2014 Breast Surgery A Companion To Specialist

Breast Surgery
A COMPANION TO SPECIALIST SURGICAL PRACTICE
Series Editors
O. James Garden
Simon Paterson-Brown
Breast Surgery
FIFTH EDITION
Edited by
J. Michael Dixon
BSc(Hons) MBChB MD FRCS FRCSEd FRCPEd(Hon)
Professor of Surgery, University of Edinburgh;
Clinical Director, Edinburgh Breakthrough Unit;
Consultant Surgeon, NHS Lothian Edinburgh Breast Unit, [email protected]
Western General Hospital, Edinburgh, UK

Edinburgh London New York Oxford Philadelphia St Louis Sydney Toronto 2014
Fifth edition © 2014 Elsevier Limited. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical,
including photocopying, recording, or any information storage and retrieval system, without permission in writing from
the publisher. Details on how to seek permission, further information about the Publisher's permissions policies and
our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency,
can be found at our website: www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright by the Publisher (other than
as may be noted herein).
First edition 1997
Second edition 2001
Third edition 2005
Fourth edition 2009
Fifth edition 2014
ISBN 978-0-7020-4959-0
e-ISBN 978-0-7020-4967-5
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
Notice
Knowledge and best practice in this field are constantly changing. As new research and experience broaden our
understanding, changes in research methods, professional practices, or medical treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any
information, methods, compounds, or experiments described herein. In using such information or methods they
should be mindful of their own safety and the safety of others, including parties for whom they have a professional
responsibility. With respect to any drug or pharmaceutical products identified, readers are advised to check the most
current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be
administered, to verify the recommended dose or formula, the method and duration of administration, and
contraindications. It is the responsibility of practitioners, relying on their own experience and knowledge of their
patients, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all
appropriate safety precautions. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or
editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability,
negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in
the material herein.
The
Publisher's
policy is to use
paper manufactured
from sustainable forests
Printed in China
Commissioning Editor: Laurence Hunter

Development Editor: Lynn Watt
Project Manager: Vinod Kumar Iyyappan
Designer/Design Direction: Miles Hitchen
Illustration Manager: Jennifer Rose
Illustrator: Antbits Ltd
(021)66485438 66485457 www.ketabpezeshki.com
Contributors Anees B. Chagpar, MD, MSc, MPH, MA

Associate Professor, Department of Surgery
Director, The Breast Center – Smilow Cancer
Hospital at Yale; Yale Comprehensive Cancer
Douglas J.A. Adamson, MBChB, MD, PhD, FRCP Center – Diversity/Disparities Program Director,
(Edin), FRCR Yale Interdisciplinary Breast Fellowship, Yale
Consultant Clinical Oncologist and Honorary University School of Medicine, New Haven, CT,
Senior Lecturer, Tayside Cancer Centre, USA
Ninewells Hospital, Dundee, UK
Daniel Xavier Choi, MD
Andrew D. Baildam, MD, FRCS United States Air Force, Air Mobility Command,
Professor of Breast and Oncoplastic 60th Surgical Operations Squadron, Travis Air
Surgery, Queen Mary, University of London, Force Base, Fairfield, California, USA
London, UK
Krishna B. Clough, MD
Nicola L.P. Barnes, MD, FRCS Chief of Surgery, Paris Breast Center, Paris, France
Specialist Registrar Breast Surgery, South
Manchester University Hospital, Manchester, Tim Davidson, ChM, MRCP, FRCS
UK Consultant Breast Surgeon, University
Department of Surgery, Royal Free Hospital,
Tom Bates, FRCS London, UK
Honorary Professor of Surgery, Centre
for Professional Practice, University of
Kent, Canterbury, UK
Nigel J. Bundred, MD, FRCS

Professor in Surgical Oncology, University of John A. Dewar, MA, FRCR, FRCP
Manchester; Consultant Surgeon, University Consultant Clinical Oncologist and Honorary
Hospital of South Manchester NHS Foundation Professor of Clinical Oncology, Ninewells
Trust, Manchester, UK Hospital and Medical School, Dundee, UK
Sussex Health Outcomes Research and
J. Michael Dixon, BSc(Hons), MBChB, MD, Education in Cancer (SHORE-C), Brighton and
FRCS, FRCSEd, FRCPEd(Hon) Sussex Medical School, University of Sussex,
Professor of Surgery, University of Edinburgh; Brighton, UK
Clinical Director, Edinburgh Breakthrough Unit;
Consultant Surgeon, NHS Lothian, Edinburgh Rosalie Fisher, MBChB, FRACP
Breast Unit, Western General Hospital, The Breast Unit, Royal Marsden
Edinburgh, UK Hospital, London, UK
Ian O. Ellis, BMBS, BMedSc, FRCPath Gerald Gui, MS, FRCS, FRCS(Ed)
Professor of Cancer Pathology and Honorary Academic Surgery (Breast Unit), Royal Marsden
Consultant Pathologist, University of NHS Trust, London, UK
Nottingham, School of Medicine, City
Hospital, Nottingham, UK Valerie Jenkins, BSc, DPhil
Sussex Health Outcomes Research and
D.Gareth R. Evans, MD, FRCP Education in Cancer (SHORE-C), Brighton and
Honorary Professor of Medical Genetics and Sussex Medical School, University of Sussex,
Cancer Epidemiology, The University of Brighton, UK
Manchester; Consultant in Medical Genetics and
Cancer Epidemiology, Central Manchester
Hospitals NHS Foundation Trust and The
Christie NHS Foundation Trust, Saint Mary's
Hospital, Manchester, UK
Lesley Fallowfield, BSc, DPhil, FMedSci vii

Professor in Psycho-oncology and Director,
Contributors City Hospital, Nottingham, UK
Monica Morrow, MD, FACS

Gabriel J. Kaufman, MD
Department of breast cancer and Chief, Breast Service, Department of Surgery;
reconstructive surgery, Paris Breast Center, Anne Burnett Windfohr Chair of Clinical
Paris, France Oncology, Memorial Sloan-Kettering Cancer
Center; Professor of Surgery, Weill Medical
College of Cornell University, New York, NY, UK
Ian Kunkler, MA, MB, BCHIR, FRCR, FRCPE
Professor of Clinical Oncology, University of
Edinburgh; Consultant in Clinical Oncology, Claude Nos, MD
Western General Hospital, Edinburgh, UK Department of breast cancer and
reconstructive surgery, Paris Breast Center,
Paris, France
Ava Kwong, MBBS(Lon), BSc(St Andrews),
FRCS, FRCS (Edin), FCSHK, FHKAM (surgery)
Clinical Associate Professor, Chief of Division of Cameron Raine, MBChB, MD, FRCS(plastics)
Breast Surgery, The University of Hong Kong, Department of Plastic Surgery, St John's
Hong Kong, China Hospital at Howden, Livingston, UK
Pamela Levack, MBChB, BMedBiol, MRCGP, Richard M. Rainsbury, MBBS, BSc, MS,
FRCP Consultant in Palliative Medicine, FRCS Consultant Oncoplastic Breast
Ninewells Hospital; Honorary Senior Lecturer in Surgeon, Royal Hampshire County Hospital,
Surgery and Molecular Oncology, University of Winchester, UK
Dundee, Dundee, UK
R. Douglas Macmillan, MBChB, MD, FRCS

Oncoplastic Breast Surgeon, Associate Clinical
Director, Nottingham Breast Institute, Nottingham
Nottingham, UK
Mark Schaverien, MBChB, MD, MSc, MEd,

FRCS(Plast)
Specialist Registrar, Department of Plastic
Surgery, Ninewells Hospital, Dundee, UK
Ian Smith, MD, FRCP, FRCPE

Professor of Cancer Medicine; Head, The
Breast Unit, Royal Marsden Hospital,
London, UK
Sarah Tang, MBBS, MA, MRCS

Specialist Registrar in Breast Surgery, Breast
viii Unit, Royal Marsden Hospital, London, UK
Alastair M. Thompson, BSc(Hons), MBChB, MD,

Emad A. Rakha, MB BCh, PhD, FRCPath FRCSEd
Clinical Associate Professor and Honorary Professor of Surgical Oncology, Dundee
Consultant Pathologist University of Cancer Centre, University of Dundee,
Nottingham, School of Medicine, City Dundee, UK
Hospital, Nottingham, UK
Steven Thrush, MBBS, FRCS(Gen Surg)
Rajendra S. Rampaul, MBChB, MD, FRCS Consultant Surgeon, Worcester Breast
Consultant Oncoplastic Breast Surgeon, The Unit, Worcestershire Royal Hospital,
Breast Centre, Woodbrook, Trinidad and Tobago Worcester, UK
John F.R. Robertson, MBChB, FRCS A. Robin M. Wilson, MBChB, FRCR, FRCP(E)
Professor of Surgery, University of Consultant Radiologist, Clinical Radiology, The
Nottingham; Consultant Surgeon, City Royal Marsden, London, UK
Hospital,
Series Editors' preface

grateful for the support and encouragement of
Laurence Hunter and Lynn Watt at Elsevier Ltd.
We trust that our original vision of delivering an up
It is now some 17 years since the first edition of to-date affordable text has been met and that read
the Companion to Specialist Surgical Practice ers, whether in training or independent practice,
series was published. We set ourselves the task of will find this Fifth Edition an invaluable resource.
meeting the educational needs of surgeons in the
later years of specialist surgical training, as well O. James Garden, BSc, MBChB, MD,
as consultant surgeons in independent practice FRCS(Glas), FRCS(Ed), FRCP(Ed),
who wished for contemporary, evidence-based FRACS(Hon), FRCSC(Hon), FRSE
information on the subspecialist areas relevant to
Regius Professor of Clinical Surgery, Clinical
their general surgical practice. The series was
Surgery School of Clinical Sciences, The
never intended to replace the large reference sur
University of Edinburgh and Honorary Consultant
gical textbooks which, although valuable in their
Surgeon, Royal Infirmary of Edinburgh
own way, struggle to keep pace with changing
surgical practice. This Fifth Edition has also had
Simon Paterson-Brown, MBBS, MPhil, MS,
to take due account of the increasing speciali
FRCS(Ed), FRCS(Engl), FCS(HK)
sation in ‘general’ surgery. The rise of minimal
Honorary Senior Lecturer, Clinical Surgery School
access surgery and therapy, and the desire of
of Clinical Sciences, The University of Edinburgh
some subspecialties such as breast and vascular
and Consultant General and Upper
surgery to separate away from ‘general surgery’,
Gastrointestinal Surgeon, Royal Infirmary of
may have proved challenging in some countries,
Edinburgh
but has also served to emphasise the importance
of all surgeons being aware of current devel
opments in their surgical field. As in previous
editions, there has been increasing emphasis on
evidence-based practice and contributors have
endeavoured to provide key recommendations
within each chapter. The eBook versions of the
textbook have also allowed the technophile
improved access to key data and content within
each chapter.
We remain indebted to the volume editors and all
the contributors of this Fifth Edition. We have en
deavoured where possible to bring in new blood to
freshen content. We are impressed by the
enthusiasm, commitment and hard work that our
contributors and editorial team have shown and
this has ensured a short turnover between
editions while maintaining as accurate and
ix
up-to-date content as is possible. We remain
Editor's preface
achieve a good outcome all you have to do is to
follow carefully the schedule you spent so much
time perfecting. But anyone who has run a mara
thon or who has edited a multi-author text knows
that it is never that simple. There are competing
I wrote in the last edition of how writing or editing demands on time as well as unexpected problems
a book is like running a marathon. At the outset to deal with. Getting all the authors to deliver
when planning to run a marathon or to edit a accord
book, you set out a program and timetable and to ing to the carefully designed schedule is never
easy, often through no fault of their own. The last numbers of potentially unnecessary interventions
few chapters, like the last few miles of a and the potential over-treatment of cancers that
marathon, are the most difficult. The Fifth Edition would not have caused problems in the patient's
has now run its course and despite ups and lifetime. This is all addressed in the new
downs and a variety of problems, it is now here revamped chapter. Benign disease is more
before you. Just as there is pride in finishing a common than breast cancer yet is often covered
marathon so I have pride in the efforts of the superficially in many texts. The chapter on benign
many who have contributed so much to bring you conditions has been expanded to include a
the best edition yet of the Breast section of the number of benign con ditions which are not
Companion Series. Although the target audience common but will not be di agnosed unless there is
is trainee and consultant surgeons with a special knowledge of them. A more detailed discussion of
interest in breast disease, I believe this book will the management of nipple dis charge including
be valuable to many more groups. Nurses ductoscopy is included in a com pletely new
working in breast clinics and operating de chapter by new authors.
partments and any doctor involved in breast clin The chapters dealing with breast-conserving sur
ics or the oncological management of breast gery and oncoplastic surgery have been
cancer should find it of great value. revamped. Written by recognised experts in their
Breast diseases and breast cancer are fortunate in fields, these chapters represent the current state
having a high public profile and being well sup of the art in breast-conserving surgery. In the last
ported financially. This irritates some colleagues edition, a chapter on mastectomy was added. Far
working in other areas but what is impressive in too many mastectomies are performed through
breast cancer are the huge strides and advances incisions that leave a poor cosmetic outcome and
in our understanding and treatment that this this excellent chapter has been reworked. It is
invest ment has produced. This means that for essential reading for all breast surgeons and now
many women breast cancer is a treatable includes Goldilocks mastectomy. The chapter on
condition rather than the death sentence it once uncommon cancers provides new and important
was. Despite these advances, there remains huge information in an area that is not well covered in
anxiety amongst women about breast cancer and many texts. One area where practice continues to
it remains women's number one fear. Reducing evolve and there have been dramatic changes
the level of fear can only be achieved through since the Fourth Edition has been the
education about the advances that have been management of the axilla. The Z0011 trial has
achieved. These new advances are all here in this divided opinion on the two sides of the Atlantic. A
Fifth Edition, which provides all the new transatlantic author presents a balanced view
of how the Z0011 trial should be interpreted and
how it has changed management of the axilla. Not
everyone has been keen to reduce the number of
axillary clearances and new trials in the UK and
the USA are planned, but what do we do in the
necessary information for doctors and nurses in meantime? Read the chapter to find out.
volved in the care of patients with breast
conditions. The chapters on imaging, including
breast screen ing and pathology, have been
revised and updated. More has been added on xi
the controversies surround ing breast screening.
Critics of screening express concern at the
Editor's preface the lat est knowledge, not only on genetics but
also on how to manage such patients surgically.
There continue to be advances in our knowledge In the USA more women than ever are choosing
of the genetics of breast cancer. A rewritten mastectomy in prefer ence to breast conservation,
chapter with the addition of a new author presents with these women often choosing bilateral
mastectomy in part because of the increasing
availability of good quality breast recon struction.
A brand new chapter with new authors covers the welcome. If there are topics we should have
advances in breast reconstruction. There have covered which have been omitted, or this text fails
been extensive revisions to the chapter on ductal to meet your expectations, please let me know. If
carcinoma in situ. Building on the success of you enjoy this book, please tell your friends and
splitting the chapters on systemic therapy into colleagues.
early, locally ad vanced and metastatic therapy, all
three chapters have been rewritten or revised. Acknowledgements
The chapter on the role of systemic therapy in
early breast cancer is authoritative as well as I would like to say a big thank you to all those who
comprehensive and would sit well in any major have contributed to this Fifth Edition. The various
text on oncology. Partial breast radiotherapy after authors past and present have been enormously
breast-conserving surgery has gained increas ing co operative and have produced consistently
popularity in the USA. Whether all older patients high-qual ity work. I would specifically like to thank
need radiotherapy after breast-conserving surgery Monica McGill, who has coordinated each chapter
continues to be debated. Controversy as to which with the authors and made the numerous
patients benefit from postoperative chest wall changes to the text as well as coping with the late
radio therapy after mastectomy abounds. There addition of numer ous new figures which, I believe,
are new ongoing trials in these areas and a new have added to the book's utility. I would also like
chapter by an author who is leading some of to thank Lynn Watt, Development Editor, for her
these trials shows that the Fifth Edition is not keen eye and endless patience with me. There
merely a rewrite but a new book. The chapter on are a number of people who have influenced me
psychosocial issues provides excellent over my career who deserve spe cial mention. I
information and advice and reflects the ex pertise first became aware of the complexi ties of the
of the authors and is an important part of the breast by working with Professor David Page, a
book. Medical negligence cases continue to pathologist from Nashville. Professor Page spent
increase and so the book finishes with some a sabbatical in Edinburgh during the year I did
sound advice and vignettes of what to watch out pathology and his enthusiasm for his topic infected
for to avoid negligence claims. Another chapter me. There are a number of surgeons who have
written by people at the top of their game and a been my heroes and who have taught me that
must-read. success only comes through hard work and
The content of this book has evolved over many dedication and these include the late John
years. Various authors over this time have been Bostwick III, who worked in Atlanta, and my friend
kind enough to share their knowledge, clinical and contemporary, Krishna Clough from Paris.
expertise and understanding because they all Professor Patrick Forrest in Edinburgh taught me
want to do better for patients and this continues to that the patient is the focus and is paramount. I
be the main motivation behind this book. It has am fortunate to have been taught and known
been reas other great individuals, too many to mention. I
suring to see people carrying, reading and quoting hope they know who they are and I thank them. I
from the Fourth Edition. I hope and trust the Fifth would like to pay a special tribute to the many
Edition will be as well received. Seeing the patients who have placed their faith and trust in
editions evolve over time brings its own rewards. me. They continue to inspire me and are the
Ultimately it will be you, the reader, who will judge reason that we all need to try harder, learn more
whether the efforts of the very many to get this and do better for them. Finally, I would like to
book to the finish line have been worthwhile. Your thank my family for their support. Writing and
feedback is always editing takes time away from those who are there
when everyone has left. Pam, my wife, has been
the rock on whom I have been able to build many
projects including this, the Fifth Edition. Oliver and
Jonathan, my sons, have helped by
xii
understanding that medicine is not just a Edinburgh
profession.
Michael Dixon
Evidence-based practice in surgery

Depending on the information available, three
grades of recommendation can be used:
Critical appraisal for developing evidence-based a. Strong recommendation, which should be
practice can be obtained from a number of followed unless there are compelling reasons
sources, the most reliable being randomised to act otherwise.
controlled clini cal trials, systematic literature b. A recommendation based on evidence of
reviews, meta-analyses and observational effectiveness, but where there may be other
studies. For practical purposes three grades of factors to take into account in decision
evidence can be used, analogous to the levels of making, for example the user of the
‘proof’ required in a court of law: guidelines
1. Beyond all reasonable doubt. Such evidence may be expected to take into account patient
is likely to have arisen from high-quality preferences, local facilities, local audit results
randomised controlled trials, systematic or available resources.
reviews or high-quality synthesised evidence c. A recommendation made where there is no
such as decision analysis, cost-effectiveness adequate evidence as to the most effective
analysis or large observational datasets. The practice, although there may be reasons for
studies need to be directly applicable to the making a recommendation in order to minimise
population of concern and have clear results. cost or reduce the chance of error through a
The grade is analogous to burden of proof locally agreed protocol.
within a criminal court and may be thought of
as corresponding to the usual standard of Evidence where a conclusion can be reached
‘proof’ within the medical literature (i.e. ‘beyond all reasonable doubt’ and therefore where
P<0.05). a strong recommendation can be given.
This will normally be based on evidence levels:
2. On the balance of probabilities. In many • Ia. Meta-analysis of randomised controlled trials
cases a high-quality review of literature may • Ib. Evidence from at least one randomised
fail to reach firm conclusions due to conflicting controlled trial
or inconclusive results, trials of poor • IIa. Evidence from at least one controlled study
without randomisation
methodological quality or the lack of evidence
• IIb. Evidence from at least one other type of
in the population to which the guidelines apply. quasi-experimental study.
In such cases it may still be possible to make
a statement as to the best treatment on the
‘balance of probabilities’. This is analogous to
the decision in a civil court where all the Evidence where a conclusion might be reached
available evidence will be weighed up and the ‘on the balance of probabilities’ and where there
verdict will depend upon the balance of may be other factors involved which influence the
probabilities. recommendation given. This will normally be based
on less conclusive evidence than that represented
3. Not proven. Insufficient evidence upon which by the double tick icons:
to base a decision, or contradictory evidence. • III. Evidence from non-experimental descriptive
studies, such as comparative studies and
case–control studies at the end of each chapter along with a short
• IV. Evidence from expert committee reports or summary of the paper's conclusions where
opinions or clinical experience of respected
applicable.
authorities, or both.
The reader is referred to Chapter 1, ‘Evidence
based practice in surgery’ in the volume, Core
Evidence which is associated with either a
Topics in General and Emergency Surgery of this
strong recommendation or expert opinion is series, for a more detailed description of this
highlighted in the text in panels such as those topic.
shown above, and is distinguished by either a
double or single tick icon, respectively. The
references associated with double xiii
tick evidence are highlighted in the reference lists
1
The role of imaging in breast diagnosis
including screening and excision of impalpable
lesions
A. Robin M. Wilson
R. Douglas Macmillan
women in the UK will develop breast cancer at
some time during their life.1
Strategies for diagnosing and managing breast
Introduction cancer are based on our current understanding of
breast disease epidemiology. Around 5% of breast
Breast cancer is a major health problem.
cancer is hereditary, mainly associated with the
Worldwide it has an increasing incidence, with
BRCA1 and BRCA2 gene defects. This type of
over 1 million newly diagnosed cases each year,
breast cancer tends to occur in younger women.
and is the common est cancer to affect women
The remaining 95% of breast cancer is sporadic
and the commonest cause of cancer death in
and its incidence increases with age. Breast
women. Breast cancer mortality in the UK is
cancer is rare under the age of 35 years and over
among the highest in the world, with
80% of breast cancer occurs in women over the
approximately 28 deaths per 100 000 women per
age of 50.
annum. This equates to around 48 000 new breast
The main causes of sporadic breast cancer are be
cancers diagnosed and 11500 deaths attributable
to breast cancer each year. Approximately 1 in 9 lieved to be environmental factors. Recognised
risk factors include early menarche, late meno women to present as soon as possible to breast
pause, nulliparity, and long-term use of the con clinics when they develop breast symptoms and
traceptive pill and hormone replacement therapy through regular breast cancer screening. Breast
(HRT). imaging is funda
As there is a poor understanding of the causes of mental to both.
breast cancer, primary prevention is currently not a
realistic or achievable option. It is known that ear
lier diagnosis of breast cancer is more likely to Imaging in symptomatic
result in a favourable outcome. Regardless of
tumour type or grade, the smaller a breast cancer
breast practice
is at the time of diagnosis, the more likely it is that
Based on mortality statistics from the 1980s and
it has not spread beyond the breast. As a result
1990s, although the UK did not have the highest
the current strategy for reducing breast cancer
incidence of breast cancer it did have the highest
mortality is to seek diagnosis as early as possible.
death rate.
Early detection and improvements in treatment
have led to a 30% reduction in breast cancer mor
tality in the UK in all age groups over the past 30
years.2
1
Early diagnosis is achieved by encouraging
Chapter 1 their breasts. The clinical and imaging
assessments in these clinics should be performed
by appropriately trained and experienced staff. All
Recognising that breast cancer diagnosis and such staff do not need to be doctors but they need
treatment required significant improvement, in 1995
to be able to give an independent opinion and be
the UK Department of Health published guidelines
for improving outcomes in breast cancer. These trained to an appropriate level.
guidelines were updated in 2002 and 2010 by the
National Institute for Clinical Excellence (NICE).3 The
guidelines emphasise the following three key issues
Breast imaging
in breast cancer care:
• accurate and timely diagnosis;
techniques Mammography
• appropriate treatment decided by accurate staging
of disease;
X-ray mammography has been the basis of breast
• appropriate follow-up of patients imaging for more than 30 years. The sensitivity of
undergoing treatment. mammography for breast cancer is age
dependent. The denser the breast, the less
Imaging is required at all three stages of this pro effective this method is for detecting early signs of
cess, and mammography and ultrasound have a breast cancer. Breast density tends to be higher
pivotal role to play. About 60% of breast cancer in younger women and increased density
is diagnosed in symptomatic breast referral clin obscures early signs of breast can
ics. These clinics follow protocols that define the cer. The sensitivity of mammography for breast
triple test, the combination of clinical assessment, can cer in women over 60 years of age
imaging (mammography and ultrasound) and core approaches 95%, while mammography can be
biopsy or needle cytology, as the required expected to detect less than 50% of breast
standard.4 ‘One-stop’ clinics are recommended at cancers in women under 40 years of age.5
which all the necessary tests required to make a
diagnosis, includ ing needle biopsy, are performed
at one clinic visit. In order to achieve the earliest 2
possible diagnosis of symptomatic breast cancer,
women are encouraged through a variety of
health promotion methods to present to these Mammography uses ionising radiation to obtain
clinics as soon as they develop any change in
an image and therefore should only be used the dense breast, the application of
where there is likely to be a clinical benefit. computer-aided detection and a number of
Consensus is that the benefits of mammography logistical advantages providing potential for more
in women over the age of 40 years are likely to far efficient mammogra phy services.10,11 The much
outweigh any oncogenic ef wider dynamic range of digital mammography
fects of repeated exposure. Screening of women means that visualisation of the entire breast
over the age of 40 by mammography is accepted density range on a single image is easily
practice. However, in symptomatic practice, there achievable. In the clinical setting, compara tive
is rarely an indication for performing studies have shown that digital mammography
mammography in women under the age of 35 performs in general as well as film/screen mam
unless there is a strong clinical suspicion of mography but is better in younger women and in
malignancy. In many centres, all women over the women with dense breasts.6,7
age of 35 presenting to breast clinics un Mammography is the basis of stereotactic breast
dergo mammography as a routine. Practice is biopsy, which can be carried out using a dedi
chang ing and ultrasound alone is being used cated prone biopsy table or by using an add-on
increasingly for the assessment of women with device to a conventional upright mammography
focal breast symp toms in women under 40 years unit. Stereotaxis is used to biopsy impalpable le
of age and even in some women aged 40–50. sions that are not clearly visible on ultrasound
Mammography is routine in all women in the (e.g. microcalcifications).
screening age group attending symptomatic
clinics who have not had a screening
Ultrasound
mammogram in the past year.
Most mammography in the UK is now carried High-frequency (≥10MHz) ultrasound is a very
out using digital image acquisition.6–9 There are effective diagnostic tool for the investigation of
major benefits from acquiring mammograms in di focal breast symptoms.12 Ultrasound does not
rect digital format.9 Compared with conventional involve
film/screen mammography, the benefits of full-field
digital mammography include better imaging of

Ultrasound is less sensitive than mammography
for the early signs of breast cancer and is
ionising radiation and is a very safe imaging tech therefore not used for popula tion screening.
nique. It has a high sensitivity for breast pathology However, ultrasound does increase the detection
and also a very high negative predictive value. 13 of small breast cancers in women who have a
dense background pattern on mammogra phy.5 In
High-resolution ultrasound easily distinguishes the screening setting there is clear evidence that
between most solid and cystic lesions and can the addition of ultrasound improves small cancer
differentiate benign from malignant lesions with a detection rates, particularly in women with dense
high degree of accuracy. However, in most
circumstances, solid lesions seen on ultrasound
breasts, but there is currently insufficient evi dence
require needle sampling for accurate diagnosis. of any mortality benefit and insufficient re sources
to allow for routine ultrasound screening of
Ultrasound is the technique of choice for the fur women with dense breasts on mammography.
ther investigation of focal symptomatic breast prob Adding ultrasound to mammography or magnetic
lems at all ages. Under 40 years of age, when the resonance imaging (MRI) screening does increase
risk of breast cancer is very low, it is usually the cancer detection but also significantly increases
only imaging technique required. Over 40, when the false-positive rate. Ultrasound is the technique
the risk of breast cancer begins to increase, it is of first choice for biopsy of both palpable and
often used in conjunction with mammography. impal
pable breast lesions visible on scanning. breast cancer, approaching 100% for invasive
Ultrasound is now used routinely to assess the ax cancer and up to 92% for ductal carcinoma in situ
illa in women with breast cancer in most units. (DCIS), but it has a high false-positive rate.18,19,20
Axillary nodes that show abnormal morphology Rapid acquisition of images facilitates
can be sampled accurately by fine-needle assessment of signal enhancement curves that
aspiration (FNA) or needle core biopsy.16,17 can be helpful in distinguishing benign from
Discussion of the role of axillary ultrasound and malignant disease. Significant overlap in the
FNA and core bi opsy can be found in Chapter 7. enhancement patterns is seen, so needle
sampling of the lesions detected is often required.
Up to 50% of patients with axillary metastatic Magnetic resonance-guided breast biopsy is
disease can be diagnosed by ultrasound combined available in a few centres but most breast lesions
with FNA or core biopsy, which may avoid the need
seen on MRM that are larger than 5mm can be
for sentinel node biopsy in some women with
pathologically proven nodal involvement.14,15
seen on ultrasound if they are clinically significant.
MRM is the best method for screening younger

Doppler ultrasound adds little to breast diag nosis
women (under 40 years) at increased risk of breast
and is not widely used. Three-dimensional cancer but, because of cost, it is unlikely to be used
The role of imaging in breast diagnosis for general population screening.21–24
ultrasound of the breast is said to increase the ac MRM is the best technique for imaging women
curacy of biopsy and the detection of multifocal with breast implants. It is also of benefit in identi
disease but is not widely available. Elastography fying recurrent disease where conventional imag
is a new application of ultrasound technology that ing and biopsy have failed to exclude recurrence.
allows the accurate assessment of the stiffness of Provided it is carried out more than 18 months
breast tissue. It is being evaluated at present and after surgery, MRI will accurately distinguish
may prove to be a useful tool in excluding signifi between scarring and tumour recurrence. MRI is
cant abnormalities, for instance in assessment of recom mended to detect multifocality prior to
asymptomatic abnormalities detected by conserva tion surgery for women with lobular
ultrasound screening. cancers and those with occult cancer on
mammography or with a significant discrepancy of
Magnetic resonance size on conventional imaging. However, it should
not be used routinely and a randomised trial
mammography (MRM) (COMICE) showed no ben efit in reducing
re-excision rates but did increase
MRM is now widely available. In order to image
the breast the patient is scanned prone and
injection of intravenous contrast is required. MRM
is the most sensitive technique for detection of
3

Chapter 1
Computed tomography
significantly the mastectomy rates for those who
had MRI. Breast MRI is of value in assessing
Computed tomography (CT) has no current
response of large or locally advanced breast proven role in primary imaging of the breast. but
cancers to neoadju vant chemotherapy. MRI of the dedicated low-dose breast CT is being
axilla can demon strate axillary metastatic disease introduced; its clinical efficacy has yet to be
but its sensitivity is not sufficient to replace determined. CT is used to diag
surgical staging of the axilla. For advanced breast
nose and stage systemic spread of breast cancer
cancer, MRI is the technique of choice for
and to assess response of metastatic disease to
assessing spinal metastatic disease.
treatment, particularly lung, pleural and liver
metastases. Some patients with breast cancer do
have their cancers di Cancer committee of experts have reaffirmed the
agnosed incidentally during a routine CT scan. mortality benefit of mammographic screening and
determined that criticisms of the mammographic
screening trials were unjustified.25–30
Isotope imaging
The mortality benefit of screening is greatest in
Breast scintigraphy is not widely used because of women aged 55–70 years.27,28 The mortality
its lack of sensitivity compared with other benefit of screening women aged between 40 and
techniques. It is used in a few centres to stage the 55 is ap proximately 20%. Screening women
breast prior to surgery and to assess response to under the age of 40 has not been shown to
treatment. Scintigraphy is widely used to provide any mortality benefit.27,28
diagnose and assess the presence of skeletal
metastatic disease. It is now often combined with
a low-resolution non-contrast CT scan. Population screening
Positron emission scintigraphy, particularly when
combined with CT, is a technology that may have Breast screening has been introduced in many
a role in future in staging breast cancer and countries over the past 25 years. In most
monitor ing response to treatment. It is widely countries, screening is recommended in all
used in the USA but elsewhere, at present, it women aged 40 and over but in countries that
continues to be regarded as a research tool and provide population-based screening, women of 50
its specific uses and indications in assessing and over are specifically targeted. Breast cancer
breast disease are still being defined. screening was introduced in the UK in 1987 and
provides screening by invita
tion, free at the point of delivery, to all women be
Breast cancer tween the ages of 50 and 70.1 Women over 70 can
attend but are not invited. Over 70% of the invited
screening Aim population need to attend for a significant overall
mortality benefit to be achieved. Women under the
The aim of breast screening is to reduce mortality age of 50 are not offered screening in the UK un
through early detection. Randomised controlled tri less they are at increased risk. A randomised trial
als and case–control studies carried out between of extending the screening invitation to the age
the 1960s and 1980s have demonstrated that range 47–73 years was started in 2010. The
popula tion screening by mammography can be results of this trial will not be available until after
expected to 2020. From 2010 screening with annual MRI in
addition to mammography has been offered to
women at very high risk of breast cancer,
4 including BRCA1 and BRCA2 gene carrriers.
reduce overall breast cancer mortality by around Method and frequency

