Canadian Journal ofPathology

Publications Agreement Number 40025049 • ISSN 1918-915X

www.cap-acp.org

Best Practice Recommendations forStandardization of Immunohistochemistry Tests

Molecular Pathology of Colorectal Cancer

Canadian Journal of

PathologyOfficial Publication of the Canadian Association of Pathologists

www.andrewjohnpublishing.com

Vol.

1,Is

sue

2•

July

200

9

CONFERENCE 2009

3Canadian Journal of PathologyJuly 2009

FRI DAY J U LY 10 1900 – 2100 Registration Open

SATU RDAY J U LY 11 0730 – 1800 Registration Open

0830 – 1200 Morning Scientific Workshops W101-W105

1330 – 1700 Afternoon Scientific Workshops W201-W205

1800 – 2000 Residents’ Symposium and Dinner

SU N DAY J U LY 12 0730 – 1800 Registration Open

0830 – 1200 Morning Scientific Workshops W301-W305

1330 – 1700 Afternoon Scientific Workshops W401-W405

1900 – 2100 President’s Reception

MON DAY J U LY 13 0730 – 1800 Registration Open

0800 – 1130 S501 Symposium – Hematopathology

0800 – 1130 S502 Symposium – Cytology

1130 – 1230 S503 – CAP Junior Scientist Award Lecture

1230 – 1330 Lunch in Exhibit Hall

1330 – 1430 S601 – CAP William Boyd Award Lecture

1430 – 1530 Platform Presentations

1530 – 1800 Poster Presentations

1800 – 2000 Exhibitors’ Wine & Cheese

1900 – 2100 Guillermo Quinonez Seminar on the Medical Humanities

Canadian Association of Pathologists60th Annual Meeting

PROGRAM OUTLINE

CONFERENCE 2009

4 Canadian Journal of Pathology July 2009

TU ESDAY J U LY 140730 – 1800 Registration Open

0800 – 0830 Anatomic Pathology Section Annual General Meeting

0830 – 1200 S701 Symposium – Anatomic Pathology

1200 – 1330 CAP Annual General Meeting Luncheon

1330 – 1700 S801 Symposium – Advanced Diagnostics

1330 – 1700 S802 Symposium – Forensic/Neuropathology

1700 – 1830 S803 Symposium – Quality in Pathology

1900 – 2200 Gala Banquet and Award Presentations at Pier 21

WEDN ESDAY J U LY 15 0730 – 1100 Registration Open

0830 – 1200 S901 Symposium – Dr. Cam Coady Slide Seminar

SCIENTIFIC WORKSHOPS

SATU RDAY J U LY 110830 – 1200 W101: Interepretation of the Non-neoplastic Mucosal Biopsies Morning of the Gastro-intestinal Tract

W102: Case Challenges in Surgical PathologyW103: Hematopathology for the Surgical Pathologist –

Potential Diagnostic PitfallsW104: Laboratory Practice Management Skills 101W105: Benign Pediatric Hematopathology: Pitfalls and Pearls

SATU RDAY J U LY 11 1330 – 1700 W201: Diagnostic Challenges in Surgical NeuropathologyAfternoon W202: Liquid Based Cytology: An Atlantic Perspective

W203: The Role of the Pathologist in Solid Organ TransplantationW204: The Role of Immunohistochemistry in Investigating

Primary Unknown MalignancyW205: Skin

SU N DAY J U LY 12 0830 – 1200 W301: Recent Advances in the Diagnosis and Classification of Kidney Morning and Urinary Bladder Tumours

W302: Glandular Lesions of the Uterine Cervix: Cytologic and Histopathologic Features

W303: Pathology of Diffuse Non-Neoplastic Lung DiseaseW304: Breast Pathology: New Entities and New Insights into Old EntitiesW305: Recent Developments in the Surgical Pathology of Salivary Glands

CONFERENCE 2009


SU N DAY J U LY 12 1330 – 1700 W401: Forensic Pathology and Sudden Death – Afternoon Looking Beyond the Standard Autopsy

W402: Fine Needle Aspiration Technique: “State of the Art”- Hands-on Workshop

W403: Controversies in the Diagnosis and Classification of Ovarian Tumours

W404: Diagnostic Approach to Selected Categories in Soft Tissue Tumours with Emphasis on Current Concepts: Interpretation of Tru-cut Biopsies and Various Immunohistochemical and Molecular Diagnostic Tests

W405: Standardized Handling and Reporting of Cancer Specimens: Colorectal and Endometrial Malignancies

SYMPOSIA AND LECTURES

SATU RDAY J U LY 11 Evening S101 Residents’ Symposium

MON DAY J U LY 13 Morning S501 Symposium – Hematopathology

S502 Symposium – Canadian Society of CytologyS503 Lecture – CAP Junior Scientist Award Lecture

Afternoon S601 Lecture – CAP William Boyd Award Lecture

Evening Guillermo Quinonez Seminar on the Medical Humanities

TU ESDAY J U LY 14 Morning S701 Symposium – Anatomic Pathology

Afternoon S801 Symposium – Advanced DiagnosticsS802 Symposium – Forensic/NeuropathologyS803 Symposium – Quality in Pathology

WEDN ESDAY J U LY 16 Morning S901 Symposium – Dr. Cam Coady Slide Seminar

CONFERENCE 2009


BUSINESS MEETINGSPreliminary Business Meetings Schedule

SATU RDAY J U LY 110800 – 2200 Pathologists Assistants Conference Day and Dinner (offsite)

0800 – 2200 Pathologists Assistants Conference Day and Dinner (offsite)

1200 – 1330 Workshop Directors Lunch

1700 – 1800 Section of Anatomic Pathology Executive Committee

1700 – 1800 CPD Committee

1900 – 2200 Residents’ Section Symposium and Dinner

SU N DAY J U LY 120800 – 1300 CAP Executive Committee, Working Lunch

0800 – 1000 RCPSC General Pathology Specialty Committee

0900 – 1100 Canadian Society of Cytology Executive Committee

1000 – 1200 RCPSC Anatomic Pathology Specialty Committee

1200 – 1400 RCPSC Anatomic and General Pathology Specialty Committees, Working Lunch

1200 – 1330 Workshop Directors Lunch

1330 – 1700 CAP Council

1700 – 1800 Section of Forensic Pathology Executive Committee and Annual General Meeting

1730 – 1830 Workload and Workforce Planning Training Session

MON DAY J U LY 130700 – 0900 National Standards Committee

1030 – 1100 Section of Hematopathology Executive Committee and Annual General Meeting

