure 21.2 Target volume for adjuvant

radiotherapy after extrapleural
pneumonectomy. A DRR showing the
overlap of PTV2 (red) with the kidney, spinal
cord and heart. It is impossible to treat this
volume to 54 Gy with conventional photon
fields while keeping to critical organ
tolerance doses.
Dose to the PTV2 can be compromised to achieve target doses to the spinal
cord, contralateral lung, contralateral kidney and liver. The minimum dose to
PTV2 should be 45 Gy and the PTV1 should receive the prescribed dose of 54 Gy.
The large circumferential CTV and multiple OAR mean the best conformal
solution is a pair of anterior and posterior photon beams with MLC shielding to
the liver, kidneys, heart and spinal cord added during the treatment course
depending on their tolerance. The shielded areas within the target volume are then
treated with electrons matched as well as possible to the photon fields.
261
MESOTHELIOMA
Table 21.1 Suggested tolerance doses for OAR in adjuvant radiotherapy for
mesothelioma*
OAR Target
Spinal cord 5 mm 50 Gy
Contralateral lung V20
15 per cent. Some centres also recommend
constraints on mean lung dose (e.g. MLD 9.5 Gy)
and V5 (e.g. V5 60 per cent)
Contralateral kidney D80 per cent 15 Gy
Ipsilateral kidney As low as possible
Heart D70 per cent 45 Gy and Dmax 60 Gy. These
limits may be exceeded if residual disease is
present close to the heart
Liver Mean dose 35 Gy
Oesophagus Less than 12 cm length in treated volume
*Taken from the MARS trial with permission of Dr Senan.
■ Complex – adjuvant radiotherapy

In view of the difficulties of producing a conformal plan that adequately treats the
PTV while achieving the dose constraints in Table 21.1, IMRT has been used in
this setting. A potential disadvantage of IMRT is the increased dose to the
remaining lung with small single centre series reporting death rates from
pneumonitis as high as 50 per cent. To reduce this risk, the gantry angles for
IMRT should be chosen to minimise exit through the contralateral lung.
The IMRT beams can also be restricted so that the superior part of the PTV is
treated with just a few beams thus avoiding exit doses to the contralateral lung and
reducing pulmonary toxicity. The inferior portion near the heart or liver is treated
with all beams to maximise the advantage of highly conformal dose in this part of
the volume.
Another possible solution is to use combined photon IMRT and electron beams –
using the electrons to produce a conformal dose at depth to the more superficial
part of the PTV and IMRT to produce conformality in other planes.
■ Conventional – prophylactic and palliative

radiotherapy
In prophylactic treatment electrons are preferred, with the energy chosen to
treat from the skin surface down to the pleura within the 90 per cent isodose.
The depth of the pleura is best measured on diagnostic CT scans but can be
estimated clinically. In practice, 9–12 MeV electrons with 5 mm tissue equivalent
bolus are frequently used. While standard cut-outs are used for drain sites, a
patient-specific cut-out is usually required for larger scars that follow more
complex procedures. If electrons are not available, 300 kV photons can be used,
but this technique will not produce adequate coverage of the tract close to the
pleural surface.
262
Verification
Palliative radiotherapy for painful masses can be achieved with a direct electron
beam or with parallel opposed photon beams. Photons are often more appropriate
given the depth of painful lesions, particularly when the mass extends around the
curved chest wall.
Dose-fractionation
■ Adjuvant
PTV1
The plan chosen must deliver a minimum dose of 45 Gy to PTV2.
Prophylactic
21 Gy in 3 daily fractions of 7 Gy given in 3 days.
Palliative
8 Gy in 1 fraction.

Cardiac function (ejection fraction must be 40 per cent) and contralateral renal
function (serum creatinine or glomerular filtration rate [GFR] estimation and
evaluation of differential renal function) should be assessed before treatment.
A dietetic assessment before radiotherapy is desirable to initiate high calorie
supplements if necessary. Patients should be advised about good skin care (gentle
washing with simple soap, and application of aqueous cream) and management of
oesophagitis (see Chapter 19). Codeine linctus or oral opiates may help alleviate
cough. 5-HT antagonists may be required to prevent emesis.
A careful discussion with the patient about the potential benefits and possible
late effects of radiotherapy is essential. Patients are treated daily with no
compensation for missed days.
■ Palliative/prophylactic radiotherapy
Patients should be instructed in skin care – the use of simple soaps and aqueous
cream to keep the skin moisturised. Bright erythema is usual when 21 Gy in
3 fractions over 3 days is used, and some long-term pigmentation of the skin is
common.
Verification
Treatment is verified using the protocol described for lung cancer in Chapter 19.
263
MESOTHELIOMA
key trials
Boutin C, Rey F, Viallat JR (1995) Prevention of malignant seeding after invasive
diagnostic procedures in patients with pleural mesothelioma. Chest 108: 754–8.
MARS (Mesothelioma and Radical Surgery) trial. Three Cycles of Chemotherapy
Followed Either by EPP and Radiotherapy or by No Surgery or Radiotherapy.
ICR, Scientific Principal Investigator Prof Julian Peto. Closed 2008.
O’Rourke N, Garcia JC, Paul J et al. (2007) A randomised controlled trial of
intervention site radiotherapy in malignant pleural mesothelioma. Radiother
Oncol 84: 18–22.
SAKK 17/04 study. Neoadjuvant Chemotherapy and Extrapleural
Pneumonectomy of Malignant Pleural Mesothelioma With or Without
Hemithoracic Radiotherapy. A randomised multicentre phase II trial. Swiss trial
aiming to recruit over 150 patients.
Information sources
Allen AA, Schofield D, Hacker F et al. (2007) Restricted field IMRT dramatically enhances IMRT
planning for mesothelioma. Int J Radiation Oncology Biol Phys 69: 1587–92.
Waite K, Gilligan D (2007) The role of radiotherapy in the treatment of malignant pleural
mesothelioma. Clinical Oncology 19: 182–7.
264
Breast 22

■ Breast radiotherapy
Adjuvant radiotherapy given following surgery for primary carcinoma of the breast
has been shown to reduce the incidence of locoregional recurrence from 30 per cent
to 10.5 per cent at 20 years and breast cancer deaths by 5.4 per cent at 20 years.
Radiotherapy is standard treatment after complete local excision of ductal
carcinoma in situ (DCIS) and current trials are evaluating its role in ‘low risk’
patients compared with surgery alone.
Clinical T1, T2 less than 3 cm, N0 invasive breast cancers are treated by wide local
excision (WLE) followed by radiotherapy with comparable local control rates to
mastectomy, both combined with axillary surgery. The tumour site, size, histological
type, grade and extent of in situ disease, as well as the size of the breast, all influence
choice of treatment, as does consideration of the expected cosmetic result and
patient preference. Radiotherapy is indicated for all patients after conservative
surgery. As yet no ‘low risk’ group has been identified where surgery alone gives
adequate local control, but ongoing trials are addressing this issue. Contraindications
to conservative surgery include multifocal breast tumours, extensive DCIS, central
tumours in a small breast and incomplete excision. Significant pre-existing cardiac or
lung disease, scleroderma and limited shoulder mobility may prevent the use of
radiotherapy. Patients with operable tumours which are 3–4 cm or more in diameter
have a higher local recurrence rate with conservative surgery and radiotherapy, and
may be offered primary systemic therapy. Long-term results of this strategy, which
aims to downstage the tumour and avoid mastectomy in many patients, are awaited.
After primary chemotherapy, indications for locoregional radiotherapy are
determined by high risk factors at presentation and preoperative clinical staging
rather than postoperative pathological staging.
Primary lymphoma of the breast is commonly high grade and treated by primary
chemotherapy followed by local radiotherapy. For malignant phylloides tumours
and sarcomas of the breast, mastectomy is the treatment of choice.
Patients with bilateral tumours are treated according to the indications for each
individual tumour site.
■ Tumour bed boost radiotherapy

After complete excision, the decision to use a ‘boost’ dose to the tumour bed
should balance the individual’s risk of local recurrence (dependent on factors such
as age, tumour grade and size, lymphovascular invasion, margin status, endocrine
receptor status, and use of systemic therapy) against the risk of late effects (e.g. cardiac
265
BREAST
or lung damage because of shallow breast tissue over heart or ribs). EORTC
22881 showed that in patients younger than 40, a boost dose of 16 Gy resulted in
a greater reduction of local failure than for other age groups, but the relative risk
reduction was similar for all ages. All patients who have microscopic tumour
present at a resection margin, and where re-excision or mastectomy is declined,
should be considered for boost radiotherapy.
■ Post-mastectomy radiotherapy
More patients are having immediate breast reconstruction after mastectomy using
either microvascular techniques or an implant. Subsequent radiotherapy may lead to
a risk of late fibrosis and outcomes are being monitored. Post-mastectomy
radiotherapy is recommended for patients with T3, T4 tumours and those with
four or more positive axillary nodes who have a high risk of local recurrence (around
30 per cent), which is reduced by at least two-thirds. The Danish Breast Cancer
Trials Group (DBCG) reported a 9 per cent absolute increase in survival rate at
15 years after post-mastectomy locoregional radiotherapy in all groups of node
positive patients. Therefore patients with one to three nodes, and for example, young
patients and those with large T2–3 tumours, grade III, oestrogen receptor negative,
lymphovascular invasion or lobular histology with an estimated 10–20 per cent local
recurrence risk at 10 years, may also be considered for locoregional radiotherapy.
Local guidelines must be developed with a threshold chosen for the level of risk of
local recurrence that merits treatment until further trial data are available.
For inoperable T3 and T4 tumours, primary systemic therapy is given before
combined local treatment with surgery and locoregional radiotherapy, the
sequence depending on tumour regression, staging and prognostic factors.
■ Lymph node irradiation

If no axillary surgery has been performed and prognostic factors are good, axillary
radiotherapy may not be indicated.
Sentinel node biopsy allows selective axillary dissection for patients with a
positive node biopsy. Current EORTC trials are comparing axillary dissection with
axillary radiotherapy after positive sentinel node biopsy. Where sentinel node biopsy
is not available, lymph node irradiation is unnecessary if an axillary dissection up
to the lateral border of the pectoralis minor (level I) is negative.
Retrospective meta-analyses have shown that in high risk patients, axillary
radiotherapy is as effective as axillary surgery in preventing axillary recurrence.
If level I axillary nodes are involved, there is a 5 per cent risk of subsequent
supraclavicular fossa (SCF) recurrence, so irradiation may be given to levels II and III
axillary and SCF nodes. When four or more nodes, a single node 2 cm or level III
nodes are involved, the risk of SCF involvement is 15–20 per cent and
radiotherapy is indicated. After axillary dissection to level III with positive nodes,
axillary radiotherapy is associated with considerable morbidity and should be
avoided unless there is known residual disease, but SCF treatment is given. Nodal
radiotherapy is indicated for locally advanced disease after primary systemic
treatment, where surgery is not possible.
Radiotherapy to the internal mammary nodes is not recommended outside a
clinical trial because of the risk of cardiac toxicity and results of EORTC trial
22922/10925 are awaited.
266
Improved adjuvant systemic therapies such as anthracyclines, taxanes and
trastuzumab alter the risk versus benefit analysis of breast and lymph node
irradiation because of the risk of cardiac toxicity. Gene expression profiling of
primary breast cancer will be used to individualise indications for radiotherapy in
the future, based on predictions of risk of locoregional recurrence.
Radiotherapy has a major role in the palliation of locally advanced and fungating
breast tumours as well as in treating symptomatic metastases at sites such as bone,
brain, skin, lymph nodes, choroid and meninges.
Prognostic factors, PS and patient preference all affect the final decision made
by the multidisciplinary team.
Sequencing of multimodality treatment

In the adjuvant setting, chemotherapy is given before radiotherapy to reduce side
effects. This may mean that radiotherapy is delayed by 4–6 months and it is not yet
known whether this will compromise local control. The optimal sequencing of
chemo- and radiotherapy was the subject of the SECRAB trial where treatment was
randomised to a sequential versus synchronous schedule of chemo- and radiotherapy,
and these results are awaited. Primary chemotherapy for operable breast cancer is
followed by surgery and then subsequent radiotherapy. For locally advanced disease,
primary chemotherapy or endocrine therapy may be followed by surgery if technically
feasible or further downstaging using locoregional radiotherapy may be attempted,
reserving surgery for excision of residual disease if restaging is clear.

Breast cancer spreads locally by direct infiltration of the surrounding parenchyma
and may extend to underlying muscle and overlying skin, including the nipple.
A dense network of lymphatics in the skin may facilitate widespread cutaneous
permeation by tumour.
Lymphatics drain laterally to the axilla, medially to internal mammary nodes and
superiorly to the supraclavicular fossa (Fig. 22.1). Lymphatic vessels from the
whole breast drain to the internal mammary nodes, which communicate with the
contralateral chain superiorly. The internal mammary nodes lie on the internal
surface of the anterior chest wall closely applied to the internal mammary artery.
Although the anatomical drainage pattern is complex, involvement by tumour is
most commonly found in the axillary lymph nodes. These are divided into levels
I–III, which are used to guide surgical axillary node dissection. Levels are
described in relation to the pectoralis minor muscle. Level I nodes lie inferolateral
to its lateral border, level II posteriorly between its medial and lateral borders, and
level III medial to the medial border of pectoralis minor adjacent to the axillary
vein and first rib. Level III nodes are continuous with the supraclavicular nodes
medially and anteriorly and also with the infraclavicular nodes.
267
BREAST
Infraclavicular
nodes
Supraclavicular
nodes
Apical nodes
Pectoralis minor
Central nodes
Lateral nodes
Subscapular nodes
Pectoral nodes
Subareolar plexus
Internal
mammary nodes I II III
(a) (b)
Figure 22.1 (a) Diagram of lymphatic drainage of the breast. (b) Transverse CT scan of the
left axilla (patient with arms up) both showing position of levels I–III axillary lymph nodes.
AV, axillary vessels; Pm, pectoralis minor; PM, pectoralis major; IP, inter-pectoral nodes.

Ideally the radiation oncologist should examine the patient preoperatively. Breast
examination includes inspection for nipple or skin retraction, discharge, ulceration
or asymmetry, and palpation for size and site of the lump and fixation to adjacent
structures. Glandular drainage areas are also assessed and TNM staging recorded
on an accurate diagram. A photograph may be used to show the exact position of
the lesion. Mammography is performed to demonstrate the tumour and to detect
calcification, multifocal or in situ disease and bilateral involvement. Ultrasound is
used to measure the size of the lesion and to guide fine needle aspiration cytology
and/or core biopsy for histology. MRI can be used to exclude multifocal disease
prior to conservative surgery, particularly for large tumours in a radiographically
dense breast and for lobular cancers. MRI is also used to monitor response to
therapy where primary chemotherapy is used. Axillary node status may be assessed
using ultrasound and guided fine needle aspiration (FNA), or, where there is
palpable disease, with CT as part of a staging procedure for more advanced disease.
Examination of the surgical specimen should define the size, site and local extent
of the primary lesion with macroscopic margins and the number and position of
axillary nodes in the specimen. Histological review determines size, type of
tumour, grade, microscopic assessment of excision margins, lymphovascular invasion,
oestrogen, progesterone and HER2 receptor status, number of lymph nodes
involved and removed, and any extracapsular extension. Many oncoplastic
techniques place the surgical scar at a distance from the tumour bed and this
relation should be shown in an accurate operative diagram. Details of the level of
any axillary dissection, any residual disease and the placement of titanium clips or
gold seeds in the tumour bed should all be recorded. When inoperable primary
tumours remain palpable after systemic therapy, they can be assessed by palpation
and ultrasound, the dimensions marked on the skin, and a photograph taken.
All patients are discussed in multidisciplinary meetings, with review of imaging
and histopathology. If the radial or superficial margins are incomplete, re-excision
268
Data acquisition
is advised, although usually the deep margin has been cleared down to pectoral
fascia. Extensive DCIS is an indication for mastectomy, but minor focal margin
involvement by DCIS may be dealt with by re-excision or a tumour bed boost,
according to risk factors. Severe lung or cardiac disease, scleroderma, other
significant comorbidity or immobility that would contraindicate radiotherapy should
be identified so that mastectomy can be considered instead of conservative surgery.
Data acquisition
■ Immobilisation
The position of the patient must remain identical for localisation on a CT scanner
or simulator and during subsequent treatment. Most commonly the patient is
treated supine using an immobilisation device which secures both arms above the
head, as this lifts the breast superiorly, reducing cardiac doses, and also provides
symmetry if contralateral breast irradiation is required later. A headrest, elbow and
armrests, knee supports and a footboard provide stability. Care must be taken at
data acquisition to adapt all the supporting devices to the individual patient’s size
and shape to maximise comfort, and so aid reproducibility for subsequent
treatment. These recorded parameters, with a system of medial and lateral tattoos
and orthogonal laser lights, ensure alignment of the patient and consistency of
set-up (Fig. 22.2). Often an inclined plane is used with fixed angle positions. This
brings the chest wall parallel to the treatment couch and may reduce the need for
collimator angulation. The inclination is limited to a 10–15° angle for 70 cm, and
17.5–20° for larger 85 cm aperture CT scanners or simulator planning. Some
centres treat the patient lying flat on the couch top, without an incline, with a
similar immobilisation system but using collimator rotation.
Figure 22.2 Large-bore CT

scanner with patient
immobilised on system
using inclined plane, arms
up, with reference points
outlined with radio-opaque
material and aligned with
laser lights.
Patients with large or pendulous breasts treated supine require a breast support,
either with a thermoplastic shell, or breast cup which can be used to bring the lateral
and inferior part of the breast anteriorly away from the heart, lung and abdomen.
It is important to avoid displacing the breast too far superiorly over the neck.
Increased erythema due to loss of skin sparing by the shell may be offset by
269
BREAST
reduced severity of skin reaction in the inframammary fold. Alternatively, patients

with pendulous breasts can be treated in the prone position, which reduces mean lung
and cardiac doses and produces a more homogeneous dose distribution (Fig. 22.3).
This may improve cosmesis, but risks under dosage at the medial and lateral
borders of the PTV close to the chest wall, and should be avoided for primary
tumours in these situations. This technique cannot be combined with lymph node
irradiation but can be used to treat bilateral tumours.
Figure 22.3 CT dose distribution to right

breast with patient in the prone position
(6 MV, [gantry 296° and 107°, weighting
100 per cent lat/105 per cent med. 15.6
[W] 19 [L]). Courtesy of Greg Rattray,
Royal Brisbane and Women’s Hospital.
Whole breast
CT scanning
Where available, CT scanning has become standard for planning breast
radiotherapy. After palpation, the breast CTV and surgical breast scar are marked
with radio-opaque material before scanning. The upper and lower limits of the CT
scan are chosen so that CT data are acquired superiorly from above the shoulder to
include the neck and inferiorly to include all of the ipsilateral lung and 5 cm below
breast tissue. CT data of the whole breast and critical structures such as lung and
heart are needed for DVH calculations and to position lymph node beams. Slice
thickness should be sufficient (usually 2–3 mm but dependent on agreed local CT
protocols) to produce good quality images for target volume and OAR definition
and to create DRRs for accurate portal image comparison. Three reference tattoos
are placed on the central slice and in the medial and lateral positions on right and
left sides so that measurements can be made to subsequent beam centres.
The volumetric CT data are exported to the treatment planning system (TPS)
and a virtual simulation package can be used to define medial and lateral tangential
beams to encompass the breast CTV. These can be adapted by viewing the
posterior border of the CTV on all CT slices to ensure coverage of the tumour bed
as delineated by titanium clips or gold seeds on CT. CLD should be less than 2 cm
to avoid symptomatic pneumonitis. The heart, especially the left anterior
descending artery, should be excluded. Where this is impossible, maximum heart
distance (MHD) must be kept to less than 1 cm. CT also helps distinguish
glandular from adipose tissue, especially at the posterolateral aspect of the breast.
If the heart cannot be excluded completely from the target volume without
compromising the tumour bed CTV, localised cardiac shielding can be introduced
270
at the dose planning stage. The final virtual simulation is performed by a radiographer,
usually with an oncologist, and diagrams, DRRs and virtual simulation rendered
images are created before the dose plan is produced (Fig. 22.4). Alternatively, the
breast CTV and PTV can be outlined on each CT image with full 3D delineation
of the target volume. This is more time consuming but has advantages where more
advanced or inoperable tumours are visualised or when inverse planned IMRT is
used. The lung is contoured in its entirety for all 3D dose planning and DVHs.
(a) (b)
(c) (d)
Figure 22.4 Virtual simulation of breast with clips in tumour bed showing (a) axial scan
with adjustment of beam border anteriorly from skin markers to avoid heart, (b) sagittal,
(c) coronal and (d) rendered image of tangential beams.

■ CTV breast
For adjuvant radiotherapy after surgical excision of tumour there is no GTV and
the whole breast is the CTV. The aim is to treat all the glandular breast tissue
down to deep fascia, but not the underlying muscle, rib cage, overlying skin or
excision scar. A CTV-PTV margin is added to account for respiration, variations in
patient position, both intra- and inter-fractionally, breast swelling and set-up
uncertainties. Each department should measure its systematic and random errors
using a verification programme comparing simulator or DRR images with EPIs.
Most departments record standard deviations for systematic errors of around
271
BREAST
2–5 mm and an additional margin of 5 mm is reported as sufficient to account for

respiratory motion. This gives a CTV-PTV margin of 1 cm for a standard breast
target volume. When implanted clips are viewed in the tumour bed at CT, the
proposed CTV and PTV margins may need to be repositioned to ensure adequate
coverage of the tumour bed. During virtual simulation, the tangential beams can
be redesigned to encompass the CT-derived CTV and PTV and to reduce the
amount of lung and heart included in the treatment volume.
■ GTV breast
For inoperable tumours and following partial regression after primary systemic
therapy where surgery is still not feasible, the gross tumour is present. This can be
defined with the patient in the treatment position using palpation, CT or
ultrasound to design boost volumes.
■ CTV-reconstructed breast or chest wall

The target volume is the skin flaps and scar and any subcutaneous tissues down to
the deep fascia overlying muscles. In locally advanced breast cancer with skin
infiltration, skin is included in the target volume. The extreme ends of the surgical
scar may be excluded medially or laterally to reduce dose to underlying heart and
lung to tolerance limits. It is important to know the site of the primary tumour
within the breast at presentation and histological details of the surgical specimen
when adjusting beams in this way at virtual simulation.
■ Simulator
Conventionally, a simulator has been used to localise the breast with the
immobilisation system described above and the patient aligned with two laterals
and a sagittal laser light. Field borders rather than target volumes are defined by
palpating the entire breast and adding a 1.5 cm margin which includes penumbra.
The superior border covers as much of the breast as possible and lies at about the
level of the suprasternal notch medially, and just below the level of the abducted
arm laterally to allow beam entry. The inferior border lies 1.5 cm below the breast,
or more if the tumour bed is situated very inferiorly. The medial border is usually
in the midline and the lateral border 1.5 cm from the lateral border of the breast.
However, these borders should be modified, both to ensure good coverage of the
tumour bed and also to reduce heart (MHD 1 cm) and lung doses (CLD
2 cm), even if in some patients this means compromising coverage of peripheral
breast tissue sited away from the tumour bed (Fig. 22.5).
Using the simulator, an isocentric technique of medial and lateral tangential fields
is constructed. The anterior border of the field in free air should be at least 1.5 cm from
the skin surface to ensure a satisfactory dose distribution. The borders of the medial
and lateral fields are then marked on the skin. Two reference tattoos are made at
medial and lateral field centres over reproducible stable sites with a third one made
on the contralateral side of the body to align with lasers to prevent rotation. An
external contour of a transverse cross-section of the patient is taken in 2D through
the centre of the fields. Where a simulator CT is available, three CT outlines may be
taken at different levels for lung correction and superior–inferior dose compensation.
272
Figure 22.5 Simulator film of left medial tangential field
with CLD, MHD and clips in the tumour bed.
Beam divergence into the lung at the posterior border of the field can be reduced by
using either independent collimators to block the posterior half of the beam, or an
appropriate gantry angle to align the opposing posterior field borders.
■ Breast tumour bed

Target volume
Using CT data, the tumour bed can be visualised in 3D by using clips placed in
pairs (to identify migration) at surgery around the wall of the surgical cavity to
mark its posterior, lateral, medial, superior and inferior borders (Fig. 22.6). The
anterior border of the cavity should also be marked with clips if the surgical scar is
not located anterior to the tumour bed. The CTV (tumour bed) then includes the
tumour bed and any seroma and a 1.5 cm margin in all directions, editing 5 mm
from the skin and lung surfaces. The CTV may be increased in any radial
dimension if excision margins are less than 5 mm. The CTV-PTV margin is chosen
as 5 mm for tumour bed boost irradiation, making a total 2 cm margin around the
tumour bed. Both CT scanning and the use of clips have been shown to improve
accuracy of localisation of the volume, depth of the tumour bed and choice of
electron energy compared with clinical assessment alone.
Commonly, boost radiotherapy to the tumour bed is given with electron
therapy. To aid treatment delivery, rendered images can be produced to show the
position of the electron beam in relation to the surface scar.
When partial breast EBRT is given, a CTV-PTV margin of 1 cm is used for the
boost with an additional 5 mm added for penumbra, and treatment is delivered
using small (‘mini’) tangential beams.
■ Axillary and supraclavicular lymph nodes

CT scanning
A CT protocol is used similar to that described for whole breast. Axillary surgical
clips may aid localisation, but uninvolved nodes are not seen on CT. CT scanning
273
BREAST
(a) (b)
(c) (d)
Figure 22.6 (a) Axial CT scan with clips in the tumour bed (dark blue), boost CTV (cyan)
and whole breast PTV (red). (b) Sagittal view. (c) 3D image (lung in green). (d) Axial CT
scan with beams for whole breast EBRT (6 MV, gantry 221° and 47°, 9.5[W] 20[L]).
can be used to design a mono-isocentric technique for combined breast and lymph
node irradiation where a single isocentre is set up at depth on the match line of
the tangential and anterior nodal fields (Fig. 22.7).
Figure 22.7 Single isocentric technique

for EBRT to treat breast, axillary and
supraclavicular lymph nodes shown with
3D rendered CT image.
Target volume
The lymphatic drainage to the axillary and supraclavicular nodes forms an irregular
volume with its upper border lying anteriorly in the supraclavicular fossa, and
extending more posteriorly at the lower border to include all groups of axillary
nodes (see Fig. 22.1, p. 268).
CT studies have shown that axillary nodes lie at a mean depth of 3–5 cm and are
anterior to the mid-axillary line. Supraclavicular nodes lie at a mean depth of 4 cm.
274
Dose solutions
Internal mammary lymph nodes lie 2–4 cm lateral and deep to the midline in the first
three intercostal spaces. CT scanning can be used to locate the internal mammary
arteries which are closely applied to the nodes to help delineate the target volume.
Studies show that level I axillary nodes may not be routinely included in the standard
breast CTV and great care must be used to delineate these nodes marked with
surgical clips in the axillary tail of the breast CTV when treatment is indicated.
The irregular target volume of the breast or chest wall and regional lymph nodes
makes it technically difficult to deliver an equal and adequate dose to all areas and
to spare the lungs, heart, brachial plexus and spinal cord.
■ Simulator
Immobilisation, patient positioning and alignment are as described for breast
radiotherapy. An anterior field is used to include level II and III axillary and
supraclavicular nodes in the target volume. The medial border is placed 1 cm lateral
to the midline or at the midline with a 10° gantry angle away from the larynx and
spinal cord. The lateral border lies at the outer edge of the head of the humerus.
The superior border extends at least 3 cm above the medial end of the clavicle, but
laterally leaves a 1–2 cm margin of skin clear superiorly to avoid excessive skin
reaction. Using a mono-isocentric technique to treat breast and lymph nodes, the
inferior border is on line with the superior border of the tangential fields through
the match line with the isocentre at depth. Shielding of the acromioclavicular joint
and humeral head is important to avoid fibrosis and maintain shoulder mobility.
Shielding to the apex of the lung should be applied with care as it may shield level
II and/or III nodes which may be part of the target volume. Where level III nodes
have been removed, an anterior field to the supraclavicular fossa nodes only is
used, with the lateral border altered to lie at the coracoid process (see Fig. 8.7,
p. 104). Placement of surgical clips may mark the level II and III axillary lymph
node areas and should be used to design the nodal field borders.
Dose solutions
■ Breast and reconstructed breast
For most patients, 6 MV (range 4–8 MV) photons are chosen as optimal.
However, with increased breast volume and separation, higher energies
(commonly 10 MV) may produce better homogeneity. Because of the increased
skin sparing of higher energy beams, care should be taken to check that superficial
cavity wall margins and scars of reconstructed breasts receive adequate dose.
■ Conformal or complex
Using virtual simulation, beams have been optimised and CT data are used to correct
for lung density. A significant number of plans fail to achieve a homogeneous 3D
dose distribution (–5 per cent, 7 per cent) when the 2D tangential technique is
calculated in 3D (Fig. 22.8). Forward planned 3D dose compensation can be
achieved using a variety of methods. A randomised clinical trial has shown that
patients treated with IMRT and improved 3D dose homogeneity have significantly
275
BREAST
better breast cosmesis (Fig. 22.9). Inverse planned dose solutions aim at
optimisation to a set of dose volume constraints and may improve homogeneity still
further. This is particularly important for the reconstructed breast.
Figure 22.8 Dose colour wash

for 3D conventional tangential
plan through isocentre, off axis
dose maximum 111.7 per cent.
Figure 22.9 Sagittal dose distributions of conventional breast radiotherapy (left)

compared with dose-compensated IMRT (right) with dose ranges.
The position of the left anterior descending coronary artery (LAD) can be seen
on CT to lie within the target volume for many patients having left sided breast
radiotherapy. Full dose to this segment of artery may be the cause of increased
cardiac mortality from left breast radiotherapy reported in the literature. Modern
planning techniques reduce dose to the heart and it is anticipated that in the future
this will translate into decreased cardiac mortality and increase in overall survival
with breast radiotherapy. With forward planned dose compensation, MLC leaves
can be used to shield the heart (Fig. 22.10) and left anterior descending coronary
artery for one or both beams, without shielding the tumour bed site which has
been marked with clips and is clearly seen in 3D with CT planning. Doses to the
contralateral breast may also be lower, reducing the risk of secondary malignancies.
276
Dose solutions
Figure 22.10 Sagittal DRR with
segmented fields for dose-compensated
IMRT with cardiac shielding. Clips seen in
the tumour bed and axilla (within shielding).
Respiratory motion may affect the dosimetry of dynamic MLC/IMRT

techniques and hence gated therapy or ABC devices to suspend respiration may
have advantages in this situation.
■ Conventional
A 2D outline with centres of field borders marked is used to prepare a dose
distribution with opposing medial and lateral tangential fields and wedges used as
missing tissue compensators. The presence of lung tissue increases dose to the
medial and lateral aspects of the breast and although the amount varies, it is
important to incorporate lung corrections. An estimation of lung tissue is marked
on the outline from the simulator film and a correction factor (range 0.2–0.3) is
applied and a dose solution produced, aiming at a homogeneous dose distribution
on a single slice of –5 per cent, 7 per cent (ICRU50). This does not give
information at superior or inferior levels of the target volume where dose
inhomogeneities of up to 10–15 per cent can occur, especially in large patients.
Ideally these patients should have at least three outlines taken through the centre,
superior and inferior levels of the volume using a simulator CT facility or camera
based outlining system. Dose distributions can then be produced at multiple levels
and tissue compensators used to improve homogeneity. Cardiac shielding using
blocks or MLC leaves can be used in both beams, although care must be taken
with inferior quadrant tumours where the tumour bed may overlie the heart. A risk
versus benefit analysis then has to be made, and the MHD reduced to less than
1 cm by altering the posterior field border or partial shielding if possible.
■ Chest wall
Conventional planning uses opposing tangential fields but dosimetry is rarely
optimal because of the thin target volume of the chest wall surrounded by air and
lung. Skin doses cannot be calculated or measured accurately and the role of bolus
to the skin remains controversial. Selective use of bolus in high risk disease after
277
BREAST
excision of local recurrence or extensive lymphovascular invasion may be

considered, usually for the first half of the treatment so that it can be removed if
the skin reaction is excessive.
Electron fields have the advantage of avoiding the lung and heart, but CT or
ultrasound should be used to measure thickness of the chest wall which may vary
throughout its volume, making choice of electron energy difficult. If the chest wall
is very convex in shape, standoff may occur at the medial and lateral field edges
with reduction of dose at these sites. Electrons to the chest wall may be combined
with photons to the axilla and supraclavicular nodes, as used in the Danish Breast
Cancer Group studies.
Where immediate breast reconstruction has taken place, conformal or IMRT
techniques should be used to optimise homogeneity of dose and bolus should be
avoided if possible to maintain good cosmesis. Higher energies (e.g.10 MV) should
be avoided because of the risk of low skin dose due to increased skin sparing.
■ Tumour bed
Electron beams are commonly used for tumour bed boost irradiation. The target
volume should be delineated and is usually 5–8 cm in diameter requiring an
electron applicator of 7–10 cm to allow for lateral penumbra. The electron energy
is chosen using CT, simulator-CT or ultrasound to measure depth of the target
volume, which should be encompassed by the 90 per cent isodose (ICRU71).
Electrons of 4–15 MeV may be required but exit doses to the heart should be
avoided. For larger volumes or where gross tumour is present, small tangential
beams with CT planning or interstitial brachytherapy may be preferable.
■ Axillary and supraclavicular lymph node irradiation

A single anterior beam alone is recommended for adjuvant radiotherapy to
supraclavicular and axillary lymph nodes. For advanced palpable axillary disease,
extensive extranodal involvement or residual axillary disease, an additional
posterior axillary beam may be needed to give adequate tumour dose. When the
axillary separation exceeds 15 cm, the MPD to the axilla for a single anterior beam
falls below 80 per cent for 6 MV photons. An adequate MPD to the axilla can be
achieved using a posterior axillary beam every day and weighted according to the
separation in the axilla (e.g. for 16–18 cm, 1:10 weighting of posterior: anterior
beam applied doses). However, the dose to Dmax increases for larger separations to
110 per cent and care must be taken to stay within the tolerance of the brachial
plexus situated at 2–3 cm depth. A dose distribution must be produced for each
patient when this technique is used. Placement of the posterior axillary beam is
difficult and should be by CT and/or clinical palpation for macroscopic tumour
and the use of surgical clips marking residual or extranodal extension of disease.
3D conformal treatment volumes may prove optimal.
■ Internal mammary node irradiation

Megavoltage anterior beams are no longer used to treat internal mammary lymph
nodes because of the exit dose to the heart. For medial quadrant disease, the
tumour bed may lie so close to the internal mammary nodes that it is impossible
to treat both target volumes homogeneously. Treatment may then have to be
278
Dose solutions
given to the primary tumour alone, by moving the tangential beam further across
the midline on to the contralateral side. Studies show that standard fields do not
encompass internal mammary nodes consistently and often overtreat normal
tissues. CT planning is therefore mandatory for internal mammary node
irradiation. Electron or combined electron/photon beams can be used to treat
internal mammary nodes as in the EORTC 22922/10925 trial protocol.
Alternatively, wide tangential fields with cardiac and lung shielding may be used,
as in the NCIC CTG MA20 trial. Care must be taken to ensure homogeneity of
dose to the primary tumour bed and a match must be made of the internal
mammary node fields to adjacent tangential breast fields.
■ Combined breast/chest wall and nodal irradiation

The inferior border of the nodal beam has to be matched to the superior border
of the tangential beams, to avoid underdosage or overdosage. This can be achieved
by half beam blocking the inferior border of the nodal beam and rotating the
collimator and couch to eliminate the divergence of the superior border of the
tangential beams at the match line.
A technique with a single isocentre at depth on the match plane uses asymmetric
collimation, but restricts the maximum wedged length of the breast tangential
beams. However, it is the preferred technique as it avoids couch and collimator
rotations with risk of collisions and errors and reduces treatment time. When nodal
irradiation is required for relapse after breast radiotherapy, a gap can be left
between fields to allow for divergence of the superior tangential beams.
■ Bilateral breast irradiation

When bilateral breast irradiation is indicated, both arms are immobilised above the
head as illustrated in Fig. 22.2 (p. 269). An appropriate gap of 1–1.5 cm should
be left in the midline between the tangential fields to avoid overlap.
When radiotherapy is later required for a primary tumour in the contralateral
breast, it is important to use the same immobilisation device as for the first tumour
treatment to keep the patient position constant. Previous radiotherapy should be
reconstructed to avoid overlap of treatment, especially in the midline and
supraclavicular regions, and dose to the underlying spinal cord should be estimated.
■ Partial breast irradiation (within clinical trials)

Studies show that around 85 per cent of local recurrences after surgery and
radiotherapy for operable breast cancer occur in the same quadrant as the primary
tumour. A risk-adapted strategy for breast radiotherapy has led to the investigation
of partial breast irradiation (PBI) treating the volume around the primary tumour
site only. Techniques include external beam radiotherapy with or without
concomitant IMRT boost, low and high dose brachytherapy, balloon catheter
brachytherapy (Ammonite device), kV X-ray applicators (Intrabeam) and
intraoperative electron therapy. Clinical trials are being carried out to test PBI
using these different modalities. Protocols defining the tumour bed, CTV and
PTV for PBI use radio-opaque markers and a careful quality assurance programme
which will ensure accurate treatment delivery. PBI should at present be restricted
to treating patients within the setting of a clinical trial.
279
BREAST
Dose-fractionation
■ Breast, reconstructed breast and chest wall
All these regimens have been tested in randomised trials with good results. The
same fractionation regimens can be used treat to DCIS, as there is no evidence that
it has a different radiosensitivity from invasive disease.
■ Breast boost irradiation
Tumour bed
Doses are prescribed using electron therapy to Dmax or using photons to the
ICRU point at the centre of the target volume; 16 Gy in 8 daily fractions has been
shown in the EORTC trial 22881 to reduce local failure by a factor of 2 compared
with no boost. 10 Gy in 5 daily fractions may be used in patients with lower risk
of local recurrence.
Incomplete excision or residual primary tumour
20–26 Gy in 10–13 daily fractions given in 2–21⁄2 weeks.
Interstitial implantation may also be considered for tumour bed boost irradiation.
Lymph node irradiation
Doses are prescribed at Dmax (e.g. at 1.5 cm for 6 MV photons).
Patients with breast cancer often live many years with metastatic disease, especially
in bone. Care must be taken to check sites of previous irradiation and to match
fields carefully to avoid overdosage and unwanted toxicity.
8 Gy single fraction for most bone metastases for relief of pain.
20 Gy in 5 daily fractions of 4 Gy given in 1 week may be used for sites such as
cervical spine, meningeal disease, and nodal masses.
36 Gy in 6 fractions of 6 Gy once or twice weekly, given in 6 weeks for fungating
primary tumours, especially in frail patients.

Treatment delivery will vary according to available technology. Where manual
wedges, physical compensators or couch rotation are used, the overall time for
each treatment fraction is longer. IMRT with wedged tangential beams and
additional MLC shaped dose compensating segments has been shown to take very
little longer to deliver than conventional tangential fields.
280
Verification
Patients are instructed to avoid abrasion of the irradiated skin when washing and to
use simple soap. Aqueous cream is applied twice daily at least 2 h before or after
treatment to keep the skin moisturised. One per cent hydrocortisone cream may be
used to relieve the irritation of dry desquamation. If moist desquamation occurs,
treatment is temporarily stopped and Atrauman gauze with a pad or hydrogel sheet
or foam dressing applied until healing occurs. Tight fitting clothes should be avoided
as much as possible to reduce friction and abrasion of the skin. Loose cotton garments
are recommended. Gentle arm exercises started after surgery are continued.
Later side effects may include breast oedema, shrinkage, pain and tenderness, rib
fracture, skin telangiectasia, symptomatic lung fibrosis, cardiac morbidity or late
malignancy when radiotherapy is combined with chemotherapy. After nodal
radiotherapy, there is a risk of arm lymphoedema, shoulder stiffness or nerve
complications.
Verification
The immobilisation device, room laser lights, set-up instructions and rendered
images are all used to ensure an identical patient position and accurate treatment
delivery. Portal imaging should be undertaken using locally agreed evidence-based
imaging protocols. This typically consists of imaging the first three daily fractions
and then weekly checks, with images being compared with the CT-generated DRR
or simulator films and a 5 mm tolerance accepted in the CLD/isocentre
position. Consideration should be given to any change in soft tissue contour where
forward planned IMRT is used to ensure the delivery of a homogeneous dose
distribution. In vivo dosimetry using a diode or TLD measurement is carried out
on day 1 in all patients to ensure delivery of the planned dose to each
field/segment. To ensure true readings, consideration should be given to the
positioning of the diode/TLD for the smaller dose compensation segments used
in forward planned techniques. IGRT can be used to match the position of
titanium clips in the tumour bed for pretreatment verification.
key trials
■ EORTC 22922/10925: Breast internal mammary-medial SCF node

irradiation for selected high risk group patients. http://astro2005.abstractsnet.
com/handouts/000156_ASTRO_Meeting__September_2005.pdf (accessed
5 December 2008).
■ EORTC 22881/10882: Boost versus No Boost tumour bed RT. See Antonini
et al. below (www.ncbi.nlm.nih.gov/pubmed/17126434).
■ IMPORT LOW and HIGH (Intensity Modulated and Partial Organ
RadioTherapy). Whole v partial breast IMRT in two risk groups.
www.icr.ac.uk/research/research_sections/clinical_trials/trials_by_disease/
breast_cancer/index.shtml (accessed 5 December 2008).
■ SECRAB: Sequencing of chemotherapy and radiotherapy in adjuvant breast
cancer. See Bowden et al. below.
■ START Trials A and B: Fractionation study of breast RT (see below).
■ SUPREMO (Selective Use of Post Operative Radiotherapy after Mastectomy):
mastectomy chest wall RT for intermediate risk patients. www.
supremo-trial.com/ (accessed 5 December 2008).
281
BREAST
Information sources
Adlard JW, Bundred NJ (2006) Radiotherapy for ductal carcinoma in situ. Clin Oncol 18: 179–84.
Antonini N, Jones H, Horiot JC et al. (2007) Effect of age and radiation dose on local control after
breast conserving treatment: EORTC trial 22881-10882. Radiother Oncol 82: 265–71.
Bowden SJ, Fernando IN, Burton A (2006) Delaying radiotherapy for the delivery of adjuvant
chemotherapy in the combined modality treatment of early breast cancer: is it disadvantageous
and could combined treatment be the answer? Clin Oncol 18: 247–56.
Dobbs HJ, Greener AJ, Driver D (2003) Geometric uncertainties in radiotherapy of breast cancer. In:
Geometric Uncertainties in Radiotherapy: Defining the Planning Target Volume. BIR Report, BIR
Publications Dept, London, UK.
Donovan E, Bleakley N, Denholm E et al. (2007) On behalf of the Breast Technology Group (UK)
Randomised trial of standard 2D radiotherapy versus intensity modulated radiotherapy in
patients prescribed breast radiotherapy. Radiother Oncol 82: 254–64.
Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) (2005) Effects of radiotherapy and of
differences in the extent of surgery for early breast cancer on local recurrence and 15 year
survival: an overview of the randomised trials. Lancet 366: 2087–106.
Goodman RL, Grann A, Saracco P et al. (2001) The relationship between radiation fields and
regional lymph nodes in carcinoma of the breast. Int J Radiat Oncol Biol Phys 50: 99–105.
Hurkmans CW, Borger JH, Pieters BR et al. (2001) Variability in target volume delineation on CT
scans of the breast. Int J Radiat Oncol Biol Phys 50: 1366–72.
Lievens Y, Poortmans P, Van den Bogaert W (2001) A glance on quality assurance in EORTC study
22922 evaluating techniques for internal mammary and supraclavicular lymph node chain
irradiation in breast cancer. Radiother Oncol 60: 257–65.
Overgaard M, Nielsen HM, Overgaard J (2007) Is the benefit of postmastectomy irradiation limited to
patients with four or more positive nodes, as recommended in international consensus reports?
A subgroup analysis of the DBCG 82 b and c randomised trials. Radiother Oncol 82: 247–53.
Owen JR, Ashton A, Bliss JM et al. (2006) Effect of radiotherapy fraction size on tumour control in
patients with early breast cancer after local excision: long term results of a randomised trial.
Lancet Oncol 7: 467–71.
Ragaz J, Olivotto IA, Spinelli JJ et al. (2005) Loco regional radiation therapy in patients with high risk
breast cancer receiving adjuvant chemotherapy: 20 year results of the British Columbia
randomised trial. J Natl Cancer Inst 97: 116–26.
Recht A, Edge SB, Solin LJ et al. (2001) Post mastectomy radiotherapy: clinical practice guidelines
of the American Society of Clinical Oncology. J Clin Oncol 19: 1539–69.
Special Issue: Radiotherapy of breast cancer (2007). Radiother Oncol 82: 243–357.
The START Trialists’ Group (2008) The UK Standardisation of Breast Radiotherapy (START) Trial B
of radiotherapy hypofractionation for treatment of early breast cancer: a randomised trial.
Lancet 371: 1098–107.
The START Trialists’ Group (2008) The UK Standardisation of Breast Radiotherapy (START) Trial A
of radiotherapy hypofractionation for treatment of early breast cancer: a randomised trial.
Lancet Oncol 9: 331–41.
Whelan T, Mackenzie R, Julian J et al. (2002) Randomised trial of breast irradiation schedules after
lumpectomy for women with lymph node-negative breast cancer. J Natl Cancer Inst 94:
1143–50.
282
Lymphomas 23

■ Hodgkin lymphoma
Results of treatment for Hodgkin lymphoma have improved greatly over the past
three decades, with 80 per cent of patients now achieving long-term cure.
Favourable prognostic factors include stage I disease, lymphocyte predominant or
nodular sclerosing histology, age 40, low erythrocyte sedimentation rate (ESR)
and no B symptoms. Higher stage, age and ESR, involvement of the mediastinum,
tumour bulk, increasing number of nodes involved, male gender, anaemia and
extranodal sites are all indicators of a worse prognosis.
For patients with early and intermediate stage disease, original treatment
regimens were based on primary extended field radiotherapy, often combined early
or at relapse with extensive and prolonged chemotherapy. Using these strategies,
15–20-year survival figures were high but mortality from second malignancies
and cardiac deaths exceeded that from the disease itself. However, results from
more recent randomised trials have shown that the use of shorter periods of
chemotherapy combined with involved field radiotherapy (IFRT) gives equivalent
survival rates to that of extended field radiotherapy with reduction of late toxicity.
Radiotherapy remains a key part of combined modality treatment to ensure
locoregional control.
For early stage I and II disease, treatment with two to four cycles of
chemotherapy, depending on prognostic factors, followed by local radiotherapy,
gives 5-year survival rates of 90–95 per cent. Studies have shown that after
chemotherapy alone, sites of relapse are predominantly in initially involved nodes.
Current data using advanced imaging techniques suggest that involved lymph node
radiotherapy (INRT) may be as effective as IFRT, and would further reduce doses
to normal tissues such as the lung, heart and breast. Smaller target volumes and
lower radiation doses are improving the therapeutic index for patients with early
stage disease. Confirmation is awaited from trials as to whether radiotherapy can be
omitted in patients who have a negative PET scan after short course chemotherapy.
For more advanced stages, intensive chemotherapy is the mainstay of treatment,
combined with radiotherapy for selected patients who do not achieve complete
remission. Relapsed Hodgkin lymphoma is treated with high dose chemotherapy
and autologous stem-cell transplantation, when possible, with 50 per cent long-
term survival rates. For other patients who relapse, combined chemotherapy and
radiotherapy can produce worthwhile responses. Radiotherapy may also be used to
palliate symptoms from lymph node or other tumour masses that are causing pain,
obstruction or bleeding.
283
LYMPHOMAS
■ Non-Hodgkin lymphoma
Low grade follicular lymphoma presents as localised stage I or II disease in
20–30 per cent of cases. INRT alone gives 50–60 per cent 15-year survival rates
and these results have not yet been improved by the addition of chemotherapy.
Stage IE and IIE MALT lymphomas of the stomach, which are Helicobacter
pylori negative or do not respond to antibiotic therapy, are treated with local
radiotherapy, with very high local control and cure rates. Other MALT sites
include thyroid, salivary glands, skin, breast, lung and orbit and, when
presenting as stage IE or IIE disease, are treated with local radiotherapy, with
70–80 per cent 10-year survival rates.
Patients with stage III and IV low grade follicular lymphoma may be
asymptomatic, so treatment with chemotherapy may be deferred until disease
progression. Response rates with conventional chemotherapy are around 60 per
cent but the disease is not curable, and more aggressive regimens with stem cell
support are considered at relapse for fit patients. The monoclonal antibody
rituximab, given in combination with chemotherapy or as maintenance after initial
treatment, has improved response rates and progression-free survival.
Diffuse aggressive B or T cell lymphomas are treated with initial chemotherapy,
combined if they are CD20 positive with monoclonal antibody therapy. This is
followed in early stage disease by radiotherapy localised to initial bulky sites of
involvement. Five-year survival rates for stage I and II disease are 80–90 per cent.
Patients with adverse prognostic factors such as increasing stage, high lactate
dehydrogenase, B symptoms and extensive or bulky disease are treated with more
intensive and prolonged chemotherapy. Radiotherapy is used for those with
residual nodal masses.
Extranodal lymphomas are usually diffuse large B cell lymphomas in type and
occur at a wide variety of sites such as the CNS including extradural tissue, head
and neck, breast, stomach, skin and bone. They are commonly treated with initial
chemotherapy followed by localised radiotherapy to the initial and/or bulky sites
of disease. When treating CNS lymphoma, care must be taken with whole brain
irradiation after chemotherapy and high-dose methotrexate. Radiotherapy may
then be omitted in older patients (60) because of the increased risk of
neurological toxicity, particularly dementia.
Splenic irradiation is rarely used now as involvement of the spleen usually
represents systemic disease, best treated by chemotherapy. Where hypersplenism
due to myelofibrosis cannot be treated by surgical splenectomy, radiotherapy may
be given.
Radiotherapy is also used sometimes in the following situations:
■ prophylactic cranial irradiation in lymphoblastic lymphoma and acute
lymphocytic leukaemia (Chapter 18)
■ total body irradiation (TBI) before allogeneic and selected autologous
transplants (Chapter 38)
■ post-transplant salvage for residual PET visualised disease
■ immune modulation with single fraction (2 Gy) for reduced intensity
conditioning
■ irradiation for CSF/CNS involvement (Chapter 18)
■ prophylactic testicular irradiation in testicular lymphoma (Chapter 38).
284
Skin lymphomas and their management are discussed in Chapter 7. Palliative
local radiotherapy is useful for relieving symptoms in patients with large
uncontrolled nodal masses or in patients who are unfit for chemotherapy.

The WHO histological classification separates nodular lymphocyte predominant
Hodgkin lymphoma from other classical Rye subtypes and divides non-Hodgkin
lymphoma into B and T/NK cell types.
Lymphomas can affect any lymph node site in the body as well as extranodal
sites such as Waldeyer’s ring, thyroid, gastrointestinal tract, CNS, testes, breast,
bone and skin. The staging system used for lymphomas is the Ann Arbor
classification, with the Cotswolds’ modification for Hodgkin lymphoma. This is
based on number of anatomical sites of involved lymph nodes above or below the
diaphragm, splenic or extranodal disease, B symptoms and the addition of bulky
disease for Hodgkin lymphoma in the Cotswolds’ modification. Spread occurs to
contiguous lymph nodes and by the bloodstream to spleen, liver, lungs, bone
marrow or skin, with extranodal sites being much more common with non-
Hodgkin lymphoma. Sites such as paranasal sinuses, testicular and vitreous
lymphomas have a predilection for CNS spread.