25% and by 35–40% in those who participate.
The optimum screening method is two-view
mammography; clinical examination of the breast
The validity of these trials was questioned in and breast self-examination have not been shown
2000–2002 but subsequent reviews by the to contribute to mortality reduction through early
Swedish combined trials group and a World Health detection and so are not recommended.9,31,32
Organisation International Agency for Research on

screening set by a central Department of Health
Advisory Committee (Table 1.1).
Women aged 50–70 in the UK are invited for The role of imaging in breast
mammography every 3 years. There has been
some concern that this screening interval is too diagnosis The screening process
long. Mammography can be expected to detect
breast cancer approximately 2 years before it
Women invited for screening attend either a static
becomes clinically apparent. The frequency of
or mobile screening unit where two-view mammog
mammographic screening is determined by the
raphy is performed. The images are then double
lead-time of breast cancer. Based on the average
read within a few days. The vast majority of
growth time of breast cancer in different ages, this
women (95%) are informed by letter within 2
means that mammographic screening should
weeks of attendance that their mammograms
ideally be carried out yearly in women aged
show no evi dence of breast cancer. Those
40–50, every 2 years in women aged 50–60 and
women in the appro priate age range will be
every 3 years thereafter. However, the UK breast
invited for screening 3 years later. They are
screening frequency trial completed in 1995 did
advised to contact their general prac titioner as
not show any predicted benefit for women aged
soon as possible if they become aware of
50–64 screened every year compared with those
any change in their breasts in the meantime. The
screened every 3 years.33 Screening once every 3
screening process includes a fully integrated
years can be expected to detect approximately
multidisciplinary assessment process for all screen
two-thirds of all breast cancers that will arise
detected mammographic abnormalities; screening
during the 3-year screening interval.
programmes should ideally retain responsibility
One-third of breast cancers present in the inter
through to definitive diagnosis. Approximately 5%
val between screens and are called interval can
of women screened are recalled for further
cers. Half of these present in the third year after
assessment of a problem identified at screening.
screening.
Some women are recalled for further assessment
of a clinical sign or symptom identified at the time
Factors affecting the effectiveness of screening but the vast majority of women are
of screening recalled because of a mammographic abnormality.
HRT increases breast density and in a proportion The most important cancers detected at
of women this reduces the sensitivity of screening are high-grade DCIS, as most cases of
this type will progress to grade 2 or 3 invasive
mammography for breast cancer.9,32–40 Up to 25%
breast cancer within the following 3 years, and
of women tak ing combined grade 2 and 3 invasive breast cancers under 10mm
oestrogen/progestogen preparations continuously in diameter, as at this size these tumours are much
show increased density on mammogra phy. This less likely to have metastasised.41,42
effect is significantly less with other HRT
preparations. As well as reducing sensitivity HRT The common types of mammographic
also reduces the specificity of mammographic abnormality and their positive predictive value for
screen ing. HRT also increases the risk of cancer are shown in Table 1.2. Well-defined
developing breast cancer.34 masses are almost always benign and do not
require recall, whereas ill-defined
masses and spiculated lesions always require
Quality assurance further assessment (Figs 1.1 and 1.2). Clustered
microcalcifi cations account for a high proportion
Breast screening programmes should have inbuilt
of recalls that result in needle biopsy. More than
quality assurance; the UK NHS Breast Screening
20% of screen-de tected breast cancer is DCIS,
Programme is subject to comprehensive quality
mostly high or intermedi ate grade, and most of
assessment and all 100 screening units across the
this type of cancer is detected by the presence of
country have to comply with nationally defined
clustered microcalcifications (Fig. 1.3). Invasive
standard guidelines. There are national targets for
cancer is usually represented on mammog raphy imaging (mammography and/or ultrasound) and
by either an ill-defined or spiculated mass. It is clinical assessment before being reassured and
essential to detect these lesions at small size as
they more commonly represent grade 2 or 3
invasive cancer. 5
Three-quarters of women recalled undergo fur ther
Chapter 1
Table 1.1 • NHS Breast Screening Programme: screening targets, June 2005 Objective
Criteria Minimum standard Target
To maximise the women to attend for • Incident screen

number of eligible (a) Rate of invasive screening ≥0.5 per 1000
women who attend cancers detected in • 80%
for screening eligible women • Prevalent screen
To maximise the ≥2.7 per 1000
number of cancers • Incident screen • Prevalent screen
detected (b) Rate of cancers ≥3.1 per 1000 >3.6 per 1000
Percentage of eligible detected that are in • Prevalent screen • Incident screen
women who attend for situ carcinoma ≥0.4 per 1000 ≥4.2 per 1000
screening • ≥70% of invited
ratio
• ≥0.85 • ≥1.0
(c) Standardised detection

Rate of invasive • ≥12 lp/mm
(b) Minimal
To maximise the cancers less than • Prevalent screen
detectable
number of small 15mm in diameter ≥2.3 per 1000
contrast:
invasive cancers detected in eligible
• Prevalent screen
women invited and
≥1.7 per 1000
screened • Incident screen
≥2.5 per 1000
• Incident screen
To achieve optimum (a) High-contrast ≥1.7 per 1000
image quality spatial resolution
5–6mm detail • ≤1.2% • ≤0.8% 0.5mm detail • ≤5% •

≤3% 0.25mm detail • ≤8% • ≤5% (c) Aim film density •
1.5–1.9
To limit radiation dose Mean settings
glandular dose per film for a • ≤2.5mGy
standard breast at clinical
women undergoing repeat • ≤3% of total examinations •
examinations <2% of total examinations
To minimise the number of Number of repeat examinations
screened who are referred for • Prevalent screen
referred for further assessment <7% • Incident
To minimise the tests* • Prevalent screen screen <5%
number of women (a) Percentage of <10% • Incident
women who are screen <7%
diagnosis of cancer after a maximum of two
(b) Percentage of women visits
screened who are placed • <0.5% • ≤0.25% • ≥80%
on early recall
To ensure that the majority Percentage of women
of cancers, both palpable who have a non-operative • >90%
and impalpable, receive a diagnosis of cancer by
non-operative tissue cytology or needle histology
Rate of benign biopsies • • Prevalent screen <1.8 per 1000
Prevalent screen <3.6 per 1000
To minimise the number of
unnecessary operative
procedures
(021)66485438 66485457
6 www.ketabpezeshki.com
Table 1.1 • (cont.) NHS Breast Screening Programme: screening targets, June 2005
Objective Criteria Minimum standard Target

• Incident screen <1.0 per
1000
To minimise diagnostic delay for
To minimise the number women who are diagnosed
Rate of cancers presenting in non-operatively
• Incident screen <2.0 per 1000
Expected standard To minimise the delay for women
of cancers in the women who require surgical assessment
screened presenting between episode
screening episodes Percentage of eligible
women whose first offered
To minimise any delay for
women who require treatment appointment is within
for screen-detected breast 36 months of their previous
cancer screen
screened women: Percentage of women who are
To ensure that women are sent their result within 2 weeks
recalled for screening at Percentage of women who
appropriate intervals (a) In the 2 years following a attend an assessment centre
normal screening episode (b) In within 3 weeks of attendance for
the third year follow ing a normal the screening mammogram
To minimise anxiety for women screening
Proportion of women for whom
awaiting the results of screening the time interval between
To minimise the interval from the non-operative biopsy and result
screening mammogram to is 1 week or less
assessment
Proportion of women for whom
the time interval between the
decision to refer to a surgeon women screened in the third year
and surgical assessment is 1
week or less • ≥90% • 100% • ≥90% •
Percentage of women who are • >90% • 100%
admitted for treatment within 2
months of their first assessment
100%
visit
• >90% • 100% • >90% •
• 1.2 per 1000 women screened

100% • >90% • 100%
in the first 2 years • 1.4 per 1000
discharged. The remaining 25% proceed to needle

biopsy and this results in a diagnosis of around six
cancers per 1000 women screened. Interval follow
up of uncertain mammographic findings is dis
couraged; the emphasis is on obtaining a definitive
diagnosis by the use of image-guided breast
Table 1.2 • Positive predictive value (PPV) for
malignancy of mammographic biopsy. Over 90% of breast cancers should be
signs diagnosed prior to first surgery. Despite advances
in breast needle biopsy techniques, a small
Sign PPV (%)
proportion of women still require open surgical
Well-defined mass <1 biopsy for diag nosis (up to 0.25% of women
Ill-defined mass 35–50 screened).
Spiculated mass 50–90
Architectural distortion 20–40
Asymmetric density <2 7
Clustered microcalcifications 15
Chapter 1
a
b Figure 1.1 • (a,b) Digital mammograms showing multiple
well-defined masses, the typical appearance of simple breast
cysts. (c) Ultrasound image showing typical features of a
simple cyst, i.e. a well-circumscribed anechoic mass with
distal acoustic enhancement.
a b
c Figure 1.2 • (a,b) Digital mammograms showing a spiculated
mass in the right breast. The appearances are typical of an invasive carcinoma. The mass contains
microcalcifications and there is evidence of skin tether. (c) Doppler ultrasound image showing the typical features of
an invasive carcinoma with an irregular mass containing abnormal central vessels.
8
Receiving an invitation for screening and attend

ing for mammography are not associated with any
significant anxiety. However, recall for further
Table 1.3 • NHS Breast Screening Programme: results

2010
assessment does cause measurable anxiety,

although this has largely subsided by 3 months.
The numbers of women who undergo open sur
gical biopsy for what proves to be benign disease
should be kept to a minimum. Considerable train
ing and investment in equipment has resulted in a
fourfold decline in benign surgical biopsies gener
ated through the screening programme, although
rates of benign biopsy vary from unit to unit. False
Figure 1.3 • Digital mammograms showing a cluster of positive recall and benign surgical biopsy are both
casting microcalcifications in the left breast. The
appearances are typical of high-grade ductal more likely in younger women.
carcinoma in situ. Overdiagnosis refers to the detection by screen
ing of breast cancers that require treatment but
The performance of the NHS Breast Screening that would never have threatened the life of the
Programme in 2010 is shown in Table 1.3. The woman. This is inevitable in screening and there
screening programme was predicted to produce a continues to be considerable debate about what
25% reduction in mortality (1750 cancers per year) proportion of screen-detected breast cancers fall
directly attributable to early detection through into this cat egory. It is likely that most low-grade
screening by the year 2010. DCIS and some special low-grade invasive
cancers represent overdiagnosis, and detection of
Adverse effects these cancers results in unnecessary treatment
and unnecessary morbid ity associated with
knowledge of the diagnosis of cancer. The fully about both the likely positive and negative
consensus view is that overdiagnosis ap plies to effects of screening. Efforts continue to reduce
no more than 10% of screen-detected breast the morbid ity of screening and to understand
cancer and that at this level this does not negate better the natural history of cancers so that
the overall mortality benefit of breast screening. potential over-treatment is minimised. Even the
Women attending for screening must be informed most ardent critics of breast
2008/2009 2009/2010
Total number of women invited (50–70) 2 642 511 2 662 298 Acceptance rate 73.8%
73.5% Number of women screened (invited) 1947.424 1 954 815 Number of women
screened (self-referral) 44,070 43 410 Total number of women screened 1 991 494 1
998 225 Number of women recalled for assessment 87,400 82 650 Percentage of
women recalled for assessment 4.4% 4.1% Number of benign surgical biopsies
1644 1519 Number of cancers detected 15 673 15 517 Number of in situ cancers
detected 3253 3064 Number of cancers less than 15mm 6460 6544 Standardised
detection ratio (invited only) 1.45 1.44
From NHS Breast Screening Programme Annual Review 2011. Available at http://www.cancerscreening.nhs.uk.
9
Chapter 1 genetic mutation are termed high risk.
screening believe screening should continue but NICE guidelines have been produced (latest
they argue it needs to be refined. This is accepted version 2006) to classify risk groups and guide
by those who organise and run screening care.43 They state that women at or near population
risk should be managed in primary care (defined as
programmes.
<3% risk for women aged 40–49). Women at
moderate risk are those with a 10-year risk of 3–8%
Screening women at between 40 and 49 years or a lifetime risk >17%,
increased risk and those at high risk are defined as those with a
10-year risk of >8% between 40 and 49 years or a
lifetime risk >30%.
Women at increased risk of developing breast
cancer due to a proven inherited predisposing
Such cut-offs are useful as guidelines for
genetic muta tion, family history (with no proven specialist referral, but most risk factors require
genetic mutation), previous radiotherapy (e.g. clinical inter pretation before risk management is
mantle radiotherapy for Hodgkin's lymphoma) or discussed with the individual.
benign risk lesions (atypical hyperplasia, lobular
Unfortunately, the problem with screening young
intraepithelial neoplasia) may be selected for
women at increased risk of breast cancer is that
screening at young age. Whether it is pos sible to
no screening test has as yet been shown to
identify other substantially increased risk groups
reduce mortality in such women. Screening in this
by summating various other epidemiological
group is therefore a management option for which
factors (e.g. age at menarche, body mass index,
an exact benefit cannot be quoted for an
age at first pregnancy, alcohol intake) continues to
individual woman. Screening should not be
be de bated. The cut-off point at which clinical
offered to those who fall below the moderate-risk
management of a woman is altered is often
cut-off.
referred to as moderate risk. Those individuals
likely or proven to be carriers of a predisposing
sociated microcalcification and no architectural
distortion (Fig. 1.4a). Such cancers often present
Methods of screening young women
symptomatically as interval cases. BRCA2-related
at increased risk: mammography
cancers are more similar to sporadic cases and
A national evaluation study of mammographic may be more likely to be detected by
screening for young women with a family history of mammography. Ultrasound screening significantly
improves sensi
tivity when there is a dense mammographic back
10 ground pattern but has a lower positive predictive
value and has not been shown to be a useful
screen ing modality. Ultrasound features of
breast cancer has been conducted (FH01 study). BRCA1 can cers are often benign or indeterminate
The study compared screening in women aged (Fig. 1.4b). If mammographic screening is
40–44 who were at moderate or high risk with the performed, it should be repeated annually in
control arm of the age trial (population-based trial women under age 50.
of screen
ing women in their forties in which the control arm Methods of screening young
was not screened); this showed a non-significant women at high risk: MRI
mortality benefit for screening. It demonstrated MRI is the most sensitive method of imaging
that screening was likely to reduce deaths from young women but has significant resource
breast cancer. The FH02 study is currently
implications.44,48 The specificity of MRI has been a
looking at the effectiveness of screening women
concern, although with second-look recall (after
aged 35–39.
which many poten tially abnormal findings may
resolve), targeted ultrasound and the slowly
Mammography has a greater positive
predictive value in young women at high risk when increasing availability of MRI-guided biopsy, this
compared with age-matched controls but lacks may be less of a problem than initially thought.
sensitivity. This may be a particular problem in The MARIBS study evaluating MRI in addition to
women with BRCA1 mutations.24,44–47 mammography and several other studies have
shown that MRI is the most sensitive screening
BRCA1-related breast cancer is usually high test for young high-risk women, but it is arguable
grade and often has a ‘pushing’ margin. It rarely whether the cancers detected are sufficient to
presents with associated DCIS. The change the outcome of these young women.49 On
mammographic features are therefore usually of a the basis of its better performance compared with
mass lesion with no as

c
mammography in increasing sensitivity, NICE has
recommended annual MRI surveillance for women
aged 30–39 years with a 10-year risk of >8% and
women aged 40–49 years with a 10-year risk
ab
>20%.43
The role of imaging in breast MRI is also recommended for high-risk women
with a dense mammographic background pattern.
diagnosis Age to start screening in young
women at increased risk
The age for starting screening should be based on
risk rather than the age of affected relatives. For
women at moderate risk, screening should start at
age 40. This can seem paradoxical if the reason
moderate risk has been established is because
there is one first-degree relative affected in their
thirties. However, if this is the only relative
affected then the individual is only at moderate
risk and the emphasis
Figure 1.4 • (a,b) Digital mammograms showing a small
circumscribed but ill-defined mass in the upper outer
right breast. (c) Ultrasound image of the same mass.
Core biopsy showed grade 3 invasive ductal carcinoma.
of management should be on reassurance rather

than screening. For women at high risk, screening
may be started at age 30. Such high-risk women
should ideally be managed in a specialist setting Needle biopsy is highly accurate in determining
where experience with MRI screening is the nature of most breast lesions.50–53 Patients
available. In the UK such women may be best with benign conditions avoid unnecessary
managed within the NHS Breast Screening surgery; carrying out
Programme. Women must be advised about the
limitations of screening at young age. This is
particularly relevant to known mutation carriers for 11
whom there is no evidence that screening
improves survival (cf. risk-reducing surgery).
Image-guided breast biopsy

Chapter 1 90 units were able to achieve the target of 70%
pre
open surgical biopsy for diagnosis should be operative diagnosis rate for cancer. By 2003, most
regarded as a failure of the diagnostic process. units had converted to automated core biopsy and
For patients who are proven to have breast all units achieved the minimum standard and the
cancer, needle biopsy pro vides accurate majority exceeded the expected standard of 90%
understanding of the type and extent of disease preoperative diagnosis rate.1
so ensuring that patients, and the doctors treating
them, are able to make informed treatment Vacuum-assisted biopsy (VAB)
choices. Needle biopsy not only provides accurate
The predominant reasons for not achieving an
in formation on the nature of malignant disease,accu rate diagnosis by needle biopsy are sampling
such as histological type and grade, but also error (missing the target) and failure to retrieve
facilitates pre treatment assessment of tumour sufficient representative material. These problems
biology. have been largely addressed by the development
of larger directional core techniques that yield
significantly greater volumes of tissue.54–56
Which biopsy technique? VAB is a very successful method for improving
the diagnostic accuracy of borderline breast
The current methods available for breast tissue lesions and lesions at sites in the breast difficult to
diagnosis are FNA cytology, needle core biopsy, biopsy using
vacuum-assisted biopsy (VAB) and open surgical
biopsy.
12
FNA versus needle core biopsy
There has been much debate about the
comparative benefits of FNA and core biopsy,50–53 other techniques. VAB has been shown to
but 14G 22-mm automated core biopsy provides understage both in situ and invasive cancer
significantly greater sensitivity, specificity and approximately half as often as conventional core
positive predictive value than FNA. Results with biopsy (typically 10% vs. 20%).57,58 The VAB
core biopsy are particularly superior to FNA in technique has a higher sen
stereotactic biopsy of microcalci
sitivity because it allows sampling of lesions at
fications and architectural distortions.
sites that are difficult to biopsy using either FNA
The overall better performance achievable with
or core biopsy and because the amount of tissue
core biopsy compared with FNA is illustrated in harvested is at least five times greater per core
the performance of the NHS Breast Screening specimen.
Programme in the UK. In 1994, using FNA as the
The indications for VAB include:
primary diagnostic technique, fewer than 10% of
• very small mass lesions; needle biopsy
• architectural distortions;
• failed ‘conventional’ core biopsy; Ultrasound guidance is the technique of
choice for biopsy of both palpable and impalpable
• small clusters of microcalcifications;
breast lesions; it is less costly, easy to perform and
• papillary and mucocele-like lesions; more accurate than freehand or other image-guided
• diffuse non-specific abnormality; techniques.50
• excision of benign lesions;
• sentinel node sampling. Ultrasound provides real-time visualisation of the
Core biopsy and VAB are now the recom biopsy procedure and visual confirmation of
mended techniques for sampling calcifications adequate sampling. Between 80% and 90% of
and mammographic architectural distortions. 59,60 breast abnormalities will be clearly visible on
For calcifications it is imperative that there is proof ultrasound and amenable to biopsy using this
of representative sampling with specimen technique. For impalpable abnormalities not
radiography. If calcification is not demonstrated on visible on ultrasound, stereotactic X-ray-guided
the specimen radiograph and the histology is biopsy is required. A few lesions are only visible
benign, then management cannot be based on on MRI and require magnetic resonance-guided
61
this result as there is a high risk of sampling error; biopsy. A number of different approaches have
the procedure must either be repeated or open been developed for this proce dure using both
surgical biopsy carried out. closed and open magnets. VAB is
Guidance techniques for breast

than X-ray guidance. These markers dissolve and
are reabsorbed in a few weeks, leaving a small
the biopsy technique recommended for magnetic metal marker in case delayed X-ray identification
resonance-guided sampling. of the biopsy site is required.
The negative predictive value of combined nor
mal mammography and ultrasound is extremely Number of samples
high; where there is a clinically palpable abnormal
ity and mammography and ultrasound are entirely A simple rule for satisfactory sampling using
normal, the likelihood of malignancy is low (<1%). needle techniques is to obtain sufficient material
However, in these circumstances it remains to achieve a diagnosis.59,60,62 For
prudent in the presence of a localised clinical ultrasound-guided core bi opsy, a diagnosis may
abnormality to carry out freehand needle biopsy be possible on a single core. Showing on
to exclude the occasional diffuse malignant ultrasound that the needle has passed through
process, such as classi cal invasive lobular the centre of the abnormality and by ex amining
carcinoma or low-grade DCIS, that may be occult the sample with the naked eye, it is usually, but
on both mammography and ultrasound. not always, possible to confirm whether a sat
For stereotactic procedures it is prudent to mark isfactory sample has been obtained. Because of
the biopsy site for future reference. Gel pellets or this and the knowledge that some lesions are
cel lulose combined with a metallic marker are heteroge neous, more extensive sampling of a
preferred and can be placed during the procedure lesion increases sensitivity. The number of core
to mark the biopsy site. These markers have the specimens obtained should reflect the nature of
advantage of be ing visible on ultrasound so that the abnormality being sampled. For
repeat biopsy or lo calisation for surgery can be ultrasound-guided biopsy where there is a
subsequently performed under ultrasound rather suspicion of carcinoma, it is recommended that
multiple core specimens are obtained.
breast lesions
As stereotactic biopsy is used for abnormalities
that are difficult to define on ultrasound and are
Wire-guided excision
therefore more difficult to sample, a minimum of
five core specimens should be obtained. Ensuring The number of impalpable, clinically occult breast
that calcification is present in at least three lesions detected by screening is increasing.
separate cores and/or five separate flecks of Accurate localisation techniques are required to
calcification are facilitate their surgical excision. The hooked wire
The role of imaging in breast diagnosis is the most commonly employed technique and
has proved very reliable but does have inherent
retrieved from the area of suspicion is essential to associated prob
en sure an accurate diagnosis. For calcifications lems. There are various designs of localisation
many prefer VAB, although some units perform wire in common use. All have some form of
14-gauge anchoring device such as a hook with a splayed
core biopsy as their initial biopsy technique. When or barbed tip. The wire is deployed under
there is diagnostic uncertainty after core, VAB can stereotactic or ultra
be used to obtain larger tissue volumes (a sound guidance within a rigid over-sheath
7-gauge VAB probe will retrieve approximately cannula, which is then removed once positioning
300mg per core). Ten- to seven-gauge is satisfac tory (Fig. 1.5). The patient is then
mammotomy probes can be used and are transferred to the operating theatre with the wire
preferred for therapeutic removal of breast lesions in situ. Most wires are very flexible and when the
such as fibroadenomas. cannula is removed the wire may assume a quite
circuitous course, es pecially after stereotactic
Biopsy results insertion when the breast is released from
compression. In a very fatty breast in which there
It is important that the results of FNA and/or core is no solid lesion or the wire has not transfixed the
needle breast biopsy are always correlated with lesion, care must be taken to avoid displacing the
the clinical and imaging findings before clinical wire. A cosmetically considered in cision is placed
management is discussed with the patient. This is near to the tip of the wire and an excision
best achieved by reviewing each case prior to any performed. Accurate wire placement is essential
surgical or other therapeutic procedure at multidis and ideally the shortest possible length of
ciplinary meetings.50
13
Surgery for clinically occult
Chapter 1
a
b Figure 1.5 • Digital mammogram showing a Reidy

localisation wire placed under ultrasound guidance transfixing and marking an impalpable mass to facilitate
surgical wide local excision (as shown in Fig. 1.4).
a
b Figure 1.6 • (a) Ultrasound showing a cluster of gel
pellets placed at the site of a previous stereotactic biopsy. The clear visibility of the pellets facilitates ultrasound
localisation for surgery of abnormalities that would normally require X-ray localisation. (b) Mammogram after
ultrasound localisation in the same case showing accurate placement of the marker wire.
planning is necessary. Inserting more than one
wire should be within the breast. In recent practice wire and bracketing the lesion with three or four
this has been greatly facilitated by the use of radio wires can be useful.
opaque and ultrasound visible markers placed at If the procedure is being performed to establish a
the time of initial stereotactic biopsy such that diagnosis, a representative portion of the lesion is
wire lo calisation can be performed under excised through a small incision, so leaving a sat
ultrasound guid ance (Fig. 1.6). In addition, for isfactory cosmetic result if the lesion proves to be
superficial lesions a skin marker may be more benign (the European surgical quality assurance
appropriate. guidelines require such diagnostic surgical
Although lesions may be clinically occult prior to excision specimens to weigh less than 30g). For
surgery, most mass lesions will be palpable intra diagnostic excisions of very small lesions, a
operatively. Procedures that can be surgically therapeutic wide excision may (after discussion
more challenging are wide local excisions for with the patient) be considered appropriate, as
DCIS with no mass lesion. In such cases, where the resulting cosmetic ef
the distribution of disease is often more eccentric,
careful excision
14

Not only can excision be guided but the margins
of a wide local excision specimen can also be
fect of removing an extra rim of normal assessed intraoperatively using ultrasound.
surrounding tissue may be insignificant. Protocols
vary for thera peutic excisions, but in general the
lesion should be excised with a 10-mm Radioisotope occult lesion
macroscopic margin of nor mal tissue. localisation
Intraoperative specimen radiography is essential,
both to check that the lesion has been re moved Radioisotope occult lesion localisation (ROLL) has
and, if cancer has been diagnosed, to ensure that been advocated as an alternative to the
adequate wide local excision has been achieved. hooked-wire technique.63 ROLL was first
Some surgeons experienced in this imaging tech described by the Milan
nique have also used intraoperative ultrasound. The role of imaging in breast diagnosis
migrate from the injection site) or low-molecular
group using 99mTc-labelled human macroaggregate weight colloid (which does migrate and is normally
albumin, using scintigraphy and a hand-held used for sentinel node biopsy). In both situations
gamma probe to guide surgical excision. The it is radiolabelled with 99mTc and injected directly
Nottingham method has modified the Milan into the lesion. The threshold of the signal proces
technique and uses radio-opaque contrast sor on the gamma detector is then adjusted so
injected with the radiolabel and immediate check that an audible signal is heard only when the
mammography (Fig. 1.7). Subsequently some probe is directly over the lesion. The probe then
centres have combined ROLL with sentinel node directs exci sion intraoperatively.
biopsy. ROLL uses essentially the same In a randomised trial of ROLL versus wire
equipment as sentinel node biopsy. It has been localisation, 2% of ROLL patients had a failed
described using macroaggregate (which does not
a
b Figure 1.7 • (a) Mammogram of the right breast showing
a small impalpable cancer. (b) Mammogram after injection of radionuclide mixed with X-ray contrast confirming
satisfactory localisation (ROLL).
15
Chapter 1 margins may be significantly easier with ROLL. In
reality there is little to choose between ROLL and
technique due to intraductal injection of wire localisation. ROLL may be a more suitable
radiolabelled colloid and dye that gave a tech nique in the localisation of non-mass lesions
ductogram appearance on check mammography (e.g. DCIS), although even for these lesions when
in both cases. 64,65
As the radio opaque dye is multiple wires are placed any advantage is small.
absorbed rapidly, both cases were suc cessfully There are various methods described for
converted to wire localisation. The main combining ROLL with sentinel node biopsy. 66–68
differences between ROLL and wire guidance Low-molecular weight colloid can be injected at a
were that both surgeons and radiologists found different site, at the same site with a different
ROLL eas ier to perform overall and patients found radiolabel, or into the tumour. With intratumoral
ROLL less painful. There has been no significant injection of 99mTc nano colloid, only one injection is
difference in accuracy of marking, operating time, required and high success rates have been
mean specimen weight, intraoperative re-excision reported. Combined with radio opaque contrast,
or second thera peutic operation in the majority of this modification of the Nottingham method has
reports, although a recent European trial reported proved simple and successful.
greater volumes of tissue were excised using
ROLL than standard wire lo calisation. Some
studies have suggested that obtaining clear
detected cancer, there should be a strong
emphasis towards the aesthetic outcome. Women
should be offered the full range of oncoplastic
Key points procedures available to achieve a good outcome,
including where appropriate the quality-of-life
benefits from breast reduction. In addition,
Oncoplastic considerations consideration should be given to whether such
procedures which reduce re-operation rates may
for screen-detected lesions be desirable to achieve an excellent cosmetic
outcome. The latter consider ation may be
Oncoplastic surgery aims to provide optimum ef
particularly relevant for DCIS, for which over 30%
fectiveness of surgical treatment for breast cancer
require further surgery follow ing an initial attempt
with minimum effect on quality of life. As the 10-
at breast-conserving surgery that fails because of
year survival of screen-detected disease is
involved excision margins. For women with larger
estimated at 87%, women do live a long time with
areas of DCIS, therapeutic mam moplasty may be
the effects of breast cancer surgery on body
the only option for achieving com plete excision
image, quality of life and self-esteem. The degree
and a satisfactory cosmetic outcome. Such
to which these outcomes are affected is strongly
procedures should be available to all women.
related to the cos metic outcome of surgery.
When assessing a woman for surgery for screen
• Breast imaging is an essential part of modern multidisciplinary breast

diagnosis. • Mammography is the technique of choice for population breast
screening.
• Screening is targeted at women aged 50–70 years and can be expected to reduce
mortality through early detection by approximately 25%.
• The aim should be to achieve as near as possible 100% non-operative diagnosis of
breast problems.
• Both palpable and impalpable breast lesions are best sampled under image
guidance. • Automated core biopsy is the sampling technique of first choice.
• Ultrasound is the guidance technique of first choice.
• Digital stereotactic core biopsy should be reserved for sampling lesions not visible on
ultrasound. • A 14-gauge core biopsy can provide a definitive diagnosis in more than 90% of
cases and should be the preferred method.
• Mammotomy can provide the diagnosis in most of the remainder.
• Stereo-guided vacuum-assisted biopsy is particularly effective for small clusters of
indeterminate microcalcifications and calcifications in sites difficult to access with core
biopsy.
16
• VAB is an effective and well-tolerated sampling technique for breast diagnosis and can also
be used to completely excise benign lesions.
• All breast needle biopsy results should be discussed at prospective multidisciplinary
meetings where the pathology results are correlated with the clinical and imaging findings. •
Accurate image-guided localisation and skills in wide local excision are required for the surgical
treatment of impalpable breast lesions.
• The aim of surgery is to excise the cancer completely and produce an excellent
cosmetic outcome.
• Oncoplastic surgery and therapeutic mammoplasty in appropriately selected women increases
the rate of excellent cosmetic outcomes.
0403. A cost comparison of full-field digital mam
mography with film-screen mammography in
References breast cancer screening. Sheffield: NHS Breast
Screening Programme Publications; 2004.
11. Gur D, Sumkin JH, Rockette HE, et al. Changes in
1. Office of National Statistics. Available from: http:// breast cancer detection and mammography recall
www.ons.gov.uk/ons/rel/cancer-unit/cancer-inciden rates after the introduction of a computer-aided de
ce tection system. J Natl Cancer Inst 2004;96:185–90.
and-mortality/2008-2010/stb-cancer-incidence-and
12. Wilson ARM, Teh W. Mini symposium: Imaging of
mortality-in-the-united-kindom--2008-2010.html;
the breast. Ultrasound of the breast. Imaging
2012.
1998;9:169–85.
2. Blanks RG, Moss SM, McGahan CE, et al. Effects
13. Lister D, Evans AJ, Burrell HC, et al. The ac curacy
of NHS breast screening programme on mortality
from breast cancer in England and Wales, 1990–8: of breast ultrasound in the evaluation of clinically
comparison of observed with predicted mortality. benign discrete breast lumps. Clin Radiol
Br Med J 2000;321:1724–31. 1998;53:490–2.
3. National Institute for Health and Clinical 14. Berg WA, Blume JD, Cormack JD, et al. ACRIN
Excellence. NICE Clinical Guideline 80: Early and 6666 Investigators. Combined screening with ultra
locally ad vanced breast cancer. 2009. Available sound and mammography vs mammography alone
from http:// in women at elevated risk of breast cancer. JAMA
guidance.nice.org.uk/CG80/NICEGuidance/pdf/ 2008;299:2151–63.
English;[accessed 23.07.12]. 15. Berg WA, Zhang Z, Lehrer D, et al. Detection of
breast cancer with addition of annual screening
4. Willett AM, Michell MJ, Lee MJR. Available from
ultrasound or a single screening MRI to mammog
Best Practice Guideline for Patients Presenting
raphy in women with elevated breast cancer risk.
with Breast Symptoms. 2010. [accessed 23.07.12]
JAMA 2012;307:1394–404.
http://www.associationofbreastsurgery.org.uk/
media/4585/best_practice_diagnostic_guidelines_ 16. Damera A, Evans AJ, Cornford EJ, et al. Diagnosis
for_patients_presenting_with_breast_symptoms.pdf of axillary nodal metastases by ultrasound guided
5. Kolb TM, Lichy J, Newhouse JH. Comparison of core biopsy in primary operable breast cancer. Br
J Cancer 2003;89:1310–3.
the performance of screening mammography,
physical examination, and breast US and evaluation 17. Houssami N, Ciatto S, Turner R, et al. Preoperative
of factors that influence them: an analysis of 27,825 ultrasound-needle biopsy of axillary nodes in invasive
breast cancer: meta-analysis of its accuracy and util ity
patient evaluations. Radiology 2002;225:165–75. 6.
in staging the axilla. Ann Surg 2011;254:243–51.
Skaane P, Young K, Skjennald A. Comparison of film
screen mammography and full-field mammography 18. Kuhl C. The current status of breast MR imaging.
with soft-copy reading in a population-based Part2. Clinical applications. Radiology
screening program: the Oslo II study. Radiology 2007;244:672–91. 19. Sardanelli F, Boetes C, Borisch
2002;225:267. 7. Pisano ED, Gatsonsis C, Hendrick B, et al. Magnetic resonance imaging of the breast:
E, et al. Diagnostic performance of digital versus film recommendations of the EUSOMA working group. Eur
mam mography for breast cancer screening. N Engl J J Cancer 2010;46:1296–316.
Med 2005;353:1773–83. 20. Kuhl CK, Schrading S, Bieling HB, et al. MRI for di
agnosis of pure ductal carcinoma in situ: a
8. James JJ. The current status of digital mammogra
prospec tive observational study. Lancet
phy. Clin Radiol 2004;59:1–10.
2007;370:485–92.
9. Committee on Technologies for the Early Detection
of Breast Cancer. Mammography and beyond:
developing technologies for the early detection of
breast cancer. Washington, DC: National Academy 17
Press; 2001.
10. Legood R, Gray A. NHSBSP Equipment Report
Chapter 1 2002;38:1458–64.
21. Kuhl CK, Schmutzler RK, Leutner CC, et al.