1100 – 1130 Canadian Society of Cytology Annual General Meeting

1800 – 2300 Canadian Chairs of Pathology, Working Dinner

1900 – 2100 Guillermo Quinonez Seminar on the Medical Humanities

TU ESDAY J U LY 140700 – 0800 Journal Editorial Board

0800 – 0830 Section of Anatomic Pathology Annual General Meeting

1200 – 1330 CAP Annual General Meeting and Luncheon

WEDN ESDAY J U LY 150700 – 0830 Media Training Workshop

0900 – 1100 Annual Meetings Committee

1300 – 1500 Public Relations Strategy Session

9Canadian Journal of Pathology

VOLUME 1, ISSUE 2 2009

About the Cover

Canadian Association ofPathologists 60th AnnualMeeting

3 Program Outline,Scientific Workshops,Symposia and Lectures,and Business Meetings

From the President’s Desk

11 Pathology in Canada:Today and TomorrowJagdish Butany, MBBS, MS,FRCPC

32 The Incoming Editor-in-Chief

Best PracticeRecommendations

14 Best PracticeRecommendations forStandardization ofImmunohistochemistryTests Canadian Association ofPathologists National StandardsCommittee/Immunohistochemistry

14 Preface

15 Use of Standard Terminology inClinical Immunohistochemistry

17 Principles/Best Practices forQuality Assurance of Clinical IHCTesting

21 Class II ImmunohistochemistryTest Principles/Best Practices

23 Proficiency Testing: Monitoringthe Quality of LaboratoryPerformance

24 Education and TrainingStandards for LaboratoryPersonnel

24 Bibliography

Features

26 Annual Meeting SurveyMartin Bullock, MD, FRCPC

28 Laboratory MedicineResidency Training2007–2008Aaron Pollett, MD FRCPC

30 Surgical Pathology: Stillthe Jewel in the Crown?Runjan Chetty, MB BCH, FRCPath,FRCPC, DPhil

33 Post-graduate Training inPathology andLaboratory Medicine:Time for a ParadigmShift?Richard G. Hegele, MD, FRCPC,PhD

37 Research in PathologyAvrum I. Gotlieb, BSc, FRCPC, MD

39 Anatomical PathologyTraining in Canada:Perspectives from JuniorResidentsZhongchuan Will Chen, MDCM,Eric K. Morgen, MD

41 Expectations of a SeniorAnatomical PathologyResidentMarie S. Abi Daoud, MD, MHSc

43 Pathology Training inCanada: A Perspectivefrom Two SeniorResidentsMichael D’Agostino, MD, MBA,Andrea Grin, MD

44 Molecular Pathology of Colorectal Cancer:Identifying the Status ofthe KRAS Gene for EGFR-Targeted TherapyDana Shor, Serge Jothy, MD,PhD, FRCPC

EDITOR-IN-CHIEF Jagdish Butany

C AP A SSOCIATION MANAGER Danièle Saintonge

MANAGING EDITORSusan Harrison

COPY EDITORSScott Bryant, Susan Harrison

ART DIRECTORBinda Traver, [email protected]

SALE S AND CIRCUL ATION COORDINATORBrenda Robinson, [email protected]

ACCOUNTINGSusan McClung

GROUP PUBLISHERJohn D. Birkby, [email protected]

––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––Canadian Journal of Pathology is published four times annually by

Andrew John Publishing Inc., with offices at 115 King Street West, Dundas, On, Canada L9H 1V1.

We welcome editorial submissions but cannot assume responsibility or commitment for unsolicited material. Any editorial material, including photographs that are accepted from an unsolicited

contributor, will become the property of Andrew John Publishing Inc.

FEEDBACKWe welcome your views and comments. Please send them to

Andrew John Publishing Inc., 115 King Street West, Dundas, On, Canada L9H 1V1.

Copyright 2009 by Andrew John Publishing Inc. All rights reserved.Reprinting in part or in whole is forbidden without express

written consent from the publisher.

INDIVIDUAL COPIE SIndividual copies may be purchased for a price of $19.95 Canadian.

Bulk orders may be purchased at a discounted price with a minimumorder of 25 copies. Please contact Ms. Brenda Robinson at

(905) 628-4309 or [email protected] for moreinformation and specific pricing.

Publications Agreement Number 40025049 • ISSN 1918-915XReturn undeliverable Canadian Addresses to:

115 King Street West, Suite 220, Dundas Ontario L9H 1V1

Canadian Journal of Pathology • Volume 1, Issue 2, 2009 Contents

This image shows an invasive ductal carcinomafrom a 53-year-old woman, 2.3 cm modifiedBloom Richardson Grade III/III (no special type).The tumour was sentinel lymph node negative,ER positive (i.e., 95% invasive tumour cell nucleistain positively using Ventana SP1), PR positive(i.e., 70% invasive tumour cell nuclei stain posi-tively using Novocastra Clone 16), HER-2/neunegative by F.I.S.H.

Immunohistochemistry details: fixation used:10% neutral buffered formalin; time to fixation:<1 hour; duration of fixation: 6–48 hours;Ventana Ultraview Detection System used; posi-tive and negative external laboratory controls andinternal normal control breast tissue stainedappropriately. (Courtesy of Dr. Bruce Youngson,Department of Pathology, University HealthNetwork, Toronto, Ontario.)

FSC Logo/Info Here

(Recycled Paper Content information inserted by Printer)


FROM THE PRESIDENT’S DESK

Welcome to the second issue of Canadian Journal ofPathology and to the 60th Annual Conference of

the Canadian Association of Pathologists. We hope thatyou enjoyed reading the introductory issue and will con-sider contributing to it. The second issue’s content ismore in line with what will be the sort of material thatyou have been waiting to see, waiting to read in yourjournal, and the type of content that we envisionedwhen we began planning Canadian Journal of Pathology.

Our specialty has been changing and continues to do soat an increasing pace. Whereas the complexity of patholo-gy cases keeps on increasing, the number of pathologystaff positions is not keeping pace with our workload. Thecomplexity of reporting, of immunohistochemistry testsfor many oncology cases, and of molecular diagnostics, aswell as cytogenetics and similar studies keeps on increas-ing. These results are all correlated and collated and pre-sented as a single diagnosis to our clinical colleagues byus, the pathologists. No wonder, then, that the same diag-nostic results that may have taken a few minutes in thepast now take considerably more time. Among the moststriking examples of these are all neoplasms, and amongthese more specifically are breast cancers and prostatecancers today, and in the future perhaps all malignantneoplasms. Most of these biopsies presently are muchsmaller than in the past, and there are more samples ofmany areas of the organ, entailing many deeper sections,immuno stains, and a far larger number of slides per casethan ever before.