Clinical examination should include all nodal sites with measurement of any
palpable involved lymph nodes by the radiation oncologist at diagnosis. This is
especially important for INRT and should be done before any chemotherapy is
given. Excision biopsy of an entire lymph node is required for accurate histological
diagnosis with fresh and fixed material. Image guided needle core biopsies are used
for inaccessible lymphadenopathy to avoid mediastinoscopy and laparotomy
whenever possible. Chest X-ray identifies gross mediastinal lymphadenopathy. CT
scan of the chest, abdomen, pelvis and neck is needed for full staging of the
primary site as well as to assess other lymph node sites and organ spread. MRI is
indicated for CNS tumours and extradural lymphomas, as well as for primary
lymphoma of bone. Radioisotope bone scans or PET scans are performed for
primary lymphoma of bone to exclude other bone lesions, as multifocal disease is
common. Bone marrow aspiration and trephine is performed in all patients with
non-Hodgkin lymphoma and selected patients with Hodgkin lymphoma.
[18F]2-fluoro-2-deoxy-D-glucose (FDG) PET-CT scanning has a sensitivity of
greater than 90 per cent for the detection of most types of non-Hodgkin
lymphoma, except MALT and small lymphocytic lymphomas where sensitivity is
less. PET-CT has the highest rate of disease detection of any imaging modality for
Hodgkin lymphoma and should be performed whenever possible in the
radiotherapy treatment position so that scans can be used for subsequent image
registration for planning radiotherapy. PET scanning changes initial staging in
10–20 per cent of patients with Hodgkin lymphoma, and if it is repeated after two
or three cycles of chemotherapy, response can be used to predict overall long-term
285
LYMPHOMAS
Figure 23.1 PET scan of

patient with stage IIA (bilateral
lower neck and mediastinum)
Hodgkin lymphoma pre- and
post-chemotherapy.
survival (Fig. 23.1). PET is more sensitive at detecting residual disease than CT,
which cannot differentiate post-treatment fibrotic mediastinal masses from active
disease. However, it should be used in combination with CT for target volume
delineation as PET – avid areas have been shown to represent only 25 per cent of
the total tumour mass seen on CT.
Data acquisition
To minimise toxicity to normal organs, and optimise dose distribution for the target
volume, 3D CT planning is used. Patients are scanned lying supine with appropriate
immobilisation devices for the anatomical site to be treated, with a system of skin
tattoos. For axillary radiotherapy, the arm is abducted and supported using an arm
pole or other restraint. For cervical nodal irradiation, a thermoplastic or vacuum
shell is used with reference marks on the shell for alignment with lasers.
As lymphomas occur at a wide variety of anatomical sites, general principles
should be applied and decisions made about the proposed treatment technique
before CT scanning to ensure that the most appropriate CT protocol is used. If
palpable masses are present, these are marked with radio-opaque material before
scanning with the patient in the treatment position. The scan should include the
whole volume of critical organs such as lung, liver, and kidneys for DVH
assessment. CT scans of 3 mm slice thickness are taken using intravenous contrast
to aid delineation of lymph nodes adjacent to vascular structures.
PET-CT and CT scans taken in the treatment position before chemotherapy are
co-registered with the post-chemotherapy CT planning scans for target volume
delineation. In addition, diagrams showing the initial lymph node measurements
at diagnosis are used.
286
Traditionally, wide or extended field radiotherapy treatments such as ‘Mantle’ and
‘Inverted Y’ were used as primary treatment for lymphomas with good cure rates
but late toxicity, particularly to the heart, lungs and breasts. With treatments now
combining chemotherapy and adjuvant radiotherapy, data have accumulated to
support the use of local radiotherapy instead of extended fields, to reduce toxicity
by sparing normal tissues.
There are three possible approaches to using smaller volume treatments. Since
lymphatic spread involves nodes adjacent to the primary site of disease first,
locoregional treatment that includes elective adjuvant irradiation of the first node
stations may be given (IFRT). The CTV includes the GTV, which is the primary
site, and adjacent nodes as CTV (a situation analogous to treatment of uninvolved
nodes in head and neck cancer). Alternatively, only the initial macroscopic volume
of disease (GTV), whether in lymph node (involved nodal INRT), or an organ
(sometimes called involved site ISRT) is treated with a GTV-CTV margin. Finally,
the residual GTV following chemotherapy, defined by PET-CT, could be the basis
of the target volume. In published reports the target volumes used may combine
aspects of these different conceptual approaches.
■ IFRT
When primary radiotherapy without chemotherapy is given (e.g. stage I follicular
non-Hodgkin lymphoma or nodular lymphocyte predominant Hodgkin
lymphoma), IFRT is used to include the uninvolved first node station in the CTV
as well as the macroscopically involved GTV (Fig. 23.2).
The CTV is designed using the initial volume of involved lymph nodes, i.e. the
initial GTV as shown on staging CT, MRI and FDG-PET-CT scans. These scans
need to be taken in the radiotherapy treatment position, whenever possible, for
accurate co-registration with subsequent CT planning scans which are taken after
chemotherapy. When primary radiotherapy is given, the GTV is present both
clinically and on imaging. After chemotherapy there may be a complete response
(PET and CT negative) with no GTV, or a partial response with a residual GTV.
The original GTV is outlined on the post-chemotherapy CT planning scan using
the co-registered initial CT and PET scans. The patient’s anatomy will have altered
with response to chemotherapy, and the principle is to cover the site of original
disease, but to reduce toxicity by excluding wherever possible normal tissues that
were displaced, but not involved by tumour. Anatomical planes act as boundaries
and the volume should not extend into adjacent bone, air or viscera.
For both IFRT and INRT of the mediastinum, the pre-chemotherapy volume is
used for the superior–inferior extension and the post-chemotherapy volume for the
axial to reduce dose to blood vessels, heart, coronary arteries and lung which may have
returned to their normal position (Fig. 23.3). Where there was definite tumour
infiltration into adjacent tissues at diagnosis, the whole volume should still be included.
The CTV for IFRT includes the clinically and radiologically involved lymph
nodes as imaged at presentation (GTV) with the adjacent lymphatics and
uninvolved first node station. CT outlining of lymph nodes uses the Ann Arbor
staging system and is aided by atlases such as those published by Gregoire for the
head and neck and Taylor for pelvic nodes.
287
LYMPHOMAS
(a) (b)
Figure 23.2 Diagrams of

opposing fields for IFRT.
(a) Unilateral cervical field for stage
IA high cervical node with shielding
to lung apex. (b) Axillary field with
(c) (d) shielding of humeral head.
(c) Neck and mediastinum with
lung shielding. (d) Groin with
shielding of testes and small
bowel.
A CTV-PTV margin is added to allow for organ motion which will vary for different
sites, for example it should be 5 mm for mediastinal nodes. A further 5–10 mm margin
is used for daily variations in set-up, the exact value determined according to individual
departmental protocols, making a total CTV to PTV margin of 10–15 mm.
■ INRT or ISRT
This target volume is based on initial macroscopic disease pre-chemotherapy (GTV),
rather than on lymph node region, and is defined using initial CT and PET scans
co-registered with subsequent CT planning scans. If the initially involved nodes
are no longer visible after a complete response, the CTV is the initial site of
involvement.
For neck, mediastinum and para-aortic nodes, a GTV-CTV margin of 20 mm in
the cranio-caudal and 10 mm in other dimensions is used. For hilar, SCF and
common iliac nodes, a GTV-CTV margin of 10 mm in the anteroposterior (AP)
and 20 mm in other directions is used. For axillary, external and internal iliac,
inguinal and femoral nodes, a GTV-CTV margin of 20 mm is used in all directions.
These variable margins are based on observation and designed to exclude initially
displaced normal structures which have returned to their usual position.
A CTV-PTV margin of 5–10 mm in 3D is added according to anatomical site
and departmental set-up error measurements.
288
(a) (b)
(c)
Figure 23.3 IFRT for a patient with stage IIA Hodgkin lymphoma (in complete remission
after chemotherapy) showing radiotherapy to the mediastinum, left SCF and left axilla.
(a) Coronal CT scan with target volumes. (b) 3D image. (c) Dose distribution using anterior
and posterior beams with MLC. Lungs green; heart yellow; spinal cord light yellow; (b) and
(c) are at different slice level to (a).
INRT may be used for adjuvant radiotherapy after chemotherapy (Fig. 23.4).
If only a partial remission is obtained, a residual GTV will be defined on
post-chemotherapy CT scans. This is used to create a boost CTV, which is the
GTV plus a 10–15 mm margin used to deliver extra irradiation to residual disease.
■ Extranodal sites
When primary radiotherapy is given for stage IE non-Hodgkin lymphoma, the GTV
is usually the whole of the involved organ such as stomach, orbit or breast with an
appropriate margin for microscopic disease to create the CTV. For a solid organ, the
volume is designed using the original GTV with a GTV-CTV margin of 15 mm in
the cranio-caudal dimension. In the AP and lateral dimensions the residual, not
initial, GTV are included with a 15 mm 3D margin to minimise normal tissue doses.
CTV-PTV margins vary according to the mobility of the organ, e.g. for stomach a
20 mm cranio-caudal margin may be used to allow for respiration (Fig. 23.5).
289
LYMPHOMAS
(a) (b)
Figure 23.4 INRT for patient with stage IA follicular non-Hodgkin lymphoma showing
treatment to unilateral right inguinal lymph nodes, showing (a) axial CT scan with target
volume and (b) conformal dose plan. Uterus yellow, left; adnexa green.
(a) (b)
(c) (d)
Figure 23.5 Patient with stage IE MALT non-Hodgkin lymphoma of stomach. (a) Axial CT
scan showing target volume. (b) 3D image. (c) Conformal plan. (d) DVH with liver (yellow),
kidneys (green) and spinal cord (light yellow).
There is no evidence yet to change the established practice of treating the whole
of Waldeyer’s ring, the parotid gland or whole brain after chemotherapy and large
margins may still be allowed (30 mm cranio-caudal) for bone lymphoma. For
splenic irradiation, CT scanning is used to localise the target volume (whole
spleen) and organs at risk which include the left kidney and stomach although
doses are below tolerance. No GTV-PTV margin is needed as even partial organ
radiotherapy is effective (Fig. 23.6).
Gross disease (GTV) is delineated with as small a margin as possible to minimise
side effects.
290
Dose-fractionation
Figure 23.6 DRR showing anterior BEV
with spleen outlined for splenic
radiotherapy showing relations to left
(green) and right (magenta) kidney.
Very occasionally extended field treatments may be used for palliation. Mantle
fields include all lymph node areas above the diaphragm and the inverted Y, infra-
diaphragmatic nodes. CT scans are performed with the patient immobilised in the
treatment position and palpable nodes marked with radio-opaque material.
Involved nodes may be contoured on each slice or virtual simulation used to design
beams to encompass disease. Shielding to the oral cavity, lungs, humeral heads,
spinal cord, breast, kidneys, bladder and bowel is obtained by appropriate MLC
configurations. A simple arrangement of anterior and posterior beams is used.
■ OAR
Normal organs such as lungs, heart, spinal cord, major blood vessels and breasts
should be outlined and PRVs created. DVH for critical organs are obtained and
tolerance restrictions applied to specific organs.
Dose solutions
Using 3D CT planning, the most appropriate conformal multi-beam plan can be
prepared. Less often anterior and posterior beams are used, sometimes unequally
weighted and shaped using MLC in 3D to spare normal tissues. Inhomogeneity
corrections are made to account for lung tissue in the mediastinum. IMRT may
have benefit at sites near the spinal cord but care must be taken to avoid increasing
dose to lungs, heart and breasts.
Dose-fractionation
■ Hodgkin lymphoma
Early stage, after complete response to chemotherapy
Advanced stage, with residual disease after chemotherapy
30 Gy in 15 daily fractions given in 3 weeks, with additional 6 Gy in 3 fractions
depending on bulk of disease.
Palliative treatment
291
LYMPHOMAS
Extended field 30 Gy in 15 daily fractions given in 3 weeks.

8 Gy given in a single fraction.
■ Non-Hodgkin lymphoma
Primary radiotherapy for stage I, IE, II, IIE low grade disease
24–30 Gy in 12–15 daily fractions given in 21⁄2–3 weeks
(or 4 Gy in 2 daily fractions).
Radiotherapy after chemotherapy for all other histologies

Palliative treatment
4 Gy in 2 daily fractions.
Splenic irradiation
10–12 Gy in fractions of 0.5–1.5 Gy given up to three times/week.
If blood counts are initially very low, treatment is started cautiously with low
doses given once a week with full blood count check before each treatment.

Laser lights, skin markers and immobilisation devices are used exactly as for CT
planning. Portal imaging or EPIs are used to verify the treatment and compare
with DRRs. In vivo dosimetry is performed at the first treatment. Where IGRT
is available, this may be useful to optimise daily localisation of the target volume.
Acute side effects depend on the particular anatomical site being treated. Skin
reactions are usually mild, unless a shell is used with loss of skin sparing. Brisker
reactions are seen in the axilla, groin and perineum. Aqueous or 1 per cent
hydrocortisone cream can be used. Temporary occipital epilation and dry mouth may
occur with upper cervical treatments, and sore throat, hoarseness and dysphagia with
lower cervical radiotherapy. Symptomatic remedies are given such as artificial saliva
and soluble analgesia. Groin and pelvic nodal treatment can cause acute cystitis and
diarrhoea. Bone marrow suppression may occur especially with large pelvic fields
after chemotherapy and blood counts are monitored for neutropenia and
thrombocytopenia. Cough due to mild pneumonitis may occur. Lhermitte’s
syndrome is rarely seen now that spinal cord shielding is used with conformal and
IMRT. Loss of fertility and early menopause may occur with pelvic treatment.
Extended mantle radiotherapy was used to irradiate axillae electively and
included much of the lateral part of the breasts. This led to an increased incidence
of second malignancies, particularly breast cancer in patients treated in young
adulthood, and lung cancer in smokers. Ten-year results have confirmed that the
use of involved field radiotherapy alone after chemotherapy with reduced dose to
breast, heart and lungs leads to fewer second malignancies. Chemotherapy is
associated with an increased incidence of leukaemia.
292
Information sources
key trials
EORTC H9-F. Chemotherapy IFRT 36 Gy v 20 Gy or no radiotherapy for
favourable HL. Available at http://groups.eortc.be/lymphoma – see H9 trial
HD 8 (GHSG). Comparison of IFRT with extended field RT after chemotherapy
for early stage Hodgkin lymphoma and HD 10 & 11 (GHSG). Comparison of two
versus four courses of ABVD IFRT 20 Gy v 30 Gy.
Klimm B, Diehl V, Pfistner B et al. (2005) Current treatment strategies of the
German Hodgkin Study Group (GHSG). Eur J Haematol 75(S66): 125–34.
Information sources
Aleman BM, Raemaekers JM, Tirelli U et al. (2003) Involved-field radiotherapy for advanced
Hodgkin’s lymphoma. N Engl J Med 348: 2396–406.
Bonnadonna G, Bonfante V, Viviani S et al. (2004) ABVD plus subtotal nodal versus involved-field
radiotherapy in early-stage Hodgkin’s disease: long term results. J Clin Oncol 22: 2835–41.
Dobbs HJ, Barrett A, Rostom AY et al. (1981) Total-body irradiation in advanced non-Hodgkin’s
lymphoma. Br J Radiol 54: 878–81.
Eich H, Mueller R, Engert A et al. (2005) Comparison of 30 Gy versus 20 Gy involved field
radiotherapy after two versus four cycles ABVD in early stage Hodgkin’s lymphoma: interim
analysis of the German Hodgkin’s Study Group Trial HD10. Int J Radiat Oncol Biol Phys 6: S1–2.
Engert A, Schiller P, Josting A et al. (2003) Involved-field radiotherapy is equally effective and less
toxic compared with extended-field radiotherapy after four cycles of chemotherapy in patients
with early stage unfavourable Hodgkin’s lymphoma. Results of the HD 8 trial of the German
Hodgkin’s Lymphoma Study Group. J Clin Oncol 21: 3601–8.
Girinsky T, van der Maazen R, Specht L et al. (2006) Involved-node radiotherapy (INRT) in patients
with early Hodgkin’s lymphoma: concepts and guidelines. Radiother Oncol 79: 270–7.
Girinsky T, Ghalibagian M, Bonniaud G et al. (2007) Is FDG-PET scan in patients with early stage
Hodgkin lymphoma of any value in the implementation of the involved-node radiotherapy
concept and dose painting? Radiother Oncol 85: 178–86.
Kaplan HS, Rosenberg SA (1966) The treatment of Hodgkin’s disease. Med Clin North Am 50: 1591–610.
Koontz B, Kirkpatrick J, Clough R et al. (2006) Combined modality therapy versus radiotherapy alone for
treatment of early-stage Hodgkin’s disease: cure versus complications. J Clin Oncol 24: 605–11.
Lee YK, Cook G, Flower MA et al. (2004) Addition of 18F-FDG PET scans to radiotherapy planning
of thoracic lymphoma. Radiother Oncol 73: 277–83.
National Institute for Health and Clinical Excellence (2008) Rituximab for the Treatment of Relapsed
or Refractory Stage III or IV Follicular Non-Hodgkin’s Lymphoma. Technical appraisal no 137.
NICE, London.
Sawyer EJ, Timothy AR (1997) Low dose palliative radiotherapy in low grade non-Hodgkin’s
lymphoma. Radiother Oncol 42: 49–51.
Shahidi M, Kamangari N, Ashley S et al. (2006) Site of relapse after chemotherapy alone in Stage I
and II Hodgkin’s disease. Radiother Oncol 78: 1–5.
Sutcliffe SB, Gospodarowicz MK, Bush RS et al. (1985) Role of radiation therapy in localised non-
Hodgkin’s lymphoma. Radiother Oncol 4: 211–23.
Yahalom J, Mauch P (2002) The involved field is back: issues in delineating the radiation field in
Hodgkin’s disease. Ann Oncol 13: 79–83.
Yahalom J (2006) Don’t throw out the baby with the bathwater: on optimizing cure and reducing
toxicity in Hodgkin’s lymphoma. J Clin Oncol 24: 544–8.
293
24 Oesophagus and
stomach

The oesophagus is arbitrarily divided into the cervical portion (from the lower
border of the cricoid cartilage to the thoracic inlet or suprasternal notch) and the
intrathoracic portion. The latter can further be divided into upper, middle and
lower thirds. Most oesophageal tumours are SCCs.
The term oesophago-gastric junction (OGJ) cancer usually includes
adenocarcinomas of the lower oesophagus and gastric cardia as well as the true
junction between the two. OGJ cancer is increasing in incidence in the Western
world. Most stomach cancers are adenocarcinomas.
The stage and location of the tumour are more important than histology in
guiding therapeutic decision making. The presence and number of involved lymph
nodes is the most important prognostic factor. In oesophageal cancer survival falls
from 70 per cent to 25 per cent when nodes are involved.
■ Oesophagus
Surgery is the principal curative therapy but only 40 per cent of patients survive for
2 years. Preoperative radiation does not improve cure rates but a meta-analysis has
shown that preoperative chemoradiation leads to a survival benefit of 13 per cent
at 2 years. However, individual phase III studies have often failed to show this
benefit and there is concern about perioperative morbidity and mortality.
Moreover, preoperative chemotherapy alone has shown a survival benefit in some
studies. The role of preoperative chemoradiation therefore remains unclear.
There is no good evidence to support adjuvant radiotherapy for any one group
of patients, but it should be considered in selected patients where the risk of local
recurrence is particularly high and exceeds that of distant metastases, e.g. with an
involved circumferential resection margin in node negative patients.
Non-surgical curative treatment has usually been used when patients are
medically inoperable or decline surgery or with T3N1 or T4 disease where surgical
cure is felt unlikely. Concomitant radiochemotherapy with cisplatin and
5-fluorouracil (5FU)-based chemotherapy improves survival compared to
radiation alone. Most regimens also include two cycles of chemotherapy either
before or after the radiochemotherapy. The 2-year survival with combined
chemotherapy and radiotherapy is very similar to that of surgery but there have
been no large trials comparing the two approaches. If the patient has small (5 cm
length) localised disease but is not able to tolerate surgery or radiochemotherapy,
radiation alone is used as potentially curative therapy.
Resection of a cervical oesophageal cancer would require a laryngo-
oesophagectomy so radiochemotherapy is the treatment of choice.
294
Many patients with oesophageal cancer are elderly, of poor PS or have metastases
at presentation. Radiotherapy is useful to palliate dysphagia or pain but rarely
relieves complete dysphagia. Intraluminal brachytherapy is also used in palliation.
■ Stomach
Surgery is the treatment of choice for localised tumours. The extent of surgery is
important. A D2 resection (extensive lymphadenectomy including nodes along the
branches of the coeliac axis) produces better local control than a D1 resection
(lymphadenectomy of perigastric nodes) though with an increase in postoperative
complications and no clear survival benefit. Many patients have a local recurrence
after surgery. A large RCT has shown adjuvant radiochemotherapy improves
3-year survival from 41 per cent to 50 per cent but only 10 per cent of patients
in the trial had a D2 resection so adjuvant treatment may be compensating for
poor surgery. Combined pre- and postoperative chemotherapy has also been
shown to improve survival compared with surgery alone and the extra benefit of
conformal postoperative radiotherapy when added to chemotherapy is not
known. Radiotherapy can be used to palliate obstructive symptoms or persistent
bleeding.

The optimal combination of surgery, radiotherapy and chemotherapy for
potentially curable oesophageal cancer is not clear. Where preoperative chemo-
radiation is used there should be a 6-week gap before surgery.

Oesophageal cancers can invade longitudinally along the oesophageal wall as ulcers
best seen at endoscopy. The lack of a serosal barrier means tumours can spread
radially to invade adjacent local structures such as the tracheobronchial tree, pleura
or recurrent laryngeal nerves.
Skip metastases – intramural tumour deposits distant from the primary – are
formed by extravasation from the network of submucosal lymphatics. They can
occur as much as 5–6 cm from the primary tumour and are found in up to a third
of patients. The lymphatic system enables longitudinal flow of lymph both
proximally and distally before it drains into adjacent nodes deep to the muscularis
layer.
The submucosal lymphatics also account for the high incidence of involved
lymph nodes and allow spread to distant nodes relatively early in the disease. The
cervical oesophagus drains first to the deep cervical nodes medial to the internal
jugular vein and then to those lateral to the vein in the supraclavicular fossa. The
thoracic oesophagus drains first to the para-oesophageal nodes and then to other
nodes in the mediastinum (see Fig. 20.2, p. 246). More inferior tumours spread
to the para-cardial nodes around the superior branch of the left gastric artery,
the left gastric nodes and the lesser curvature nodes before spreading to the
coeliac axis. However, surgical series show 28 per cent of upper third tumours
have involved gastric nodes and 33 per cent of lower third tumours have already
295
OESOPHAGUS AND STOMACH
spread to the superior mediastinum. The risk of nodal spread also increases with
T stage.
The stomach is divided arbitrarily into the fundus (cardia), body and pylorus
(antrum). As in the oesophagus there are extensive submucosal lymphatics that
facilitate intramural spread of tumour and make it difficult to predict the location
of involved nodes. Lymphatic drainage is initially to the perigastric nodes on the
greater and lesser curvatures and then to the nodes along the branches of the
coeliac axis (left gastric, hepatic, splenic and coeliac) (Fig. 24.1). Proximal
tumours can also spread to the para-oesophageal nodes.
Coeliac nodes Hepatoduodenal

Perigastric nodes on lesser curvature Left gastric nodes nodes
Common hepatic
Perigastric nodes on greater curvature Splenic nodes
nodes
(a) (b)
Figure 24.1 Lymphatic drainage of the stomach. (a) Perigastric nodes. (b) Stomach
removed to show nodes along the branches of the coeliac axis.

■ Oesophagus
Most tumours are diagnosed at endoscopy and the superior and inferior extent is
usually documented as the distance from the incisor teeth. Assuming an incisor to
carina distance of 25 cm may help to register endoscopic information with the
planning CT but this distance is more variable than previously thought. All patients
should have contrast-enhanced CT scan of the neck, thorax and abdomen to assess
the primary tumour and to look for enlarged nodes. Axial CT slices enable radial
spread to be assessed. Longitudinal extent can also be estimated with CT, particularly
when coronal and sagittal reconstructions are performed. Endoscopic ultrasound
(EUS) is the most accurate way to assess superior and inferior extent of the primary
tumour and can detect intramural skip metastases. The EUS information should be
described with reference to a point (e.g. carina) that can be seen on the planning CT
scan so that the position of the tumour can be accurately defined.
296
A bronchoscopy or vocal cord assessment should be performed if invasion of the
tracheobronchial tree or recurrent laryngeal nerves is suspected.
Curative treatment volumes should ideally cover all lymph node sites at
significant risk of harbouring microscopic metastases and would necessarily be very
large. However, much of the evidence for lymphatic spread comes from relatively
old surgical series. Modern staging with EUS, which has high sensitivity, and with
PET-CT, which has high specificity, enables high risk nodes to be identified more
accurately. A combination of CT, EUS and PET-CT will perhaps produce the
most accurate nodal staging.
■ Stomach
The surgeon, oncologist and pathologist should meet to discuss the most likely sites
of recurrence if adjuvant radiotherapy is to be considered, in view of its complexity.
Data acquisition
■ Simulator
For palliation, treatment volumes can be defined in the simulator. The patient lies
supine with arms by their sides as lateral or oblique beams are not used. The GTV
is defined from endoscopic and CT information. Barium can be used to determine
the superior extent of the tumour on fluoroscopy. No CTV is defined. The PTV
is the GTV with a 2 cm superoinferior margin and a 1 cm axial margin. In effect,
the beam edges are defined as 2–3 cm proximal and distal to the GTV with a 2 cm
lateral margin.
A similar technique is used for palliative radiotherapy for stomach cancer with
barium and CT information used to define the position of the tumour.
■ Immobilisation
Patients undergoing conformal radiotherapy are planned and treated lying supine
with arms above the head, ideally immobilised with a vacuum-formed polystyrene
bag. For cervical oesophagus tumours, a thermoplastic or Perspex shell should
be created in the same way as for hypopharyngeal tumours.
■ CT scanning
CT slices 3–5 mm thick are obtained from the hyoid bone to the inferior border
of the kidneys with intravenous contrast if possible. For adjuvant stomach
radiation the volume scanned should extend from the carina to the iliac crests.

■ Oesophagus
First, the superior and inferior extent of the GTV is defined by the bulk of disease
seen on the diagnostic and planning CT scans and by the information from EUS
and endoscopy. If these two modalities produce discordant results, the GTV
297
should encompass the greatest apparent extent of tumour superiorly and inferiorly
from all investigations. The axial extent of the GTV is defined using CT
information although EUS may help to indicate enlarged nodes that need to be
included. If the GTV is more than 10 cm long, a palliative approach should be
strongly considered.
A CTV is then produced by expanding the GTV by 2 cm longitudinally and
1 cm axially. This is edited in the axial plane to take account of possible routes of
spread (e.g. edited off the spinal column and aorta). The superior and inferior
slices are edited to ensure they reflect possible spread along the axis of the
oesophagus.
The para-oesophageal nodes in the superior mediastinum are posterior to the
trachea and to the left and right of the oesophagus, and more inferiorly they
include nodes adjacent to the thoracic duct and aorta. In practice they will be
included in the CTV as defined above.
There is no consensus about the inclusion of clinically negative nodes in the
CTV although there are now data to support a low rate of recurrence in adjacent
nodes when there is no prophylactic nodal radiotherapy. Many centres still
recommend irradiating high risk nodal sites particularly in N1 disease and would
include nodes in the supraclavicular fossae bilaterally for cervical oesophageal
tumours, adjacent mediastinal nodes for mid-oesophageal tumours, and the left
gastric nodes (along the lesser curve of the stomach) and coeliac nodes for OGJ
tumours.
The CTV is grown isotropically by 5 mm axially and 10 mm superoinferiorly to
produce the PTV, although these margins can be individualised in each centre if
the internal margin and set-up errors have been measured (Fig. 24.2).
(a) (b)
Figure 24.2 Volume definition for a T3N1 carcinoma of the thoracic oesophagus.
(a) GTV, CTV and PTV. (b) Superior axial slice to show the isotropically expanded CTV
(orange) which does not cover the oesophagus and the edited CTV (cyan) which does.
■ Stomach
For adjuvant treatment the CTV should include the tumour bed with a 2 cm margin
proximally and distally, and the perigastric nodes on the greater and lesser curvature.
The phase III trial which showed a survival advantage for chemoradiation included
the coeliac, local para-aortic, splenic, hepato-duodenal and pancreatico-duodenal
nodes and most authors therefore recommend these are treated. For tumours of the
298
Dose solutions
fundus, the lower para-oesophageal nodes should be included but the pancreatico-
duodenal nodes can be omitted. For antral tumours the splenic nodes can be
omitted. A 10 mm margin is added isotropically to produce the PTV.
The liver, kidneys and heart are contoured as OAR.
Dose solutions
■ Conformal
Oesophagus
Thoracic oesophageal volumes are best treated with an anterior and two posterior
oblique beams or with four equispaced beams. Both these arrangements spare the
spinal cord but deliver more dose to the lung than opposing AP beams (Fig. 24.3).
Angles for the lateral beams are chosen to avoid dose to the spinal cord PRV. The
weight applied to each beam is altered until a plan is produced which adequately
covers the PTV while not exceeding tolerance doses to the spinal cord PRV, lungs
or heart. The plan can be individually tailored depending on comorbidity. For
example in a patient with respiratory disease it may be appropriate to choose a
solution that further minimises lung dose at the expense of being closer to (but
not exceeding) tolerance of spinal cord PRV and heart. A superoinferior wedge
may be required in the anterior and posterior beams when more proximal tumours
are being treated to compensate for the inclined plane of the oesophagus which is
closer to the anterior chest wall in the upper thorax.
In some patients a two-phase technique is used with opposing AP beams for half
to two-thirds of the treatment and a three- or four-beam arrangement for the
remainder. This reduces the mean lung dose and can be especially useful if surgery
is planned after chemoradiation because respiratory complications are common
with tri-modality therapy.
Tumours of the cervical oesophagus are much more anterior so the anterior
beam is supplemented by right and left anterior oblique beams. A superoinferior
wedge is again required.
The spinal cord PRV dose should be kept below 40 Gy in view of the likely length
of cord adjacent to the PTV and the use of concomitant chemotherapy. The heart
V40 should be less than 30 per cent and the lung V20 less than 25 per cent.
Stomach
Anterior and posterior opposing beams were used to cover the target volume in the
Macdonald trial but more conformal volume-based techniques are now also described
using five coplanar or four non-coplanar beams. In addition to the critical organ doses
specified for oesophageal cancer, the liver V30 should be below 60 per cent and two-
thirds of one kidney (and ideally both) should be below 20 Gy.
■ Complex
IMRT has been used in both oesophageal and adjuvant stomach cancer. IMRT
with seven coplanar beams has significant theoretical advantages in sparing normal
tissues in stomach cancer but there is little published clinical data to support its use
compared with conformal radiotherapy for either tumour type.
299
(a) (b)
(c)
2400.0
2200.0
2000.0 A-P Beams
1800.0
3 Beam plan
1600.0 4 Beam plan
Volume [cm3]
1400.0 V20
1200.0
1000.0
800.0
600.0
400.0
200.0
0.0
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
20.0
22.0
24.0
26.0
28.0
30.0
32.0
34.0
36.0
38.0
40.0
42.0
44.0
46.0
48.0
50.0
52.0
54.0
56.0
58.0
60.0
(d) Dose(Gy)
Figure 24.3 Conformal beam arrangements for a T3N1 carcinoma of the thoracic
oesophagus. (a) Plan with anterior and posterior beams – minimal lung dose but high
dose to the cord and heart. (b) Three-beam plan for the same volume – low cord dose,
higher lung dose than a four-beam solution. (c) Four-beam plan for the same volume.
(d) DVHs for both lungs minus PTV for each plan treated to 50 Gy in 25 fractions.
300
Brachytherapy
Dose-fractionation
■ Oesophagus
Definitive radiochemotherapy
50 Gy in 25 daily fractions with cisplatin-5FU given in 5 weeks reduced to
45 Gy in 25 daily fractions if radiochemotherapy is given preoperatively.
Curative radiotherapy
or 20 Gy in 5 fractions of 4 Gy given in 1 week.
■ Stomach
45 Gy in 25 daily fractions of 1.8 Gy given in 5 weeks with concomitant 5FU and
leucovorin.

Mucositis is the predominant symptom in patients undergoing oesophageal
radiotherapy and should be proactively managed with analgesia and dietary advice.
High calorie supplement drinks are useful to maintain adequate oral intake.
Many patients receiving adjuvant radiotherapy for stomach cancer experience
grade 3 or 4 toxicity and nausea, lethargy and haematological effects are common.
Patients should be weighed and assessed weekly throughout treatment and should
be given prophylactic antiemetics (5-HT antagonist). It is essential to ensure
adequate nutritional intake before treatment begins.
Verification
Ideally the treatment isocentre on a DRR from the CT simulation is compared
with portal images of the isocentre on the treatment machines using electronic
portal imaging. Off-line correction protocols are used for standard conformal
treatment. Images are taken on days 1–3 and weekly thereafter with a correction
made if the mean error in any one plane is 5 mm.
Brachytherapy
High dose rate brachytherapy can be used to palliate dysphagia although
expandable metal stents are the treatment of choice for this symptom. A single
fraction of 10–15 Gy is prescribed at 1 cm from the central axis of the source.
301
Brachytherapy has also been advocated in addition to curative radiation or for local
recurrence after definitive radiotherapy.
key trials
Cooper JS, Guo MD, Herskovic A et al. (1999) Chemoradiotherapy of locally

advanced esophageal cancer: long-term follow-up of a prospective randomized
trial (RTOG 85–01). Radiation Therapy Oncology Group. JAMA 5 281: 1623–7.
Macdonald JS, Smalley SR, Benedetti J et al. (2001) Chemoradiotherapy after
surgery compared with surgery alone for adenocarcinoma of the stomach or
gastroesophageal junction. N Engl J Med 345: 725–30.
Randomized Phase III Trial of Adjuvant Chemotherapy or Chemoradiotherapy in
Resectable Gastric Cancer (CRITICS) (currently recruiting patients). Dutch
colorectal cancer group/Netherlands Cancer Institute. M Verheij, principal
investigator.
Information sources
Arnott SJ, Duncan W, Gignoux M et al. (2005) Oesophageal Cancer Collaborative Group.
Preoperative radiotherapy for oesophageal carcinoma. Cochrane Database Syst Rev (4):
CD001799.
Button MR, Morgan CA, Croydon ES et al. (2008) Study to determine adequate margins in
radiotherapy planning for esophageal carcinoma by detailing patterns of recurrence after
definitive chemoradiotherapy. Int J Radiat Oncol Biol Phys. 73: 818–23.
Cancer Care Ontario Gastrointestinal Cancer Evidence-based Series and Practice Guidelines.
Available at: www.cancercare.on.ca/english/home/toolbox/qualityguidelines/diseasesite/
gastro-ebs/ [cited 2008 October 11] (accessed 28 July 2008).
Gebski V, Burmeister B, Smithers BM et al. (2007) Survival benefits from neoadjuvant
chemoradiotherapy or chemotherapy in oesophageal carcinoma: a meta-analysis. Lancet Oncol
8: 226–34.
Scottish Intercollegiate Guidelines Network (2006) Management of Oesophageal and Gastric
Cancer. National Clinical Guideline no. 87. Available at: www.sign.ac.uk/pdf/sign87.pdf
Smalley SR, Gunderson L, Tepper J et al. (2002) Gastric surgical adjuvant radiotherapy consensus
report: rationale and treatment implementation. Int J Radiat Oncol Biol Phys 52: 283–93.
Wong RK, Malthaner RA, Zuraw L et al. (2003) Cancer care Ontario practice guidelines initiative
gastrointestinal cancer disease site group. Combined modality radiotherapy and chemotherapy
in nonsurgical management of localized carcinoma of the esophagus: a practice guideline. Int
J Radiat Oncol Biol Phys 55: 930–42.
302
Pancreas and liver 25
Pancreas
Cancer of the pancreas is the fifth (USA) or sixth (UK) commonest cause of cancer
death. With locally advanced disease (40 per cent), median survival is 6–10
months, and after successful complete resection (R0) alone for early stage disease,
11–15 months. After surgery, locoregional relapse occurs in 75 per cent. Forty
per cent of patients have metastases at presentation and untreated, median survival
is 3–6 months and at 1 year is only 1 per cent.
Most tumours are adenocarcinomas (90 per cent). They usually present late
because of an insidious and asymptomatic onset, especially those in the body
(15 per cent) or tail (10 per cent). Lesions in the head (75 per cent) may cause
obstruction of the bile ducts with jaundice but usual presenting symptoms are
anorexia, nausea and vomiting, fatigue and weight loss. CA19.9 is a useful
biochemical marker for monitoring treatment but it is not useful for screening as
it is negative in 10–15 per cent of cases.
Surgical resection is a prerequisite for long-term control but is only feasible in
about 20 per cent of patients, of whom only 30 per cent will actually have an
adequate R0 resection margin of 1 mm. Surgery can only be considered if there
is no evidence of metastatic disease (commonly in the liver or peritoneum), no
involvement of the coeliac, hepatic or superior mesenteric arteries, no obstruction
of the confluence of the superior mesenteric and portal veins, and minimal portal
vein involvement (180° involvement over 1 cm). The general PS of the patient
must be good.
Adjuvant chemotherapy with 5FU or gemcitabine after resection has been
shown to improve 5-year survival in this group by 10–20 per cent. The addition
of adjuvant radiotherapy has not yet improved outcomes.
For patients with borderline resectable tumours, neoadjuvant chemotherapy or
chemoradiotherapy are currently recommended to try to achieve resectability. For
locally advanced, unresectable tumours, initial chemotherapy with good evidence
of response may be followed by chemoradiotherapy.
For patients with metastatic or progressive disease in spite of adjuvant or
neoadjuvant treatment, palliation with gemcitabine and/or short course
radiotherapy is given. The evidence for these treatments is based on small
phase I and II trials with poor methodology and quality control (especially for
radiotherapy). New trials with neoadjuvant or adjuvant treatment using complex
and conformal radiotherapy in combination with drugs with a radiosensitising
effect, such as 5FU, and new agents such as bevacizumab are underway.
303
PANCREAS AND LIVER

After complete resection, chemotherapy is given adjuvantly. For other patients there
are various regimens for combining chemotherapy with radiotherapy. Commonly,
high dose radiotherapy is given with chemotherapy (5FU, gemcitabine/
bevacizumab) but since there are no studies which have yet shown statistically
significant benefit in terms of overall survival, it is recommended that patients should
be entered into collaborative clinical trials, with accurate staging, standardised
chemotherapy and carefully defined and optimised radiotherapy, organised from
centres with special expertise in the management of these rare tumours.

The pancreas is a retroperitoneal structure lying within the four parts of the
duodenum. Developmentally, it forms from ventral and dorsal outgrowths of the
foregut which fuse around the vessels that become the superior mesenteric artery
and vein. The head of the pancreas lies anterior to the inferior vena cava overlying
the first to third lumbar vertebral bodies. The body passes obliquely to the left
overlying the aorta, the left psoas muscle and the splenic artery and vein, while the
tail extends in front of the left kidney to the hilum of the spleen (Fig. 25.1).
Portal vein
Left suprarenal
Bile duct
Left colic flexure
Right suprarenal
Level of first
Head of pancreas lumbar vertebra
Right colic flexure
Uncinate process of pancreas
Superior mesenteric vein
Superior mesenteric artery Left ureter
Figure 25.1 Relationships of the pancreas.

The patient’s general medical condition and performance must be assessed.
Significant adverse prognostic factors which should be taken into account in
planning treatment are haemoglobin 12 g/dL and weight loss of 5 per cent in
the previous 3 months. Risk factors for pancreatic cancer, which will also increase
the complication rate after treatment, include smoking and alcohol use, obesity
and poor diet, chronic pancreatitis and previous peptic ulcer disease.
EUS, contrast-enhanced CT scanning and MRI can all be used to demonstrate
the primary tumour. CT and MRI are equivalent for assessing extent of local
disease, but neither is good at assessing margins of extension accurately, detecting
peritoneal spread or predicting lymph node involvement. Percutaneous biopsy
should be avoided if resection is planned, and EUS is preferred to investigate and
biopsy small tumours. Transabdominal ultrasound is performed if the patient
304
Pancreas
presents with biliary obstruction. Otherwise a contrast-enhanced CT scan with
water (1 L orally) and intravenous contrast is used for staging with images acquired
with a 1–1.25 mm slice thickness, to determine involvement of visceral arteries or
the portal vein, lymph node enlargement, bile duct obstruction and metastases.
Dynamic contrast-enhanced MRI can also be used to delineate primary tumour,
vascular involvement and liver metastases. FDG-PET-CT has limited usefulness as
it cannot distinguish between tumour and inflammation at the tumour margins
where the main uncertainty in GTV definition occurs, due to a surrounding
desmoplastic stromal reaction and infiltrating growth pattern.
Immobilisation
The patient lies supine in a vacuum moulded bag with arms above the head in arm rests.
CT scanning
CT scans are acquired as described above with a slice thickness of 5 mm at 2–5 mm
interslice intervals from the top of the liver or top of T11 to cover lymph nodes,
to the lower border of L3 and/or kidneys. Appropriate marks on the bag and the
patient’s skin are used to align lasers. CT images are usually used for planning
(Fig. 25.2) although CT-MRI fusion may be appropriate in some cases if
additional information is derived from an MR scan.
Figure 25.2 Axial CT scan showing

tumour of pancreatic body. T, tumour;
SMA, superior mesenteric artery; BD, bile
duct; D, duodenum.

Because of the difficulty in determining tumour margins, radiologist and radiation
oncologist should consult closely to delineate from scans the GTV which includes
any enlarged regional lymph nodes of 1.5 cm (GTV-T and -N).
The CTV should include visible tumour and surrounding oedema. This is an
area of considerable uncertainty and margins must be individually designed.
PTV margins are anisotropic with 5–10 mm in the AP direction, 2–4 mm in the
transverse plane and 15–30 mm cranio-caudally to take account of organ
movement with respiration or gut motion, as well as set-up variations. If breath
holding techniques or gating are used, cranio-caudal margins may be reduced.
305
PANCREAS AND LIVER
OAR include the spinal cord, kidneys, liver and small bowel. Dose-limiting small
bowel tolerance is reflected in the recommended dose-fractionation. These OAR
should be outlined on all slices for DVH assessment. A PRV may be added, but in the
transverse plane compromise may be needed to maintain these dose limits to OAR.
■ Dose solutions
Conformal
Using 6–10 MV photons, CT forward planning with volumes shaped with MLC may
be used to minimise dose to kidneys, liver, spinal cord and small bowel (Fig. 25.3).
Figure 25.3 Dose distribution for

treatment of tumour of pancreatic body
(shown in Fig. 25.2).
Complex
Complex plans may not be cost-effective at present outside trials because of poor
tumour outcomes, but full inverse planning using dose constraints to OAR and
IMRT with a five- (three anterior/two lateral) or seven-beam plan may be optimal
to reduce dose to the small bowel.
Conventional
For palliative or simple low dose treatments, AP/PA beams may give a satisfactory dose
distribution, or a plan with an anterior and two lateral wedged beams may be used.
Radical (in combination with chemotherapy with
gemcitabine or 5FU)
45–50.4 Gy in 25–28 fractions of 1.8 Gy given in 5–512⁄ weeks.
Palliative
This dose may be used to palliate pain or in association with more intensive
chemotherapy.

A 5-HT antagonist antiemetic should be prescribed before treatment begins. Decline
in performance status may be noted within 14 days of the start of treatment and
306
Liver and biliary tumours
nausea, diarrhoea and pain may continue to cause problems, or, if there is a tumour
response, may improve during treatment. Antacids and analgesics may also be needed.
Severe mucositis or even ulceration of the stomach or duodenum may occur. Special
attention must be given to maintaining adequate nutrition and hydration. Concomitant
chemotherapy may result in bone marrow suppression. It is sometimes difficult to
separate side effects from treatment from those due to progressive disease. Patients
should be seen weekly for close monitoring and symptom management.
■ Verification
EPIs should be taken on the first 5 days of treatment, for comparison with DRR
images obtained after CT planning, and then repeated once weekly after any
necessary adjustments have been made. In vivo dosimetry with TLD or diodes
should be used to ensure treatment dose is delivered as planned.

Primary liver tumours are the fifth commonest cancers in the world, being four
times more frequent in Asia than in Europe and the USA. Chronic liver infection,
an ageing population and increasing obesity contribute to an increasing incidence.
Liver tumours
Radical removal of liver tumours, where feasible, is the treatment of choice. For
early small volume disease (3 cm), resection may be possible. More widespread
involvement by tumour may necessitate liver transplantation.
For inoperable disease, thermal ablation by various means or chemotherapy may
be used. Response rates of 25 per cent can be achieved but there is no
improvement in survival, although intrahepatic chemotherapy with embolisation
may produce a slightly higher response rate.
Conventional and conformal EBRT have shown some responses which increase
with dose given, but radiation hepatitis is a dose limiting factor. This may be
partially overcome with complex treatment planning and stereotactic delivery of
radiation, and leads to 5–10 per cent long-term control.
Treatment with iodine-131 labelled antiferritin antibodies or lipiodol has shown
no survival benefit. Yttrium-90 is being investigated.
Cholangiocarcinoma
Gall bladder cancer has also been treated with EBRT with or without intraoperative
radiotherapy (IORT) (15 Gy) and 5FU, with good outcomes for patients with
previous complete resection or microscopic residual disease (80 per cent 2-year
local control).
For inoperable disease, EBRT may be followed by a boost from IORT or
brachytherapy if feasible, with prolongation of time to recurrence of 4 months and
2-year survival of 21 per cent. There is conflicting evidence about whether
increasing dose improves outcomes. Stereotactic radiotherapy is now being studied
for localised inoperable disease.
307
PANCREAS AND LIVER
Liver metastases
Whole liver irradiation doses must not exceed 30 Gy if radiation-induced liver
disease (radiation hepatitis) is to be avoided.
Advances in conformal EBRT and IMRT have made it possible to deliver focal
high dose radiotherapy treatments to individual metastases in patients with good
prognosis disease when surgical excision is not feasible. Response rates may exceed
50 per cent with high doses of EBRT (60 Gy) or stereotactic radiosurgery
(26 Gy) giving 80 per cent local control at 18 months.
The use of helical tomotherapy may make it possible to give highly focused
IMRT to metastases at the same time as the primary tumour.

Primary hepatic tumours may be classified as unifocal expansive, infiltrating or
multifocal (50 per cent). They invade the portal vein and spread thence to the
lungs. They present commonly with non-specific symptoms of abdominal pain,
malaise, fever and weight loss, although rupture and haemorrhage may occur.
Staging is by contrast-enhanced ultrasound, CT or MRI. Characteristically in the
arterial phase there is hypervascularity with wash-out in the portal venous phase
which must be seen in two different techniques of imaging to confirm the diagnosis.
Biopsy is only done if scans are not characteristic.
Expected toxicity from treatment can be predicted from the Child–Pugh
classification of cirrhosis which takes into account levels of bilirubin, prothrombin
time, albumin, and presence of ascites and encephalopathy to classify into grades
A, B and C (worst).

Liver tumours or metastases
The GTV may be defined using PET-CT scans or MR images taking into account
all respiratory phases. A CTV margin of 5 mm and CTV-PTV margin of a further
5 mm are contoured.
Cholangiocarcinoma
GTV is determined by preoperative CT. CTV should include the gall bladder fossa
and adjacent liver and regional nodal areas. To allow for movement with
respiration, margins for CTV to PTV should be about 20 mm cranio-caudally,
8 mm AP and 9 mm left to right.
■ Dose solutions
These treatments require complex solutions with stereotaxis or tomotherapy
and should be carried out in centres with appropriate equipment and expertise.
Cholangiocarcinoma
Inoperable disease
54 Gy in 30 daily fractions of 1.8 Gy given in 6 weeks
15 Gy as IORT boost.
308
or
EBRT 36–55 Gy + brachytherapy 15 Gy in 3 daily fractions of 5 Gy HDR or 25 Gy
at 1 cm LDR.
Stereotactic radiotherapy is now being studied for localised inoperable disease.
Liver metastases
20–30 Gy in 13–20 daily fractions of 1.5 Gy may occasionally be used to relieve
painful hepatomegaly.
Stereotactic radiosurgery of 26 Gy or tomotherapy delivering 40 Gy in 5 fractions
of 8 Gy are being investigated for single or few liver metastases and for treatment
of primary tumours.

Nausea and vomiting are controlled with 5-HT antagonist antiemetics with or
without steroids. Bilirubin, prothrombin time and albumin must be monitored.
Whole liver irradiation doses must not exceed 30 Gy if radiation induced
liver disease (radiation hepatitis) is to be avoided. This presents 2 weeks to 2 months
after radiotherapy with ascites, hepatomegaly, confusion and jaundice with raised
levels of alkaline phosphatase, prothrombin time and thrombocytopenia. The
pathological basis is veno-occlusive disease and the outcome is fatal in 10–20 per cent
of patients.
■ Verification
DRRs derived from CT planning are compared with EPIs or on board imaging,
which is recommended for these complex treatments.
key trials
Pancreas
Neoptolemos JP, Stocken DD, Friess H et al. (2004) European Study Group for
Pancreatic Cancer. A randomised trial of chemoradiotherapy and chemotherapy
after resection of pancreatic cancer. N Eng J Med 350: 1200–10.
Neuhaus P, Riess H, Post S et al. (2008) CONKO-001: Final results of the
randomised, prospective, multicenter phase III trial of adjuvant chemotherapy
with gemcitabine versus observation in patients with resected pancreatic cancer
(PC). J Clin Oncol 26(20 suppl): abstract LBA4504.
Oettle H, Post S, Neuhaus P et al. (2007) Adjuvant chemotherapy with
gemcitabine vs. observation in patients undergoing curative-intent resection of
pancreatic cancer: a randomised controlled trial. JAMA 297: 267–77.
Regine WF, Winter KA, Abrams RA et al. (2008) Fluorouracil vs. gemcitabine
chemotherapy before and after fluorouracil-based chemoradiation following
resection of pancreatic adenocarcinoma: a randomised controlled trial. JAMA
299: 1019–26.
309
PANCREAS AND LIVER
Information sources
■ Pancreas
Ghaneh P, Costello E, Neoptolemos JP (2007) Biology and management of pancreatic cancer. Gut
56: 1134–52.
Small W Jr, Berlin J, Freedman GM et al. (2008) Full-dose gemcitabine with concurrent radiation
therapy in patients with nonmetastatic pancreatic cancer: a multicenter phase II trial. J Clin
Oncol 20: 942–7.
Sultana A, Tudur-Smith C, Cunningham D et al. (2007) Meta-analyses of chemotherapy for locally
advanced and metastatic pancreatic cancer. J Clin Oncol 25: 2607–15.
Van Laethem J, Van Cutsem E, Hammel P et al. (2008) Adjuvant chemotherapy alone versus
chemoradiation after curative resection for pancreatic cancer: feasibility results of a randomised
EORTC/FFCD/GERCOR phase II/III study (40013/22012/0304). ASCO Meeting Abstracts: 4514.
■ Liver and biliary tract

Graco C, Catalano G, Grazie AD et al. (2004) Radiotherapy of liver malignancies. From whole liver
irradiation to stereotactic hypofractionated radiotherapy. Tumori 90: 73–9.
Lawrence TS (2008) Radiotherapy for intrahepatic cancers: the promise of emerging sophisticated
techniques. J Support Oncol 6: 14–15. Available at www.SupportiveOncology.net.
Pugh RN, Murray-Lyon IM, Dawson JL (1973) Transection of the oesophagus for bleeding
oesophageal varices. Br J Surg 60: 646–9. (Child–Pugh classification).
Tse RV, Hawkins M, Lockwood G (2008) Phase 1 study of individualised stereotactic body
radiotherapy for HCC and intra-hepatic cholangiocarcinoma. Clin Oncol 26: 657–64.
310
Rectum 26

Colorectal cancer is the second most common cancer in the UK in terms of incidence
and mortality. Both incidence and cure rates are increasing. One-third of colorectal
tumours arise in the rectum, more commonly in men.
Most rectal cancers are adenocarcinomas which are discussed here. Subtypes
which have a poorer prognosis include signet ring, and mucinous adenocarcinomas
(associated with microsatellite instability and hereditary non-polyposis colorectal
cancer [HNPCC]). Medullary carcinoma may be associated with a better prognosis.
Rare types of rectal cancer include small cell carcinoma, carcinoids, lymphoma,
sarcoma and squamous cell carcinoma. Squamous cell carcinomas arising from the
transitional area between rectum and anal verge are classified and treated as anal
cancer.
Surgery is the mainstay of treatment and gives 5-year overall survival rates of
93 per cent (stage I), 72–85 per cent (stage II), 44–83 per cent (stage III) and
8 per cent (stage IV). The 5-year survival in stage IV disease with resection of liver
metastases after primary treatment is 36 per cent. The most significant improvement
in treatment has been the widespread use of total mesorectal excision (TME)
which has reduced local recurrence rate to less than 10 per cent, compared with
30 per cent with older surgical techniques.
In selected T1,N0,M0 tumours 3 cm in diameter that are not poorly differen-
tiated, endocavitary local contact radiotherapy with the Papillon technique using a
low energy 50 kV machine has produced good results and new equipment for this
technique is being developed. HDR brachytherapy can also be used for small,
localised rectal tumours.
Pre- and postoperative adjuvant therapies are used in high-risk cases to
improve resectability, reduce local recurrence, increase sphincter preservation,
and improve overall survival. Based on the same evidence, practices differ. In
North America, resectable cancers are treated with postoperative chemoradio-
therapy if they are found to be T3/4 or node positive. Some centres in Europe
advocate short course preoperative radiotherapy for all resectable rectal cancers as
it provides a quick, effective, practical and cheap method of delivering neo-
adjuvant radiotherapy. Alternatively long course preoperative chemoradiotherapy
311
RECTUM
can be given to selected patients with non-resectable tumours or resectable

tumours with MRI features predictive of subsequent incomplete excision:
■ primary tumour or lymph node is threatening the circumferential resection margin
(CRM) i.e. within 1–2 mm of mesorectal fascia
■ involvement or extension beyond the mesorectal fascia
■ involved lymph nodes outside the mesorectal fascia
■ extramural vascular invasion (EMVI).
Additional high risk factors for local recurrence are:
■ tumours in the lower third of the rectum which often require abdominoperineal
resection (APER) with risk of poor lateral margin clearance with consequent
higher local recurrence rates
■ large anterior quadrant tumours, where sphincter-preserving surgery may be
difficult.
Preoperative chemoradiotherapy is the subject of ongoing studies but it is used
more often in these poor risk tumours. All patients are discussed in a multidisciplinary
team meeting between surgeons, pathologists, radiologists and oncologists to plan
multimodality treatment for each individual patient on the basis of risk factors.
Palliation may be achieved for unresectable and locally advanced T4 tumours with
radiotherapy alone. Alternatively long course chemoradiotherapy may be given to
try to downstage the tumour for resection. Hypofractionated weekly radiotherapy
may be used to palliate local symptoms in patients unfit for longer
chemoradiotherapy schedules. Local recurrence is less of a problem with widespread
use of TME and preoperative radiotherapy. It may cause neuropathic sacral and
sciatic pain with bladder and bowel dysfunction. Chemoradiotherapy or radio-
therapy alone may offer effective short-term palliation for patients with local
recurrence who have not previously been irradiated. Patients who have received
radiotherapy previously may be considered for re-irradiation, but long-term
toxicity data are lacking and small bowel toxicity may occur.
An algorithm for the management of rectal cancer is shown in Figure 26.1. The
type of surgical resection needed depends on the site of the cancer. Upper third
rectal tumours can be removed by a high anterior resection and TME, mid and lower
third tumours can be removed by anterior resection with TME, and low rectal
tumours less than 5 cm from the anal verge or those with sphincter involvement
require an APER. TME and sphincter-preserving surgery can only be performed if
the surgical resection margin below the tumour is 1 cm away from the sphincter.
The main RCTs that have influenced practice are shown in Table 26.1.

Surgery is performed 1 week after short course preoperative EBRT. Surgery is
delayed until 6–10 weeks after long course preoperative chemoradiotherapy. After
12 weeks, radiation fibrosis presents problems at surgery. The standard
chemotherapy used with radiotherapy for rectal cancer is 5FU as bolus, continuous
312
Carcinoma of
the rectum
Preoperative staging
Upper third Middle third Lower third
T1 – T3a T1 – T3a,
T3 ( 5 mm from MRF) T1 surgery
surgery
surgery
T3b T3b T2
SCRT or ChemoRT SCRT or ChemoRT SCRT or ChemoRT
T3 (5 mm from MRF) T3c,

T3c – T4 ChemoRT T3 – T4 ChemoRT
T4 ChemoRT
Key:
T3a,<5mm extramural tumour invasion
T3b, 5–10mm
T3c, >10mm
MRF, Mesorectal fascia
SCRT, Short course preop Radiotherapy
ChemoRT, Long course preop Chemoradiotherapy
Figure 26.1 Algorithm for management of rectal cancer.
Table 26.1 Main RCTs which have influenced management of rectal cancer
Trial Randomisation Results
GTSG Study; N Engl Surgery 9-year OS 27 per cent, LR 25 per cent

J Med (1985) Surgery POCRT 9-year OS 54 per cent*, LR 10 per cent*
Mayo NCCTG; N Engl S PORT 5-year OS 40 per cent, LR 25 per cent
J Med (1991) S POCRT 5-year OS 55 per cent*, LR 15 per cent*
Swedish Rectal Trial; Surgery 5-year OS 48 per cent, CSS 65 per cent,
N Engl J Med (1997) LR 27 per cent
Preop SCRT S (25 Gy/5#) 5-year OS 58 per cent*, CSS 74 per cent*,
LR 11 per cent*
Dutch Rectal Trial; S (TME) 2-year OS 81.8 per cent, LR 8.2 per cent
N Engl J Med (2001) Preop SCRT 2-year OS 82.0 per cent, LR 2.4 per cent*
S(TME) (25 Gy/5#)
(Continued)
313
RECTUM
Table 26.1 Continued
German Rectal Trial; Pre-op LCCRT S (TME) 5-year OS 76 per cent, LR 6 per cent*
N Engl J Med (2004) S (TME) POCRT 5-year OS 74 per cent, LR 13 per cent
CR07 Rectal Trial; Preop SCRT S (TME) 3-year OS 80.8 per cent, DFS 79.5 per cent,
ASCO 2006 LR 4.7 per cent*
S (TME) selective 3-year OS 78.7 per cent, DFS 74.5 per cent,
POCRT LR 11.1 per cent
Polish Study; Br Preop SCRT S(TME) 4-year OS 67.2 per cent, DFS 58.4 per cent,
J Surg (2006) LR 9 per cent
Preop LCCRT S(TME) 4-year OS 66.2 per cent, DFS 55.6 per cent,
LR 14.2 per cent
*p 0.05.
S, surgery; PORT, postoperative radiotherapy; POCRT, postoperative chemoradiotherapy; preop LCCRT, preoperative
long course chemoradiotherapy; preop SCRT, preoperative short course radiotherapy; TME, total mesorectal excision;
OS, overall survival; CSS, cause-specific survival; DFS, disease-free survival; LR, local recurrence.
infusion or orally. Studies are underway comparing different combination regimens

with new agents such as capecitabine and oxaliplatin.