Mammography, breast ultrasound and magnetic
resonance imaging for surveillance of women at 18
high familial risk for breast cancer. J Clin Oncol
2005;23:8469–76.
22. Warner E, Plewes DB, Shumak RS, et al. 34. Million Women Study Collaborators. Breast cancer
Comparison of breast magnetic resonance and hormone replacement therapy in the Million
imaging, mammogra phy, and ultrasound for Women Study. Lancet 2003;362:419–27.
surveillance of women at high risk for hereditary Report of significantly increased risk of breast cancer in
breast cancer. J Clin Oncol 2001;19:3524–31. women taking HRT in the UK.
23. Stoutjesdijk MJ, Boetes C, Jager GJ, et al. 35. Perrson I, Thurfjell E, Holmberg I. Effect of
Magnetic resonance imaging and mammography estrogen and estrogen–progestin replacement
in women with a hereditary risk of breast cancer. J regimes on mam mographic breast parenchymal
Natl Cancer Inst 2001;93:1095–102. density. J Clin Oncol 1997;15:3201–7.
24. Brekelmans CTM, Seynaeve C, Bartels CCM, et 36. Sendag F, Cosan Terek M, Ozsener S, et al.
al. Effectiveness of breast cancer surveil lance in mammo graphic density changes during different
BRCA1/2 gene mutation carriers and women with postmeno pausal hormone replacement therapies.
high familial risk. J Clin Oncol 2001;19:924–30. Fertil Steril 2001;76:445–50.
25. Olsen O, Gotzsche PC. Cochrane review on screen 37. Evans A. Hormone replacement therapy and mam
ing for breast cancer with mammography. Lancet mographic screening. Clin Radiol 2002;57:563–4. 38.
2001;358:1340–2. Litherland JC, Stallard S, Hole D, et al. The ef fect of
26. Olsen O, Gotzsche PC. Systematic review of hormone replacement therapy on the sen sitivity of
screening for breast cancer with mammography. screening mammograms. Clin Radiol 1999;54:285–8.
Available at
39. Kavanagh AM, Mitchell H, Giles GG. Hormone re
http://image.thelancet.com/extras/fullreport.pdf
placement therapy and accuracy of mammographic
27. WHO handbook on cancer prevention. 7th ed screening. Lancet 2000;355:270–4.
Lyons: IARC Press; 2002. 40. Litherland JC, Evans AJ, Wilson ARM. The effect of
A comprehensive review of all the available data on the hormone replacement therapy on recall rate in the
effectiveness of breast cancer screening in reducing National Health Breast Screening Programme. Clin
breast cancer mortality. Radiol 1997;52:276–9.
28. Nystrom L, Andersson I, Bjurstam N, et al. Long 41. Evans AJ, Pinder SE, Ellis IO, et al. Screen
term effects of mammographic screening: update detected ductal carcinoma in situ (DCIS):
overview of the Swedish randomised trials. Lancet over-diagnosis or obligate precursor of invasive
2002;359:909–19. disease? J Med Screen 2001;8:149–51.
Long-term follow-up of the combined Swedish trials
42. Evans AJ, Burrell HE, Pinder SE, et al. Detecting
showing significant mortality benefit after more than 20
which invasive cancers at mammographic
years.
screening saves lives? J Med Screen
29. Tabar L, Vitak B, Tony HH, et al. Beyond ran 2001;8:86–90.
domised controlled trials: organised
43. National Institute for Clinical Excellence. Guidance
mammographic screening substantially reduces
on cancer services. Familial breast cancer.
breast carcinoma mortality. Cancer
London: National Institute for Clinical Excellence;
2001;91:1724–31.
2006.Available from http://www.nice.
30. Duffy S, Tabar L, Chen HH, et al. The impact of org.uk;[accessed 30.10.12].
organised mammographic screening on breast car
44. Robson M. Breast cancer surveillance in women
cinoma mortality in seven Swedish counties.
with hereditary risk due to BRCA1 or BRCA2 muta
Cancer 2002;95:458–69.
tions. Clin Breast Cancer 2004;5:260–8.
31. Hackshaw AK, Paul EA. Breast self-examination
45. Tilanus-Linthorst M, Verhoog L, Obdeijn I-M, et al.
and death from breast cancer: a meta-analysis. Br
A BRCA1/2 mutation, high breast density and
J Cancer 2003;88:1047–53.
prominent pushing margins of a tumour indepen
32. Smith RA, Saslow D, Sawyer KA, et al. American dently contribute to a frequent false-negative mam
Cancer Society guidelines for breast can cer mography. Int J Cancer 2002;102:91–5.
screening: update 2003. CA Cancer J Clin 46. Warner E, Plewes DB, Hill KA, et al. Surveillance of
2003;53:141–69. BRCA1 and BRCA2 mutation carriers with
33. Breast Screening Frequency Trial Group. The fre magnetic resonance imaging, ultrasound,
quency of breast cancer screening: results from mammography, and clinical breast examination.
the UKCCCR randomized trial. Eur J Cancer JAMA 2004;292:1317–25.
47. Hamilton LJ, Evans AJ, Wilson ARM, et al. Breast 2004;59:895–902.
imaging findings in women with BRCA1- and 48. Kriege M, Brekelmans CTM, Boetes C, et al.
BRCA2-associated breast cancer. Clin Radiol Efficacy of MRI and mammography for breast

62. Fishman JE, Milikowski C, Ramsinghani R, et al.
US guided core-needle biopsy of the breast: how many
cancer screening in women with a familial or specimens are necessary? Radiology
genetic predisposition. N Engl J Med 2003;226:779–82.
2004;351:427–37. 49. Leach MO, Boggis CR, Dixon 63. Luini A, Zurrida S, Paganelli G, et al. Comparison
AK, et al. Screening with magnetic resonance imaging of radioguided excision with wire localisation of
and mammog raphy of a UK population at high familial occult breast lesions. Br J Surg 1999;86:522–5.
risk of breast cancer: a prospective multicentre cohort 64. Rampaul RS, Bagnall M, Burrell H, et al.
study (MARIBS). Lancet 2005;365:1769–78. Radioisotope for occult lesion localisation: results
50. Teh W, Wilson ARM. Definitive non-surgical breast from a prospective randomised trial of ROLL
diagnosis: the role of the radiologist. Clin Radiol versus wire guidance in occult lesions of the
1998;53:81–4. breast. Br J Surg 2004;91:1575–7.
51. Britton PD. Fine needle aspiration or core biopsy. 65. Rampaul RS, Burrell H, Macmillan RD, et al.
Breast 1999;8:1–4. Intraductal injection of the breast: a potential pitfall
of radioisotope occult lesion localisation. Br J
52. Britton PD, McCann J. Needle biopsy in the NHS Radiol 2002;76:425–6.
Breast Screening Programme 1996/7: how much
and how accurate? Breast 1999;8:5–11. 66. Patel A, Pain SJ, Britton P, et al. Radioguided
occult lesion localisation (ROLL) and sentinel node
53. Vargas HI, Agbunag RV, Khaikhali I. State of the art biopsy for impalpable invasive breast cancer. Eur J
of minimally invasive breast biopsy: principles and Surg Oncol 2004;30:918–23.
practice. Breast Cancer 2000;7:370–9.
67. Tanis PJ, Deurloo EE, Valdes Olmos RA. Single in
54. Heywang-Kobrunner SH, Schaumloffel U, Viehweg tralesional tracer dose for radio-guided excision of
P, et al. Minimally invasive stereotaxic vacuum core clinically occult breast cancer and sentinel node.
breast biopsy. Eur Radiol 1998;8:377–85. Ann Surg Oncol 2001;8:850–5.
55. Brem RF, Schoonjans JM, Sanow L, et al. 68. Gray RJ, Giuliano R, Dauway EL, et al.
Reliability of histologic diagnosis of breast cancer Radioguidance for nonpalpable primary lesions and
with stereotactic vacuum-assisted biopsy. Am Surg sentinel lymph node(s). Am J Surg
2001;67:388–92. 2001;182:404–6.
56. Parker SH, Klaus AJ, McWey PJ, et al.
Sonographically guided directional
vacuum-assisted breast biopsy using a handheld
device. AJR Am J Roentgenol 2001;177:405–8.
57. Kettritz U, Rotter K, Murauer M, et al. Stereotactic
vacuum biopsy in 2874 patients: a multicenter
study. Cancer 2004;100:245–51.
58. Brenner RJ, Bassett LW, Fajardo LL, et al.
Stereotactic core needle breast biopsy: a
multi-institutional pro spective trial. Radiology
2001;218:866–72.
59. Wilson ARM. Therapeutic applications of vacuum

biopsy. Breast Cancer Res 2008;4:31.
60. Wilson R, Kavia S. Comparison of large-core vac
uum-assisted breast biopsy and excision systems.
Recent Results Cancer Res 2009;173:23–41.
61. Kuhl CK, Morakkabati N, Leutner CC, et al. MR
imaging-guided large-core (14-gauge) needle
biopsy of small lesions visible at breast MR
imaging alone. Radiology 2001;220:31–9.
19
2
Pathology and biology of breast cancer
Rajendra S. Rampaul
Emad A. Rakha
John F.R. Robertson
Ian O. Ellis
Management of women with breast carcinoma has
undergone significant changes over the past 20
Introduction years. The pathology and biology of breast cancer
influ ences diagnosis, selection of primary and reproducible. A prognostic index is defined as
adjuvant treatment, formulation of follow-up quantitative set of values based on results of a
protocols, prog nosis and provision of counselling prognostic model. There are several reasons for
and reassurance. Screening and public education the use of such prognostic in
have accounted for a major shift in the number dices in breast cancer, which include the ability:
and type of the breast car cinomas detected today.
1. to separate patients into groups with signifi
Cancers are now of smaller size and more often
cantly differing survival probabilities;
lymph node negative. There is an increasing need
2. to separate patients into groups that include
to discriminate accurately the risk of recurrence
and select appropriate adjuvant systemic a ‘cured’ group and a group with poor survival;
therapies. Modern clinical practice employs a 3. to place a sufficient percentage of cases into
signifi cant input from histopathological data to each group;
assist in the decision-making process for 4. to be applicable for all operable breast
selecting treatments. This can be achieved by cancers – small, screen detected as well as
identifying accurate prog nostic and predictive symptomatic and young age;
factors. A prognostic factor is defined as any 5. to be prospectively validated;
patient or tumour characteristic that is predictive 6. to be capable of use in all units and to
of the patient's outcome. Outcome is usually be inexpensive.
measured in terms of cancer-specific survival or In this chapter common pathological features of
disease-free survival. A predictive factor is defined breast cancer are reviewed and their role in
as any patient or tumour characteristic that is guiding patients' management considered.
predic tive of the patient's response (outcome) to a
specified treatment. The factors currently
employed in breast cancer prognostication and Traditional factors
prediction each possess independent prognostic
information and predictive power and may be Lymph node stage
combined into an index, making
Involvement of local and regional lymph nodes by
metastatic carcinoma is one of the most important
20 prognostic factors in breast cancer. The revised
it more ‘user friendly’, informative and

0.6
TNM staging system for breast cancer has con

Pathology and biology of breast
cancer Lymph node t

i
lymph node stage is considered as a

stage time-dependent
firmed that the absolute number of axillary lymph

1.0
nodes involved in metastatic cancer (‘positive lymph
nodes’) is one of the most important prognostic fac
i
0.8 b
tors in breast cancer.1 Lymph node stage has been a
factor; the longer the tumour has been growing the

used consistently as a guide for therapy. However,
0.4
b
o Studies show that overall survival is

decreased as the number of nodes that are involved
r
more likely it is that lymph nodes are involved by

increases.5–7
P
spread. It has also been reported that, when taken

Optimal management of the axilla is discussed in
Chapter 9.
Sentinel node biopsy or blue dye directed

axillary sampling has been shown to provide
accurate prognostic information and has an
extremely low rate of lymphoedema.2 Of 1275
patients undergoing axillary sampling, only 0.04%
0.2 suffered symptomatic complaints of arm swelling
alone, lymph node stage is incapable of defining ei and lymphoedema.
0.0 0 2 50 100 150 200 50
ther a cured group or one with close to 100% mor Survival in months
Node negative Figure 2.1 • Overall survival for patients in the
1–3 nodes positive NTPBCS according to lymph node stage (P < 0.0001).
>3 nodes positive
A refinement of lymph node sampling is pro vided

by the technique of sentinel lymph node (SLN)
biopsy. There have been several studies to
examine the validity of this technique. The
ALMANAC trial provides level I evidence for UK
based practice.3
p<0.0001
From a pathological perspective, there are unre
tality from breast cancer. solved questions about how to optimally assess
The clinical assessment of nodal status (as in the sentinel node. Studies have shown that the
TNM classification) is unreliable.2,3 Palpable nodes mean
may be enlarged because they show benign number of sentinel nodes is close to 3. As the num
reactive changes whilst nodes bearing tumour de ber of nodes removed during SLN biopsy is fewer,
posits can be impalpable. Although axillary so pathologists are now challenged to process these
nography is a moderately sensitive and fairly few nodes optimally. Several methods have been
specific technique in the diagnosis of axillary met studied: routine paraffin histology; intraoperative
astatic involvement,4 histological examination of frozen section; immunohistochemistry and more
the lymph nodes from the axilla should be carried intensive methods such as serial sectioning and
out in all patients with primary operable invasive molecular methods including polymerase chain
breast cancer. It is well known that patients who reaction (PCR) and reverse transcription–PCR
have histologically confirmed lymph node in (RT–PCR).
volvement have a significantly poorer prognosis At present there is no consensus for the optimal
than those without nodal metastases. The 10-year handling of the SLN in the laboratory. Most cen
survival is reduced from 85% for patients with no tres have developed their own ‘in-house’
nodal involvement to 40% for those with involve protocols, which are invariably tied to the
ment of multiple nodes. Prognosis worsens the institution's research ambitions. This ad hoc
greater the number of nodes involved, and the approach can influence in terpretation of results,
level of nodal involvement can provide useful in as the amount of material as sessed between
formation; metastasis to the higher level nodes centres is not uniform. In a number of large
(level II or III) in the axilla and particularly those at studies, using different histopathological
the apex (level III) carries a worse prognosis (Fig. techniques, the SLN false-negative rate (defined
2.1). as how often the SLN is negative for malignancy
when cancer is present in the rest of the axilla)
varies be tween 0% and 11%.
21
In Nottingham the MRC ALMANAC protocol is
employed to assess all sentinel node biopsies.
Chapter 2 false-positive rate due to the sensitivity of the
method and potential for contamination. In
These nodes are cut into slices about 3 mm thick, addition, it is not possible with PCR to determine
taken perpendicular to the long axis of the node to whether the DNA comes from cells that are
maximise the assessment of the marginal si nus, viable. An advantage of haematoxylin/ eosin
with one node per cassette. The majority of nodes sections and immunohistochemistry is that the
can be completely embedded in one cas sette. morphology of the cells can be examined and ma
Larger nodes have alternate slices embed ded lignancy confirmed. A major unresolved question
and may require more than one cassette. Large, is
obviously involved nodes have one section taken.
Intraoperative frozen sections or imprint cytology
of axillary lymph nodes have been advocated by 22
some authors. Conventional frozen sections have
a high false-negative rate of between 10% and
30%. More intensive intraoperative assessment whether carcinoma detected only by PCR or RT–
with se PCR has the same prognostic significance as
rial sections and immunohistochemistry has been haema toxylin/eosin positivity.
described,8 but is time consuming and labour in The detection rate of micrometastases in axillary
tensive. Frozen section is best applied to selected lymph nodes has been reported to range from 9%
cases; for example, if the node is macroscopically to 46%.10 Studies have used serial sectioning with
abnormal and this is confirmed histologically to be or without immunohistochemical stains to detect
metastatic carcinoma, further axillary surgery can mi
be performed immediately. Some studies have crometastatic foci and these methods have
found low false-negative rates of 2–3% with intra increased detection rates.
operative imprint cytology,9 but not all have been The European Working Group for Breast
able to achieve this level of accuracy. Lymph node Screening Pathology11 has formulated working
status can be assessed peroperatively using ultra guidelines for the assessment and pathological
sound combined with FNA or core biopsy, and this work-up of SLNs in breast cancer. From a
reduces the need for preoperative frozen sec tion literature review available at that time, the
and imprint cytology to assess axillary nodes committee concluded that it was not possible to
intraoperatively.
determine the significance of micro
RT–PCR is more sensitive than immunohisto
metastasis or isolated tumour cells. They noted
chemistry at detecting metastatic tumour in axil that approximately 18% of cases are associated
lary nodes. Two types of methods have been used with other nodal (non-sentinel node) metastases.
to detect tumour cells with molecular techniques. False negative rates are most often determined
First, a genetic defect such as chromosomal rear via im munohistochemistry (IHC). However, at
rangement or mutation can be used. The problem present it is recommended that an intensive
with this is that no single genetic defect is seen in work-up is not justified on a population level. The
all breast carcinomas. The second method is to committee did suggest multilevel assessment
use a molecular marker that is present in tumour and, where resources permit, intraoperative
cells but not in the adjacent tissue. To identify a assessment, although the re sults of Z0011 (see
marker that has this specificity and is expressed Chapter 7) suggest that some node-positive
in the ma jority of tumours is difficult. It may be patients may not require axillary dis section and
necessary to use a panel of markers. A major this brings the whole issue of intraop erative
problem with PCR is the potentially high assessment into question. The current view is that
IHC should not be used routinely to assess disease-free survival rate of greater than 90%.14
axillary lymph nodes. Tumour size is a time-dependent prognostic factor
that depends on the period between tumour de
velopment and detection, and on the balance be
Tumour size tween tumour cell proliferation and death (tumour
growth rate). It is well known that the association
Tumour size is one of the most powerful predic between increasing tumour size and increasing
tors of prognostic factor in breast cancer.12,13 The number of positive lymph nodes and worse out
frequency of nodal metastases in patients with come is highly statistically significant.13 Therefore,
tumours <10 mm is 10–20%,13 and node-negative the main aim of population screening with mam
patients with tumours <10 mm have a 10-year mography is to detect smaller tumours that are
Pathology and biology of breast Survival by size

cancer 0–2 cm
0.1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
2.01–5 cm
0.0 168 180
Months
0 12 24 36 48 60 72 84 96 108 120 132 144 156
Figure 2.2 • Overall survival according to size.
Modern pathologists have recognised that invasive
carcinomas can be divided according to their
likely to have a better outcome than those that
degree of differentiation. There are two ways to
present symptomatically (of a larger size). Patients
achieve this: (i) by allocating a histological type
with smaller tumours have a better long-term sur
according to the architectural pattern of the
vival than those with larger tumours (Fig. 2.2).
tumour; (ii) by assigning a grade of differentiation
Estimation of tumour size has assumed particular
based on semi
importance since the introduction of population
quantitative evaluation of structural characteristics.
screening. In most studies the frequency of axillary
lymph node metastasis in small <15 mm (so-called Certain histological types of invasive carcinoma
minimally invasive carcinoma (MIC)) invasive carry a favourable prognosis. Tubular, mucinous,
invasive cribriform, medullary and tubulolobu lar
carcinomas is 15–20%,15–17 compared with over
types, together with rare tumour types such as
40% in tumours measuring 15 mm or more.18
adenoid cystic carcinoma, adenomyoepithelioma
During the prevalent round of breast screening
and low-grade and squamous carcinoma, have all
even more favourable results are obtained, with
been reported to have a more favourable outcome
the frequency of axillary lymph node metastasis
than invasive carcinoma of no special type (ductal
ranging from 0% to 15%.19–22 The Nottingham
NST). Undoubtedly, assessment of histological
Tenovus Primary Breast Cancer Study (NTPBCS) type provides prognostic information in breast
has generated data that suggest the cut-off point cancer. However, this effect is relatively small in
of 10 mm is not the best discriminator for MIC.
multivari ate analysis27 when compared with the
Life-table analysis of survival curves found no dif
prognostic value of histological grade; histological
ference between tumours measuring up to 9 mm
type may prove to be more useful in increasing
and those measuring 10–14 mm. This indicates our under standing of the biology of breast
that <15 mm may be a more realistic watershed in
cancers.28 Invasive lobular carcinoma (ILC) is of
defining small, invasive, carcinomas of good
particular importance. It comprises approximately
prognosis. It is clear that pathological tumour size
5–15% of breast cancers and appears to have a
is a valuable prognostic factor and it has become
distinct biology. It is less common than invasive
an important quality assurance measure for breast
carcinoma of no spe cial type, also called invasive
screening.21,23–26 It is also used in part to judge the
ductal carcinoma (IDC).
ability of radiologists to detect small impalpable
invasive carcinomas.
23
Differentiation
Chapter 2 progressive behavioural characteristics with nearly
linear survival curves that crossed those of IDC after
approximately 10 years of follow-up, thus eventually
exhibiting a worse long-term outcome. Interestingly,
Recently, Rakha et al.29 examined a large
ILC showed a better response to adjuvant hormonal
group of 5680 breast tumours (415 patients (8%)
therapy (HT), with improvement in survival in
with pure ILC and 2901 (55.7%) with IDC (no special
patients who received HT compared with matched
type)) and demonstrated that, compared to IDC,
patients with IDC.
patients with ILC tended to be older and have
tumours that were more frequently of lower grade
(typically grade 2; 84%), hormone-receptor positive
(86% compared to 61% in IDC), of larger size and
with absence of vascular invasion. More patients Tumour grade
with ILC compared with IDC were placed in the
good Nottingham Prognostic Index group (40%
compared to 21% in IDC). ILC showed indolent but The Nottingham method (Elston and Ellis) is a
modification and improvement of previously exist
• Grade 1 – well differentiated = score 3–5
ing morphological grading systems and is able to
points. • Grade 2 – moderately differentiated =
provide greater objectivity of grade.30
score 6–7 points.
• Grade 3 – poorly differentiated = score
8–9 points.
It should be noted that grade is valuable

1.0
irrespective of histological type.
0.9 There is a highly significant correlation between
tu mour grade with long-term prognosis (Fig. 2.3);
0.8 pa tients with grade 1 tumours have a 93% chance
0.7 of surviving 10 years after diagnosis, whereas in
patients with grade 3 tumours this is reduced to
70%. It has now been shown conclusively that the
Nottingham method, with its more objective
In brief, assessment of grade considers three
criteria, has ex cellent reproducibility when used
tumour characteristics: tubule formation, nuclear
by experienced pathologists.
pleomor phism and mitotic counts. A numerical
Interestingly, grade is not included in the recent
scoring sys tem of 1–3 is used for each factor
individually. The three scores are added together revision of the TNM staging system of breast
to produce scores of 3–9 on which an overall cancer as its value is questioned in certain
tumour grade is assigned: settings such as lobular type.
0.2
1
2
0.1
l
0.6
0.5
3
0.4 0.0
0 12 24 36 48 60 72 84 96 108 120
0.3 132 144 156 168 180 Months
Grade
Figure 2.3 • Overall survival according to grade.
24
examina
tion. In patients whose axillary nodes are tumour
free on histological examination there is a correla
The value of grade was addressed by Rakha et al.31
in a large study of grade and overall survival in a tion between the presence of lymphovascular inva
large and well-characterised consecutive series of sion (LVI) and early recurrence.
operable breast cancer (n = 2219 cases), with a
long-term follow-up (median 111 months) using the
Recently, Lee et al.37 assessed the prognostic
Nottingham histological grading system. Histological
value of LVI in a group of 2760 patients with node
grade was strongly associated with both breast
negative breast cancer with long-term follow-up
cancer specific survival (BCSS) and disease-free
(median 13 years). This study demonstrated a strong
survival (DFS) in the whole series as well as in
association between poor histological grade and
different subgroups based on tumour size (pT1a,
younger age with LVI-positive cancers. LVI was
pT1b, pT1c and pT2) and lymph node (LN) stages
prognostically significant and was independent of
(pN0 and pN1 and pN2). The authors were able to
grade, size and type for overall survival (see Fig.
demonstrate differences in survival between
2.4).
different individual grades (1, 2 and 3). In
multivariate analyses histological grade was an
independent predictor of both BCSS and DFS.
Molecular/predictive factors
Additionally, the usefulness of routinely Despite the overall association between molecular
assessing grade in invasive lobular cancer was markers with prognosis and outcome, they are lim
examined by this group in 4987 patients,32 of whom ited in their ability to capture the nuances of the
517 were pure ILC cases. The majority of ILC was complex cascade of events that drive the clinical
of classical type or mixed lobular variants (89%).
be haviour of breast cancer. Morphological factors
Most ILC cases were moderately differentiated
(grade 2) tumours (76%), while a small proportion of are unable to predict response to systemic
tumours were either grade 1 or 3 tumours (12% treatments, and tumours of apparently
each). There were significant associations between homogeneous morpho logical characteristics still
histological grade and other clinicopathological vary in response to ther apy and have distinct
variables of poor prognosis such as larger tumour
size, positive lymph node status, vascular invasion, outcomes. The use of strategies ranging from
oestrogen receptor and androgen receptor hormone therapy to chemotherapy and recently to
negativity, and p53 positivity. Multivariate analyses novel receptor-directed therapies and vaccines
showed that histological grade was an independent relies on the expression of predictive factors by
predictor of BCSS and disease-free interval.
the cancer, either individually (e.g. oestro gen
receptors (ERs), progesterone receptors (PRs),
Vascular invasion human epidermal growth factor receptor (HER-2)
The presence of tumour emboli in vascular and

lym phatic spaces has emerged as an important It has been suggested that it can provide prognos
1.0
prognos tic factor. Several studies have now
tic information as powerful as lymph node stage.
shown that the presence of vascular invasion Reproducibility is the limiting factor in its wide 0.8
correlates closely with local and regional lymph
spread adoption and routine clinical assessment. In
node involvement.33,34 a Nottingham study of assessing vascular invasion,
Pathology and biology of breast cancer the
can be obtained as long as strict criteria are used.