The future looks challenging and exciting, with manychanges predicted, some of which are here already. Toborrow a phrase, “The future is here!”

The examination and study of tumours in other tissueshas continued to increase. The evaluation of the sentinellymph nodes has resulted in another significant load ofwork. Thyroid tumours that were once largely diagnosedon H & E stains today entail trays and trays of slides fullof immunohistochemistry for most thyroid cancers.

Other tumours are no exceptions. It is for these reasonsthat the complexity constantly increases; add to this theneed to perform multiple deeper sections on each slide,and one realizes how time consuming each report hasbecome. The increasing complexity of tumours and theever-smaller endoscopic and needle biopsies of earlylesions are pushing the need for subspecialization andsubspecialty consultations. These also emphasize the needfor continuing professional development or, as some callit, lifelong learning!

The first report of the National StandardsCommittee/Immunohistochemistry, specifically the stan-dards for class 2 tests (FDA classification), which havesuch significant prognostic value for our patients withbreast cancer, is released in this issue. This committee’schairs, Dr. Emina Torlakovic and Dr. Blake Gilks, didtremendous work in organizing the first national confer-ence on immunohistochemistry techniques (Vancouver,British Columbia, April 2–3, 2009). The conference, co-sponsored by CAP and CPAC (Canadian Partnershipagainst Cancer), had over 100 registrants, and the evalua-tions were uniformly positive. The plan is to organizemore of these conferences, every 2 years. The next onewill likely be held in Montreal.

The sudden appearance of the H1N1 virus in humanbeings in several countries has once again raised thespectre of serious viruses gone berserk, with the possibil-ity of numerous deaths in this pandemic. Fortunately,laboratories today, especially since the SARS outbreak inour country and in others, have had improvements inequipment and in skilled personnel. The CanadianPublic Health Institute in Winnipeg, Manitoba, was ableto sequence the gene, and hopefully there will soon be avaccine against this virus; although with this virus’s rep-utation to mutate and to cross species, causing symptomsin both species, it may be much more difficult to raisethe vaccine. The virus continues to cause problems inmany countries around the world, though it seems to beless virulent than anticipated, except in those with

Pathology in Canada: Today and Tomorrow

P a t h o l o g y i n C a n a d a : To d a y a n d To m o r r o w


other pre-existing illnesses.

As we are getting ready to go to press comes the news ofpotential breast cancer diagnostic problems in Quebec,perhaps similar to those that occurred in Newfoundland.These are related to estrogen and progesterone receptortesting. This raises the spectre once again of problems inpathology diagnoses, the media attention on laboratories,and the urgent need for uniform standards for laboratoryaccreditation, external quality assurance programs, appro-priate equipment, and good staffing of laboratories acrossthe nation – something that we have been pushing forsome time now.

Residency programs in most provinces run well and pro-duce pathologists with good training and good futureprospects for themselves, their institutions, and the spe-cialty. The Royal College specialty committees in anatom-ical pathology and general pathology have begun to takenote of the need to respond to the changing trends inpathology training, pathology examinations, and residen-cy requirements. With these changes in place, we shouldbe able to produce the pathologist of the future – onewho is well trained in immunohistochemistry and has agood knowledge of molecular diagnostics to be able tobuild on in the future as new molecular diagnostic testsarrive. We will also produce the pathologist who is wellversed in the use of digital technologies and who will beable to work with the new paradigms in pathology diag-nostics. The next generation of pathologists will soon,more than likely, become the pioneers of “Star Trek” med-icine. They will have to learn and develop greater skills inworking with others, especially those in similar diagnosticspecialties such as microbiology, infectious diseases, anddiagnostic radiology. They will also likely have to remainresponsive to changes in the environment. The CanMEDSphysician competency framework and the RCPSCMaintenance Certification Programs should make it easi-er to respond to future changes in our work environ-ments. Will these programs be enough? Or will more beneeded? At the least, we need new modalities of CPD,with a greater emphasis on user-driven learning pro-grams.

All the new skills cannot be learned in the present resi-dency program; in fact, residency training will continueto evolve and change with the times. Some of these addi-tional skills, including diagnostic skills in subspecialties,will need the provision of greater funding for fellowships,with available positions in those areas of expertise and thedesire among pathology residents and the employing

institution to send staff for higher and subspecializedtraining. One of the options may be to revamp thepathology residency program, to make it competencybased rather than numbers or time based. With the ongo-ing evolution of pathology diagnostics and their increas-ing complexity, the onus will be on us as individuals toconstantly upgrade our skills.

Today, diagnostic radiology appears to play a moreimportant role in diagnostics. These professionals seem tohave an ever-increasing ability to electronically enhancewhat is seen and to offer a higher diagnostic speed, skill,and accuracy. We the pathologists, the makers of thedefinitive tissue diagnosis, must continue to work andevolve to allow this tissue diagnostic specialty to remainthe gold standard for treatment to which all other modal-ities will be compared. This applies also to maintainingthe quality of autopsy as of surgical pathology.

We cannot afford to abandon our position or even dele-gate our responsibility to our patients by allowing othersto do what we do best, still wish to do, and have the great-est degree of expertise in. This does not mean that wemust practise hegemony; however, we are the ones besttrained and experienced in tissue diagnostics and whocan, in fact, proffer the most accurate, appropriate, andtimely diagnoses. We must continuously ensure that theresults are the most accurate possible, given in the mostappropriate time and with the appropriate amount ofdetail. In the absence of these, we may well run the risk ofbeing sidelined in the future.

The role of laboratories, of pathologists, and of laborato-ry staff in maintaining standards and quality is unques-tionable. Our reputation in this regard has until recentlybeen second to none. However, the critical role of thatlarge part of pathology, the autopsy, has been decliningeverywhere. This is unfortunate, and we must makeefforts to reverse it. Its role in clinical quality, education,and training is unrivalled and must be maintained.

With this issue, I have fulfilled the responsibilities Iundertook with regard to the Journal, namely to bring outa good and appropriate journal with a guarantee that theprocess will continue, and the hope that it will prosperand grow in the next few years. Under Dr. GodfreyHeathcote and his editorial staff, the Journal will, I amsure, flourish. (See "The Incoming Editor-in-Chief," p. 32.) The new editorial board will be meeting for thefirst time in Halifax, Nova Scotia, and I am sure they willbe able to resolve residual concerns and will, in fact, bethe finest talent. The Publications Committee chaired by

B u t a n y


Dr. Runjan Chetty will set editorial policy and be chargedwith maintaining continuity of editorial staff.