The rectum extends from the external sphincter to the recto-sigmoid junction. It is
divided into a lower third 3–6 cm, middle third 5–6 cm to 8–10 cm, and upper third
8–10 to 12–15 cm from the anal verge (Fig. 26.2). The upper third of the rectum is
surrounded by peritoneum on the anterior and lateral surfaces and is retroperitoneal
posteriorly. At the recto-vesical or recto-uterine pouch, the rectum becomes
completely retroperitoneal and follows the curve of the sacrum entering the anal
canal at the level of the levator ani. The mesorectum contains the blood supply and
lymphatics for the upper, middle and lower rectum. The location of a rectal tumour
is defined by the distance from the lower edge of the tumour to the anal verge.
Figure 26.2 Sagittal T2-weighted MRI of a

middle third rectal cancer showing division
of the rectum into lower, middle and upper
thirds. b, bladder; sp, symphysis pubis; as
anal sphincter.
314
Flexible endoscopes can overestimate this distance compared with measurements
with rigid endoscopes, which therefore may be preferred since the distance from the
anal sphincter to the tumour is critical for planning sphincter sparing surgery.
Rectal carcinomas arise in the mucosa and may be exophytic, ulcerated or annular,
when they may produce obstruction of the lumen. Tumours extend through the wall
of the serosa to invade surrounding organs such as the bladder, prostate and vagina,
with direct extension into the presacral region in advanced cases. The lymphatic
drainage of the rectum is to the mesorectal lymph nodes (contained in mesorectal
fascia), mesenteric lymph nodes (along inferior mesenteric artery), lateral lymph
nodes (along middle rectal, obturator and internal iliac vessels) and external iliac
lymph nodes. There is also drainage to hypogastric and presacral lymph nodes as
shown in Fig. 26.3. Spread toward the anal margin may be associated with inguinal
lymph node involvement.
Mesenteric nodes
External
iliac nodes
Hypogastric
nodes
Presacral
nodes
Rectum
Mesorectal
nodes
Figure 26.3 Lymphatic drainage of the rectum.

The standard staging procedure includes a history, examination and digital rectal
examination (DRE) to determine the distance of the tumour from the anal verge,
sphincter involvement and clinical fixation. Full blood count, renal and liver
function tests and CEA are performed. A raised CEA suggestive of metastases
predicts a poor prognosis. Examination of the entire colon with a barium enema
or colonoscopy to exclude a synchronous primary is essential. CT scanning is used
to assess the pelvis, abdomen and liver, with CT or CXR to assess the lungs.
Endorectal ultrasound or MRI are used to stage the primary rectal tumour. The
European multi-institutional MERCURY study has shown the central role of MRI
in staging rectal cancers. Accuracy (92 per cent), specificity (98 per cent), positive
and negative predictive values (73 and 93 per cent, respectively) confirm its
315
RECTUM
usefulness. In patients in whom an MRI predicts a clear CRM, 94 per cent will
have a clear margin at surgery. If it predicts probable involvement or shows tumour
less than 1 mm from the CRM, it is highly likely that there is involvement.
The AJCC and UICC 6th edition TNM staging system 2002 is currently used
for evaluating rectal cancer. Patients are staged clinically and radiologically before
surgery to define the need for neoadjuvant therapy, and pathologically staged
postoperatively to define the need for adjuvant therapy. Nodal involvement carries
a poorer prognosis but it is recognised in the staging system that patients with
T1–T2, N1 rectal cancer have a better prognosis than other stage III rectal
cancers. Disease staged N0 after surgery has a better prognosis when 12–14 lymph
nodes are identified. A close or positive CRM is a poor prognostic indicator.
Data acquisition
Patients lie in the prone position. The small bowel can be displaced anteriorly by
the use of devices such as a bellyboard, which allows it to fall forwards into the
bellyboard aperture (Fig. 26.4). Modern bellyboard devices are more comfortable,
improve immobilisation and reduce set-up errors in the prone position. A full
bladder protocol is used for planning and treatment as this displaces small bowel
superiorly. A radio-opaque marker is placed on the anal verge and a DRE performed
to determine the distance from the anal verge marker to the inferior edge of the
tumour. The anterior border of the vagina can be identified using a radio-opaque
tampon. Oral contrast may be used to help identify small bowel. Treatments are
planned using CT-based virtual simulation or conformal CT planning. A planning
CT scan is performed with 3 mm slices from the level of L5 to 2 cm below the anal
marker. Low rectal tumours are better defined on MRI than on CT. However at
present staging MRI is performed with the patient supine. In the future there may
be a role for co-registration of planning CT with prone MR scans. The use of FDG-
PET for rectal cancer planning is currently being investigated. If a simulator has to
be used, bony landmarks help to define the target volumes.

■ Preoperative radiotherapy
The target volume includes the primary tumour, adjacent lymph nodes and the
presacral region. The inferior mesenteric nodes are not included because they are
removed at later surgery. It has been shown that extension of the treatment volume
above L5 increases late complications significantly. Treatment may be planned by
CT defined target volumes or using standard bony landmarks as in the CR07 trial.
CT defined target volumes

Target volumes are based on the known patterns of local recurrence in rectal cancer.
The GTV includes all gross tumour seen on the planning CT scan with reference to
information from diagnostic endoscopy, MRI and DRE. Any involved lymph
nodes, extrarectal extension, or extranodal deposits seen on MRI should be
included. The CTV includes the primary tumour, mesorectum from the sacral
promontory to insertion of levator ani into the rectal wall, and the posterior
316
(a) (b)
(c)
Figure 26.4 (a) Bellyboard for prone immobilisation. (b) CT plan with bellyboard.
(c) CT plan without bellyboard.
presacral space. The inferior pelvic subsite, including the anal perineum, anal
sphincter, perianal and ischiorectal space, is included if the tumour is 6 cm from
the anal margin or if the tumour invades the anal sphincter. The mesorectal and
lateral lymph nodes are included in all patients. The obturator nodes are included
when the tumour is 10 cm from the anal margin, and the external iliac lymph
nodes included if there is anterior organ involvement. The inguinal lymph nodes
may be included if tumour invades the lower third of the vagina, or if there is
major tumour extension into the internal and external anal sphincter.
317
RECTUM
In principle, the PTV is created by adding a margin of 1–1.5 cm to the CTV. In

practice, a margin may be added to the GTV to create the PTV directly ensuring
the above CTV sites are included. The margins used to grow the GTV to PTV
are 3–5 cm superiorly (limited by the L5/S1 junction), 2–3 cm inferiorly, 2 cm
anteriorly, and posteriorly to cover the sacrum.
A boost to the primary tumour can be added. The phase 2 PTV then includes
the tumour mass with a 2 cm 3D margin. Small bowel is demonstrated by oral
contrast and contoured as an OAR.
Conventional planning
The following bony landmarks can be used to define field borders:
■ lateral: 1 cm outside bony pelvis
■ posterior: 1 cm behind sacrum to include sacral hollow and presacral lymph nodes
■ superior: sacral promontory, L5/S1 border as defined on lateral sagittal view
■ anterior: 2–3 cm anterior to sacral promontory and including the anterior
vaginal wall in females. Care must be taken to include the whole rectum which
may be more anteriorly placed. This can be seen on CT, at virtual simulation or
on diagnostic MR scans
■ inferior: 3 cm below the inferior edge of the tumour. For lower third tumours
the border should lie below the anal marker to cover the perineum.
■ Postoperative radiotherapy
Close collaboration with the surgical team and pathologists is essential to aid
planning with a full description of the extent of residual tumour, anatomical
location of clips demarcating the tumour bed, and sites of close margins. The target
for postoperative radiotherapy is the tumour bed, adjacent lymph nodes, presacral
region and any residual tumour. Clinical examination is used to localise the level of
the anastomosis in patients after anterior resection or the perineal scar following
APER. Surgical procedures such as reconstruction of the pelvic floor and absorbable
mesh slings can be used to reduce the amount of small bowel in the pelvis.
Because the GTV has been removed and the CTV is large, complex solutions
and conformal CT planning are used to minimise the amount of normal tissue
(especially small bowel) in the target volume, but the target volume is planned
with reference to standard bony landmarks:
■ lateral: 1 cm outside bony pelvis
■ posterior: 1 cm behind sacrum to include sacral hollow and presacral lymph
nodes
■ superior: sacral promontory, L5/S1 border as defined on lateral sagittal view
■ anterior: 2 cm anterior to sacral promontory and including the anterior vaginal
wall
■ inferior: 3 cm below the inferior edge of the anastomosis or to the inferior
obturator foramen, whichever is lower. For patients who have had an APER the
perineal scar is marked to ensure it is covered.
An optional phase 2 volume is individualised to target the tumour bed and any
residual tumour with a 2 cm 3D margin. This is best defined on CT using all the
operative and pathological information available.
318
Dose solutions
Dose solutions
Dose distributions with high energy beams (10 MV) are often better than with
lower energies. A three- or four-beam arrangement includes a direct posterior
beam with reduced weighting, and either two lateral or posterior oblique wedged
beams depending on the shape of the patient (Fig. 26.5).
(a) (b)
(c)
Figure 26.5 Conformal plan using direct posterior and lateral beams shown on CT slice:
(a) axial, (b) coronal and (c) sagittal.
Posterior oblique beams at a gantry angle of 45–60° will produce a rounded

volume with some anterior spread of dose, whereas lateral beams will give a sharp
cut off anteriorly, reducing small bowel dose. Angles of less than 45° are not used
as this causes overlap of the beams posteriorly, increasing the skin reaction in the
natal cleft. With advanced disease where there is bladder or anterior abdominal
wall involvement, a four-beam arrangement or AP opposing beams may need to
be used to increase the dose anteriorly.
319
RECTUM
3D conformal CT planning allows for DVH assessment of the normal tissues

and conformal shielding with MLC to reduce the dose to normal tissues, most
importantly the small bowel. The TD 5/5 for one-third of the small bowel volume
is 50 Gy. This is based on data showing a 3 per cent incidence of late small bowel
obstruction with doses of 45Gy delivered to 664 cm3, about one-third of the
small bowel volume. When planning conformally, the volume of small bowel
receiving over 45 Gy should be kept to the minimum possible and no small
bowel should receive over 50 Gy. Care must be taken to identify small bowel that
may be stuck in the sacral hollow postoperatively.
When planning radiotherapy for low rectal tumours or postoperatively following
APER, the target volume includes the perineum or perineal scar. External bolus
over these areas may be needed to ensure adequate dose inferiorly.
Where possible the anal canal should be excluded from the radiation beams to
preserve function of the anal sphincter. The ano-rectal junction can be identified
using 3D conformal planning or virtual CT simulation.
IMRT planning has limited value at present. In the future, better definition of
volumes with the possibility of dose escalation may make it useful if stage-
dependent CTV targets can be defined.
For palliative treatment, a smaller volume can be used which covers the rectal
tumour, sacrum and involved soft tissue and local lymph nodes only to minimise
the amount of small bowel treated. Treatment can be planned in the same way as
described above using CT planning or diagnostic films to identify bony landmarks
to create a target volume.
Dose-fractionation
■ Preoperative radiotherapy
Short course
Long course
Phase 1
Phase 2 (optional)
5.4–9 Gy in 3–5 daily fractions of 1.8 Gy.
Phase 1
Phase 2 (optional)
5.4–9 Gy in 3–5 daily fractions of 1.8 Gy.
Palliative long course chemoradiotherapy may be used for maximal local control
for inoperable rectal cancers where prolonged survival is possible.
320
Verification
Phase 1
Phase 2 (optional)
5.4–14.4 Gy in 3–8 daily fractions of 1.8 Gy
or a hypofractionated regimen can be used
30–36 Gy in 5–6 fractions of 6 Gy once weekly given in 5–6 weeks.

The patient is treated each day in the prone position with a full bladder. A low
residue diet is advised and the patient seen weekly to assess toxicity. Erythema and
desquamation of the skin can occur in the perineal and sacral areas. Nursing
assessment, hydrocolloid dressings, nutritional support and analgesia are important.
Diarrhoea should be treated with loperamide hydrochloride as appropriate. If small
bowel acute radiation toxicity is suspected with abdominal pain and localised
peritonism, the patient should be rested from treatment and the radiotherapy plan
and sites of small bowel reviewed. A minority of patients receiving short course
preoperative radiotherapy develop an acute sensory neuropathy, which can be
alleviated by reduction in the treatment volume to the level of S2/3.
Verification
Patients are set up daily, using sagittal and lateral tattoos (over the iliac crests) and
lasers to prevent lateral rotation.
Electronic portal imaging is used to verify the treatment set-up and the EPIs are
compared with the DRRs to ensure the isocentre, beams and MLC shielding
configuration are correct. Set-up can be monitored with EPIs according to local
protocols during treatment with correction for displacements outside tolerance
levels of 5–8 mm. With future advances, image guided radiotherapy with kV and
cone beam CT may play a greater role in rectal cancer radiotherapy planning.
Bosset JF, Calais G, Mineur L (2005) Enhanced tumorocidal effect of

key trials
chemotherapy with preoperative radiotherapy for rectal cancer: preliminary

results – EORTC 22921. J Clin Oncol 23: 5620–7.
Braendengen M, Tveit KM, Berglund A et al. (2008) Randomised phase III study
comparing preoperative radiotherapy with chemoradiotherapy in nonresectable
rectal cancer. J Clin Oncol 26: 3687–94.
Bujko K, Nowacki MP, Nasierowska-Guttmejer A et al. (2006) Long-term results of a
randomised trial comparing preoperative short-course radiotherapy with preoperative
conventionally fractionated chemoradiation for rectal cancer. Br J Surg 93: 1215–23.
Gastrointestinal Tumor Study Group (1985) Prolongation of the disease-free
interval in surgically treated rectal carcinoma. N Engl J Med 312: 1465–72.
Kapiteijn E, Marijnen CA, Nagtegaal ID et al. (2001) Dutch Colorectal Cancer
Group. Preoperative radio therapy combined with total mesorectal excision for
resectable rectal cancer. N Engl J Med 345: 638–46.
321
RECTUM
Krook JE, Moertel CG, Gunderson LL et al. (1991) Effective surgical adjuvant
therapy for high-risk rectal carcinoma. Mayo NCCTG. N Engl J Med 324:
709–15.
MERCURY Study Group (2006) Diagnostic accuracy of preoperative magnetic
resonance imaging in predicting curative resection of rectal cancer: prospective
observational study. BMJ 333: 779.
Sauer R, Becker H, Hohenberger W et al. (2004) German Rectal Cancer Study
Group. Preoperative versus postoperative chemoradiotherapy for rectal cancer.
N Engl J Med 351: 1731–40.
Sebag-Montefiore D, Quirke P, Steele RJ et al. (2006) CR07: Pre-operative
radiotherapy vs. selective post-operative chemo-radiotherapy for patients with
rectal cancer. ASCO abstract. Pro ASCO 24, 18S, abstract 3511.
Swedish Rectal Cancer Trial (1997) Improved survival with preoperative
radiotherapy in resectable rectal cancer. N Engl J Med 336: 980–7.
Information sources
Borger JH, van den Bogaard J, de Haas DF et al. (2008) Evaluation of three different CT simulation
and planning procedures for the preoperative irradiation of operable rectal cancer. Radiother
Oncol 87: 350–6.
Elective Clinical Target Volumes in Anorectal Cancer: An RTOG Consensus Panel Contouring Atlas.
Available at www.rtog.org.
Letschert JG, Lebesque JV, Aleman BM et al. (1994) The volume effect in radiation-related late small
bowel complications: results of a clinical study of the EORTC Radiotherapy Cooperative Group
in patients treated for rectal carcinoma. Radiother Oncol 32: 116–23.
MRC Clinical Trials Unit. Results of CR07. Available at www.ctu.mrc.ac.uk/studies/CR07.
Myint AS (2007) Radiotherapy for early rectal cancer. Clin Oncol 19: 637–8.
Nagtegaal ID, Quirke P (2008) What is the role for the circumferential margin in the modern
treatment of rectal cancer? J Clin Oncol 26: 303–12.
National Institutes of Health Consensus Conference (1990) Adjuvant therapy for patients with colon
and rectal cancer. JAMA 264: 1444–50.
National Institute for Health and Clinical Excellence. Improving Outcomes in Colorectal Cancer.
Available at: www.nice.org.uk/csgcc.
Nicholls J (2008) The multidisciplinary management of rectal cancer. Colorectal Dis 10: 311–13.
Roels S, Duthoy W, Haustermans K et al. (2006) Definition and delineation of the clinical target
volume for rectal cancer. Int J Radiat Oncol Biol Phys 65: 1129–42.
322
Anus 27

Anal cancer is rare with approximately 500–600 new cases diagnosed each year in
the UK. It is more common in women and the median age at diagnosis is 60. Anal
margin tumours, however, are more common in men and at an earlier age. Primary
chemoradiotherapy is the current treatment of choice.
Eighty per cent of all anal cancers are epidermoid squamous cell carcinomas,
which are discussed here. Most cases are preceded by high grade anal
intraepithelial neoplasia (AIN), but only 1 per cent per year of patients with AIN
develop invasive cancer. Anal margin tumours are often keratinising and well
differentiated while canal tumours are usually non-keratinising and poorly
differentiated. Upper canal tumours may be mixed squamous cell and
adenocarcinoma of transitional, basaloid or cloacogenic type. Rare types of
primary anal cancer include adenocarcinomas, small cell carcinomas, melanoma,
lymphoma and leiomyosarcoma.
The risk factors associated with anal cancer are human papillomavirus (16, 18)
infection (found in over 80 per cent of cases), genital warts, other sexually
transmitted diseases, immunosuppression, ano-receptive intercourse and tobacco
smoking. It is associated with human immunodeficiency virus (HIV) infection but
not correlated with degree of immunosuppression and is not considered an
acquired immune deficiency syndrome (AIDS) defining illness.
With radical chemoradiotherapy, 5-year survival rates are 90 per cent for T1,
80 per cent for T2, 45–55 per cent for T3/4 and 65–75 per cent overall.
The initial studies of Nigro demonstrated the curative potential of
chemoradiotherapy for anal cancer. Further studies confirmed the benefit of
adding mitomycin C (MMC) and 5FU to radiotherapy for anal cancer and three
large RCTs in the1990s confirmed benefits (Table 27.1).
Radical chemoradiotherapy can be considered for all patients with non-
metastatic anal cancer who are fit for radical treatment. It is associated with an
increased risk of toxicity in patients with HIV infection, inflammatory bowel
disease and after previous pelvic surgery or renal transplantation. Patients with
HIV and immunosuppression with a CD4 count 200 develop significant toxicity
and morbidity with chemoradiotherapy. The dose of radiotherapy and
chemotherapy may have to be modified but this reduces control rates. Patients
with inflammatory bowel disease tolerate pelvic radiotherapy poorly with a high
risk of late toxicity. Patients who have had previous pelvic surgery often have
adhesions and risk increased dose to small bowel and late complications. The
323
ANUS
Table 27.1 RCTs of chemoradiotherapy for anal cancer
UKCCR ACT 1 (n 585) RT alone 3-year LC 39%, CSS 61%, OS 58%

RT 5FU/MMC 3-year LC 61%,* CSS 72%,* OS 65%
EORTC (n 103) RT alone 3-year LC 55%, OS 65%
RT 5FU/MMC 3-year LC 65%,* OS 70%
RTOG (n 291) RT 5FU 4-year LC 64%, DFS 50%, OS 65%
RT 5FU/MMC 4-year LC 83%,* DFS 67%*, OS 67%
*p 0.05.
RT, radiotherapy; OS, overall survival; LC, local control; CSS, cause-specific survival; DFS, disease-free survival.
multidisciplinary team should carefully consider the risks and benefits of radical
surgical resection versus radical chemoradiotherapy.
Surgery is now most commonly performed for failure after chemoradiotherapy.
Local excision may be considered for node negative anal margin tumours less than
2 cm in diameter without anal canal involvement providing the tumour can be
completely excised without damage to the sphincter. After incomplete excision,
adjuvant chemoradiotherapy is advised.
Palliative radiotherapy can be considered for patients who present with metastatic
disease or have such poor performance status that they will not tolerate radical
chemoradiotherapy.

Current chemoradiotherapy protocols recommend that the first fraction of
radiotherapy should be given after commencement of an intravenous infusion of
5FU chemotherapy. This is given in the first and last weeks of radiotherapy with
other agents such as MMC or cisplatin. Trials are studying the best combination
of drugs with radiotherapy and whether maintenance chemotherapy is
advantageous.

The anal canal is 3–4 cm long and extends from the anorectal ring (top of ‘surgical’
anal canal), composed of upper fibres of the internal sphincter, to the anal margin
as shown in Fig. 27.1. The anal margin consists of the area of skin around the anal
orifice. Tumours of the anal margin are slow growing and tend to infiltrate locally
within the perineum, with late spread to lymph nodes. Tumours of the anal canal
most commonly arise in the transitional zone just above the dentate line, and tend
to spread proximally in the submucosa to the distal rectum. Tumours invade
324
anteriorly into the vagina and uncommonly to the prostate, laterally to the ischio-
rectal fossa, and posteriorly along the ano-coccygeal ligament to the coccyx.
Anterior spread to the prostate may be limited by Denonvilliers’ fascia, and inferior
spread may be limited by the suspensory ligaments.
Rectum
Dentate line
‘Surgical’
anal canal Internal
sphincter
External
sphincter
Anal margin
Figure 27.1 Anatomy of the anus.
Lymphatic drainage from the anal margin and perianal skin is to the superficial
inguinal and femoral lymph nodes, and thence to the external iliac nodes. The
lymphatic drainage of the anal canal, proximal to the dentate line, is to the superior
rectal, superior haemorrhoidal, hypogastric, obturator, internal iliac and presacral
lymph nodes (Fig. 27.2). Tumours distal to the dentate line drain to superficial
inguinal nodes as well as to pararectal nodes.
Node at origin of
inferior mesenteric
artery
External iliac nodes

Presacral nodes
Hypogastric nodes
Superior
Obturator node haemorrhoidal nodes
Anus Pararectal nodes
Inguinal nodes
Figure 27.2 Lymphatic drainage of the anus.
325
ANUS

A full history should elicit symptoms of local tumour and possible spread, and
record comorbidities, PS and fitness for radical treatment. Clinical examination
includes inspection of the perineal skin, DRE to assess the site and extent of the
primary tumour, and vaginal examination to detect involvement of the posterior
vaginal wall. Endoscopy may be performed to exclude a synchronous rectal or
colonic tumour. HIV testing should be done in patients at risk. EUA allows
detailed assessment, and biopsies of the tumour and of any enlarged inguinal
lymph nodes 1 cm.
Thirty per cent of patients have enlarged inguinofemoral lymph nodes, of which
50 per cent are involved and 50 per cent show reactive inflammatory change only.
Further local staging is performed with endorectal ultrasound and MR scan of the
pelvis, and CT scan of the chest and abdomen to exclude distant metastases. FDG-
PET scanning is useful in primary tumour staging and to assess residual, recurrent
and metastatic disease.
Tumours of the anal canal and margin are staged according to the TNM and
AJCC 2002 classification. The terms anal margin and perianal skin are
synonymous and cancers at this site are classified as skin tumours. Poor prognostic
factors are metastases, lymph node spread, higher T stage, older age, poor PS and
anaemia (haemoglobin 10 g/dL).
Data acquisition
Patients are treated in the prone position, which displaces small bowel superiorly
to reduce toxicity and allows easy visualisation of the anal verge for the application
of radio-opaque markers and perianal bolus. The small bowel can also be displaced
anteriorly by the use of devices such as a bellyboard, which allows it to fall forwards
into the bellyboard aperture (Fig. 26.4, p. 317). Modern bellyboard devices are
more comfortable, improve immobilisation and reduce set-up errors in the prone
position. The AP separation is less in the prone position than supine.
Treatment is planned with CT scanning, virtual simulation or conventional
simulator films. Oral contrast is used to delineate small bowel on CT scans or
simulator films. Thin lead wire or a soft flexible catheter containing contrast delineates
the anal canal. All palpable disease in the anal margin or lymph nodes is marked with
radio-opaque material for identification on CT scans or simulator films. Surface
boundaries of the inguinal region are marked on the skin with wire for simulation.

Current trials such as the RTOG 0529 trial of IMRT for anal cancer define CT-
based target volumes for anal cancer (see website below). In the UK the ACT II
RCT defines the target volumes in terms of field borders for a two-phase shrinking
field technique, which has ensured consistency of treatment between cancer centres.
■ CT planning
The phase 1 target volume includes all macroscopic primary tumour (GTV-T) and
involved lymph nodes (GTV-N), and all microscopic disease at risk in the pelvis,
326
inguinal and femoral regions (CTV-TN). The phase 2 target volume includes all
macroscopic primary tumour (GTV-T) and involved lymph nodes if present
(GTV-N) with a 10–15 mm margin for the CTV. Both these target volumes
should be planned at the initial planning visit as the initial macroscopic disease may
respond rapidly to treatment.
The extent of tumour in the anal canal is often difficult to determine on
planning CT images or simulator films. It may be necessary to use the clinical
information about the size of the tumour, its length and distance of the inferior
limit from the anal verge to contour the GTV. Using radio-opaque wire on the
anal verge and rectal contrast, the distance of the superior extent of the anal
tumour from the anal verge can be confirmed. Diagnostic MR scans which show
the extent of tumour in three dimensions can be fused with the planning CT scan,
or measurements from MRI can be transferred onto the CT planning scan or
simulator films.
CTV-PTV margins of 10–15 mm allow for patient positioning and set-up errors.
OAR include the small bowel, bladder and vagina. Transplanted kidneys may lie in
the pelvis and should be excluded from the treatment volume or repositioned.
■ Conventional planning
Phase 1 volume (all tumours):
■ Superior border: 2 cm above inferior aspect of the sacroiliac joints is the standard
border. However, the superior border should be enlarged to include a minimum
margin of 3 cm above the upper extent of GTV-T or GTV-N.
■ Lateral border: to include both inguinal nodal regions – in practice this border
lies lateral to the femoral head.
■ Inferior border: 3 cm below the anal margin (for disease confined to the anal
canal only) or 3 cm below the most inferior extent of tumour (for anal margin
tumours)
These borders are shown in Fig. 27.3.
2 cm
3 cm
Figure 27.3 Phase 1 treatment borders for all
anal tumours (from ACT II). Reproduced with
Marker on anal verge permission from Dr D Sebag-Montefiore on
(canal only tumour) behalf of the ACT II triallists.
327
ANUS
Phase 2 volume
The volume depends on the presence or absence of positive nodes in the
inguinofemoral or pelvic regions and is defined from CT scanning. Lymph nodes
that are likely to contain tumour should be treated to the full phase 2 dose.
Phase 2 volume for lymph node negative cases (N0)

All borders allow 3 cm around the GTV defined at initial planning as shown in
Fig. 27.4 for anal canal tumours and Fig. 27.5 for anal margin tumours.
3 cm 3 cm
Figure 27.4 Phase 2 treatment borders for
N0 anal canal tumours (ACTII). Reproduced
3 cm 3 cm 3 cm 3 cm
with permission from Dr D Sebag-Montefiore
3 cm 3 cm
on behalf of the ACT II triallists.
Figure 27.5 Phase 2 treatment borders for N0

3 cm
anal margin tumours (ACTII). Direct field with 3 cm
margins superiorly, inferiorly and lateral to GTV.
Reproduced with permission from Dr D Sebag-
3 cm
Montefiore on behalf of the ACT II triallists.
Phase 2 volume for lymph node positive disease (N)

Figure 27.6 shows borders for treatment of GTV with a 3 cm margin and MLC or
lead shielding to exclude normal tissue and reduce toxicity.
3 cm
3 cm
Figure 27.6 Phase 2 treatment borders for
3 cm
N anal tumours (ACTII). Reproduced with
3 cm
permission from Dr D Sebag-Montefiore on
behalf of the ACT II triallists.
328
Dose solutions
Dose solutions
Conformal/conventional
Phase 1
Beam arrangements are chosen to cover the PTV as defined above, but commonly
anterior and posterior opposing beams are used with MLC shielding to normal
tissues. With conventional planning, standard borders are used as described above.
High photon energies (10–20 MV) give more satisfactory dose distributions,
especially for large patients. Bolus is used between the buttocks and over the
perineum to ensure adequate dose to anal margin tumours and tumours within
2 cm of the anal verge. Superoinferior dose compensation may be needed to
improve dose homogeneity.
Phase 2 (node negative)
A planned volume with three or four beams is used for anal canal tumours with no
involved lymph nodes. A posterior and two wedged lateral beams are used to cover
the PTV with a fourth anterior beam added if it improves dose distribution. In the
future, with better definition of target volumes, IMRT with sparing of normal
tissues may be used.
For patients with disease confined to the anal margin only, a direct photon
beam, or electron therapy with energy chosen to encompass the CTV within the
90 per cent isodose, is used. Photons may give better coverage at depth with less
lateral scatter than with electrons.
This is best planned in the simulator under direct vision. Tumours that extend
onto the anal margin and anal canal tumours that extend to within 2 cm of the anal
verge require bolus to the perianal skin to ensure they are not underdosed. Small
bolus bags or a custom-made wax block can be used.
Phase 2 (node positive)
For anal canal tumours with involved lymph nodes, anterior and posterior opposing
beams are used in the same way as for phase 1 with MLC shielding to as much small
bowel and bladder as possible. CT virtual simulation can be used to create a 3 cm
margin around the GTV-T and involved lymph nodes (GTV-N) with MLC shaping.
As discussed above the need for postoperative treatment following surgery is now
uncommon.
In cases where there has been incomplete excision or involved circumferential
margins, the patient can be treated with chemoradiotherapy in a similar fashion to that
described above for lymph node negative cases. The phase 2 borders are based on the
site of the tumour bed, using information from the operation notes and preoperative
imaging. For small tumours less than 2 cm with close margins or incomplete excision,
single-phase conformal planning or interstitial brachytherapy may be used.
In patients who present with metastatic disease the initial treatment is with
systemic chemotherapy if they are fit. If the volume of metastatic disease is small,
329
ANUS
many patients benefit from additional chemoradiotherapy to achieve local control

in the pelvis. This can be planned as outlined above for radical chemoradiotherapy,
or as a smaller volume single-phase treatment of GTV plus a 3D margin of 3 cm.
In patients who are unfit for radical treatment, small volume radiotherapy to
conventional palliative doses may be used.
■ Interstitial brachytherapy
Both LDR iridium wire and HDR afterloading perineal implants can be used to treat
anal cancers. Brachytherapy alone can be used for small T1 tumours and as a boost
following radical chemoradiotherapy with local control rates of 80–90 per cent. It
can also be used for palliative treatment of recurrent or locally advanced tumours.
Dose-fractionation
■ Curative or adjuvant chemoradiotherapy
Phase 1
Phase 2
19.8 Gy in 11 daily fractions of 1.8 Gy given in 21⁄2 weeks
or
Single phase
50.4 Gy in 28 daily fractions of 1.8 Gy given in 6 weeks.
Low dose chemoradiotherapy

30 Gy in 15 daily fractions given in 3 weeks to GTV 3 cm margin with
continuous 5FU.

Patients are treated prone either flat on the couch or using a bellyboard
immobilisation system with a full bladder. It is important that set-up for anal
margin tumours is checked on the first day by the radiotherapist to ensure that
there is adequate cover of the primary tumour inferiorly.
This is one of the few radical treatments where grade 3 and 4 toxicity is expected.
The perineal and inguinal tissues are particularly sensitive to irradiation and skin
reactions are often brisk and painful. Regular review and use of hydrocolloid
dressings, nutritional support, analgesia and antidiarrhoeal medication are essential.
During concomitant chemotherapy, patients should receive appropriate
antiemetics, and prophylactic antibiotic cover is advised for the duration of
treatment. Blood tests are monitored regularly for myelosuppression and any sign
of infection treated promptly.
330
Information sources
Verification
Lasers and skin tattoos are used to align the patient each day and lateral tattoos
over the iliac crests help to prevent lateral rotation. Electronic portal imaging is
used to verify the treatment set-up and the EPIs are compared with the DRR
images. Set-up can be monitored with EPIs according to local protocols during
treatment with correction for displacements outside tolerance levels of 5–8 mm.
Potential changes to radiotherapy target volumes and dose escalation, IGRT with
kV and cone beam CT may come to play a greater role in anal cancer radiotherapy.
key trials
Ajani JA, Winter KA, Gunderson LL et al. (2008) Fluorouracil, mitomycin, and
radiotherapy vs fluorouracil, cisplatin, and radiotherapy for carcinoma of the anal
canal: a randomised controlled trial. JAMA 23: 299: 1914–21.
Bartelink H, Roelofsen F, Eschwege F et al. (1997) Concomitant radiotherapy
and chemotherapy is superior to radiotherapy alone in the treatment of
locally advanced anal cancer: results of a phase III randomised trial of
the European Organization for Research and Treatment of Cancer
Radiotherapy and Gastrointestinal Cooperative Groups. J Clin Oncol
15: 2040–9.
Flam M, John M, Pajak TF et al. (1996) Role of mitomycin in combination with
fluorouracil and radiotherapy, and of salvage chemoradiation in the definitive
nonsurgical treatment of epidermoid carcinoma of the anal canal: results of a
phase III randomised intergroup study. J Clin Oncol 14: 2527–39.
Glynne-Jones R, Meadows H, Wan S et al. (2008) EXTRA—A multicenter phase
II study of chemoradiation using a 5 day per week oral regimen of capecitabine
and intravenous mitomycin C in anal cancer. Int J Radiat Oncol Biol Phys 72:
119–26.
James R, Meadows H, Wan S (2005) ACT II: the second UK phase III anal
cancer trial. Clin Oncol 17: 364–6.
Nigro ND, Vaitkevicius VK, Considine B Jr (1974) Combined therapy for cancer
of the anal canal: a preliminary report. Dis Colon Rectum 17: 354–6.
UKCCCR Anal Cancer Trial Working Party. UK Co-ordinating Committee on
Cancer Research (1996) Epidermoid anal cancer: results from the UKCCCR
randomised trial of radiotherapy alone versus radiotherapy, 5-fluorouracil, and
mitomycin. Lancet 348: 1049–54.
Information sources
ACT II Trial. Available at: www.ucl.ac.uk/cancertrials/trials/actii/index.htm (accessed 8 December 2008).
Brachytherapy (2005) In: Hoskin P, Coyle C (eds) Radiotherapy in Practice. Oxford University Press,
Oxford.
Charnley N, Choudhury A, Chesser P et al. (2005) Effective treatment of anal cancer in the elderly
with low-dose chemoradiotherapy. Br J Cancer 92: 1221–5.
Elective Clinical Target Volumes in Anorectal Cancer: An RTOG Consensus Panel Contouring Atlas.
Available at: www.rtog.org/
331
28 Prostate

Prostate cancer is now the commonest cancer in men, accounting for almost
25 per cent of all new male cancer diagnoses and is the second most common cause
of cancer related death in men. There has been a huge rise in the recorded incidence
of prostate cancer with the use of prostate-specific antigen (PSA) and surgery for
benign prostatic hypertrophy (BPH), but this is not reflected in increased mortality
rates. There are almost 32 000 new cases and 10 000 deaths from prostate cancer a
year in the UK. However, most men die with their prostate cancer rather than from
it and management must balance the potential toxicity of active treatment, with the
chances of benefit in a disease with a long natural history.
Advances in diagnosis and screening policies with the use of PSA have led to a
stage migration so that prostate cancer is now detected at earlier stages with better
prognostic features.
The most common type of prostate cancer is adenocarcinoma (95 per cent).
Tumours are graded using the Gleason scoring system which evaluates architectural
details of individual cancer glands and describes five distinct growth patterns from
Gleason 1 (well differentiated) to Gleason 5 (poorly differentiated). The two
commonest growth patterns seen are summated to give a final Gleason score (GS)
ranging from 2 (1 1) to 10 (5 5). There are other rarer types of prostate
carcinoma such as ductal, intralobular acinar, small cell and clear cell. This chapter
focuses on the treatment of the common adenocarcinoma. There is less evidence to
guide treatment of other pathological subtypes. Prostate cancer is staged using the
AJCC and TNM staging.
Tumours are stratified by T stage, GS and PSA into three prognostic groups of
low, intermediate and high risk:
■ low risk: T1–T2a and PSA 10 ng/mL and GS
6
■ intermediate risk: T2b or PSA 10–20 ng/mL or GS 7
■ high risk: T2c–T4 or PSA 20 ng/mL or GS 8–10.
The risk is higher in an intermediate risk patient who has GS 4 3 than with
GS 3 4.
Patients can be offered appropriate treatment options by a multidisciplinary
team, according to stage of disease, prognostic risk group and estimated survival
taking into account performance status and comorbidity.
EBRT, interstitial brachytherapy and surgery are options for the curative treatment
of localised prostate cancer with equivalent outcomes but different side effects.
332
The options for treatment of low, intermediate and high risk prostate cancer are
listed in Box 28.1. Factors that influence the choice include PS, other medical
illnesses, likelihood of progression to symptomatic disease, life expectancy, morbidity
of treatment (particularly on sexual function) and patient preference.
Box 28.1 Treatment options for localised prostate cancer

Low risk: (T1–T2a), PSA 10 ng/mL, GS
6)
■ Active surveillance
■ Prostate iodine-125 brachytherapy
■ Radical prostatectomy (open/laparoscopic/robotic)
■ Radical EBRT 6 months HT
■ Watchful waiting
Intermediate risk (T2b or PSA 10–20 or GS 7)
■ Prostate125I brachytherapy (T2b GS 34 [in 50 per cent tumour sample] and
PSA 15)
■ Radical prostatectomy (open/laparoscopic/robotic)
■ Radical EBRT with 6 months HT (WPRT)
High risk:( T2c–T4 or PSA 20 or GS 8)
■ Radical EBRT with 2–3 years HT (WPRT)

■ Hormone therapy alone
HT, hormone therapy; WPRT, whole pelvis radiotherapy.
In general patients should have a life expectancy of greater than 10 years before
radical treatment is recommended. Men younger than 75 with other major illnesses
such as ischaemic heart disease and diabetes may not live 10 years and men over
75 with no other illnesses may live into their nineties. There is evidence that radical
radiotherapy in fit men over 75 is very well tolerated and just as effective. It is the
role of the multidisciplinary team to identify patients who will benefit from radical
treatment and counsel the patient on the different options available. Radical treatments
with surgery, EBRT or brachytherapy have similar outcomes and the patient
should be informed of the different side effects of each treatment. Factors that
influence the final choice for an individual patient include lower urinary tract
symptoms, sexual function, likelihood of infertility, and risks of anaesthesia and
surgery. The benefits and risks of surgery, EBRT and brachytherapy are shown in
Table 28.1.
The results of radiotherapy are assessed by monitoring the PSA, DRE and symptoms
and signs of metastases. A PSA nadir of 2 within 2 years is associated with long-
term control. Biochemical failure is defined by a rise in PSA by 2.0 ng/mL above the
nadir level following radiotherapy, according to the international RTOG-ASTRO
333
PROSTATE
Table 28.1 Benefits and risks of radical treatment options for localised disease
Treatment Advantages Disadvantages
EBRT Curative Daily treatment for 6–8 weeks

No operation Combined with hormone therapy
No anaesthetic with side effects
No hospital admission Early side effects:
Maintain usual activities • Cystitis and diarrhoea
during treatment • Fatigue
PSA monitoring to detect Late side effects:
recurrence (more difficult to • Rectal toxicity
interpret than after surgery) • RTOG G2 12 per cent G3 1 per cent
• Bladder toxicity
• RTOG G2 10 per cent
• Impotence 50 per cent
• Infertility
Brachytherapy Curative Needs anaesthetic
1–2 day-case visits Early side effects:
Quick recovery • Urinary retention 10 per cent
No hormone therapy • Local discomfort
PSA monitoring to detect • Urgency and cystitis
recurrence (can take 2 years • Impotence 25–50 per cent
for PSA to fall) • Proctitis 2–5 per cent
May remain fertile Late side effects:
• Bladder toxicity
• RTOG G2 30 per cent
• Urethral stricture 10 per cent
• Impotence 25 per cent
• Incontinence 1 per cent
• Rectal bleeding 24 per cent
Surgery Curative Risks of major surgery
Also treats BPH and symptoms 3–7 days as in-patient
Pathological assessment 6 weeks recovery
of whole prostate May leave cells that have
PSA falls quickly following breached the capsule
surgery allowing monitoring Early side effects:
• Incontinence (recovers in 6 weeks)
• Risks of PE/DVT
• Mortality 0.16–0.66 per cent
Late side effects:
• Incontinence 2–10 per cent
• Impotence 5–60 per cent
• Stricture 0.5–9 per cent
• Infertility
Active Avoid unnecessary treatment ‘Worry’ while closely monitored
surveillance No side effects Treatment may be needed at later date
Only an option for cancers with a low
risk of progression
334
Phoenix consensus. Published nomograms are very useful for assessing potential
benefits from treatment. The Memorial Sloane Kettering pretreatment nomogram
gives a 5-year progression-free probability following radical EBRT from 70–90 per cent
for low risk to 50–70 per cent for intermediate risk and 20–50 per cent for high
risk prostate cancer.
■ Pelvic lymph node radiotherapy

Selected patients may benefit from radiotherapy to the pelvic lymph nodes with
the prostate and seminal vesicles. A randomised trial has shown that patients with
a risk of lymph node involvement between 15 per cent and 35 per cent may benefit
from pelvic lymph node irradiation. However, although pelvic radiotherapy has
shown a trend to improved 5-year progression-free survival, but not overall survival,
this benefit is only seen in patients receiving neoadjuvant, concurrent and adjuvant
hormone therapy. This has to be balanced with the increased toxicity associated
with WPRT. With modern techniques, including IMRT and IGRT, this balance
may become more favourable.
■ Adjuvant and salvage radiotherapy

Adjuvant radiotherapy is not given routinely when PSA levels are undetectable
after surgery to avoid overtreatment of the majority of patients already cured. If
the PSA is persistently raised following surgery, it may indicate local or metastatic
disease. After negative staging investigations, metastases are still likely if the PSA
doubling time is 9 months, and local radiotherapy is therefore not recommended.
Seminal vesicle or lymph node involvement, especially with a high Gleason score,
also correlates with risk of metastases and makes radical local radiotherapy
inappropriate. If there is a persistently raised PSA with a doubling time 9 months
and there are positive margins at the site of extracapsular extension, local radiotherapy
should be considered. Results are better with early salvage radiotherapy when the PSA
is 1.2 ng/mL (or 0.2 with the supersensitive assay). In patients in whom PSA later
rises on three consecutive occasions (ASTRO definition of biochemical failure),
salvage radiotherapy is also offered. Clinical trials of radiotherapy may be appropriate
for patients with high risk factors for local recurrence but undetectable PSA.
Prostate and pelvis
Patients with extensive local disease in the prostate and/or pelvis may benefit from
a course of palliative EBRT to relieve symptoms. Bleeding from the prostate may
also be alleviated by EBRT.
Bone and lymph node metastases

EBRT is an excellent treatment for palliation of pain from symptomatic bone
metastasis. A single fraction of 8 Gy is very effective with few side effects.
A fractionated course of EBRT may be used to treat spinal cord compression, and
is given postoperatively following orthopaedic fixation of pathological fractures.
Symptoms from lymph node and visceral metastases can also be relieved by EBRT.
335
PROSTATE
Radionuclide therapy
Patients with metastatic bone disease and multiple painful sites can be difficult to
treat with EBRT to localised areas. The radionucleotides strontium and samarium
have been shown to be beneficial in reducing symptoms from bone metastases.
This treatment is only considered for patients with good bone marrow reserve, no
imminent risk of cord compression and no nodal or visceral metastases.
Breast buds
Gynaecomastia is a major problem for patients on long-term anti-androgen and
oestrogen therapy. Superficial X-ray or electron therapy (9–12 MeV) to the breast
bud area using a 7–9 cm diameter circular field has been shown to reduce the
incidence significantly when used prophylactically. In established cases, radiotherapy
can reduce symptoms.

■ EBRT and hormone therapy
Hormone therapy before, during and after EBRT (neoadjuvant, concurrent and
adjuvant androgen deprivation [NCAD]) has been proven to increase local control,
disease-free survival and overall survival for selected patients with prostate cancer.
Androgen deprivation reduces the size of the prostate by 30 per cent and the
number of tumour cells, possibly through synergistic apoptotic mechanisms.
Six months NCAD can be used with EBRT for low and intermediate risk disease
to reduce the target volume and allow safer dose escalation. Treatment must be
started at least 2 months before radiotherapy. NCAD with 2–3 years continuing
adjuvant androgen deprivation is beneficial for high risk disease (Table 28.2).
Table 28.2 Summary of EBRT target volumes and duration of hormone therapy
according to risk group
Risk group Low Intermediate High
Target volume Prostate BSV Prostate Prostate

BSV/SV SV
WPRT WPRT
Duration of None or 6 months 6 months 2–3 years
hormone therapy
BSV, base of seminal vesicle; SV, seminal vesicle; WPRT, whole pelvic radiotherapy. The SV is included if the risk of
SV involvement is greater than 15 per cent using the Roach formula: SV risk PSA [(GS6) 10].
Standard androgen deprivation is achieved by cyproterone acetate 100 mg three

times daily or bicalutamide 50 mg once daily for 3 weeks starting 1 week before the
first LH-releasing hormone agonist/antagonist (LHRHa) injection. The LHRHa is
continued monthly for 6 months and then 3 monthly for 2–3 years in high risk cases.
■ EBRT and brachytherapy boost

HDR and LDR brachytherapy may be used as a boost after EBRT to escalate dose
to the prostate.
336
■ EBRT and chemotherapy
The combination of chemotherapy and EBRT for high risk prostate cancer is
currently under investigation.

The prostate gland lies between the pubic symphysis and the anterior rectal wall
and is closely applied to the bladder neck and seminal vesicles (Fig. 28.1). The
lymphatics drain from the prostate to the obturator, presacral, internal, external,
common iliac and para-aortic lymph nodes (Fig. 28.2).
(a) (b)
Figure 28.1 Anatomy of prostate shown on T2-weighted MRI. (a) Axial (TZ transitional
zone, PZ peripheral zone). (b) Sagittal (pb, penile bulb; b, bladder; sp, symphysis pubis;
r, rectum; df, Denonvilliers’ fascia).
Para-aortic
nodes
Presacral
nodes
Common
iliac nodes
Internal
iliac nodes External
iliac nodes
Bladder
Obturator
Seminal nodes
vesicles Prostate
Figure 28.2 Lymphatic drainage of the prostate.

337
PROSTATE

Prostate cancer is confirmed by multiple biopsies under transrectal ultrasound
guidance or transperineally using a perineal template.
The extent of primary disease is assessed by MRI of the prostate and MRI or CT
scan of the pelvis and abdomen to detect locoregional and distant spread. Prostate
cancer can extend through the prostate capsule, especially posteriorly, at the apex,
and at the junctions between the base and bladder neck and seminal vesicles.
Denovilliers’ fascia reduces spread towards the rectum. Extraprostatic extension is
shown on MRI in Fig. 28.3. Spread occurs to the seminal vesicles and lymph nodes
with increasing tumour stage and histological grade. The risk of microscopic
seminal vesicle and lymph node involvement can be assessed by nomograms or
Roach formulae, both based on the Partin data.
Figure 28.3 T2-weighted MRI showing T3

prostate cancer with extraprostatic extension
on left.
Roach formulae:
Percentage risk of seminal vesicle involvement PSA [(GS 6) 10]
Percentage risk of LN involvement 2/3 PSA [(GS 6) 10]
Patients in the intermediate and high risk groups also need a bone scan
except those with a PSA 20.
Data acquisition
■ Immobilisation
A planning CT scan is obtained in the treatment position. Patients are treated
supine rather than prone as this has been shown to produce less prostate motion,
reduce doses to normal organs at risk, and is more comfortable for the patient.
A bladder filling protocol should be used to maintain a constant bladder filling –
‘comfortably full’. Patients are asked to empty the bladder and drink 200 mL of
water 20 min before the scan and before treatment each day. A completely full
bladder has been shown to displace small bowel away from the treated volume, but
338
Data acquisition
it also leads to greater variation in prostate position and is not recommended. The
rectum should be empty for treatment as a full rectum also leads to greater
variation in prostate position. Patients should be advised on a low residue diet and
if they have a full rectum at the time of planning CT scan should receive further
dietary advice before repeating the scan. The use of laxatives, suppositories, enemas
and endorectal balloons during treatment continues to be investigated.
An immobilisation system using a head pad combined with individually
adjustable knee and ankle supports provides a high degree of accuracy without the
need for further pelvic immobilisation (Fig. 28.4).
Figure 28.4 Immobilisation

for treatment of prostate
cancer.
■ CT scanning
With the patient immobilised in the treatment position following bladder and rectal
protocols, a radiotherapy planning CT scan is performed. Skin reference tattoos are
placed anteriorly on the midline of the symphysis pubis and laterally over the hips
and aligned with lasers to prevent lateral rotation. Radio-opaque markers are placed
on the skin to locate the tattoos on the CT scans. The CT scan is taken with 3–5 mm
slices from the mid sacroiliac joint to 1 cm below the anus/ischium to include the
prostate, seminal vesicles, rectum and bladder, and is extended superiorly to L3 if
the pelvic lymph nodes are to be treated. No oral or rectal contrast is used, but
intravenous contrast may aid delineation of the pelvic lymph nodes. At the time of
the planning CT scan, the size of the rectum and bladder should be assessed. If
the bladder is empty or the rectum is 4 cm in AP diameter at the level of the
prostate base, the scan should be repeated after implementing the bladder and
rectal protocols until the desired parameters are met. Inappropriately large rectal
volumes have been shown to reduce local control rates by moving the prostate out
of the planned PTV.
CT data are then transferred to a radiotherapy planning computer for outlining
and target volume definition. To improve target definition, MR scans of the pelvis
can be incorporated into radiotherapy planning protocols. Outlining studies have
shown that the size of the prostate is overestimated on CT compared with MRI,
which defines the apex of the prostate better and is associated with less image
degradation with fiducial markers. Solutions to overcome the geometrical distortion
and shift artefact seen with MRI are currently being investigated. CT-MRI image
339
PROSTATE
registration (Fig. 28.5) is useful for defining the contour of the prostate especially
at the apex and for IGRT with fiducial markers.
Figure 28.5 Co-registration of CT and MR

images for target volume definition.