Lymphatic and vascular invasion is considered to
be a valuable surrogate for lymph node stage in
cases where nodes have not been removed for y
t
i
l
issue of reproducibility was addressed specifically. 0.4

all inter-observer agreement on histological features
and an 85.8% overall agreement in the classification
0.2
of vascular invasion. Other studies have reported a
Vascular invasion
i Negative
0.6 Positive
b
Of 1704 cases, a subset of 400 cases was

examined
p<0.0001
by two or more pathologists. There was a 77% over
similarly high concurrence between pathologists.35,36 The assessment of vascular invasion is

subjective but there is good evidence that a high rate of concurrence
0 50 100 150 200 250
Survival in months
Figure 2.4 • LVI and breast cancer-specific survival. 25

Chapter 2
Knowledge of the expression of hormone
and epidermal growth factor receptor (EGFR)) or receptors and their subcellular regulatory
globally (e.g. high-throughput gene expression pathways is one of the classic examples of the
assays). These molecular markers are used not use of translational medi cine in breast cancer to
only to guide treatments but are also being used further diagnosis and treat ment of this disease.
to moni The degree of ER expression is used to predict
tor response and detect relapse. It is envisaged an individual's response to hormone therapy. Both
that molecular predictive markers may form the ERs and PRs are steroid receptors that are
basis of tailoring an individual's adjuvant therapy located in the cell nucleus. Oestrogen and proges
based on the genetic fingerprint of their cancer. terone are considered to diffuse into cells, or be
trans ported to the nucleus. Genes, regulated by
steroid receptors, are involved in controlling cell
Oestrogen receptors (ERs)/ growth, and it is currently believed that these
progesterone receptors effects are the most relevant to ERs and that ER
(PRs) expression influ ences the behaviour and
treatment of breast cancer.38 Elucidation of staining and multiplies the two. The Allred score
downstream effect on genes that are known to be categorises the percentage of cells (scored from 0
influenced by hormones has led to the inclusion of to 5) with the intensity (scored from 0 to 3) and
these ER-regulated genes in high-density adds these two scores to give a numerical score
oligonucleotide array panels.39 This may better de from 0 to 8. Others simply estimate the
fine those pathways that are endocrine responsive percentage of positively stained tumour nuclei. It
and may lead to improved therapies with reduced is important to appreciate that all methods are
side-effects and a greater potential for cure.40 In subjective and, at best, semiquantita tive. ER/PR
unselected patients, approximately 30% will re status in the histopathological report has a clearly
spond to endocrine therapy. A tumour with both defined role as a predictive factor for the
ER and PR expression has a 78% chance of response of systemic endocrine therapy. ERs and
respond ing to hormone therapy while those PRs should be assessed on all breast tumour
cancers that are ER/PR negative respond rarely, if specimens. As these assays have become
ever.41–43 Due to their close relationship with simpler, and less costly, they are now available for
histological grade, ERs are not of independent all patients. Progress still needs to be made in
prognostic significance.43 standardisation of assay technique, objectivity
and reproducibility.
Methods of measuring ERs and PRs
There are a multitude of methods available; how Type 1 growth factor receptors
ever, in clinical practice most are based on two dis
tinct strategies. The first is an older method based Activation of growth factor receptors was first
on ligand-binding methods, i.e. radiolabelled found to be important in human cancers by iden
steroid ligand is used to detect the receptor. The tifying the homology between the viral oncogene
second relies on the recognition of the receptor v-erbB and EGFR. Sequence similarity between
protein by specific antibodies. Several studies the erbB-1 gene that encodes EGFR resulted in
examined the correlation between assay and the iso lation of a second growth factor receptor,
outcome/response and studies the hu man ortholog neu, HER-2 (erbB-2).
Screening of genomic DNA and messenger RNAs
with probes allowed isolation of two additional
26 relatives of the human ERB1 gene. They were
subsequently named erbB-3 (HER-3) and erbB-4
(HER-4).
have suggested that IHC analyses have a number To date there are four known receptors that
of advantages.42 IHC is currently the most belong to this growth factor receptor family,
commonly used method because it can be HER-1–4, and ligands for some of these have
performed on paraffin embedded material and been identified. The family works in a co-receptor
allows assessment of the ex pression specifically dimerisation type ac
in either invasive or in situ cancer. tivation pathway.
The first two members of this family have been
Interpretation of assays studied extensively in breast cancer. There are lim
The optimal way to score IHC for ERs and PRs re ited data on the role of HER-3 and HER-4 as well
mains controversial. One method is to use an ‘H’ as their ligands. This family of signalling molecules
score (histo score), which takes into account the is currently of immense clinical interest as the first
fre quency of positive cells as well as intensity of

assessed. Findings from this review
demonstrated a great het erogeneity of study
two members are targets against which novel bio design, levels of cut-offs for posi tive and negative,
receptor agents have been developed, including and, not surprisingly, differences in its prognostic
Iressa®, Tarveca® and Herceptin®. value in these various studies.
EGFR
EGFR expression has been well studied in breast Recently, Tsutsui et al.47 reported on a large series
cancer. Consequent on major differences in study of 1029 patients (with adjuvant therapy) and found
designs, the EGFR assay used and confounding EGFR to be of independent prognostic value
factors such as adjuvant therapy, there is no irrespective of nodal status. In contrast, studies by
Rampaul et al.48 and Ferrero et al.49 (without
consen
adjuvant therapy) showed EGFR to be of no
sus on its prognostic value. With the development prognostic value (these latter series also
of tyrosine kinase inhibitor (TKI)- and incorporated grade and had longer follow-up).
EGFR-directed biotherapies, there is now a
growing impetus to not only standardise assays The predictive value of this marker for hormone
for detecting EGFR but also to delineate resistance or responsiveness is better defined.
accurately its prognostic and predictive potential. EGFR tumours are considered more resistant to
endocrine therapy, and there exists an inverse
Methods of testing relationship with EGFR-negative cancers (they
Several techniques directed at DNA, RNA, protein are more likely ER positive), being more often
(and functionally active (activated) as well) or se sensitive to endo
rum can be employed to identify EGFR expression crine manipulation. Though some studies have
in breast cancer. been able to demonstrate that EGFR status is a
Of all such methods, IHC is the most practical. predic tor of tamoxifen failure and even response
The advantages of IHC in EGFR as in HER-2 test rates,50 there are conflicting data from
ing include reproducibility, ease of interpretation well-designed level II studies51 that have shown
and the low cost compared to other techniques. It no value of EGFR in predicting the efficacy of
can also be employed on archival tissue. Protein tamoxifen in high-risk postmenopausal women.
lev els can be quantified by Western analysis or Perhaps the most exciting use of EGFR status
enzyme immunoassay; however, architecture of and indeed the driving impetus for clinicians and
tissue is lost in these procedures and there may scien tist to measure it accurately is its putative
be contamination by normal tissue cells or ductal role in defining those who should respond to
carcinoma in situ. Serum EGFR levels have been novel EGFR directed therapies.
examined in breast carcinoma; however, few data
are available.44,45 Due to the paucity of relevant HER-2
data, it is difficult to draw conclusions on the HER-2 is an important target in the development
current value of measuring EGFR in the serum of of a variety of new cancer therapies, which include
patients. monoclonal antibody (mAb)-based therapy, small
molecule drugs directed at the internal tyrosine
Prognostic and predictive value kinase portion of the HER-2 oncoprotein, and vac
The prognostic value of EGFR in breast cancer cines. The most widely known HER-2-directed ther
has been examined in several large studies. apy is trastuzumab (Herceptin; Genentech, South
There remains no consensus on its prognostic San Francisco, CA, USA). Trastuzumab is a
value. The lack of a clear picture has been human ised recombinant mAb that specifically
attributed to several factors: lack of a standard targets the HER-2 extracellular domain. There are
assay (monoclonal vs. polyclonal), lack of a a variety of techniques available to determine
cut-off for positivity and great variation in study HER-2 status in breast cancer, some of which are
designs (size, follow-up and influence of ad employed for re search purposes only. In
juvant therapy). Klijn et al.46 reviewed data from diagnostic pathology labo ratories HER-2 status is
over 5000 patients where EGFR had been assessed routinely either by IHC, which assesses
expression of the HER-2 oncoprotein, and hybridisation (CISH) and silver-enhanced in situ hy
fluorescence in situ hybridisation (FISH), which bridisation (SISH).
measures the number of HER2 gene copies per
chromosome 17 or gene amplification.
Modifications of ISH using colorimetric detection
27
are being developed, including chromogenic in situ
Chapter 2 College of American Pathologists (CAP) 52
specified a threshold of >30% strong circumferen
Methods of HER-2 testing tial membrane staining for a positive result. If both
uniformity and a homogeneous, dark circumferen
Current guidelines for HER-2 testing52 specify the
tial pattern are seen, resultant cases are likely to
methods that are suitable to detect either the
be amplified by FISH as well as positive for
HER-2 protein (by IHC) or gene amplification
HER-2 pro tein expression. The equivocal range
(using FISH or other in situ methods). Guidelines
for IHC (score 2+), which may include up to 23%
stress the need for stringent, reproducible and
of samples, is defined as complete membrane
consistent criteria for testing.
staining that is either non-uniform or weak in
Immunohistochemistry for HER-2 testing Among intensity but with obvious circumferential
the methods in use for determining HER-2 status, distribution in at least 10% of cells.
IHC is the most widely used. In studies em ploying
various commercially available antibod ies, a wide
variety of sensitivity and specificity in fixed 28
paraffin-embedded tissues is seen.53,54 Antigen
retrieval techniques are currently not standardised
and they introduce the potential for false-positive Equivocal or inconclusive results should be tested
staining. Nonetheless, IHC possesses many by FISH. Consistent with previous guidelines, a
advan tages to support its widespread adoption: (i) nega tive HER-2 test is defined as either an IHC
it al lows for the preservation of tissue architecture result of 0 or 1+ for cellular membrane protein
and so can be used to identify local areas of expression (no staining or weak, incomplete
overexpres sion within a heterogeneous sample, membrane staining in any proportion of tumour
and can dis tinguish between HER-2 positivity in cells), though up to 5% of 1+ on IHC may be FISH
in-situ and invasive cancer; (ii) it is applicable to positive.
routine patient samples, facilitating use as a
diagnostic test, and this allows prospective and In situ hybridisation for HER-2 testing
retrospective research stud ies of HER-2 status to FISH and CISH measure directly the number of
be undertaken. HER2 genes per chromosome 17, and when there
Two Food and Drug Administration (FDA)- is a chromosome centromeric enumeration probe
approved IHC tests for determining HER-2 status (CEP) included, the copy number of chromosome
are available: HercepTest (DAKO, Carpeteria, CA, 17 gene amplification is defined as an increase in
USA), based upon a polyclonal antibody; and HER-2/CEP17 ratio above 2.0. ISH results are
CB11 (Pathway, Ventana Medical Systems, semiquantitative, counting the number of signals
Tucson, AZ, USA), based upon a monoclonal in non-overlapping interphase nuclei of the lesion
antibody. The National Comprehensive Cancer using either single-colour (HER-2 probe only, e.g.
Network guide lines55 classify an IHC score of 0 or Ventana Inform) or dual-colour hybridisation (us
1+ as represent ing HER-2-negative status, 3+ as ing HER-2 and chromosome 17 centromere
positive, while 2+ is equivocal. Positive staining is probes simultaneously (e.g. Abbott, Chicago,
defined as strong, con tinuous membranous USA, DAKO, Copenhagen, Denmark, etc.), the
expression of HER-2 in at least 10% of tumour latter making it easier to distinguish true HER-2
cells. However, a joint report from the American amplification from chromosomal aneuploidy. ISH
Society of Clinical Oncology (ASCO) and the allows simultane
ous morphological assessment, where evaluation sidered amplified and both include an HER2 gene
of gene amplification can be restricted to invasive probe and a chromosome 17 probe. The INFORM
carcinoma cells. Many studies have compared test (Ventana Medical Systems) requires that at
FISH and IHC in the evaluation of HER-2 and least 5.0 gene copies of HER2 be present if a
have dem sample is to be considered amplified as this kit
onstrated concordance between the two uses a single HER2 gene probe without a
techniques of up to 91%. Two studies have shown chromosome 17 probe.
that FISH predicts more accurately HER-2 Recommended guidelines for HER-2
positivity than IHC when applied to molecularly assessment in the UK have recently been
characterised breast cancers.56,57 updated and the reader is referred to these
Three FISH tests are FDA approved for selecting guidelines for further details.58
patients for treatment with trastuzumab. The Path In brief, a two-phase testing algorithm based on
Vysion (Vysis Inc., Downers Grove, IL, USA) and IHC assay as the primary screen with FISH being
PharmDx (Dako) tests require a ratio (HER-2 to re served for equivocal cases is currently
CEP17) of 2.2 or greater for the sample to be con recommended.

recent stud ies have, however, shown a
consistently poorer prognosis when comparing
This is based on evidence showing very good HER-2-positive versus
concordance between IHC and FISH results on -negative small node-negative breast cancers.
breast carcinomas from 37 laboratories when Rilke et al.61 reported on the prognostic signifi
tested in experienced reference centres.59 cance of HER-2 expression and its relationship
with other prognostic factors. Using specimens
Chromogenic in situ hybridisation (CISH/SISH) from 1210 consecutive patients treated between
CISH and SISH are colorimetric methods to de 1968 and 1971 at a single institution (National
tect gene amplification that can be viewed using a Cancer Institute of Milan), with no systemic
standard light microscope. Concordance between adjuvant therapy and 20-year follow-up, they
FISH and SISH, for the validation of HER2 gene found overexpression of HER-2 in 23% and
status, is about 96% (kappa = 0.754). showed a negative impact on survival of
node-positive but not node-negative pa tients.
Prognostic significance and association with Analysis of HER-2 in relation to the presence of
other prognostic factors lymphoplasmacytic infiltrate (LPI; favourable
60
The seminal work by Slamon et al. in 1987 prognosis) and nodal status demonstrated that in
showed that HER2 gene amplification indepen node-negative, LPI-negative patients, HER-2 over
dently predicted overall survival (OS) and disease expression showed the same level of correlation
free survival (DFS) in a multivariate analysis in with poor prognosis as with those patients with
node-positive patients. Since then most large stud nodal metastasis. However, in the patients with
ies have confirmed this relationship in multivariate node negative disease and LPI positivity, HER-2
analysis. It is now well established that there is a over expression correlated with good prognosis.
significant correlation between HER-2 overexpres Some
sion/amplification and poor prognosis for patients Pathology and biology of breast cancer
with nodal metastasis. There was until recently no
consensus on the prognostic value of HER-2 in studies have reported a prognostic value for
node-negative breast cancer patients, a group HER-2 in node-negative patients in selected
most often diagnosed through screening and subgroups,62–64 whereas others have shown no
represent ing a subgroup that could potentially correlation.65,66
benefit from appropriate adjuvant therapy. More Mirza et al.66 have published a systematic review
of prognostic factors in node-negative disease. mours.81,82 The 20-year update of the Naples GUN
Data for HER-2 showed a lack of standardisation Trial81 found that HER-2 overexpression not only
of as says and no association with survival. predicted resistance to tamoxifen, but that HER-2-
However, HER-2 status was shown to be of positive patients had a worse outcome on
independent prognostic significance in several tamoxifen
large studies with long-term follow-up, with therapy compared with those who were untreated.
worstened prognosis for those node-negative Several studies have also shown a reduction in
patients with HER-2-positive signalling.67,68,69 response rates to hormonal therapy. Metastatic
There is at present no agreement on the associa breast cancer that overexpressed HER-2,
tion between HER-2 and other prognostic factors. measured by high plasma levels of extracellular
Several studies have shown a lack of association domain, dem onstrated a substantial reduction in
between HER-2 status and tumour size,60,70,71 al response rate to hormonal therapy.84 Other
though some do report a correlation.61,66,72–76 studies have failed to find an association or even
a trend between HER-2 status and response to
Prediction of response to therapy: hormonal hormonal therapy.85–87 Elledge et al.87 examined
therapy the response to tamoxifen in 205 tumours with
Transfection of normal breast cancer cells with the ER-positive disease. In HER 2-positive compared
HER2 gene has been shown to result in to HER-2-negative patients, they found no
acquisition of oestrogen-independent growth that significant evidence for a poorer re sponse, time to
is insensitive to tamoxifen.77,78 treatment failure or survival. In a more recent
A number of clinical studies, using various study, the relationship between HER-2
end-points such as time to relapse, more rapid overexpression and response to tamoxifen was
spread to other sites, and DFS or OS, have
reported an association between HER-2 positivity
and resistance to hormonal therapy.79–83 Some 29
reports have described specific re
sistance to tamoxifen in HER-2-overexpressing tu
Chapter 2 tumours and basal phenotype are not
interchangeable but rather dis tinct entities.
Ninety-one per cent of TNP tumours display a
examined in the adjuvant setting in 741 (650 ER significant association with the basal-like centroid
positive, 91 ER negative/PR positive) of the total and about 20% of non-TNP tumours clus ter
of 1572 patients in the CALGB 8541 Trial who together with TNP tumours in the ‘basal-like’
had HER-2 measured.88 Tamoxifen significantly cluster; these data support the view that the major
improved response, DFS and OS, irrespective of ity of, but not all, TNP tumours have a basal-like
HER-2 status. However, tamoxifen was not ran phenotype and that the majority of, but not all,
domised within this trial and all patients received basal-like tumours have a TNP phenotype. 89 Triple
one of three regimens of doxorubicin. Thus these negative breast cancers have a relapse pattern
data on tamoxifen resistance have limitations to that is very different from hormone-positive breast
their interpretation. can cers: the risk of relapse is much higher for the
With regards to HER-2 and prognosis, not only first 3–5 years but drops sharply and substantially
is there clinical value in its positive expression but be low that of hormone-positive breast cancers
also in its absence – in so-called ‘triple-negative after that. This relapse pattern has been
phe notype’ cancers (TNP), i.e. cancers that are recognised for all types of triple-negative cancers
HER-2, ER and PR negative. It also denotes a for which suffi cient data exist, although the
biologically different subgroup. The recognition of absolute relapse and survival rates differ across
basal phe notype in these triple-negative cancers subtypes.
is of growing importance, with several lines of Previous studies have shown that the expression
evidence support ing the view that triple-negative of ‘basal markers’ (i.e. CK5/6, CK14, CK17 and/ or
EGFR) is associated with a poor prognosis, re cancers are more frequently ‘medullary-like’
gardless of hormone receptor expression. The ex carcinomas and high grade compared to those in
pression of basal markers (basal cytokeratins and patients without this genetic alteration. Cancers
EGFR) in TNP tumours (core basal phenotype) associated with BRCA1 mutations have a
also correlates with a worse prognosis and identi significantly higher mitotic rate, a larger proportion
fies a clinically distinct subgroup within the TNP of tumour with a continuous push ing margin and
group. Moreover, it should be noted that identifi more lymphocytic infiltration than sporadic breast
cation of a subgroup of tumours solely based on cancers.90,91 These are also likely to be less
the lack of expression of immunohistochemistry positive for ERs and PRs and are more of ten
(for example, TNP) risks misassignment based on aneuploid, have a high S-phase fraction, show
technical artefacts.89 greater accumulation of p52 protein and
expression of basal cytokeratins (CK5/6, CK14
and CK17) than do sporadic breast cancers.92–97
30 However, none of these features alone or in
combination can be used to identify a cancer as
being from a BRCA1 gene mutation carrier.
Pathological features reported in BRCA2-related
An in-depth discussion of these issues is beyond
breast cancer have been less con sistent. Marcus
the scope of this work and the reader is directed to
et al.92 noted a significantly higher proportion of
detailed reviews by Rakha et al.89
tubulolobular cancers in BRCA2 mu tation carriers
than in other patients. However, it has been
Histopathology of patients reported in another study that tubular carcino mas
are less common in BRCA2 mutation carriers.94
with BRCA1 and BRCA2 Armes et al.97 investigated the histological pheno
mutations types of breast carcinoma in women under 40
years of age with and without BRCA1 and BRCA2
The cancers that develop in patients with a genetic germ line mutations. It was found that the
predisposition to breast cancer as a result of mu pleomorphic variant of invasive lobular carcinoma
tations in the breast cancer susceptibility genes was more common in patients with BRCA2
BRCA1 and BRCA2 are of great clinical interest. mutations. Others have reported that
Identification of histological features that could in BRCA2-related cancers tend to be of high
dicate a genetic predisposition would be useful in histological grade,94,98 whereas others have not
providing an insight into the function of these noted any significant difference in grade between
genes and may aid in identifying those in whom BRCA2-related cancers and controls.99
screen ing for genetic mutations would be useful.
There is a general agreement that BRCA1-related

sequencing of the entire human genome has
permitted the introduction of high-throughput
High-throughput molecular technologies that allow survey of thousands of
genes and their products in a single assay.
techniques as prognostic Together with powerful analytical tools this has
and predictive tools in breast opened new av enues for classifying breast cancer
into biologically and clinically distinct groups
cancer based on DNA copy number alterations and gene
expression patterns.
Genomics Cancer development is driven by the
Mechanisms of genetic aberrations The accumulation of DNA changes within the genome.
DNA repair defects lead to a genome-wide transcription start sites of approximately half of all
genetic instability and this can drive further cancer human genes. DNA hypomethylation has been
progression. Genomics shown to correlate with chromosome instability.
(the study of the human genome), Chromatin architecture remodelling is essential for
transcriptomics (the study of the transcriptome gene transcription; thus the prognostic significance
(mRNAs)) and pro teomics (the analysis of the of quantitative chromatin changes.100,101
protein complement of the genome) are the three
branches of molecu lar biology that are being
Assessment of genomic status
explored in an effort to improve understanding,
in breast cancer
diagnosis, prognostication and provision of new
therapeutic targets for the treatment of breast Several high-throughput molecular techniques
cancer. have been developed to assess the status of the
The genetic fingerprint provides information on genome in a given cell population, in terms of
normal cellular processes and morphological/phe copy number, DNA sequencing and structure.
notypic expression. When this message is altered, Large-scale DNA or gene copy number
it forms the nidus for the development and progres alterations can be assessed using loss of
sion of cancer. These alterations can be caused heterozygosity (LOH) and compara
by DNA mutations, chromosomal aberrations, tive genomic hybridisation (CGH) studies as well
epigen etic modification and protein interactions. as other techniques. LOH can be used to identify
DNA mutations can lead to a change in gene chro mosomal regions where allelic losses are
function. These mutations are often due to base more fre quent. In one study,102 a panel of 150
substitutions that may directly cause a stop polymorphic microsatellite markers from
codon. As a result the gene is only partly throughout the whole genome was used to
transcribed and any functional protein production identify such regions. These data were correlated
terminated. Chromosomal in stability can be due with clinicopathological fea tures and showed four
to DNA modification, mutation or viral genome specific loci that correlated with lymph node
integration. Changes in chromo some numbers metastasis (11q23–24, 13q12, 17p13.3 and
are also seen and cancers can be an euploid 22q13). Microsatellites can also be used to check
(60–90 chromosomes) or near diploid (46 the ability of cells to repair DNA rep lication errors.
chromosomes). In solid tumours virtually all chro Microsatellite instability (MSI) can be employed to
mosomal rearrangements are unbalanced, the net identify both genetic and epigenetic modification.
result being loss or gain in certain parts of the chro It has been shown that both genetic and
mosome. By screening the genome, current epigenetic alterations of hMSH2 and hMLH1
losses and gains can be identified. This contribute to genomic instability and tumour pro
information can be used to investigate regions of gression in sporadic breast cancer.
overlap to reveal Chromosome-based CGH can be used to
Pathology and biology of breast cancer identify the loss of one or both copies of a given
gene, as well as regions of amplification. In
the genes that are involved in malignant transfor addition, the technique provides information on
mation and cancer progression. In some the number of copies of any part of each
situations, the DNA copy number and code chromosome throughout the whole ge
remain intact but the accessibility of this DNA for nome. There have been several studies using
transcription is af fected by epigenetic
modifications. This process can occur through
CG
DNA methylation. It directly silences genes and
interferes with the binding of transcrip tion factors
by changing the chromatin structure around the H
gene. Global hypomethylation is a char acteristic
feature of the genome of a cancer cell. However,
some sequences can be hypermethylated, such
as CpG islands, which tend to lie around the
31

Chapter 2 tion with each one representing one gene.
Broadly speaking, molecular profiling of breast
in breast carcinoma,103,104 and those with clinical cancers by gene expression microarrays can be
follow-up have led to the identification of genetic performed in one of two ways: unsupervised or
changes related to prognosis. However, the resolu supervised analysis. Unsupervised analysis is
tion of every chromosome-based CGH technique used mainly for partitioning tumour samples into
was limited to approximately 10 million base pairs groups or classes on the basis of gene
(Mbp). This limitation makes it difficult to link copy expression profiles regardless of other features.
number changes to the genes involved. Another The main objective of
limi tation to this technique is the need to perform
karyo
typing for target identification in every experiment. 32
Therefore, array CGH has been developed to over
come the problems of chromosome-based CGH.
Array CGH does not require a normal metaphase this approach is to determine whether discrete sub
spread but rather an array of DNA fragments sets of cancers can be defined on the basis of
(100bp to 100kb) and their precise chromosomal gene expression profiles, and whether new
locus. This approach can provide resolutions up classes can be identified (class discovery) that
to 1Mb. Albertson et al. utilised this technique to may have clinical significance and to develop a
map the recurrent breast cancer amplicon at chro new molecular tax onomy. Supervised analysis, on
mosome 20q12.3. They were able to demonstrate the other hand, is used to allocate tumours to
that what was previously described as a single am specific groups based on clinical or pathological
plicon is in fact two distinct amplicons (ZNF217 features (e.g. clinical out come or response to
and CYP24). A novel amplicon at 17q21.3 that is therapy). There are two main types of supervised
implicated in the amplification and overexpression analysis: class comparison and class prediction.
of the HOXB7 gene in breast cancer has also The former aims to identify the transcriptomic
been recently characterised using array CGH.105 differences between two classes of tumours,
whereas the development of a ‘gene signa ture’ is
Assessment of gene expression the ultimate goal of the latter.
in breast cancer In breast cancer, the class discovery studies
In gene expression profiling, each gene is usually were pioneered by the Stanford group,106 which
represented by a single element (created with a proved the principle that breast cancer could be
complementary DNA (cDNA) or oligonucleotide classified into molecularly distinct groups based
for the gene studied) and with high-density oligo upon gene ex
nucleotide cDNA microarray technology many pression profiles and their similarity to normal cell
thousands of gene-specific mRNAs can be counterparts. Multiple independent studies have
measured in parallel in a single tissue sample. con firmed and expanded the original results.
The principle of gene expression cDNA array is According to these studies, breast cancer was
comparable to array CGH, other than that it uses classified into two main classes (ER-positive and
cDNA gener ER-negative tu mours) and each one can be
ated from mRNA. Similar to the two main types of classified into multiple molecularly distinct
expression microarrays (cDNA microarrays and subclasses. For example, the ER-negative
oligonucleotide microarrays), serial analysis of tumours encompass three subgroups, one
gene expression (SAGE) can provide profiling of overexpressing HER-2, one with tumours ex
the tran pressing genes characteristic of
scriptome by taking a raw count of sequence tags, basal/myoepithelial cells (basal-like cancer), and
each representing a transcript in an RNA popula one with a gene expres sion profile similar to
normal breast tissue. cDNAs have also been used teins are expressed by a cancer, or whether they
to distinguish cancers with BRCA1/BRCA2 are functional, if expressed. Post-translational
mutations107,108 and to determine ER status,109 modifi cations such as glycosylations or
lymph nodes status110,111 and prog nostic phosphorylation of proteins affect their expression
subgroups in node-negative breast cancer.112 and function at the protein level, which cannot be
These recent studies demonstrate the ability of detected by as sessing the status of the genome
cDNA to have a direct translational use in clinical or transcriptome in a given cell population.
practice. Therefore, proteomics aims to assess changes at
the protein level in which the proteome is
Proteomics displayed. Proteomics is quickly evolv ing to
provide supportive and critical information
The genetic code does not inform on which pro
throughput molecular
to data generated through genomic approaches in techniques
high-throughput formats. This can be achieved by
It is envisaged that gene expression profiles may
using different techniques such as
be able to guide decisions on the choice of
two-dimensional polyacrylamide gel
hormonal, chemotherapeutic or targeted agents
electrophoresis, isotope-coded affinity tagging,
for each indi vidual in the future. Presently, one of
surface-enhanced laser desorption ionisation and
the best ex amples of the use of genomics and
matrix-associated laser desorption ionisation–time
proteomics in clinical practice is the use of HER-2
of flight.
expression/ amplification to select patients for
These global expression profiling approaches
trastuzumab. Such profiles are useful to identify
yield candidate proteins and related genes that
prognostically significant genes, which studies
require verification through application of other
have demonstrated can perform better than
techniques. Tissue microarrays (TMAs) provide a
traditional markers.104,114 In metastatic disease,
method for high-throughput protein expression
expression patterns can be used to establish the
analysis of large cohorts of archival samples that
origin of the metastatic le sion in patients with
can be readily linked to clinicopathological and
more than one primary. Transcription profiling has
long-term follow-up databases. TMAs have expe
also been shown to dif ferentiate accurately breast
dited the validation of the prognostic and
cancers with germ-line mutations in
predictive significance of several candidate
BRCA1/BRCA2 from those without such
biomarkers. The technique allows for a composite
mutations. Such findings, if validated in other
slide of up to 1000 cores of tissue to be
constructed into one paraffin block. The
advantages are obvious in screening for novel
studies, open the way for molecular phenotyping
protein expression. There are concerns on the
in high-risk families. In addition, gene expression
validity of the cores as being representative
profiling has been used to develop expression pre
samples and indeed this aspect has not been well
dictors (class prediction studies) that can be used
studied, with only one study examining a large
for many types of clinical management decisions,
sample size with correlation to corresponding
including risk assessment, diagnostic testing, prog
larger histology sections.103 This technology and
nostic stratification and treatment selection. There
its application have been extensively reviewed.113
are two types of class prediction studies in breast
cancer: prognostic and predictive class prediction.
Prognostic class prediction includes (i)
Clinical use of the high poor-prognosis gene signatures that can
discriminate between a good and a poor outcome poor-prognosis group.101,109 MammaPrint® assay,
by comparison between highly aggressive and which is based on the Amsterdam 70-gene gene
less aggressive primary tumours, and (ii) signature, is currently used as a molecular
recurrence score gene signatures that define diagnos
tumours based on the risk of disease relapse. The tic test for breast cancer prognostication and pre
predictive class prediction includes predictors of re diction. Another predictor calculates a recurrence
sponse to therapy.115,116 score on the basis of the expression of 21 known
An example of a prognostic class prediction genes, with the use of RT–PCR in formalin-fixed,
study is that reported by van't Veer et al. paraffin-embedded tissue. The ‘Oncotype DX as
Expression pro files of 117 primary breast cancers say’ has been developed by researchers at
were compared with known prognostic factors Genomic Health. Oncotype DX is currently used
and matched with 5 years of follow-up data. as a diag nostic test to quantify the likelihood of
Twenty-five thousand genes were used to disease re currence in women with early-stage
generate expression profiles, which separated the breast cancer and assesses the likely benefit from
tumours into two groups. Group 1 developed tamoxifen and certain types of chemotherapy.
distant metastasis in 34% and in group 2 70% Oncotype DX em ploys a mathematical algorithm
developed distant metastases. From the called the recur rence score (range 1–100) to
25000-gene set, 70 genes had great accuracy in calculate continuous risk for relapse and death for
predicting recurrent disease. Multivariate analysis patients receiving
with grade, size, LVI, age and ER showed the poor
prognosis microarray profile to be an independent
predictor of recurrent disease. This approach was 33
further tested in 295 patients and again the use of
gene profiling was able to accurately identify a
Chapter 2 tients based on the hypothesis that features of the
molecular programme of normal wound healing
adjuvant tamoxifen, and has recently been shown might play an important role in cancer metastasis.
in a large population-based case–control study to They found that this gene signature can improve
be an effective predictive test for ER-positive, the risk stratification of early breast cancer over
that provided by standard clinicopathological fea
tures. A gene expression signature of hypoxia re
Nottingham Prognostic Index (NPI), which is sponse, derived from studies of cultured mammary
epithelial cells' ‘hypoxia gene signature’, showed
based on three factors using the following
a strong predictor of clinical outcomes in breast
formula: ( )
cancer.119
NPI Pathological tumour size cm 0.2 = ⋅
Although proteomics has a long history that pre
node-negative breast cancer patients treated or un dates profiling at the RNA level, difficulties in pro
+ tein identification and the lack of reproducibility of
treated with tamoxifen and no chemotherapy.117 + several assay platforms have limited the utility of
In addition, a number of microarray studies have such profiling systems. In breast cancer research,
() however, proteomics has begun to take on a role
in the monitoring of response, prediction of resis
Lymph node stage 1,2 or 3 ( )
tance and relapse in patients treated with novel
Histological grade 1,2 or 3
bio directed therapies.120
been published which identified other prognostic
signatures with clinical significance. For example,
Change et al.118 have used a ‘wound-response Clinical use of prognostic
gene expression signature’ to stratify breast factors in patient
cancer pa
management
Prognosis in breast cancer depends on the Percentage in prognostic group
presence 1980–86 1990–99
Arbitrary cut-off points of 3.4 and 5.4 are used to
divide patients into six prognostic groups: EPG 12 14
excellent (EPG), good (GPG), moderate (MPG) I GPG 19 21
and II, poor (PPG) and very poor (VPG) (Tables MPG I 29 28
2.1–2.3, Figs 2.5 and 2.6). MPG II 24 22
The NPI provides extremely powerful prognostic in PPG 11 11
formation within the NTPBCS and was able to VPPG 5 4
dem onstrate its utility and reproducibility in
Prognostic groups: EPG, excellent prognosis; GPG, good
studies from other centres. In this respect Henson prognosis; MPG, moderate prognosis; PPG, poor
et al.,18 in a ret rospective analysis of prognostic prognosis; VPG, very poor prognosis.
data in over 22000 women from the SEER
Programme of the National Cancer Institute in the
Table 2.2 • Differences in 10-year survival for
USA, confirmed that a
each prognostic group
Table 2.1 • NPI showing percentage in each prognostic 10-year

group
of spread of disease and on the number of intrinsic biological
EPG 96 2
inherent aggressive ness of the characteristics, some of which
P (log rank)
tumour. The latter depends on a survival
(%) ±95% CL
been the approach in deriving the
have already been evaluated, such as
morphological features, growth rate and hormone
responsiveness. Accurate prognostication is now 34
required on an in dividual patient basis and this GPG 93 4 0.14 MPG I 82 4 <0.0001 MPG II
can only be achieved by using a Prognostic Index 75 4 0.0007 PPG 53 8 <0.0001 VPG 39 12
that includes both time dependent and biological
0.003 All 80 2
factors. The best way to obtain such an index is
to take potential factors that have been shown to Prognostic groups: EPG, excellent prognosis; GPG, good
prognosis; MPG, moderate prognosis; PPG, poor
have some value in uni variate analysis and prognosis; VPG, very poor prognosis.
submit them to multivariate analysis. This has CL, confidence limit.