I take this opportunity to thank the members of the exec-utive who have worked tirelessly with me over the pastthree years, those who while located at the college officehave given their best and helped us unstintingly. Theadministrative staff at the college have consistently beenworking to the best of their capabilities and the best thatanyone can demand of them. They are an integral and avery large part of CAP.

Every one of the members of the executive through theseroughly three years has given his or her best for you, themembers of CAP. I am indebted to each of them for theircontinued and unstinted support through these tumul-tuous times. I could not have asked for a better or morehelpful team. In particular, I must thank Dr. DonaldCook, who, even though facing difficulties in the labora-tories in St. John’s, nevertheless stayed on for more than 2years past his term and did this graciously and pleasantly.

I take this opportunity also to thank everyone of you, themembers of CAP, who have helped in whichever form,whether it be in speaking at short courses or workshopsor participating in them as discussants, those who havespoken at the evening sessions, and those who are contin-uing to do so. All those of you who have attended thesesessions have been a tremendous help since we could nothave put on these courses and symposia without you.

Lastly, an ancient Chinese proverb that most of us arefamiliar with says, “May you always live in interestingtimes.” I can tell you honestly that the past 3 years havebeen interesting, and more. While I do wish that my suc-cessor too lives in interesting times, I do hope for all oursakes that the times will be less tumultuous.

I thank you all for having given me the opportunity toserve you for these past many months and the opportuni-ty to be your president for the past 2 years and 10months. I will always cherish these years and your sup-port. You gave me and your executive the opportunity todo our best and to do what was necessary as eventsevolved. We all believe that we lived up to your expecta-tions. I wish you all good health and every success.

I must thank a few others without those help it would nothave been possible to spend the many hours I have onthis aspect of my work. First among these are my wife andmy family for their faith and boundless support. Next, myinstitution, from the president down to our technical staff

and, most notably, my chief, Dr. Sylvia Asa, who oftenprovided valuable criticism and advice. And, finally, myuniversity department chair, Dr. Avrum Gotlieb, whoseexperience and expertise I was always able to call upon intimes of crises.

I would like to welcome you to the 60th AnnualConference of the Canadian Association of Pathologists.We hope that the program that the Local ArrangementsCommittee, chaired by Dr. Laurette Geldenhuys and herentire hardworking group, as well as the ContinuingProfessional Development Group and the AnnualScientific Meeting Committee have prepared represents awhole new experience and the beginnings of a new annu-al conference. This year, we have tried ever harder tomake this an annual meeting that you will not forget – ameeting that you will remember for its educational valueand for the friendships you had, friendships you cement-ed, and new friendships you make here in Halifax. I amheartened by the increasing support that the residentmembers show by attending in good numbers. We areplanning to change the format of the annual meetingbased on the survey we undertook earlier this year. Wethank you for having participated in the survey. Yourresponses are our guide to the format of future annualconferences.

Dr. Godfrey Heathcote has a team in place that will carrythe Journal forward for the next 3 years. They includemany from among you and others in this country whoare leaders in your fields, all masters of our art and teach-ers to boot. In this issue, you will find letters from someyoung residents talking about their hopes, aspirations,and concerns. These views are coming from those whowill be most directly affected by current changes andshould be of significant concern to every one of us.

This has been an eventful period in my life and an event-ful quarter since the publication of the last issue of theCanadian Journal of Pathology! We hope that you enjoyedthe first issue, and will enjoy this, and others to come,even more.

Ever has it been that love knows not its own depth untilthe hour of separation.

– Kahlil Gibran

Jagdish Butany

PrefaceImmunohistochemistry (IHC) and immunocytochem-istry (ICC) assays are highly complex diagnostic analysesused to aid in the accurate identification and biologicalcharacterization of tissue types in neoplastic and non-neoplastic diseases. Although currently applied mainly tothe diagnosis of neoplasms, some of these tests provideinformation of important prognostic and predictive valuein selected human neoplasms, and as such are often criti-cal for the appropriate and effective treatment of thepatients. These assays require specialized training in theselection of the appropriate tissue fixation and process-ing, preparation of the IHC/ICC slides, selection of con-trols, pre-treatment, detection systems, reagents, andextensive training in test selection and the interpretationof results. This document addresses several topics relevant

to test quality and recommends standards for appropriateIHC/ICC test development, performance, and interpreta-tion in diagnostic pathology and laboratory medicine.

With the understanding that medical knowledge andtechnology are constantly evolving, these recommenda-tions comprise current principles and best practices forquality assurance in clinical IHC testing that anticipatethe need for expansion and improvement of test options.The recommendations seek to assist Canadian patholo-gists and clinical IHC laboratories in the developmentand introduction of appropriate quality assurance proce-dures to do the following:

• Promote development and implementation of stan-dards for quality assurance

• Provide a standardized approach to tissue handling,

BEST PRACTICE RECOMMENDATIONS


Best Practice Recommendations for Standardization of

Immunohistochemistry TestsCanadian Association of Pathologists, National Standards Committee/Immunohistochemistry; Emina Emilia Torlakovic,MD, PhD; Robert Riddell, MD, FRCPath, FRCPC; Bryan Hewlett, ART (CSMLS), MLT (CMLTO); Diponkar Banerjee, MBChB,FRCPC, PhD; Hala El-Zimaity, MD, MS, FRCPC; Gregory Flynn, MD, FRCPC; Dragana Pilavdzic, MD, FRCPC; Peter Dawe, MS;Anthony Magliocco, MD, FRCPC; Penny Barnes, MD, FRCPC; Richard Berendt, MD, FRCPC; Donald Cook, MD, FRCPC; BlakeGilks, MD, FRCPC; Gaynor Williams, MD, PhD; Bayardo Perez-Ordonez, MD, FRCPC; Bret Wehrli, MD, FRCPC; Paul E.Swanson, MD; Christopher N. Otis, MD; Søren Nielsen, HT, CT; Mogens Vyberg, MD; Jagdish Butany, MBBS, MS, FRCPC