■ Prostate ⴞ seminal vesicles
It is not possible to define the GTV accurately with current imaging techniques.
It is therefore standard practice to define a CTV that includes the whole prostate
and any possible extracapsular extension, with either the base of, or the entire
seminal vesicles. The risk of involvement of the seminal vesicles is defined using the
Roach formulae and target volume chosen accordingly.
Low risk of seminal vesicle involvement

PSA [(GS 6) 10] 15 per cent
CTV to include prostate base of seminal vesicles.
Moderate/high risk of seminal vesicle involvement

T3 or (PSA [(GS 6) 10]) 15 per cent
CTV to include prostate and seminal vesicles.
Prostate outlining starts on the mid gland slice along the fat plane between the
prostate and pelvic floor muscles and along Denovilliers’ fascia posteriorly. Defining
the apex can be difficult and this can be aided with reference to diagnostic MR scans
or by image registration as discussed above. The base of the seminal vesicles is
included in the CTV for all patients. This is defined as 1–2 cm of central seminal
vesicles proximal to the base of the prostate often at the same level as the middle lobe
that bulges into the bladder. When the entire seminal vesicles are to be included
in the CTV, the distal ends may have to be excluded if the seminal vesicles wrap
around the prostate, to keep the rectal dose within safe limits. Studies have shown
that 90 per cent of seminal vesicle involvement is limited to the proximal 2 cm.
The PTV is defined with a 3D margin around the CTV to include an internal
margin accounting for physiological variations in the shape, position and size of the
340
prostate, and a set-up margin to compensate for uncertainties in patient position
and set-up during planning and treatment. The set-up margins can be measured
with verification studies and quality assurance programmes. The standard margin is
10 mm grown isotropically around the CTV. To limit the dose to the rectum, the
posterior margin is reduced to 5 mm if verification studies allow, and is reduced
further for a phase 2 volume when needed, to keep within rectal dose constraints.
If the rectal volume cannot be reduced to 4 cm in diameter at the base of the
prostate, the margins from CTV to PTV may need to be increased posteriorly.
Image guidance techniques discussed later will allow further safe reduction of
these margins.
Single phase
CTV 74 Prostate base of the seminal vesicles (SV) (or whole SV).
PTV 74 CTV74 10 mm sup/inf/right/left/ant and 5–10 mm post.
Two phase
Phase 1
CTV 56 Prostate base SV (or whole SV).
Phase 2
CTV 74 Prostate only.
■ Pelvic lymph nodes

Selected patients with a risk of lymph node involvement 15–35 per cent, according
to the Roach formula, may benefit from pelvic lymph node irradiation.
Roach formula
Risk of lymph node involvement 2/3 PSA [(GS 6) 10]
Clinical target volume for these patients includes prostate, seminal vesicles and
the common, internal and external iliac, presacral, hypogastric and obturator lymph
nodes. The lymph node areas should be outlined according to published guidelines
and using contrast-enhanced vessels and a pelvic lymph node atlas (Fig. 28.6).
The prostate and seminal vesicles can be outlined as CTV-T and the lymph node
areas as CTV-N as follows.
Phase 1 Pelvis and prostate

CTV T 46 Prostate and SV.
CTV N 46 Lymph nodes areas.
PTV TN 46 CTV T 46 10 mm sup/inf/right/left/ant and 5–10 mm post.
CTV N 46 5–10 mm.
Phase 2 Prostate SV
CTV T 56 Prostate SV.
PTV 56 CTV T 56 10 mm sup/inf/right/left/ant and 5–10 mm post.
341
PROSTATE
Figure 28.6 CTV and PTV for pelvic node

irradiation shown on three level axial CT
scans (blood vessels – yellow).
Phase 3 Prostate
CTV T 74 Prostate only.
PTV 74 CTV T 74 10 mm sup/inf/right/left/ant and 2–5 mm post.
■ Prostate bed
After prostatectomy there is no GTV. The CTV is the prostate bed as best defined
on the CT planning scan. The histopathology report of the prostatectomy specimen
and preoperative MRI can be helpful. The CTV is based on an estimation of the
location of the prostate preoperatively and is centred on the vesico-urethral junction.
Surgical clips within the prostate bed are included in the CTV but those in the
anatomical position of vessels are not. The preoperative location of the seminal
vesicles is included if pathologically involved or the risk of seminal vesicle
involvement according to the Roach formula is 15 per cent.
342
Dose solutions
CTV
Inferior border
5 mm cranial to superior border of the penile bulb.
Anterior border
Posterior aspect of symphysis pubis (2 cm above the vesico-urethral anastomosis).
Posterior 1/3 of bladder wall (2 cm above anastomosis).
Posterior border
Anterior rectal wall.
Lateral border
Medial border of obturator internus and levator ani muscles.
Superior border
Base of SV if uninvolved and risk 15 per cent.
Distal ends of SV if involved or risk 15 per cent.
The PTV is the CTV 1 cm isotropically.
■ OAR
The OAR are the rectum, bladder, nerves of the prostatic plexus lying adjacent to the
penile bulb, small bowel and femoral heads. The rectum is outlined from the inferior
level of the ischial tuberosities and at least 1 cm below the PTV to the recto-sigmoid
junction above the PTV to give a length of approximately 12 cm. Consideration of
small bowel in the target volume is important when pelvic nodes are treated.
Dose solutions
■ Conformal
Using forward planning, an optimal dose distribution is calculated. Beams with
customised MLC shielding are chosen to include the PTV within the 95 per cent
isodose and minimise the dose to the OAR. The beams, dose weighting and
wedging are optimised. A standard approach is to use a technique with an anterior
and two wedged posterior oblique beams as shown in Fig. 28.7. In some cases,
wedged lateral beams may spare more normal rectum, and four or six coplanar
beam arrangements may reduce doses to the OAR further. When treating the
pelvic lymph nodes, an anterior and two wedged lateral beams are used, with a
posterior beam if necessary.
The dose to OAR is assessed by DVHs. Plans are reviewed to minimise hot-spots
in OAR. The acceptable dose constraints for OAR are shown in Table 4.1 (p. 49).
■ Complex
The increased conformity of IMRT using inverse planning techniques ensures
better coverage of the target volume, sparing of OAR, and safer dose escalation. It
has significant advantages when irradiating the pelvic lymph nodes with the ability
to conform to the nodes and spare normal bowel in the pelvis (Fig. 28.8).
343
PROSTATE
(a)
(b)
Figure 28.7 (a) Axial and (b) sagittal dose distributions of conformal treatment for
prostate cancer. Bladder yellow; rectum brown.
Figure 28.8 Tomotherapy IMRT plan for treatment of (a) pelvic lymph nodes and (b)
prostate cancer.
344
Dose-fractionation
There are many IMRT planning and delivery systems. Fixed delivery ‘step and
shoot’ techniques use five equally spaced coplanar fields with the MLC set in
various positions as defined by the inverse treatment plan to create the beam
modulation needed. Alternatively, dynamic therapy or tomotherapy may be used.
■ Conventional
Palliative treatment to the prostate may be given using opposing anterior and
posterior beams and MLC shielding, planned by CT virtual simulation or with a
conventional simulator.
Dose-fractionation
Prostate SV
Single phase
Two phase
Phase 1: prostate SV 56 Gy in 28 daily fractions given in 51⁄2 weeks.
Phase 2: prostate 18 Gy in 9 daily fractions given in 11–13 days
With conformal planning and IMRT the dose to the prostate SV can be
increased to 64 Gy with a 10 Gy prostate boost, if OAR dose constraints are met.
Prostate SV pelvis
Phase 1: Pelvic nodes prostate SV 46 Gy in 23 daily fractions given in 41⁄2
weeks.
Phase 2: Prostate SV 10 Gy in 5 daily fractions given in 1 week.
Phase 3: Prostate 18 Gy in 9 daily fractions given in 11–13 days.
Alternative fractionation schedules (to the prostate alone)

50 Gy in 16 daily fractions of 3.1 Gy (Manchester regimen) given in 31⁄2 weeks.
Prostate bed
Extensive symptomatic prostatic disease
36 Gy in 6 weekly fractions of 6 Gy given in 6 weeks.
Bone pain
20 Gy in 5 daily fractions in 1 week.
345
PROSTATE
Lymph node or visceral disease

20 Gy in 5 daily fractions given in 1week.
Breast bud radiotherapy

Prophylaxis 8 Gy single fraction.
Palliation 12 Gy in 2 daily fractions given in 2 days.

The patient is treated supine with a comfortably full bladder and after rectal voiding.
Immobilisation systems are used with anterior and lateral laser lights to align midline
and lateral skin tattoos to prevent lateral rotation. The isocentre is marked with
reference to the anterior tattoo over the pubic symphysis.
Severe skin reactions are rare but may occur over the sacrum and natal cleft. If
loose stools develop, a low residue diet is advised to prevent diarrhoea. Loperamide
hydrochloride may be used for diarrhoea with care to avoid constipation. Mild
proctitis and tenesmus are common and if severe, may be treated with steroid or
local anaesthetic suppositories. Urinary symptoms of frequency, urgency and slow
stream are common. Patients should remain well hydrated and a mid-stream urine
should be examined to exclude infection. Patients with pre-existing obstructive
problems may be helped by α-blockers such as tamsulosin. Cystitis may be helped
by potassium citrate, cranberry juice or simple analgesia.
After radiotherapy, the incidence of grade 3 chronic intestinal sequelae requiring
hospitalisation for diagnosis and/or minimal intervention is 3 per cent. Most cases
occur within 3–4 years of treatment. A gastroenterologist should investigate rectal
bleeding, diarrhoea, urgency and tenesmus. Argon plasma coagulation (APC), a non-
contact thermal coagulation technique, applied endoscopically, can be used to
treat chronic radiation proctitis. In very severe cases, formalin or a combination of
APC and topical formalin can be useful.
The incidence of impotence is 30–50 per cent and may be reduced by excluding
the penile bulb from treatment.
Verification
Electronic portal images are taken for the first 3–5 days of treatment and compared
with DRRs from the planning CT scan, using bony landmarks, the beam edges and
centre. The set-up error is calculated from any differences and may be corrected
according to local protocol.
IGRT may improve accuracy of treatment delivery. The use of radio-opaque fiducial
markers within the prostate allows variations arising from prostate movement to be
identified and incorporated into the local protocol. Commercial systems such as
Acculoc are available and have been shown to improve targeting of dose. Less
invasive methods include the BATS ultrasound system and cone-beam CT scans
taken before treatment to localise the prostate.
Diodes or TLD measurements on the first treatment day are used to measure
dose delivered.
346
Prostate brachytherapy
Prostate brachytherapy is an excellent option for patients with localised low risk
prostate cancer. Patients with intermediate risk prostate cancer can be treated with
brachytherapy alone, but as the risk of disease outside the prostate capsule increases,
EBRT alone or in combination with brachytherapy should be considered.
Indications for brachytherapy:
■ life expectancy 10 years
■ biopsy-confirmed adenocarcinoma prostate
■ low risk disease: T1–T2a, PSA 10, GS 6
■ intermediate risk disease: T1–T2a, PSA 15, GS 3 4 (low volume)
■ prostate volume 50 mL (dynamic techniques can treat up to 90 mL).
It is important to select patients with no significant urinary outflow obstruction since
they are at increased risk of urinary retention and morbidity following brachytherapy.
Ideally patients should have an international prostate symptom score (IPSS) of 12
and urine flow rate Qmax of 15 mL/s.
■ LDR permanent seed brachytherapy

The most frequently used isotope is iodine-125. An alternative is palladium-103,
which has a higher dose rate. The Seattle technique involves a planning study
under general anaesthesia with the patient in the lithotomy position. A transrectal
ultrasound (TRUS) probe is inserted into the rectum attached to a stepping unit
that can advance or retract the probe (Fig. 28.9). Attached to the TRUS is a
template, the coordinates of which are transposed onto the ultrasound images of
the prostate. Serial ultrasound sections with 5 mm slices of the prostate from the
base to the apex are captured onto the planning computer. The prostate, urethra
(identified with the aid of aerosolised jelly) and rectum are outlined on the ultrasound
images. A margin of 2–3 mm is added in all directions except posteriorly to form
the PTV. A computer program is used to calculate the exact number and position of
seeds required. It will also calculate where needles must be inserted in the template
to achieve the desired distribution of seeds in the prostate. For treatment a second
anaesthetic is needed, and with the patient and prostate in exactly the same position
as in the planning study, needles are inserted through the template, and seeds
implanted into the prostate using a stranded or loose seed technique.
It is difficult to replicate the planning position and the prostate moves when
needles are inserted. To overcome this, the technique has now been modified into
a single anaesthetic, single stage inverse planned procedure. Intraoperative dosimetry
can now be planned from needle location. The Pötter technique involves inserting
needles into the prostate at 1 cm intervals with a peripheral:central ratio of 3:1. The
needle position is registered on the planning software with the ultrasound slices as
above. Once the prostate volume has been outlined, the computer generates a plan
within 45 s. Advanced dynamic dose calculation can capture seed position in real
time continuously updating dosimetry as seeds are inserted. Any deviations from
the plan can be corrected in real time before the end of the procedure.
Current standard reporting of dose is based on dosimetry from a CT scan
performed 4 weeks after seed implantation when oedema has settled. Doses to the
prostate, rectum and urethra are calculated. The V100 is the percentage of the
347
PROSTATE
Needle
Prostate
Rectum
Probe
Figure 28.9 Technique for insertion of radioactive seeds into the prostate. Courtesy of
Prostate Cancer Charity.
prostate volume that has received the 100 per cent prescription dose and should
be 90 per cent. The D90 is the dose received by 90 per cent of the prostate
volume and should be at least 90 per cent of the prescribed dose. The prostate
V150 and V200 give a measure of the homogeneity of the dose distribution. The
maximum rectal dose should be kept as low as possible. If it is less than 140 per cent
of the treatment dose, the risk of proctitis is less than 1.2 per cent. The maximum
urethral dose should be kept as low as possible, and should be less than 150 per cent
of the treatment dose.
Using the same technique, LDR brachytherapy can be used for a boost treatment
following EBRT to the prostate and pelvis.
■ HDR afterloading brachytherapy

HDR afterloading brachytherapy is performed under general anaesthesia using a
stepping iridium-192 HDR machine. In contrast to LDR permanent seed implants,
there is no calculated pre- or post-implant dosimetry, and dosimetry is defined by
catheter position and dwell times. Large fraction sizes can be delivered which may
have a biological advantage in prostate cancer. The technique reduces radiation
exposure to the oncologist and other staff. The main use is as a boost treatment in
conjunction with external beam radiotherapy. Afterloading catheters are inserted
into the prostate through the perineum under TRUS guidance using a template
348
that is sutured or fixed with adhesive to the patient. A CT scan is performed of the
patient with the catheters in place and the dosimetry calculated on a 3D planning
system. Several high dose fractions are then delivered in the HDR afterloading
room and the patient remains an inpatient with an indwelling urinary catheter until
the end of treatment. HDR monotherapy is performed in exactly the same way.
Monotherapy
LDR 145 Gy (I-125), 125 Gy (Pa-103).
HDR 34–36 Gy in 4 fractions.
Combination therapy
LDR
EBRT: prostate pelvis 45 Gy in 25 daily fractions of 1.8 Gy given in 5 weeks.
Brachytherapy: 110 Gy (I-125), 90 Gy (Pa-103).
HDR
EBRT: prostate pelvis 45 Gy in 25 daily fractions of 1.8 Gy given in 5 weeks.
Brachytherapy: 17 Gy (Ir-192) in 2 fractions given in 24 hours.
key trials
Bolla M, Collette L, Blank L et al. (2002) Long-term results with immediate

androgen suppression and external irradiation in patients with locally advanced
prostate cancer (an EORTC study): a phase III randomised trial. Lancet 360:
103–6.
Bolla M, van Poppel H, Collette L et al. (2005) Postoperative radiotherapy after
radical prostatectomy: a randomised controlled trial (EORTC trial 22911). Lancet
366: 572–8.
D’Amico AV, Manola J, Loffredo M et al. (2004) 6-month androgen suppression
plus radiation therapy vs. radiation therapy alone for patients with clinically
localised prostate cancer: a randomised controlled trial. JAMA 292: 821–7.
Horwitz EM, Bae K, Hanks GE et al. (2008) 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
26: 2497–504.
Lawton CA, DeSilvio M, Roach M 3rd et al. (2007) An update of the phase III
trial comparing whole pelvic to prostate only radiotherapy and neoadjuvant to
adjuvant total androgen suppression: updated analysis of RTOG 94–13, with
emphasis on unexpected hormone/radiation interactions. Int J Radiat Oncol Biol
Phys 69: 646–55.
Pollack A, Zagars GK, Starkschall G et al. (2002) Prostate cancer radiation dose
response: results of the MD Anderson phase III randomised trial. Int J Radiat
349
PROSTATE
Information sources
de Crevoisier R, Tucker SL, Dong L et al. (2005) Increased risk of biochemical and local failure in
patients with distended rectum on the planning CT for prostate cancer radiotherapy. Int J
Radiat Oncol Biol Phys 62: 965–73.
Jackson ASN, Sohaiby SA, Staffurth JN et al. (2006) Distribution of lymph nodes in men with
prostatic adenocarcinoma and lymphadenopathy at presentation: a retrospective radiological
review and implications for prostate and pelvis radiotherapy. Clin Oncol 18: 109–16.
Kupelian PA, Langen KM, Willoughby TR et al. (2008) Image-guided radiotherapy for localized
prostate cancer: treating a moving target. Semin Radiat Oncol 18: 58–66.
Memorial Sloane Kettering Prostate Cancer Nomogram. Available at www.mskcc.org
National Institute for Health and Clinical Excellence. (2008) Prostate Cancer: Diagnosis and
Treatment. Clinical guidance 58. London, NICE. Available at www.nice.org.uk (accessed
9 December 2008).
Prostate Brachytherapy Video. Available at www.nyprostate.org
Prostate cancer risk management programme. Available at www.cancerscreening.nhs.uk/prostate/
Roach M 3rd, Hanks G, Thames H Jr et al. (2006) Defining biochemical failure following radiotherapy
with or without hormonal therapy in men with clinically localised prostate cancer:
recommendations of the RTOG-ASTRO Phoenix Consensus Conference. Int J Radiat Oncol
Biol Phys 65: 965–74.
Taylor A, Rockall AG, Powell ME (2007) An atlas of the pelvic lymph node regions to aid
radiotherapy target volume definition. Clin Oncol 19: 542–50.
350
Bladder 29

In the UK there are approximately 10 000 new cases of bladder cancer a year. The
majority (85 per cent) are superficial bladder cancers confined to the mucosa with
a 5-year survival rate of 80–90 per cent. However, 70 per cent recur and of these,
10–15 per cent develop muscle invasive bladder cancers within 5 years. Carcinoma
in situ is usually more aggressive with up to 50 per cent progressing to muscle
invasive cancers within 5 years. In the past 30 years, 5-year survival rates for muscle
invasive bladder cancer have increased from 40 per cent to 60 per cent.
There has been a decrease in the use of radiotherapy as a single modality
treatment and an increase in combined radical cystectomy and radiotherapy. There
are no prospective randomised trials of radical cystectomy versus radiotherapy.
When the effects of selection bias, stage migration due to clinical versus pathological
staging and differences in prognostic factors are taken into account, there is little
current evidence of a benefit of surgery over radiotherapy. Despite this, many still
consider radical cystectomy to be the gold standard. Radical cystectomy involves en bloc
removal of the pelvic lymph nodes and pelvic organs anterior to the rectum–bladder,
urachus, and visceral peritoneum, with prostate and seminal vesicles in men and
ovaries, fallopian tubes, uterus, cervix and vaginal cuff in women. Surgical
techniques have improved including use of lower urinary tract orthotopic
reconstructions. The best surgical series report 5-year recurrence free rates of
80 per cent for T2, N0, 55 per cent for T3, N0 and 30 per cent for Tx, N1 bladder
cancer. Review of the world literature on outcomes following EBRT reports 5-year
survival of 10–59 per cent for T2, 10–38 per cent for T3 and 0–16 per cent for T4
bladder cancers. A recent UK study has shown 5-year survival rates of 56.8 per cent
after radical radiotherapy and 53.4 per cent after radical cystectomy in patients
with invasive cancer, even though the patients undergoing radical cystectomy were
significantly younger.
Ninety per cent of bladder cancers are urothelial (transitional cell) carcinomas
(TCC), and 5 per cent are SCC. In areas where schistosomiasis is endemic, SCC
constitutes up to 80 per cent of cases. Fewer than 5 per cent of cases are primary
adenocarcinoma arising from the urachal remnant, small cell carcinoma, sarcoma,
carcinosarcoma, lymphoma or melanoma. TCC is graded as G1 (well differentiated),
G2 (moderately differentiated) and G3 (poorly differentiated). The tumour grade is
an important predictor of tumour behaviour.
Bladder cancer is staged using the AJCC and TNM staging based on data from EUA
and imaging. Tumours originate in the bladder epithelium and infiltrate deeply into the
muscle layers, penetrating through the bladder wall into perivesical fat and adjacent
pelvic organs. Lymphatic spread is first to the hypogastric, obturator, internal, external
351
BLADDER
and common iliac lymph nodes and thence to the para-aortic and inguinal nodes. The
risk of lymph node metastases relates to the depth of tumour invasion: 20 per cent for
T1, 30 per cent for T2a and 60 per cent for T2b. More than a third of patients treated
radically develop distant metastases within 18 months. Stage and grade of the tumour
are the most important prognostic factors. The presence of extravesical spread (T3b)
and lymph node involvement are poor prognostic factors. Nodal involvement is usually
considered as an indication for palliative treatment in the UK.
Radical radiotherapy is an option for patients with T2a–T3b bladder cancer alone
or in combination with chemotherapy. Good prognostic features are younger age,
lower tumour stage, no pelvic lymph node involvement, normal renal function, no
hydronephrosis, normal haemoglobin and complete transurethral resection of
bladder tumour (TURBT). Patients must have good bladder function, no history
of previous pelvic radiotherapy, surgery or infections and no inflammatory bowel
disease. Cystectomy is preferable in patients with poor bladder function, a tumour
within a bladder diverticulum, diffuse bladder involvement, multiple tumours,
extensive carcinoma in situ, hydronephrosis, large tumours 5 cm with
extravesical mass, and for SCC and adenocarcinoma. With good patient selection
and accurate staging, radical radiotherapy is a good non-surgical curative option
for invasive bladder cancer with bladder preservation.
Radical radiotherapy has been considered in the past for pT1 G3 high risk
superficial bladder cancer as an alternative to cystectomy. A recent RCT has shown
radiotherapy to be no better than conservative treatment in these patients.
There is no proven benefit to irradiating the pelvic lymph nodes, and chemotherapy,
which is being increasingly used, may be adequate to treat metastatic disease when
local radiotherapy is given.
Studies of preoperative neoadjuvant radiotherapy have shown no benefit and there
is no evidence for postoperative adjuvant radiotherapy, which is avoided because
of the high risk of bowel toxicity.
Palliative radiotherapy is useful for the relief of symptoms such as haematuria and
pain for patients with T4 bladder tumours, pelvic nodal disease, bone and other
distant metastases. Palliative short courses of pelvic radiotherapy may be appropriate
for elderly patients with T2 and T3 bladder cancer who have significant comorbidities
precluding radical treatment. They can be managed with repeated transurethral
resection of a bladder tumour (TURBT), and palliative radiotherapy to the
bladder and pelvis to control symptoms.

■ Neoadjuvant chemotherapy
Neoadjuvant chemotherapy before radical cystectomy or radiotherapy has been given
with the aim of improving bladder preservation rates and survival. A meta-analysis has
352
shown that neoadjuvant platinum-based chemotherapy increases 5-year overall
survival significantly from 45 per cent to 50 per cent.
Fit patients with adequate renal function and no hydronephrosis should be
considered for neoadjuvant chemotherapy.
■ Concurrent chemoradiotherapy
Concurrent chemoradiotherapy is currently being investigated and appropriate
patients, such as those with solitary TCC tumours, no extensive carcinoma in situ
and good PS, should be entered into clinical trials.
■ Adjuvant chemotherapy
Pathological staging is used to select patients for adjuvant chemotherapy after
surgery. However this sequencing delays the systemic treatment of micrometastatic
disease. There is no definite proof of benefit yet but studies are continuing.

The bladder occupies the anterior portion of the pelvic cavity just superior and
posterior to the pubic bone (Fig. 29.1).The base of the bladder is separated from
the rectum by the vas deferens, seminal vesicles and ureters in men, and by the
uterus and vagina in women. As the bladder distends, the neck remains fixed and
the dome rises into the pelvic cavity. The bladder mucosa is lined with transitional
epithelium. It appears smooth when full and has numerous folds when empty.
Rectum
Bladder
Seminal
Symphysis
vesicle
pubis
Prostate Denonvilliers
fascia
Bulb of penis
Figure 29.1 Sagittal section though pelvis to show prostate and bladder anatomy.

Patients are investigated with cystoscopy, TURBT, and bimanual pelvic examination
under anaesthesia and urine cytology. Biopsies of the tumour should include
muscle, and define the histological type, grade and depth of invasion. Random
biopsies identify the presence of carcinoma in situ.
Diagrams of findings showing tumour location and extent at cystoscopy are
very useful to aid planning. Ultrasound and intravenous urethrogram (IVU) are
353
BLADDER
performed to detect other urothelial tumours and hydro-ureter secondary to

obstruction of the ureteric orifices. MR scans show the local extent of disease in
the pelvis and bladder better than CT. Alternatively, a CT scan of the abdomen
and pelvis can be done. Chest X-ray will exclude large lung metastases. PET-CT is
being investigated and may become the investigation of choice to stage for lymph
node involvement and metastases.
Data acquisition
A CT scan is performed with the patient supine in the treatment position with arms
folded across the chest with ankle supports to stabilise the legs and pelvis, and a knee
support for comfort. The bladder should be emptied immediately before scanning
and treatment, to reduce the volume irradiated and doses to normal tissues. The
rectum should be empty to reduce organ motion and interfractional variations.
A small volume of oral contrast is given 1 h before the planning CT scan to show the
small bowel. The scan is performed with 3–5 mm slices from the lower border of L5
to the inferior border of the ischial tuberosities. An anterior tattoo is placed over the
pubic symphysis and two lateral tattoos over the iliac crests to prevent lateral rotation
with radio-opaque markers used for location on the scan. It is important to confirm
on the planning scan that the patient has been able to empty the bladder. If there is
a persistent large amount of residual urine on repeat CT scans, the patient requires
intervention, which may include medication, a transurethral resection of the prostate
(TURP) or as a last resort a urinary catheter. MRI is beneficial in planning partial but
not whole bladder treatments and MRI/CT fusion may be useful.

The current UK standard is treatment of the whole bladder and any extravesical
component of the tumour.
The GTV is the primary bladder tumour, which is difficult to define on CT
alone but MRI/CT fusion may help. Partial bladder approaches, which treat the
whole bladder to a lower dose and boost the tumour alone, are currently being
investigated. Detailed information from cystoscopy, bladder maps and diagnostic
CT and MRI scans is needed for this technique.
The standard approach is to define the CTV as the GTV (primary tumour and
any extravesical spread) and the whole bladder. In patients with tumours at the
bladder base, the proximal urethra and in men the prostate and prostatic urethra
are included on the CTV.
The PTV is the CTV with a 1.5–2 cm margin (Fig. 29.2). Studies have shown
that the most significant bladder movement is in the cranial and AP directions and
is random in time and direction. Bladder movement has been shown to occur in up
to 60 per cent of patients as a displacement of over 1.5 cm between the bladder wall
and the 95 per cent isodose. It is sensible therefore to grow the CTV by 1.5 cm to
form the PTV around the normal outside bladder wall, but to grow the CTV by
2 cm to form the PTV around the primary bladder tumour and any extravesical
spread. Tumours lying in the cranial part of the bladder are most at risk of moving
out of the target volume and some authors recommend a margin of 3 cm from CTV
to PTV around these bladder tumours. With improved image guidance these
354
Dose solutions
margins can be reduced. Care has to be taken with small bowel that may be stuck
down to bladder tumours with extravesical spread. In these cases where small bowel
is in close contact with tumour, a repeat planning CT scan can be performed to
check whether the small bowel is mobile or the dose needs to be reduced.
Figure 29.2 CT axial slice

with CTV whole bladder,
PTV and OAR. Rectum,
yellow; femoral heads,
green.
OAR should be outlined including rectum, femoral heads and small bowel.
Recommended dose constraints are: rectum V50 60 per cent, V60 50 per cent;
femoral heads V50 50 per cent; small bowel V45 250 cm3.
For palliation of T4 tumours and pelvic nodal disease, the PTV must encompass
the primary disease and its extension into the pelvis. If this volume is excessively
large it can be reduced to cover only the area causing symptoms.
Dose solutions
■ Conformal
Forward planning is used to optimise a 3D conformal plan, usually with10–15 MV
photon beams, such as an anterior and two-wedged lateral or posterior oblique
beams. The angle between the posterior oblique beams should be chosen to
minimise dose to the rectum (Fig. 29.3).
Figure 29.3 Axial CT slice showing

conformal plan for whole bladder
irradiation.
To minimise normal tissue irradiation, conformal MLC shielding is used. Doses

to the rectum, femoral heads and small bowel should be kept as low as possible
355
BLADDER
(see above). This is assessed by DVH constraints as shown in Figure 29.4, and the
volume of small bowel receiving 45 Gy kept to a minimum.
Dose (Gy) PTV

Dose 83.3%
10.0
12.0
14.0
16.0
18.0
20.0
22.0
24.0
26.0
28.0
30.0
32.0
34.0
36.0
38.0
40.0
42.0
44.0
56.0
58.0
60.0
0.0
2.0
4.0
6.0
8.0
(45.83 Gy)
Volume 100.00%
100.0 100.0
90.0 90.0
PTV
80.0 80.0
70.0 70.0
Femoral
60.0 60.0
heads
50.0 Rectum 50.0
40.0 40.0
30.0 30.0
20.0 20.0
10.0 10.0
0.0 0.0
100.0
0.0
3.0
6.0
9.0
12.0
15.0
18.0
21.0
24.0
27.0
30.0
33.0
36.0
39.0
42.0
45.0
48.0
51.0
54.0
57.0
60.0
63.0
66.0
69.0
72.0
75.0
78.0
81.0
84.0
87.0
90.0
93.0
96.0
102.0
105.0
108.0
Relative dose (%)
Figure 29.4 DVH for conformal plan for bladder cancer showing PTV, rectum and femoral
heads.
With CT planning, a three-field technique as described above can be used for

palliative treatments. In patients with extensive disease involving the rectum and
pelvis, AP opposing beams can be designed using virtual simulation (Fig. 29.5).
Figure 29.5 Virtual CT

simulation showing DRR
with volume for palliative
treatment of bladder
cancer.
■ Complex
IMRT with inverse planning has the potential to reduce the dose to normal tissues,
allow the delivery of a synchronous boost needed for partial bladder irradiation
and permit dose escalation to the tumour. However, IMRT for this tumour site
requires excellent immobilisation, with IGRT to locate and minimise PTV at the
time of treatment. It has been shown that without IGRT an isotropic margin of
3 cm is required, but with IGRT this can be reduced to 12 mm.
356
Verification
■ Conventional
Where CT planning is not available, the treatment can be simulated conventionally
and the bladder visualised with contrast inserted into the bladder via an indwelling
urinary catheter.
Dose-fractionation
■ Curative
Whole bladder
■ Palliative
Whole bladder
21 Gy in 3 fractions given on alternate days in 1 week
or
36 Gy in 6 fractions of 6 Gy given once weekly for 6 weeks.
In most cases 21 Gy in 3 fractions on alternate days over 1 week is as effective
as longer schedules for palliation as shown by the MRC BA09 trial. A weekly
hypofractionated regimen of 6 Gy weekly for 6 fractions has been shown to
effectively palliate symptoms in patients unfit for radical treatment and may be
preferred by some patients.

Before radiation starts the patient should be made as fit as possible. Urinary
infection should be treated, anaemia should be corrected to haemoglobin
12 g/dL and a low residue diet advised. Radiation cystitis is common; infection
should always be excluded and a high fluid intake advised. Catheterisation should
be avoided if possible to minimise the risk of infection. Mild proctitis and lethargy
are common. Late side effects include fibrosis and shrinkage of the bladder,
haematuria due to bladder telangiectasia, late bowel damage, vaginal dryness and
stenosis in women and impotence in men.
Verification
The patient is immobilised as described above with an empty bladder and rectum
and aligned using an anterior laser to check the midline and two lateral lasers to
prevent rotation. The isocentre is marked with reference to the tattoo over the
pubic symphysis for set-up. If an isocentre shift has to be performed, this is verified
by comparison of EPI or simulator films with DRRs. TLD or diode measurement
of dose delivered is performed on the first day of treatment.
The bladder is mobile and can change shape as well as position during treatment.
Verification is therefore very important. The current standard is EPI comparing
bony anatomy with the AP and lateral DRRs daily for the first 3–5 days, and then
357
BLADDER
once weekly correcting for systematic errors. Verification that allows visualisation of
soft tissues at the time of treatment delivery, such as kV cone beam imaging, MV
imaging with tomotherapy and MV imaging of gold seed markers cystoscopically
implanted into the bladder, will significantly improve the accuracy of radiotherapy
to the bladder. ART, which modifies treatment plans to account for variations in
individual patients, or gated radiotherapy based on delivery of treatment only when
the bladder is in the correct position may be useful techniques.
key trials
Advanced Bladder Cancer Meta-analysis Collaboration (2003) Neoadjuvant

chemotherapy in invasive bladder cancer: a systematic review and meta-
analysis. Lancet 361: 1927–34.
Advanced Bladder Cancer (ABC) Meta-analysis Collaboration (2005) Adjuvant
chemotherapy in invasive bladder cancer: a systematic review and meta-analysis
of individual patient data. Advanced Bladder Cancer (ABC) Meta-analysis
Collaboration. Eur Urol 48: 189–99.
Duchesne GM, Bolger JJ, Griffiths GO et al. (2000) A randomised trial of
hypofractionated schedules of palliative radiotherapy in the management of
bladder carcinoma: results of Medical Research Council trial BA09. Int J Radiat
Harland SJ, Kynaston H, Grigor K et al. (2007) A randomised trial of radical
radiotherapy for the management of pT1G3 NXM0 transitional cell carcinoma of
the bladder. J Urol 178: 807–13.
Tester W, Caplan R, Heaney J (1996) Neoadjuvant combined modality program
with selective organ preservation for invasive bladder cancer: results of Radiation
Therapy Oncology Group phase II trial 8802. J Clin Oncol 14: 119–26.
Information sources
Alonzi R, Hoskin P (2005) Novel therapies in bladder cancer. Clin Oncol 17: 524–38.
Bladder Cancer WebCafé. Available at: www.blcwebcafe.org
Harris SJ, Buchanan RB (1998) An audit and evaluation of bladder movements during radical
radiotherapy. Clin Oncol 10: 262–4.
Kotwal S, Choudhury A, Johnston C et al. (2008) Similar treatment outcomes for radical cystectomy
and radical radiotherapy in invasive bladder cancer treated at a United Kingdom specialist
treatment centre. Int J Radiat Oncol Biol Phys 70: 456–63.
Logue J, McBain CA (2005) Radiation therapy for muscle-invasive bladder cancer: treatment
planning and delivery. Clin Oncol 17: 508–13.
McLaren DB, Morrey D, Mason MD (1997) Hypofractionated radiotherapy for muscle invasive
bladder cancer in the elderly. Radiother Oncol 43: 171–4.
NCRI Trials Portfolio. Available at: www.publincrn.org.uk
Redpath AT, Muren LP (2006) CT-guided intensity-modulated radiotherapy for bladder cancer:
isocentre shifts, margins and their impact on target dose. Radiother Oncol 81: 276–83.
Sengelov L, Von der Maase H (1999) Radiotherapy in bladder cancer. Radiother Oncol 52: 1–14.
Shipley WU, Kaufman DS, Tester WJ et al. (2003) Overview of bladder cancer trials in the Radiation
Therapy Oncology Group. Cancer 97(8 suppl): 2115–19.
358
Testis 30

Testicular cancer is the most common malignancy in young men, affecting about
2000 a year in the UK. Ninety per cent of patients present with low stage I–IIB
disease. Eighty per cent have clinical stage I disease confined to the testis with
normalised tumour markers after orchidectomy. The cure rate for clinical stage I
testis cancer is expected to be 100 per cent. To achieve this, patients should be
managed by experienced health professionals, using consensus guidelines such as
those from the European Germ Cell Cancer Consensus Group (EGCCCG, see
below), which outline the current standards of care.
Over 95 per cent of testicular cancers are germ cell tumours. Half of these are
seminoma and half are non-seminomatous germ cell tumours (NSGCTs). NSGCTs
are usually a mixture of teratoma, embryonal carcinoma, teratocarcinoma,
choriocarcinoma and yolk sac tumour. Other rare testicular tumours include non-
Hodgkin lymphoma, Sertoli and Leydig cell tumours. The treatment of non-
Hodgkin lymphoma is discussed elsewhere and the management of Sertoli and
Leydig cell tumours is surgical.
In seminoma the pathological risk factors for recurrence are tumour size 4 cm
and rete testis invasion, whereas in NSGCT these are vascular and lymphatic
invasion, the presence of embryonal carcinoma and the absence of yolk sac tumour.
Patients are staged according to the TNM classification of the UICC, and also
according to the International Germ Cell Cancer Collaboration Group
(IGCCCG) prognostic groups.
■ Seminoma
In stage I seminoma there are now three standard options for management
following radical orchidectomy.
■ surveillance
■ para-aortic nodal radiotherapy (consider ‘dog leg’ radiotherapy to include the
ipsilateral iliac and inguinal nodes in patients who have had previous
inguinal/scrotal surgery or have distortion of lymphatic pathways)
■ single course of carboplatin chemotherapy.
Surveillance studies have shown that only 15–20 per cent of patients relapse.
Using the prognostic factors of rete testis involvement and tumour size 4 cm, the
relapse rate is 32 per cent in patients with both risk factors, 15 per cent in patients
with one risk factor and 12 per cent with no risk factors. Both adjuvant para-aortic
nodal radiotherapy and a single course of carboplatin reduce the risk of relapse to
359
TESTIS
3–4 per cent. Surveillance involves intensive and prolonged restaging investigations
over up to 10 years and is reliant on patient compliance, but may be useful where
immediate treatment is contraindicated.
The following trials have influenced practice:
■ MRC TE10: this trial randomised patients between treatment with para-aortic
nodal radiotherapy or a dog leg field to a dose of 30 Gy in 15 fractions with no
significant differences in outcome.
■ MRC TE18: this trial randomised patients undergoing para-aortic nodal
radiotherapy to receive 30 Gy in 15 fractions or 2 Gy in 10 fractions with no
significant differences in outcome, but less toxicity with the lower dose.
■ MRC TE19: this trial randomised patients to receive one cycle of carboplatin
(AUC 7) or para-aortic nodal radiotherapy. At 7 years’ follow-up, there was no
difference in outcome.
In stage IIA seminoma, radiotherapy offers a high cure rate of 85–90 per cent.
The current guidelines recommend radiotherapy to the para-aortic and ipsilateral
iliac nodes to a dose of 30 Gy. In stage IIB seminoma, most patients receive
chemotherapy, but current guidelines still consider an option of radiotherapy to
the para-aortic and ipsilateral iliac nodes to a dose of 36 Gy.
Patients who relapse after initial treatment should be managed by experienced
teams and entered into prospective randomised trials. The incidence of relapse
after chemotherapy alone is low and radiotherapy may have a role for some
patients with small and localised relapses. Radiotherapy may be used palliatively in
patients with chemo-resistant disease.
Patients with testicular intraepithelial neoplasia (TIN) are treated with a
surveillance strategy or orchidectomy in the UK. In Europe, radiotherapy to the
testis of 20 Gy in 10 fractions over 2 weeks is also used.
■ NSGCT
Stage I patients are currently managed by orchidectomy and surveillance. Patients
with more advanced or relapsed disease are treated with chemotherapy.
Radiotherapy can have a palliative role in chemo-resistant disease for bulky
inoperable disease, or cerebral, lymph node or bone metastases.

Regional spread is predominantly lymphatic. From the testis it follows the
testicular arteries to the para-aortic, renal hilar and retro-crural lymph nodes with
involvement of the contralateral nodes occurring in 15–20 per cent. The typical
‘landing zone’ of left sided tumours is the left renal hilar and para-aortic region,
and of right sided tumours is the inter-aortocaval and paracardiac area, as shown
in Fig. 30.1.
The lymphatics from the scrotum drain to the inguinal lymph nodes but may be
distorted by hernia repair, orchidopexy, scrotal surgery or pelvic infection.
Malignant teratomas also have a propensity for early vascular spread to the lungs
and liver.
360
Renal hilar
nodes
Para-aortic
nodes
Iliac nodes
Figure 30.1 Lymphatic drainage

of the testis.

The standard assessment should include a full clinical examination of the
contralateral testis, lymph nodes areas (especially supraclavicular nodes and
abdomen) and breasts to exclude gynaecomastia. Routine tests should include
tumour markers AFP, β-hCG and lactate dehydrogenase, a baseline luteinising
hormone (LH), follicle-stimulating hormone (FSH) and testosterone, chest X-ray,
and CT scan of the chest, abdomen and pelvis with intravenous contrast. MRI is
equally sensitive but less commonly used except for imaging brain metastases.
FDG-PET is useful in assessing the significance of residual masses shown on CT
after treatment. Patients should be advised on sperm storage.
Data acquisition
A CT scan of the abdomen is taken with the patient lying supine in the treatment
position with arms by his sides. The head and legs are restrained with head rest,
ankle and knee supports. Midline and lateral tattoos are used with laser lights to
align the patient and prevent lateral rotation. Patients can be planned
conventionally in a simulator, or using a CT scan for virtual simulation of the fields,
which gives information on the soft tissue anatomy, and position of the renal pelvis.

■ CT scanning
For stage I, the CTV includes lymph node areas at high risk of microscopic
involvement i.e. para-aortic, renal hilar and retro-crural nodes bilaterally. Normal
nodes are not visible on CT scans but intravenous contrast can be used to enhance
renal blood vessels to help locate hilar nodes. If there has been previous scrotal
surgery, the target volume also includes ipsilateral pelvic and inguinal lymph nodes
(identified by contrast enhanced pelvic blood vessels) and the ipsilateral scrotal sac.
361
TESTIS
For stage II disease, GTV is defined by outlining enlarged (1 cm) lymph
nodes. CTV includes the nodal areas as for stage I, ensuring an adequate margin
around the GTV. Both kidneys must be identified and outlined as critical normal
organs, taking care to exclude a horseshoe kidney. Fig. 30.2 shows how virtual CT
simulation can be used for accurate targeting of lymph nodes and design and
shaping of fields to avoid normal kidneys.
(a) (b)
Figure 30.2 Virtual simulation for para-aortic nodal irradiation for seminoma of left testis.
(a) Anterior DRR with kidneys outlined. (b) Coronal multiplanar reconstruction with MLC
shielding to the kidneys.
■ Conventional simulation
Conventionally, the target volume for para-aortic nodal radiotherapy is defined by
standard field sizes on a simulator:
■ superior: lower border of T10
■ inferior: lower border of L5
■ lateral: to borders of transverse processes of vertebrae with inclusion of ipsilateral

renal hilum.
The above field margins ensure inclusion of the para-aortic, renal hilar and retro-
crural lymph nodes. The kidneys must be identified at planning by an IVU done
in the simulator, and a horseshoe kidney excluded. As much normal kidney as
possible should be excluded from the field using shielding.
In patients where the ipsilateral iliac nodes or ipsilateral iliac and inguinal nodes
are to be treated, a larger rectangular field (‘dog leg’) is defined and shielding added
to protect the kidneys, bladder and bowel as shown in Fig. 30.3. It is important to
ensure adequate inclusion of lymph nodes lying at the mid level of the fifth lumbar
vertebra between the para-aortic and pelvic nodes. Lead shields 1 cm thick are used
to protect the remaining testis from scattered irradiation and preserve fertility.
Measurements using TLDs have shown that the radiation dose to the testis can be
reduced to 0.5–0.7 Gy. If scrotal irradiation is necessary because of previous scrotal
surgery or tumour involvement of the tunica vaginalis, a scrotal field using electron
therapy is used to treat the scrotal sac and lower inguinal nodes on the affected side.
362
Dose-fractionation
This field is matched to a tattoo on the inferior border of the dog leg field. A lead
cut-out is made to shield the penis and remaining testis.
Figure 30.3 Virtual simulation

for para-aortic, iliac and
inguinal ‘dog leg’ irradiation for
seminoma of right testis.
Dose solutions
Conformal CT planning is being developed. CT based fields tend to be wider than
traditional fields and provide improved dosimetry to vessel-based target volumes.
A high proportion of failures are marginal relapses which may be avoided by
individualised CT-based planning. In practice, opposing anterior and posterior
beams with individual shielding of renal parenchyma by MLC may give the best
target volume coverage with lowest integral dose for this low dose treatment.
Conventionally, anterior and posterior fields are treated isocentrically on a linear
accelerator. Perspex templates are used to define shielding to critical structures
when extended dog leg fields are used. Calculations of the dose distributions for
dog leg fields are made as for other irregular fields.
Low dose radiotherapy may give good palliation for metastases in sites such as
brain and bone where chemotherapy is not appropriate.
Dose-fractionation
■ Seminoma stage I
■ Seminoma stage IIA

■ Seminoma stage IIB

■ TIN
■ Palliative
363
TESTIS

The patient lies supine on the treatment couch with lasers used to align anterior
and lateral tattoos. When dog leg irradiation is used, a template is needed to place
lead shielding, or this is done more commonly now with MLC. Nausea and
vomiting are common side effects and can be reduced with prophylactic
antiemetics. If the patient has a history of peptic ulceration, a proton pump
inhibitor should be prescribed with antiemetics. The skin reaction is mild. Patients
should be offered sperm storage before treatment. They can be reassured that the
para-aortic treatment does not result in any significant dose to the testis nor impair
fertility in the long term. Dog leg fields and scrotal irradiation using testicular
shielding do, however, give some dose (0.7 Gy and 1.5 Gy respectively) and
patients should be counselled accordingly.
Verification
Treatment verification is carried out with EPI taken daily for the first 3 days and
any systematic errors corrected. Diodes or TLD are used on the first treatment to
check dose to isocentre and remaining testis.
key trials
Fosså SD, Horwich A, Russell JM et al. (1999) Optimal planning target volume for
stage I testicular seminoma: a Medical Research Council randomised trial. Medical
Research Council Testicular Tumour Working Group, TE10. J Clin Oncol 17: 1146.
Jones WG, Fosså SD, Mead GM et al. (2005) Randomised trial of 30 versus 20 Gy
in the adjuvant treatment of stage I testicular seminoma: a report on Medical
Research Council Trial TE18, European Organisation for the Research and
Treatment of Cancer Trial 30942 (ISRCTN18525328). J Clin Oncol 23: 1200–8.
Oliver RT, Mason MD, Mead GM et al. (2005) MRC TE19 collaborators and the
EORTC 30982 collaborators. Radiotherapy versus single-dose carboplatin in
adjuvant treatment of stage I seminoma: a randomised trial. Lancet 366: 293–300.
Information sources
Aparicio J, Germà JR, García del Muro X et al. (2005) Risk-adapted management for patients with
clinical stage I seminoma: the second Spanish Germ Cell Cancer Cooperative Group study.
J Clin Oncol 23: 8717–23.
Huddart R, Kataja V (2008) ESMO Guidelines Working Group. Testicular seminoma: ESMO clinical
recommendations for diagnosis, treatment and follow-up. Ann Oncol 19(suppl 2): ii49–51.
Krege S, Beyer J, Souchon R et al. (2008) European consensus conference on diagnosis and
treatment of germ cell cancer: a report of the second meeting of the European Germ Cell
Cancer Consensus Group (EGCCCG): part II. Eur Urol 53: 497–513.
Martin JM, Joon DL, Ng N et al. (2005) Towards individualised radiotherapy for Stage I seminoma.
The Testicular Cancer Resource Centre. Available at: www.acor.org/TCRC/ (accessed 10 December
2008).
Warde P, Specht L, Horwich A et al. (2002) Prognostic factors for relapse in stage I seminoma
managed by surveillance: a pooled analysis. J Clin Oncol 20: 4448–52.
364
Penis 31

Penis cancer is rare, with approximately 360 new cases per year in the UK, and
should be treated in specialised centres. Over 95 per cent of tumours are SCCs.
Penile cancer is staged by the AJCC TNM classification of 1997. In addition a
commonly used staging system is that proposed by Jackson (Table 31.1).
Table 31.1 Staging system for carcinoma of the penis proposed by Jackson*
Stage Characteristics
I Tumour confined to glans and/or prepuce

II Tumour extends onto shaft of penis
III Tumour with malignant, but operable, inguinal lymph nodes
IV Inoperable primary tumour extending off the shaft of the penis
or inoperable groin nodes or distant metastases
*From: Jackson SM (1966) The treatment of carcinoma of the penis. Br J Surg 53: 33–5. Reprinted with permission
from Blackwell.
With advances in reconstructive surgical techniques, most patients (85 per cent)
are now treated with penile preserving surgery. The overall 5-year survival is
52 per cent, ranging from 66 per cent to 29 per cent in lymph node negative
and positive disease, respectively. The local control rate with radical surgery is
90 per cent, and with brachytherapy or EBRT it is 65–85 per cent. Radiotherapy
has a role in small early tumours as an alternative to penile preserving surgery
in highly selected cases and is used palliatively in advanced inoperable cases.
Palpable mobile inguinal nodes are found in 30–50 per cent of patients at
presentation, but half of these are due to infection rather than tumour. If tumour is
confirmed by cytology or histology, an ipsilateral modified radical inguinal node
dissection is performed with a dynamic sentinel node study on the contralateral side.
Electron beam radiotherapy can be used for very small stage I superficial tumours.
Selected early stage I and II tumours 4 cm in size, and where there is invasion of
the corpora cavernosa 1 cm, can be treated with interstitial brachytherapy.
EBRT can be used for selected stage II and III tumours in patients unfit for or
refusing surgery.
Adjuvant nodal radiotherapy can be used to reduce the risk of local recurrence
after nodal dissection. Patients with a single positive node with extracapsular
365
PENIS
extension can be considered for adjuvant radiotherapy to the ipsilateral groin.

Patients with involved multiple or bilateral inguinal nodes can be considered for
adjuvant radiotherapy to the involved groin(s) and adjacent pelvic nodes.
Palliative radiotherapy and chemotherapy can be used for advanced inoperable
primary tumours or to treat fixed or fungating inguinal nodes or distant metastases.

The penis is composed of two corpora cavernosa and the corpus spongiosum.
Distally the corpus spongiosum expands into the glans penis, which is covered by
a skin fold known as the prepuce. The prepuce and skin of the penis drain to the
superficial inguinal lymph nodes. The glans penis and corpora have a very rich
lymphatic supply, which drains to the deep inguinal and external iliac lymph nodes
as shown in Fig. 31.1.
External
iliac
nodes
Deep
inguinal
Superficial
nodes
inguinal
nodes
Figure 31.1 Lymphatic

drainage of the penis.

After histological confirmation of primary nodal disease, patients should have a
full assessment including endoscopic examination of the urethra, cystoscopy, chest
X-ray and CT scan of the abdomen and pelvis. In advanced cases, MRI of the
pelvis and bone scans may be needed to complete staging.
Data acquisition
Patients are treated supine with arms folded comfortably across the chest.
Treatment is planned with the box technique as described below, or using a
planning CT scan with intravenous contrast of the inguinal and pelvic lymph node
regions. Scans are taken from the lower border of L5 to 5 cm inferior to the ischial
tuberosities using 3–5 mm slices.
366
Target volume definition and dose solutions
Target volume definition and dose solutions
Very small superficial stage I tumours on the prepuce or glans, unsuitable for
surgery or brachytherapy, can be treated with electron therapy as for SCC on the
skin. A lead cut-out is made to treat the tumour with a 2 cm margin. An appropriate
electron energy and thickness of bolus are chosen to give a skin dose of 100 per cent
and ensure the tumour is covered at depth by the 90 per cent isodose.
For larger tumours, the CTV is the clinical and radiological GTV with a 2 cm
margin proximally and distally, which in practice gives a PTV that covers the entire
penis. To immobilise the penis, it is covered with a gauze Tubigrip and placed in a
central cavity of a wax block within a hinged Perspex box. The box is closed, the gauze
drawn up through a hole in the top and a wax plug is inserted to maintain the penis
in position. The wax block allows even build up, and a homogeneous dose is obtained
using opposing lateral fields treating isocentrically with 4 MV or 6 MV photons. The
testes and groins are shielded from scattered irradiation by a lead sheet (Fig. 31.2).
Figure 31.2 EBRT technique for

carcinoma of the penile shaft.
Where the penis is short or retracted and cannot be treated with this technique,
a CT planned volume is used to treat it with wax bolus and shielding to the testes.
The typical arrangement is two anterolateral oblique 6 MV photon beams.
For inguinal and pelvic node adjuvant radiotherapy, pelvic blood vessels shown
on a contrast-enhanced planning CT scan are used to define the nodal CTV and a
3D margin of 7–10 mm added to form a PTV (see Fig. 35.2, p. 405).
Conventionally, radiotherapy of the inguinal regions is defined by a field that
extends laterally to the midpoint of the femoral neck, medially to the midline,
inferiorly 2 cm below the inferior border of the ischial tuberosity and superiorly to a
line joining the top of the anterior superior iliac spine to the pubic symphysis. Direct
anterior or opposing anterior and posterior beams are used, with central shielding to
cover the bladder and MLC shaping. If an anterior electron field is used, an
additional margin will be needed to account for the wider penumbra of electrons.
In advanced cases, radiotherapy can be used to palliate pain and bleeding with
a simple beam arrangement to the affected area.
■ Interstitial brachytherapy
Circumcision must be performed before implantation. The implant is performed
under general anaesthesia following insertion of a urinary catheter. The planning
target volume is defined as the GTV (palpable and visible tumour) with a margin
of 1–2 cm. For small superficial tumours, a single plane implant may be
367
PENIS
considered, but for the majority, the entire circumference of the penis is enclosed
in a two-plane implant. It is usual to use two or three sources in each plane and to
maintain parallelism by the use of a small template (Fig. 31.3a). The separation
between sources is usually 12–15 mm, but it may sometimes be up to 18 mm as
determined by the thickness of the shaft of the penis.
(a) (b)
Figure 31.3 Technique for implantation of carcinoma of shaft of penis with iridium wires.
(a) Rigid needles inserted through template. (b) Iridium wires held in position by template.
c, catheter.
The implant is usually performed with the axis of implantation at right angles to the
axis of the penis. Only a small part of the needle track is within the penis, and at the
apex of the glans one or more needles may be entirely free in air but held in position
by the template. The active length of iridium wire is usually 4–5 cm, which means that
there is active wire extending through the skin on either side of the lesion (Fig. 31.3b).
Care must be taken not to pass the wires through the urethra, which can normally
be avoided by making sure the catheter moves freely. In order to prevent radioactive
material from coming into contact with the skin of the thigh or testes, the shaft of the
penis is supported erect in a foam block. For those patients in whom it is important
to preserve fertility, a 2 mm thick sheet of lead is placed over the thighs and testes.
Dose-fractionation
■ Primary tumour
Radical EBRT or electrons
Brachytherapy
65 Gy to the 85 per cent reference isodose in 6–7 days.
368
Information sources
■ Lymph nodes
EBRT or electrons
■ Palliative

For EBRT the patient is set up with the wax/Perspex block defining the area to
be treated. Where no block is used, set-up uses lasers to align lateral pelvic tattoos.
Systemic antibiotics and circumcision may be needed prior to radiotherapy to
control infection and prevent radiation-induced oedema and urethral obstruction.
Treatment is poorly tolerated. Pain and difficulty passing urine are common.
A high fluid intake is encouraged and midstream urine specimens sent regularly to
exclude infection. Complications of both EBRT and brachytherapy include tissue
swelling, moist desquamation, secondary infection, penile ulceration (8 per cent),
penile necrosis (3–16 per cent) and urethral strictures (10–45 per cent).
Verification
With the block technique, treatment verification is done daily by ensuring
adequate inclusion of the penis within the block and coverage by the treatment
beams. EPI is used to verify other pelvic EBRT treatments with TLD or diode
measurements. The anterior oblique field technique is checked using orthogonal
EPI to verify the isocentre.
Information sources
Hadway P, Smith YC, Corbishley C et al. (2007) Evaluation of dynamic lymphoscintigraphy and
sentinel lymph-node biopsy for detecting occult metastases in patients with penile squamous
cell carcinoma. BJU Int 100: 561–5.
Jackson SM (1966) The treatment of carcinoma of the penis. Br J Surg 53: 33–5.
Mistry T, Jones RWA, Dannatt E et al. (2007) A 10-year retrospective audit of penile cancer
management in the UK. BJU Int 100: 1277–81.
Radiographics. Imaging of penile neoplasms. Available at: http://radiographics.rsnajnls.org/
cgi/content/full/25/6/1629 (accessed 15 December 2008).
Sarin R, Norman AR, Steel GG et al. (1997) Treatment results and prognostic factors in 101 men
treated for squamous carcinoma of the penis. Int J Radiat Oncol Biol Phys 38: 713–22.
Smith Y, Hadway P, Biedrzycki O et al. (2007) Reconstructive surgery for invasive squamous
carcinoma of the glans penis. Eur Urol 52: 1179–85.
Solsona E, Algaba F, Horenblas S et al. (2004) European Association of Urology Guidelines.
Available at: www.uroweb.org/fileadmin/tx_eauguidelines/penile (search via Google and click on
[pdf] Erectile Dysfunction; accessed 15 December 2008).
St Georges Specialist Network Penis Cancer Guidelines (2007) Available at: www.sussexcancer.
net/professionals/clinicalgroups/tumourgroups/urology/guidelines.asp (accessed 11 December
2008).
369
32 Cervix

Cervical cancer is the second commonest female malignancy worldwide and is
more common in developing countries. Its pathogenesis is associated with
persistent papilloma virus infection, smoking, multiple sexual partners, parity and
early coitus and childbirth, and 80–90 per cent are SCCs. Adenocarcinomas are
commonly of endocervical type and are increasing in incidence. Anaplastic small
cell tumours are aggressive with less than 50 per cent survival rate for stage I
disease. Metastases from colon and breast and direct spread from uterine malignancies
are also seen.
The choice of treatment for carcinoma of the cervix depends on the stage of the
disease (FIGO), tumour volume and histology, probability of lymph node involvement
and age and PS of the patient. Intraepithelial disease (cervical intraepithelial neoplasia
[CIN]) is treated with superficial ablative techniques. Microinvasive stage Ia1
disease (no greater than 3 mm deep stromal invasion and no greater than 7 mm
diameter) is treated with vaginal hysterectomy or excisional cone biopsy in young
patients to preserve fertility.
■ Early stage disease

Stage Ia2 (3–5 mm deep stromal invasion and no greater than 7 mm diameter) and
Ib1 (4 cm or less in size) are treated with radical hysterectomy and bilateral pelvic
lymphadenectomy. In selected patients with tumours less than 2 cm who are keen
to preserve fertility, radical vaginal trachelectomy (removal of cervix, the upper
part of the vagina, and parametrial tissue) and laparoscopic pelvic lymphadenectomy
may be carried out.
Stage Ib1 and IIa (tumour 4 cm or less) are treated equally effectively with
radical surgery or combined radical EBRT and brachytherapy with both treatments
giving 80–90 per cent 5-year survival rates.
Stage Ib2 or IIa tumours (4 cm tumour size) have deep stromal invasion and
an increased risk of parametrial and lymph node involvement. Surgery is performed
in selected patients with large tumours, lymphovascular space invasion (LVSI) and
adenosquamous or high grade histology, followed by postoperative radiotherapy.
Ovarian transposition may minimise the chances of radiation-induced menopause.
Primary radical radiotherapy is preferable for other stage Ib2 and IIa patients.
This avoids the increased morbidity seen when surgery is followed by postoperative
radiotherapy. In patients who have residual macroscopic disease, positive surgical
margins, parametrial involvement and positive lymph nodes, concomitant
chemotherapy with radiotherapy improves local control and survival rates compared
with postoperative radiotherapy alone, but late morbidity is also increased.
370
■ Locally advanced disease
Primary radiotherapy is the treatment of choice for locoregionally advanced disease
with a careful balance of EBRT and brachytherapy to maximise dose to tumour
and avoid normal tissues. Five-year survival rates with radiotherapy alone for stage
IIb, IIIb and IVa disease are 65–75 per cent, 35–50 per cent and 15–20 per cent,
respectively. Studies of combined chemoradiation with cisplatin 5FU have
shown a 30–50 per cent decrease in risk of death from cervical cancer compared
with extended field radiotherapy alone. However, patients with poor PS, impaired
renal function and para-aortic node disease were excluded, making the results
applicable to a selected population only. Similar good results have been shown
with concurrent mitomycin and 5FU, but neoadjuvant chemotherapy shows no
benefit when given before radiotherapy. Data from randomised trials show that
cisplatin-based chemotherapy improves survival, especially in stage II disease. It is
commonly delivered as a single agent, weekly during EBRT.
Where a vesico- or recto-vaginal fistula is present, a urinary diversion procedure
or defunctioning colostomy should be performed prior to radiotherapy.
The role of para-aortic node irradiation is controversial with no survival advantage
shown by the EORTC trial, but another study has reported 25–50 per cent long-
term survival after treatment for microscopic disease but with increased morbidity.
■ Stage IVb metastatic disease

Short course pelvic radiotherapy is successful in relieving bleeding and pelvic pain
in patients with metastatic disease. Palliative radiotherapy is used to relieve pain
from bone secondaries and symptoms from brain and nodal metastases.
■ Occult disease
Patients who have cervical cancer detected as an incidental finding at simple
hysterectomy for other reasons are usually treated postoperatively with chemo-
radiation and vault brachytherapy as appropriate.