ARR
Table 2.3 • Relative risk reduction by NPI

Pathology and biology of breast 1980–86 ± 95% CL 1990–99 ± 95% CL RRR
(death)
(death)
cancer %
EPG 88 6 96 2 0.67 8 GPG 72 8 93 4 0.75 21 MPG I 61 6 82 4 0.54 21 MPG II 42 6 75 4

0.57 35 PPG 14 8 53 8 0.45 39 VPG 12 10 39 12 0.31 27 All 55 80 0.56 25
Prognostic groups: EPG, excellent prognosis; GPG, good prognosis; MPG, moderate prognosis; PPG, poor prognosis; VPG, very
poor prognosis.
ARR, absolute risk reduction; CL, confidence limit; RRR, relative risk reduction.
1.0
0.9
n
i
it
r
o PPG
0.1
r
EPG
0.8
0.7
0.6
GPG
MPGI
0.5
VPG
0.4
0.0
18016815614413212010896847260483624120 Months
Figure 2.5 • Survival by NPI (1980–86).
combination of lymph node stage and

histological grade improved prediction of
prognosis. In a similar way to the NTPBCS,
MPGII Chevallier et al.121 have identified ‘young’ age,
0.3 tumour size and histological grade as fac
tors to be added to lymph node stage in the
0.2 prediction of recurrence; these factors were
used to divide lymph node-negative patients
into three prognostic groups.
One of the strengths of the NPI is the fact that
it has been verified prospectively in the
NTPBCS. Further confirmation of its value has
been provided by its validation in two large in breast cancer include Adjuvant!Online
multicentre studies in (AO)124 and the St Gallen criteria,125 and more
volving nearly 11 000 patients in total.122,123 Such recently Predict (a new UK prognostic model
that predicts survival following surgery for
invasive breast can cer).126 This tool is a
studies demonstrate the inherent power of the population-based validation of the prognostic
path ological factors used in the NPI, which has model PREDICT for early breast cancer.
become the most widely used index for the
management of
patients with breast cancer, certainly in the UK. 35
Other pathology-based prognostic indices used
Chapter 2 1.0
0.2
0.9
0.1
18016815614413212010896847260483624120 Months
0.8
Figure 2.6 • Survival by NPI (1990–99).
0.7
g
n
i
i
Adjuvant!Online
(http://www.adjuvantonline.com) was developed
in the USA and published in 2001; users can
input information on a patient's age, ER status,
EPG tumour grade, tumour size and number of
GPG positive nodes, and obtain predictions of 10-year
OS (the likelihood of being alive 10 years after
the diagnosis of breast cancer was first carried
MPGI MPGII
out), BCSS (the likelihood of not dying of breast
can
v cer within 10 years of diagnosis) and event-free
r
u survival (EFS; the likelihood of surviving 10 years

without recurrence (local, regional or distant), a
s
second primary breast cancer, or death from

n
breast cancer), both with and without any

it
r
proposed ad
o
r
juvant therapy. The performance of AO has been
P
evaluated in cohorts of patients in Germany,
0.6
Canada and the UK. It has an advantage over
the NPI in that it integrates treatment, and for
PPG
this rea
0.5
son AO is the programme most commonly used
0.4 by oncologists.
The current St Gallen-derived algorithm for selec
tion of adjuvant systemic therapy for early breast
cancer patients includes tumour size and grade,
nodal status, menopausal status, peritumoral ves
VPG sel invasion, endocrine status and HER-2 status.
Predict, which is similar in concept to AO but is
0.3 based on UK patients, has been developed at
the University of Cambridge. This tool is an = 0.02), ER (P = 0.02), PR sta tus (P = 0.02),
alternative mitotic index (P = 0.01) and tumour ploidy (P =
0.04). Survival from initial therapy cor related
significantly with ER (P = 0.01) and PR status (P
36 = 0.04).127 In these patients, histological grade,
mitotic index, tumour ploidy, and ER and PR
status of the primary tumour may predict
response and prognosis.
updated method of assessing probability of These authors also explored the prognostic
survival for UK patients. value of certain factors as a composite index to
All of the above discussion applies to primary guide decision-making in metastatic disease.128
op erable breast cancer. Prognostic factors have The ad vanced breast cancer (ABC) index
also been explored in locally advanced and comprises the grade, ER status, site of initial
metastatic breast cancer, albeit not to the same metastasis and disease-free interval. This index
extent. In a small study of locally advanced was tested both ret rospectively and
pancreatic cancer patients (n = 60) treated with prospectively and shown to sepa rate groups
either tamoxifen or radiation therapy and with where the survival between each was highly
crossover upon failure, response to therapy significant (P < 0.001).
correlated significantly with histological grade (P

Breast cancer histopathology minimum dataset.

Figure 2.7 • Breast cancer histopathology minimum dataset. DCIS, ductal carcinoma in situ.
37
Chapter 2
Conclusion Contents of the final surgical

pathology report: the
Modern management of breast cancer requires a
multifaceted understanding of both the clinico
minimum dataset
pathological features and the molecular portraits
The Royal College of Pathologists' minimum da
of the cancer. Emerging evidence is suggesting taset for breast cancer was originally developed in
that genomic approaches to prognosticating recognition that certain histopathological features
patients can be complementary to standard of both in situ and invasive carcinoma are directly
prognostic tools. Existing combinations of such related to clinical outcome and may therefore be
factors are either largely clinical (e.g. NPI) or im portant in deciding the most appropriate
purely molecular (e.g. Oncotype DX), but a treatment, including extent of surgery and use of
hybridised index is likely to yield more and choice of adjuvant therapy. In addition,
information. Updating the traditional NPI and histopathological features can be used to monitor
incorporating well-established prognostic factors breast screening programmes, the success of
such as grade, stage and tumour size with which is reflected by more favourable prognostic
biological factors have produced NPI+. This new features of the cancers detected and also
tool, which combines updated methods for assess changing patterns of disease, par ticularly
ment of histological grade and tumour stage with identified by cancer registries.
biological class determination, has recently been The minimum set of data should be used by pa
shown to provide significant prediction of outcome thologists reporting all breast cancers, both screen
in all biological classes.129 detected and those presenting symptomatically.
The need to accurately prognosticate each The Royal College of Pathologists minimum data
patient continues to drive research efforts to set has been approved by the NHS Breast
unravel the ‘perfect’ prognostic index. This Screening Programme and the European
information will al low us to use adjuvant therapies Commission Working Group, the British
to influence and improve such prognosis for a Association of Surgical Oncologists, the British
better outcome and hopefully cure. Breast Group and the United Kingdom
Association of Cancer Registries (Fig. 2.7).
Key points
• Histopathological features such as size, grade and lymph node stage are key to
prognosticating and guiding adjuvant therapy in breast cancer patients.
• The type 1 growth factor family (e.g. EGFR and HER-2) are becoming key molecular features
in the design of novel biologically targeted therapies.
• The combination of standard pathological features and each cancer's genetic fingerprint will
enable us in the future to individually tailor each patient's therapy.
38
References breast cancer: the ALMANAC Trial. J Natl Cancer
Inst 2006;98(9):599–609.
1. Singletary SE, Allred C, Ashley P, et al. Revision of 4. Alvarez S, Anorbe E, Alcorta P, et al. Role of so
the American Joint Committee on Cancer staging nography in the diagnosis of axillary lymph node
system for breast cancer. J Clin Oncol metastases in breast cancer: a systematic review.
2002;20:3628–36. Am J Roentgenol 2006;186:1342–8.
2. Rampaul RS, Mullinger K, Macmillan RD, et al. 5. Rampaul RS, Evans AJ, Ellis IO, et al. Long term
Incidence of clinically significant lymphoedema as regional recurrence and survival after axillary
a complication following surgery for primary nodal sampling for breast cancer. Eur J Cancer
operable breast cancer. Eur J Cancer 2003;1(4):23(abstr.).
2003;39(15):2165–7. 6. Shukla HS, Melhuish J, Mansel RE, et al. Does lo
3. Mansel RE, Fallowfield L, Kissin M, et al. cal therapy affect survival rats in breast cancer?
Randomized multicenter trial of sentinel node bi opsy Ann Surg Oncol 1999;6(5):455–60.
versus standard axillary treatment in operable

from age 40 to 48 years. Breast 2011;20(6):525–8.
15. Foulkes WD, Grainge MJ, Rakha EA, et al. Tumor
7. Macmillan RD, Rampaul RS, Lewis S, et al. size is an unreliable predictor of prognosis in
Preoperative ultrasound-guided node biopsy and basal-like breast cancers and does not correlate
sentinel node augmented node sample is best prac closely with lymph node status. Breast Cancer Res
tice. Eur J Cancer 2004;40(2):176–8. Treat 2009;117(1):199–204.
8. Veronesi U, Galimberti V, Zurrida S, et al. Sentinel 16. O'Dwyer PJ. Editorial. Axillary dissection in pri
lymph node biopsy as an indicator for axillary mary breast cancer; the benefits of node clearance
dissection in early breast cancer. Eur J Cancer warrant reappraisal. Br Med J 1992;302:360–1.
2001;37(4):454–8. 17. Blamey RW. Clinical aspects of malignant disease.
9. Veronesi U, Paganelli G, Viale G, et al. Sentinel In: Elston CW, Ellis IO, editors. Systemic pathology.
lymph node biopsy and axillary dissection in breast The Breast. 3rd ed. London: Churchill Livingstone;
cancer: results in a large series. J Natl Cancer Inst 1998. p. 501–13.
1999;91(4):368–73. 18. Henson DE, Ries L, Freedman LS, et al.
10. Schwartz GF, Guiliano AE, Veronesi U; Consensus Relationship among outcome stage of disease and
Conference Committee. Proceeding of the consensus histologic grade for 22,616 cases of breast cancer.
conference of the role of sentinel lymph node bi opsy in Br Cancer Res Treat 1991;22:207–19.
carcinoma or the breast April 19–22, 2001, 19. Kutianawala MA, Sayed M, Stotter A, et al. Staging
Philadelphia, PA, USA. Breast J 2002;8(3):124–38. the axilla in breast cancer: an audit of lymph-node
11. Cserni G, Amendoeira I, Apostolikas N, et al. retrieval in one UK regional centre. Eur J Surg
Discrepancies in current practice of pathological Oncol 1998;24:280–2.
evaluation of sentinel lymph nodes in breast can cer. 20. Steele RJC, Forrest APM, Gibson T. The efficacy of
Results of a questionnaire based survey by the lower axillary sampling in obtaining lymph node
European Working Group for Breast Screening status in breast cancer: a controlled randomized
Pathology. J Clin Pathol 2004;57(7):695–701. trial. Br J Surg 1985;72:368–9.
12. Fisher ER, Fisher B, Sass R, et al. Pathologic Pathology and biology of breast cancer
findings from the National Surgical Adjuvant
Breast Project (Protocol No. 4). XI. Bilateral breast 21. Dixon JM, Dillon P, Anderson TJ, et al. Axillary
cancer. Cancer 1984;54:3002–11. sampling in breast cancer: an assessment of its ef
13. Carter CL, Allen C, Henson DE. Relation of tu mor ficacy. Breast 1998;7:206–8.
size, lymph node status, and survival in 24,740 breast 22. Cabanas RM. An approach for the treatment of pe
cancer cases. Cancer 1989;63:181–7. nile carcinoma. Cancer 1977;39:456–66.
14. Bennett RL, Evans AJ, Kutt E, et al. Pathological 23. Royal College of Radiologists. Quality assurance
and mammographic prognostic factors for screen guidelines for radiologists. 1990 NHS BSP
detected cancers in a multi-centre randomised, con Publications no. 15.
trolled trial of mammographic screening in women
24. Royal College of Radiologists. Quality assur ance
guidelines for radiologists. 1997 NHS BSP 2002;41(3A):154–61.
Publications no. 15. January. 31. Rakha EA, El-Sayed ME, Lee AH, et al. Prognostic
25. European Commission. European guidelines for sig nificance of Nottingham histologic grade in invasive
quality assurance in mammography screening. 2nd breast carcinoma. J Clin Oncol 2008;26(19):3153–8.
ed Luxembourg: Office for Official Publications of 32. Rakha EA, El-Sayed ME, Menon S, et al. Histologic
the European Communities; 1996. grading is an independent prognostic factor in inva
26. Mansel RE, Goyal A, Newcombe RG. ALMANAC sive lobular carcinoma of the breast. Breast Cancer
Trialists Group. Internal mammary node drain age Res Treat 2008;111(1):121–7.
and its role in sentinel lymph node biopsy: the 33. Pinder SE, Ellis IO, Galea M, et al. Pathological
initial ALMANAC experience. Clin Breast Cancer prog nostic factors in breast cancer III. Vascular
2004;5(4):279–84. invasion: relationship with recurrence and survival
27. Rosen PP, Groshen S, Saigo S, et al. Pathological in a large study with long-term follow-up.
prog nostic factors in stage I (TIN0M0) and stage II Histopathology 1994;24(1):41–7.
(TINIM0) breast carcinoma: a study of 644 patients 34. Truong PT, Yong CM, Abnousi F, et al.
with median follow-up of 18 years. J Clin Oncol Lymphovascular invasion is associated with re
1989;7:1239–51. duced locoregional control and survival in women
28. Parker C, Rampaul RS, Pinder SE, et al. with node-negative breast cancer treated with
E-Cadherin as a prognostic indicator in primary mastectomy and systemic therapy. J Am Coll Surg
breast cancer. Br J Cancer 2001;14(85):1958–63. 2005;200(6):912–21.
29. Rakha EA, El-Sayed ME, Powe DG, et al. Invasive 35. Haybittle JL, Blamey RW, Elston CW, et al. A prog
lobular carcinoma of the breast: response to nostic index in primary breast cancer. Br J Cancer
hormonal therapy and outcomes. Eur J Cancer 1982;45:361–6.
2008;44(1):73–83.
30. Elston CW, Ellis IO. Pathological prognostic fac
tors in breast cancer I. The value of histological 39
grade in breast cancer: experience from a large
study with long-term follow-up. Histopathology
Chapter 2 Mod Pathol 2005;18(1):26–35.
43. Dowsett M, Allred C, Knox J, et al. Relationship
between quantitative estrogen and progesterone
36. Ellis IO, Bell J, Todd J, et al. Evaluation of immu
receptor expression and human epidermal growth
noreactivity with monoclonal antibody NCRC-II in
factor receptor 2 (HER-2) status with recurrence in
breast carcinoma. Br J Cancer 1987;56:295–9.
the Arimidex, Tamoxifen, Alone or in Combination
37. Lee AH, Pinder SE, Macmillan RD, et al. Prognostic trial. J Clin Oncol 2008;26(7):1059–65.
value of lymphovascular invasion in women with
44. Kumar RR, Meenakshi A, Sivakumar N. Enzyme
lymph node negative invasive breast carcinoma.
immunoassay of human epidermal growth factor
Eur J Cancer 2006;42(3):357–62.
(hEGFR). Hum Antibodies 2001;10:143–7.
38. Gee JM, Eloranta JJ, Ibbitt JC, et al.
45. Eberhard DA, Huntzicker E, et al. Epidermal
Overexpression of TFAP2C in invasive breast
Growth Factor receptor immunohistochemistry:
cancer correlates with a poorer response to
assay selection and amplification to breast
anti-hormone therapy and re duced patient
cancers. ASCO 2002;no. 1791 (Abstr.).
survival. J Pathol 2009;217(1):32–41.
46. Klijn JGM, Berns PMJ, Schmitz PI, et al. The clini
39. Sarwar N, Kim JS, Jiang J, et al. Phosphorylation
cal significance of epidermal growth factor receptor
of ERalpha at serine 118 in primary breast cancer
in human breast cancer: a review of 5232 patients.
and in tamoxifen-resistant tumours is indicative of
Endocr Rev 1992;13:3–17.
a complex role for ERalpha phosphorylation in
breast cancer progression. Endocr Relat Cancer 47. Tsutsui S, Ohno S, Murakami S, et al. Prognostic
2006;13(3):851–61. value of epidermal growth factor and its
relationship to the ER status of 1029 patients with
40. Gee JM, Shaw VE, Hiscox SE, et al. Deciphering
breast cancer. Breast Cancer Res Treat
antihormone-induced compensatory mechanisms
2002;71:67–75.
in breast cancer and their therapeutic implications.
Endocr Relat Cancer 2006;13(Suppl. 1):S77–88. 48. Rampaul RS, Pinder SE, Robertson JF, et al.
EGFR expression in operable breast cancer: is it of
41. Dowle CS, Owainati A, Robins A, et al. The prog
prognos tic significance? Clin Cancer Res
nostic significance of the DNA content of human
2004;10(7):2578.
breast cancer. Br J Surg 1987;74:133–6.
49. Ferrero JM, Ramaioli A, Largillier R, et al.
42. Putti TC, El-Rehim DM, Rakha EA, et al. Estrogen
Epidermal growth factor receptor expression in 780
receptor-negative breast carcinomas: a review of
breast cancer patients: a reappraisal of the
morphology and immunophenotypical analysis.
prognos tic value based on an eight-year median
follow-up. Ann Oncol 2001;12(6):835–41. Netw 2006;4(Suppl. 3):S1–24.
56. Bartlett JM, Going JJ, Mallon EA, et al. Evaluating
HER2 amplification and overexpression in breast
40 cancer. J Pathol 2001;195:422–8.
57. Press MF, Slamon DJ, Flom KJ, et al. Evaluation of
HER-2/neu gene amplification and overexpression:
comparison of frequently used assay methods in a
50. Nicholson RI, McCelland RA, Gee JMW, et al. molecularly characterized cohort of breast cancer
Epidermal growth factor receptor expression in specimens. J Clin Oncol 2002;20:3095–105.
breast cancer. Association with response to
58. Walker RA, Bartlett JM, Dowsett M, et al. HER2
endocrine therapy. Breast Cancer Res Treat
testing in the UK: further update to recommenda
1994;29:117–25.
tions. J Clin Pathol 2008;61(7):818–24.
51. Knoop A, Bentzen SM, Nielsen MM, et al. Value of
59. Dowsett M, Bartlett J, Ellis IO, et al. Correlation
epidermal growth factor receptor, HER-2, p53 and
between immunohistochemistry (HercepTest) and
steroid receptors in predicting the efficacy of
fluorescence in situ hybridization (FISH) for HER-2
Tamoxifen in high risk postmenopausal breast can
in 426 breast carcinomas from 37 centres. J Pathol
cer patients. J Clin Oncol 2001;19:3376–84. 52.
2003;199:418–23.
Wolff AC, Hammond ME, Schwartz JN, et al. American
60. Slamon DJ, Clark GM, Wong SG, et al. Human
Society of Clinical Oncology/College of American
breast cancer: correlation of relapse and survival
Pathologists guideline recommendations for human
with amplification of the HER-2/neu oncogene.
epidermal growth factor receptor 2 test ing in breast Science 1987;235(4785):177–82.
cancer. J Clin Oncol 2007;25:118–45. 53. Ratcliffe N,
61. Rilke F, Colnaghi MI, Cascinelli N, et al. Prognostic
Wells W, Wheeler K, et al. The combina tion of in situ
significance of HER-2/neu expression in breast can
hybridization and immunohistochemi cal analysis: an cer and its relationship to other prognostic factors.
evaluation of Her2/neu expression in Int J Cancer 1991;49(1):44–9.
paraffin-embedded breast carcinomas and adjacent
62. Viani GA, Afonso SL, Stefano EJ, et al. Adjuvant
normal-appearing breast epithelium. Mod Pathol trastuzumab in the treatment of her-2-positive early
1997;10:1247–52. breast cancer: a meta-analysis of published
54. Busmanis I, Feleppa F, Jones A, et al. Analysis of random ized trials. BMC Cancer 2007;7:153.
cerbB2 expression using a panel of 6 commercially 63. Carter P, Presta L, Gorman CM, et al.
available antibodies. Pathology 1994;26:261–7. Humanization of an anti-p185HER2 antibody for
55. Carlson RW, Moench SJ, Hammond ME, et al. human cancer therapy. Proc Natl Acad Sci U S A
HER2 testing in breast cancer: NCCN Task Force 1992;89:4285–9.
report and recommendations. J Natl Compr Canc

breast cancer on the basis of hormonal and HER-2
status. Oncologist 2010;15(10):1043–9.
64. Disis ML, Knutson KL, Schiffman K, et al. Pre 69. Chia S, Norris B, Speers C, et al. Human epider
existent immunity to the HER-2/neu oncogenic mal growth factor receptor 2 overexpression as a
protein in patients with HER-2/neu overexpressing prognostic factor in a large tissue microarray se
breast and ovarian cancer. Breast Cancer Res ries of node-negative breast cancers. J Clin Oncol
Treat 2000;62:245–52. 2008;26:5697–704.
65. Ward RL, Hawkins NJ, Coomber D, et al. Antibody 70. Kraus MH, Popescu NC, Amsbaugh SC, et al.
immunity to the HER-2/neu oncogenic protein in Overexpression of the EGF receptor-related proto
patients with colorectal cancer. Hum Immunol oncogene erbB-2 in human mammary tumor cell
1999;60:510–5. lines by different molecular mechanisms. EMBO J
1987;6:605–10.
66. Mirza AN, Mirza NQ, Vlastos G, et al. Prognostic
factors in node-negative breast cancer: a review of 71. Ring CJ, Blouin P, Martin LA, et al. Use of transcrip
studies with sample size more than 200 and follow up tional regulatory elements of the MUC1 and ERBB2
more than 5 years. Ann Surg 2002;235(1):10–26. genes to drive tumour-selective expression of a pro
drug activating enzyme. Gene Ther 1997;4:1045–52.
67. Bernards R, Destree A, McKenzie S, et al. Effective
tumor immunotherapy directed against an onco 72. Yu DH, Hung MC. Expression of activated rat neu
gene-encoded product using a vaccinia virus vector. oncogene is sufficient to induce experimental metas
Proc Natl Acad Sci U S A 1987;84:6854–8. tasis in 3T3 cells. Oncogene 1991;6:1991–6.
68. Amar S, McCullough AE, Tan W, et al. Prognosis 73. Bria E, Cuppone F, Fornier M, et al. Cardiotoxicity
and outcome of small (≤1 cm), node-negative and incidence of brain metastases after adjuvant
trastuzumab for early breast cancer: the dark side Expression of ras p21, p53 and C-erb-2 in advan vced
of the moon? A meta-analysis of the randomized breast cancer and response to first line hor monal
trials. Breast Cancer Res Treat therapy. Br J Cancer 1995;72:1259–66.
2008;109(2):231–9. 84. Pegram MD, Konecny GE, O'Callaghan C, et al.
74. Chang H, Riese 2nd DJ, Gilbert W, et al. Ligands Rational combinations of trastuzumab with che
for ErbB-family receptors encoded by a motherapeutic drugs used in the treatment of breast
neuregulin-like gene. Nature 1997;387:509–12. cancer. J Natl Cancer Inst 2004;96(10):739–49.
75. Juhl H, Downing SG, Wellstein A, et al. HER-2/ neu 85. Muss H, Berry D, Thor A, et al. Lack of interaction
is rate-limiting for ovarian cancer growth. of tamoxifen (T) use and ErbB-2/HER-2/neu (H)
Conditional depletion of HER-2/neu by ribozyme expression in CALGB 8541: a randomized adjuvant
targeting. J Biol Chem 1997;272:29482–6. trial of three different doses of cyclophosphamide,
76. Deshane J, Siegal GP, Wang M, et al. doxorubicin and fluorouracil (CAF) in node-positive
Transductional efficacy and safety of an primary breast cancer (BC). Proc Am Soc Clin
intraperitoneally delivered adenovirus encoding an Oncol 1999;18:68a.
anti-erbB-2 intracellular single-chain antibody for 86. Paik S, Bryant J, Park C, et al. ErbB-2 and
ovarian cancer gene therapy. Gynecol Oncol response to doxorubicin in patients with axillary
1997;64:378–85. lymph node positive, hormone receptor-negative
77. Schmidt M, Hynes NE, Groner B, et al. A bivalent breast cancer. J Natl Cancer Inst
single-chain antibody-toxin specific for ErbB-2 and 1998;90:1361–70.
the EGF receptor. Int J Cancer 1996;65:538–46. 87. Elledge RM, Green S, Ciocca D, et al. HER-2 ex
Pathology and biology of breast cancer pression and response to tamoxifen in estrogen
receptor-positive breast cancer: a Southwest Oncology
Group Study. Clin Cancer Res 1998;4(1):7–12.
78. Park JW, Hong K, Carter P, et al. Development of
anti-p185HER2 immunoliposomes for cancer ther apy. 88. Ravin PM, Green S, Albain V, et al. Initial report of
Proc Natl Acad Sci U S A 1995;92:1327–31. the SWOG biological correlative study of CerbB-2
expression as a predictor of outcome in a trial com
79. Chen SY, Yang AG, Chen JD, et al. Potent antitu paring adjuvant CAF with tamoxifen (T) alone. Proc
mour activity of a new class of tumour-specific killer Am Soc Clin Oncol 1998;17:97a.
cells. Nature 1997;385:78–80.
89. Rakha EA, Reis-Filho JS, Ellis IO. Basal-like
80. Newby JC, Johnston SRD, Smith I, et al. breast cancer: a critical review. J Clin Oncol
Expression of epidermal growth factor and C-erb-2 2008;26(15):2568–81.
during the development of tamoxifen resistance in
human breast cancer. Clin Cancer Res 90. Johannsson OT, Idvall I, Anderson C, et al. Tumour
1997;3:1643–51. biological features of BRCA1-induced breast and
ovarian cancer. Eur J Cancer 1997;33:362–71.
81. Nicholson RI, McCelland RA, Finlay P, et al.
Relationship between EGF-R, C-erb-2 protein 91. Chappuis PO, Nethercot V, Foulkes WD. Clinico
expression and Ki67 immunostaining in breast pathological characteristics of BRCA1- and
cancer and hormone sensitivity. Eur J Cancer BRCA2-related breast cancer. Semin Surg Oncol
1993;29A:1018–23. 2000;18:287–95.
82. Giai M, Roagna R, Ponzone R, et al. Prognostic
and predictive relevance of C-erb-2 and ras
expression in node positive and negative breast 41
cancer. Anticancer Res 1994;14:1441–50.
83. Archer SG, Eliopoulos SA, Spandidos D, et al.
Chapter 2 95. Robson M, Gilewski T, Haas B, et al. BRCA
associated breast cancer in young women. J Clin
Oncol 1998;16:1642–9.
92. Marcus JN, Watson P, Page DL. Hereditary
breast cancer. Pathobiology, prognosis and 96. Karp SE, Tonin PN, Bégin LR, et al. Influence of
BRCA 1 and BRCA 2 linkage. Cancer BRCA 1 mutations on nuclear grade and
1996;77:967. oestrogen receptor status of breast carcinoma in
Ashkenazi Jewish women. Cancer
93. Marcus JN, Page DL, Watson P, et al. BRCA 1
1997;80:435–41.
and BRCA 2 hereditary breast cancer
phenotypes. Cancer 1997;80:543. 97. Armes JE, Egan AJM, Southey MC, et al. The
histologic phenotypes of breast carcinoma occur
94. Breast Cancer Linkage Consortium. Pathology of
ring before age 40 in women with and without
familial breast cancer: differences between breast
BRCA 1 or BRCA 2 germline mutations. Cancer
cancer in carriers of BRCA 1 or BRCA 2 muta tional
1998;83:2335–45.
and sporadic cases. Lancet 1997;349:1505.
98. Agnarsson BA, Jonasson JG, Björnsdottir IB, et
al. Inherited BRCA 2 mutation associated with expres sion profiles in hereditary breast cancer. N
high grade breast cancer. Breast Cancer Res Eng J Med 2001;344:539–48.
Treat 1998;47:121–7. 109. van't Veer LJ, Dai H, van de Vijer MJ, et al.
99. Marcus JN, Watson P, Page DL, et al. BRCA 2 he Expression profiling predicts outcomes in breast
reditary breast cancer phenotype. Breast Cancer cancer. Breast Cancer Res 2002;5:57–8.
Res Treat 1997;44:275–7. 110. Gruvberger S, Ringner M, Chen Y, et al.
100. Baak JP, Colpaert CG, van Diest PJ, et al. Oestrogen receptor status in breast cancer is
Multivariate prognostic evaluation of the mi totic associated with remarkably distinct gene
activity index and fibrotic focus in node negative expression profiles. Cancer Res
invasive breast cancers. Eur J Cancer 2001;61:5979–84.
2005;41(14):2093–101. 111. West M, Blanchette C, Dressman H, et al.
101. Baak JPA, Vooiji GP, Brugal G. Nuclear image cy Predicting the clinical status of human breast
tometry: quantitation of chromatin pattern, steroid cancer by using gene expression profiles. Proc
receptor content and Ki-67. In: Baale JPA, editor. Natl Acad Sci U S A 2001;98:11462–7.
Manual of quantitative pathology in cancer diag 112. Ahr A, Kam T, Solbach S, et al. Identification of
nosis and prognosis. Heidelberg: Springer-Verlag; high risk breast cancer by gene expression profil
1991. p. 232–43. ing. Lancet 2002;359:131–2.
102. Palcic B, Garner DM, MacAulay CE. Image cytom 113. van de Vijver MJ, He YD, van't Veer LJ. A gene
etry and chemoprevention in cervical cancer. J expression signature as a predictor of survival in
Cell Biochem 1995;23(Suppl.):43–54. breast cancer. N Engl J Med
103. Nagahata T, Hirano A, Utada Y, et al. Correlation 2002;347:1999–2009.
of allelic losses and clinicopathologic factors in 114. Leong A.S.-Y., Zhuang Z. The changing role of pa
504 primary breast cancers. Breast Cancer thology in breast cancer diagnosis and treatment.
2002;9:208–15. Pathobiology 2011;78(2):99–114.
104. Hermsen M.A.J.A., Baak JPA, Weiss J, et al. 115. Bertucci F, Viens P, Hingamp P, et al. Breast
Genetic analysis of 513 lymph node negative cancer revisited using DNA array-based gene
breast carci nomas by CGH and relation to expression profiling. Int J Cancer
clinical, pathologic, morphometric and DNA 2003;103(5):565–71.
cytometric prognostic factors. J Pathol
1998;186:356–62. 116. Perou CM, Sorlie T, Eisen MB, et al. Molecular
portraits of human breast tumours. Nature
105. Janssen EAM, Baak JPA, Guervos MA, et al. In 2000;406:747–52.
lymph node-negative breast cancer, specific chro
mosomal aberrations are strongly associated with 117. Habel LA, Shak S, Jacobs MK, et al. A population
high mitotic activity and predict outcome more ac based study of tumor gene expression and risk of
breast cancer death among lymph node-negative
patients. Breast Cancer Res 2006;8:R25.
118. Chang HY, Nuyten DS, Sneddon JB, et al.
Robustness, scalability, and integration of a
42 wound-response gene expression signature in pre
dicting breast cancer survival. Proc Natl Acad Sci
U S A 2005;102:3738–43.
119. Chi JT, Wang Z, Nuyten DS, et al. Gene expres
curately than grade, tumour diameter, and oestro sion programmes in response to hypoxia: cell type
gen receptor. J Pathol 2003;201(4):555–61. 106. specificity and prognostic significance in human
Hyman E, Kauraniemi P, Hautaniemi S, et al. Impact cancers. PLoS Med 2006;3:e47.
of DNA amplification on gene expression patterns in 120. McCelland CM, Gullick WJ. Identification of
breast cancer. Cancer Res 2002;62: 6240–5. surrogate markers for determining drug activity
107. Hu Z, Fan C, Oh DS, et al. The molecular using proteomics. Biochem Soc Trans 2003;31(6):
portraits of breast tumors are conserved across 1488–90.
microarray platforms. BMC Genomics 2006;7:96.
108. Hedenfalk I, Duggan D, Chen Y, et al. Gene