Emina Emilia Torlakovic, MD, PhD, is with the University of Saskatchewan, Saskatoon SK; Robert Riddell, MD, FRCPath, FRCPC, is with MountSinai Hospital, Toronto ON; Bryan Hewlett, ART (CSMLS), MLT (CMLTO), and Gregory Flynn, MD, FRCPC, are with QMPLS, Toronto ON; DiponkarBanerjee, MBChB, FRCPC, PhD, is with PHSA Laboratories, Vancouver BC; Hala El-Zimaity, MD, MS, FRCPC is with McMaster University, HamiltonON; Dragana Pilavdzic, MD, FRCPC, is with Sir Mortimer B. Davis Jewish General Hospital, Montreal QC; Peter Dawe, MS is with the CanadianCancer Society, St. John’s NL; Anthony Magliocco, MD, FRCPC is with the Tom Baker Cancer Centre, Calgary AB; Penny Barnes, MD, FRCPC, iswith the QEII Health Sciences Centre, Halifax NS; Richard Berendt, MD, FRCPC is with the Cross Cancer Institute, Edmonton AB; Donald Cook,MD, FRCPC, is with St. Clare’s Mercy Hospital, St. John’s NL; Blake Gilks, MD, FRCPC, is with the Vancouver Hospital and Health Science Centre,Vancouver BC; Gaynor Williams, MD, PhD, is with the University of Manitoba, Winnipeg MB; Bayardo Perez-Ordonez, MD, FRCPC, and JagdishButany, MBBS, MS, FRCPC, are with the University Health Network, Toronto ON; Bret Wehrli, MD, FRCPC, is with London Health Sciences Centre,London ON; Paul E. Swanson, MD, is with the University of Washington Medical Center, Seattle, WA; Christopher N. Otis, MD, is with BaystateHealth and Turfts University School of Medicine, Springfield, MA; Søren Nielsen, HT, CT, and Mogens Vyberg, MD, are with the Institute ofPathology, Aalborg Hospital/Aarhus University Hospital, Aalborg, Denmark.

Correspondence may be directed to Emina Emilia Torlakovic, MD, PhD: [email protected].

The proposed scope and content included here were addressed by the CAP National Standards Committee/Immunohistochemistry (NSC/IHC)members and external consultants and were presented to the CAP Executive Committee. This is a living document and will continue to evolve asmore data and information become available. The opinions expressed and arguments employed herein do not necessarily reflect the officialviews of the organization or its members.

To r l a k o v i c e t a l .


test performance and test interpretation

• Facilitate introduction of newly developed qualityassurance frameworks and maintenance of stringentstandards for test performance and interpretation forprognostic and predictive tests, the results of whichare used for stratification of patients for appropriatetherapies

• Increase public and professional confidence in thequality of immunohistochemical testing

These recommendations for quality control (QC) andquality assurance (QA) in clinical IHC provide the basisfor reasoned QC/QA in the clinical IHC laboratory and toensure both the accuracy of the tests and their interlabo-ratory reproducibility. This document addresses almostexclusively indirect IHC methods performed on formalin-fixed/paraffin-embedded tissues used by most practicinganatomic pathologists and hematopathologists.

This document includes recommendations on good labo-ratory practices, designed to ensure that appropriate qual-ity processes are considered; It does not prescribe the useof specific reagents, methodologies, or laboratory equip-ment.

In summary, this document provides a framework forlaboratory measures for clinical IHC, which should be ofhelp to pathologists and medical laboratory technologists,as well as organizations involved either in the develop-ment or the implementation of laboratory quality assur-ance programs for the practice of clinical immunohisto-chemistry.

The Canadian Association of Pathologists is neither a reg-ulatory nor a licensing body. These standards are pro-posed and recommended for use as one of the tools toachieve the above objectives in the area of high complexi-ty laboratory testing.

Use of Standard Terminology in ClinicalImmunohistochemistryThe use of standard terminology will improve communi-cation between pathologists and laboratory staff andensure appropriate test classification, which will furtherdetermine the level of appropriate quality control (QC)and quality assurance (QA) measures that need to beimplemented by clinical IHC laboratories.

Proposed Terminology

Clinical IHC laboratory – Any diagnostic pathology labo-ratory employing immunohistochemical tests for the pur-pose of diagnosing and/or characterizing human disease,the results of which will be evaluated by a pathologist andincorporated into the pathology reports.

Immunohistochemical test (IHC tests) – Tests thatemploy immunoassays to produce colour change co-localizing with an epitope of interest in tissue sections.IHC also encompasses testing on cell blocks or clot speci-mens prepared from cytologic and hematologic materials.While the majority of the IHC tests use immunoenzymat-ic detection methods and in particular horseradish perox-idase (HRP) and the chromogen diaminobenzidine(DAB) to demonstrate a positive reaction, IHC tests alsomay utilize other methods of detection (e.g., immunoflu-orescence, alkaline phosphatase, and others).

Immunocytochemical tests (ICC tests) – Tests thatemploy immunoassays to produce colour change co-localizing with an epitope of interest in cytologicalsmears, cytospin preparations, or monolayer prepara-tions. ICC tests often use alkaline phosphatase-baseddetection systems (with red chromogens), although peroxidase-based techniques are also commonlyemployed. Since processing of cytological samples is oftensubstantially different than in IHC tests, ICC tests requiredifferent QC/QA measures with an emphasis on the useof appropriate positive and negative controls preparedunder the same conditions.

Pre-analytical variables of IHC tests – Any and all steps intissue processing, including intraoperative tissue han-dling/treatment (prolonged ischemia, delayed fixation,etc.), type and length of fixation, decalcification, and ele-ments of tissue handling. The pre-analytical component isconcluded at microtomy and the placement of the tissuesection on pre-treated glass slides.

Analytical variables of IHC tests – The analytical vari-ables phase begins with the handling of the cut slides in aclinical IHC laboratory. It is completed with the cover-slipping of the stained slides.

Post-analytical variables of IHC tests – Interpretationand reporting of the results, which also includes interpre-tation of positive and negative control results.

Class I IHC/ICC tests – Any IHC or ICC test that is inter-preted in the context of histo- or cyto- morphology andclinical data. This class of in vitro medical tests includesthe great majority of IHC and ICC tests most of whichare used for determination of cell differentiation (e.g.,cytokeratins, vimentin, S-100, CD45). Any IHC/ICC testthat is reported as a stand-alone result to a clinician forprognostic or predictive purposes is by definition class II(see below).

Class II IHC/ICC tests1 – Any IHC or ICC test that is notdirectly confirmed by routine histopathologic or cytologicinternal and external control specimens. These tests are

ordinarily reported as independent diagnostic informa-tion to the ordering clinician. Claims regarding clinicalutility associated with these data must be widely acceptedand supported by valid scientific evidence. These testresults are often used to determine patient management.Examples of class II tests are those intended for semi-quantitative measurement of an analyte, such as hormonereceptors in breast cancer.