Cervical cancer arises at the junction of the columnar epithelium of the endocervix
and the squamous epithelium of the ectocervix. It spreads directly into cervical
stroma presenting either as exophytic tumour protruding into the vagina or
endocervical tumour expanding the cervix. Superiorly, spread is into the lower
uterine segment, inferiorly to the vagina and laterally from paracervical spaces into
the broad and utero-sacral ligaments and pelvic sidewalls. Rectal and bladder
mucosal involvement is uncommon at presentation and occurs late (Fig. 32.1).
The lymphatic plexus is rich, connecting with the lower uterine segment and
laterally to the paracervical, parametrial, obturator, and internal and external iliac
nodes (Fig. 32.2). Posterior tumours may spread to presacral, common iliac and
para-aortic nodes along the ovarian vessels. Stage Ib tumours have 15–25 per cent
risk of pelvic, and 5–15 per cent risk of para-aortic, nodal involvement, depending
on tumour size, depth of stromal invasion and presence of LVSI. Blood spread is
late with metastases seen in lungs, extrapelvic nodes, liver and bone as well as
direct extension from para-aortic nodes to the lumbar vertebrae.
371
CERVIX
(a) (b)
Figure 32.1 Sagittal MR scans of stage IVa SCC cervix with bladder invasion,
biopsy-proven mucosal disease at cystoscopy, (a) before and (b) after radical EBRT
and brachytherapy.
Common
iliac node Internal
iliac node
External
iliac node Presacral
nodes
Obturator
node
Uterus
Cervix
Bladder Rectum
Figure 32.2 Lymphatic drainage of the cervix. Reproduced with permission from
Johnston TB, Davies DV, Davies F (1958) Grays Anatomy, 32nd edn. Edinburgh:
Churchill Livingstone.

Clinical assessment includes vaginal examination to inspect and palpate the extent of
tumour in the cervix and vagina, and a bimanual examination to determine mobility
and enlargement of the uterus. Vaginal spread is visualised using a speculum, and
rectal examination defines involvement of parametrial tissues, utero-sacral ligaments
and fixation to pelvic sidewalls. EUA is carried out in all patients, ideally jointly with
surgical colleagues, with cervical biopsy, endocervical curettings and accurate assess-
ment of the lesion according to the FIGO staging system. Insertion of radio-opaque
372
Data acquisition
markers such as gold grains into the lateral and AP extent of tumour can be used to
aid target volume delineation and verify daily treatment delivery. Cystoscopy is
performed to look for invasion of the bladder base mucosa, and proctoscopy where
rectal invasion is suspected. CT scanning of the chest and abdomen is performed
for staging and to assess the renal tract, and to detect ureteric obstruction or
hydronephrosis, which may need stent placement prior to treatment. MRI is the
modality of choice for detecting depth of cervical and uterine invasion and parametrial
spread. CT and MRI detection of pelvic and para-aortic nodes has a low sensitivity,
as nodes are designated involved when the short axis diameter is greater than 1 cm.
When MRI is combined with ultra-small particles of iron oxide (USPIO) there is
increased sensitivity of pelvic lymph node detection down to 3 mm. PET has a high
sensitivity for detecting both nodal and distant metastases.
Data acquisition
■ CT scanning
With the change from 2D to 3D planning, CT scanning is now recommended for
data acquisition. Patients are scanned supine with arms on the chest, knees and lower
legs immobilised, and anterior and lateral tattoos marked with radio-opaque material
aligned with lasers to prevent lateral rotation. For obese patients, a prone bellyboard
immobilisation system may be used to allow small bowel to fall anteriorly away from
the target volume. Clinical examination is made in the treatment position and
inferior extent of tumour in the vagina marked with radio-opaque material. A protocol
is used to maintain a constant bladder filling – ‘comfortably full’ – as variations in
bladder volume have been shown to influence mobility and position of the uterus
and cervix. For example, patients are asked to empty the bladder and drink 200 mL
water 20 min before the scan and treatment each day.
CT scans are taken from the first lumbar vertebra to 5 cm beyond the vaginal
introitus. Intravenous contrast is used to outline pelvic blood vessels to be used as
surrogates for pelvic nodes in CTV delineation. It may also enhance the GTV-T.
Oral contrast may be given to outline small bowel if IMRT is to be given.
EUA findings are used to optimise localisation of tumour margins along with
diagnostic MRI and/or PET-CT scans co-registered with planning CT scans to
define the AP part of the volume in particular (Fig. 32.3).
■ Conventional
Conventionally, the simulator is used to acquire patient data with the patient
supine with arms on the chest, knee and lower leg immobilisation or alpha cradles
to prevent pelvic rotation, and aligned using orthogonal laser beams with anterior
and lateral tattoos marked with radio-opaque material. For obese patients, a prone
bellyboard immobilisation system may be used to allow small bowel to fall
anteriorly away from the target volume. Palpation of the primary tumour is carried
out with the patient in the treatment position and the inferior border marked with
radio-opaque material.
A protocol is used to maintain a constant bladder filling – ‘comfortably full’ – as
variations in bladder volume have been shown to influence mobility and position
of the uterus and cervix.
373
CERVIX
(a) (b)
(c)
Figure 32.3 (a) CT, (b) MRI and (c) CT/MRI fusion showing GTV for carcinoma of the
cervix. Courtesy of Professor R Reznek, Institute of Cancer, Barts and the London School
of Medicine and Dentistry.
AP and lateral simulator films are taken. All clinical, surgical and histological
data must be used as appropriate, along with any diagnostic CT or MRI information,
showing tumour extent and the position of the kidneys, to define individualised
treatment beams. Standard field borders using bony anatomical landmarks have
been shown to include unnecessary normal tissue and may miss primary tumour or
lymph nodes.
The upper border of the individualised treatment beam is at the lower margin of
L4 to include distal common iliac nodes. The inferior border is 3 cm below the most
inferior disease in the vagina as palpated or seen on MRI. Lateral borders are 2 cm
outside the bony pelvic sidewalls. The anterior border must encompass the GTV-T
as well as the common iliac nodes and is usually placed through the anterior third of
the symphysis pubis. The posterior border is 2 cm from the GTV-T including
posterior extension of tumour, utero-sacral ligaments and upper presacral nodes and
is commonly situated 0.5 cm posterior to the anterior border of the S2/3 vertebral
junction. Individualised shielding is employed in the anterior beam to the superior
corners to exclude small bowel. Shielding to the inferior corners to protect femoral
heads may mask the external iliac nodes and should be used sparingly. Lateral beams
374
have shielding to sacral nerve roots posteriorly. For simulation of para-aortic nodal
beams, intravenous contrast is required to localise kidneys for shielding.

Primary radiotherapy is delivered to the primary cervical tumour (GTV with a
margin for local microscopic spread [CTV-T]) as well as to potential sites of pelvic
lymph node involvement (CTV-N), with or without concurrent chemotherapy.
An initial course of EBRT is given to CTV-T and CTV-N to shrink the bulky
primary tumour and make the geometry of subsequent intracavitary brachytherapy
to CTV-T optimal. When intracavitary treatment is not feasible, e.g. due to
obstruction of the cervical os, an EBRT boost is given to the residual GTV-T, with
a 1–1.5 cm margin. If multiple uterine fibroids are present, primary surgery may be
indicated, as brachytherapy may not be feasible.
CTV-T includes the primary GTV-T with potential microscopic spread to
cervix, uterus, parametrial tissues, upper vagina, and broad and proximal utero-
sacral ligaments. If posterior extension of cervical tumour is present, the entire
utero-sacral ligaments and upper presacral nodes should be included. If following
chemoradiation there is residual macroscopic disease on repeat imaging, an
individualised EBRT boost to this disease in pelvic sidewall or nodes is performed,
ensuring no overlap with brachytherapy doses.
CTV-N includes the pelvic lymph nodes, i.e. obturator, internal, external and
common iliac, and upper presacral nodes. These are delineated by identifying
contrast-enhanced pelvic blood vessels on each CT scan and using a 7 mm margin
to create a 3D CTV (Fig. 32.4).
Figure 32.4 T2-weighted MRI

after USPIO administration
showing lymph nodes outlined in
yellow. CTV delineation using
7 mm margin around contrast-
enhanced pelvic blood vessels
and nodal contours. (Blue lines
show range of 3–15 mm margins.)
Reproduced with permission from
Taylor A et al. (2005) Int J Radiat
A typical CTV to PTV margin of 15–20 mm is used around the CTV-T to allow
for organ motion of cervix and uterus and measured set-up uncertainties. For
CTV-N organ motion occurs to a lesser extent and a 7–10 mm CTV to PTV
margin is typically sufficient for set-up variations (Fig. 32.5).
375
CERVIX
(a) (b)
(c) (d)
Figure 32.5 T2b carcinoma of the cervix: (a) sagittal T2-weighted MRI, (b) sagittal,
(c) axial and (d) coronal CT scans with CTV and PTV for primary tumour and nodes
(CTV-N in purple, PTV-N in green).
Virtual simulation can be used to define beams and shielding, using 3D tumour
and normal organ information (Fig. 32.6). For 3D conformal radiotherapy and
IMRT, tumour and normal organs are also outlined in 3D. Normal tissues to be
contoured include bladder, rectum, small and large bowel, and the femoral heads.
The GTV has been resected at surgery so the CTV-T comprises the surgical
tumour bed including the vaginal vault and parametrial tissues. CTV-N contains
obturator, internal, external and common iliac nodes. CTV-T is delineated using
preoperative CT and MRI, operative diagrams and histological findings to include
sites at high risk of recurrence. Pelvic lymph nodes (CTV-N) are delineated using
CT contrast-enhanced blood vessels with a 7 mm margin. A CTV-PTV margin of
10–15 mm is used for CTV-T and a 7–10 mm margin for CTV-N to allow for
organ motion and set-up errors.
Normal tissues to be contoured include bladder, rectum, small and large bowel,
and the femoral heads.
376
(a) (b)
Figure 32.6 Virtual simulation of pelvic beams: (a) anterior BEV and (b) lateral BEV
showing MLC shielding to sacral nerve roots.
■ Para-aortic radiotherapy
CT scanning is performed from the diaphragm to pelvis with intravenous contrast
given to enhance blood vessels and kidneys and to locate the para-aortic nodes.
Virtual simulation is used to design anterior and posterior beams, from the upper
border of L1 to the L5/S1 junction, avoiding the kidneys laterally with MLC
shielding and in continuity with pelvic beams when common iliac nodes are
involved (Fig. 32.7). 3D volume outlining permits the use of IMRT.
Figure 32.7 Virtual simulation of treatment

beams for para-aortic and pelvic node
irradiation with MLC shaping to shield
kidneys.
Symptoms of bleeding or pain from locally advanced cervical cancer can be
alleviated with small volume, short-course radiotherapy. CT planning is used as
described above, to localise the GTV alone with a 1.5–2 cm margin for the PTV.
An arrangement of three to four beams is used to reduce toxicity as much as
possible and can be produced after virtual simulation using 3D CT data.
377
CERVIX
Dose solutions
All patients selected for primary radiotherapy are treated by EBRT first to the
primary tumour and pelvic lymph nodes. Shrinkage of the GTV leads to better
geometry and closer application of the intracavitary sources, which deliver higher
radiation doses to the tumour. The balance of dose from EBRT and intracavitary
radiation is modified according to the stage and bulk of disease, and the risk or
presence of pelvic lymph node involvement.
■ Conventional
High-energy photons (10–16 MV) are used for EBRT to spare superficial tissues
and improve dose distributions. Conventionally, anterior, posterior and two lateral
beams are usually used to spare the rectum with decreased weighting of the
posterior beam (see Fig. 33.4, p. 389).
■ Conformal
Using 3D dose planning, MLC shaping is designed to spare normal tissues, where
possible. Similarly, standard shielding can be used on simulated anterior, posterior
and lateral beams to reduce dose to small bowel and sacral nerve roots.
■ Complex
IMRT techniques using five or seven photon beams may reduce dose to small
bowel, rectum and bladder by shaping treatment to the pelvis. IMRT may also
help to limit dose to pelvic bone marrow in patients undergoing chemoradiation.
Organ motion studies have shown that the cervix and uterus position varies with
both bladder and rectal filling by up to 20 mm. IGRT may be used to localise the
target volume on a daily basis with a full IMRT implementation programme to
ensure that it is delivered accurately and safely.
Brachytherapy
The principles of brachytherapy, differences between LDR, MDR and HDR, and
ICRU Report No. 38 are described in detail in Chapter 5. Brachytherapy allows
delivery of a very high dose to the central tumour volume to obtain maximal local
control without exceeding the tolerance of surrounding normal tissues. It is feasible
because the normal uterus and vaginal vault are relatively radio-resistant and there
is rapid fall-off of dose at a distance from the cervix, protecting the adjacent rectum,
bladder and small bowel. Gynaecological brachytherapy can be delivered at low,
medium or high dose rate, or with pulsed brachytherapy, which uses many, closely
spaced radiation fractions or pulses to mimic continuous LDR treatment.
■ GEC-ESTRO recommendations for image-based

3D brachytherapy
This method relies on CT and/or MR imaging rather than 2D orthogonal
radiographs. Doses are prescribed to volumes rather than reference points. The
378
Brachytherapy
GEC-ESTRO group has recently published recommendations on target volume
concepts and plan evaluation using DVHs. GTV, high risk (HR) CTV and
intermediate risk (IR) CTV are defined with the bladder, rectum, sigmoid colon and
small bowel as OAR (Fig. 32.8). The total dose, including EBRT dose, should be
isoequivalent to 80–90 Gy in 2 Gy fractions for the HR CTV, and 60 Gy for the IR
CTV. Doses of 87 Gy to the HR CTV have been associated with improved local
control compared with doses below 87 Gy. Doses for 2 mL of tissue volume (D2cc)
for the OAR are calculated at 2 Gy per fraction. Isoequivalent doses of 80–90 Gy for
the bladder and 70–75 Gy for the rectum and sigmoid colon are generally accepted.
Figure 32.8 CT scan of

patient with HDR
brachytherapy applicators in
situ and isodoses. HR (high
risk) CTV cyan, IR
(intermediate risk) CTV
magenta.
■ The Manchester system for gynaecological

brachytherapy
This system relies on prescribed doses to defined reference points. Manchester
point A is defined as a point 2 cm lateral to the central uterine canal and 2 cm
superior to the lateral fornix in the plane of the uterus. It lies within the
paracervical tissues near the uterine artery and ureter, and was chosen to take into
account the tolerance of adjacent dose-limiting normal structures. In practice, the
prescribed dose of irradiation is specified at a point that is 2 cm above and 2 cm
lateral to the flange of the intrauterine tube at the external os. This point is taken
to be equivalent to Manchester point A. Point B is defined as being 3 cm lateral to
point A, i.e. 5 cm from the midline on the lateral axis, and it is used to provide an
indication of dose to the distal parametria. The ICRU Report No. 38 specifies
reference points for the bladder and rectum. The bladder point is posterior to the
catheter balloon filled with 7 mL of contrast solution. The rectal point is 0.5 cm
posterior to the most posterior packing or posterior sarface of the ovoids.
■ Applicators used in gynaecological brachytherapy

A number of different applicators are available for use with an intact uterus.
A central intrauterine tube is used with a vaginal applicator, which can consist of a
cylinder, ovoids or a ring. Applicators have been developed that contain holes for
interstitial needles, which may be used to produce better coverage of inner
parametrial disease.
379
CERVIX
Vaginal vault brachytherapy can be delivered using a vaginal cylindrical applicator

or vaginal ovoids. Both are available in varying diameters/sizes. Ovoids produce a
better dose distribution if coverage of the parametrial soft tissues is required,
whereas cylindrical applicators mainly treat the vaginal mucosa. Doses are
prescribed to 0.5 cm from the surface of the applicator.
■ Example of LDR/MDR applicator insertion technique

After examination under general or spinal anaesthesia and urinary catheter insertion,
the uterine cavity is measured with a sound. The cervical canal is dilated and a central
uterine tube of appropriate length is inserted. The vaginal applicators are then
positioned and fixed to the tube. Vaginal packing is used to distance OAR from the
applicator and to prevent rotation of the applicators. The applicators can be further
secured using labial sutures or bandaging. 2D or 3D imaging can then be performed.
■ Fractionated HDR delivery technique

The principles of applicator insertion are the same as those for LDR/MDR.
A Schmidt’s sleeve can be placed or sutured into the cervical canal, which may
facilitate reinsertion of the applicators without anaesthesia for subsequent fractions.
A rectal retractor can be used instead of or as well as vaginal packing to create
distance between the applicators and rectum. A perineal bar is another method
used to secure the applicator position.
Dose-fractionation
Stage IB2 and IIA
EBRT
45 Gy in 25 daily fractions of 1.8 Gy given in 5 weeks followed by intracavitary
brachytherapy.
Intracavitary brachytherapy
LDR
27 Gy to point A single insertion.
or
HDR
14 Gy in 2 fractions given in 5–8 days to point A.
Stage IIB or above
EBRT
50.4 Gy in 28 daily fractions of 1.8 Gy given in 51⁄2 weeks followed by intracavitary
brachytherapy.
LDR
22.5–25 Gy to point A single insertion.
380
Dose-fractionation
HDR
21 Gy in 3 fractions over 5–8 days to point A.
Concurrent chemotherapy (weekly cisplatin 40 mg/m2) is given for both high risk
early stage disease and locally advanced tumours unless patients are medically unfit
for chemotherapy or have a GFR 50 mL/min. Overall treatment time should not
exceed 56 days including brachytherapy and should ideally be 49 days or less.
■ EBRT boost to central tumour when brachytherapy

not feasible
15 Gy in 8 daily fractions
or
20 Gy in 11 daily fractions.
Total dose 65 Gy (in 2 Gy equivalent).
■ EBRT boost to residual macroscopic disease

5.4 Gy in 3 daily fractions of 1.8 Gy
or
EBRT
50.4 Gy in 28 daily fractions of 1.8 Gy in 51⁄2 weeks if macroscopic residual disease.
Concurrent chemotherapy is administered if there are positive surgical margins,
positive pelvic nodes or parametrial involvement and treatment time should not
exceed 56 days.
Intracavitary brachytherapy to vaginal vault

LDR
15 Gy at 0.5 cm from surface of applicator.
20 Gy at 0.5 cm from surface of applicator given in a single application if macroscopic
residual vault disease.
HDR
8 Gy at 0.5 cm from surface of applicator in 2 fractions
or
12 Gy in 3 fractions if positive vaginal surgical margin.
■ Para-aortic node radiotherapy

381
CERVIX
Whole pelvis or para-aortic nodes
20–30 Gy in 5–10 daily fractions given in 1–2 weeks.
8–10 Gy in 1 fraction for haemostasis.

Patients are aligned with tattoos and laser lights using an identical set-up to that
for localisation in the CT scanner or simulator. When given concurrently,
chemotherapy is given every 7 days with radiotherapy delivered 1–2 h later with
strict scheduling compliance. Overall treatment time should be as short as possible
and delays avoided so that for SCC it does not exceed 56 days and ideally is 49
days or less. These are category 1 patients and any unscheduled gaps in treatment
are rectified by treating at the weekend or by using 2 fractions in 1 day with a
minimal interval of 6 h. Intracavitary brachytherapy should follow EBRT as
planned without delay.
Patients are reminded to follow the bladder filling protocol to ensure
reproducibility of treatment and reduce side effects. Acute side effects may be
increased when radiotherapy is combined with concurrent chemotherapy.
Diarrhoea may occur during EBRT of the pelvis, and is treated with loperamide
hydrochloride and a low residue diet. If diarrhoea worsens and/or abdominal pain
occurs, treatment may need to be suspended. Urinary frequency and dysuria may
occur, and a urine specimen should be taken to exclude infection. Rectal bleeding
and painful defecation may occur. Patients undergoing para-aortic nodal
irradiation should receive prophylactic antiemetics to prevent nausea and
vomiting. A good fluid intake should be encouraged and patients should avoid
smoking. Severe perineal or natal cleft skin reactions are treated with 1 per cent
hydrocortisone cream. Studies have shown that haemoglobin should be
maintained at 12 g/dL or above, as locoregional recurrence rates are higher in
anaemic patients. Premenopausal patients will lose ovarian function, and hormone
replacement therapy may be given to improve menopausal symptoms and prevent
osteoporosis. Vaginal hydration gels and dilators are used to help reduce vaginal
shortening and stenosis. Psychosexual counselling may be helpful. Chronic fatigue,
urinary or faecal urgency or incontinence and pelvic pain are uncommon late
effects of pelvic radiotherapy and require special care.
Verification
Portal films or EPIs are taken on the first 3–5 treatment days and compared with
DRRs or simulator images within a verification strategy protocol. Any systematic
errors of 5 mm are identified and corrected and weekly EPIs then taken. Exit
dosimetry using diodes is performed on the first day in all patients.
382
Information sources
Key trials
Keys HM, Bundy BN, Stehman FB et al. (1999) Cisplatin, radiation and adjuvant
hysterectomy compared with radiation and adjuvant hysterectomy for bulky
stage IB cervical carcinoma. N Engl J Med 340: 1154–61.
Morris M, Eifel PJ, Lu J et al. (1999) Pelvic radiation with concurrent
chemotherapy compared with pelvic and para-aortic radiation for high-risk
cervical cancer. N Engl J Med 340: 1137–43.
Peters WA, Liu PY, Barrett RGW et al. (1999) Cisplatin, 5-fluorouracil plus radiation
therapy are superior to radiation therapy as adjunctive therapy in high risk, early
stage carcinoma of the cervix after radical hysterectomy and pelvic
lymphadenectomy: report of a phase III inter group study. Gynecol Oncol 72: 443.
Rose PG, Bundy BN, Watkins EB et al. (1999) Concurrent cisplatin-based
radiotherapy and chemotherapy for locally advanced cervical cancer. N Engl J
Med 340: 1144–53.
Whitney CW, Sause W, Bundy BN et al. (1999) Randomised comparison of
fluorouracil plus cisplatin versus hydroxyurea as an adjunct to radiation therapy
in stage IIB-IVA carcinoma of the cervix with negative para-aortic lymph nodes:
a Gynecologic Oncology Group and Southwest Oncology Group study. J Clin
Oncol 17: 1339–48.
Information sources
Cancer Backup (2007) Pelvic radiotherapy in women: possible late effects. Available at:
www.cancerbackup.org.uk
Green J, Kirwan J, Tierney J et al. (2005) Concomitant chemotherapy and radiation therapy for
cancer of the uterine cervix. Cochrane Database Syst Rev CD002225.
Haie-Meder C, Pötter R, Van Limbergen E et al. (2005) Recommendations from Gynaecological
(GYN) GEC-ESTRO Working Group (I): concepts and terms in 3D image based 3D treatment
planning in cervix cancer brachytherapy with emphasis on MRI assessment of GTV and CTV.
Pötter R, Haie-Meder C, Van Limbergen E et al. (2006) Recommendations from Gynaecological
(GYN) GEC ESTRO Working Group (II): concepts and terms in 3D image based treatment
planning in cervix cancer brachytherapy: aspects of 3D imaging, radiation physics, radiobiology,
and 3D dose volume parameters. Radiother Oncol 78: 67–77.
Sedlis A, Bundy BN, Rotman MZ et al. (1999) A randomised trial of pelvic radiation therapy versus no
further therapy in selected patients with stage IB carcinoma of the cervix after radical hysterectomy
and pelvic lymphadenectomy: a Gynecologic Oncology Group Study. Gynecol Oncol 73: 177–83.
Taylor A, Rockall AG, Reznek RH et al. (2005) Mapping pelvic lymph nodes: guidelines for
delineation in intensity-modulated radiotherapy. Int J Radiat Oncol Biol Phys 63: 1604–12.
Taylor A, Rockall AG, Powell ME (2007) An atlas of the pelvic lymph node regions to aid
radiotherapy target volume definition. Clin Oncol 19: 542–50.
383
33 Uterus

A total abdominal hysterectomy with bilateral salpingo-oophorectomy is the
mainstay of treatment for all tumours of the corpus uteri, of which
adenocarcinoma of the endometrium of endometrioid type is the most common.
Histological staging and prognostic factors are determined from the surgical
specimen and are used to guide adjuvant treatment.
For low risk patients (stage IA and IB, grade I–2) relapse-free survival is 95
per cent with surgery alone, and neither lymphadenectomy nor adjuvant EBRT is
indicated. Vaginal vault brachytherapy may be considered in some patients with
stage IB grade 2 disease where there is lymphovascular space invasion (LVSI).
For intermediate risk patients (with stage IA–IB grade 3 without LVSI, IC
grade I and 2, stage IIA grade I–2, age 60 or over) there is a 15–25 per cent risk
of locoregional relapse after surgery, which can be reduced to around 5 per cent
with postoperative pelvic radiotherapy. The PORTEC-2 trial randomised these
patients between postoperative pelvic radiotherapy and vaginal brachytherapy to
see if local control can be maintained, with a reduction in toxicity of treatment and
improvement in quality of life. Vaginal vault brachytherapy may be the treatment
of choice if no difference in overall pelvic failure and survival is shown.
High risk patients include those with two or more of the following risk factors:
stage T1C (invasion of more than 50 per cent of myometrium), grade 3 disease,
and LVSI. For these patients lymph node involvement is around 35 per cent and
overall 5-year survival 55–60 per cent. Other poor prognostic features are serous
papillary or clear cell histology, which often present with advanced disease, and
involvement of the cervix (stage II). Bilateral pelvic lymph node dissection
(BPLND) is performed for accurate staging but has not been shown to confer any
survival advantage for high risk early stage patients. Pelvic radiotherapy after
surgery improves locoregional control. However, high risk patients and those with
locally advanced disease die of metastatic spread, so the addition of concurrent and
adjuvant chemotherapy is being investigated in a randomised trial (PORTEC-3).
A vaginal brachytherapy boost is given in addition to EBRT to improve local
control in patients with histological involvement of the cervical stroma.
The behaviour of serous tumours is more like that of primary ovarian cancer, so
omentectomy is performed in addition to BPLND. Chemotherapy may play a role
in adjuvant treatment in addition to EBRT.
Radiotherapy may be used as a primary treatment for more advanced or
inoperable disease or in the rare instance when the patient is unfit for surgery. Five-
year survival rates of 70–75 per cent are obtained for stage II, 50 per cent for stage III
and 25 per cent for stage IV disease.
384
Rarer tumours include carcinosarcomas, leiomyosarcomas, endometrial stromal
sarcomas (which are mainly low grade), and squamous cell carcinomas. Surgery is
the mainstay of treatment for these tumours, with adjuvant pelvic EBRT given
with proven benefit for carcinosarcomas and SCCs and for local control only in
selected cases of leiomyosarcomas and endometrial stromal sarcomas.
Sequencing of multimodality treatment

Multicentre trials are currently assessing the value of concurrent chemotherapy
with radiotherapy, followed by adjuvant chemotherapy with platinum drugs and
taxanes, for high risk and advanced stage endometrial carcinoma. These regimens
are being compared with pelvic radiotherapy alone to establish outcomes in terms
of locoregional relapse and survival, as well as the toxicity profile and effect on
quality of life. Chemotherapy can also be given before pelvic radiotherapy.

The uterus is supported by the levator ani muscles and is commonly anteverted and
inclined forwards at an angle of 90° to the axis of the vagina (Fig. 33.1). The
myometrial wall of the body of the uterus is lined by endometrium and covered
externally by a reflection of the peritoneum. The base of the bladder is closely
applied to the anteroinferior part of the uterus, and posteriorly the pouch of
Douglas lies between the posterior fornix and the rectum. Loops of small bowel
may lie in the pouch of Douglas, limiting the dose of irradiation which can be given.
Uterus
Pouch of
Douglas
Bladder
Symphysis Rectum
pubis
Figure 33.1 Regional anatomy of the uterus.
Tumour may infiltrate locally through the myometrium and parametrium to

involve other pelvic organs including into the peritoneal cavity, or it may extend
inferiorly to the endocervix and vagina. Lymphatic drainage from the corpus passes
through the broad ligament to the obturator, external iliac and presacral lymph
nodes, and via the round ligament to the superficial inguinal lymph nodes
(Fig. 33.2). Spread may also occur in the ovarian vessels to the para-aortic nodes,
and by retrograde lymphatic spread to the lower third of the vagina, particularly
posterior to the urethra. If the endocervix is involved, lymphatic spread occurs
385
UTERUS
through paracervical and parametrial pathways to pelvic lymph nodes. Serous

tumours spread like ovarian tumours trans-coelomically. The most common sites
for distant metastases are the lungs, liver, bone and brain.
Presacral Para-aortic
nodes nodes
Obturator External
node iliac nodes
Inguinal
nodes
Figure 33.2 Lymphatic drainage of the uterus.
Assessment of nodal involvement by CT and MRI are both dependent on size

criteria (usually a short axis lymph node diameter greater than 1 cm), but the
sensitivity is only 40–70 per cent. MRI with intravenous contrast agent USPIO
(ultra-small particles of iron oxide) improves the sensitivity of detecting pelvic
nodal metastases down to 3 mm with maintained specificity. PET has a high
sensitivity for detecting both nodal and distant metastases.

Patients commonly present with postmenopausal bleeding. They first undergo a
transvaginal ultrasound examination. Assessment then involves hysteroscopy and
biopsy to make a histological diagnosis. MRI is more sensitive than CT at
identifying the presence of myometrial invasion (Fig. 33.3) and determining its
depth (e.g. whether greater than 50 per cent), and cervical involvement, as well as
at detecting pelvic and para-aortic lymph nodes for accurate preoperative staging.
All patients with moderate or high grade tumours on biopsy should therefore
undergo MRI at diagnosis to select patients for referral to specialist centres for
BPLND. Assessment of patients for adjuvant radiotherapy should take into
account all surgical and histological information including stage, grade,
histological subtype, presence of LVSI, and positive lymph nodes or positive
washings collected at surgery. Inguinal lymph nodes are assessed by palpation,
although involvement of enlarged nodes can only be proved by histological or
cytological examination. It is uncommon and related to invasion of the lower third
of the vagina.
CT scan of the chest and abdomen is performed for high grade uterine sarcomas
to exclude distant metastases because of the risk of early haematogenous spread.
386
Data acquisition
Figure 33.3 Sagittal T1-weighted MR
scans showing a T1c endometrial
carcinoma infiltrating into the outer half of
the myometrium.
Data acquisition
■ CT scanning
Most patients referred for radiotherapy require adjuvant pelvic EBRT following
surgery. With the change from 2D to 3D planning, CT scanning is now recommended
for data acquisition. Patients are scanned in the treatment position supine with arms
on the chest, knees and lower legs immobilised. Anterior and lateral tattoos are aligned
with lasers to prevent lateral rotation and marked with radio-opaque material. For
obese patients, the prone bellyboard may be used to allow small bowel to fall anteriorly
away from the target volume. A protocol is used to maintain a constant bladder filling
‘comfortably full’ to push the small bowel superiorly. For example, patients are asked
to empty the bladder and drink 200 mL water 20 minutes before the scan and
treatment each day. The introitus is marked with radio-opaque material.
CT scans are taken from the superior border of L3 to 5 cm beyond the vaginal
introitus. Intravenous contrast can be used to visualise primary tumour and uterus
if still in situ and pelvic blood vessels, and thereby design target volumes for pelvic
nodes most accurately. In this situation image registration with diagnostic MRI is
especially useful to localise gross tumour in the uterus and/or nodes. Oral contrast
is given to outline small bowel if IMRT is to be given.
■ Conventional
Conventionally the simulator is used to acquire data with the patient supine with
arms on the chest, knee and lower leg immobilisation or alpha cradles to prevent
pelvic rotation. The patient is aligned using orthogonal laser beams with anterior
and lateral tattoos marked with radio-opaque material. For obese patients, a prone
bellyboard immobilisation system may be used to allow small bowel to fall
anteriorly away from the target volume. A protocol is used to maintain constant
bladder filling ‘comfortably full’ as described above.
The vaginal vault is marked with a radio-opaque tampon for adjuvant treatment.
In the rare cases where surgery has not been feasible, palpation of the primary
tumour is carried out with the patient in the treatment position and the inferior
border marked with radio-opaque material.
387
UTERUS
AP and lateral simulator films are taken. All clinical, surgical and histological
data must be used, with any diagnostic CT or MRI information showing tumour
extent, to design individually placed beam borders. The use of standard borders
using bony anatomical landmarks has been shown to include unnecessary normal
tissue and may miss primary tumour or lymph nodes.

CT planning
There is no GTV postoperatively and CTV-T includes the vaginal vault and
parametrial tissues. CTV-N includes the obturator, external and internal iliac nodes
and additional common iliac lymph nodes when indicated. CTV-T and -N are
delineated using preoperative CT and MRI, operative diagrams and histological
findings to include sites at high risk of recurrence. Guidelines have been proposed
for standardised nodal CTV definition using CT scans with contrast-enhanced
pelvic blood vessels (surrogate for lymph nodes) with a 7 mm margin.
A typical CTV-PTV margin of 10–15 mm is added for CTV-T and 7 mm for
CTV-N to allow for organ motion and set-up errors, and is individualised depending
on known results of departmental set-up variations.
Normal tissues to be contoured include bladder, rectum, small bowel and the
femoral heads.
Conventional
The superior border is usually at the L5/S1 junction to include external and internal
iliac nodes. If these nodes are positive, the border is at the lower margin of L4,
individualised to include distal common iliac nodes. The inferior border is placed to
include the upper half of the vagina for adjuvant treatment, or 3 cm below the most
inferior disease in the vagina as palpated or seen on MRI. Lateral borders are 2 cm
outside the bony pelvic sidewalls. The anterior border must encompass the CTV-N
as well as GTV-T (if present) and is placed through the anterior third of the
symphysis pubis. The posterior border is commonly situated 0.5 cm posterior to the
anterior border of the S2/3 vertebral junction, but this is varied according to surgical
and histological findings. Individualised shielding is employed in the anterior beam
to the superior corners to exclude small bowel. Shielding to the inferior corners to
protect femoral heads may mask the external iliac nodes and should be used sparingly.
Lateral beams have shielding to sacral nerve roots posteriorly (Fig. 32.6b, p. 377).
A tattoo is placed at the centre of the volume with two lateral pelvic tattoos to
aid alignment and reproducibility of the patient position.
Non-surgical patients with stage I and II disease may have multiple comorbidities
which limit the volume that can be treated radically. The GTV for uterine tumours
is defined on clinical examination and with MRI. The CTV-T includes the whole
uterus, cervix, ovaries, parametrium and upper half of vagina. Involved nodes are
detected by CT, MRI or MRI with USPIO. If the whole vagina is involved, the
388
inguinal nodes may need to be included in the CTV-N. A CTV-PTV margin of
10–15 mm is commonly used to allow for set-up uncertainty and physiological
movement of the corpus and bladder.
For patients with locally advanced inoperable stage III/IV disease, the treatment
intent is often palliative. Neoadjuvant chemotherapy may be considered followed
by EBRT individualised to the gross extent of tumour in uterus, cervix, parametrial
tissues, any vaginal extension and involved lymph nodes with a 2 cm margin.
OAR
The bladder, rectum, femoral heads and small bowel are delineated, and the
sigmoid colon if utero-vaginal brachytherapy is used.
Vault brachytherapy can be delivered using either a vaginal cylindrical applicator
(Fig. 33.4) or vaginal ovoids. Both are available in varying diameters/sizes. Ovoids
will treat the upper third of the vagina in most patients whereas cylindrical
applicators can be loaded to treat any length of vagina required.
(a)
Figure 33.4 (a) Vaginal cylinder

applicator for HDR brachytherapy
with (b) dose distribution from an
applicator 3 cm (W) 3 cm (dwell
length) prescribed at 0.5 cm from
(b) surface.
389
UTERUS
A central uterine tube vaginal applicator is commonly used to treat the

primary tumour in the unoperated patient.
See Chapter 32.
Dose solutions
■ Conventional
A ‘brick’ arrangement with four beams is commonly chosen (anterior, posterior and
lateral opposing wedged beams), or where possible the posterior beam is omitted
to reduce rectal dose. Higher energy photons (8–16 MV) are commonly employed
and individual shielding used in superior corners of the anterior beam to exclude
small bowel and reduce normal tissue dose. Care should be taken that any shielding
of inferior corners to exclude femoral heads does not also shield external iliac nodes.
Shielding to the sacral nerve roots may be used posteriorly in the lateral beams.
■ Conformal
With 3D target volume localisation, 3D dose planning can be used with individual
shaping of each beam using MLC or shielding blocks to spare normal tissues. One
anterior, two lateral opposing and one partially weighted posterior beam are
commonly used (Fig. 33.5). Conformal radiotherapy and systematic target volume
definition significantly reduces dose to the rectum and bladder compared with
conventional solutions, as well as avoiding geographical miss.
(a) (b)
Figure 33.5 Postoperative radiotherapy for endometrial carcinoma. (a) DRR with anterior
BEV. (b) Axial slice colourwash of the dose distribution using four-beam arrangement.
■ Complex
Pelvic IMRT (Fig. 33.6) has been shown to reduce the dose to bladder, rectum
and small bowel by 20–50 per cent. This is especially important in the
390
Dose-fractionation
Figure 33.6 Stage IIIc carcinoma
of endometrium with inoperable
pelvic nodes – IMRT plan for
postoperative EBRT.
postoperative situation, with reduction in acute and late morbidity. It may help to
limit dose to pelvic bone marrow in patients undergoing chemoradiation.
Dose-fractionation
External beam irradiation
45–50.4 Gy in 25–28 daily fractions of 1.8 Gy given in 5–51⁄2 weeks.
Vaginal vault brachytherapy as boost after EBRT for patients

with cervical involvement
LDR
10–15 Gy at 0.5 cm from the surface of the applicator.
HDR
8 Gy in 2 fractions at 0.5 cm from the surface of the applicator given at least
2 days apart.
Vaginal vault brachytherapy as sole adjuvant treatment for

selected intermediate risk patients
LDR
30 Gy at 0.5 cm from the surface of the applicator.
HDR
22 Gy in 4 fractions at 0.5 cm from the surface of the applicator given at least
2 days apart.
Unoperated stage I and II disease
EBRT
45 Gy in 25 daily fractions of 1.8 Gy given in 5 weeks followed by intracavitary
irradiation.
391
UTERUS
Intracavitary irradiation
LDR 20 Gy to point A single insertion
or
HDR 12 Gy in 3 fractions.
Inoperable stage III disease

EBRT
50.4 Gy in 28 daily fractions of 1.8 Gy given in 51⁄2 weeks followed by intracavitary
irradiation.
Intracavitary irradiation
LDR 22.5 Gy to point A single insertion
or
HDR 14 Gy in 2 fractions.
EBRT boost
Rarely, if intracavitary treatment is not feasible EBRT may be followed by an
additional dose to the GTV using a conformal plan.
15–20 Gy in 8–11 daily fractions of 1.8 Gy given in 11⁄2–2 weeks.
Intracavitary radiotherapy alone for primary disease, if unfit

for EBRT
LDR
50 Gy given in 2 fractions in 15 days.
HDR
33 Gy in 5 fractions given in 15 days.
20 Gy in 5 fractions given in 1 week.
8–10 Gy as a single fraction for haemostasis.

The patient is treated using the same position, immobilisation system and bladder
filling protocol as for localisation, aligned using three skin tattoos and orthogonal
laser lights.
Acute side effects may be increased when radiotherapy is combined with
concurrent chemotherapy. Diarrhoea or abdominal discomfort may occur during
EBRT, and is usually controlled by a low residue diet and loperamide hydrochloride.
If these symptoms are severe, treatment may be interrupted to allow recovery. Care
of the perineal skin is advised, but erythema and desquamation are now uncommon
with 10–16 MV beams, unless the lower vagina is affected. Severe perineal or natal
cleft skin reactions are treated with 1 per cent hydrocortisone cream and Intrasite gel.
392
Information sources
Less commonly rectal discomfort or bleeding, urinary frequency or dysuria may
occur and a urinary specimen should be taken to exclude infection.
A good fluid intake should be encouraged.
The risk of lymphoedema is increased when EBRT is given to patients who have
undergone BPLND. Premenopausal patients will lose ovarian function, and
hormone replacement therapy may be given to improve menopausal symptoms
and prevent osteoporosis. Psychosexual counselling may be helpful, as well as
vaginal hydration gels and dilators to improve vaginal function. Chronic fatigue,
urinary or faecal urgency or incontinence and pelvic pain are uncommon late
effects of pelvic radiotherapy and require special care.
Verification
Portal images are taken on the first 3 days of treatment using EPIs or portal films
and compared with simulator or DRR images with a verification strategy protocol.
Any systematic errors of 5 mm are identified and corrected, and weekly EPIs
then taken. Exit dosimetry using diodes is performed on the first day in all patients.
Treatments in large patients may show a high random error and then daily portal
imaging may be necessary.
key trials
Creutzberg CL, van Putten WL, Warlam-Rodenhuis CC et al. (2004)

Postoperative Radiation Therapy in Endometrial Carcinoma (PORTEC) Trial.
Outcome of high-risk stage IC, grade 3, compared with Stage I endometrial
carcinoma patients. J Clin Oncol 22: 1235–41.
Kitchener H, Redman CW, Swart AM et al. (2006) ASTEC – A Study in the
Treatment of Endometrial Cancer: a randomised trial of lymphadenectomy in the
treatment of endometrial cancer. Gynecol Oncol 101: S21 (abstract).
Nout RA, Putter H, Jürgenliemk-Schulz IM et al. (2008) Vaginal brachytherapy
versus external beam pelvic radiotherapy for high-intermediate risk endometrial
cancer: results of the randomised PORTEC-2 trial. J Clin Oncol 26(suppl):
abstract LBA5503.
Information sources
Churn M, Jones B (1999) Primary radiotherapy for carcinoma of the endometrium using external
beam radiotherapy and single line source brachytherapy. Clin Oncol 11: 255–62.
Kong A, Johnson N, Cornes P et al. (2007) Adjuvant radiotherapy for stage I endometrial cancer.
Cochrane Database Syst Rev (2): CD003916.
Small W Jr, Mell LK, Anderson P et al. (2008) Consensus guidelines for delineation of clinical target
volume for intensity-modulated pelvic radiotherapy in postoperative treatment of endometrial
and cervical cancer. Int J Radiat Oncol Biol Phys 71: 428–34.
393
34 Vagina

Primary carcinomas of the vagina are rare, with 80 per cent occurring in women
over 60 years as SCCs. Clear cell adenocarcinomas of the vagina are seen in young
women who are exposed to oestrogen therapy in utero. More commonly, tumours
found in the vagina are an extension from primary cervical, vulval or urethral
tumours, local spread from the rectum or bladder or metastatic lymphatic or
vascular deposits from endometrial, breast or lung carcinomas. Surgery is the
mainstay of treatment for malignant melanomas of the vagina, which have a high
incidence of distant metastases and poor overall survival rates.
Vaginal intraepithelial neoplasia (VAIN) is treated with local laser ablation,
excision or radiotherapy. Surgery is preferred in younger patients with intraepithelial
‘field change’ as this can be associated with cervical or anal cancers. It also avoids
a radiation-induced menopause. Older patients with multifocal or multiply recurrent
disease, or those who are unsuitable for organ-preserving surgery, can be equally
effectively treated with vaginal brachytherapy.
Radiotherapy is the treatment of choice for most patients with vaginal tumours,
because radical surgery may cause loss of function of bladder and/or rectum.
Small, superficial, stage I lesions less than 3 cm in size are treated with brachytherapy
alone. All other stages of the disease require either combined EBRT and
brachytherapy or EBRT alone to treat the primary tumour and lymph nodes.
Concurrent chemotherapy with cisplatin is used for locally advanced disease.
Radiotherapy gives overall survival rates of 44–77 per cent, 34–48 per cent,
14–42 per cent, and 0–18 per cent for stage I, II, III and IV disease, respectively.
Most relapses after radiotherapy occur locally and cure rates are dose dependent.
Maximal conformality of treatment is therefore an advantage. Palliative EBRT is
given to control bleeding or pain from advanced primary tumours or vaginal
metastases. Intracavitary or interstitial brachytherapy can also be used for palliation
of vaginal mucosal disease and is particularly useful in patients with malignant
melanoma.
Surgery has a role in selected patients with stage I disease, and usually involves
a radical upper vaginectomy and bilateral pelvic lymphadenectomy with radical
hysterectomy if the uterus has not previously been removed. Adjuvant
radiotherapy should be considered if surgical margins are close or positive. Pelvic
exenteration with vaginal reconstruction is sometimes chosen for more advanced
stages, residual or recurrent disease. Surgery is the treatment of choice if a vesico-
vaginal or recto-vaginal fistula is present at diagnosis. Ovarian transposition before
radiotherapy can help to preserve ovarian function in the premenopausal patient.
394
More than 50 per cent of vaginal tumours arise in the upper third of the vagina
and from the posterior vaginal wall and spread laterally into paravaginal tissues
(Fig. 34.1). They invade directly into the adjacent bladder and rectum. Lymphatic
spread from the upper part of the vagina is to the lower cervix, obturator and
hypogastric nodes, from the posterior wall to presacral and para-rectal deep pelvic
nodes and from the anterior wall to the lateral pelvic wall nodes. The lymphatics
of the lower third of the vagina drain by vulval lymphatics to the inguino-femoral
and iliac nodes. Lower third tumours can also directly invade the urethra or anus.
Of patients with stage II disease, 25–30 per cent have positive pelvic nodes,
necessitating EBRT as well as brachytherapy.
(a) (b)
Figure 34.1 T2-weighted MR images of a patient with stage II clear cell adenocarcinoma
of the vagina arising in the right posterior wall of the upper third. (a) Axial and (b) sagittal
views.

Clinical examination of the entire vaginal mucosa and cervix is made by colposcopy
and multiple biopsies taken from any suspicious areas as well as the primary
tumour. For patients with invasive carcinoma, staging and assessment must include
cystoscopy, hysteroscopy, endometrial curettage, cervical biopsy, bimanual
examination and sigmoidoscopy where indicated. The inguino-femoral region is
examined for lymphadenopathy and needle aspiration of palpable nodes performed
to obtain cytology. CT or MRI can be used to detect ureteric obstruction and
lymph node involvement, which do not influence the FIGO staging system but are
used to plan treatment. Tumour stage (FIGO) is the most important prognostic
factor, with tumour grade, volume extent within the vagina and presence of
lymphovascular invasion all influencing survival. Adenocarcinomas and tumours of
the lower third of the vagina have a worse prognosis.
395
VAGINA
Data acquisition
■ CT scanning
Patients are scanned supine with legs abducted where appropriate to reduce
perineal skin reactions. Movement is limited by leg immobilisation with anterior
and lateral tattoos used to prevent lateral rotation. Radio-opaque markers are
placed by clinical examination in the treatment position at the inferior extent of
vaginal tumour to define the inferior GTV margin and at the introitus (Fig. 34.2).
Variations in bladder volume have been shown to influence mobility and position
of the uterus, cervix and vagina and a protocol should therefore be used to
maintain a constant bladder filling – ‘comfortably full’. Patients are asked to empty
the bladder and drink 200 mL water 20 min before the scan and before treatment
each day. Patients should be encouraged to empty their bowels regularly before
CT scanning and daily treatment. Intravenous contrast is used to enhance pelvic
blood vessels, which are used as surrogates for pelvic lymph nodes when designing
the CTV-N. CT scans are acquired from L3 to 5 cm below the introitus.
Figure 34.2 DRR with radio-opaque

marker at the introitus to show anterior BEV.

■ CT planning
GTV-T is the primary vaginal tumour as defined by clinical examination, EUA,
and co-registered MRI or CT scans. CTV-T includes the entire vagina, cervix and
surrounding paravaginal tissues. For tumours of the lower third, the introitus is
also included.
CTV-N includes different pelvic lymph nodes below the common iliac nodes
depending on the site of the primary tumour:
■ for tumours of the upper two-thirds of the vagina – obturator, external and
internal iliac, presacral and para-rectal lymph nodes
■ for tumours of the lower third of vagina – inguino-femoral and distal external
iliac nodes
■ for tumours involving the posterior vaginal wall – presacral and para-rectal deep
pelvic nodes.
396
Dose solutions
These are delineated by identifying contrast-enhanced pelvic blood vessels on
each CT scan and using a 7 mm margin to create a 3D CTV-N.
A typical CTV to PTV margin of 10–15 mm is used around the CTV-T to allow
for organ motion of cervix and vagina and measured set-up uncertainties. For
CTV-N, organ motion occurs to a lesser extent and a 7 mm CTV to PTV margin
is usually sufficient for set-up variations (Fig. 34.3). Bladder, rectum, small bowel
and femoral heads are all outlined as OAR. For brachytherapy:
CTV-T GTV-T 2 cm margin. (34.1)
After EBRT, brachytherapy is used to deliver a further localised high-dose boost

to residual GTV for stage I and II disease. When volume of residual tumour
precludes brachytherapy (e.g. stage III disease), repeat imaging is used to define a
new GTV-T. Further EBRT is given using a conformal plan with a 15–20 mm
margin for the CTV-T and a further 10 mm margin for the PTV.
■ Conventional
Patient positioning, immobilisation and bladder protocol are used as described
above. Where CT is not available for planning, all clinical, surgical, histological
and radiological data must be used to define tumour extent before locating the
beam borders using a simulator. Palpation of the primary tumour is carried out with
the patient in the treatment position and the inferior tumour extent and introitus
marked with radio-opaque material. AP and lateral simulator films are taken.
For tumours of the upper two-thirds of the vagina, the CTV includes the whole
vagina, cervix and obturator, external and internal iliac, presacral and para-rectal
lymph nodes. The superior border of the beam is at the L5/S1 junction, lateral
borders are 20 mm lateral to the bony pelvic sidewall and the inferior border is at
the introitus. The anterior border should encompass the GTV-T and iliac nodes
superiorly and is usually at the anterior one-third of the symphysis pubis. The
posterior border is 20 mm from the GTV-T including posterior extension of
tumour and internal iliac nodes and is commonly situated 5 mm posterior to the
anterior border of the S2/3 vertebral junction. When tumours involve the
posterior wall of the vagina, presacral and para-rectal deep pelvic nodes are
included, and the border should be placed posterior to these nodes and the GTV.
For tumours of the lower third of the vagina the volume includes the whole
vagina, introitus, para-vaginal tissues, inguino-femoral and distal external iliac
nodes. The superior border is at the upper acetabulum to include inguinal nodes,
inferior border 30 mm below the introitus and lateral borders cover the femoral
heads to include femoral nodes.
Dose solutions
■ EBRT
For all tumours of the vagina, a conformal plan is used (Fig. 34.3) with MLC or
standard shielding to avoid normal organs, especially small bowel and posterior
rectum. IMRT may provide the best solution for pelvic nodal treatments.
397
VAGINA
Figure 34.3 Axial CT scan of

a patient with a vaginal
tumour showing CTV and
PTV of primary and nodes
(CTV purple, PTV green) with
dose distribution.
For conventional treatment of upper vaginal tumours, four MV beams are
commonly used, e.g. anterior, posterior and two lateral; for lower third tumours,
where CTV-N includes inguinal nodes, anterior and posterior opposing beams can
be used, unequally weighted to spare the rectum and increase dose anteriorly to
the inguino-femoral regions.
■ Brachytherapy
Cylindrical vaginal applicators can be used to deliver intracavitary brachytherapy
for VAIN or superficial tumours that are less than 5 mm deep, either at diagnosis
or after EBRT. Where tumours are more than 5 mm deep, either at diagnosis or
after EBRT, interstitial brachytherapy is required. Interstitial implants using the
Paris system are described in Chapter 5. Brachytherapy for tumours of the upper
third of the vagina is delivered with the same technique as for cervical carcinomas,
with a central intrauterine tube and vaginal applicators, as discussed in Chapter 32.
A brachytherapy boost to tumours of the middle and lower third of the vagina,
limited to the posterior wall, can be delivered using cylindrical applicators in both
the vagina and rectum.
■ Palliation
For palliation of locally advanced fixed and fungating vaginal tumours, either
EBRT or, when technically feasible, brachytherapy can be given. Radiotherapy is
limited to the GTV with a 15 mm CTV margin to reduce toxicity using a
conformal plan where possible or CT-simulated opposing beams.
Dose-fractionation
■ EBRT
Primary tumour and lymph nodes
45–50.4 Gy in 25–28 daily fractions of 1.8 Gy given in 5–51⁄2 weeks .
External beam boost if brachytherapy not feasible

15–20 Gy in 8–11 daily fractions of 1.8 Gy given in 11⁄2–2 weeks.
398
■ Brachytherapy
Brachytherapy alone
LDR
For VAIN, 60 Gy using intracavitary technique at 5 mm from surface of applicator.
For superficial invasive disease, 65–70 Gy using intracavitary technique at 5 mm
from surface of applicator in two insertions approximately 2 weeks apart.
For invasive disease, 65–70 Gy to 85 per cent isodose using an interstitial
technique according to the Paris system.
HDR
33 Gy in 6 fractions given in 10–19 days.
Using intracavitary technique, prescribed at 5 mm from the surface of applicator.
Using interstitial technique, prescribed to 85 per cent isodose using Paris system.
Brachytherapy after EBRT

LDR
15–20 Gy at 5 mm from the surface of applicator or to point A if using intracavitary
technique.
If there are applicators in the vagina and rectum, 15–20 Gy are prescribed at the
MPD between the surfaces of the two applicators.
20–25 Gy to 85 per cent isodose using Paris system using interstitial technique.
HDR
11 Gy in 2 fractions or 16.5 Gy in 3 fractions using either intracavitary or
interstitial techniques.
EBRT
20 Gy in 5 daily fractions given in 1 week
or
8–10 Gy as a single fraction for haemostasis.
LDR brachytherapy 20–30 Gy.
HDR brachytherapy 18–24 Gy in 3–4 fractions.