121. Chevallier B, Mossen V, Dauce JP, et al. A
prognos tic score in histological node negative
breast cancer. Br J Cancer 1990;61:436–40.
122. Brown JM, Benson EA, Jones M. Confirmation of
a long term prognostic index in breast cancer.
Breast 1993;2:144–7.
123. Balslev I, Axesson CK, Zedelev K, et al. The
Nottingham Prognostic Index applied to 9,149 pa
tients from the studies of the Danish Breast
Cancer Cooperative Group (DBCG). Br Cancer
Res Treat 1994;32:281–90.
124. Ravdin PM, Siminoff LA, Davis GJ, et al.
Computer programme to assist in making
decisions about ad juvant therapy for women with
early breast cancer. J Clin Oncol
2001;19:980–91.
125. Goldhirsch A, Glick JH, Gelber RD, et al. Meeting
highlights: international expert consensus on the
primary therapy of early breast cancer 2005. Ann

Oncol 2005;16:1569–83.
126. Wishart GC, Azzato EM, Greenberg DC, et al.
PREDICT: a new UK prognostic model that pre dicts
survival following surgery for invasive breast cancer.
Breast Cancer Res 2010;12(1):R1.
127. Robertson JF, Ellis IO, Pearson D, et al.
Biological factors of prognostic significance in
locally advanced breast cancer. Breast Cancer
Res Treat 1994;29(3):259–64.
128. Robertson JF, Dixon AR, Nicholson RI, et al.
Confirmation of a prognostic index for patients
with metastatic breast cancer treated by endo
crine therapy. Breast Cancer Res Treat 1992;
22(3):221–7.
129. Ellis IO, et al. Improvement of the Nottingham
Prognostic Index (NPI +). Cancer Res (In Press).
43
3
Nipple discharge and the role of ductoscopy in
breast diseases
Sarah Tang
Gerald Gui
with nipple discharge demonstrated an overall inci
dence of underlying breast malignancy of 18.7%,3
Introduction which is higher than the figure reported in many
individual series.
After breast lumps and mastalgia, spontaneous Bilateral multiduct discharge can be stimulated
nipple discharge forms the next most common by nipple manipulation in the majority of premeno
symptom, comprising 3–8% of referrals to symp pausal women. The production of such physiologi
tomatic breast clinics.1,2 The majority of nipple cal fluid is exploited by researchers of the
discharge is benign, with up to 20% being associ intraductal approach for cytological, molecular
ated with an underlying malignancy.3 Conventional and protein based studies. This fluid is more likely
methods of investigating nipple discharge include to be released from the ducts if the nipples are
mammography, ultrasound and smear cytology, repeatedly stimu lated by chafing against clothing
each of which have recognised limitations. and during physical activities such as jogging.
Standard operations such as microdochectomy Squeezing of the nipples to elicit such discharge
and major duct excision are undirected perpetuates symptoms. This is because the
procedures that carry a risk of leaving nipple ducts are plugged by keratin and repeated
undiagnosed pathology in the breast. stimulation or squeezing removes the kera tin
Mammary ductoscopy allows direct visualisa tion plugs and allows the fluid that is normally pres ent
of the duct epithelium to locate the lesion pre in the ductal tree to leak on to the surface of the
cisely and to map the three-dimensional anatomy. nipple. The fluid in physiological discharge ranges
Endoscopic instruments for diagnostic biopsy and from clear to white to yellow to green to black.
therapeutic excision are available. The ability to The most common physiological nipple
visualise normal or benign ductal structures may discharge is lactation. Ongoing milky discharge
facilitate conservative management of may occur up to 2 years following a pregnancy
symptomatic nipple discharge and enable and is a normal phenomenon. Galactorrhoea can
targeted excision of vi sualised lesions or also result from prolactin-secreting pituitary
indeterminate areas. adenomas, medication that influences the
oestrogen, progesterone or pro
lactin pathways, hypothyroidism and recreational
Causes of nipple discharge drugs such as marijuana. Commonly prescribed
medical drugs that may cause nipple discharge
The causes of nipple discharge are wide ranging.
are summarised in Table 3.1.
The majority of discharge is physiological,
hormone related or results from benign breast
change. A re cent meta-analysis that included over
3000 women
44
Nipple discharge and the role of ductoscopy in breast diseases
Nipple discharge arising in relation to
Table 3.1 • Common drugs that mimic galactorrhea and inflammation and irritation of the lining of the
the likely underlying mechanisms of action ducts may be blood stained and contain acute
inflammatory cells.
Mechanism of action Medication
Persistent spontaneous discharge from a single
Dopamine receptor blockade nip ple orifice is usually indicative of specific
Antidepressants: pathology involving that duct. Benign intraductal
Selective serotonin papillomas account for about 80% of single-duct
reuptake bloodstained nipple discharges.2,4,5 Papillomas
inhibitors (citalopram,
give rise to nipple discharge of varying colour but
fluoxetine, paroxetine,
as these structures can bleed intermittently,
sertraline)
papillomas are often as sociated with blood
Tricyclic antidepressants
staining. Recent bleeding may manifest as frank
Antipsychotics:
Risperidone blood, but stagnant bloodstained fluid can be dark
Butyrophenones (haloperidol) red, brown or black.
Phenothiazines Periductal mastitis is a chronic and recurrent in
(chlorpromazine) flammatory condition associated with smoking in
Thioxanthenes younger women. Purulent nipple discharge can be
(chlorprothixene, flupenthixol) seen in later stages when the duct ruptures and an
Anti-emetics: abscess or mammary duct fistula has developed.
Metoclopromide Invasive breast cancer is an uncommon cause of
Domperidone spontaneous nipple discharge as the proliferating
Dopamine depletion Methyldopa mass usually obliterates the duct lumen. Ductal
Reserpine car cinoma in situ or an extensive area of in situ
Monoamine oxidase dis
inhibitors
colour range seen in physiological discharge.
Inhibition of dopamine release discharge. Bloodstained nipple
ease in association with an
Codeine Heroin discharge should be distin
invasive breast cancer may
Morphine
predispose to bloodstained nipple
Histamine receptor blockade Cimetidine
Famotidine guished from bleeding from the nipple surface that
Ranitidine may occur with Paget's disease.
tissue and lactotrophs
Stimulation of breast
Oral contraceptives
Assessment
loss of these keratin plugs and sphincter
Mechanism unknown Verapamil
relaxation include massage and warm
comfortable environments such as in a bath or un
Duct ectasia and benign breast change are com der bedsheets. Cysts do not usually communicate
mon causes of multiduct discharge. In duct directly with breast ducts. It has been suggested
ectasia, the breast ducts become tortuous and that inflammation related to a cyst may result in
dilated, pre disposing to fluid accumulation. This erosion into a duct system but this is unproven.
fluid may dis charge spontaneously or upon The colour of nipple discharge associated with
manipulation when the sphincters distal to the duct ectasia and fibrocystic change can vary from
lactiferous sinuses relax and the keratin plugs are clear to white, yellow, grey, green, brown or black.
displaced. Common situ ations that may trigger
This is the same further investigation as appropriate and
consideration of diagnostic surgical excision.
A thorough breast problem orientated history is
taken including a drug history, and a complete clini Bilateral physiological multiduct discharge aggra
cal examination should also be performed. vated by manipulation in younger patients usually
resolves when the triggering factors are removed.
Pathological nipple discharge is considered to be Bilateral and profuse milky discharge in the
discharge arising from a single duct that is younger population should be investigated by
persistent (defined as more than twice per week).
At age over 50 years, the presence of blood in the measuring
discharge and the presence of a clinical lump
increase significantly the risk of associated
malignancy,3,6 and it is recommended that these
patients are fully investigated by conventional 45
imaging techniques. Normal imaging should direct
Chapter 3 have underlying malignancy.7
Ultrasound can reliably diagnose duct ectasia
serum prolactin and if elevated, supplemented by and can identify discrete intraductal lesions.
magnetic resonance imaging (MRI) of the pituitary Papillomas below a threshold size of 1–2mm may
gland. Thyroid function tests may be indicated if not be vis ible on ultrasound imaging.
there are appropriate clinical features. Ultrasound-guided core biopsy can be used to
The colour of nipple discharge is not a reliable obtain a tissue diagnosis of any lesions visualised
method of distinguishing between physiological, but papillary lesions are of ten assessed as an
be nign or malignant aetiologies. indeterminate B3 lesion and so usually still
require excision for a definitive tissue diagnosis. A
However, a recent meta-analysis found vacuum-assisted ultrasound-guided mammotome
a higher incidence of underlying malignancy in biopsy can be both diagnostic and therapeutic.
patients with bloodstained nipple discharge Fine-needle aspiration of papillary le sions is
compared to those with non-bloody discharge (of usually unhelpful as cytology rarely resolves
any colour description; odds ratio (OR) = 2.27,
diagnostic uncertainty.
95% confidence interval (CI) = 1.32–3.89) or when
compared to those with serous discharge alone Ductography is less widely used but is an inves
3
(OR = 2.49, 95% CI = 1.25–4.93). It is therefore tigation that gives clear delineation of the breast
recommended that all patients with bloodstained anatomy. A small amount of radio-opaque contrast
nipple discharge are investigated fully.
Mammography is indicated in women over the

46
age of 35 years and may demonstrate a mass or
micro calcification in association with nipple
discharge of malignant aetiology. Even in younger
women with bloodstained discharge, is instilled into a cannulated nipple duct. Accurate
mammography can demon strate calcification in detection of small filling defect(s) caused by pap
women with widespread duc tal carcinoma in situ illomas and duct narrowing or obstruction from
(DCIS). Other mammographic features may also malignant change can be recognised. However,
be evident in patients with be nign nipple duc tography does not provide a tissue diagnosis
discharge. Duct ectasia may occasion ally be and the precise position of the area of interest
visible on mammography. The sensitivity of within the breast is not usually available to the
mammography in detecting pathology in nipple surgeon, although it is possible to target the area
discharge is 57.1% (positive predictive value of abnor mality with a wire localisation procedure.
(PPV) of 16.7% and negative predictive value The use of MRI to evaluate the ductal tree is
(NPV) of 91.4%).7 In a study of 306 patients with gaining inter est but is not a standard investigation
nipple discharge who had normal mammography of nipple dis charge. In a comparative study of 163
and ul trasound, 10% were subsequently found to patients with nipple discharge who had normal
mammography and ultrasound, ductography was
found to have a PPV of 19% and an NPV of 63%. enhanced techniques of obtaining nipple fluid
MRI was per formed in 52 patients and found to such as ductal lavage to increase the cell yield,
have a PPV of 56% and an NPV of 87%.7 which can improve the performance of cytological
Smear cytology of nipple discharge may show assessment.
the presence of malignant cells, papillary cells, red
blood cells, benign duct epithelium or foam cells of
macrophage origin. The available literature dem
Conventional surgery
onstrates sensitivities of smear cytology ranging The indications for surgical intervention are persis
from 11.1% to 16.7%8,9 and specificities ranging tent (defined as >2 per week) single-duct
from 66.1% to 96.3%. In Simmons et al.'s study of discharge in a woman over the age of 50 years,
108 pathological specimens for nipple discharge, guided further by the results of abnormal
the PPV of smear cytology was 50% and the NPV investigations. Pressure over a discrete lesion
was 76.5%.8 These data arise from an era before such as a symptomatic pap illoma often results in
mammary ductoscopy when surgery providing the duct-specific discharge. The target lesion may be
pathology used as the gold standard could not be skin marked or wire localised if
targeted, so some of the pathologies present may impalpable, to ensure excision of the target lesion.
have been missed at surgery, hence lowering the Current surgery for pathological nipple discharge
PPV of cytology. A further limitation arises when consists of major duct excision and
insufficient epithelial cells are available for a reli microdochectomy.
able diagnosis. This may be overcome by

techniques.
Major duct excision in the absence of an imaging Despite the limitations, these operations do form
ab normality is an undirected operation where a the current standard surgical practice. Surgery to
disc or wedge of retroareolar tissue is excised. the breast ducts should be avoided in women of
The amount of breast tissue that is excised is at childbearing age who wish to breast feed, as surgi
the discretion of the surgeon's clinical judgment cal injury or scarring are likely to lead to restricted
but often includes a considerable amount of or failed lactation.
tissue that does not need to be resected for a
benign process. The greater the amount of tissue Mammary ductoscopy and
resected, the higher the risk of sur gical
complications including altered sensation, hae nipple discharge
moseromas, poor aesthetic outcome from volume
loss and nipple ischaemia. Microdochectomy aims Spontaneous nipple discharge lends itself well to
to resect a single duct guided by passing a mammary ductoscopy (MD) as the ductal orifice is
lacrimal probe down the offending duct system. readily identified and coexisting duct ectasia can
The branch ing system of the ductal tree makes aid scope passage. Positive endoscopic findings
this procedure imprecise and the surgeon is blind can be used to guide surgery resulting in targeted
throughout the operation as to the location, depth excision. Visualisation of remaining ducts can rule
and number of lesions that form the target of the out coex
resection. The risk of perforating the cannulated isting pathology and avoid unnecessary resection
duct with the probe is high and if the surgeon of normal parenchyma. MD is not widely available
performs a wide resection around the probe then and its role as an established procedure remains
it in effect becomes a much bigger procedure. to be defined.
Proximal lesions further than 3cm from the nipple MD by an experienced endoscopist can be more
base are likely to be missed by these undirected accurate than mammography in identifying exten
sive in situ disease (65% vs. 50%).10 Ling et al.
found that MD-guided intraductal biopsy had a
significantly higher papilloma detection rate than The duct orifices are identified by nipple fluid ex
undirected miocrodochectomy (92.6% vs. 40.7%, pression by warming, massage or gentle pressure.
P<0.05).11 Normal findings at MD may endorse a The orifice may be dilated and some systems
conservative approach or alternatively facilitate a require a working shaft to be inserted, through
recommendation for duct transaction alone to man which the scope is passed.
age symptomatic nipple discharge.
c
History and technology
Microendoscopic technology has advanced and
earlier limitations of poor optical resolution and
problems with access because of the large calibre
of scopes have been overcome. The development
of working channels within microendoscopes has d
made it possible to biopsy lesions under direct b
vision using cytology brushes and microbiopsy
forceps. Accessories for localisation, such as a
self-retaining hookwires or rhomboids that are im
mobilised by expansion within the duct, can be
e
passed down the working channel to demarcate
the area of interest for open therapeutic excision.
Current scopes can be flexible or rigid, with di Figure 3.1 • The dissembled components of a 0.9- mm
ameters that range from 0.7 to 1.2mm (Fig. 3.1). LaDuscope® (PolyDiagnost GmbH, Pfaffenhofen,
Microendoscopes magnify up to 60 times to pro Germany) comprising the fibreoptic scope (a),
duce high-quality images with saline for insufflation disposable two-port cannula (b), cannula sheath (c),
shifter (d) and protective metal sheath for the fibreoptic
and irrigation of the ducts. More recent develop for use during sterilisation (e).
ments include autofluorescence technology,12
which may help to distinguish between benign,
premalig nant and malignant lesions.
47
Technical considerations
Chapter 3 by the scope creates a false passage that can be
recognised by the yellow cavernous honeycomb
MD, when associated with a standard surgical appearance of adipose tissue.
procedure, is often performed under general an
aesthetic. Diagnostic MD can be performed as an
Intraduct appearances
outpatient procedure with topical local anaesthetic
applied to the nipple, followed by periareolar in There is general consensus on the intraductal ap
filtration and/or infusion down the cannulated pearances of common lesions from studies that
duct, the latter facilitating relaxation of the muscle have used histological correlation (Figs 3.2 and
sphincters.13,14 Methylene blue dye, marking wires, 3.3). Malignant lesions are more likely to display
clips and transillumination15 may be used for ori haem orrhagic characteristics, streaking, fissuring
entation, to localise lesions, guide surgical and irregularity of the ductal walls or present with
excision and facilitate pathological a lu minal mass that might obstruct the duct. In
assessment.15–17 contrast, benign non-papillomatous lesions have
Complications of MD are uncommon and include a smooth,
pain, inflammation and infection. Duct perfora tion level surface without haemorrhagic features.4,18–21
MD sensitivity at detecting papillomas is high at
96–97%11,19 with a PPV of 73%.21 Benign duct
strictures may on occasion limit MD access.21 Pathological considerations
Sensitivity for detecting DCIS and invasive
disease is lower, with reports ranging from 41% to The diagnostic sensitivity of MD can be improved
81%.19,22 The presence of an extensive by incorporating ductal lavage cytology,24 which
non-invasive compo nent associated with an generates information from the proximal ducts that
invasive cancer increases significantly the may be out of reach of the ductoscope. In a study
likelihood of abnormal ductoscopic findings (71% of 415 women, the PPV of 80% of MD alone in
detecting DCIS was improved to 100% by the addi
vs. 16%).23
tion of lavage cytology.25 The cell count can be fur
ther enhanced using fine brushes passed down
the working channel to exfoliate intraluminal
lesions under direct vision, yielding up to 33000
cells.26,27 A cell-rich endoscopy specimen allows
better dis crimination between benign cells, mild to
severe atypia and malignant cells. A normal duct
on MD is infrequently associated with abnormal
cytology.14
Reliable methods of intraductal biopsy are es
sential if MD is to achieve a histological diagnosis
Figure 3.2 • A still image from mammary
ductoscopy showing a large papilloma within a duct without surgical excision. Various techniques have
with otherwise normal epithelium. been developed using vacuum-assisted biopsy
sys tems, endobaskets and grasping clips.11,23,28
The intraductal breast biopsy (IDBB) system
48 pioneered by Hunerbein and Matsunaga consists
of a 0.7-mm gradient index endoscope covered
by an external metal sheath containing a side
opening aperture near its tip.23,28 This technique
yields diagnostic material in 89–92% of cases
with a sensitivity of 76.2% and a specificity of
100% for papillomas.10,23 Using bi opsy clips, Ling
et al. were able to produce diagnos tic samples in
90% of patients.11 Biopsy techniques can be
extended to perform therapeutic endoscopic
papillomectomy.29,30 In two studies, IDDB stopped
symptomatic nipple discharge with therapeutic ef
ficacies of 77.6% and 95.4%.28,29 Carcinomas are
associated with a lower diagnostic biopsy yield
Figure 3.3 • A still image from mammary (58.3%).28 It is postulated that carcinomas are
ductoscopy showing DCIS and invasive disease in
branching ducts.

generally located more peripherally in the terminal

ducts and can be difficult to access by MD.28
Limitations
In the context of nipple discharge, the majority of
le sions are found within 5cm of the nipple but up
cancer
to 37% may be proximal and therefore out of the
reach of a 6-cm ductoscope.15,21 Unlike other MD as an adjunct to breast conservation sur gery
applications for microendoscopy such as the for cancer is technically feasible, with high rates
salivary gland, where there is a sizeable of tumour visualisation.13 In a subsequent
single-duct orifice, breast ducts open via multiple publication, Dooley successfully identified the
tiny orifices at or just below the nipple surface. malignant lesion on MD in 150 out of 201 patients
There is variability in the number of duct sys tems, (74.6%), found additional lesions out side the
with one anatomy study demonstrating 29 ducts planned excision site in 83 cases (41%) and
arising from 15 orifices.31 This study also found decreased the positive margin rate from 23.5% to
that some ducts branch within the nipple itself and 5%.32 Louie et al. reported lower tu mour
may not be accessed readily by lavage cannulae visualisation rates in 6 of 14 (43%) pa tients and
or MD. did not show any benefit of MD in reducing
positive margin rates.22 The role of duct
endoscopy in breast cancer management is cur
Key points rently limited outside clinical trials.
Ductoscopy and breast
• Nipple discharge is the third most common presenting symptom to breast

clinics. • More than 80% of nipple discharge is benign.
• Mammography and ultrasound have a low sensitivity and specificity for diagnosing the cause
of nipple discharge.
• Nipple smear cytology has a low sensitivity and positive predictive value.
• The risk of an underlying malignancy is increased if the nipple discharge is spontaneous, single
duct, persistent (occurs more than twice per week), bloody, age >50 years and associated
with a palpable abnormality.
• Mammary ductoscopy allows direct visualisation of the ductal epithelium and can be used to
direct surgical resection of target lesions while limiting resection of normal breast tissue. • The
development of microbiopsy forceps, cytology brushes and intraductal localisation wires has
enhanced the role of mammary ductoscopy as an interventional diagnostic and therapeutic
procedure.
malignancy when compared to serous or coloured
discharge.
References 4. Kapenhas-Valdes E, Feldman SM, Cohen J-M, et

al. Mammary ductoscopy for evaluation of nipple
dis charge. Ann Surg Oncol 2008;15:2720–7.
1. Baitchez G, Gortchev G, Todorova A, et al.
5. Khan SA, Mangat A, Rivers A, et al. Office ductos
Intraductal aspiration cytology and galactography
copy for surgical selection in women with pathologic
for nipple discharge. Int Surg 2003;88:83–6.
nipple discharge. Ann Surg Oncol 2011;18:3785–90.
2. Okazaki A, Hirata K, Okazaki M, et al. Nipple 6. Dolan RT, Butler JS, Kell MR, et al. Nipple
discharge disorders: current diagnostic manage discharge and the efficacy of cytology in
ment and the role of fibre-ductoscopy. Eur Radiol evaluating breast can cer risk. Surgeon
1999;9:583–90. 2010;8:252–8.
3. Chen L, Zhou WB, Zhao Y, et al. Bloody nipple dis This retrospective study of 313 patients with nipple dis
charge is a predictor of breast cancer risk: a meta charge found that duct cytology was diagnostic of only
analysis. Breast Cancer Res Treat 50% of underlying breast carcinoma but identified four
2012;132(1):9–14. This is the first meta-analysis of risk factors as having a significant association with breast
carcinoma: (a) age >50 years (P<0.0001); (b) bloody
bloodstained nipple dis charge in relation to breast cancer
nipple discharge (P< 0.008); (c) presence of a breast lump
incidence. It showed a significant association between
(P<0.0001); and (d) single-duct discharge (P<0.006). This
bloodstained nipple discharge and underlying
provides evidence in line with current expert opinion and
contributes to the strong recommendation made in
conjunction with Ref. 3.
49
Chapter 3
7. Morrogh M, Morris EA, Liberman L, et al. The pre
dictive value of ductography and magnetic
resonance imaging in the management of nipple
discharge. Ann Surg Oncol 2007;14:3369–77.
8. Simmons R, Adamovich T, Brennan M, et al.
Nonsurgical evaluation of pathologic nipple dis
charge. Ann Surg Oncol 2003;10:113–6.
9. Kooistra BW, Wauters C, van de Ven S, et al. The
diag nostic value of nipple discharge cytology in
618 con secutive patients. Eur J Surg Oncol
2009;35:573–7.
10. Dobowy A, Raubach M, Topalidis T, et al. Breast
duct endoscopy: ductosocpy from a diagnostic to
an interventional procedure and its future
perspective. Acta Chir Belg 2011;111:142–5.
11. Ling H, Liu GY, Lu JS, et al. Fiberoptic ductoscopy 50
guided intraductal biopsy improve the diagnosis of
nipple discharge. Breast J 2009;15:168–75.
12. Jacobs VR, Paepke S, Schaaf H, et al. 20. Rose C, Bojahr B, Grunwald S, et al. Ductoscopy
Autofluorescence ductoscopy: a new imaging tech based descriptors of intraductal lesions and their
nique for intraductal breast endoscopy. Clin Breast his topathologic correlates. Onkologie
Cancer 2007;8:619–23. 2010;33:307–12.
13. Dooley WC. Endoscopic visualization of breast 21. Moncrief RM, Nayar R, Diaz L, et al. A comparison
tumors. JAMA 2000;284:1518. of ductoscopy-guided and conventional surgical ex
14. Dooley W, Francescatti D, Clark L, et al. Office cision in women with spontaneous nipple
based breast ductoscopy for diagnosis. Am J Surg discharge. Ann Surg 2005;241:575–81.
2004;188:415–8. 22. Louie LD, Crowe JP, Dawson AE, et al.
15. Dooley WC. Routine operative breast endoscopy Identification of breast cancer in patients with
for bloody nipple discharge. Ann Surg Oncol pathologic nipple discharge: does ductoscopy
2002;9:920–3. predict malignancy? Am J Surg 2006;192:530–3.
16. Escobar PF, Baynes D, Crowe JP. Ductoscopy 23. Hunerbein M, Dubowy A, Raubach M, et al.
assisted microdochectomy. Int J Fertil Womens Gradient index ductoscopy and intraductal biopsy of
Med 2004;49:222–4. intra ductal breast lesions. Am J Surg 2007;194:511–4.
17. Zhu X, Xing C, Jin T, et al. A randomized controlled 24. Denewer A, El-Etribi K, Nada N, et al. The role and
study of selective microdochectomy guided by duc limitations of mammary ductoscopy in man
toscopic wire marking or methylene blue injection. agement of pathologic nipple discharge. Breast J
Am J Surg 2011;201:221–5. 2008;14:442–9.
18. Shen KW, Wu K, Lu JS, et al. Fiberoptic ductos 25. Shen KW, Wu J, Lu JS, et al. Fiberoptic ductoscopy
copy for patients with nipple discharge. Cancer for breast cancer patients with nipple discharge.
2000;89:1512–9. Surg Endosc 2001;15:1340–5.
19. Matsunaga T, Ohta D, Misak T, et al. Mammary duc 26. Khan SA, Baird C, Staradub VL, et al. Ductal
toscopy for diagnosis and treatment of intraductal lavage and ductoscopy: the opportunities and the
lesions of the breast. Breast Cancer 2001;8:213–21. limita tions. Clin Breast Cancer 2002;3:185–91.
27. Johnson-Maddux A, Ashfaq R, Cler L, et al.
Reproducibility of cytologic atypia in repeat nipple
duct lavage. Cancer 2005;103:1129–36.
28. Matsunaga T, Kawakami R, Namba K, et al.
Intraductal biopsy for diagnosis and treatment of in
traductal lesions of the breast. Cancer 2004;39:863.
29. Bender O, Balci F, Yuney E, et al. Scarless en duct anatomy in the human nipple: three dimen
doscopic papillectomy of the breast. Onkologie sional patterns and clinical applications. Breast
2009;32:94–8. Cancer Res Treat 2007;106:171–9.
30. Kamali S, Bendoer O, Aydin MT, et al. Ductoscopy 32. Dooley WC. Routine operative breast endoscopy
in the evaluation and management of nipple dis during lumpectomy. Ann Surg Oncol
charge. Ann Surg Oncol 2010;17:778–83. 2003;10:38–42.
31. Rusby JE, Brachtel EF, Michaelson JS, et al. Breast
4
Breast-conserving surgery: the balance
between good cosmesis and local control
J. Michael Dixon
women presenting symptomatically to breast clin
ics and those who are diagnosed through
screening programmes have small breast
Introduction cancers, which are suitable for breast-conserving
surgery.
The aim of local treatment of breast cancer is to
achieve long-term local disease control with the
The major advantages of breast-conserving
minimum of local morbidity. The majority of
treatment are as follows:
women where data were available: the relative
• breast-conserving treatment produces an
acceptable cosmetic appearance in the majority of risk reduction for mas tectomy versus
women with breast cancer;1 breast-conserving surgery was 0.04 (95% CI
• breast-conserving treatment results in lower levels −0.04 to +0.12).4
of psychological morbidity, with less anxiety and
depression and improved body image, sexuality
Originally it was thought that local therapy had
and self-esteem, compared with mastectomy;2,3 little influence on overall survival but it is clear that
• two systematic reviews have shown equivalence local failure is responsible, at least in part, for some
in terms of disease outcome for breast-conserving patients developing metastatic disease.6,7
treatment and mastectomy.4,5
It is thus important in patients selected for breast
One of these reviews (search date 1995) anal conserving surgery to minimise local recurrence
ysed data from six randomised controlled trials while at the same time achieving a good cosmetic
that compared breast conservation treatment with outcome.
mastectomy.4 A meta-analysis of data from five of
these six trials involving 3006 women found no sig
nificant difference in the risk of death at 10 years
Selection of patients for
(odds ratio 0.91, 95% confidence interval (CI) breast conservation
0.78–1.05). The sixth randomised trial used differ
ent protocols. In the second systematic review, Traditionally, single cancers clinically measuring
nine randomised controlled trials involving 4981 4cm or less, without signs of local advancement,
women have been managed by breast-conserving
randomised to mastectomy or breast-conserving treatment (Box 4.1). Different units have different
treatment were included in the analysis.5 A meta size crite
analysis of these nine trials found no significant ria and many units have a tumour size cut-off for
difference in the risk of death over 10 years: the breast-conserving surgery of 3cm or less clinically.
relative risk reduction for breast-conserving sur
gery compared with mastectomy was 0.02 (95%
CI −0.05 to +0.09).5 There was also no difference 51
in the rates of local recurrence in the six
randomised controlled trials involving 3006
Chapter 4 *None of these are absolute contraindications.
†
Following a fully informed discussion of the pros and cons
of breast-conserving surgery vs mastectomy.
Box 4.1 • Indications and contraindications
for breast-conserving surgery
Indications
• T1, T2 (<4 cm), N0, N1, M0
Increasing tumour size does not equate with
• T2>4 cm in large breasts
increasing local recurrence rates and limiting breast
• Single clinical and mammographic lesion conserving surgery to cancers below a certain size
Relative contraindications* is illogical.
• T4, N2 or M1
• Patients who prefer mastectomy† Clinical tumour size overestimates actual tumour
• Clinically or radiologically evident size. There is a much better correlation between
multifocal/multicentric disease pathological tumour size and the size measured
• Collagen vascular disease on imaging, with ultrasound assessment being
• Large or central tumours in small breasts more accurate than mammographic
• Women with a strong family history of measurements.8
breast cancer or who are proven BRCA1 Magnetic resonance imaging (MRI) appears better
and BRCA2 mutation carriers than ultrasound in assessing disease extent,
particu larly in invasive lobular carcinoma.9 The they were reported to have a high reported
problem with MRI is that it has a low specificity incidence of in-breast recurrence14,15 and so have
and a low positive predictive value in that only usually been treated by mastectomy, combined in
two-thirds of le sions identified by MRI as appropriate patients with immediate
suspicious of malignancy are subsequently reconstruction. Recent evi dence has, however,
confirmed as malignant.10 The role of MRI in demonstrated similar rates of local recurrence for
assessing patients for breast-conserving surgery both patients with unifocal and with multifocal and
has been investigated in a randomised study and even multicentric disease.16,17 If it is feasible to
the conclusions of this study were that routine use excise the separate cancers in differ ent parts of
of MRI is not worthwhile.8,11 MRI did not re duce the breast and produce an acceptable cosmetic
the rate of incomplete excisions and was not outcome then such patients should no lon ger be
associated with a reduction in short-term local treated routinely by mastectomy. Satisfactory
recurrence but did significantly increase the mas rates of local control following breast-conserving
tectomy rate in patients who were otherwise treatment for multifocal or multicentric cancers are
consid ered good candidates for breast-conserving achieved providing all disease is excised to clear
surgery. It is the balance between tumour size as margins.16,17 Early studies on multifocal and
assessed by imaging and breast volume that multicentric cancers often failed to achieve clear
determines whether margins and this explains the high rates of local
recurrence reported in these early series.15
Patients with bilateral cancers can also be treated
52 by bilateral breast conservation. The rates of
breast-conserving surgery vary significantly
between countries and within countries. These
rates are clearly influenced as much by the views
a patient is suitable for breast-conserving surgery
of the surgeon treating the patient as the
rather than tumour size per se. Options for
availability of radiotherapy locally. Failure to offer
patients with tumours considered too large,
breast-conserving surgery to suit able and
relative to the size of the breast, for
appropriate patients has become a medico legal
breast-conserving treatment include neoadjuvant
issue. If a patient who fulfils the criteria for
systemic therapy to shrink the tumour, an
breast-conserving surgery is treated by
oncoplastic procedure (see Chapter 6) involving
mastectomy then the exact reasons for the
either transfer of tissue into the breast or
decision to proceed to mastectomy should be
remodelling of the breast with surgery to the oppo
recorded legibly in the patient's notes. Some
site breast to obtain symmetry (see Chapter 6).12,13
patients choose mastectomy in preference to
In a patient with small breasts, excision of even a breast-conserving surgery but do so based on an
small tumour may produce an unacceptable inadequate understanding that out comes for the
cosmetic result.
two treatments are identical. In one series of
Patients with multiple tumours in the same breast patients choosing mastectomy rather than
have not previously been considered good candi
dates for breast-conserving treatment because