Qualitative IHC tests – Test results that are interpretedonly as “positive” or “negative.” Some quantitation may beinvolved, since a cutoff point/threshold for interpretingthe result as positive is often quantitatively defined (i.e.,> 10% reactive cells designated as a positive result). Thecellular localization of the evaluated epitope/antigen mustbe taken into account when interpreting these tests. Thesetests can be optimized without reference control materialby using appropriately selected positive and negative con-trols. However, the calibration of these tests needs to bevalidated and verified, both in-house and further by par-ticipation in external QC/QA programs.

Quantitative IHC tests – The test results are interpretedand reported according to an accepted scoring scheme.The tests are usually interpreted in a semi quantitativemanner (from 0 to 3+), but may also be a subject to eval-uation by image analysis or manual cell counts. It isassumed that the tests are optimized and calibrated insuch a manner that the intensity of staining, percent posi-tive cells, and distribution of staining proportionallyreflects the levels of target antigen expression. These testscannot be run without prior calibration against referencecontrol material or a specified tissue equivalent.Optimally, the performance of class II quantitative IHCtests will be informed by national/international consensusguidelines that address issues relating to and performanceguidelines for pre-analytical, analytical, and post-analyti-cal variables.

Prognostic IHC tests – The results of these tests inde-pendently forecast clinical outcome. They may be eitherqualitative or quantitative. They are considered class IIIHC tests.

Predictive IHC tests – The results of these tests independ-ently predict response to a particular therapy. They mayeither be qualitative or quantitative. They are considered

class II IHC tests. These tests are currently limited to onlythose for which targeted therapies are developed (e.g.,HER2/neu in breast carcinoma, CD117 in gastrointestinalstromal tumour, CD antibodies in lymphomas/leukemias).

Controls – A device, solution, lyophilized preparation, cellline(s), or human tissues intended for use in the qualitycontrol process.

Calibration of IHC tests – Calibration of the IHC tests isbased on results obtained by analysis of appropriatelyselected positive and negative controls or designated ref-erence controls. Calibration of qualitative tests (class I) isbased on the use of semiquantitative controls that includeat least one tissue sample with weak expression of the tar-get antigen and one sample with moderate or strongexpression of the target antigen. A single tissue fragmentmay also be suitable if it predictably represents cells withlow and high antigen expression. For details on controldesign see positive controls (below). Calibration of prog-nostic and/or predictive IHC tests (class II) is based onthe use of reference, previously validated control materi-als (calibrated control samples, which may consist oftumour samples, histoids, matrix models, or cell lines)with predetermined and reproducible levels of target anti-gen expression. Samples negative for the antigen of inter-est should be included. Both commercial and homemadecalibrated control samples are acceptable if scientificallyvalidated. Optimal calibration may or may not be equiva-lent to maximal sensitivity of the tests, since some testsmay intentionally be calibrated so that they do not detectvery low levels of expression of certain antigens/epitopes.

Antibody optimization of IHC tests – Antibody optimiza-tion is just one part of an analytical component, in whichit will be demonstrated that a certain clone or a specificlot of the primary antibody accurately and reproduciblydetects its target epitope. This is not equal to “validation”or “verification” of a clinical IHC tests (see below).

Primary antibody selection – Clinical IHC tests are oftennamed according to the antigens/epitopes they detect.This should not be equated with primary antibodies thatare available for detection of the antigen/epitope ofchoice. Antibody reactivities may have been proposed bythe commercial provider or based on initial published lit-erature regarding the characteristics of a particular anti-

C l i n i c a l I m m u n o h i s t o c h e m i s t r y S t a n d a r d s : R e c o m m e n d a t i o n s


1Note: Very few IHC tests are currently in clinical use as class II IHC tests. However, the same test can be designated as class I and/or class IItest depending on how the results of testing are interpreted. CD117 positivity in acute leukemia is an additional marker of myeloid differentia-tion (class I); CD117 positivity in a stromal gastrointestinal tumour may be used for the stratification of the patient for imatinib therapy (class II).Bcl-2 positivity in the Bcl-6+ germinal center cells helps support the diagnosis of follicular lymphoma (class I); Bcl-2 expression in ER+ breastcarcinoma may be considered as a good prognostic marker (class II). ER-positivity may be used in evaluation of metastatic carcinoma to sug-gest possible primary site (class I), but ER-positivity in breast carcinoma is often used to stratify the patients for hormonal therapy (class II).



body, but the tissue distribution of antibody reactivitywith a given clone may expand as greater experience isgained and either a wider than previously recognized dis-tribution of the target epitope is defined or cross-reac-tivites with non-target antigens is encountered. Sincemany clinical laboratories usually do not have the meansto extensively evaluate a particular primary antibodybeyond internal validation and optimization as definedelsewhere in this document, a selection of a given primaryantibody should be based on mature published literatureor the results of external QA/QC programs.

Validation and verification of clinical IHC tests – A“valid” assay performs as designed to detect the specifiedantigen. A “verified” assay detects the antigen as designedin a specified tissue/specimen type. The process of valida-tion or verification of class I and class II IHC test is verydifferent. Validation and verification of class I tests is usu-ally performed by using in-house samples of positive andnegative controls. Detailed knowledge of antigen/epitopedistribution and levels of expression in different tissues isrequired for appropriate selection of controls. In contrast,an ideal validation of class II IHC test should be based onreference material from completed prospective random-ized studies, though it may also be achieved if the test andreagent set produce results on local samples that are sub-stantially equivalent to the originally validated IHC pro-tocol. Concordance of ≥95% for both positive and nega-tive results with reference laboratory results or other ref-erence method (FISH for HER2/neu) is recommended.The number of test samples that is required for test vali-dation is determined by power analyses based on the pro-posed concordance rate and known characteristics ofboth the calculation to be used and the expected “passrate.” The same or different cut off point may be used for“pass” or “fail” by external quality assurance (EQA) pro-grams that provide proficiency testing for class II tests.Concordance at this level usually parallels a kappa-valueof ≥0.80 or “perfect or near perfect” agreement with a ref-erence laboratory or method. This is a desirable target fortests, as the results are intended to be used to determinebest therapies for cancer patients. Some EQA programsmay provide test samples that also quantitatively andqualitatively support IHC test validation and verification.Participation in such programs provides appropriate sup-port for initial and continuous re-validation of these tests.