Patients are treated with the same bladder and bowel protocol and immobilisation
devices as at CT scanning or simulation and aligned using three tattoos and laser
lights to avoid lateral pelvic rotation. Care must be taken to check that the inferior
border of the treatment encompasses the distal extent of tumour.
Acute perineal and natal cleft skin reactions are common and can be severe, and
are treated with 1 per cent hydrocortisone cream. If moist desquamation occurs,
treatment may need to be suspended and Intrasite gel used to promote healing,
with diamorphine for pain control. Diarrhoea, abdominal pain, cystitis, proctitis
399
VAGINA
and rectal bleeding may occur. Infections are excluded, loperamide hydrochloride
prescribed and a low residue diet advised, as appropriate. Patients should be
warned in advance of the 4–10 per cent risk of vesico-vaginal or recto-vaginal
fistulae, which may occur particularly with more advanced tumours invading
bladder or rectal wall. Patients should be encouraged to use vaginal rehydration
gels and dilators to maintain vaginal function once treatment is completed
as vaginal fibrosis can lead to narrowing and shortening of the vagina. There is an
11 per cent risk of necrosis of the femoral heads at 5 years when opposing anterior
and posterior beams are used in an elderly population to treat inguinal nodes.
Verification
Portal films or EPIs are taken daily for the first 3–5 days to calculate any systematic
errors. Corrections are made to reduce the systematic error to 1 mm and a
weekly imaging protocol is then followed. Exit dosimetry is carried out using
silicon diodes on the first treatment day to verify dose delivered.
Information sources
Chyle V, Zagars GK, Wheeler JA et al. (1996) Definitive radiotherapy for carcinoma of the vagina:
outcome and prognostic factors. Int J Radiat Oncol Biol Phys 35: 891–905.
Mock U, Kucera H, Fellner C et al. (2003) High-dose-rate (HDR) brachytherapy with or without
external beam radiotherapy in the treatment of primary vaginal carcinoma: long term results and
side-effects. Int J Radiat Oncol Biol Phys 56: 950–7.
Perez CA, Grigsby PW, Garipagaoglu M et al. (1999) Factors affecting long term outcome of
irradiation in carcinoma of the vagina. Int J Radiat Oncol Biol Phys 44: 37–45.
Pingley S, Shrivastava SK, Sarin R et al. (2000) Primary carcinoma of the vagina: Tata Memorial
Hospital experience. Int J Radiat Oncol Biol Phys 46: 101–8.
400
Vulva 35

Patients with carcinoma of the vulva commonly present with early stage T1, T2,
N0 squamous cell carcinoma at a mean age of 70 years. Surgery is the mainstay of
treatment, ranging from wide local excision (WLE) for early stage disease to
radical vulvectomy for large or multicentric lesions, those within 1 cm of the
midline or involving the labia minora. Overall 5-year survival rates of 95 per cent
for stage I and 85 per cent for stage II disease are obtained. Unilateral or bilateral
inguino-femoral lymph node dissection (LND) is recommended where the
primary tumour invades more than 1 mm into the stroma and is chosen according
to the size and laterality of the primary tumour. For small, lateralised tumours,
defined as lying with a medial margin 1 cm from the midline, contralateral lymph
node metastases are rarely seen in the absence of an ipsilateral metastasis, so
unilateral LND may be performed. Where groin nodes are involved at surgery,
radiotherapy is given to the inguino-femoral and pelvic lymph nodes, as
randomised trial evidence has shown that pelvic radiotherapy is superior to elective
pelvic node dissection. Lymph node metastasis is the single most important
prognostic factor and results in a 50 per cent reduction in long-term survival.
■ Node negative
Adjuvant radiotherapy to the vulva alone is given to improve local control in
patients undergoing WLE where the tumour shows a histological margin of less
than 5 mm, re-excision is not possible and nodes are negative. If macroscopic
residual disease is shown in the vulva in node negative patients after WLE, a radical
local excision or vulvectomy is performed to obtain complete clearance. If this is
not possible, radiotherapy is given with or without chemotherapy.
■ Node positive
Postoperatively, adjuvant EBRT is given to inguino-femoral and pelvic lymph
nodes when there are one or more macroscopically involved nodes, two or more
microscopically involved nodes, or evidence of extracapsular spread.
Primary radiotherapy chemotherapy may be used for patients who are unfit
for surgery or with locally advanced disease, with or without subsequent surgery.
Studies are examining the role of more conservative surgery combined with
irradiation in selected patients with good prognosis in order to preserve bladder
and/or rectal function and improve quality of life.
Palliative radiotherapy is given for fungating disease, pain or bleeding at the
primary site or to the inguino-femoral regions. There is no role for radiotherapy in
401
VULVA
patients with vulval intraepithelial neoplasia (VIN), which represents a field change
and is treated with surgery.

For stage III and IVA disease, primary concurrent chemoradiation with cisplatin
5FU in patients with adequate renal function (GFR 50 mL/min) is currently
being investigated, followed 4–6 weeks later by surgery. No randomised trial
data are available yet comparing this with primary radiotherapy alone. Toxicity is
undoubtedly greater for the combined treatment but the response rates are
promising.

The vulva consists of the mons pubis, labia majora, labia minora, clitoris and
Bartholin’s glands. Most tumours of the vulva involve the labia majora. SCC
spreads to the superficial inguinal and then deep femoral nodes in about 20–30
per cent of patients presenting with early stage disease. Of these, 20 per cent are
shown to have pelvic lymph node metastases in addition. Tumour involvement of
the midline structures of the vulva, e.g. clitoris, may lead to bilateral groin node
spread, which is also present in 25–30 per cent of the patients with a positive
ipsilateral groin node. Occasionally there is direct spread to the pelvic nodes via
internal pudendal vessels. Direct local spread to the vagina, urethra, anus, bladder,
rectum and pelvic bones is less common than lymph node spread. Blood-borne
metastases to lung and bone are uncommon.

Careful clinical examination of the vulva and whole perineum is essential, with
multiple biopsies of any suspicious areas performed as well as a deep biopsy of the
primary tumour to assess depth of invasion. Excision of the entire lesion is not
advised as planning of subsequent treatment may then be difficult. Inspection of
the cervix and cervical cytology should also be carried out because of the multi-
centric pattern of disease.
SCCs account for 90 per cent of vulval tumours and, rarely, melanomas,
adenocarcinomas and basal cell carcinomas are seen. Extent of the primary tumour
is assessed by clinical examination and MRI (Fig. 35.1). The inguino-femoral
region is examined for lymphadenopathy. Fine needle aspiration of palpable nodes
is performed (with or without ultrasound guidance) to distinguish between
malignancy and infection. CT scanning can be used to stage pelvic lymph nodes,
but MRI with ultra small particles of iron oxide (USPIO) shows increased
sensitivity for detecting disease in normal sized nodes. Sentinel node biopsy is being
investigated and results correlated with full inguino-femoral lymphadenectomy.
Data acquisition
Patients are immobilised supine with legs abducted as much as possible, on a CT
scanner or simulator using lateral tattoos to prevent pelvic rotation. Where CT is
402
Target volume defination
(a) (b)
Figure 35.1 Stage III carcinoma of the vulva showing disease around the urethral catheter
suggestive of urethral involvement. (a) Axial and (b) sagittal T2-weighted MR scans.
available, it can be used to identify the primary tumour and inguino-femoral

and pelvic lymph nodes, using intravenous contrast to outline blood vessels. CT is
especially useful for measuring depth of lymph nodes when selecting electron
beam energies.
Using a simulator for locally advanced disease, clinical examination in the
treatment position is essential with radio-opaque material used to mark limits of
the macroscopic tumour and lymph nodes. The introitus should be marked with
radio-opaque material in all patients to aid localisation of the inferior border of the
CTV or field margin.

When the margin of WLE is less than 5 mm, if re-excision is not possible and
bilateral inguino-femoral LND is negative, radiotherapy is given to the vulva
alone. This is usually done with the patient in the lithotomy position, immobilised
using stirrups and poles, using detailed EUA findings and histopathology results
to define the target volume including the depth. The GTV has been excised and
the CTV includes the excision scar and remaining vulval tissues.
Following surgical excision and when nodal irradiation is indicated, the CTV
includes the remaining vulva, inguino-femoral nodes, and distal external and
internal iliac lymph nodes. Inguinal, femoral and iliac blood vessels can be used as
surrogates for nodes and a 7 mm margin created around vessels to define CTV-N.
The CTV-PTV margin used is dependent on measured set-up errors within each
department.
If the simulator is used, the superior border is placed above the acetabulum to
include distal external and internal iliac nodes, the inferior border 2 cm inferior to
the vulval marker, and lateral borders defined by palpation to the outer inguinal
ligament at the anterior superior iliac spine or to cover the femoral heads.
403
VULVA
For locally advanced disease, the GTV of the primary tumour and lymph nodes
are outlined with a 2 cm margin for the CTV-TN. The CTV for elective nodal
irradiation is delineated as described above.
Dose solutions
Radiotherapy to the vulva alone is delivered with the patient in the lithotomy
position, using electron therapy with good apposition of the applicator and bolus
to ensure adequate skin dose to macroscopic disease.
For photon therapy to the vulva and lymph nodes, the patient lies supine. The
target volume is irregular, lying anteriorly at the vulva and inguinal nodes, with
deep extension to the femoral and pelvic lymph nodes. Treatment is usually given
with conformal radiotherapy using four beams (Fig. 35.2) as this reduces toxicity
compared with anterior and posterior opposing photon beams since bowel and
rectum can be shielded. If opposing fields are used, unequal weighting, such as 2:1
anterior to posterior, may be used to increase dose to the anterior structures but
this arrangement may give unacceptable hot-spots. Bolus may be needed to ensure
adequate dose to gross primary tumour or positive surgical margins, particularly
with the skin sparing effect of higher energy beams.
(a) (b)
(c) (d)
Figure 35.2 Treatment of T3N2G2 locally invasive carcinoma of the vulva and
inguino-femoral and pelvic nodes showing GTV, CTV and PTV with bolus (pink), on
(a) lateral and (b) anterior DDRs and (c) sagittal and (d) axial CT scans with 4 beam
configuration. Courtesy of Dr Frances Calman and Marium Naeem.
404
Dose-fractionation
For locally advanced disease, additional treatment is given to the primary tumour
and palpable lymph nodes with a 2 cm margin. This can be done with reduced
anterior and posterior opposing photon beams or a perineal electron beam
localised to the primary vulval site and delivered in the lithotomy position.
Inguino-femoral nodes lie at a depth of 5–8 cm and a direct electron beam can be
used for this boost treatment with energy chosen using CT planning. Electron
therapy may be optimally combined with photons to ensure adequate treatment at
depth. Care should be taken to avoid overlap between electron beams used to treat
the vulva and boost treatment to the groin nodes.
Interstitial radiotherapy may be useful as a boost treatment in selected patients,
particularly if the lower vagina is involved by tumour.
Dose-fractionation
■ Adjuvant radiotherapy to vulva alone, node negative,
surgical margin 5 mm
Electron therapy
■ Adjuvant radiotherapy to vulva alone, node negative,

residual macroscopic disease
Electron therapy
Boost to residual disease

Total dose
■ Adjuvant radiotherapy to vulva, inguinal and pelvic

nodes, node positive
EBRT
Electron therapy
Boost to inguinal nodes if multiple positive nodes or extracapsular spread:
15 Gy in 8 fractions of 1.875 Gy in 11⁄2 weeks.
Total dose
60–65.4 Gy in 33–36 daily fractions given in 61⁄2 weeks.
405
VULVA
■ Primary radiotherapy to primary tumour and nodes

(may be given as concurrent chemoradiation for
inoperable, locally advanced disease)
EBRT
Subsequently surgery is performed if possible or further radiotherapy to the
primary tumour and involved nodes given.
Electron therapy
Boost to primary tumour and palpable nodes:
15–20 Gy in 8–11 daily fractions of 1.8 Gy given in 21⁄2 weeks
or
interstitial therapy to give a boost dose of 20 Gy to primary tumour.
Total dose
60–65.4 Gy in 33–36 daily fractions.
■ Recurrent disease
EBRT
Electron therapy
Individualised boost:
15–20 Gy in 8–11 daily fractions of 1.8 Gy
or
interstitial therapy to give a boost dose of 20 Gy to primary tumour.
Total dose
60–65 Gy in 33–36 daily fractions.
Recurrent disease in the inguinal region only, if patient not fit for pelvic
radiotherapy.
50 Gy in 20 fractions using an individualised combination of photons and
electrons.
Photon or electron therapy
8–10 Gy in a single fraction for haemostasis.
406
Information sources
Care must be taken in aligning the patient, using immobilisation of the lower
limbs and lateral tattoos to prevent pelvic rotation.
Acute reactions of the vulval skin are common and can be severe. They should be
treated with 1 per cent hydrocortisone cream. If moist desquamation occurs,
treatment may need to be suspended and Intrasite gel used to promote healing with
diamorphine given for pain relief. Urinary frequency and dysuria, and proctitis and
diarrhoea are other possible side effects, which should be treated symptomatically.
Late vulval fibrosis and atrophy may occur and vaginal rehydrating gel can be
helpful. Lymphoedema can occur in up to 30 per cent of patients when inguino-
femoral surgery and radiotherapy are combined. Urethral stenosis can also occur as
a late effect. Limiting the total vulval dose to 65 Gy or less reduces the risk of skin
necrosis. There is an 11 per cent risk of necrosis of the femoral heads at 5 years, if
opposing anterior and posterior beams are used in an elderly population. This is
reduced by conformal planning.
Verification
A series of daily portal or EPI films is taken to assess the systematic and random
error rates and a correction strategy agreed. Checks must be made that all visible
and palpable disease is encompassed as planned by the dose solution chosen. Exit
dosimetry is carried out using silicon diodes or TLD on the first day of treatment
to verify dose delivered.
Information sources
Barnes EA, Thomas G (2006) Integrating radiation into management of vulva cancer. Semin Radiat
Oncol 16: 168–76.
Busch M, Wagener B, Schaffer M et al. (2000) Long term impact of post operative radiotherapy in
carcinoma of the vulva FIGO I/II. Int J Radiat Oncol Biol Phys 48: 213–18.
de Hullu JA, van der Avoort IAM, Oonk MHM et al. (2006) Management of vulvar cancers. Eur J
Surg Oncol 32: 825–31.
Homesley HD, Bundy BN, Sedlis A et al. (1986) Radiation therapy versus pelvic node resection for
carcinoma of the vulva with positive groin nodes. Obstet Gynecol 68: 733–40.
IGCS. Guidelines for Vulva Cancer. Available at: www.kenes.com/igcs/posters/P01d.htm (accessed
12 December 2008).
Montana GS, Thomas, GM, Moore DH et al. (2000) Preoperative chemo-radiation for carcinoma of
the vulva with N2/N3 nodes: a Gynaecologic Oncology Group study. Int J Radiat Oncol Biol
Phys 48: 1007–13.
Pecorelli S, Ngan HYS, Hacker NF (2006) Staging classifications and clinical practice guidelines of
gynaecologic cancers. Suppl Int J Gynecol Obstet 70: 207–312.
Tyring SK (2003) Vulvar squamous cell carcinoma: guidelines for early diagnosis and treatment. Am
J Obstet Gynecol 189: S17–23.
407
36 Sarcoma

Bone and soft tissue sarcomas are uncommon tumours (1 per cent of all
cancers). More than 50 per cent arise in the extremities, 35 per cent in trunk,
abdomen and retroperitoneum, and 10–15 per cent in other miscellaneous sites.
Staging systems are shown in Table 36.1.
Table 36.1 Staging systems for soft tissue sarcomas
Tumours may be staged using the TNM system and grading

I Grade 1–2 T1a–T2b
II Grade 3–4 T1a–1b, T2a
III Grade 3–4 T2b–T3
IV Any grade, any T, with metastases to nodes
or elsewhere or more commonly as

I Low grade of any size or depth
II High grade of 5 cm of any depth
III High grade 5 cm deep
IV Any grade with metastases
Grading takes into account mitotic count (3, 3–20 or 20 mitoses /10 high
power fields), necrosis and size.
A large number of tumour types (50) are recognised by histological examination,

immunocytochemistry, cytogenetics and molecular pathology techniques but the
histological subtype is of relatively little significance in planning radiotherapy. Tumour
size and anatomical site, which affect resectability, and tumour grade, are the most
significant prognostic factors. Deep rather than superficial tumour situation, and
recurrence after previous complete excision confer a worse prognosis.
Osteosarcoma and chondrosarcoma are treated with surgery with or without
chemotherapy. They are radio-resistant tumours and if surgery is impossible,
consideration should be given to referring the patient to an appropriate centre for
proton therapy. If radiotherapy has to be given, local control requires high doses
even after surgery or adjuvant chemotherapy.
Ewing’s sarcoma and rhabdomyosarcoma are discussed in Chapter 37.
Surgery is the most effective treatment to ensure cure of soft tissue sarcomas,
and the first intervention should remove all tumour with a wide margin. If
tumours cannot be completely excised (without amputation of the limbs)
radiotherapy will improve local control rates.
408
After complete excision of low grade stage I tumours with a 1 cm clear margin,
no further treatment is needed. It is not yet clear whether radiotherapy is
advantageous after complete excision of small (5 cm) high grade tumours.
Postoperative radiotherapy is given after complete excision of high grade
tumours which are more than 5 cm in diameter, and for all grades and sizes of
tumours after incomplete excision.
Inoperable tumours are treated with initial radiotherapy, with surgery
subsequently if marked tumour shrinkage makes this possible. In some sites,
surgery will never become feasible and radical radiotherapy is then the treatment
of choice. Chemotherapy may also be given initially in this situation to try to
reduce tumour bulk.
All patients who present with locally recurrent disease or whose initial excision
has been performed without proper staging and assessment will require
postoperative radiotherapy with or without re-excision.
Desmoid tumours which arise in young people should be treated by wide local
excision. However, sometimes relentless local progression of symptomatic disease
or repeated recurrence after inadequate excision may need to be treated with
radiotherapy. Giant cell tumours of bone should be treated surgically unless this is
impossible. Lymphoma of bone is usually treated with chemotherapy alone. Bone
lesions of Langerhans’ cell histiocytosis are sensitive to low doses of radiotherapy
but since their malignant nature is uncertain, curettage or intralesional steroids are
preferred.
Radiation induced soft tissue sarcomas may be treated with radiotherapy as part of
the planned management if appropriate. Such decisions must always be taken by the
multidisciplinary team who are able to balance risks and benefits for individual patients.
Palliation may be achieved with shorter regimens and higher dose/fraction,
although doses of less than 40 Gy in 3 weeks are unlikely to be adequate to relieve
symptoms or prevent local progression.

Chemotherapy may be used adjuvantly after surgery to try to prevent metastases in
high risk disease and has been shown to produce marginal benefit for local control
and survival. It may be given as first treatment to shrink inoperable primary tumours.
The timing of radiotherapy in relation to surgery is important. Preoperative
therapy has the advantage of a well-defined tumour volume with no risk of spillage
of tumour cells which would require wider treatment margins. Blood supply is
uninterrupted. For retroperitoneal tumours, preoperative radiotherapy may reduce
toxicity to the bowel which is displaced by tumour but postoperatively falls into
the surgical bed that has to be treated.
Clinical trials have shown that this approach produces comparable or better
tumour control rates than postoperative EBRT. It may increase the probability of
delay in wound healing. If preoperative EBRT is given, an interval of 4–6 weeks
should elapse between EBRT and surgery.
A boost to the tumour bed by brachytherapy at the time of surgery or
subsequently may improve local control rates. Intraoperative electron beam
therapy has been used for retroperitoneal sarcoma and produces better local
control rates and less bowel toxicity but with high rates of neuropathy.
409
SARCOMA

As soft tissue sarcomas may occur in any part of the body, detailed protocols must
consider site-specific factors as well as sarcoma-related ones. Relevant anatomical
details can be found in the various other chapters of this book. Spread is commonly
haematogenous to lungs or through lymphatics to regional lymph nodes.

Careful clinical examination of the whole body is followed by plain X-ray for bone
tumours to localise the tumour and assess the stability of the bone. For all patients,
the tumour is scanned with ultrasound and/or CT scanning, and a biopsy is done
at the same time under imaging control to plan treatment preoperatively. Both CT
and MR scans are used diagnostically to assess relation of tumour to bone (CT) and
to other soft tissues, most importantly nerves and blood vessels (MRI) (Fig. 36.1).
CT scans are used to rule out metastases in the lungs, which are the commonest site
of metastatic spread except in epithelioid sarcoma (regional nodes), and
angiosarcoma where subdermal spread makes treatment volume definition difficult.
(a) (b)
Figure 36.1 Preoperative MR scan of pleomorphic sarcoma in left vastus lateralis in
(a) coronal and (b) axial views.
Data acquisition
■ Immobilisation
Immobilisation techniques will vary with different tumour sites, but for the common
tumours in upper or lower limbs, individually prepared vacuum bags or Perspex
shells are used. Laser lights are used to minimise rotational movement, and cranio-
caudal movement should be prevented by appropriate foot or hand restraints.
410
■ CT planning
If possible, preoperative and pre-radiotherapy CT scans should be taken with the
limb in the same position, and pre- and postoperative CT and/or MRI co-registered
for planning. However, changes in muscle configuration following surgery and the
position of the scar will alter appearances significantly and it is essential for surgeon
and radiation oncologist to plan jointly all treatments for optimal local control and
functional outcome.

Postoperatively, a CTV must be designed which includes the initial GTV, any
likely sites of tumour dissemination from surgery (such as scar) and a margin to
encompass potential microscopic spread, which will vary for different tumour types
from 20 mm to 50 mm (Fig. 36.2). Clips placed at surgery may be helpful. With
improved imaging, a GTV-CTV margin of 20 mm may be adequate. Optimum
treatment volume sizes are being studied (in the VORTEX trial). Incomplete
excision of very large tumours (which may be 15–20 cm long in the limbs) may
make the CTV planned this way prohibitively large, and it may not be possible to
encompass scars and excision margins in full. EBRT may then be restricted to areas
of bulky residual disease, for example around vessels or nerves.
(a)
Figure 36.2 Axial CT scans: (a)

preoperative showing same tumour
(T); and (b) postoperative with CTV
(b) and PTV.
A CTV-PTV margin of 5–10 mm is added for set-up variability. In sites (such as

ribs) where respiration is important, techniques appropriate to these sites should
be used. OAR will also vary depending on the primary tumour site, and tolerance
doses should be noted from relevant chapters. PRVs depend on balancing relative
risks of recurrence or normal tissue toxicity in individual cases.
411
SARCOMA
Dose solutions
■ Conformal
As long as 3D imaging and planning are available, all treatments should be planned
conformally, even though in some cases, beam arrangements may still be simple
(Fig. 36.3).
(a)
(b)
Figure 36.3 (a) Axial and (b) sagittal dose distribution for treatment of tumour.
■ Complex
IMRT is particularly useful for soft tissue sarcomas arising in sites such as the
retroperitoneum where a steep dose gradient (for example between tumour and
kidney) is needed or where a concave or convex dose distribution will spare vital
tissues such as spinal cord. If respiratory movement is significant, a method of
gating should be considered (see Chapter 2).
412
Dose-fractionation
■ Conventional
Treatment plans will need to be individualised although simple arrangements of
opposing or wedged pairs of beams may be used for limb lesions. Care should be
taken in this case to leave a strip of normal tissue unirradiated on one or both sides
of the limb to prevent late lymphoedema from fibrosis of lymphatic channels.
Radiation to a joint should be avoided if possible.
■ Brachytherapy
Catheters are inserted into the tumour bed at the time of surgery, thus ensuring
that they are correctly placed using the Paris rules (Chapter 5). They are not
loaded until the sixth postoperative day to permit wound healing. With low dose
rate iridium wire, 45 Gy are given over 4–6 days as the only radiotherapy, or as a
boost of 15–20 Gy before 45–50 Gy EBRT. This approach may be particularly
useful for re-irradiation treatments.
Dose-fractionation
■ Postoperatively
This dose is recommended for maximal tumour control although it will be
associated with a risk of functional impairment from fibrosis. Delivery of some of
this dose by brachytherapy may be beneficial (15–20 Gy).
■ Preoperatively
10–16 Gy in 5–8 daily fractions given in 5–10 days as a postoperative boost if the
surgical margins are positive.
■ Inoperable tumours using conformal, IMRT with

concomitant boost, or shrinking field techniques
GTV
66 Gy in 33 daily fractions given in 6 1⁄2 weeks.
CTV
Doses may have to be adjusted to give as high a dose as is compatible with
surrounding normal tissue tolerance.
■ Palliative treatments
40–45 Gy in 15–17 daily fractions of 2.67 Gy given in 3–31⁄2 weeks if normal tissue
tolerance permits.
■ Langerhans’ cell histiocytosis

8–10 Gy in 4–5 daily fractions given in 1 week.
413
SARCOMA
■ Giant cell tumour of bone

45 Gy in 25 daily fractions of 1.8 Gy given in 4–5 weeks.
Verification
Simulator films or DRRs are compared with port films or EPIs and doses are
verified with lithium fluoride dosimeters or diodes.

For lesions in the upper part of the adductor compartment in the thigh or the
sacral region, testis and natal cleft should be excluded or shielded.
Skin reactions should be prevented by good skin care and reactions treated with
aqueous or 1 per cent hydrocortisone cream. For retroperitoneal sarcomas,
antiemetics will be needed and loperamide for diarrhoea in pelvic/abdominal
tumours. Neuropathy may occur if high doses are used. Late effects of muscle
fibrosis can be reduced by physiotherapy to encourage movement during and for
6–8 weeks after treatment. Pathological fracture is commoner after postoperative
than preoperative radiotherapy (9 per cent compared with 2 per cent).
key trials
O’Sullivan B, Davis AM, Turcotte R et al. (2005) Preoperative versus

postoperative radiotherapy in soft-tissue sarcoma of the limbs: a randomised
trial. Lancet 359: 2235–41.
Pisters PW, Harrison LB, Leung DH et al. (1996) Long-term results of a
prospective randomised trial of adjuvant brachytherapy in soft tissue sarcoma.
J Clin Oncol 14: 859–68.
Rosenberg SA, Tepper J, Glatstein E et al. (1982) The treatment of soft-tissue
sarcomas of the extremities: prospective randomised evaluations of (1) limb-
sparing surgery plus radiation therapy compared with amputation and (2) the
role of adjuvant chemotherapy. Ann Surg 196: 305–15.
VORTEX, UK-opened 01/02/06. A Randomised Phase 3 Trial Comparing a
Two-Phase Conventional Radiotherapy Treatment with that of a Single Phase
Treatment to Selectively Spare Normal Tissue and Increase Limb Function
without Compromising Local Control in Soft Tissue Sarcomas.
Yang JC, Chang AE, Baker AR et al. (1998) Randomised prospective study of
the benefit of adjuvant radiation therapy in the treatment of soft tissue sarcomas
of the extremity. J Clin Oncol 16: 197–203.
Information sources
Pervais N, Colterjohn N, Farrokhyar F et al. (2008) A systematic meta-analysis of randomised controlled
trials of adjuvant chemotherapy for localised soft tissue sarcoma. Cancer 113: 573–81.
Sarcoma Meta-analysis Collaboration (1997) Adjuvant chemotherapy for localised resectable soft-
tissue sarcoma of adults: meta-analysis of individual data. Lancet 350: 1647–54.
Weiss SW, Goldblum JR (2007) Enzinger and Weiss’s Soft Tissue Tumors. Edinburgh: Elsevier.
414
Paediatric tumours 37
General considerations
Children who are cured of cancer have a long life expectancy and treatment-related
complications may be more severe than in adults. Although many paediatric
tumours are very radio-sensitive, current protocols of treatment are designed to try
to reduce toxicity of treatment for those who can be cured and to increase intensity
of chemotherapy for those with poor outcome. Radiotherapy is therefore only used
where it cannot effectively be replaced by surgery and chemotherapy. In very young
children, even when radiotherapy will improve cure rates, it is often now deferred
until the child is older and toxicity will be reduced because of greater maturity of
organs and tissues.
Normal tissues in the period of development are more radio-sensitive than adult
normal tissues, and the relative sparing effect of low dose per fraction for late
morbidity is therefore exploited using doses per fraction of 1.2–1.8 Gy. Sequential
studies in various tumour types have led to reductions in total dose and target
volumes. Nevertheless, radiotherapy is still an important component of curative
therapy for many tumour types and remains a valuable tool for easy and effective
palliation.

Radiotherapy is usually given after chemotherapy and/or surgery and therefore the
effects expected from radiotherapy alone may be modified.

All patients must be treated in specialist centres, using agreed national or
international protocols. Management must be discussed at an appropriate
multidisciplinary team meeting. As new drugs with potential synergistic or additive
effects with radiotherapy are used, radiation doses may need to be reduced further
for safety. Baseline assessment of any normal organ function that may be affected
by treatment must be undertaken to be used in monitoring long-term effects.
Immobilisation
Play therapists and specially trained nurses and radiographers are essential for
preparation of all children for radiotherapy. Close collaboration is needed between
paediatric and radiation oncologists. For very young children, adequate
415
PAEDIATRIC TUMOURS
immobilisation may be difficult to achieve without anaesthesia, which increases the

complexity of organising treatment. Dedicated and experienced anaesthetists and
designated theatre sessions are needed to assure that treatments can be given safely
and at the appropriate times. With adequate preparation, most children older than
3 years will learn to cope with the treatment satisfactorily without anaesthesia.
Immobilisation is achieved with thermoplastic or Perspex shells or vacuum
moulded bags as appropriate. Play specialists may help experienced mould room
staff in making adjustments to normal practice to allow children who were initially
apprehensive to become more at ease with the process.
CT scanning
With the patient immobilised in the treatment position, CT scans are acquired
through the region of interest with 3–5 mm slice thickness. Diagnostic images may
be co-registered with planning scans, which should include not only the whole
tumour but the whole of any relevant OAR.
■ Dose solutions
Treatment should always be as conformal as possible using CT and MRI as
appropriate if available. However, other considerations such as maintaining
symmetry and minimising dose outside the treatment volume may lead to simple
beam arrangements. IMRT solutions may produce better dose distributions for
some tumours, and better sparing of OAR (for example in the head and neck where
parameningeal sarcomas may be very close to the visual apparatus). However, they
may result in increased low doses to larger volumes. There is concern about the
potential risk of further increasing incidence of second malignancies.
It is important to maintain symmetry of growth of the musculoskeletal system
by treating vertebrae across their whole width including the transverse processes
and avoiding unfused epiphyses wherever possible. Treatment to one side of the
neck only may result in asymmetry of muscle growth. Sometimes, taking all these
factors into account, treatment with simple opposing MLC shaped beams may
remain the best choice. Full DVHs of OAR must be obtained (see Chapter 4).
Children should be treated by specialist teams working in collaboration with larger
international and national groups. Details of treatment prescriptions are given in
constantly revised protocols of the UK, European and American collaborative
groups, which should always be consulted.
If anaesthesia is needed, treatment should be given early in the morning to ensure
that the necessary period of fasting before general anaesthesia happens during the
night to avoid mealtimes. Attention must be paid to maintaining good nutritional
status especially if treatment causes anorexia, nausea or vomiting. Central line
access is needed. Medication for specific side effects for each tumour site is detailed
in relevant chapters. Appropriate adjustments in dosage are made according to the
weight of the child. The radiation oncologist should be involved in the follow-up
of any child treated with radiotherapy to document late toxicity, the expression of
which may be modified by continuing changes in chemotherapy practice.
416
Wilms’ tumour
■ Verification
EPIs are taken on the first day of treatment and for 3 days and sometimes weekly
throughout treatment to ensure reproducibility, especially if the child is awake
during treatment. Diodes or TLDs are used to verify doses delivered.
Wilms’ tumour
Wilms’ tumour occurs in about 8 per million children (80–90 cases per year in the
UK), 70 per cent before the age of 4 and 90 per cent before age 7. The majority of
Wilms’ cases are sporadic but there is a strong association between Wilms’, aniridia,
genitourinary malformation and mental retardation in the WAGR (Wilms’ tumour,
aniridia, genitourinary abnormalities and mental retardation) syndrome, and in
overgrowth syndromes such as Beckwith–Weidemann and hemihypertrophy.
Results of a series of randomised studies in Europe and the USA confirmed that:
■ preoperative radiotherapy reduces the risk of tumour rupture with increase in

disease-free survival, but not overall survival
■ preoperative chemotherapy is as effective as preoperative radiotherapy
■ 8 weeks’ chemotherapy is adequate for stage I disease and short treatments are
as good as long ones
■ two drugs (vincristine and actinomycin) are better than one
■ addition of a third drug (doxorubicin) improves outcome for poor prognosis disease.
Both actinomycin and doxorubicin are radio-sensitisers, so should not be given at

the same time as radiotherapy.
Treatment of this tumour is multimodal with surgery and chemotherapy, and in
some cases radiotherapy. Actual sequencing of treatment varies between American
and European studies, which have also used slightly different staging systems. The
risk of biopsy tract seeding led the European groups to advocate chemotherapy
without biopsy but American groups have been concerned about the 10 per cent
incidence of misdiagnosis based on imaging alone.
Treatments are stratified by stage, histological subtype and by treatment
response into low, intermediate and high risk groups. Preoperative radiotherapy
has been replaced by chemotherapy. Postoperative radiotherapy at a low dose is
given to patients with stage III favourable histology disease.
Those with unfavourable histology are treated with a higher dose. Those with
intermediate group disease are currently treated with radiotherapy and
chemotherapy with or without doxorubicin. If there is residual lung tumour after
preoperative chemotherapy in a child with metastases, whole lung radiotherapy
should be given. Radiotherapy is used for high risk primary tumour whatever the
chemotherapy response. This is given to the whole abdomen and pelvis if there is
diffuse tumour or gross rupture. Radiotherapy may also be used to treat liver
metastases which are unresectable or show incomplete response to chemotherapy,
and to treat bone and brain metastases regardless of chemotherapy response.
Survival rates are high – 90 per cent for stages I and II and 70 per cent for stage
IV, so there has been a major emphasis on reducing treatment and its toxicity and
417
PAEDIATRIC TUMOURS
on identifying disease with a poor prognosis for intensification of treatment.

Current studies are concerned with developing case adapted treatment based on
prognostic factors.

The commonest presentation is with an asymptomatic abdominal mass (75 per cent),
with pain (44 per cent), fever and haematuria occurring less commonly. The
diagnosis must be differentiated from neuroblastoma by measurement of
catecholamines and ultrasound and MRI. Staging includes chest X-ray and/or CT
scan of the chest and measurement of renal function by DMSA scan.

GTV is defined as the tumour volume after chemotherapy but before surgery using
co-registered MR-CT scans where available (Fig. 37.1). The CTV is the GTV
10 mm. Care is taken to irradiate the vertebrae symmetrically taking the volume
across the midline. PTV is individualised taking into account departmental
measurements of systematic errors.
Figure 37.1 Fused CT/MR images of

Wilms’ tumour of left kidney.
■ Dose solutions
Conformal
With virtual simulation or 3D definition of the GTV, AP/PA beam arrangements
with MLC shaping are often still appropriate to produce symmetrical irradiation of the
vertebrae, avoid the contralateral kidney and minimise whole body doses (Fig. 37.2).
Complex
IMRT is rarely needed and conformal treatment is often adequate.
Stage III favourable
10.8 Gy in 6 fractions of 1.8 Gy given in 11⁄2 weeks.
Stage III intermediate

418
Wilms’ tumour
(a) (b)
Figure 37.2 (a) Coronal DRR showing target volume and OAR for treatment of Wilms’
tumour of left kidney. (b) Axial CT scan showing AP/PA beam arrangement.
Boost
10.8 Gy in 6 daily fractions of 1.8 Gy if node positive or macroscopic residual disease.
Stage II, III high risk

25.2 Gy in 14 fractions of 1.8 Gy daily given in 21⁄2 weeks.
Boost
10.8 Gy in 6 fractions of 1.8 Gy to macroscopic residual disease.
Whole abdomen and pelvis

21 Gy in 14 daily fractions of 1.5 Gy given in 21⁄2 weeks with kidney shielding at 12 Gy.
Whole lung (Fig. 37.3)

15 Gy in 10 daily fractions of 1.5 Gy given in 2 weeks with lung correction.
Possible boost
10.8–14.4 Gy in 6–8 daily fractions of 1.8 Gy to residual disease after surgery.
Partial liver
Dose constraints to OAR

■ Whole lung – 15 Gy
■ Heart – 15 Gy
■ Ovary/testis – as low as possible (5 Gy)
■ Liver – 20 Gy to 50 per cent of total liver volume
If multiple areas need to be treated (i.e. flank plus lungs and/or liver) the
treatments should be simultaneous to avoid the overlap of fields which would
occur if the different areas were to be treated sequentially.
419
PAEDIATRIC TUMOURS
(a) (b)
Figure 37.3 Whole lung irradiation. (a) Coronal DRR. (b) Axial CT scan showing AP/PA
beam arrangement. All images courtesy of Dr M Gaze.
Neuroblastoma
Neuroblastoma is the most common extracranial solid tumour in children, with
80 per cent presenting before the age of 4 and 90 per cent before age 10 with a
median of 22 months.

Tumours are now grouped by image-defined risk factors and histology, and survival
is influenced by age and stage with overall 5-year survival rates of 44 per cent.
Histology ranges from ganglioneuroma, which resembles normal organ
architecture and requires little if any treatment, to undifferentiated neuroblastoma
with MYCN amplification, which carries a poor prognosis. Neuroblasts are known
to be radio-sensitive in vitro but show variable sensitivity in vivo. There are few
well-controlled trial data to define the role of radiotherapy but review of studies
undertaken showed reduction of risk of local relapse of between 22 per cent and
81 per cent. A study of chemotherapy alone versus chemotherapy and radiotherapy
has shown improved overall survival rates when both modalities are used (41 per cent
vs. 73 per cent). Indications for radiotherapy are shown in Table 37.1.
As these tumours may arise from a number of different sites, individualised
planning must take into account all relevant normal tissue tolerances and possible
late side effects.
Table 37.1 Role of radiotherapy in treatment of neuroblastoma
Radiotherapy No radiotherapy
Age 18 months ? Stage IV MYCN 10 Localised disease MYCN 10; stage IV
MYCN 10
Age 18 months Localised disease ? Localised disease MYCN 10
MYCN 10; stage IV
420
Neuroblastoma
Multimodality treatment includes initial chemotherapy followed by surgery, often
high dose chemotherapy and consideration of radiotherapy by EBRT, finishing
with immunotherapy with retinoic acid.
Studies are now looking at combinations of chemotherapy (topotecan),
MIBG (meta-iodobenzylguanidine) and EBRT and patients are referred to
specialised centres for these combinations. High dose chemotherapy is now
preferred to TBI for stage 4 (International Neuroblastoma Staging System, INSS)
or stage M (International Neuroblastoma Risk Group Staging System, INRGSS)
disease.
Thirty-five per cent of neuroblastomas arise in the adrenal gland, 30 per cent from
paraspinal ganglia, 19 per cent from the posterior mediastinal sympathetic chain
and the rest in the neck or other sites. They originate from the primitive adrenergic
neuroblast of neural crest tissues. Prognosis depends on histological type and
differentiation and measurement of MYCN amplification and loss of heterozygosity
(LOH) on chromosome 1p and 11q. Diagnosis is made on the basis of tumour
imaging and measurement of urinary catecholamines, with biopsy to confirm
histological subtype and for estimation of molecular markers of prognosis. MYCN
is not just a marker of prognosis but also determines risk stratification and
therefore influences the selection of treatment.

The GTV is defined as the post-chemotherapy but presurgical volume shown by
CT scanning.
CTV GTV 10 mm
PTV CTV 5 mm or according to departmental protocols.
Consideration of potential growth impairment may require CTV volumes to
extend across the midline to preserve symmetry and care must be taken since some
organs may be displaced following surgery. Beam sizes will be at least 6 cm.
■ Dose solutions
Simple opposing beam arrangements may give adequate tumour coverage and
minimal normal tissue dose, but shaping should be achieved with MLC if available.
Because of multiple constraints and the proximity of vital organs, inverse planning
and use of IMRT may be necessary to achieve an optimal plan.
Radioactive labelled MIBG can be a useful treatment if scanning shows good
uptake in tumour; 4 Gy to the whole body is given in two doses, the first
determined by body weight (444 MBq/kg). The actual whole body dose (WBD)
achieved is then measured and the second dose adjusted to achieve the final
desired WBD of 4 Gy.
421
PAEDIATRIC TUMOURS
Rhabdomyosarcoma
Seventy per cent of rhabdomyosarcomas occur before the age of 10 with the peak
incidence at 2–5 years of age. Radiotherapy improves local control rates compared
with chemotherapy alone and is still necessary for cure in many patients.
Concern about the late effects of radiotherapy has led to much reduced use,
although it is still important for local control for tumours of unfavourable
histology or site or those larger than 5 cm. Primary chemotherapy after initial
biopsy is tailored to known prognostic factors and many combinations of drugs
and alternating regimens are used. Surgery may be used for residual easily
resectable macroscopic disease. Where complete remission of disease is obtained
with chemotherapy, radiotherapy is deferred until relapse. If complete remission
cannot be achieved by cosmetically and functionally satisfactory surgical excision
after chemotherapy, radiotherapy may be given as part of the initial treatment.
Treatment decisions are always influenced by the need to minimise effects on
normal tissues and studies are still underway to refine the indications for each
component of the multimodality treatment.

These tumours may arise in many different sites in the body. Site of origin is
a significant prognostic factor. Favourable sites include the orbit, head and neck
(non-parameningeal) and paratesticular tumours. Unfavourable sites are
parameningeal, extremities, bladder/prostate, abdomen, pelvis, retroperitoneum
and trunk.
Histological subtype is also of prognostic significance with alveolar and
pleomorphic histology being associated with poorer outcomes than embryonal types.
Staging is according to the American Intergroup Rhabdomyosarcoma Study
definition:
1 – completely resected localised disease
2 – gross resection, microscopic residual
3 – incomplete resection, gross residual disease
4 – metastatic disease.
Risk groups are defined based on staging, site, size, age and histological type.
■ Data acquisition, target volume definition and

dose solutions
CT scanning and MRI are used to plan treatment with the patient in an
appropriate immobilisation device for the tumour site. General principles of
planning are applied depending on the site involved (see relevant chapters).
For treatment of residual disease after chemotherapy with or without surgery,
the GTV is taken as the pre-chemotherapy volume in most cases. Exceptions
include where a mass protruded into a body cavity without invasion (e.g. chest
wall lesion extending into pleural cavity displacing lung) and has now shrunk
back.
422
Ewing’s sarcoma
Some tumour sites such as orbit and limb are electively irradiated because of a
high risk of local relapse. GTV for treatment of relapse is defined by the volume
before re-induction chemotherapy. A CTV margin of 5 mm is allowed. PTV
margin is chosen according to site and departmental protocols.
The most conformal plan should be designed with careful MLC shielding of
normal tissues. Brachytherapy may be considered for vaginal tumours.
Treatment of primary disease residual after chemotherapy
Boost if indicated for bulky residual disease after chemotherapy/surgery.
9 Gy in 5 daily fractions of 1.8 Gy given in 1 week to macroscopic residuum.
Ewing’s sarcoma
Chemotherapy is essential for control of systemic disease which is the commonest
cause of failure. All patients are treated within national or international protocols
which use primary multi-agent chemotherapy, followed wherever possible by surgical
excision. Radiotherapy is an effective agent, which has produced similar local control
rates to surgery in randomised trials. Surgery is preferred for selected sites where
complete excision is feasible. Radiotherapy may be added after incomplete excision
and is a useful treatment for palliation of local disease when a child presents with
metastases or after failure of chemotherapy. Longer-term trials are needed to
determine optimal combinations of local treatment. Local radiotherapy to the
tumour site gives results equivalent to irradiation of the whole bone.
Ewing’s sarcoma in bone and its soft tissue variant, peripheral primitive
neuroectodermal tumour, are characterised by a diagnostic (t11; 22) (q24; q12)
chromosomal translocation. They occur commonly in the limbs or chest wall
(formerly known as Askin tumours) and pelvis, and present with pain, swelling and
sometimes systemic symptoms of fever and weight loss. They are diagnosed, as
other sarcomas, by MRI (which clearly demonstrates the extent of soft tissue
involvement which is a common finding), CT scanning and biopsy. Staging
investigations must rule out metastases in bone, lungs, bone marrow, lymph nodes
and other soft tissues. Lesions at a distance from the primary tumour may be found
(skip lesions). Volume of tumour and site are important prognostic factors, with
bulky pelvic lesions carrying the worst prognosis.
■ Data acquisition, target volume definition and dose

solutions
Planning is carried out using the relevant principles outlined in site-specific
chapters. MRI co-registered with CT scans gives the best definition of tumour
423
PAEDIATRIC TUMOURS
extent. GTV is defined as the pre-chemotherapy extent of disease. CTV is created

by adding a margin of 2–3 cm around the tumour for a first phase of treatment and
a boost given to the GTV with a smaller CTV margin of 2 cm. Since extension is
primarily along the marrow cavity, it may be appropriate to create anisotropic
margins with the greatest expansion in the long axis of the bone. Conformal
planning is used to give optimal dose distribution with sparing of normal tissues
including a strip of tissue along a limb to prevent oedema. Small bowel and
bladder must be outlined when treating large pelvic lesions.
Radiotherapy alone for local control
Boost
Radiotherapy after surgery


Principles are discussed in relevant site specific chapters. Careful multidisciplinary
surveillance is important during treatment. Diarrhoea and frequency of micturition
may occur with pelvic irradiation. Gentle exercise should be encouraged during
treatment of limb lesions to minimise fibrosis.
Lymphomas
There has been a gradual reduction in use of radiotherapy for lymphomas in
childhood with the development of new effective chemotherapy combinations and
increasing sophistication in diagnosis on the basis of molecular markers. In
Hodgkin lymphoma, benefit has been shown in the DHL-HD studies for IFRT of
20–25 Gy in some groups. The UK HD3 study compared chemotherapy and
radiotherapy in stage 1 disease and showed equivalence in terms of tumour
control. In view of the high risk of second tumours after radiotherapy,
chemotherapy is now preferred. Radiotherapy is used to treat PET positive residual
disease after chemotherapy; it may be given for palliation or as salvage therapy after
chemotherapy failure.

■ GTV 1 pre-chemotherapy disease extension
■ GTV 2 boost to post-chemotherapy disease extent (PET )
■ CTV GTV 5 mm margin
■ PTV according to tumour site based departmental protocols and adjacent
critical tissues
424
Information sources
■ Dose solutions
■ Conventional – AP/PA opposing beams.
■ Conformal – outline all nodes on CT planning scans, use MLC shaped beams.
■ Complex – this is not usually appropriate or necessary but may be considered for
bulky abdominal disease or that in proximity to vital organs.
PTV 1
(1.5 Gy fractions may be used for very young children or very large volumes).
PTV2
The testes should always be shielded during treatment of infra-diaphragmatic
disease.
Langerhans’ cell histiocytosis

This is a disease in which it is believed that a common, as yet unidentified,
pathogen triggers an aberrant immune response leading to lesions in single sites,
or in multiple sites in the same system (for example bone), or several systems (skin,
bone, lung, brain, etc.). These lesions are known to be radio-sensitive but immune
modulating systemic treatment is now preferred to radiotherapy, which has a
minor role for palliation of severe pain. For localised disease requiring treatment
where surgery is impossible, or to abort severe functional deficits, as in the
treatment of spinal cord compression, doses of 6–10 Gy in 1.5 Gy per fraction have
been recommended.
Information sources
Websites of various international and national collaborative groups are recommended for up-to-date
details of treatment protocols:
Children’s Cancer and Leukaemia Group (CCLG) (UK) – www.cclg.org.uk
Children’s Oncology Group (COG) – www.childrensoncologygroup.org
International Society of Paediatric Oncology (SIOP) – www.siop.nl
National Cancer Institute (NCI) – www.cancer.gov/clinical trials; clinical trials home page of the
National Cancer Institute (accessed 15 December 2008).
Pinkerton R, Matthay K, Shankar AG (eds) (2007) Evidence-based Pediatric Oncology, 2nd edn.
Blackwell Publishing, Oxford.
425
38 Systemic irradiation
Indications for systemic irradiation

■ Total body irradiation
The indications for allogeneic bone marrow or stem cell transplantation with total
body irradiation (TBI) as part of the conditioning are diminishing as more
effective high dose chemotherapy regimens have been developed and the late
effects of radiation are more widely recognised, especially in children.
Current indications for TBI as part of the conditioning regimen include the
treatment of young adults (40–45 years) with:
■ high risk acute myelocytic leukaemia (AML) in first remission

■ second remission AML (standard risk)
■ second remission acute lymphocytic leukaemia (ALL) if there is a human
leucocyte antigen (HLA) compatible sibling donor
■ first remission ALL with CNS involvement or Philadelphia chromosome
positivity
■ low grade lymphoma after chemotherapy failure
■ childhood AML/ALL in second or subsequent remissions.
The aim of treatment is to reduce the number of any residual malignant cells and
to permit engraftment of the re-infused donor marrow or stem cells. For tumour
control, the total dose of radiation should be as high as possible within the limits
of normal tissue tolerance. Ease of engraftment is related to the type of marrow or
stem cells infused. Less immune suppression is needed when the patient’s own cells
are re-infused than when fully compatible donor cells are used. High doses of TBI
appear to facilitate engraftment of cells that are not fully compatible, although this
type of procedure is less commonly used now.
■ Half body irradiation

High dose single fraction half body irradiation (HBI) is sometimes given as
palliative treatment for widespread painful bony metastases from carcinomas,
usually of the breast, bronchus or prostate, and can afford good symptomatic
relief, although survival is not prolonged.
Fractionated systemic irradiation is now rarely used for non-Hodgkin lymphoma
because new drugs are more effective. If used, sequential HBI is better tolerated
than TBI and is therefore preferred.
426
Data acquisition
Assessment of disease
Patients are treated according to national or international protocols and the
radiation oncologist must have access to these and to discussions of the relevant
multidisciplinary team to ensure the radiation is given appropriately, at the correct
time. Remission of the disease should be obtained where possible before TBI and
stem cell transplant is used. For patients with ALL, ‘boosts’ may be given in
conjunction with TBI to sanctuary ‘sites’, either prophylactically or when there has
been a previous relapse in these areas.
Patients with lymphoma who achieve a good response after chemotherapy but
who have residual tumour at sites of initial bulky disease have also been treated
with local boosts to these areas. However, TBI is less commonly used for these
patients now. TBI given after previous mediastinal or cranial irradiation will be
associated with a higher rate of complications.
In other situations, boosts may be given safely and most conveniently in a few
treatments before TBI. Total doses are determined according to the age of the
patient, the time interval since previous irradiation and the type and amount of
previous chemotherapy. Young age, short time interval (less than 6 months) and
high doses of chemotherapy are generally associated with increasing toxicity.
Because the chemotherapy component of conditioning will vary from one drug
(cyclophosphamide, melphalan) to various drugs in combination, the oncologist
must remain familiar with local haematological protocols and take into account any
potential interactions with radiotherapy. The doses specified here as within normal
tissue tolerance are when TBI is used in combination with cyclophosphamide.
Data acquisition
■ Conventional
TBI
A vacuum bag is prepared in which the patient can lie comfortably on their side
during treatment. The patient’s height and the required beam dimensions are
determined and measurements of body thickness (separation) are taken at multiple
sites (head, neck, upper and lower lung, mediastinum, abdomen, pelvis, thigh,
ankle) to calculate dose distribution and define dose prescription point at the
maximum lung dose.
HBI
The patient lies in the supine position with head supported on a comfortable head
rest. Beams extend from the head or the feet to cover the upper or lower half of
the body. The other border of the field is determined by the site of any disease,
but is usually at the level of the iliac crests. Lateral tattoos are placed over the iliac
crests and used to match beams if necessary.
■ Complex
If treatment is to be given with IMRT or tomotherapy techniques, CT scans of the
whole or half body should be obtained.
427
SYSTEMIC IRRADIATION

There is no GTV if the patient is treated in complete clinical remission and the
CTV is the whole body. Since with a conventional technique, beams of radiation
are used which extend beyond the patient at all points, a PTV is not defined. If
additional radiation is to be given to sites at high risk of microscopic residual
disease such as the testes, brain or mediastinum, which results in a cumulative dose
higher than that which can be given to the whole body, these may be designated
as additional CTVs with a dose specified. OAR such as the lungs, kidneys and eyes
may be shielded in some techniques and for complex treatment solutions should
be outlined to obtain DVHs and design MLC shielding.
Dose solutions
■ TBI – conventional
For treatment, the patient lies on one side facing a linear accelerator placed at an
extended distance to allow an adequate single beam to cover the whole body. The
collimator is rotated so that the maximum size of the diagonal of the beam (usually
about 140 cm) can be obtained. A Perspex shield placed at the edge of the treatment
couch between the patient and the machine provides skin build up (Fig. 38.1). As
leukaemic cells may infiltrate the skin, skin sparing is undesirable. The arms may
be positioned across the chest to act as bolus and help reduce lung doses. AP–PA
opposing beams are used to achieve an adequate MPD.
Figure 38.1 Patient

positioned at
extended focus–skin
distance for total
treatment with
dosemeters
positioned on back
for measurement of
upper and lower
lung, mediastinal,
abdominal and
pelvic doses.
A number of different TBI regimens have been shown to be safe and effective,
using the endpoints of engraftment, tumour control and incidence of
pneumonitis, the most common dose-limiting toxicity. From radiobiological first
principles, a TBI regimen should use the highest tolerable dose to control tumour,
and an LDR or fraction size to minimise normal tissue damage. Variability in
parameters of the tumour, growth characteristics, radio-sensitivity and differences
in the number of occult residual cells at the time of TBI make it difficult to
determine the best schedule for any individual patient. However, study data from
428
Dose-fractionation
Seattle, and the UK Medical Research Council protocols, suggest that doses of
12–14.4 Gy are safe and effective. Beam arrangements are not critical although it
is desirable to treat the whole body as a single volume.
Opinion is divided about the need to achieve a homogeneous dose distribution.
Some centres attempt to obtain homogeneity using tissue compensators for areas
such as the feet, between the legs, and neck where doses will otherwise be high.
There are, however, usually no consequences of these higher doses, which are
within tolerance of the relevant areas of the body. Some groups attempt to shield
lungs and kidneys as pneumonitis is a common and severe complication, and renal
function needs to be preserved as much as possible since many nephrotoxic drugs
are used in these patients. However, shielding at extended FSD in a patient who is
not well immobilised may lead to undesired shielding of potential sites of leukaemic
cell infiltration and has not been shown to reduce rates of lung complications.
■ TBI – complex
With more modern linear accelerators, field in field compensation, using open
fields with segmentation of the beam to give higher or lower doses where required,
and full IMRT (as for example may be delivered with tomotherapy) may permit an
optimised whole body dose distribution with relative shielding of some organs. CT
planning is then needed for adequate dosimetry and to determine how much dose
attenuation is needed for specific areas. However, leukaemic cells may involve any
organs of the body and doses need to be as high as tolerable to afford maximal
leukaemic cell kill. There is so far no clear evidence that these more sophisticated
techniques have led to improvements in outcome.
■ Half body – conventional

Treatment is given with a linear accelerator working at an extended FSD to give an
adequate beam size. The patient lies supine on the couch with legs bent if necessary,
and treatment is given with opposing lateral beams. If the patient is short, an
adequate field size may be obtained with them lying prone and then supine on a
mattress on the floor using anterior and posterior beams. With this technique, the
upper part of the body is treated with a gantry angle of 0° while the lower half is
treated with the gantry rotated 24° away from the junction, to minimise overlap of
beams and give a slightly larger beam size. If both halves of the body are to be
treated, the two beams should be matched as discussed in Chapter 2.
Dose-fractionation
■ TBI
Using a 6–10 MV linear accelerator beam at extended FSD, or rotational therapy:
■ Children
14.4 Gy in 8 fractions of 1.8 Gy specified as the maximum lung dose, treating
twice daily over 4 days with as long an interval as possible between fractions
(minimum 6 h).
429
■ Adults
12 Gy in 6 fractions of 2 Gy specified as the maximum lung dose, treating twice daily
for 3 days with as long an interval as possible between fractions (minimum 6 h).
Where there has been no previous cranial or testicular irradiation and a full
prophylactic dose is indicated, a boost may be given in the 3 days preceding total
body irradiation:
■ Cranial boost
Opposing lateral beams are used without immobilisation or shielding, using Reid’s
baseline (external outer canthus of eye to external auditory meatus) since
subsequent cataract is likely anyway from TBI.
5.4–6 Gy in 3 daily fractions.
■ Testicular boost
A single direct beam of orthovoltage radiation or electrons is used with the penis
taped as far out of the field as possible and the legs apart to avoid the skin of the
thighs.
5.4–6 Gy in 3 daily fractions.
■ Mediastinal boost (bulk disease with residuum after chemotherapy – no previous
radiation)
10–10.8 Gy in 5–6 daily fractions.
■ HBI (single fraction)

Upper half body
6 Gy MPD at the centre of the beam in the mediastinum (with lung correction)
given in a single fraction.
Lower half body

8 Gy MPD in a single fraction specified at the midpoint of the pelvis.
■ Fractionated HBI
3 Gy in 10 daily fractions given in 2 weeks (0.3 Gy/fraction).
If treatment to the other half of the body is to be given subsequently, an interval
of 6–8 weeks should elapse to allow time for bone marrow recovery.