Breast-conserving surgery: the balance between good cosmesis and local control
influenced surgeons selecting patients for breast
breast-conserving surgery, over half of patients did conserving surgery because of their perceived im
not know that mastectomy and breast-conserving pact on local recurrence. These include young age
surgery produces identical rates of survival.18 (under 35–39 years), the presence of an extensive
A range of clinical and pathological factors have in situ component associated with an invasive
tumour, grade 3 histology and widespread
lymphatic/vascular invasion. These are margin of normal surrounding breast tissue.
considered in detail below. Controversy has surrounded which incisions give
the best cosmetic results. The predominant
orientation of collagen fibres in the skin was de
Breast-conserving surgery scribed by Langer25 and these skin crease lines
around the breast are essentially circular (Fig. 4.1).
Two surgical procedures have been studied exten
Subsequent work by Kraissl26 demonstrated that
sively: quadrantectomy and wide local excision.
the lines of maximum resting skin tension run in a
Quadrantectomy was based on the belief that the
more transverse orientation across the breast
breast is organised into segments, with each seg
(Fig. 4.1). In general, scars that are parallel both
ment draining into its own major duct, and that in
to the lines of maximum resting skin tension and
vasive cancer spreads down the duct system
to the orientation of collagen fibres are quickest to
towards the nipple.19 Both of these premises are
heal and produce the best cosmetic outcomes,
incorrect.
with the lowest rates of scar hypertrophy and ke
loid formation.
A single major subareolar duct does not
drain a localised segment of tissue but can drain
widespread areas of the breast. Incisions that follow the lines of maximum
resting skin tension produce the most cosmetically
The effectiveness of quadrantectomy relates to acceptable scars.
the large amount of tissue excised around the tu
mour rather than to the removal of a cancer and A knowledge of Langer's lines and Kraissl's lines
its draining duct. In an early randomised study of thus allows a surgeon to make incisions that en
lumpectomy or quadrantectomy, a significantly hance the cosmetic outcome of breast-conserving
greater number of patients who had lumpectomy surgery. It has been tradition to place an incision
had incomplete local excisions.20 Not surprisingly, to excise a cancer directly over the lesion, but this
therefore, local recurrence was more common can result in an unsightly scar, particularly if the
after lumpectomy than after quadrantectomy, cancer is high and medial. In such instances plac
although survival was no different.20 Other
non-randomised studies have shown similar rates
of local recur
rence for both quadrantectomy and wide local ex
cision, providing margins of excision are clear.21
Quadrantectomy or segmental excisions are no
longer appropriate breast-conserving options be
cause they produce identical rates of local recur
rence to wide excision but produce a significantly
poorer cosmetic outcome compared with wide
excision.22
Patients having breast-conserving surgery are

adequately treated by wide local excision and do
not require either a segmental or quadrantic ab
excision.23,24 Figure 4.1 • The direction of Langer's lines (a) and
Special technical details: wide lines of maximum resting skin tension in the breast (b)
(so called dynamic lines of Kraissl).
local excision
The aim of wide local excision is to remove all
invasive and any ductal carcinoma in situ with a 53
Chapter 4
ing the scar some distance below in the skin
crease lines and tunnelling up to the lesion cision through the breast should be approximately
produces a bet ter cosmetic result. Cancers close
to the nipple and even those some distance away
in the upper half of the breast can be excised 54
through circumareolar incisions. A cancer low in
the breast close to the in framammary fold can
similarly be excised through an incision placed in
1cm beyond the limit of the palpable mass. Having
the fold. A cancer in the up per outer quadrant can
incised through the breast tissue, dissection contin
be excised easily through an axillary incision.
ues under the cancer. In the majority of patients it
Excising skin directly overlying a cancer is only
is necessary to divide the whole thickness of
necessary if the skin is involved. It is not
breast tissue down to the pectoral fascia, to
necessary to remove dimpled or tethered skin.
ensure that there is an adequate margin of tissue
The cosmetic result after breast-conserving
removed deep to the cancer. If the lesion is
surgery is influenced by the amount of skin
superficial, and there is a significant amount of
excised, with poor results being obtained in those
breast tissue deep to the cancer, it may not be
patients who have most skin removed. In patients
necessary to remove full thickness of breast
with direct skin involvement, the aim should be to
tissue. Likewise if the lesion is deep, more tissue
minimise the skin excised and to remove only
can be left superficially on the skin flaps. Having
enough skin to get mi croscopically clear margins.
reached the deep margin, which is usually the
pectoral fascia, the breast tissue and cancer are
Routine excision of skin when performing a wide
excision cannot be justified.23
lifted from this fascia. It is not necessary to excise
pectoral fascia unless it is tethered to the tumour
or the tumour is involving it. If a carcinoma is
Limiting the length of incision is also important, infiltrating one of the chest wall muscles, then a
as longer incisions produce significantly poorer portion of the affected muscle should be excised
cos metic outcomes. A knowledge of the depth of beneath the tumour, the aim being to remove suf
the cancer within the breast provided by ficient muscle to get beyond the limits of the
preoperative or intraoperative breast ultrasound cancer. Having dissected under the cancer, the
can be valuable when planning the extent of cancer and surrounding tissue are grasped
excision. For instance, if a cancer is 2cm deep between the finger and thumb of the
within the breast, then at least 1cm of fat and non-dominant hand and excision of the cancer at
subcutaneous tissue can be left on the skin flaps; the other margins is completed. The specimen
leaving this tissue improves the cosmetic should be orientated immediately follow
outcome. Whatever incision is used, it is
ing excision with Liga-clips, sutures or metal mark
important to have discussed the position of all
ers, prior to specimen radiography and submission
scars with the patient prior to surgery.
to the pathologist.27 Metal markers or Liga-clips
Having made the skin incision, the skin and subcu
are preferred because they can be seen on
taneous fat are dissected off the breast tissue.
specimen ra diography. Routine X-ray of
Care should be taken when elevating skin not to
orientated specimens is recommended because it
remove subcutaneous fat unnecessarily as thin
has been shown to help the surgeon confirm the
skin flaps give a poor postoperative cosmetic
target lesion has been excised and allows
result. Where the cancer is close to the skin,
assessment of completeness of excision at the
hydrodissection in filtrating 1 in 400000 adrenaline
radial margins.27 If the specimen radiograph
in saline can help to separate the skin and
shows the cancer or any associated microcalcifica
subcutaneous fat from the breast tissue and
tion is close to a particular margin, then further
breast fat, and facilitates skin el evation over the
tissue can and should be removed from the mar
cancer. Skin flaps beyond the edge of the cancer
gin of concern, before being orientated and sent to
for at least 1–2cm are mobilised. This allows the
pathology.
fingers of the non-dominant hand to be placed
A number of studies have evaluated the use of
over the palpable cancer. The breast tissue is
cavity shavings and bed biopsies, but few have
then divided beyond the fingertips. The line of in
compared these with standard assessment of mar shavings routinely is that significant amounts of
gins. A minority of surgeons continue to take cav extra breast tissue can be removed, particularly if
ity shavings and bed biopsies routinely. Neither the whole cavity is shaved; this unfortunately can
has been shown to be reliable indicators of lo cal adversely affect cosmetic outcome. Most
recurrence. A major concern of taking cavity

simus dorsi miniflap.12,13 Drains are not necessary
importantly, however, centres who do not use after wide local excision and should not be used
these techniques report excellent local control routinely. They do not protect against haematoma
rates, which have continued to fall over time. 28,29 formation and increase infection rates. Breast skin
Wide excision with standard examination of mar wounds should be closed in layers with
gins thus provides sufficient information on mar gin absorbable sutures, finishing with a subcuticular
status for clinical use. Bed biopsies or cavity suture.
shavings are only of value and warranted where
there is concern at operation that one particular Staples and interrupted sutures do not produce
margin is involved. satisfactory results and are not an acceptable
New devices that use ultrasound or rely on the method of wound closure in the breast.
electrical properties of the tissue are in trials and
are being used to assess whether they can detect Complications of wide excision include hae
margin involvement and facilitate increased rates matoma, infection, incomplete excision, seroma
of com and poor cosmetic results. Haematoma requiring
plete excision at the first operation. evacuation is uncommon but occurs in approxi
Having excised the cancer from the breast, sutur mately 2% of patients. Infection requiring
ing the defect in the breast without mobilisation of treatment affects 5–10% and is more common
the breast tissue usually results in distortion of the when combined with an axillary dissection.
breast contour. Defects in the breast are best Incomplete excision rates should be in the range
closed by mobilising surrounding breast tissue 10–25%. The most common problem following
from the overlying skin and subcutaneous tissue surgery by wide local excision is a poor cosmetic
and in some patients mobilisation from the result. Factors influencing cosmetic outcome and
underlying chest wall is required. Large defects methods of avoiding this are consid ered in detail
(>10% breast volume) that are left open fill with below.
seroma, which often ab sorbs later, following
which scar tissue develops and contracts and can
result in an ugly distorted breast. Following Excising impalpable cancers
large-volume excisions, after mobilisation it is
usually possible to close the defect in the breast Impalpable lesions can be localised prior to
tissue by a series of interrupted absorbable surgery using one of a number of different
sutures. Given the success of lipofilling or techniques, in cluding skin marking, injection of
lipomodelling for improving some patients with blue dye, carbon or radioisotope, insertion of a
poor cosmetic outcomes following hooked wire, or in traoperative ultrasound.
breast-conserving surgery and radiotherapy, a Although excising an im palpable cancer is easier
number of surgeons are exploring whether if the skin incision is made directly over the
immediate lipofilling after wide local exci sion can cancer, remote incisions produce the best
improve outcomes, particularly for women with cosmetic results. Most impalpable small cancers
small breasts. Very large defects require onco should thus be approached through a cosmetically
plastic breast reshaping (see Chapter 6) or a latis placed incision. The location of the lesion can be
de termined in a number of ways: (i) the surgeon
can calculate the position of the breast lesion ters the cancer. Wires that are marked with beads
from the mammogram taken after wire insertion; or that change in diameter, or have a guide that
(ii) the ra diologist or the surgeon can mark the can be placed over the wire, help the surgeon to
skin overlying the lesion using ultrasound; or (iii) if determine exactly how far along the wire the
radioactivity has been injected to localise the lesion is situated. Ideally the hook of the wire
cancer, the surgeon can use a gamma probe to should be 1cm through the lesion rather than
localise the area of maxi mum radioactivity. After within its centre. The direc tion of the wire on the
making an appropriately sized and cosmetically preoperative mammogram is not always a reliable
placed skin incision, this is deepened. If a wire is guide to the course of the wire through the breast.
in place, then dissection con tinues towards the Once the wire is in place standard
wire in the plane between the breast and mammographic views are not always pos sible
subcutaneous fat so that the wire can be located and thus a lesion that is apparently lateral to
some distance before it enters the lesion. For
instance, if a mammographic abnormality has
been localised in the craniocaudal position, then it
55
helps to identify the wire superiorally before it en
Chapter 4 conserving surgery
the entry point of the wire may not be lateral on Until recently there was a reported large variation
the check craniocaudal film once the compression between different centres in recurrence rates
from the breast has been released. The aim is to follow ing breast-conserving surgery combined
remove the mammographic lesion with a 1-cm with whole breast radiotherapy for invasive breast
clear radio carcinoma. Over 80% of all local recurrences
logical margin and in most women to excise tissue were reported to be located adjacent to the site of
up to subcutaneous fat and down to pectoral initial excision. This is no longer true and an
fascia. As for palpable lesions, all specimens increasing percentage of ‘re
should be orientated with Liga-clips or metal
markers, or secured to an orientated grid so that
an orien tated specimen radiography can be to be a group of older patients with low-risk
performed. Radiography is best performed in an cancers (completely excised, node negative and
X-ray machine designed specifically to X-ray hormone re ceptor positive on hormone treatment)
specimens, such as a Bioptics® machine. There and women of any age whose cancers have an
have been conflicting re ports about whether extremely good prognosis (small grade 1 or
compressing the specimen af fects the incidence special type cancers that are completely excised,
of subsequent positive margins as reported by the node negative and hormone receptor positive on
pathologist. Orientated specimen radiography hormone treatment) whose rates of local
improves the rate of complete excision of recurrence without radiotherapy are acceptable.
impalpable cancers.27 Cooperation between sur Following whole-breast radiotherapy, it is possible
geon and pathologist is required so that the area to increase the local dose of radiotherapy by
of concern can be identified and assessed by the boosting the tumour bed. This reduces local re
pa thologist to ensure adequacy of excision. The currence rates, particularly in younger women, al
majority of wide excisions of palpable and im though there are cosmetic penalties associated
palpable cancers are performed under general with the use of boost.31
anaes thesia, but it is possible to perform these The rates of in-breast tumour recurrences
procedures under local anaesthesia. continue to fall over time28,29 (Fig. 4.2). Whereas a
1% an nual rate of in-breast cancer events was
formerly considered acceptable, rates are now
Factors affecting local less than 0.5% per annum. The rates of in-breast
recurrence after breast tumour events continue at a similar rate for at
least 20 years. This needs to be borne in mind prob lem in older patients (>65 years).
when considering surveil lance programmes for Recurrence is less frequent in women with large
such patients. breasts but whether this relates to the larger
excisions
Patient-related factors 0.50

l
Local recurrence following breast-conserving o

e
therapy is significantly more common in younger

l
patients.32–35 e
n
In contrast, local recurrence is much less of a

v
e therapy delivered most
i
r
t
0.25 0.00
r
currences’ in treated breasts are m

r
actually second pri u
a
to the whole breast in a dose of p<0.00001
u
l
c
u
4000–5000cGy given over 3–5
e
weeks continues to be used in
maries.25 Megavoltage radiation 02468 10
reduces the rate of whether localised 1993-1998 1999-2006
patients after 5y 6.5 4.4 3.9 1.7
breast-conserving local recurrence and radiotherapy
10y 12.4 9.2
surgery because ra improves30overall delivered 6.2
survival. Studies 4.1
diotherapy both
continue to evaluate 1981-1989 1990-1992
either during or within a few days of surgery is as
Figure 4.2 • Local recurrence rates in Edinburgh over
effective as whole-breast radiotherapy. As yet it four separate time periods showing a significant and
has not been possible to identify groups of continued fall in local recurrence rates over time. (Data
patients who do not require radiotherapy. unpublished, courtesy of Gill Kerr, Edinburgh Cancer
However, there is likely Centre.)
56

population.37,38
that can be performed in these patients or to
alterations in steroid metabolism (fat is known to Tumour-related factors
be an impor tant site of conversion of androgens
to oestrogens) is uncertain.36 A family history of Tumour location, tumour size, the presence of
breast cancer, particu larly carriage of a mutation skin or nipple retraction, and the presence or ab
in one of the breast can cer genes, predisposes a sence of axillary node involvement have not been
patient to an increased rate of second primary shown consistently to predict for local recurrence
cancers in both the treated and contra lateral after breast-conserving surgery.39–42 The hormone
breast unless these women undergo a prophy receptor status of a breast cancer does not seem
lactic oophorectomy, when local recurrence rates to exert any influence on local control
fall to levels similar to those of the general rates.31–35,39–42 In-breast tumour recurrence has
been reported to be more common in human diotherapy and tamoxifen. A recent update of this
epidermal growth factor receptor (HER2)-positive study reported an exceedingly small rate of recur
cancers. rence in patients randomised to radiotherapy and
tamoxifen, and acceptably low rates of annual re
Tumour size currence in patients treated with either tamoxifen
Size is not significantly associated with local recur alone or radiotherapy alone. Higher rates of recur
rence. Only 3 of 28 series that have examined the rence were seen in patients who received neither
relationship of tumour size and occurrence have ra diotherapy nor tamoxifen. In these low-risk
shown any significant association between tumour cancers treatment by radiotherapy alone or
size in breast recurrence.43,44 A large study from tamoxifen alone may therefore produce an
Boston44 demonstrated that cancers over 4cm in acceptable rate of long term control.
size that were treated by breast conservation sur
gery had a rate of recurrence similar to that of Histological type
smaller cancers (Table 4.1). There are few data relating histological tumour
type to recurrence. Invasive lobular cancer has
Tumour grade not been reported to be associated with a higher
A number of reports have analysed the recurrence rate than so-called invasive ‘ductal’
relationship between tumour grade and local carcinoma.45–48
recurrence.
One study did suggest that patients with invasive
lobular carcinoma who developed local recurrence
Table 4.1 • Size of tumour related to local recurrence
were more likely to develop multifocal recurrence42
Size (cm) Local recurrence (%) but this has not been confirmed by others.
Patients with invasive lobular cancer appear more
0–1 21
likely than patients with no special-type tumours
1.1–2 8 to have an incomplete excision. This is explained
2.1–3 13 in part by the underestimation of tumour extent by
3.1–4 17 mam mography and ultrasound and the inability of
4.1–5 4 sur geons to feel the extent of the cancer at
Data from Eberlein TG, Connolly JN, Schnitt JS et al. operation. Patients with invasive lobular cancer
Predictors of local recurrence following conservative breast on core biopsy should be warned of an increased
surgery and radiation therapy. The influence of tumour size.
Arch Surg 1990; 125:771–9.
likelihood of posi tive margins and where the
extent of disease is not easy to assess on
The lowest rates of local recurrence are reported
mammography and/or ultrasound, MRI may be
in grade 1 tumours.
valuable.9
Although some series report a higher recurrence
Lymphatic/vascular invasion
rate in grade 3 compared with grade 2 cancers,
Increased local failure rates have been reported
this is by no means universal.31–35,39–42 The rela
in most, but not all, series in patients with histo
tive risk of local recurrence between grade 1 and
logical evidence of lymphatic/vascular invasion
grade 2/3 cancer is approximately 1.5. The British
(LVI).31–35,39–42,45–48 Of concern, the percentage of
Association of Surgical Oncology undertook a trial
that randomised patients with node-negative grade
1 or special-type cancers to no further treatment,
tamoxifen alone, radiotherapy alone or both ra 57
Chapter 4 the rate of local breast recurrence compared with
tumours with no evidence of this feature.
tumours reported to have LVI varies widely be

LVI is more common in and around the cancer of
tween different series by up to a factor of 4.
younger women (<35 years) compared with
cancers in older women (>50 years).
Carcinomas with LVI have approximately double
patients with both multifocal and multicentric
Extensive in situ component cancers.16,17
The histological factor that was initially thought to
be associated with an increased rate of local recur The most important surgical-related factor is
rence was the presence of an extensive in situ completeness of excision. Current practice is to aim
com ponent (EIC) within and surrounding an for at least microscopically disease-free margins.
invasive cancer. A tumour is defined as having Ideally, there should be a clear rim of normal tissue
(≥1mm) around the carcinoma at all radial
EIC if 25% or more of the tumour mass is
margins.28,29
non-invasive and non invasive carcinoma is also
present in the breast tissue surrounding the
invasive cancer.48 EIC was consid ered not only a Treatment-related factors
predictor of local recurrence but also of residual
disease within the breast following an incomplete Controversy has surrounded how much extra
wide excision. Early reports indicated that local tissue around a cancer needs to be removed and
recurrence rates were three to four times higher in what con stitutes an involved or positive margin.
Some studies have defined a positive or involved
cancers with EIC.31–35,39–42,45–48 The major ity of
margin as disease at the margin, others consider
these studies did not, however, take account of
disease within 1mm or even disease within 2mm
margins and the authors did not perform multivari
of the margins as involved. Conversely, negative
ate analysis. Providing clear margins are obtained,
margins or uninvolved mar gins have been
it is now known that there is no increased rate of
variably defined as no tumour at the margin, or
local recurrence in patients with EIC.49–51 There is
tumour within 1, 2, 5 and 10mm of the margin.
an in teraction between age and EIC, with younger
Whatever definition has been used, almost all
women being more likely to have EIC. It has been
studies have reported an increased rate of local
suggested that the higher frequency of EIC in
recurrence in patients with positive, non-negative
younger women might explain some of the
or involved margins. When comparing patients
increased rate of local re currence seen in
with in volved margins to those with uninvolved or
younger women.42
negative margins, the relative risk of local
Multiple tumours recurrence varies between 1.4- and
9-fold.28,29,31–35,39–42,45–47,49–54 This is despite, in
Patients with macroscopically multiple cancers
were formerly considered to have an increased almost half of these series, patients with involved
risk of lo cal recurrence compared with a patient or close margins having received higher doses of
with a uni focal cancer but patient numbers in such radiotherapy than patients with clear mar gins. In
early series on which this evidence is based were the few studies which found that margins are not
important predictors of local recurrence, the dose
small.14,15 If multifocality is identified by the
of radiotherapy delivered to the tumour bed
pathologist or there are two cancers that are
ranged from 65 to 72cGy, i.e. a dose range that is
adjacent, then acceptable local recurrence rates
effective without surgery.51–54 In one study, patients
can be obtained provided that all margins of
excision are clear of disease. This led some to with involved margins who underwent re-excision
attempt breast-conserving surgery for patients to achieve a negative margin had a zero local
with multiple separate tumours in the same recurrence rate; this compares with a 22% local
breast. Provided that the cancers are excised recurrence rate in those patients who had
completely and clear margins obtained then re-excision but still had non-negative margins
reports indicate a satisfactory rate of local control (P=0.001).50 Surveys in the UK and USA have
can be demonstrated that approximately 50% of
surgeons aim for a margin of more than 2mm,
whereas 50% of surgeons are happy with a
58 margin of 2mm or less.55 There is thus no consen
sus about what constitutes an adequate margin of
excision for breast-conserving surgery. A
systematic review of margins and local recurrence
obtained with breast-conserving surgery in
was con ducted by Singletary, who reported that excision (1 or 2mm).28
some of the lowest rates of local recurrence were
in centres that had used narrow margins of

Table 4.2 • Local recurrence rates (%) at 5 years in patients from Boston48 and Stanford50 subdivided by margin
status and the presence (EIC+) or absence (EIC−) of an extensive in situ component
Boston Stanford
EIC+ EIC− EIC+ EIC−
Margins
Positive/non-negative 37 7 21 11 Close 0 5
Negative 0 2 0 1
breast-conserving
A recent large comprehensive meta-analysis of excisions should consider their protocols. Studies
mar gins has confirmed that wider margins do not have looked at the presence of lobular carci noma
reduce local recurrence.29 Wider margins remove in situ57 and atypical ductal hyperplasia58 at the
more tissue and so impact on cosmetic outcomes. margins of excision. Neither of these features
The meta-anal ysis confirmed an increased rate of signifi cantly increases local recurrence rates and
local recurrence for both involved and close so there is no need for re-excision if the
(<1mm) margins.29 Compared with no ink on pathologist reports these features alone at any of
cancer cells a 1-mm mar gin was associated with the margins of excision. Age interacts with
a significantly lower rate of local recurrence. margins. Involved margins have their greatest
Recent data from Edinburgh show there is no impact on local recurrence in younger rather than
need to re-excise if anterior or posterior margins in older women.53
are <1mm, providing full thickness of breast tissue There is a direct interaction between EIC and
has been excised and boost radiotherapy given. margins (Table 4.2). Patients with EIC and positive
Unpublished data from Edinburgh show similar margins who proceeded to radiotherapy without
local recurrence rates whether the distance to the re-excision had a 37% local recurrence rate in a
nearest margin was in the range 1–5mm or se
5–10mm, con firming the results from the ries from Boston48 and a 21% recurrence rate in a
meta-analysis that nar row 1-mm margins are Stanford series.50 In contrast, patients with EIC
sufficient. and negative margins, on primary or re-excision,
had a zero local recurrence rate. These two
A recent meta-analysis concluded that wider studies demon
margins do not reduce rates of local recurrence. strate that patients with EIC do not have an
Incomplete excision, i.e. tumour at a margin, does,
however, result in an unacceptable rate of local increased rate of local recurrence providing the
control.28,29
A 1-mm margin is sufficient. margins of ex cision are clear of invasive and in
situ cancer.
A recent large series of patients treated by breast Patients undergoing re-excision for close or in
conserving surgery with a ≥1mm margin re ported volved margins have only a 30% incidence of re
rates of local recurrence of <0.5% per year (Fig. sidual cancer in re-excised tissue.59 More than two
4.2).56 The majority of these so-called recur rences foci of microscopic margin involvement in the origi
were in fact second primaries. In conclusion, a 1- nal wide excision appears to be associated with a
or 2-mm margin is adequate and produces accept greater incidence of residual cancer (65%)
able rates of long-term control. Any centre compared with fewer than two foci, which had a
excising margins wider than 1 or 2mm on much lower rate of residual cancer. If there was
no residual dis ease at re-excision then there was residual disease in the re-excision specimen.59
a 4% local failure rate at 4.7 years compared with Further stud ies in this area are required. If it were
a 13% failure rate in patients who had residual possible to identify a group of women who do not
disease at re-excision. Patients younger than age benefit from re-excision, this would have great
50 in this study were also more likely to have clinical utility.
disease in the re-excision speci men. The
conclusion was that the majority of pa tients who Adjuvant systemic therapy
undergo re-excision do not benefit from this
procedure. Patients with lucent breasts and a Aromatase inhibitors, tamoxifen and chemother
well-defined lesion appear particularly unlikely to apy, in the presence of radiotherapy, reduce local
benefit from re-excision, if margins are close or fo
cally positive, whereas younger women with dense
breasts are much more likely to benefit from 59
further surgery, as many of these women will have
Chapter 4
recurrence rates after breast-conserving

surgery.60,61 In the absence of radiotherapy,
aromatase inhibi tors, tamoxifen or chemotherapy
alone do not pro duce satisfactory rates of local
control apart from
in low-grade, node-negative cancers.62 The
interval between surgery and radiotherapy may
be impor tant and there are suggestions that the
rates of lo cal recurrence increase if radiotherapy
is delayed. The sequencing of radiotherapy and
chemotherapy is the subject of ongoing trials but
it appears that giving both synchronously rather
than sequentially improves local recurrence.
a
Factors influencing cosmetic
outcome after breast
conserving surgery
There is a great variation in different series in the
number of patients with good to excellent cosmetic
results after breast-conserving surgery (Fig. 4.3).
b Figure 4.3 • Examples of excellent (a) and poor (b)

cosmetic results from breast-conserving surgery and
radiotherapy.
60
The importance of a good cosmetic outcome

is based on studies that have shown a significant reduce both breasts through reduction-type
correlation between poor cosmetic outcome and incisions. Better cosmetic results fol lowing
increased levels of anxiety, depression, poor body
image, problems with sexuality and low self
breast-conserving surgery and radiotherapy are
esteem.2 obtained in medium- and moderate-sized breasts;
achieving a good cosmetic outcome can often be
a problem in small breasts.1
Patient factors
Tumour factors
There is conflicting evidence about whether age
influences cosmetic outcome, with some studies Increasing tumour size means that increasingly
claiming that older women have worse cosmetic large amounts of tissue have to be removed. The
results than younger women.1 One problem in volume of tissue excised is the most important fac
younger women is that as the patient ages the nor tor relating to cosmetic outcome and so not sur
mal contralateral breast tends to increase in size, prisingly patients with larger tumours tend to have
whereas the treated breast tends not to and some worse cosmetic results than women with smaller
treated breasts shrink. A treated breast that is sym cancers.64,65 There are some data indicating that in
metrical immediately following treatment can thus surgery for small and impalpable cancers, a dispro
become asymmetrical over time. portionately large amount of normal surrounding
There is a trend towards increased fibrosis in breast tissue is removed to ensure that all the af
larger breasts, which leads to poorer cosmetic re fected tissue is excised. The role of the surgeon is
sults than seen in smaller breasts.63 For this to excise the cancer completely in as small a
reason large-breasted women may be best volume as possible. Removing large volumes of
treated by an operation that excises their cancer tissue when excising small screen-detected
and at the same time reduces the volume of the cancers is not satis factory surgical practice.
breast (therapeutic mammaplasty). The best
option to achieve symme try in such women is to

Location of tumour
Cosmetic outcomes tend to be better if the tumour
is located in the upper outer quadrant.66 Studies
have shown that major downward displacement of
the nipple occurs when surgery is performed on tu
mours located in the inferior half of the breast. This
can be corrected at the time of initial surgery by de
epithelialising a crescentic portion of skin above the
nipple to re-centre it on the new breast mound (see
Chapter 6). If the tumour is central and the nipple–
areola complex needs to be removed, then this can
have a major effect on cosmetic outcomes.1 This is significantly increasing the rate of local recurrence
why central tumours were at one time considered compared with more peripherally situated
a relative contraindication to breast-conserving sur cancers.67 Good cosmetic out comes are possible
gery. However, studies have suggested that in some women with this ap proach (Fig. 4.4). In
excision of central cancers not directly involving women with moderate-sized breasts who have
the nipple– areola complex can be treated by wide cancers that directly involve, or are very close to
excision and nipple preservation, without the nipple, the nipple and/or areola can be
excised in continuity with the cancer and the skin
closed by a purse-string suture (Fig. 4.5). Another
The extent of surgical excision or the volume
option is to rotate a dermoglandular local flap of resected breast tissue is the most important
from the lower part of the breast to fill the de fect factor affecting cosmesis.1,64
(Fig. 4.6). This produces very satisfactory cos
metic outcomes.68 In women with central The poorer cosmetic results obtained with quadran
superficial cancers in larger breasts, the nipple tectomy, even in the most experienced hands, com
can be excised as part of a reduction-type pared with wide excision are well documented and
procedure with direct closure, or the defect are related to the much larger volumes of tissue re
caused by excising the nipple and areola can be moved by quadrantectomy.69
filled with a new island of skin developed on an
Even more critical than the volume of tissue re
inferior dermoglandular pedicle.
sected is the percentage volume of the breast
excised. There is a highly statistical correlation
between cos metic outcome and percentage
volume of the breast excised (Fig. 4.8), with
excisions of less than 10% of breast volume
generally being associated with a good cosmetic
outcome, whereas excisions over 10% often
produce a poor cosmetic result (Fig. 4.9). Where it
is clear that more than 10% of breast volume
needs to be excised in order to remove the
cancer, then consideration should be given to one
of a number of procedures that can improve the fi
nal cosmetic result. These include volume replace
ment with a myocutaneous or local lipocutaneous
Figure 4.4 • Good cosmestic results from excision of a flap,12,13 volume replacement using immediate lipo
subareolar cancer left breast via a circumareolar
incision.
filling following the tumour excision, an oncoplastic
reduction procedure (therapeutic mammoplasty),
Figure 4.5 • Result after a central excision of a
cancer and use of a purse-string suture to close the neoadjuvant drug therapy or a mastectomy with or
central defect. without immediate reconstruction.
This can only be performed if there is at least 9cm

of skin between the margin of skin excision and 61
the inframammary fold (Fig. 4.7).
Surgical factors
Chapter 4
a
nipple
and produce a satisfactory cosmetic outcome without
major breast distortion. This procedure has been called
central quadrantectomy. The nipple–areola complex is
excised and a portion of skin inferior is
marked out. An incision around the circular skin island is
made and the remaining skin around the island is
de-epithelialised. A full-thickness incision is then made in
the breast and the skin island is rotated to fill the central
b defect. Staples are useful to position the flap. When the
Skin flap is deemed to be in an optimal position, the staples
island are removed and the wound closed in two layers with
absorbable sutures. (b) Final result from a right wide local
excision Grissotti flap and nipple reconstruction.
De-epithelialised skin
Figure 4.6 • (a) How to excise a central cancer under the
c
b
d Figure 4.7 • (a) Patient prior to operation – cancer
under right nipple evident by asymmetry with right nipple flatter and right nipple higher than left. (b) Preoperative
markings showing the area around the nipple that will be excised. (c) Operative view of the island of skin that is
mobilised on a de-epithelialised inferior dermoglandular flap. (d) Final result
after radiotherapy prior to nipple reconstruction.
62
(
node biopsy, primarily because of the volume of tis
sue removed, which results in an axillary deficit, but
s
l
15
e P<0.0001
m
lo
V
Breast-conserving surgery: the balance

between good cosmesis and local control
Axillary surgery
30
also because removing all nodes increases the risk
of
10
Axillary clearance does appear to be associated

with
25
a worse cosmetic outcome compared with sentinel breast oedema.63,64,71

20
5
Figure 4.8 • Percentage of breast excised compared
with body image score. Percentage of breast excised
calculated by measuring total weight of excision and
estimating breast volume (from initial diagnostic
craniocaudal mammogram). Body image score based
on patient-administered questionnaire of 15 questions
(score runs from 15, the best possible score, to 60,
the worst and highest possible score). Data from a
series of 120 patients treated in the Edinburgh Breast
Unit.
Postoperative complications Breast-conserving surgery