Principles/Best Practices for Quality Assurance ofClinical IHC TestingScope: Internal

QC/QA standards are based on performance of daily pos-itive and negative controls. Therefore, the scope of this

section is focused on the selection and evaluation of posi-tive and negative controls for clinical IHC tests. Such con-trols are calibrated according to recommended standards.Most of the published literature refers to class II tests, forwhich specific guidelines have been published or are inpreparation. However, for all other class II tests and thegreat majority of class I tests, there are no publishedguidelines. A useful source to consult is the continuouslyupdated published literature at PubMed Search (seehttp://www.ncbi.nlm.nih.gov/sites/entrez). Among fewothers, the NordiQC organization also posts some usefulspecific recommendations for control tissues, antibodyselection and optimal methodology (see http://www.nordiqc.org/Techniques/Recommended_control_tissue.htm).

Standardization in clinical immunohistochemistry relieson consensus regarding an “optimal result” as well asoptimal/standardized selection of positive controls.Therefore, appropriate selection of the positive controls iscritical for the introduction and validation of IHC tests inthe clinical IHC laboratory and monitoring of daily IHCruns.

Positive Controls

Positive controls consist of tissue samples that contain anantigen of interest that can be detected by using primarymonoclonal or polyclonal antibodies designed to bind tothe selected epitope(s) in either fresh, frozen, or formalin-fixed/paraffin-embedded (FF/PE) samples. Positive con-trols are valid only if they are fixed and prepared in thesame manner as the tissue samples that are tested in theassay.

Selection of Materials/Tissues for PositiveControls

1. Positive controls for fresh air-dried (or briefly fixed inethanol or methanol or cytological spray fixative)cytological preparations must be only fresh air-dried(or fixed in the same fixative) cytological prepara-tions derived from previously characterized patients’samples (i.e., pleural fluid with metastatic melanoma)or previously characterized cell lines.

2. Positive controls for frozen tissue samples must bepreviously characterized frozen patients’ samples orfrozen cell blocks from previously characterized celllines.

3. Positive controls for formalin-fixed/paraffin-embed-ded (FF/PE) tissues must be previously characterizedpatients’ samples or FF/PE cell blocks from previouslycharacterized cell lines. The former of these two ispreferred.

4. The use of normal tissues with predictable antigen

expression, rather than tumour samples with variableexpression of antigen, is highly recommended in theselection of positive controls, although neoplastic tis-sues may be considerable value, indeed even required,in selected settings (ALK – anaplastic lymphomakinase, for example).

5. Inclusion of non-expressor cells/tissues with expectednegative results in a given positive control is highlyrecommended. The inclusion of such tissues providesa means for detecting unintended antibody cross-reactivity to cells or cellular components and repre-sents a “specific negative control.”

6. Positive controls for acetic zinc formalin (AZF)-fixed/paraffin-embedded decalcified specimens aretissues fixed in AZF, embedded in paraffin and decal-cified by using the same decalcifying procedure as forpatients’ samples. Similarly, any other type of tissueprocessing creates a need for positive controlsprocessed exactly the same way. It is not only inap-propriate to use positive controls which are processeddifferently than tested samples, but it may be diagnos-tically misleading.2

Positive Controls Design

Qualitative IHC Tests Need SemiquantitativePositive Controls

Semiquantitative positive controls are created by inclu-sion of tissues that optimally show predictable high, inter-mediate, and low levels of expression of the tested epi-tope. Samples with no reactivity for the given epitopeshould also be included. Such controls should ideally con-tain tissue with low levels of expression and either inter-mediate or high level if both of the latter are not avail-able. Small tissue arrays with several tissue cores with var-ious expression levels of the epitopes are recommended as

best practice. Alternatively, a single tissue fragment is suf-ficient if it reproducibly contains such representativeareas. Examples: (a) benign tonsillar tissue contains man-tle zones with low level of CD23 expression and germinalcenters with follicular dendritic cells with high level ofCD23 expression; (b) appendix contains mucosa withcrypts that demonstrate graded expression of Bcl-2; and(c) liver tissue exhibits low levels of CK8 (or low molecu-lar weight cytokeratin) expression in the hepatocytes andhigh levels in the bile ducts.

Quantitative IHC Tests (HER2, ER, PR) Need TrueQuantitative Positive Controls That Are CalibratedAccording to Reference Material/Standard

So called “reference material” can be designed by usingappropriately validated IHC kits or cell lines. It can alsobe created by using appropriately validated tumour sam-ples. See also “Validation and Verification of clinical IHCtests” (Recommended Terminology).

Types of Positive Controls

External Positive Controls

External positive controls are previously characterizedpositive tissue samples or cell lines that are tested in par-allel with patients’ samples. For IHC tests in general, onesuch external positive control is sufficient per run.However, for clinical IHC testing, it is recommended thatthe appropriate positive control be placed on a slidetogether with the patient material. This external controlsample needs to be indelibly identified as such.Automated IHC platforms/instruments can in someinstances have pipetting failure with random skipping ofa specimen. In some cases, where an internal control isnot present or is not informative, such instrument failurecannot be detected by any other means, other than havingthe external positive control placed on the same slide. It is



2If positive controls are not decalcified as the tested sample, they are much more likely to produce good positive signals with methods opti-mized on non-decalcified samples. False negative results are not uncommon with such practice. It continues to be a challenge for reference lab-oratories to perform IHC tests on tissue samples from different laboratories, for which the reference laboratory may not have appropriate posi-tive controls. This is particularly true for bone marrow specimens, for which tissue processing protocols vary widely.

3It is wrongly assumed that such internal controls are always better than external controls and that when internal control is present, no externalcontrol is needed. In the first place, not every tested antigen has a normal internal counterpart that can be used for this purpose. Internal con-trols are critical in evaluation of tissue preservation; however, this is useful only when an internal control is expected to be positive, but the testproduces negative result. In such cases, if the lesion/tumour is also negative, a false negative result cannot be ruled out. Internal controlsrarely, if ever, provide semiquantitative information regarding different target antigen levels of expression, which are essential for properly cali-brated IHC testing. A notable exception is the presence (or intentional inclusion) of normal breast tissue with tumour tissue to represent a refer-ence point for interpretation of breast cancer markers results. For some particular applications (e.g., CK5/6 staining of basal cells in prostatebiopsies) the internal positive and negative control are sufficient and, in fact, superior to external controls, as they also allow assessment offactors related to tissue fixation and processing.



not known how often this machine failure occurs (nopublished data are available); however, it can be statedthat “unpublished experience of reference laboratories”strongly favours this approach to positive controls. Thisapproach is also extremely useful in the clinical settingwhen an “unexpected negative” result is encountered bythe pathologists. The presence of an external positive con-trol on the same slide as patient’s sample will greatlydecrease number of repeat tests and will demonstrate thatanalytical component of the IHC testing was valid.