During the first fraction of TBI, diodes or lithium fluoride dosemeters are placed
on the front and back of the patient in the positions shown in Figures 38.1 and
38.2. These remain in place throughout the first fraction. Before the next fraction
is given, measured doses are analysed to ensure that they are within tolerance. The
prescription treatment dose is specified as the maximum lung dose as this is the
dose-limiting normal tissue. For each fraction, treatment is delivered with
opposing anterior and posterior beams by rotating the couch.
430
Verification
Figure 38.2
Dosemeters placed
on front of patient in
positions
corresponding to
those in Figure 38.1.
Although these patients are immunosuppressed and at risk of infection, special

sterile precautions have been shown to be unnecessary. Premedication with steroid
and antiemetics (5-HT antagonist) given 30 min before each fraction will reduce
nausea and vomiting. Parotid swelling is common within the first 24 h of
irradiation but subsides spontaneously. Subsequent dry mouth and disturbance of
taste may persist for up to a year. Mild diarrhoea occurs from 4–5 days after the
start of treatment. Reversible hair loss, if not already present, starts after 10–14
days. Recovery of engrafted cells begins from day 7–21. A somnolence syndrome
of anorexia, lassitude, nausea or headache may occur from 6 to 8 weeks after
treatment and is self-limiting. Sterility may be expected but has not proved
absolutely inevitable and hormone replacement therapy will be needed for many
patients. Cataract is usual after several years.
Pneumonitis with cough and breathlessness from radiation alone occurs in 5
per cent of patients treated with this schedule. However, chest symptoms are
common. Their aetiology is multifactorial and effective treatment depends on
accurate diagnosis. The incidence of second malignancies after TBI and stem cell
transplantation increases with each year of survival and lifetime surveillance is
essential.
Verification
Doses to this irregular volume are measured by placing lithium fluoride
dosemeters or diodes in anterior and posterior pairs in various parts of the body
and checking dose distributions in the first fraction of the treatment. Cover of the
body is assured by checking the light beam on the wall behind the patient.
key trials
Thomas ED, Clift RA, Hersman J et al. (1982) Marrow transplantation for acute
nonlymphoblastic leukemia in first remission using fractionated or single-dose
irradiation. Int J Radiat Oncol Biol Phys 8: 817–21.
431
Information sources
Aristeia C, Tabiliob A (1999) Total-body irradiation in the conditioning regimens for autologous stem
cell transplantation in lymphoproliferative diseases. Oncologist 4: 386–97.
Kim TH, Khan FM, Galvin JM (1980) A report of the work party: comparison of total body irradiation
techniques for bone marrow transplantation. Int J Radiat Oncol Biol Phys 6: 779–84.
Leiper AD (1995) Late effects of total body irradiation. Arch Dis Child 72: 382–5.
Soule BP, Simone NL, Savani BN (2007) Post-transplant events – pulmonary function following total
body irradiation (with or without lung shielding) and allogeneic peripheral blood stem cell
transplant. Bone Marrow Transplant 40: 573–8.
Wheldon TE, Barrett A (2001) Radiobiological modelling of the treatment of leukaemia by total body
irradiation. Radiother Oncol 58: 227–33.
432
Appendix
List of abbreviations
5FU 5-fluorouracil EORTC European Organisation for
5-HT 5-hydroxytryptamine Research and Treatment of
ABC active breathing control Cancer
AFP -fetoprotein EPID/EPI electronic portal imaging device/
AP/PA anteroposterior/posteroanterior image
APER abdomino-perineal resection of EUA examination under anaesthesia
rectum EUS examination under ultrasound
ART adaptive radiotherapy control
BED biological effective dose FDG [18F]2-fluoro-2-deoxy-D-glucose
-hCG subunit of human chorionic FIGO Federation Internationale de
gonadotrophin Gynecologie et d’Obstetrique
CEA carcinoembryonic antigen FNA fine needle aspiration/aspirate
CFRT conformal radiotherapy FSD focus skin distance
CLD central lung distance GEC-ESTRO Groupe Européenne de
CLND completion lymph node Curiétherapie – European
dissection Society of Therapeutic
CNS central nervous system Radiation Oncology
CSF cerebrospinal fluid GFR glomerular filtration rate
CSRT craniospinal radiotherapy GTV gross tumour volume
CT computed tomography Gy gray
CTV clinical target volume HBI half body irradiation
DAHANCA Danish head and neck cancer HDR high dose rate
group HER2 antigen expressed in some
DCIS ductal carcinoma in situ breast cancers, of prognostic
Dmax dose at the depth of maximum significance in determining
build up response to Herceptin
DMSA dimercaptosuccinic acid–kidney HNPCC hereditary non-polyposis
scan colorectal carcinoma
DRE digital rectal examination HVL half value layer
DRR digitally reconstructed ICRU International Commission on
radiograph Radiation Units
DVH dose–volume histogram IFRT involved field radiotherapy
EBRT external beam radiotherapy IGRT image-guided radiotherapy
EBV Epstein–Barr virus IMRT intensity-modulated
EGFR epithelial growth factor receptor radiotherapy
ENT ear, nose and throat INRT involved node radiotherapy
433
APPENDIX
IORT intraoperative radiotherapy % DD percentage depth dose
IVU intravenous urogram PET positron emission tomography
kV kilovolts PRV planning organ at risk volume
LDR low dose rate PS performance status
LENT-SOMA Late Effects on Normal PTV planning target volume
tissues – Subjective, Objective, RCT randomised controlled trial
Management and Analytic RPA recursive partition analysis
scales RTOG Radiation Therapy Oncology
LOH loss of heterozygosity Group
LVSI lymphovascular space invasion SCC squamous cell carcinoma or
MALT mucosa-associated lymphoid spinal cord compression
tissue SF surviving fraction
MeV mega electron volts SPECT single photon emission
MIBG meta-iodobenzylguanidine computed tomography
MLC multi-leaf collimation/collimator SSD source skin distance
MPD midplane dose TBI total body irradiation
MRI magnetic resonance imaging TLD thermo-luminescent
MRS magnetic resonance dosemeters/dosimetry
spectroscopy TNM Tumour, Nodes, Metastases
MV mega volts TPS treatment planning system
MYCN antigen expressed in TSEBT total skin electron beam
neuroblastoma therapy
NCI-CTC National Cancer Institute – UICC Union Internationale Contre le
Common Toxicity Criteria Cancer
NTCP normal tissue complication USPIO ultra-small particles of iron
probability oxide
OAR organs at risk WBD whole body dose
OPG orthopantogram WHO World Health Organization
PCI prophylactic cranial irradiation WLE wide local excision
PEG percutaneous endoscopic
gastroscopy
434
Index
Note:
Abbreviations
BCC – basal cell carcinoma
PNET – primary neuroepithelial tumour
SCC – squamous cell carcinoma
SVCO – superior vena caval obstruction
Other abbreviations included in the index are explained on pages 433–434.
Page numbers in bold refer to figures; those in italics refer to tables
abdominal pain 321, 382, aesthesioneuroblastoma conventional planning 327,