0
after neoadjuvant therapy
Significant numbers of patients are diagnosed
with large or locally advanced breast cancer,
many of whom are treated initially by
neoadjuvant chemo
Development of a haematoma, seroma or postop

0 10 20 30 40 50 Body image score
60
erative infection tends to increase the chances of
the patient getting a poor cosmetic result.1
therapy or endocrine guiding principles when
40
therapy. Up to a half of performing breast
20 these patients will conservation surgery after
e subsequently become completion of
g
0 candidates for neoadjuvant therapy.
P <0.001
a
n
breast-conserving First, it is im
e
surgery. To improve portant to know the exact
complete excision rates
c
r
Excellent/good site of the tumour within
and minimise cosmetic the breast, as significant
e
P Moderate/poor
100 94% 80 deformity in these numbers of patients with
40% patients there are some
60
≤ 10% > 10% the number of re excisions that a patient can have
Breast excised to achieve com plete removal of all invasive and in
Figure 4.9 • Percentage of good/excellent results in situ disease,70 but with the greater number of
patients subdivided according to whether 10% or less re-excisions more tissue is removed and so the
or more than 10% of breast volume was excised by likelihood of a good cosmetic result decreases.
breast conserving surgery.
For patients who require multiple excisions to get
clear margins then con sideration should be given
Re-excision and number of procedures to correcting any vol ume deficit prior to delivery of
Re-excision of the tumour bed has a negative im radiotherapy or considering subsequent
pact on cosmesis.64 This is mainly as a contralateral symmetrising surgery.
consequence of the increased total volume of HER2-positive and triple negative cancers will
tissue excised from the breast. There is no limit to have a complete pathological response. All
patients un dergoing neoadjuvant chemotherapy ing neoadjuvant endocrine therapy is therefore
who might be candidates for subsequent usually successful at excising all disease in one op
breast-conserving surgery should therefore have eration and few patients have involved margins.
one or more tumour markers placed centrally prior In contrast, a significant number of patients after
to or early during treatment. The patterns of neoadjuvant chemotherapy have a diffuse pattern
response to neoadjuvant chemotherapy and of response to chemotherapy, with reduction in
neoadjuvant endocrine therapy differ.72 The most tumour cellularity but without significant shrink age
common form of pathological change in patients of volume. This diffuse tumour may be impal pable
undergoing neoadjuvant endo and more than 25% of patients undergoing
crine therapy is central scar formation,72 which
results in concentric reduction in tumour size and 63
tumour volume. Breast-conserving surgery follow
Chapter 4
breast-conserving surgery after neoadjuvant che

motherapy will have incomplete excision of dis
ease. All patients undergoing breast-conserving
surgery after neoadjuvant chemotherapy should be
warned of this rate of incomplete excision and the
possible need for a further operation.
MRI following neoadjuvant chemotherapy is the

best of the currently available imaging methods to
assess extent of disease and is the best predictor of
whether a cancer is suitable for breast-conserving
Figure 4.10 • Result from wide local excision right of
surgery.73 a cancer in the upper inner quadrant via a
circumareola
any effect on cosmetic outcome, whereas some
Radiotherapy stud ies have suggested that chemotherapy has a
negative impact on the cosmetic outcome of
Increasing doses of radiotherapy, particularly with breast-conserving surgery.1
the addition of boost, have a detrimental effect on
cosmetic outcomes.31,63,64 Long-term follow-up is
necessary to assess cosmetic outcome; 3 years af Treatment of poor cosmetic
ter treatment, radiotherapy effects tend to stabilise. results after breast-conserving
Fibrosis is a late effect of radiotherapy and pro surgery
duces breast retraction and contour distortion. The
treated breast tends not to increase in size to the Prevention is better than treatment, so minimis ing
same extent as the opposite untreated breast, so volume, closing breast defects and immediate
pa tients as they age can develop asymmetry lipofilling (Fig. 4.10) are much better options than
even when the initial cosmetic result was trying to correct asymmetry once it has developed.
excellent. Boost has a negative impact on Options for improving poor cosmetic outcomes in
cosmesis because it produces intense fibrosis clude increasing the volume of one or both breasts
and unsightly skin changes, includ ing
telangiectasia.31
Other treatment effects 64

incision with immediate lipofilling and
postoperative radiotherapy.
Tamoxifen and aromatase inhibitors have little if
of the injected fat survives. Multiple episodes of
lipofilling combined with scar release or scar
if the main problem is essentially loss of volume
excision are often required to correct significant
rather than significant radiation fibrosis. While
breast deformity and asym
placement of silicone prosthesis or prostheses has
metry (Fig. 4.11).76 If the problem is simply one of
been described, they are successful only for
patients with little or no deformity and absence of asymmetry and the treated breast is a satisfactory
shape but smaller than the normal contralateral
marked skin changes.74 Swelling following implant
breast, then the contralateral breast can be
insertion in a radiotherapy-treated breast is often
reduced. If the treated breast is shrunken,
marked and takes many months to settle. Some
misshapen and scarred, then another option is to
patients develop marked capsular contraction
excise part or the whole of the treated breast and
after implant place ment following radiotherapy but
reconstruct ei
there are no data on whether this incidence is
ther part of the breast or the whole breast.
greater than in women having augmentation with
no prior radiotherapy. The use of implants has Pedicled myocutaneous latissimus dorsi or
been largely superseded by lipofilling. Fat is transverse rectus abdominis myocutaneous
aspirated from the abdomen and thighs and (TRAM) flaps offer an opportunity to excise
unsightly areas of skin and/
centrifuged, washed or filtered.75,76
or breast distortion and scarring, and provide one
Having removed the oil and blood it is then
option to regain symmetry (Fig. 4.12).
injected as microdroplets into the area of
distortion and/or asymmetry. Approximately 40%

Pre Lipofilling After Lipofilling Scar Release + 2 nd Fill
Figure 4.11 • Patient with a defect from a previous wide local excision and radiotherapy before and after lipofilling
and then scar release and lipofilling of the left breast.
b c
Figure 4.12 • Patient with a poor cosmetic result after breast-conserving surgery before (a) and after (b,c) partial
breast reconstruction with a pedicled latissimus dorsi myocutaneous flap.
65
Chapter 4 and the aim of primary treatment is to avoid local
Significance and treatment

‘recurrences’ in treated breasts are now second
of local recurrence primary cancers. Local recurrence within the first
5 years is asso ciated with a worse long-term
Local recurrence rates of 0.5% or less per year
outlook than recurrence thereafter.7,77–79 The role
after breast-conserving treatment are now
of systemic therapy following mastectomy for an
achievable.
apparently localised breast recur rence appears to
be of benefit.79 Uncontrollable local recurrence is
An isolated local breast recurrence does
not appear to be a threat to survival, but breast
uncommon after breast conservation, but when it
recurrence is a predictor of distant disease,7,77,78 does occur it is difficult to treat.
recurrence if at all possible. Extended hormonal treatment with letrozole following completion of 5 years of
tamoxifen
local recurrence if the initial recurrence occurs
Isolated recurrences of the breast can be treated
within the first 5 years of treatment.78 Until
by re-excision or mastectomy.77–79 Re-excision
recently 80% of local recurrences in the
alone is as sociated with a high rate of subsequent conserved breast occurred at the site of the origi
nal breast cancer, with 90% of these local the rate of contralateral breast cancer
recurrences following breast-conserving surgery development.79
being invasive. This is no longer true and an
Prolonging adjuvant hormonal therapy beyond 5
increasing percentage of
years has a significant impact on the rate of subse
reduces' in breast. Local recurrence is reduced by
letrozole compared with tamoxifen given as quent local relapse in postmenopausal patients
adjuvant therapy to postmenopausal women with with hormone receptor-positive breast cancer.
ER positive breast cancer (Table 4.3). Extended
recurrences by almost two-thirds and also reduces
Table 4.3 • Efficacy end-points in 4922 patients enrolled into the BIG 1-98 trials
Letrozole (n = 2463) Tamoxifen (n = 2459)
No. of patients % No. of patients %

Disease-free survival events 352 14.3 418 17.0 Local 19 0.8 38 1.6 Contralateral breast 14
0.6 26 1.1 Regional 13 0.5 11 0.5 Distant 182 7.4 212 8.6 Deaths without cancer 60 2.4 48
2.0 Deaths (overall survival events) 194 7.9 211 8.6 Systemic failures 331 13.4 374 15.2
Modified from Coates AS, Keshaviah A, Thurlimann B et al. Five years of letrozole compared with tamoxifen as initial adjuvant
therapy for postmenopausal women with endocrine-responsive early breast cancer: update of study BIG 1–98. J Clin Oncol 2007;
25(5):486–92. Published by the American Society of Clinical Oncology.
Key points
• For patients with single breast cancers, survival outcomes from breast-conserving treatment
are equivalent to that of mastectomy.
• Radiotherapy (after breast-conserving surgery) reduces the rate of local recurrence and
improves overall survival. No subgroup of patients has yet been identified that can avoid
radiotherapy. • The major surgical factor influencing local recurrence is completeness of
excision, and clear margins (≥1mm) must be obtained when performing breast-conserving
surgery.
• Wider margins (>5mm) do not appear to achieve better local control rates than narrow
margins (≥1mm).
• Younger patients have an increased rate of local recurrence after breast-conserving
surgery. Conversely, older patients have a lower rate of local recurrence.
66
• Tumour grade, EIC and LVI have a limited influence on the rate of local recurrence. Patients
with these factors should not be denied breast-conserving surgery, providing the cancer can
be excised to clear margins.
• There is a direct correlation between cosmetic outcome after breast-conserving surgery and
psychological morbidity, with better cosmetic outcomes being associated with less anxiety
and depression and better body image and self-esteem.
• The most important factor influencing cosmetic outcome after breast-conserving surgery is the
percentage volume of breast excised. Removing more than 10% of the breast volume
dramatically increases the number of women having a poor cosmetic outcome.
• Patients who develop local recurrence after breast-conserving surgery, particularly in the
first 5 years, are at increased risk of having systemic relapse.
• Isolated local recurrences after breast-conserving surgery are usually treated by
mastectomy, although re-excision is possible, particularly if the recurrence develops more
than 5 years after treatment or the patient has not received radiotherapy to the breast.
• Prolonging hormonal therapy beyond 5 years in postmenopausal women with hormone
receptor positive breast cancer reduces the rate of subsequent ‘in-breast recurrence’ and the
rate of contralateral breast cancer development.
surgery were analysed. Meta-analysis found no sig
nificant difference in the risk of death over 10 years for
patients treated by mastectomy or breast-conserving
References surgery. The authors also found no significant difference
in the rates of local recurrence in the six trials where data
were available. Level I evidence.
1. Sharif K, Al-Ghazal SK, Blamey RW. Cosmetic as
sessment of breast-conserving surgery for primary 6. Fortin A, Larochelle M, Laverdière J, et al. Local
breast cancer. Breast 1999;8:162–8. failure is responsible for the decrease in survival
2. Al-Ghazal SK, Fallowfield L, Blamey RW. for patients with breast cancer treated with
Comparison of psychological aspects and patient conserva tive surgery and postoperative
satisfaction following breast conserving surgery, radiotherapy. J Clin Oncol 1999;17:101–9.
simple mastectomy and breast reconstruction. Eur 7. Fisher B, Anderson S, Fisher E, et al. Significance
J Cancer 2000;36:1938–43. of ipsilateral breast tumour recurrence after
3. Shain WS, d'Angelo TM, Dunn ME, et al. lumpec tomy. Lancet 1991;338:327–31.
Mastectomy versus conservative surgery and radia 8. Turnbull LW, Brown SR, Olivier C, et al. Multicentre
tion therapy: psychological consequences. Cancer randomized controlled trial examining the cost
1994;73:1221–8. effectiveness of contrast-enhanced high field
4. Early Breast Cancer Trialists' Collaborative Group. magnetic resonance imaging in women with
Effects of radiotherapy and surgery in early breast primary breast cancer scheduled for wide local
cancer: an overview of the randomised trials. N excision (COMICE). Health Technol Assess
Engl J Med 1995;333:1444–55. 2010;14(1):1–182.
This review analyses data on 10-year survival from six 9. Mann RM, Hoogeveen YL, Blickman JG, et al. MRI
randomised controlled trails comparing breast conser compared to conventional diagnostic work-up in
vation with mastectomy. Meta-analysis of data from five the decision and evaluation of invasive lobular
of the randomised trials (3006 women) found no differ carci noma of the breast: a review of existing
ence in the risk of death at 10 years. Where more than literature. Breast Cancer Res Treat
half of node-positive patients in both the mastectomy 2008;107:1–14.
and breast-conserving groups received adjuvant nodal 10. Houssami N, Ciatto S, Macaskill P, et al. Accuracy
radiotherapy, both groups had similar survival rates. In and surgical impact of MRI in breast cancer stag
contrast, where less than half of node-positive women in ing: systematic review and meta-analysis in
both groups received adjuvant nodal radiotherapy, sur detection of multifocal and multicentric cancer. J
vival was better for the breast-conserving surgery group Clin Oncol 2008;26(19):3248–58.
(overall risk vs. mastectomy 0.69, 95% CI 0.5, 90% 0.57). This is a review of the accuracy of MRI in breast can cer
Level I evidence. and shows that MRI has a high sensitivity but poor
5. Morris AD, Morris RD, Wilson JF, et al. Breast con specificity.
serving therapy versus mastectomy in early stage
breast cancer: a meta-analysis of 10 year survival.
Cancer J Sci Am 1997;3:6–12.
67
In this review nine randomised controlled trials involving
4981 women potentially suitable for breast-conserving

Chapter 4 12. Dixon JM, Venizelos B, Chan P. Latissimus dorsi
mini-flap: a technique for extending breast conser
vation. Breast 2002;11:58–65.
11. Turnbull L, Brown S, Harvey J, et al. Comparative
13. Raja MAK, Straker VF, Rainsbury RM. Extending
effectiveness of MRI in breast cancer (COMICE)
the role of breast-conserving surgery by immediate
trial: a randomized controlled trial. Lancet 2010;
vol ume replacement. Br J Surg 1997;84:101–5.
375(9714):563–71.
This is the only randomised trial of MRI in breast 14. Fisher ER, Sass R, Fisher B, et al. Pathologic find
conserving surgery. This study shows that routine MRI in ings from the national surgical adjuvant breast
breast-conserving surgery is not indicated. project (protocol 6): relation of local breast re
currence to multicentricity. Cancer 1986;57:
1717–24. early stage breast cancer treated with breast
15. Kurtz JM, Jacquemier G, Amalaric R, et al. Breast conservation therapy. Am J Surg
conserving therapy for macroscopically multiple 2002;184:383–93.
cancers. Ann Surg 1990;212:38–44. Ten-year follow-up of a cohort of patients treated by
breast conservation reporting patterns of local, regional
16. Lim W, Park EH, Choi SL, et al. Breast conserv ing and systemic recurrence. Study reports low annual rate
surgery for multifocal breast cancer. Ann Surg of breast tumour recurrence with a ≥1mm margin.
2009;249(1):87–90.
29. Houssami N, Macaskill P, Marinovich ML, et al.
17. Gentilini O, Botteri E, Rotmensz N, et al. Meta-analysis of the impact of surgical margins on
Conservative surgery in patients with multifocal/ local recurrence in women with early-stage
multicentric breast cancer. Breast Cancer Res invasive breast cancer treated with
Treat 2009;113(3):577–83. breast-conserving therapy. Eur J Cancer
18. Ballinger S, Mayer KF, Lawrence G, et al. Patients' 2010;46(18):3219–32.
decision-making in a UK specialist centre with high A meta-analysis of margins that demonstrates that wide
mastectomy rates. The Breast 2008;17:574–9. margins are unnecessary in breast-conserving surgery. It
shows that both involved and close <1mm margins are
19. Holland DR, Connolly JL, Gelman R, et al. The
associated with a significant increase in local recur
presence of an extensive intraductal component
(EIC) following a limited excision predicts for prom rence. This meta-analysis concludes that a 1-mm mar
inent residual disease in the remainder of the gin is sufficient.
breast. J Clin Oncol 1990;8:113–8. 30. Early Breast Cancer Trialists' Collaborative Group.
20. Veronesi U, Volterrani F, Luini A, et al. Effect of radiotherapy after breast-conserving sur
Quadrantectomy versus lumpectomy for small size gery on 10-year recurrence and 15-year breast can
breast cancer. Eur J Cancer 1990;16:671–3. cer death: meta-analysis of individual patient data
for 10801 women in 17 randomised trials. Lancet
21. Ghossein NA, Alpert S, Barba J, et al. Importance
2011;378(9804):1707–16.
of adequate surgical excision prior to radiotherapy
A meta-analysis demonstrating the value of radiother apy
in the local control of breast cancer in patients
after breast-conserving surgery in both reducing
treated conservatively. Arch Surg 1992;127:411–5.
recurrence and improving overall survival.
22. Sacchini V, Luini A, Tana S, et al. Quantitive and
31. Bartelink H, Horiot JC, Poortmans P, et al.
qualitative cosmetic evaluation after conserva tion
Recurrence rates after treatment of breast cancer with
treatment for breast cancer. Eur J Cancer
standard radiotherapy with or without addi tional
1991;27:1395–400.
radiation. N Engl J Med 2001;345:1378–87.
23. NIH Consensus Conference. Treatment of early
32. Kurtz JM. Factors influencing the risk of local recur
stage breast cancer. JAMA 1991;265:391–5. 24.
rence in the breast. Eur J Cancer 1992;28:660–6.
Fisher B, Wolmark N, Fisher ER. Lumpectomy and
axillary dissection for breast cancer: surgical, patho 33. Mate TP, Carter D, Fischer DB, et al. A clinical and
logical and radiation considerations. World J Surg histopathological analysis of the results of conserva
1985;96:692–8. tion surgery and radiation therapy in stage I and stage
II breast carcinoma. Cancer 1986;58:1995–2002.
25. Langer K. Zur anatomie und physiologie der Haut.
Huber Die Spaltbarkiet Der Cutis. S-b-Akad Wiss 34. Zafrani B, Viehl P, Fourqhet A, et al. Conservative
Wein 1861;44:19–46. treatment of early breast cancer: prognostic value
of the ductal in situ component and other
pathological variables on local control and sur
vival. Long term results. Eur J Cancer Clin Oncol
68 1989;25:1645–50.
35. Ryoo MC, Kagan AR, Wollin M. Prognostic factors
for recurrence and cosmesis in 393 patients after
radiation therapy for early mammary carcinoma.
26. Kraissl CJ. The selection of appropriate lines for
Radiology 1989;172:555–9.
elective surgical incisions. Plast Reconstr Surg
1951;8:1–28. 36. Chauvet B, Simon JM, Reynaud-Bougnoux A, et al.
Récidives mammaires après traitment conservateur
27. Nedelman R, Dixon JM. Marking of specimens in
des cancers du sein: facteurs prédictifs et significa tion
patients undergoing stereotactic wide local excision
pronostique. Bull Cancer 1990;77:1193–205.
for breast cancer. Br J Surg 1992;79:55.
28. Singletary SE. Surgical margins in patients with

Cancer 1996;78:1921–8.
37. Pierce L, Levin A, Rebbeck T, et al. Ten-year 50. Smitt MC, Nowels KW, Zdeblick MJ, et al. The im
outcome of breast-conserving surgery (BCS) and portance of the lumpectomy surgical margin status
radiotherapy (RT) in women with breast cancer in long term results of breast conservation. Cancer
(BC) and germline BRCA 1/2 mutations: results 1995;76:259–67.
from an international collaboration. Breast Cancer 51. Solin LJ, Fowble BL, Schultz DJ, et al. The sig
Res Treat 2003;82:S7. nificance of the pathology margins of the tumour
excision on the outcome of patients treated with de
38. Rebbeck TR, Lynch HT, Neuhausen SL, et al.
finitive irradiation for early stage breast cancer. Int
Prophylactic oophorectomy in carriers of BRCA1
J Radiat Oncol Biol Phys 1991;21:279–87.
or BRCA2 mutations. N Engl J Med
2002;346(21):1616–22. 52. Spivack B, Khanna MM, Tafra L, et al. Margin sta
tus and local recurrence after breast-conserving
39. Fowble BL, Solin LJ, Schultz DJ, et al. 10 year
sur gery. Arch Surg 1994;129:952–7.
results of conservative surgery and irradiation for
stage I and II breast cancer. Int J Radiat Oncol Biol 53. Wazer DE, Jabro G, Ruthazer R, et al. Extent of
Phys 1991;21:269–77. margin positivity as a predictor for local recurrence
after breast conserving irradiation. Radiat Oncol
40. Haffty BG, Fischer D, Rose M, et al. Prognostic fac
Investig 1999;7:111–7.
tors for local recurrence in the conservatively
treated breast cancer patient: a cautious 54. Schmidt-Ullrich R, Wazer DE, DiPetrillo T, et al.
interpretation of the data. J Clin Oncol Breast conservation therapy for early stage breast
1991;6:997–1003. carcinoma with outstanding ten year locoregional
control rates: a case for aggressive therapy to the
41. Locker AP, Ellis IO, Morgan DAL, et al. Factors influ
tu
encing local recurrence after excision and radiotherapy
mour bearing quadrant. Int J Radiat Oncol Biol
for primary breast cancer. Br J Surg 1989;76:890–4.
Phys 1993;27:545–52.
42. Kurtz JM, Jacquemier G, Amalric R, et al. Risk fac
55. Vallasiadou K, Young OE, Dixon JM. Current prac
tors for breast recurrence in premenopausal and
postmenopausal patients with ductal cancers treated tices in breast conservation surgery: results of a
by conservation therapy. Cancer 1990;65:1867–78. questionnaire. Br J Surg 2003;90:44.
43. Asgiersson KS, McCulley SJ, Pinder SE, et al. Size 56. Montgomery DA, Krupa K, Jack WJL, et al.
of invasive breast cancer and risk of local Changing pattern of the detection of locoregional
recurrence after breast-conservation therapy. Eur relapse in breast cancer: the Edinburgh
J Cancer 2003;39:2462–9. experience. Br J Cancer 2007;96:1802–7.
44. Eberlein TG, Connolly JN, Schnitt JS, et al. 57. Abner AL, Connolly JL, Recht A, et al. The rela tion
Predictors of local recurrence following between the presence and extent of lobular car
conservative breast surgery and radiation therapy. cinoma in situ and the risk of local recurrence for
The influence of tumour size. Arch Surg patients with infiltrating carcinoma of the breast
1990;125:771–9. treated with conservative surgery and radiation ther
45. Jacquemier RG, Kurtz JM, Amalric R, et al. An as apy. Cancer 2000;88:1072–7.
sessment of extensive intraductal component as a 58. Fowble B, Hanlon AL, Patchefsky A, et al. The
risk factor for local recurrence after presence of proliferative breast disease with atypia
breast-conserving surgery. Br J Cancer does significantly influence outcome in early-stage
1990;61:873–6. invasive breast cancer treated with conservative
46. Fourquet A, Campan F, Zafrani B, et al. Prognostic surgery and radiation. Int J Radiat Oncol Biol Phys
factors of breast recurrence in the conservative man 1998;42:105–15.
agement of early breast cancer: a 25 year follow up. 59. Swanson GP, Rynearson K, Symmonds R.
Int J Radiat Oncol Biol Phys 1989;17:719–25. Significance of margins of excision on breast
47. du Toit RS, Locker AP, Ellis IO, et al. An evaluation cancer recurrence. Am J Clin Oncol
of differences in prognosis, recurrence patterns 2002;25:438–41.
and receptor status between invasive lobular and 60. Rose MA, Henderson IC, Gellman R, et al.
other invasive carcinomas of the breast. Eur J Premenopausal breast cancer patients treated with
Surg Oncol 1991;17:251–7. conservative surgery, radiotherapy and adjuvant
48. Schnitt SJ, Connolly JL, Kettry U. Pathologic chemotherapy have a low risk of local failure. Int J
findings on re-excision of the primary site in breast Radiat Oncol Biol Phys 1989;17:717–21.
cancer pa tients considered for treatment by 61. Goss PE, Ingle JN, Martino S, et al. A randomized
primary radiation therapy. Cancer trial of letrozole in postmenopausal women after
1987;59:675–81. five years of tamoxifen therapy for early-stage
49. Gage I, Schnitt SJ, Nixon AJ, et al. Pathologic mar breast cancer. N Engl J Med 2003;349:1793–802.
gin involvement and the risk of recurrence in pa 62. Forrest P, Stewart HJ, Everington D, et al.
tients treated with breast-conserving therapy. Randomised controlled trial of conservative therapy
for breast cancer: 6-year analysis of the Scottish Cancer 1977;39:917–23.
trial. Lancet 1996;348:708–13.
63. Prosnitz LR, Goldenberg IS, Packard RA, et al.
Radiation therapy as initial treatment for early
stage cancer of the breast without mastectomy.
69
Chapter 4
64. Wazer DE, DiPetrillo T, Schmidt-Ullrich R, et al.
Factors influencing cosmetic outcome and
complication risk after conservative surgery and
radiotherapy for early stage breast carcinoma. J Clin
Oncol 1992;10:356–63.
65. Dewar JA, Benhamou S, Benhamou E, et al.
Cosmetic results following lumpectomy, axillary
dissection and radiotherapy for small breast
cancers. Radiother Oncol 1988;12:273–80.
66. Liljegren G, Holmberg L, Westman G, et al. The
cosmetic outcome in early breast cancer treated
with sector resection with or without radiotherapy.
Eur J Cancer 1993;29A:2083–9.
67. Haffty BG, Wilson LD, Smith R, et al. Subareolar
breast cancer: long-term results with conservative
surgery and radiation therapy. Int J Radiat Oncol
Biol Phys 1995;33:53–7.
68. Petit JY, Garusi C, Greuse M, et al. One hundred
and eleven cases of breast conservation treatment
with simultaneous reconstruction at the European
Institute of Oncology. Tumori 2002;88:41–7.
69. Amichetti M, Busana L, Caffo O. Long-term cos
metic outcome and toxicity in patients treated with
quandrantectomy and radiation therapy for early stage
breast cancer. Oncology 1995;52:177–81.
70. Coopey S, Smith BL, Hanson S, et al. The safety of
70
multiple re-excisions after lumpectomy for breast
cancer. Ann Surg Oncol 2011;18(13):3797–801.
71. Sneeuw KA, Aaronson N, Yarnould J, et al.
Cosmetic and functional outcomes of breast 72. Thomas JS, Julian HS, Green RV, et al.
conserving treatment for early stage breast Histopathology of breast carcinoma following
cancer. 1. Comparison of patients' ratings, neoadjuvant systemic therapy: a common associa
observers' ratings and objective assessments. tion between letrozole therapy and central scarring.
Radiother Oncol 1992;25:153–9. Histopathology 2007;51:219–26.
73. Manton DJ, Chaturvedi A, Hubbard A, et al.
Neoadjuvant chemotherapy in breast cancer: early
response prediction with quantitative MR imaging
and spectroscopy. Br J Cancer 2006;94:427–35.
74. Schaverien MV, Stutchfield BM, Raine C, et al.
Implant-based augmentation mammaplasty fol
lowing breast conservation surgery. Ann Plast Surg
2011; Feb 21. Epub ahead of print.
75. Delay E, Garson S, Tousson G, et al. Fat injection
to the breast: technique, results and indications
based on 880 procedures over 10 years. Aesthetic
Surg J 2009;29:360–78.
76. Rigotti G, Marchi A, Stringhini P, et al. Determining
the oncological risk of autologous lipoaspirate graft
ing for post-mastectomy breast reconstruction.
Aesthetic Plast Surg 2010;34:475–80.
77. Haffty BG, Fischer D, Beinfield M, et al. Prognosis Cancer 1991;27:240–4.
following local recurrence in the conservatively 79. Anderson EDC. Treatment of breast recurrence
treated breast cancer patient. Int J Radiat Oncol after breast conservation. In: Dixon JM, editor.
Biol Phys 1991;21:293–8. Breast cancer: diagnosis and management.
78. Kurtz JM, Jacquemier G, Amalric R. Is breast con London: Elsevier; 2000. p. 1–5.
servation after local recurrence feasible. Eur J

2014 Breast Surgery A Companion To Specialist

Uploaded by

Document Informationclick to expand document informationSurgical book for surgeons guide.

Document Informationclick to expand document information

Copyright:

Available Formats

2014 Breast Surgery A Companion To Specialist

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2014 Breast Surgery A Companion To Specialist

Uploaded by

Copyright:

Available Formats

Breast Surgery

A COMPANION TO SPECIALIST SURGICAL PRACTICE

Western General Hospital, Edinburgh, UK

Commissioning Editor: Laurence Hunter

(021)66485438 66485457 www.ketabpezeshki.com

Contributors Anees B. Chagpar, MD, MSc, MPH, MA

Nigel J. Bundred, MD, FRCS

Lesley Fallowfield, BSc, DPhil, FMedSci vii

Monica Morrow, MD, FACS

R. Douglas Macmillan, MBChB, MD, FRCS

Mark Schaverien, MBChB, MD, MSc, MEd,

Ian Smith, MD, FRCP, FRCPE

Sarah Tang, MBBS, MA, MRCS

Alastair M. Thompson, BSc(Hons), MBChB, MD,

(021)66485438 66485457 www.ketabpezeshki.com

Series Editors' preface

(021)66485438 66485457 www.ketabpezeshki.com

Evidence-based practice in surgery

(021)66485438 66485457 www.ketabpezeshki.com

MRM is the best method for screening younger

(021)66485438 66485457 www.ketabpezeshki.com

reduce overall breast cancer mortality by around Method and frequency

(021)66485438 66485457 www.ketabpezeshki.com

Criteria Minimum standard Target

To maximise the women to attend for • Incident screen

(c) Standardised detection

5–6mm detail • ≤1.2% • ≤0.8% 0.5mm detail • ≤5% •

The role of imaging in breast diagnosis

Objective Criteria Minimum standard Target

• >90% • 100% • >90% •

• 1.2 per 1000 women screened

discharged. The remaining 25% proceed to needle

Receiving an invitation for screening and attend

Table 1.3 • NHS Breast Screening Programme: results

assessment does cause measurable anxiety,

(021)66485438 66485457 www.ketabpezeshki.com

of management should be on reassurance rather

Image-guided breast biopsy

Guidance techniques for breast

(021)66485438 66485457 www.ketabpezeshki.com

b Figure 1.5 • Digital mammogram showing a Reidy

(021)66485438 66485457 www.ketabpezeshki.com

• Breast imaging is an essential part of modern multidisciplinary breast

21. Kuhl CK, Schmutzler RK, Leutner CC, et al.

(021)66485438 66485457 www.ketabpezeshki.com

59. Wilson ARM. Therapeutic applications of vacuum

(021)66485438 66485457 www.ketabpezeshki.com

TNM staging system for breast cancer has con

cancer Lymph node t

lymph node stage is considered as a

firmed that the absolute number of axillary lymph

tors in breast cancer.1 Lymph node stage has been a

factor; the longer the tumour has been growing the

o Studies show that overall survival is

more likely it is that lymph nodes are involved by

spread. It has also been reported that, when taken

Sentinel node biopsy or blue dye directed