392 186, 190 327–8, 328
accelerated fractionation 25, afterloading technique see CT planning 326–7
42 brachytherapy curative radiotherapy
head and neck cancers airway obstruction, SVCO 67 323–4, 329, 330
91–2 alcohol abuse 88 data acquisition and
accelerated hyperfractionation α-fetoprotein 67, 225, 361 immobilisation 326
42, 92, 93, 94 alternative therapies 5–6 dose-fractionation 330
Acculoc 346 alveoli, upper/lower, tumours dose solutions 329–30
ACT II trial, anal cancer 324, 124 electron therapy 329
326 anaemia 107, 382 histological types 323
actinomycin 417 anaesthesia, for radiotherapy indications for radiotherapy
active breathing control (ABC) 216, 416 323–4
11, 24 anal canal invasion/spread 324
acute effects, of radiotherapy anatomy 324, 325 key trials 324, 326, 331
on tissues 33, 45 exclusion, rectal cancer lymph node negative/
DVHs 47, 47 therapy 320 positive disease 328,
high α/β ratio values 33, 34 anal cancer 311, 323–31 328, 329
acute lymphocytic leukaemia 10–20 MV radiotherapy organs at risk (OAR) 327
(ALL) 426, 427 329 palliative radiotherapy 324,
adaptive radiotherapy (ART) anal canal tumours 323, 329–30
14–15, 24 324, 325, 326, 327 phase 1 volume, borders
bladder cancer 358 anal margin tumours 323, 327, 327, 329
adenocarcinoma see individual 324, 325, 326, 329 phase 2 volume borders
tumour sites assessment of primary 328, 328, 329
adenoid cystic carcinoma disease 326 postoperative radiotherapy
paranasal sinuses 186 beam arrangements 329 324, 329
salivary glands 178, 180, brachytherapy 56, 330 risk factors and survival
181 clinical/radiological anatomy rates 323
adverse effects see acute 324–5, 325 sequencing of multimodality
effects; late effects; side conformal/conventional therapy 324
effects therapy 329 staging 326
435
INDEX
anal cancer (continued) assessment and subtypes opposing (parallel) 19
target volume definition 72 single 19, 26
326–8 contraindications to wedged 19, 201, 203,
treatment delivery and radiotherapy 72 252, 253
patient care 330 data acquisition 72 see also individual tumour
verification of dose 331 dose-fractionation 78 types
analgesia 108, 236 dose solutions 75, 75–8 beam divergence 20, 21,
anal intraepithelial neoplasia electron therapy 73, 74, 105, 273
(AIN) 323 77–8 beam junctions 20, 20–1
anal margin 324, 325 eyelids 197 bellyboard devices 316, 316,
tumours 323, 324, 325, indications for radiotherapy 326, 330, 373
326, 329 71–2 benign prostatic hypertrophy
anatomy, clinical 6 low and high risk types 71, (BPH) 332
androgen deprivation, prostate 73 benzydamine hydrochloride
cancer 336, 336 patient immobilisation 73 (Difflam) 108
angiosarcoma 410 shielding 73, 73–4, 74 bicalutamide 336
cutaneous 85 superficial radiotherapy 73, bilateral pelvic lymph node
antiemetics 211, 301, 306, 309 75–7 dissection (BPLND)
testicular cancer calculation (sample) 384, 393
radiotherapy 364 76–7 biliary tumours 307–9
total body irradiation 431 dose-fractionation 78 see also
antioxidants 5, 52 dose solutions 75, 75–7 cholangiocarcinoma
anus stand-off and correction biological effective dose (BED)
anatomy/lymphatic drainage factors 76, 76 35
324–5, 325 surface dose rate and SSD calculation 48
tumours see anal cancer applicators 76, 77, 77 bladder
anxiety in patients 95 target volume definition 73 anatomy 353, 353
Argon plasma coagulation treatment delivery and cervical cancer spread
(APC) 346 patient care 78–9 371, 372
arm exercises 281 basal dose rate 58, 58 empty, bladder cancer
arteriovenous malformations BATS ultrasound system 346 radiotherapy 354
25 B cell lymphomas 84, 284 mobile organ 357
Askin tumour 423 beam, mixed (photons and size assessment in prostate
aspirin 6 electrons) 23, 262 cancer therapy 339
astrocytoma 205 beam arrangements telangiectasia 357
see also glioma anterior neck field 103, tolerance dose 50
ataxia telangiectasia 45 104, 105 bladder cancer 351–8
axillary lymph nodes 267, 268 co-axial opposing lateral adjuvant radiotherapy 352
breast cancer see breast 28 assessment of primary
cancer direct anterior beam, orbital disease 353–4
depth 274 tumours 198, 199 beam arrangements 355,
levels I-III 267, 268, 275 dual field, cutaneous 355–6
in lymphoma 286 lymphoma 84 carcinoma in situ 351
gaps between 20, 20 clinical/radiological anatomy
back scatter factor (BSF) 76, IMRT 23–4 353, 353
77 matching techniques conformal therapy 355,
basal cell carcinoma (BCC) 103–5, 129 355–6
71–9 multiple beams 19 conventional therapy 357
436
Index
curative radiotherapy 352 dose specification 26 dose-fractionation 380,
data acquisition 354 spine see vertebrae, 381
dose-fractionation 357 metastatic tumour Manchester system 56,
dose solutions 355–7 tumours with 68, 335 57, 379
dose–volume histograms pain see bone pain head and neck cancers 95
356, 356 sarcoma see sarcomas, soft high dose rate (HDR) 54
grade and staging 351 tissue and bone interstitial 54–5
immobilisation and bone marrow anal cancer 56, 330
preparation 354, 357 conditioning 426, 427 Manchester system 56,
IMRT/complex therapy suppression 292 57, 379
356 transplantation 426, 431 penile cancer 367–8,
incidence 351 bone pain 68–70 368
indications for radiotherapy data acquisition 69 intracavitary 55
351–2 dose-fractionation 70, cervical cancer 378–80
key trials 358 335, 345 dose reporting 59–60
muscle invasive, and dose solutions 69–70 intraocular tumours 196
lymphatic spread indications for radiotherapy iridium-192 see iridium-192
351–2 68–79 wire implants
organs at risk (OAR) 355, palliation in prostate cancer IR(ME)R practitioner,
355 335, 345 authorisation 61
palliative radiotherapy 352, target volume definition 69 isotopes 55
357 bone scans 212 legislation 60–1
prognostic factors 352 bowel management 339, lip cancers 113, 113
sequencing of multimodality 396 low dose rate (LDR) 54
therapy 351, 352–3 brachytherapy 54–61 LQ model application 35
survival 351 3D image-based, cervical Manchester system 56,
target volume definition cancer 378–9 57, 379
354–5 advantages 54 medium dose rate (MDR)
treatment delivery and afterloading technique 55, 54
patient care 357 57–9 mould 55
verification of dose 357–8 prostate cancer 348–9 nasopharyngeal cancer 156
virtual simulation 356, 356 SCC of hand 80 non-small cell lung cancer
bladder filling protocol, basal dose rate 58, 58 256
radiotherapy cervical cancer see cervical oesophageal cancer 301–2
cervical cancer 373, 382 cancer oral cavity tumours see oral
prostate cancer 338–9 clinical uses and sites 55, cavity tumours
uterine tumours 387 56 Paris system 56, 57–9, 413
vaginal tumours 396 contraindications 54 prostate cancer see prostate
blindness prevention 46 definition 54 cancer
blood tests 28 delivery systems 54–5 rectal cancer 311
Bloom’s syndrome 45 disadvantages 54, 132 ‘remote afterloading’ 55
bone dose reporting 59–60 sarcomas 409, 413
Ewing’s sarcoma 423–4 dosimetry 56–7 seed implantation into
giant cell tumours 409, EBRT with, oral cavity prostate 347–8, 348
414 tumours 124 squamous cell carcinoma
lymphoma 285, 289, 409 gynaecological 378–80 80
metastases applicators used uterine tumours 389, 389,
dose solutions 19 379–80, 389, 389–90 389–90, 391, 392
437
INDEX
brachytherapy (continued) breast, lymphatic drainage electron therapy 273, 278
vaginal boost 384 267, 268, 274 field borders of breast 272
vaginal tumours see vaginal breast bud radiotherapy, immobilisation 269,
tumours prostate cancer 336, 269–70, 272
vaginal vault 380, 381, 394 346 indications for radiotherapy
dose-fractionation 391 breast cancer 265–82 265–7
endometrial α/β ratio values 34 inoperable (T3/T4) 266,
adenocarcinoma 384, adjuvant radiotherapy 265, 268, 272
389–90, 391 278 internal mammary nodes
vulval tumours 56, 405 assessment of primary 278–9
brain disease 268–9 key trials 266, 281
metastases see cerebral axillary node involvement large breasts 269–70, 275
metastases 266, 268 locally advanced 272
tolerance doses 48, 49, 50 CT scanning 273–4, lung shielding 275, 279
whole, irradiation see whole 274 lymph node irradiation
brain radiotherapy dose-fractionation 280 266, 275, 278–9
brainstem, tolerance dose dose solutions 278 target volume definition
49, 50 target volume 274–5 274–5
brain tumours beam divergence, reduction lymphoma (primary) 265
beam junctions 20 273 mono-isocentric technique
choroid plexus carcinoma beam junctions 20 for breast/lymph node
211 bilateral tumours 265, 279 irradiation 20, 274,
craniopharyngioma 222–4 brachytherapy 56 274, 275, 279
ependymoma 217–18 implant calculation 59 organs at risk (OAR) 270
germ cell tumours see germ cardiac shielding 270–1, palliative radiotherapy 267,
cell tumours, intracranial 276, 277, 277, 279 280
germinomas 224–7 chest wall dose solution partial breast irradiation
gliomas see glioma 277–8 273, 279
infratentorial PNET see clinical/radiological anatomy post-mastectomy
infratentorial PNET 267, 268 radiotherapy 266
key trials 229 complex therapy and IMRT radiotherapy
lymphoma 220–1 275–7, 276, 280, 281 contraindications 269
medulloblastoma see conformal therapy 275–7 reconstruction after 272
medulloblastoma conservative surgery 265 dose solution/fractionation
meningioma 218–20 contralateral breast therapy 275, 280
metastatic 205, 212, 276–7 immediate,
227–9 conventional therapy/2D conformal/IMRT 278
see also cerebral outlines 272, 277 respiratory motion and 277
metastases CT scanning see computed sequencing of multimodality
oligodendroglioma 211 tomography (CT) treatment 267
sellar region 222–4 data acquisition 269–71 side effects of radiotherapy
stereotactic radiotherapy 25 whole breast, CT 281
suprasellar region 225 270–1, 271 simulator use 269, 272–3,
supratentorial 217–18 dose-fractionation 280 273, 275
WHO classification 205, dose solutions 275–9 isocentric technique 272
205 10MV 275, 278 lymph node treatment
whole brain irradiation see ductal carcinoma in situ 275
whole brain radiotherapy (DCIS) 265, 269, 280 spread 267
438
Index
supraclavicular lymph nodes repopulation 33, 35, 39, 40 applicators/insertion
273–5, 278 sterilisation 32, 36 379–80
target volume definition turnover 33 dose-fractionation 380,
272–5 see also normal tissue 381
treatment delivery and cell killing, radiation-induced fractionated HDR delivery
patient care 280–1 32, 36 380
tumour bed linear quadratic (LQ) model image-based 3D
dose solution/fractionation 26 (GEC-ESTRO) 378–9
278, 280 cell survival curves 32, 33 Manchester system 379
target volume 273 LQ model and 34–5 clinical/radiological anatomy
tumour bed boost melanoma cells 80 371, 372
radiotherapy 265–6, ratio α/β 32, 33, 34, 40 conformal therapy 378
273, 280 central lung distance (CLD) conventional planning/
verification of dose 281 19, 270, 272 simulator 373–5
virtual simulation 270–1, central nervous system (CNS) CT scanning 373
271, 275 in acute leukaemia 221–2 data acquisition 373–5
wide local excision (WLE) brain tumours see brain dose-fractionation 380–2
265 tumours dose solutions 378
breath-holding 11, 24, 250, key trials 229 early stage disease (stage Ia
305 lymphoma 211, 220–1, to IIa) 370, 371, 380
buccal mucosa 284 high-energy (10–16 MV)
brachytherapy 56, 132, spinal tumours see spine conventional therapy
132, 133 cerebral metastases 205, 378
tumours 124, 131, 132, 212, 227–9 immobilisation 373
133 assessment and data IMRT/complex therapy
acquisition 228, 228 378
CA19.9 303 dose solutions and indications for radiotherapy
caesium-137 55, 55 fractionation 229 370–1
calibration data 29 indications for radiotherapy key trials 383
Candida infections 108 227–8 locally advanced (stage IIb
carboplatin 93, 359, 360 lung tumours with 241, to IVa) 371
carcinoembryonic antigen 242 microinvasive (stage Ia1)
(CEA) 4, 315 malignant melanoma 81 370
cardiac shielding 270–1, target volume definition occult disease 371
276, 277, 277, 279 228–9 organs at risk (OAR) 379
cataract, induction 201 cervical cancer 370–83 palliative radiotherapy 371,
‘catch-up’ treatments 40 adenocarcinoma 370 377, 382
cauda equina syndrome 62 adjuvant radiotherapy 376, para-aortic radiotherapy
cell(s) 381 377, 377, 381–2
density, dose–cure anaplastic small cell tumours primary radiotherapy
relationship 37 370 375–6, 380
doubling times 40 assessment of primary shielding 374
number and treatment disease 372–3 spread 371, 372, 372
number relationship beam arrangements and stage IVb metastatic disease
36–7, 38 borders 374 371
proliferative sterilisation 32 boost EBRT 375, 381 survival rates 371
redistribution (proliferative brachytherapy 56, 371, target volume definition
cycle) 39 378–80 374, 375–7, 376
439
INDEX
cervical cancer (continued) cervix target volume in breast
brachytherapy 379 anatomy/lymphatic drainage cancer 272
primary radiotherapy 371, 372 chest X-ray 285
375, 375–6 tumours see cervical cancer children see paediatric
treatment delivery and cetuximab 94 tumours
patient care 382 CFRT (conformal radiotherapy) cholangiocarcinoma
treatment time 382 see conformal dose solutions and dose-
verification of dose 382 treatment fractionation 308–9
virtual simulation 376, 377, CHART fractionation schema indications for radiotherapy
377 42, 241 307
cervical cord compression chemoradiotherapy 5 target volume definition
62, 65 anal cancer 323, 324, 330 308
cervical intraepithelial neoplasia bladder cancer 353 chondrosarcoma 408
(CIN) 370 head and neck cancers choroid 197
cervical nodes, in head and 93–4 melanoma 196, 203
neck cancers laryngeal cancer 165, 166 metastases 196, 198, 203
adjuvant therapy 90, 123, oral cavity tumours 124 choroid plexus carcinoma
128 oropharyngeal cancers 211
curative therapy 90 134 circumcision 367, 369
delineation 99–102, 100, lung tumours 244, 251 cirrhosis 308
101, 102, 102 oesophageal cancer 294 cisplatin
extracapsular spread 99, pancreatic cancer 303 head and neck cancers
102 prostate cancer 337 92, 93–4, 158, 186,
laryngeal cancers 165, rectal cancers 311–12 294
170, 170 vulval tumours 401 vaginal/cervical carcinomas
levels 99, 101, 102 chemotherapy 5 371, 394
N1, N2 or N3 disease 90, adjuvant, bladder cancer classification systems 1–2
99 353 claustrophobia 95
nasopharyngeal cancer breast cancer 265, 267 clinical evaluation 1
158, 162 cervical cancer 371, 382 clinical target volume see CTV
node-negative (N0) 89, 99, lymphomas 220, 283, 284 (clinical target volume)
123, 128, 134–5, 139 nasopharyngeal cancer clips, surgical see radio-
oral cavity tumours 123, 156, 158 opaque clips
128 neoadjuvant, bladder cancer cobalt-60 isotope 55
oropharyngeal cancers 352–3 cobalt-60 machine, dose
134–5, 139 neuroblastoma 421 planning 18
paranasal sinus tumours radiotherapy with see cochlea, tolerance dose 49
187 chemoradiotherapy ‘cold’ schedule 34
prophylactic therapy 89, spinal cord compression cold triangle 20, 21
90, 91, 128 therapy 64 colorectal cancer 311
salivary gland tumours SVCO 67 see also rectal cancers
177, 183 see also induction columella, tumours 117, 118
see also individual tumour chemotherapy commissure tumours 111
sites chest wall completion lymph node
cervical nodes, in lymphomas convexity, standoff with 278 dissection (CLND) 86
286 dose-fractionation 280 complex treatment 19, 23–4
cervical spine lesions, dose planning in breast ‘field in field’ arrangement
immobilisation 69 cancer 277–8, 279 209
440
Index
see also IMRT medulloblastoma 212, 213 thyroid cancer 232, 233,
Compton scattering 15 meningioma 218, 219 237
computed tomography (CT) mesothelioma 259, 260 uterine tumours 386, 387,
15–16 MRI images co-registration 388
3D see three-dimensional 16 vaginal tumours 396, 397,
(3D) planning multi-slice scanners 16 398
4D scan, lung cancer 250 nasal tumours 118 virtual simulation 17
anal cancer 326–7, 327 nasopharyngeal cancer 160 vulval tumours 402, 403
bladder cancer 354 oesophageal cancer 296, whole brain 207
breast cancer 15, 268 297 see also three-dimensional
axillary/supraclavicular oral cavity tumours 125, (3D) planning
lymph nodes 273–4, 126 concomitant boost schedule,
274 oral contrast, bladder cancer head and neck cancers
immobilisation 269, 269 354 91–2, 92
whole beast treatment orbital tumours 199–200 conformal treatment 18,
270–1 organ motion reduction 21–3
cervical cancer 373, 374, protocols 16 anal cancer 329
375 oropharyngeal cancer 137, bladder cancer 355,
cone-beams, prostate 138 355–6
cancer 346 paediatric tumours 416 breast cancer 275–7
contouring 16–17 pancreatic cancer 304, cervical cancer 378
contrast media 15–16 305, 305 coplanar non-standard
conversion into relative paranasal sinus tumours configurations 22–3
electron densities 19 187, 188 definition/description 21–2
data transfer network 16, penile cancer 366 dose calculations 18, 19
17 PET with see under positron dose planning/solutions
diagnostic aperture 15 emission tomography 18, 21–3
ear tumours 115 (PET) glioma 209, 210
Ewing’s sarcoma 423 prostate cancer 338, head and neck tumours
fine needle biopsies 258, 339–40, 340 88, 103–5, 104, 105
268 brachytherapy 347 high/low beam energies
gliomas 207, 208 protocols 16 23
head and neck cancer 96, for radiotherapy planning hypopharyngeal cancers
97, 107 14, 15–16 152–3, 153
nodal volume delineation rectal cancer 315, 316 intracranial germ cell
99–102, 100, 101, 102, salivary gland tumours tumours 226
102 179, 180 laryngeal cancer 171–3
hypopharyngeal cancers sarcoma 410, 411 liver tumours, primary 307
149, 149 simulator 18 lung cancer see lung cancer
immobilisation for 14, 14 ‘slow scan’, lung cancer and medulloblastoma 213,
infratentorial PNET 212, lung motion 249–50 214–15
213 spinal 207 mesothelioma 261, 261
intracranial germ cell tumours spinal cord compression mixed beams 23
225, 225, 226 64, 65, 65 nasopharyngeal cancer
laryngeal cancer 169, 169 SVCO 67 162–3
lung tumours 241, 246, target localisation 14–15 oral cavity tumours 128–9
248, 248, 249, 250 testicular cancer 361 orbital tumours/disease
lymphoma 285, 286 thymic tumours 238 201
441
INDEX
conformal treatment Cotswolds’ modification of Ann dose-fractionation 83
(continued) Arbour classification dual field arrangement 84
oropharyngeal cancer 141, 285 total skin electron beam
141–2 couch, rotation, minimising therapy 82–3, 83
pancreatic cancer 306, 306 dose heterogeneity cyclophosphamide 427
paranasal sinus tumours 20, 430 cyproterone acetate 336, 336
191, 191–2 cough 292 cystectomy, radical 351, 352
parotid gland tumours management 263 cystitis, radiation 346, 357
182–3, 183 cranial boost 430 cystoscopy 373
pituitary adenoma 223–4 cranial nerve palsy 116, 158
planning see computed cranial radiotherapy DAHANCA 5–85 trial 94, 109
tomography (CT); three- prophylactic, small cell lung DAHANCA accelerated
dimensional (3D) cancer 242, 256 radiotherapy, head and
planning whole brain see whole brain neck tumours 91–2,
prostate cancer 343, 344 radiotherapy 92
rectal cancers 319, 319, craniopharyngioma 222–4 Danish Breast Cancer Trials
320 cystic recurrences 224 Group (DBCG) 266,
salivary gland tumours data acquisition and dose 278
182–3, 183 solutions 223–4 data acquisition 15–18
sarcomas, soft tissue and dose-fractionation 224 3D planning see three-
bone 412, 412 craniospinal radiotherapy dimensional (3D)
thyroid gland 235 (CSRT) 211 planning
tonsillar cancer 141, 141–2 ependymoma 218 contouring 16–17
uterine tumours 390 intracranial germ cell simulator use 17–18
vaginal tumours 397 tumours 224 virtual simulation see virtual
vulval tumours 404, 404 medulloblastoma and simulation
Wilms’ tumour 418, 419 infratentorial PNET 211 see also specific imaging
conformity index 13 cricoid cartilage 100 modalities
conjunctiva 198 CTV (clinical target volume) deafness 215, 216
lymphoma 196, 198, 199, 5, 9, 10 conductive hearing loss
203 consensus guidelines 10 184
contact lenses, lead 74, 74 CTV-N 10, 11, 99 decision-making 1
contouring 16–17 CTV-N50 10 Denonvilliers’ fascia 325,
contraindications to CTV-T 10, 11 337, 338
radiotherapy 72, CTV-T50 10 dental assessment, head and
79–80, 269 CTV-TN 10, 11 neck cancers 107
brachytherapy 54 see also individual tumours depth dose (DD), percentage
contrast media, CT 15–16 CTV-PTV margin 11, 12 18
conventional simulators calculation/formula 12 basal cell carcinoma 75, 75
17–18 CT for 14 electron therapy 21, 22
conventional treatment, dose population-derived standard desmoid tumours 409
planning 18, 19–21 12 diarrhoea, management 321,
cornea cure, dose relationship see 346, 382, 392
tolerance dose 49 dose–cure relationship TBI 431
ulceration 201 cutaneous lymphoma 82–4 DICOM 3 and DICOM RT
corpora cavernosa 366 B cell 84 protocols 16, 17
corpus spongiosum 366 dose-fractionation 84 diets/dietician role
corticosteroids see steroids T cell 82–3, 83 ear tumours 116
442
Index
esophageal cancer 301 minimising 20–1 complex treatment 19,
head and neck cancer 108 normal tissue see normal 23–4
mesothelioma 263 tissue see also IMRT
prostate cancer therapy optimal 18 conformal treatment 18,
planning 339 prostate cancer 27 21–3
thyroid cancer 236 proton therapy 25 conventional treatment 18,
digitally reconstructed verification 23, 28 19
radiograph (DRR) 16, dose-fractionation 25–6, 42 levels 18–19
28 2Gy 25, 34, 35, 36, 37 see also individual tumour
digital rectal examination (DRE) head and neck cancers types
315, 316 88, 91, 92 dose–volume histograms
anal cancer 326 accelerated 25, 42 (DVHs) 13, 43, 46–8,
Dmax 18, 26 altered (hyperfractionation 47
DNA damage 32 and acceleration) 42, bladder cancer 356, 356
repair 39, 45 92, 93, 94 brachytherapy 56
doctor’s delineation error 10, comparisons, LQ model see ear tumours 116
16 linear quadratic (LQ) IMRT/complex therapy
‘dog leg’ radiotherapy 359 model 343, 344, 345
testicular cancer 362, 363, effects on tumour and lung cancer 47, 47, 253,
364 acute/late effects 42 254
dose future directions 43 oesophageal, in lung cancer
specification 26–8 paediatric tumours 416 therapy 254–5
total 34, 40 tissue sparing by 35 oesophageal cancer 300
variations 41 see also accelerated oropharyngeal cancers
dose–cure relationship 36–8, fractionation; 143
37, 38 hyperfractionation; paranasal sinus tumours
cure probability, calculation hypofractionation; 194
36–7, 37 individual tumours prostate cancer 27, 343
increased dose, limitations dose-limiting effects 33, 34, rectal cancer 320
44 40 specific organs at risk
radiosensitivity and cell dosemeters, total body 49–50
number effects 37–8, irradiation 428, 429, dosimetry
38 430–1 brachytherapy 56–7
dose distribution dose planning see dose Manchester system 56, 57
brachytherapy 56 solutions/planning Paris system 56, 57–9
DVHs see dose–volume dose prescription point 26 protocols 30
histograms bone pain management 70 ‘double trouble’ 41, 43
electron therapy 21, 22 spinal cord compression 65 Douglas, pouch of 385
homogeneity, IMRT for 41 dose prescriptions see doxorubicin 417
ICRU reference point 26 dose-fractionation; dose dry mouth 292
maximum/minimum target solutions/planning; Dukes’ classification 3
doses 26 individual tumour types dysphagia 142, 256
non-uniformity/ dose reporting, brachytherapy brachytherapy for 301–2
inhomogeneous 26 59–60
beam junctions 20, 20–1 dose–response relationship ear tumours 114–17
brachytherapy 56 33 3D conformal planning 115
corrections 19 dose solutions/planning assessment of primary
electron therapy 21, 22 18–24 disease 114
443
INDEX
ear tumours (continued) bone pain 70 epiphyses, tolerance dose 49
beam arrangements breast cancer 273, 278 epithelial tissues,
115–16, 116 dose distribution 21, 22 dose–response
clinical/radiological anatomy effective treatment depth relationship 33
114 21, 22 epithelioid sarcoma 410
CT scanning 115 head and neck cancers equivalent uniform dose (EUD)
data acquisition 114–15 105, 106 48
dose-fractionation 116 intraoperative, sarcomas errors, systematic and random
dose solutions 115–16 409 12, 28
dose–volume histograms lip cancers 112 erythema
116 mesothelioma 258, 259, acute 79
immobilisation 114 261, 262 basal cell carcinoma
indications for radiotherapy nose/nasal tumours 118 treatment 79
114 penile cancer 365 breast cancer, reduction
organs at risk (OAR) 115 percentage depth dose 269–70
recurrence risk 115 21, 22 see also skin reactions
sequencing of multimodality photon beams with 23, ethmoid bone 198
therapy 114 262 ethmoid sinuses, tumour 97,
target volume definition 115 squamous cell carcinoma 188, 190
treatment delivery and 80 European LENT-SOMA
patient care 116–17 total skin see total skin classification 6
tumour spread and lymph electron beam therapy European Organisation for
nodes 114, 119 (TSEBT) Research and
verification of dose 117 vaginal tumours 405, 406 Treatment of Cancer
effective treatment depth, emergency radiotherapy (EORTC) 6, 266
electron beam therapy 62–70 Ewing’s sarcoma 423–4
of BCC 78 haemorrhage 70 external auditory canal
elderly indications 62 tumours 114, 115
basal cell carcinoma therapy spinal cord compression eye(s)
72 62–6 anatomy 197, 197–8
cutaneous angiosarcoma superior vena caval care, head and neck
85 obstruction 67–8 cancers 108
glioblastoma multiforme 206 endocavitary local contact shielding 74, 74
lip cancers 111 radiotherapy, rectal eye drops, lubricating 194,
electron beam radiotherapy cancers 311 204
see electron therapy endometrium 385 eyelids
electronic portal images (EPIs) carcinoma see uterine basal cell carcinoma 197
16, 28, 292 tumours squamous cell carcinoma
anal cancer 331 endoscopic ultrasound (EUS) 197
bladder cancer 357–8 296 eye shield, spade-shaped
prostate cancer 346 anal cancer 326 74, 74
rectal cancer 321 cervical cancer 373
uterine tumours 393 pancreatic cancer 304 face
electron therapy 21 endothelial cell proliferation 45 basal cell carcinoma 71,
anal cancer 329 enteral feeding 236 72, 73
applications 21 ependymoma 217–18 cutaneous angiosarcoma
BCC see basal cell epidermal growth factor 85
carcinoma (BCC) receptor (EGFR) 94 Fanconi’s anaemia 45
444
Index
femoral head clinical/radiological treatment delivery and
necrosis 400, 407 anatomy 225, 225 patient care 211
shielding 388 data acquisition 225 verification of dose 211
tolerance dose 49 dose-fractionation 227 gliomatosis cerebri 211
fibroblastic proliferation 45 dose solutions 226–7 glottic larynx, tumours
fibrosis 45 indications for anatomy 166, 167
‘field in field’ arrangement, radiotherapy 224 squamous cell carcinoma
glioma 209 metastatic 227 165, 166, 168
FIGO (Federation Internationale target volume definition dose solutions 171,
de Gynecologie et 226, 226 172, 174
d’Obstetrique) 2, 3, non-germinomatous 211, TNM staging 166, 167
372, 395 227 see also laryngeal cancer
‘five R’s’, radiobiology 39, testicular see testicular GORTEC very accelerated
39–40 cancer regimen 92, 92
floor of mouth, tumours 123, germinomas 224, 227 grading of tumours 3–4
124, 125 giant cell tumours of bone Graves’ disease 197, 199,
adjuvant radiotherapy 130 409, 414 200, 203
brachytherapy 132 gingko biloba 6 GTV (gross tumour volume)
fluoroscopy, lung cancer 249 glial tumours 211 5, 9, 9
5-fluorouracil (5-FU) 294, see also glioma defining/contouring 16–17
324, 371 glioblastoma, radioresistance GTV-M 9
focus skin distance (FSD) 21, 38 GTV-N 9, 11
216, 429 glioblastoma multiforme 206 GVT-T 9
forced expiratory volume in one glioma 205–11 GTV-CTV margin 10, 11
second (FEV1) 241 assessment of primary guidelines 6–7
fractionation of doses see disease 207 gums, shielding 74
dose-fractionation clinical/radiological anatomy gynaecomastia 336
‘functional reserve’ 35 206
fungal infections 108, 132 complex radiotherapy 209 HADS (Hospital Anxiety and
conformal therapy 209, 210 Depression Scale) 4
ganglioneuroma 420 data acquisition 207 haematuria 357
gaps, in treatment, avoidance dose comparisons 206 haemoglobin, low 107, 382
40, 106, 382 dose-fractionation 210–11 haemopoietic tissues,
gastric cancer see stomach dose solutions 209, 210 dose–response
cancer early vs delayed relationship 33
GEC-ESTRO group 56 radiotherapy 206 haemorrhage 70
gemcitabine 303 high grade (stage III–IV) dose-fractionation for 70
gene expression profiling 267 206, 208, 208, 210 hair loss 227
genetic profiles, prognostic immobilisation 207 permanent 211, 216,
factor 5 low grade (stage I-II) 227
genetic susceptibility, normal 205–6, 207, 208, 210 tolerance dose 49
tissue damage 45 organs at risk (OAR) 209 reversible 431
geniohyoid muscle 100 palliative radiotherapy 209, half beam blocking 20
germ cell tumours 211 half body irradiation (HBI) 426
germinomas 224, 227 prognosis/survival 205–6 bone pain 70
intracranial 224–7 spinal cord 207 data acquisition 427
assessment of disease target volume definition dose-fractionation 430
225 207–9, 208 dose solution 429
445
INDEX
half body irradiation (HBI) opposing beams 103, PTV 102–3, 105
(continued) 104, 106 treatment delivery and gaps
hand, squamous cell tissue equivalent bolus 106
carcinoma 80 105 tumour shrinkage, effects
hard palate, tumours 123, electron therapy 105, 106 107
124, 178 immobilisation 95–6, 96, unknown primary tumour
head and neck cancers 107 sites, nodal involvement
88–110 IMRT 24, 106 135, 137, 140
3D data acquisition 95–6 indications for radiotherapy verification of dose 106–7
altered fractionated 89–91 see also individual cancer
regimens 88, 91 induction chemotherapy 94 sites
beam arrangements initial patient assessment hearing loss, conductive 184
103–5, 104, 105 88 heart
beam junctions 20 inter-fractional motion 107 shielding 270–1, 276, 277,
brachytherapy 95 key trials 109, 120 277, 279
cervical node involvement outcome predictive factors tolerance dose 49
see cervical nodes 107 hemithorax, irradiation 258
concomitant chemotherapy palliative therapy 90–1, 92, hepatitis, radiation 309
92, 93–4 93 high-dose point 41
conformal therapy 88, patient care 107–9 hilar nodes, lung cancer
103–5, 104, 105 posterior node treatment 245–6
CT and MRI scanning 96, 105, 106 histiocytosis, Langerhans’ cell
97, 107 postoperative radiotherapy 409, 413, 425
dose–cure relationship 38 102, 119–20, 120 histological tumour types,
dose-fractionation 91–2, primary, curative or adjuvant prognostic factor 4
91–3, 92 therapy 89 HIV infection 323, 326
2Gy fraction standard recurrent, neck, Hodgkin lymphoma 283
88, 91, 92 brachytherapy 56 assessment 285
accelerated and re-irradiation 95 children 424–5
concomitant boost side effects of radiotherapy dose-fractionation 291–2
91–2, 92 107–9 indications for radiotherapy
altered (hyperfractionation squamous cell carcinoma 283
and accelerated) 92, 79, 106, 119 involved field radiotherapy
93, 94 dose solutions and (IFRT) 283, 287–8,
hyperfractionation 92, 93 fractionation 120, 120 288, 289
hypofractionation 92, 93 postoperative neck palliative radiotherapy 283
very accelerated irradiation 119–20, 120 PET scanning 285, 286
(GORTEC) 92, 92 target volume definition prognostic factors 283
dose solutions 103–6 96–103 second malignancies 44,
anterior neck field 103, cervical lymph node CTV 51
104, 105 99 staging 285
complex/IMRT 106 CTV 97, 98 target volume definition
conventional/conformal GTV 96–7 287–9
103–5, 104, 105 lymph node delineation see also lymphoma
low neck 103, 104, 99–102, 100, 101, 102, hormone replacement therapy
105 102 382
matching techniques OAR and PVR 103 hormone therapy, prostate
103–5 postoperative CTV 97, 99 cancer 336, 336
446
Index
Ho’s staging, nasopharyngeal dose-fractionation 154–5 Improving Outcomes Guidance
cancer 156, 157 dose solutions 152–4 (IOG) 6–7
Ho’s technique, immobilisation 149 IMRT (intensity-modulated
nasopharyngeal cancer IMRT 154, 155 radiotherapy) 19, 21
160 indications for radiotherapy disadvantages 262
‘hot’ schedule 34 146–7 dose planning 23–4
β-human chorionic key trials 155 dose–volume histograms
gonadotrophin (β-HCG) palliative radiotherapy 147, 343, 344, 345
4, 67, 225, 361 155 ‘double trouble’ and 41,
hydrocephalus 222 CTV 152 43
hyoid bone 100, 136 PTVs 152 forward planned/segmental
hyperfractionation 25, 42 sequencing of multimodality 23, 275–6, 276
accelerated 42 therapy 147 inverse planning 23, 276,
head and neck tumours simulator use 150 276, 343, 344, 345
92, 93 spread of tumours 147–8 photon and electron beam
extreme, low dose rate target volume definition combination 262
brachytherapy 54 150–2 plan verification 23
head and neck tumours TNM staging 146 rotational 24
92, 93 treatment delivery and second malignancy risk after
lung tumours 241 patient care 155 51
hypersplenism 284 hypopharynx, anatomy 136, seven-beam coplanar
hypofractionation 25, 42 144–8, 148 technique 154
bladder cancer palliation hypoxic cells, reoxygenation specific tumours
357 39, 41 bladder cancer 356
head and neck tumours hypoxic sensitisers, breast cancer 275–7,
92, 93 radiotherapy with 94 276, 280, 281
melanoma 80 hysterectomy, abdominal cervical cancer 378
hypopharyngeal cancers 384 glioma 209
146–55 head and neck cancers
adjuvant radiotherapy 147, ICRU reference point 26 106
155 IGRT (image-guided hypopharyngeal cancers
CTV 151–2 radiotherapy) 15, 24 154, 155
assessment of primary bladder cancer 356 intracranial germ cell
disease 148–9, 149 cervical cancer 378 tumours 226–7
clinical/radiological anatomy prostate cancer 346 laryngeal cancer 173
147–8, 148 iliac lymph nodes 362, 366 lung cancer 255
conformal therapy 152–3, illustration, scheme for vii medulloblastoma and
153 imaging see individual infratentorial PNET
conventional radiotherapy modalities 214
154 immobilisation of patient mesothelioma 262
CT 149, 149 13–15 nasopharyngeal cancer
curative for T1T2 N0 for thoracic radiotherapy 163
tumours 146, 154 248–9, 249 oesophageal cancer
CTV for 150, 151 see also patient positioning; 299
curative for T3–4 N+ individual tumours oral cavity tumours 129
tumours 146–7, 154 immobilisation systems orbital tumours 203
CTV 150–1, 151 13–14 oropharyngeal cancers
data acquisition 149–50 impotence 346 142–4, 144, 144
447
INDEX
IMRT (intensity-modulated vulval carcinoma 401 intraoperative
radiotherapy) inguinal tissue, radiosensitivity radiotherapy (IORT),
(continued) 330 cholangiocarcinoma
pancreatic cancer 306 inguinofemoral lymph nodes 307
paranasal sinus tumours anal cancer 326 intraparotid nodes 178, 179
192, 192, 193 vaginal tumours 395, 397, nasal vestibule tumour
parotid sparing 47, 106, 405–6 spread 119, 120
129, 142 vulval carcinoma 401, 402, paranasal sinus tumour
prostate cancer 343, 404 spread 190
344, 345 inhomogeneity, dose see intravenous urethrogram (IVU)
rectal cancer 320 under dose distribution 353–4
salivary gland tumours inner canthus, BCC ‘inverted Y’ technique,
183 penetration 72 lymphoma 287, 291
stomach cancer 299 inner ear, as OAR 182 involved field radiotherapy
thyroid gland 236 intensity-modulated (IFRT), lymphoma
uterine tumours 390–1, radiotherapy (IMRT) see 283, 287–8, 288, 289
391 IMRT children 424
vaginal tumours 397 internal mammary artery involved node radiotherapy
Wilms’ tumour 418 267, 268, 275 (INRT), lymphomas
total body irradiation and internal mammary nodes 287, 288–9
429 267, 268 Hodgkin lymphoma 283,
induction chemotherapy breast cancer involvement 287–8, 289
head and neck cancers 94 266 non-Hodgkin lymphoma
hypopharyngeal cancers 147 depth 275 284
laryngeal cancer 166 dose solutions 278–9 iodine-125 55, 55
inferior pelvic subsite, rectal internal margin 10–12 prostate cancer
cancer 317 International Classification of brachytherapy 347
inflammation, acute, Disease for Oncology iodine-131 307
management 52 version 3 (ICD-O-3) 2 iridium-192 wire implants
inflammatory bowel disease International Classification of 54–5, 55, 57–9
323 Disease (ICD) version breast implant calculation
infratentorial PNET 211–17 10 2 59
assessment of disease and International Commission on HDR afterloading prostate
anatomy 212–13 Radiation Units (ICRU) brachytherapy 348–9
conformal therapy 213 9, 9, 26 implantation technique
data acquisition 213 intestine, chronic sequelae of 57–9, 132, 132–3
dose-fractionation 216 radiotherapy 346 oral cavity tumours 132,
dose solutions 214–16 intracranial pressure, raised 132–3
indications for radiotherapy 207, 216 penile cancer 368, 368
212 intraocular tumours Paris system 57–9, 58
target volume definition brachytherapy 196 volume treatment and
213–14 dose solutions 201 separation of source
treatment delivery and enucleation 196, 201 58
patient care 216 indications for radiotherapy iris 197
inguinal lymph nodes 315, 196 irradiated volume 13
317, 360, 362, 366 target volume definition isocentre technique 20
penile cancer 365, 367, 369 200 breast cancer 20, 274,
vaginal tumours 395 see also orbital tumours 274, 275, 279
448
Index
head and neck cancers data acquisition 168–9 see also shielding
105 dose-fractionation 175 left anterior descending
isodose distributions 13, 18, dose solutions 171–5 coronary artery (LAD)
19 early (T1–2, N0) 165, 168, 276
curves 43 170, 171, 175 legislation, brachytherapy
electron therapy 21 immobilisation 168 60–1
isodoses, computer-derived, IMRT/complex therapy 173 lens, tolerance dose 49, 51
brachytherapy 58–9 indications for radiotherapy lentigo maligna 81
isoeffective schedules 34 165–6 in situ 80, 81
isotopes, brachytherapy 55 key trials 175 lentigo maligna melanoma
nodal CTVs 170, 170 81
Kallman-relative seriality model nodal involvement 165, leptomeningeal tumours 205,
46 167, 170, 170, 171 212
Kaposi’s sarcoma 85 palliative therapy 166 leukaemia
keloids 86 sequencing of multimodality acute, CNS irradiation
kidney therapy 166 221–2
avoidance 362 simulator use 168–9 total body irradiation 426,
tolerance dose 49, 51 target volume definition 429
170–1 Leydig cell tumours 359
labia majora, tumours 402 TNM staging 166, 167 Lhermitte’s syndrome 292
lacrimal fossa 197 treatment delivery and LH-releasing hormone
lacrimal gland 197 patient care 175 agonist/antagonist
carcinoma 196, 203 tumour spread 166–7, (LHRHa) 336
damage 201 169, 170 lidocaine, head and neck
tolerance dose 49, 51 see also glottic larynx, cancers 108
lactate dehydrogenase 67, tumours lifestyle factors, affecting
361 larynx radiotherapy 6
lamina papyracea 198 anatomy 166, 167 linear accelerator, dose
Langerhans’ cell histiocytosis nasendoscopy 168, 168 planning 18
409, 413, 425 late effects, of radiotherapy on linear quadratic (LQ) model
language therapy 108, 116 tissues 33, 44, 45 26, 34–5
laryngeal cancer 165–76 dose limiting effects 34, 40 dose–volume histograms
adjuvant radiotherapy 165, ‘double trouble’ and 41, 43
171, 173, 173, 175 43 equations, cell survival 32,
advanced (T3–4, N) 165, DVHs 47, 47 33
173, 174, 175 low α/β ratio values 33, lip cancer 111–13
assessment of primary 34, 40 assessment of primary
disease 168, 168 measurement and scales for disease 112
beam arrangements 52 brachytherapy 56, 113,
168–9, 171–3, 172, reduction by dose reduction 113
173 44 clinical/radiological anatomy
clinical/radiological anatomy tissue type effect 45 111, 112
166–7 treatment 52 dose-fractionation 113
conformal therapy 171–3 lead masks 73, 73 dose solutions/planning
conventional radiotherapy lip cancers 113 112
171, 175 lead shielding electron therapy 112
CT scanning 169, 169 basal cell carcinoma 73, indications for radiotherapy
curative 170, 175 73–4 111
449
INDEX
lip cancer (continued) tolerance dose 49, 51 non-small cell (NSCLC)
sequencing of multimodality whole lung radiotherapy 241–2, 243–4, 248,
therapy 111 417, 419, 420 253
squamous cell 111, 113 lung cancer 241–57 dose-fractionation
255
target volume definition adjuvant radiotherapy 241, HDR brachytherapy 256
112 244, 251, 255 OAR volume 45
treatment delivery and assessment of primary organ motion 249–50
patient care 113 disease 246–8 CTV-PTV margin 11,
verification of dose 113 cerebral metastases 241, 251–2
lithium fluoride TLDs 221 242 see also respiratory
liver clinical/radiological anatomy motion
partial irradiation in Wilms’ 245–7 palliative radiotherapy 241,
tumour 419 complex therapy 255 242, 244, 252, 255,
tolerance dose 50, 51 conformal therapy 252–5 256
whole liver irradiation 308, contralateral beam prophylactic cranial
309 253 irradiation 242, 256
liver metastases three-field plan 252, sequencing of multimodality
dose-fractionation 309 253, 253 therapy 243–5, 251
target volume definition two-phase technique simulator use 249
308 253, 254 skip metastases 246
treatment delivery and CT and CT-PET 241, 246, small cell (SCLC) 242,
patient care 309 248, 248, 249, 250 245, 253
whole liver irradiation 308, curative radiotherapy dose-fractionation 255
309 241–2, 252, 255 spread and satellite tumour
Wilms’ tumour 417 GTV, CTV and PVT nodules 245
liver tumours, primary 307–9 250–2, 251 target volume definition
assessment and data acquisition 248–50 250–2
classification 308 distal collapse/consolidation TNM staging system 246,
conformal therapy 307 vs 245, 245, 250 247
dose solutions and dose- dose-fractionation 255–9 treatment delivery and
fractionation 308, 309 dose solutions 19, 252–5 patient care 256
indications for radiotherapy dose–volume histograms verification of dose 256
307 47, 47, 48, 253, 254 Lyman-probit model 46
target volume definition 308 electron density 19 lymphadenopathy
treatment delivery and hyperfractionated (CHART) floor of mouth tumours with
patient care 309 protocol 241 125
loperamide hydrochloride immobilisation 248–9, 249 nasopharyngeal cancer
346, 382, 392, 400, indications for radiotherapy 158
414 241–3 oropharyngeal cancers
lumbar cord compression key trials 241, 256 134
62, 65 left lower node 246 see also individual lymph
lung lymph node involvement nodes
anatomy 245, 246 245–6, 246, 248, 248 lymphatic drainage
fibrosis 47, 51 metastases 245–6, 246, anus 325, 325
metastases 410, 417 248, 248 breast 267, 268, 274
normal, OAR volume 45 MLC shielding 252 cervix 371, 372
shielding, in breast cancer neoadjuvant radiotherapy floor of mouth and tongue
treatment 275, 279 241 124
450
Index
oesophagus 295 dose solutions 291 anal cancer 326, 327
penis 366, 366 extranodal 284, 285 bladder cancer 354
prostate gland 337, 337 target volume definition breast cancer 268
rectum 315, 315 289–90, 290 cervical cancer 373, 374
stomach 296, 296 Hodgkin’s see Hodgkin data transfer and contouring
testis 360, 361 lymphoma 16
thyroid gland 232, 233 IFRT 283, 287–8, 289 Ewing’s sarcoma 423
uterus 385, 386 opposing fields 287, gliomas 207, 208
lymph node(s) 288 head and neck cancers
delineation, head and neck indications for radiotherapy 96, 97
cancers 99–102, 100, 283–5 hypopharyngeal cancers
101 INRT/ISRT 283, 284, 287, 149, 150
excision biopsy 285 288–9, 290 intracranial germ cell
T cell cutaneous lymphoma ‘inverted Y’ technique 287, tumours 225, 225,
83 291 226
see also individual nodes key trials 293 lymphoma 285
lymph node metastases low grade follicular 284 medulloblastoma 212
bladder cancer 351–2 MALT 196, 284, 290 meningioma 218, 219
ear tumours 114, 119 ‘Mantle’ technique 287, nasopharyngeal cancer
head and neck tumours 291, 292 160
80, 119, 120 orbital 196, 202 oral cavity tumours 125
hypopharyngeal cancers organs at risk (OAR) 291 oropharyngeal cancers
148 oropharyngeal 135 137
lip cancers 111, 112 paediatric 424–5 pancreatic cancer 304
nasal tumours 117 palliative radiotherapy 283, paranasal sinus tumours
prostate cancer 335 285, 290–1, 292 187, 188
squamous cell carcinoma side effects of radiotherapy prostate cancer 338, 339,
80, 119, 120 292 340
see also individual nodes spinal cord compression rectal cancer 314, 315,
and tumours therapy 66 316
lymphoedema 393, 407 staging systems 285 salivary gland tumours
lymphoma 283–93 systemic irradiation 426, 179, 180
assessment of primary 427 sarcoma 410, 410
disease 285–6 target volume definition spinal cord compression
B cell cutaneous 84 287–91, 289, 290 63, 63
bone 285, 289, 409 T cell cutaneous 82–3 thyroid cancer 232
breast 265 testicular 359 uterine tumours 386, 387,
clinical/radiological anatomy treatment delivery and 387
285 patient care 292 malignant melanoma see
CNS 211, 220–1, 284 WHO classification 285 melanoma, malignant
conjunctival 196, 198, see also non-Hodgkin MALT lymphoma 284, 290
199, 203 lymphoma conjunctival 196
cutaneous see cutaneous lymphovascular space invasion mammography 268
lymphoma (LVSI) 370, 384 Manchester system,
data acquisition 286 gynaecological
diffuse large B cell 84, machine checking 29–30 brachytherapy 56, 57,
220–1, 284 magnetic resonance imaging 379
dose-fractionation 291–2 (MRI) 15 mandible, hotspots in 182
451
INDEX
‘mantle’ technique, lymphoma verification of dose 217 prophylactic radiotherapy
287, 291, 292 megavoltage (MV) imaging 258, 259, 261, 262,
Masaoka clinical staging 14, 15, 358 263
system, thymic tumours megavoltage (MV) radiotherapy sequencing of multimodality
237 21, 23, 26 therapy 259
mastectomy 266, 269 see also electron therapy simulator use 260
matching techniques, head melanoma, malignant 80–1 target volume definition
and neck cancers choroidal 196, 203 260–1
103–5 dose-fractionation 81 treatment delivery and
maxillary sinus tumours 185, recurrence rates 81 patient care 263
187 sinonasal 186 verification of dose 263
target volume definition skin 80–1 methotrexate, high dose
189, 189, 190 vaginal 394 (intrathecal) 220,
see also paranasal sinus meningioma 218–20 284
tumours conformal and complex MIBG, radioactive 421
maximum heart distance therapy 218 micrometastases, TNM staging
(MHD), breast cancer menopausal symptoms 382 system 3
270, 272 MERCURY study, MRI in rectal microscopic tumour spread,
maximum target dose 26 cancer 315 dose–cure relationship
medial pterygoid muscle 126 Merkel cell carcinoma 86 37
mediastinal boost, TBI 430 mesorectum 314 middle ear tumours 114
mediastinal lymphadenopathy mesothelioma 258–64 midplane dose (MPD) 28
67 adjuvant radiotherapy 258, minimum target dose 26
assessment, CT 248 260, 263 MLC shielding 13, 23
lung cancer 245, 246, 246 complex therapy 262 anal cancer 329
lymphomas 285 conformal therapy 261, bladder cancer 355–6
mediastinal lymph nodes 261 breast cancer 276
246, 246, 295 assessment of primary cervical cancer 377, 377
mediastinal tumours, lung disease 259 laryngeal cancer 171,
cancer 241 clinical/radiological anatomy 172
medulloblastoma 211–17 259 lung cancer 252
assessment of disease and conformal therapy 261, mesothelioma 261
anatomy 212–13 261 oral cavity tumours 129
beam arrangements 214, CT scanning 259, 260 paranasal sinus tumours
214, 215 data acquisition 260 191
beam junctions 20 dose-fractionation 263 prostate cancer 343
complex radiotherapy 214 dose solutions 261, 261–3 see also shielding
conformal therapy 213, immobilisation 260 Mohs’ micrographic surgery
214–15 indications for radiotherapy 72
data acquisition 213 258–9 mould brachytherapy 55
dose-fractionation 216 key trials 264 mouth, floor, tumours see floor
dose solutions 214–16 MLC shielding 261 of mouth, tumours
indications for radiotherapy organs at risk (OAR) 260 mouth bite 95, 139, 160
212 tolerance doses 261, paranasal sinus tumours
target volume definition 262 188
213–14 pain, sites of 258, 259 MRC TE10, TE18 and TE19
treatment delivery and palliative radiotherapy trials 360, 364
patient care 216 258–9, 261, 262–3 mucosa, tolerance dose 51
452
Index
mucosal associated lymphoid curative radiotherapy 156 irradiation, postoperative
tissue (MALT) data acquisition 160 119–20
lymphomas see MALT dose-fractionation 163 tumours see head and neck
lymphoma dose solutions 162–3 cancers
mucosal reactions, head and immobilisation 160 neoadjuvant, concurrent and
neck cancers 108 IMRT/complex therapy adjuvant androgen
mucositis 163 deprivation (NCAD)
head and neck cancers indications for radiotherapy 336, 336
94, 107, 108 156–8 nerve roots, irritation 63
oesophageal cancer 301 key trials 163 neuroblastoma 420
oral cavity tumours 132 late effects of radiotherapy children 420–1
pancreatic cancer 307 158, 164 olfactory 186, 190
salivary gland tumours 184 metastases 159 neuroblasts, radiosensitivity
thyroid cancers 236 nodal involvement 158, 420
multi-leaf collimation (MLC) 162 neutron therapy, salivary gland
in IMRT 23–4 organs at risk (OAR) 161 tumours 183
see also MLC shielding recurrent disease nimorazole 94
muscle fibrosis 414 radiotherapy 156, 158 non-Hodgkin lymphoma 284
muscle growth, asymmetrical sequencing of multimodality dose-fractionation 292
416 therapy 158 half body irradiation 426
mutism syndrome 216 simulator use 160 indications for radiotherapy
mycosis fungoides (MF) 82–3 staging systems 156, 157 284–5
myelofibrosis 284 survival rates 156 involved lymph node
mylohyoid muscle 100 target volume definition radiotherapy 284,
160–1 288, 289, 290
nasal cavity, tumours 117, treatment delivery and target volume definition
190 patient care 163 287–91
TNM staging 185 tumour spread 158, 159 see also lymphoma
nasal vestibule, cancer 117 WHO classification (types) non-small cell lung cancer see
see also nose/nasal tumours 156 lung cancer, non-small
nasendoscopy nasopharynx cell (NSCLC)
larynx 168, 168, 170 anatomy 135–6, 158–9, normal tissue
paranasal sinus tumours 159 children, development and
188 shielding in acute leukaemia 415
nasopharyngeal cancer (NPC) 221 damage 44
156–64 sinus tumours invading dose–response relationship
aetiological factors 156 190 33
assessment of primary natal cleft skin reactions 382, radiotherapy effects 44, 45
disease 160 392, 399 volume effects on response
brachytherapy 56, 156 National Institute for Health 35–6
chemosensitivity 158 and Clinical Excellence normal tissue complication
chemotherapy 156, 158 (NICE) 7 probability (NTCP)
clinical/radiological anatomy NCI-CTC 3 (Common Toxicity 35–6, 44
158–9, 159 Criteria version 3) 6 nose/nasal tumours 117–19
conformal therapy 162–3 neck assessment of primary
conventional radiotherapy contour, conformal therapy disease 117
163 and 152 clinical/radiological anatomy
CT scanning 160 dissection 90, 123, 135 117
453
INDEX
nose/nasal tumours treatment delivery and data acquisition 126
(continued) patient care 301 dose-fractionation 131
data acquisition 117–18 two-phase technique 299 dose solutions 128–30
dose-fractionation 118 verification of dose 301 immobilisation 126
dose solutions 118, 118 oesophago-gastric junction IMRT 129
electron therapy 118 (OGJ) cancer 294 indications for radiotherapy
immobilisation 117 oesophagus 122–4
indications for radiotherapy anatomy 294, 295 in situ carcinoma 123
117 strictures, lung cancer N0 neck radiotherapy 123,
sequencing of multimodality therapy 254–5 128
therapy 117 tolerance dose 50 N+ neck radiotherapy
target volume definition 118 olfactory neuroblastoma 186, 123, 128, 129
treatment delivery and 190 nodal levels,
patient care 119 oligodendroglioma 211 recommendations for
verification of dose 119 optic chiasm 197 127
nostril, vaseline use for 79 dose limit for 191, 201 palliative radiotherapy (T3,
pituitary adenoma T4 tumours) 124,
obturator nodes 315, 317 relationship 222, 223 126, 128, 131
oesophageal cancer 294–302 tolerance dose 49, 51 sequencing of multimodality
adjuvant radiotherapy 294 optic nerve 197 therapy 124
assessment of primary blindness prevention 46 shielding (MLC) 129
disease 296–7 dose limit for 191, 201 simulator use 126
brachytherapy 301–2 paranasal sinus tumour target volume definition
cervical 294, 295, 299 conformal therapy 191 126–8, 127
clinical/radiological anatomy tolerance dose 49, 50 TNM staging 125
295–6 oral cavity tumours 122, treatment delivery and
conformal therapy and beam 122–33 patient care 132
arrangements 299, 300 adjuvant radiotherapy for oral hygiene 108–9, 193
CT scanning 296, 297 local disease 123, orbit, anatomy 197, 197–8
data acquisition 297 128, 131 orbital tumours 196–204
dose-fractionation 301 assessment of primary anterior, shielding in acute
dose solutions 299, 300 disease 125 leukaemia 221
dose–volume histograms beam arrangements 129 assessment of primary
300 brachytherapy 122, 123, disease 198
immobilisation 297 124, 131 beam arrangements 198,
indications for radiotherapy indications/disadvantages 199, 201, 201, 202
294–5 132, 133 benign disease affecting
key trials 302 iridium-192 hairpins 197, 199, 200, 203
lymphatic spread 295, 297 132, 132–3 conformal planning 201
multimodality therapy 294 clinical/radiological anatomy CT scanning 199–200
palliative radiotherapy 295, 122, 124–5, 125 data acquisition 198–200,
297, 301 conformal therapy 128–9 199
simulator use 297 conventional techniques for dose-fractionation 201,
skip metastases 295 129 203
target volume definition CT scanning 125, 126 dose solutions 201, 201–3
297–8, 298 curative, for local disease extraocular tumours
thoracic 294, 295, 299, (T1/early T2) 122–3, involving 196–7, 200,
300 126–7, 131 201–3
454
Index
immobilisation 198 tolerance doses see simulator use 138
IMRT 203 tolerance doses sites 136
indications for radiotherapy see also under individual squamous cell carcinoma
196–7 tumours 134, 137
intraocular tumours involving oropharyngeal cancer 134–45 target volume definition
196, 200, 201 adjuvant radiotherapy 144 138, 139–41
key trials 204 CTV 140 PTV 141, 142, 143, 144
lymphoma 196, 202 assessment of primary treatment delivery and
organs at risk (OAR) 200 disease 137–8 patient care 145
palliative radiotherapy 198, clinical/radiological anatomy unknown primary tumour
199 135–7 sites, nodal involvement
paranasal sinus tumours conformal therapy 141, 135, 137, 140
extending 187 141–2 CTV for 140, 140–1
posterior, in irradiation for CT scanning 137, 138 oropharynx, anatomy 135–6,
acute leukaemia 221 curative radiotherapy 135, 136
postoperative radiotherapy 139–40, 144 orthopantomogram (OPG)
196, 201, 201, 203 data acquisition 138 107, 125, 160
proton and plaque therapy dose-fractionation 144–5 osteosarcoma 408
204 dose solutions 141–4 outcomes
pseudotumour 197 dose–volume histograms adverse, measurement and
sarcoma 196, 203 (DVHs) 143 scales for 52
simulator use 198–9, 199 immobilisation 138 recording systems 6
target volume definition IMRT 142–4, 144 ovarian transposition 394
200 indications for radiotherapy ovary, tolerance dose 50,
PTV 201–3, 202 134–5 51
tolerance dose of tissues key trials 145 ovoids, vaginal 380, 389
49, 50 lateral wall tumours 136,
treatment delivery and 139 paediatric tumours 415–25
patient care 204 lymphoma 135 assessment of primary
verification of dose 204 multimodality therapy 134 disease 415
orchidectomy 359 N0 tumours 134, 139 CT scanning 416
organ motion 10–12 N1 tumours 134–5 data acquisition 415–16
IGRT to reduce 24 N2/3 tumours 135 dose and late effects 44
larger PTVs and PRVs 45 neck radiation 134–5 dose-fractionation 416
PTV and 10, 24, 45 nodal involvement 135, dose solutions 416
reduction, CT protocols 16 137, 139 Ewing’s sarcoma 423–4
respiratory/lung see CTVs 139, 139–40 glioma 206, 209
respiratory motion organs at risk (OAR) 142, immobilisation 415–16
vocal cords and 170 144 indications for radiotherapy
organs at risk (OAR) 13, palliative radiotherapy 135, 415
44–53 145 Langerhans’ cell
children 416 posterior wall tumours histiocytosis 425
conformal therapy 22 136, 139 lymphomas 424–5
CT outlining 15–16 primary tumour radiation medulloblastoma and
definition 45 134 infratentorial PNET
doses limited by, in Wilms’ prognostic features 135 212, 214, 216
tumour 419 prophylactic radiotherapy neuroblastoma 420–1
systemic irradiation 428 134 organs at risk (OAR) 416
455
INDEX
paediatric tumours non-Hodgkin lymphoma key trials 309
(continued) 285 multimodality therapy 303,
rhabdomyosarcoma 422–3 oesophageal cancer 295, 304
sequelae of radiotherapy 297, 301 neoadjuvant radiotherapy
415, 416 oral cavity tumours 124, 303
sequencing of multimodality 126, 128, 131 organs at risk (OAR) 306
therapy 415 oropharyngeal cancer 135, palliative radiotherapy 303,
total body irradiation (TBI) 145 306
429 pancreatic cancer 303, risk and prognostic factors
treatment delivery and 306 304
patient care 416 paranasal sinus tumours survival rates 303
verification of dose 417 186, 192 target volume definition
Wilms’ tumour see Wilms’ penile cancer 366, 367, 305–6
tumour 369 treatment delivery and
pain rectal cancers 312, 320–1 patient care 306–7
abdominal 321, 382, 392 sarcomas 409 verification of dose 307
bone see bone pain spinal cord compression Papillon technique 311
head and neck cancers 108 48 para-aortic nodal radiotherapy
pelvic, cervical cancer 371 squamous cell carcinoma 359
palate 80 cervical cancer 371, 377,
hard, tumours 123, 124, superior vena caval 377, 381–2
178 obstruction 67–8 patient care after and side
soft, tumours 136 testicular cancer 360, 363 effects 382
palladium-103 55, 347 thyroid gland 234, 235, testicular cancer 360, 362,
palliative radiotherapy 62–70 236 362–3
anal cancer 324, 329–30 uterine tumours 389 uterine tumours 385, 392
bladder cancer 352, 357 vaginal tumours 394, 399 para-aortic nodes, anatomy
bone pain 68–70 vulval carcinoma 406 360, 361, 385, 386
breast cancer 267, 280 see also emergency parallel opposing beams, dose
cervical cancer 371, 377, radiotherapy planning 19
382 Pancoast tumour 244 paralysis 45, 46, 48
conventional simulators pancreas, anatomy 304, 304 paranasal sinuses, anatomy
17–18 pancreatic cancer 303–7 187, 187
glioma 209, 211 assessment of primary paranasal sinus tumours
haemorrhage 70 disease 304–5 185–95
half body irradiation 426 clinical/radiological anatomy adenoid cystic cancer 186
head and neck tumours 304, 304 adjuvant radiotherapy 97,
90–1 conformal therapy 306, 186, 192
Hodgkin lymphoma 283 306 assessment of primary
hypopharyngeal cancers CT scanning 304, 305, disease 188
147, 152, 155 305 beam arrangements 191,
laryngeal cancer 166 data acquisition 305 191–2
lung cancer 241, 242, dose-fractionation 306 clinical/radiological anatomy
244, 252, 255, 256 dose solutions 306 187, 187
lymphoma 290–1, 292 immobilisation 305 conformal therapy 191,
malignant melanoma 81 IMRT 306 191–2
mesothelioma 258–9, 261, indications for radiotherapy conventional radiotherapy
262–3 303 192
456
Index
data acquisition 188–9 see also salivary gland assessment of primary
dose-fractionation 192 tumours disease 366
dose solutions 191, 191–2 parotid nodes, ear tumour clinical/radiological anatomy
dose–volume histograms spread 114 366, 366
194 Paterson–Parker tables 56, 57 curative radiotherapy 365
immobilisation 188 pathological classification 2 data acquisition 366
IMRT 192, 192, 193 pathological fractures 414 dose-fractionation 368–9
indications for radiotherapy patient care, during treatment dose solutions 367, 367–8
185–6 28 electron beam radiotherapy
lymphatic spread and see also individual tumours 365
metastases 185, 187, patient contouring 16–17 indications for radiotherapy
190 patient documentation 30 365–6
melanomas 186 patient positioning 13–15, 14 interstitial brachytherapy
organs at risk (OAR) 189, organ motion prevention 367–8, 368
190–1 24 palliative radiotherapy 366,
palliative radiotherapy 186, recording/documenting 14 367, 369
192 see also immobilisation of stage I tumours 367
primary radiotherapy 186 patient staging 365, 365
sequencing of multimodality patients’ wellbeing 4 target volume definition
therapy 186–7 pelvic exenteration 394 367–8
simulator use 189 pelvic lymphadenectomy, treatment delivery and
squamous cell carcinoma bilateral 394 patient care 369
186 pelvic lymph node(s) verification of dose 369
target volume definition bilateral dissection (BPLND) wax block technique 367,
189–91 384, 393 367, 369
treatment delivery and in cervical cancer 373 penile preserving surgery 365
patient care 193–4 in vulval carcinoma 401, penis, anatomy/lymphatic
verification of dose 194 404 drainage 366, 366
parapharyngeal space, pelvic lymph node peptic ulceration 364
tumours 158 radiotherapy performance status 4
Paris system, dosimetry 56, penile cancer 367 perigastric nodes 296, 296
57–9, 413 prostate cancer 335, 343 perineal tissue
parotid gland dose-fractionation 346 radiosensitivity 330
anatomy 178 target volume definition skin reaction 382, 392,
sparing, IMRT 47, 106, 341–2, 342 399
129, 142 pelvic pain, cervical cancer peripheral nerves, tolerance
swelling, in TBI 431 371 doses 48, 49, 50
tolerance dose 49 pelvic radiotherapy Perspex screen 82
parotid gland tumours anal cancer 323 Perspex shell 13–14
177–84 cervical cancer palliation construction method and
conformal therapy 182–3, 382 materials 95
183 prostate cancer 335, gliomas 207
sites 178 341–2 head and neck cancers
spread and nodal vulval carcinoma 401 95, 96, 126
involvement 178, Wilms’ tumour 419 laryngeal cancer 168
179 penile cancer 365–9 medulloblastoma and
target volume definition adjuvant radiotherapy infratentorial PNET
180, 180–2, 181 365–6 213
457
INDEX
Perspex shell (continued) pneumonitis, radiation 253, primary cutaneous follicle centre
nasopharyngeal cancer 160 429, 431 lymphoma (PCFCL) 84
paranasal sinus tumours Poisson statistics 36 primary cutaneous marginal
188 PORTEC-2 trial 384, 393 zone lymphoma
Perspex shield, total body positioning of patient see (PCMZL) 84
irradiation (TBI) 428 patient positioning primitive neuroepithelial tumour
phantom measurements 29 positron emission tomography (PNET) see infratentorial
pharyngeal wall, posterior (PET) 15 PNET
147 cervical cancer 373 principles of planning 9–31
tumours 147, 148, 153 cervical node involvement in data acquisition 15–18
photons, in mixed beams 23, head/neck cancer 90 dose-fractionation 25–6
262 Hodgkin lymphoma 285, dose solutions 18–24
picture archiving and 286 dose specification 26–8
conservation systems hypopharyngeal cancers immobilisation 13–15
(PACS) 6 150 patient care during
pineal gland tumours 225 lymphomas 285 treatment 28
pituitary adenoma 222–4 PET-CT quality assurance 29–30
conformal and complex lung tumours 241, 248, target volumes 9–13
therapies 223–4 248, 249, 250 verification of dose 28
data acquisition and dose lymphomas 285, 286 see also specific topics
solutions 223–4 oropharyngeal cancers proctitis 346, 357
dose-fractionation 224 137, 137 prognostic factors 4–5
pituitary function tests 194 pancreatic cancer 305 prostate cancer 332–50
pituitary gland, tolerance dose uterine tumours 386 active surveillance, benefits
49 postcricoid mucosa, tumours and risks 334
planning 147, 148 adenocarcinoma 332
3D see three-dimensional posterior fossa, radiotherapy adjuvant radiotherapy 335
(3D) planning 214, 215, 216 assessment of primary
checks/checking 30 posterior fossa syndrome 212 disease 338
information required before ‘post-gap acceleration’ 40 brachytherapy 55, 56,
1–8 postoperative neck irradiation 347–9
principles see principles of 119–20 benefits and risks 334
planning postoperative radiotherapy boost, EBRT with 336,
radiobiology and see cell optic nerve region 46 348
killing, radiation-induced; see also specific tumours dose-fractionation 349
radiobiology Pötter technique, dynamic dose calculation
planning organ at risk volume brachytherapy 347 347
(PRV) 13, 45 predictive indicators 5 HDR afterloading 348–9
planning target volume see predictive tools 5 indications 347
PTV (planning target prelaryngeal (Delphian) node LDR permanent seed
volume) 167 347–8, 348
plaque therapy, orbital pre-planning 1–8 Pötter technique 347
tumours 204 primary cutaneous B cell Seattle technique 347
plasmacytoma, spinal cord lymphoma (PCBCL) 84 single stage inverse
compression 66, 66 primary cutaneous diffuse planned procedure
pleomorphic adenoma 178, large B cell lymphoma 347
180, 184 leg type (PCLBCL LT) V100, V150 and V200
pleural effusions 258 84 347–8
458
Index
clinical/radiological anatomy radical treatment, indications IGRT to reduce organ
337, 337 333 motion and 24
conformal therapy 343, Roach formulae 338, 341 organ motion increasing
344 salvage radiotherapy 335 10, 45
conventional therapy 345 seminal vesicle involvement see also individual cancers
CT scanning 338, 339–40, 335, 340–1, 345 PUVA, total skin electron beam
340 sequencing of multimodality therapy with 82
curative radiotherapy therapy 336–7 pyriform fossa
332–5, 345 single phase radiotherapy anatomy 147, 148, 148
results, assessment 341 tumours 146, 147, 148,
333, 335 surgery, benefits and risks 149, 149, 154
data acquisition 338–40 334
dose distribution 27 target volume definition quality assurance 29–30
dose-fractionation 345–6 340–3 dosimetry protocols 30
dose solutions 27, 343–5 pelvic lymph nodes machine checks 29–30
dose–volume histograms 341–2, 342 patient documentation 30
27, 343 prostate and seminal planning checks 30
EBRT, benefits and risks vesicles 340–1 staffing 30
334 prostate bed 342–3 quality control, machines 29
EBRT and hormone therapy treatment delivery and quality of life measures 6
336, 336 patient care 346
EBRT with brachytherapy two phase radiotherapy radiation dose see entries
boost 336, 348 341 beginning dose
EBRT with chemotherapy verification of dose 346 radiation hepatitis 309
337 prostatectomy 334, 342 radiation injury 45
extraprostatic extension prostate gland radiation pneumonitis 253,
338 anatomy/lymphatic drainage 429, 431
grades and staging 332 337, 337, 353, 353 Radiation Therapy Oncology
immobilisation 338–9, 339, outlining, target volume Group (RTOG) 6
346 definition 340 ‘radiobiological dose’ 43
incidence, prognosis and prostate-specific antigen (PSA) radiobiology 32–43
early detection 332, 67, 332, 333 acute/late effects on tissues
333 protocols and guidelines 6–7 33
indications for radiotherapy proton therapy 25 ‘double trouble’ and 41, 43
332–6 meningioma 218 ‘five R’s’ 39, 39–40
key trials 349 orbital tumours 204 future directions for 43
localised PRV 13, 45 linear quadratic (LQ) model
low/intermediate/high risk psychosexual counselling 34–5
types 333, 340 382, 393 target volume effects 35–6
treatment options 333, pterygopalatine fossa, tumours see also cell killing,
333, 334 extending into 188, radiation-induced
organs at risk (OAR) 343 198 radiochondritis 79
palliative radiotherapy PTV (planning target volume) radionecrosis, treatment 52
335–6, 345 9, 10 radionuclide therapy, prostate
dose-fractionation conformal therapy 21 cancer 336
345–6 CTV-PTV margin see radio-opaque clips, target
pelvic lymph node CTV-PTV margin volume definition 99
radiotherapy 335 DVHs for evaluating 47, 47 lung cancer 250
459
INDEX
radio-opaque clips, target curative radiotherapy 311 renal hilar lymph nodes 360,
volume definition data acquisition 316 361
(continued) dose-fractionation 320–1 reoxygenation 39, 41
mesothelioma 260 dose solutions 319–20 repair, DNA damage 39, 45
radio-opaque markers 12, endocavitary local contact repopulation, cell 39, 40
15, 24, 28 radiotherapy 311 resection, residual tumour
cervical cancer 372–3 HDR brachytherapy 311 after, classification 3
prostate cancer 339 key trials 313–14, 321–2 resistant tumours, dose–cure
vaginal tumours 396 local recurrence, risk factors relationship 38, 38
radioresistance, dose–cure 312 respiratory gating 250
relationship 38, 38 location 314 respiratory motion
radiosensitising drugs 303, lower third, tumours 312 breast cancer treatment and
417 management, algorithm 277
radiosensitivity 39, 39–40 312, 313 lung cancer
abnormal, conditions with palliative radiotherapy 312, CTV-PTV margin 11,
45 320–1 251–2
dose–cure relationship postoperative radiotherapy data acquisition and
37–8, 38 311, 318, 319, 320 249–50
radiotherapy, influence of other phase 2 volume 318 reduction 24, 250
treatments on 5–6 preoperative radiotherapy Response Evaluation Criteria in
radium-228 55 311, 316–18, 320 Solid Tumours (RECIST)
random errors 12, 28 re-irradiation 312 6
randomised controlled trials sequencing of multimodality retina, tolerance dose 49, 51
(RCT) 7 therapy 312, 314 retinoblastoma 196
IMRT 24 target volume definition retinopathy 201
rectal cancers 313–14 316–18 retro-crural lymph nodes 360
see also trials TNM staging 316 retromolar trigone, cancers
ratio α/β 32, 33, 34, 40 total mesorectal excision 123, 124
recording/documenting (TME) 311, 312 retroperitoneal tumours
checking/quality assurance treatment delivery and (sarcomas) 409, 412,
30 patient care 321 414
patient position 14 verification of dose 321 retropharyngeal nodes 99,
treatment outcomes 6 recto-vaginal fistula 371, 400 158
rectal bleeding 382 rectum rhabdomyosarcoma 422–3
rectal cancers 311–22 anatomy/lymphatic drainage rhenium 55
adjuvant radiotherapy 314, 314–15, 315 rhinotomy, lateral 188
311–12 OAR in prostate cancer Roach formulae, prostate
assessment of primary 343 cancer 338, 341
disease 315–16 size assessment in prostate RTOG 0529 trial 324, 326
beam arrangements 319, cancer 339 RTOG 9003 schedule,
319 tolerance dose 50 hyperfractionation, head
bony landmarks and field recursive partitioning analysis and neck tumours 92,
borders 318 (RPA) classification, 93
boost radiotherapy 318 melanoma 81 ‘rules of thumb’ 34, 36
clinical/radiological anatomy redistribution, of cells 39 ruthenium-106/106 55
314, 314–15, 315 regression of tumours 32
conformal therapy 319, Reid’s baseline 430 saliva, head and neck cancers
319, 320 remission, clinical 32 108–9
460
Index
salivary gland, sparing, IMRT clinical/radiological anatomy standard vs unfamiliar, LQ
24 410 model 35
salivary gland tumours 177–84 conformal therapy 412, total time between 35
adenoid cystic carcinoma 412 see also dose-fractionation
178, 180, 181 conventional therapy 413 schistosomiasis 351
adjuvant radiotherapy 177, CT planning 411 Schmidt’s sleeve 380
181, 183, 183 data acquisition 410–11 scrotum
assessment of primary dose-fractionation 413–14 irradiation 362, 364
disease 179 dose solutions 412, lymphatics 360
clinical/radiological anatomy 412–13 Seattle technique,
178 Ewing’s sarcoma 423–4 brachytherapy 347
conformal therapy 182–3, immobilisation 410 second malignancies after
183 IMRT/complex therapy radiotherapy 44, 51
conventional radiotherapy 412 after paediatric tumours
183 indications for radiotherapy 416
CT scanning 179, 180 408–9 lymphomas 292
data acquisition 179–80 inoperable tumours 409, squamous cell carcinoma
dose-fractionation 183–4 413 79, 80
dose solutions 182–3 intraoperative electron total body irradiation 431
high grade tumours 177, therapy 409 SECRAB trial 267, 281
180 key trials 411, 414 sellar region tumours
immobilisation 179 organs at risk and PRVs 222–4
IMRT 183 411 seminal vesicles, in prostate
indications for radiotherapy palliative radiotherapy 409, cancer 335, 345
177–8 413 target volume definition
key trials 184 postoperative radiotherapy 340–1
local/metastatic spread 409, 413 seminoma
177, 178, 183 preoperative therapy 409, ‘dog leg’ radiotherapy
low grade tumours 177 413 359, 362, 363, 364
neutron therapy 183 radiation-induced 409 dose-fractionation 363
organs at risk (OAR) 182 recurrent (local) 409 dose solutions 363
pleomorphic adenoma sequencing of multimodality indication for radiotherapy
178, 180, 184 therapy 409 359–60
sequencing of multimodality staging 408, 408 key trials 360
therapy 178 surgery 408–9 palliative radiotherapy 360,
simulator use 180 target volume definition 363
target volume definition 411, 411 radiosensitivity 38
180, 180–2, 181 treatment delivery and risk factors for recurrence
treatment delivery and patient care 414 359
patient care 184 verification of dose 414 stage I 359–60
verification of dose 184 scalp, cutaneous stage IIA/IIB 360
samarium 336 angiosarcoma 85 treatment delivery and
sarcomas, soft tissue and scars, keloids, radiotherapy for patient care 364
bone 408–14 86 virtual simulation, para-aortic
assessment of primary schedule, treatment 34, 40–1 nodal irradiation 362,
disease 410 cervical cancer 382 362
brachytherapy boost 409, ‘double trouble’ and 41, 43 sensory neuropathy, acute
413 ‘hot’ and ‘cold’ 34 321
461
INDEX
sentinel node biopsy, breast sinonasal tumours see small cell lung cancer see lung
cancer 266 paranasal sinus cancer, small cell
Sertoli cell tumours 359 tumours (SCLC)
serum markers 3 sinuses, tumours see smoking 88, 107, 242
set-up margin 12 paranasal sinus soft palate, tumours 136
set-up variations 12 tumours soft tissue sarcomas see
SF-36 4 skin sarcomas, soft tissue
shielding 20 conformal therapy, dose and bone
basal cell carcinoma 73, optimisation 23 solitary plasmacytoma 66
73–4, 74 tattoos see tattoo, skin somnolence 216, 222, 431
cardiac, breast cancer skin care source skin distance (SSD)
treatment 270–1, 276, breast cancer radiotherapy 76, 77
277, 277, 279 281 speech and language therapy
cervical cancer 374 mesothelioma radiotherapy 108, 116
eyes 74, 74 263 spermatogenesis, tolerance
IFRT in lymphomas 288 skin dermatomes 64 dose 50
medulloblastoma and skin reactions sperm storage 364
infratentorial PNET lymphomas 292 sphenoid sinus, tumours 190
214 management in head and spinal cord 62
oral cavity tumours 129 neck cancers 109 glioma 207
squamous cell carcinoma prostate cancer 346 late effects of radiotherapy
73 sarcomas 414 45, 46, 48
testes and groins 367, 369 see also erythema head and neck irradiation
testicular cancer skin tumours 71–87 103
radiotherapy 362 BCC see basal cell lung cancer radiotherapy
total body irradiation and carcinoma (BCC) and 252
428, 429 cutaneous angiosarcoma mesothelioma radiotherapy
see also MLC shielding 85 and 260
short-distance therapy see dose specification 26 PRV 45, 48, 299
brachytherapy Kaposi’s sarcoma 85 tolerance dose 46, 48, 49
side effects lymphoma see cutaneous spinal cord compression
acute, recording system lymphoma 62–6
6 melanoma 80–1 3D CT planning 65, 66
monitoring 28 Merkel cell carcinoma 86 clinical features 62–3
scoring systems 28 non-melanoma 71–80 clinical/radiological anatomy
see also acute effects; late SCC see squamous cell 64
effects carcinoma (SCC) data acquisition 64
simulator-CT 18 skip metastases dose-fractionation 66
simulators lung cancer 246 dose solutions 65, 65–6
3D see three-dimensional oesophageal cancer 295 investigations 63, 63
(3D) planning; virtual skull, base, nodal volume multimodality treatment
simulation delineation guidelines sequence 63–4
conventional 17–18 100 outcome determinants and
single-field treatment, dose small bowel prognosis 62
planning 19 acute radiation toxicity 321 palliative therapy, dose 48,
single photon emission displacement, anal cancer 66
computed tomography radiotherapy 326 surgical decompression
(SPECT) 15 tolerance dose 50 63–4
462
Index
target volume definition positive and negative 76, verification of dose 301
64–5 76 straight leg raising, limitations
treatment delivery and Stanford technique, modified 63
patient care 66 82, 83 strontium-90 55, 55, 336
tumours causing 62, 64 stem cells submandibular salivary gland
spinal radiotherapy, dose–response relationship tumours 178, 182
medulloblastoma and 33 subparotid glands 179
infratentorial PNET repopulation 33 superficial radiotherapy
213 survival/survival curves 33, BCC see basal cell
spine 34 carcinoma (BCC)
bone metastases see stereotactic frames 13, 218 keloids 86
vertebrae, metastatic stereotactic radiosurgery, liver superior mediastinal nodes
tumour tumours 309 232, 233
ependymoma 217–18 stereotactic radiotherapy 25 superior sulcus (Pancoast)
glioma 206 cerebral metastases tumours 244
tolerance dose 46, 48, 49 227–8 superior vena cava, stenting
splenic irradiation 284, 290, sterility 431 67, 241
291 sternocleidomastoid muscle, superior vena caval obstruction
dose-fractionation 292 spread into 99, 102 (SVCO) 67–8, 241
squamous cell carcinoma steroids clinical features 67
(SCC) 79–80 breast cancer skin care data acquisition 67
anaplastic 79 281 dose-fractionation 68
bladder 351 head and neck cancers dose solutions 68
contraindications for 108 indications for radiotherapy
radiotherapy 79–80 high dose 67
cutaneous 79–80 spinal cord compression multimodality treatment
dose-fractionation 80 63 sequencing 67
epidermoid, of anus 323 SVCO 67 target volume definition
eyelids 197 thyroid cancer 236 68, 68
glottic larynx 165 St John’s wort 5–6 supraclavicular fossa (SCF),
head and neck cancers see stomach breast cancer
head and neck cancers lymphatic drainage 296, recurrence 266
indications for radiotherapy 296 supraclavicular lymph nodes
79–80 target volume definition 274, 275
lip 111, 113 298–9 CT scanning, in breast
lung 245 stomach cancer 294–302 cancer 273–4
metastases 79, 80 assessment of primary depth 274
oropharyngeal 134, 137 disease 297 supraglottic larynx, tumours
paranasal sinuses 186 clinical/radiological anatomy 167
rectal 311 296, 296 squamous cell carcinoma
shielding 73 data acquisition 297 165
supraglottic larynx 165 dose-fractionation 301 dose solutions 171
vulval 402 dose solutions 299 TNM staging 166
staffing, quality assurance 30 indications for radiotherapy see also laryngeal cancer
staging of tumours 2–4 295 surgery, influence on
stand-off key trials 302 radiotherapy 5
convex chest wall and 278 treatment delivery and surgical decompression, spinal
correction factors 76, 76 patient care 301 cord 63–4
463
INDEX
survival curve 32, 33 cervical cancer 373, 382 indications for radiotherapy
survival probability 36 prostate cancer 339, 346 359–60
surviving fraction (SF) 32, 36 testicular cancer 361, 364 key trials 360, 364
SF2 36, 39–40 uterine tumours 387, 388 lymphatic spread 360, 361
swallowing problems 108 T-bar device 248, 249 lymphoma 359
systematic errors 12 T cells, cutaneous lymphoma palliative radiotherapy 363
systematic target volume (STV) 82–3, 83 stage I disease 359, 361
12 TD5/5 tables 46, 49–50, 320 stage II disease 360
systemic irradiation 426–32 TD50/5 tables 46, 49–50 staging 359
assessment of disease for tears, drainage 197–8 target volume definition
427 temozolomide 206, 211 361–3
data acquisition 427 temporal bone, resection treatment delivery and
dose-fractionation 429–30 114, 116 patient care 364
dose solutions 428–9 temporomandibular joint (TMJ), verification of dose 364
half body see half body excessive doses virtual simulation 362, 362,
irradiation (HBI) 182–3 363
indications 426 tenesmus 346 testicular intraepithelial
target volume definition teratoma, malignant 360 neoplasia (TIN) 360,
428 testes 363
total body see total body hormone production, TG43 (Task group 43) 56
irradiation (TBI) tolerance dose 50 therapeutic ratio 44–5
treatment delivery and lymphatic drainage 360, thermo-luminescent dosimetry
patient care 428, 429, 361 (TLD) 28
430–1 prophylactic irradiation 284 breast cancer therapy 281
verification of dose 431 shielding 362, 364, 367, CNS irradiation for acute
369 leukaemia 221
tampon, radio-opaque 316, tolerance dose 50, 51 medulloblastoma and
387 testicular boost, total body infratentorial PNET 217
target volumes 9–13 irradiation 430 prostate cancer 346
CT protocols for delineating testicular cancer 359–64 thermoplastic shell 14, 221
16 assessment of primary thoracic radiotherapy
CTV see CTV (clinical target disease 361 curative and palliative 242
volume) clinical/radiological anatomy immobilisation for 248–9,
defining 9–13 360, 361 249
effects on tissue response conventional simulation three-dimensional (3D)
35–6 362–3 conformal therapy see
GTV see GTV (gross tumour data acquisition 361 conformal treatment
volume) ‘dog leg’ radiotherapy three-dimensional (3D)
PTV see PTV (planning 362, 363, 364 planning
target volume) dose-fractionation 363 bladder cancer 355,
see also individual tumours dose solutions 363 355–6
taste problems, head and neck EGCCCG guidelines 359, DVHs 46–8, 47
cancers 108–9 360 ear tumours 115
tattoo, skin 15 germ cell tumours 359 head and neck cancers
anal cancer 331 non-seminomatous 359, 95–6
bladder cancer 354, 357 360 nasal tumours 118
breast cancer 269, 270, seminomas see nasopharyngeal cancer
272 seminoma 160
464
Index
oropharyngeal cancer 138 CT scanning 232, 233, clinical (cTNM) 2–3
rectal cancer 318, 319, 237 GTV classification by 9
320 data acquisition 233 head and neck cancer,
spinal cord compression dose-fractionation 236 nodal staging 89, 90
treatment 65, 65 dose solutions 235–6 hypopharyngeal cancers
SVCO therapy 67 Hürthle cell 231 146
uterine tumours 387, 390 immobilisation 233 laryngeal cancer 166, 167
see also computed IMRT 236 lung tumours 246, 247
tomography (CT); indications for radiotherapy nasal cavity tumours 185
conformal treatment 231 nasopharyngeal cancer
thymic tumours 237–40 medullary cancer 231, 156, 157
adjuvant radiotherapy 237, 232, 236 oral cavity tumours 125
239, 239 palliative radiotherapy 234, paranasal sinus tumours
clinical/radiological anatomy 235, 236 185
238 sequencing of multimodality pathological (pTNM) 2–3
data acquisition and therapy 232 penile cancer 365, 365
assessment 238 simulator use 233 rectal cancers 316
dose-fractionation 239 target volume definition 234 T0, T1–4 classification 3
dose solutions 238 treatment delivery and tolerance, factors affecting
indications for radiotherapy patient care 236 36
237 verification of dose 236–7 tolerance doses 36, 46–8
Masaoka clinical staging well-differentiated specific organs 48–51,
system 237 (papillary/follicular) 49–50
preoperative radiotherapy 231, 232 mesothelioma therapy
237, 239 dose-fractionation 236 261, 262
sequencing of multimodality dose solutions 235 tomotherapy 24
therapy 237 GTV, CTV and PTV 234, helical, liver metastases
target volume definition 235 308
238 multimodality therapy medulloblastoma and
treatment delivery and 232 infratentorial PNET
patient care 240 recurrence and spread 214
thymus, anatomy 238 231, 232, 234 see also IMRT (intensity-
thyroid cancer 231–7 thyroid cartilage 100 modulated
adjuvant radiotherapy 231, thyroidectomy 232 radiotherapy)
235, 235 thyroid eye disease (Graves’ tongue tumours 124, 125
anaplastic cancer 231, disease) 197, 199, anterior tumours 124, 132
232, 235 200, 203 base tumours 136
dose-fractionation 236 thyroid gland, lymphatic brachytherapy 56
dose solutions 235, drainage 232, 233 IMRT 142, 142
235, 236 tissue target volume definition
GTV, CTV 234 normal see normal tissue 139
multimodality therapy regeneration 35 brachytherapy 56, 132
232 see also cell(s) hairpin technique 54–5
assessment of primary tissue equivalent bolus, head lateral tumours 124
disease 232 and neck cancers 105 tonsillar cancer 98, 136
clinical/radiological anatomy TNM (tumours, nodes, conformal therapy 141,
232 metastases) staging 141–2
conformal therapy 235 system 2–4 MRI 138
465
INDEX
total body irradiation (TBI) schedule see schedule, tumour markers 3
children 429 treatment intracranial germ cell
complex therapy 429 treatment plan 41, 43 tumours 225
data acquisition 427 treatment planning system prognostic factor 4
dose distribution (TPS) 19, 43, 270 superior vena caval
homogeneity 429 trials obstruction 67
dose-fractionation 429–30 anal cancer 324, 326, 331 two-dimensional therapy
dosemeters 428, 429, bladder cancer 358 planning
430–1 breast cancer 266, 281 dose calculations 18, 19,
dose solutions 428–9 cervical cancer 383 45, 46
indications 426, 427 CNS/brain tumours 229 tolerance doses 46
patient positioning 428, head and neck squamous
428 cell carcinoma 109, UICC (Union Internationale
total reference air kerma 120 Contre le Cancer) 2
(TRAK) 60 hypopharyngeal cancers ultra-small particles of iron
total skin electron beam 155 oxide (USPIO) 373,
therapy (TSEBT) laryngeal cancer 175 375, 386, 402
82–3, 83 lung cancer 241, 256 ultrasound
adverse effects 82 lymphoma 293 breast cancer 268
dose-fractionation 83 mesothelioma 264 endoscopic see endoscopic
T cell cutaneous lymphoma nasopharyngeal cancer 163 ultrasound (EUS)
82, 83 oesophageal/stomach transabdominal, pancreatic
tracheitis 175 cancer 302 cancer 304–5
tracheostomy 175 orbital tumours 204 transrectal 347
training/teaching, clinical oropharyngeal cancer 145 urethra, iridium wire avoidance
anatomy 6 pancreatic cancer 309 368
transabdominal ultrasound, prostate cancer 349 urinary diversion 371
pancreatic cancer rectal cancers 313–14, urinary tract obstruction 346,
304–5 321–2 347
transitional cell carcinoma 351 salivary gland tumours 184 uterine tumours 384–93
transnasal oesophagoscopy sarcomas, soft tissue and 3D dose planning 387, 390
148, 149 bone 411, 414 adjuvant radiotherapy 388,
transrectal ultrasound (TRUS) seminoma 360 391
347 TBI and bone marrow assessment of primary
transurethral resection of transplantation 431 diseases 386, 387
bladder tumour testicular cancer 360, 364 beam arrangements 387
(TURBT) 352 uterine tumours 393 ‘brick’ 390, 390
treated volume (TrV) 13 trismus, reduction 193 brachytherapy 56, 389–90
treatment tumour applicators 389, 389–90
availability, as prognostic shrinkage, radiotherapy boost therapy 391
factor 4–5 causing 5 dose-fractionation 391
avoidance of gaps 40, stasis 32 vaginal vault 384,
106, 382 tumour cells see cell(s) 389–90, 391
influence of other modalities tumour control probability clinical/radiological anatomy
on radiotherapy 5–6 (TCP) 44 385, 385–6
number, tumour cell number tumour cure, dose relationship conformal therapy 390
relationship 36–7 see dose–cure conventional high energy
objective 36 relationship photon therapy 390
466
Index
conventional planning/ fibrosis 400 vaginectomy 394
simulator use 387–8 rehydration 382, 400 valleculae (glossoepiglottic
CT scanning 386, 387, vaginal intraepithelial neoplasia fossae) 136, 136
388 (VAIN) 394, 397 Van Herk’s formula, CTV-PTV
data acquisition 387–8 vaginal trachelectomy 370 margin 12
dose-fractionation 391–2 vaginal tumours 394–400 vascular-directed therapy, for
dose solutions 390–1 assessment of primary late effects 52
immobilisation 387 disease 395 veno-occlusive disease 309
IMRT/complex therapy brachytherapy 56, 398 verification (dose) 28
390–1, 391 after EBRT 399 vertebrae, metastatic tumour
indications for radiotherapy interstitial 394, 398 62, 64
384–5 intracavitary 394, 398 beam junctions and 20, 21
infiltration and spread 385 target volume definition see also spinal cord
key trials 393 397 compression
organs at risk (OAR) 389 clear cell adenocarcinoma vesico-vaginal fistula 371, 400
palliative radiotherapy 389, 394, 395 virtual simulation 17
392 clinical/radiological anatomy bladder cancer 356, 356
para-aortic nodal 395, 395 breast cancer 270–1, 271,
radiotherapy 385, 392 conformal therapy 397 275
primary radiotherapy conventional planning/ cervical cancer 376, 377,
388–9, 391–2 simulator use 397 377
rare tumours 385 CT planning/scanning spinal cord compression
risk factors and prognostic 396–7, 398 treatment 65, 65
factors 384 data acquisition 396 testicular cancer 362, 362,
sarcomas 386 dose-fractionation 398–9 363
sequencing of multimodality dose solutions 397–8 vocal cords 168
therapy 385 immobilisation 396 paralysis 148
serous tumours 384, 386 IMRT 397 squamous cell cancer
survival 384 indications for radiotherapy 168, 170
target volume definition 394 see also glottic larynx,
388–90 lower third of vagina 395, tumours
treatment delivery and 397 VORTEX trial 411, 414
patient care 392–3 lymphatic spread 395 vulval fibrosis 407
verification of dose 393 melanoma 394 vulval intraepithelial neoplasia
uterus, anatomy/lymphatic MV beams 398 (VIN) 402
drainage 385, 385–6, organs at risk (OAR) 397 vulval skin reactions 407
386 palliative radiotherapy 394, vulval tumours 401–7
uvea 197 399 adjuvant radiotherapy 401,
metastases 196 target volume definition 405–6
uvula, tumours 136 396–7 assessment of primary
treatment delivery and disease 402
V20 concept 47, 47, 48 patient care 399–400 brachytherapy 56, 405
vacuum moulded bags 14 upper third of vagina 395, boost therapy 398, 405,
vagina 397 406
care, in cervical cancer verification of dose 400 clinical/radiological anatomy
radiotherapy 382 vaginal vault, brachytherapy 402
cervical cancer spread see under conformal therapy (4 beams)
372 brachytherapy 404, 404
467
INDEX
vulval tumours (continued) wax compensator 214 medulloblastoma 214–15,
data acquisition 402–3 wax nostril plugs 117, 118, 215, 216
dose-fractionation 405–6 188 whole ventricular radiotherapy
dose solutions 404–5 weight loss, head and neck (WVRT), intracranial
electron therapy 405, 406 cancers 108 germ cell tumours
immobilisation and position whole abdomen irradiation, 226, 226
402–3 Wilms’ tumour 419 Wilms’ tumour 417–19
indications for radiotherapy whole body dose (WBD), conformal therapy 418,
401–2 neuroblastoma 421 419
locally advanced 405 whole body fixation frames dose solutions and
node negative disease 401 14 dose-fractionation
node positive disease whole brain radiotherapy 216 418–19
401–2 acute leukaemia 221, indications for radiotherapy
palliative radiotherapy 221–2 417–18
401–2, 406 cerebral metastases 227, organs at risk (OAR) 418,
recurrent disease 406 228, 228, 229 419, 419
sequencing of multimodality CNS lymphoma 221 target volume definition
therapy 402 dose-fractionation 229 418, 418
simulator use 403 medulloblastoma and World Health Organization
survival rates 401 infratentorial PNET (WHO)
target volume definition 214–15, 215, 216 brain tumour classification
403–4 metastatic malignant 205, 205
total vulval dose 407 melanoma 81 lymphoma classification
treatment delivery and side effects 227 285
patient care 407 whole liver irradiation, liver nasopharyngeal cancer
urethral involvement metastases 308, classification 156
(stage III) 402, 403 309 thymic tumours classification
verification of dose 407 whole lung radiotherapy, 237
Wilms’ tumour 417,
WAGR syndrome 417 419, 420 xerostomia 129, 142, 160,
wax block, penile cancers whole spine radiotherapy 163
367, 367, 369 216 X-ray beam attenuation 19
468

Practical Radiotherapy Planning, Fourth Edition

Uploaded by

Copyright:

Available Formats

Practical Radiotherapy Planning, Fourth Edition

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

Practical Radiotherapy Planning, Fourth Edition

Uploaded by

Copyright:

Available Formats

Practical

Ann Barrett MD, FRCP, FRCR ■

British Library Cataloguing in Publication Data

Library of Congress Cataloging-in-Publication Data

ISBN 978 034 0927731

Commissioning Editor: Gavin Jamieson

Typeset in 10/12 Galliard by Macmillan Publishing Solutions

2 Principles of radiotherapy planning 9

3 Radiobiology and treatment planning 32

4 Organs at risk and tolerance of normal

6 Emergency and palliative radiotherapy 62

8 Head and neck: general considerations 88

9 Lip, ear, nose and treatment of the

10 Oral cavity 122

11 Oropharynx cancer and unknown

15 Salivary glands 177

16 Sinuses: maxilla, ethmoid and

18 Central nervous system 205

19 Thyroid and thymoma 231

24 Oesophagus and stomach 294

37 Paediatric tumours 415

38 Systemic irradiation 426

Appendix: List of abbreviations 433

treatment. This is essential to allow effective cancer registration and comparisons

undifferentiated tumours; with Gx used when it is not possible to determine

Influence of other treatments on radiotherapy

Systems for recording outcomes of treatment

Protocols and guidelines

Target volume definition

Figure 2.1 ICRU target volume definitions showing

■ Gross tumour volume

■ Clinical target volume

■ Planning target volume

■ Organ motion/internal margin

■ Set-up variations/set-up margin

PTV margin 2.5Σ 0.74 σ

Machine Energy (MV) FSD (cm) 10 ⴛ 10 cm field*

Dmax depth (cm) % DD at 10 cm

Cobalt-60 1.25 100 0.5 58.7

*Maximum field size at 100 cm is 40  40 cm.

Three levels of dose planning and treatment are described here:

Figure 2.5 Technique for matching two

Figure 2.7 9 MeV electron

growth algorithm software and departmental protocols, taking into consideration

Patient care during treatment

Radiation and cell kill

Cell survival curves

SF Exp (αd  βd2) (3.1)

0.2 Large α/β

Acute and late effects on tissues

The linear quadratic model for tissues

D1 (α/β d1) D2 (α/β d2) (3.2)

Radiation dose and tumour cure probability

Table 3.1 Radiobiological factors controlling response to fractionated radiotherapy (the

Radiobiological Mechanism of effect on Clinical relevance

Radiosensitivity Intrinsic radiosensitivity differs Can account for variable

Avoidance of gaps during treatment

Treatment plan, schedules and

Fractionation Rationale Mode of delivery Effect on tumour Effect on Effect on

Relative dose (%)

*Maximum field size at 100 cm is 40 40 cm.

SF Exp (αd βd2) (3.1)

0.2 Large α/β

Oesophagus V50 32

1.5 1.23 1.13

Lesions 3 cm diameter (superficial radiotherapy [SRT] 80–140 kV)