Internal Positive Controls

Internal positive controls are tissues in the patient’s samplethat contain the target antigen within normal tissue ele-ments, in addition to the tissue elements to be evaluated.3

Internal positive controls are very useful if the tissue stud-ied is expected to show at least some degree of expressionof targeted epitope. This rule can be applied only if thetest tissue is expected to demonstrate target antigenexpression. Even if this is the case, the pathologists shouldconsider the variation between normal and tumourexpression of most antigens. Many diagnostic criteriarefer to IHC tests results based on detection of low levelsof antigen(s), which are used to establish proper diagnosis(e.g., CD15 in Hodgkin lymphoma, cytokeratin in smallcell carcinoma), which may not be demonstrated ifappropriate positive controls with exact/similar low levelsare not used. Such controls are not useful if the tissueused for the study shows only aberrant tissue (“tumouronly”).

Negative Controls

Specific Negative Controls/Negative TissueControl

Specific negative controls or negative tissue controls arethose tissues that are known not to contain the antigenof interest. This type of negative control enables detec-tion of unintended antibody cross-reactivity to cells orcellular components and they may be a portion of apatient sample (“internal negative control”), which par-allels the concept of “internal positive control.” Specificnegative controls are used to document no reaction incells/tissues that are known to have no expression of thetested epitope(s). If non-specific negative controls arenegative and specific negative control positive, the falsepositive result is due to variables associated with the pri-mary antibody. Occasionally polyclonal antibodies maybe contaminated with other antibodies due to impureantigen used to immunize the host animal. This problemmay be detected by use of specific negative controls. Seealso “Non-Specific Negative Controls.”

Nonspecific Negative Controls/Negative ReagentControls

Negative reagent controls are used to confirm the speci-ficity of the test and to assess the degree of nonspecificbackground staining present by omitting the primaryantibody. Commonly, the primary antibody is replaced byone of the following: (1) antibody diluent, (2) samespecies nonimmune immunoglobulin of the same dilu-tion and immunoglobulin concentration, (3) an irrele-vant antibody, or (4) buffer. Non-specific negative con-trols can detect unintended background staining. Themain cause of non-specific background staining is non-immunological binding of the specific immune sera byhydrophobic, ionic, and electrostatic forces to certain siteswithin tissue sections. This form of background stainingis usually uniform. Prolonged fixation in formalin orother aldehyde-based fixatives should be avoided as itmay produce non-specific background. This backgroundstaining from overfixation can be remedied by postfixationwith Bouin’s, Zenker’s, or B5 fixatives, but this is not usefulin daily practice. Endogenous peroxidase activity is foundin many tissues and can be detected by reacting fixed tissuesections with DAB substrate, which is routinely eliminatedby pretreatment of the tissue section with hydrogen perox-ide prior to incubation of the primary antibody.

Non-specific negative controls are sections prepared fromthe same block of patient material that is used for clinicaltesting. The purpose of negative reagent controls is todetect either the lack of specificity of the test or nonspe-cific background staining, and for this reason has beenreferred to as the “methodology control.” Negative resultsin negative controls represent medical evidence that stain-ing for a particular epitope in the test tissue is not a falsepositive result due to variables other than the primaryantibody. Therefore, negative controls are critical for dailyQC/QA documentation in clinical immunohistochem-istry. The following principles should be adhered to:

1. Negative tissue controls need to be processed in thesame manner as the slides for specific IHC tests,including various epitope retrieval procedures.

2. The number of negative reagent controls is deter-mined by the number of different pretreatment pro-cedures: one negative reagent control should be pre-pared for each methodologic variation employed in agiven clinical case. For example, if three different epi-tope retrieval procedures are used (for example –HIER in citrate buffer, HIER in EDTA, and proteasedigestion), three negative controls (one processedHIER in citrate buffer, one with HIER in EDTA, andone with protease digestion) must be prepared.

3. Negative controls are typically run by omission of thespecific primary Ab in the protocol and replacementby an appropriate (presumably) non-reactive moiety.For monoclonal primary antibodies, the optimalchoice is an antibody of the same isotype, present inthe same Ig concentration as the test primary anti-body, using the same diluent, but non-reactive withhuman epitopes. Specially prepared commerciallyavailable negative controls may be used. For polyclon-al antibodies, negative reagent controls should be adilution of Ig fractions of whole serum ofnormal/non-immune serum of the same animalsource. Mouse ascites fluid can also be used as a nega-tive control. Finally, the primary antibody may bereplaced with cell culture medium (McCoy’s tissueculture medium is commonly used).4 Importantly,none of the non-specific negative controls are able todetected an undesirable/unexpected cross-reactivityof the primary Ab with some epitopes. In addition,when two or more antibodies are applied to serialsections, which is often the case in clinical laboratory,negative stain areas of one slide represent the negativecontrol for other antibodies. This combined approachis recommended.

4. Negative results with negative controls do not ensurethe specificity of the IHC tests in all cases. Therefore,when unexpected reactivity of the primary Ab isencountered, false-positive results need to be consid-ered.5

EXCEPTIONS

Class I tests. When a panel of antibodies is used for tissueanalysis, the results of other tests in the panel may pro-vide sufficient negative control information so that noadditional negative controls are needed. This approachinvolves proactive interpretation of “negative control”results by the pathologist who orders the panel. Whenordered, panels may not be sufficient or appropriate toserve as negative controls and, therefore, additional nega-tive controls could be ordered by the pathologists.

Class II tests. If published guidelines address the type ofnegative controls to be used for the particular class IItests, the guidelines should be followed. If negative con-trols are not described by the published guidelines, prin-ciples suggested in items 1 to 4 above should be followed.

Documentation of Positive and Negative ControlsResults

Note: In establishing retention requirements, care shouldbe taken to comply with provincial and federal regula-tions as these may exceed the following CAP NSC/IHCrecommendations.

The retention of laboratory documentation should bemaintained in such manner that it demonstrates that thetest(s) were performed correctly on the correct patient,that the reagents and equipment used to perform thetest(s) were operating correctly and that pathologists weregiven correct information, which would allow correctinterpretation of the IHC results. This information islargely included in the laboratory QA/QC records. Therecommendation on retention of laboratory documenta-tion on positive and negative controls in IHC laboratoriesis addressed here only from the aspect of internalaudit/review and its use for troubleshooting and does notaddress requirements for accreditation or legal purposes,which may vary in different jurisdictions/provinces.

1. Every clinical IHC laboratory needs to keep dailyrecords of all positive and negative controls and theirresults.6 Review and sign off of the positive and nega-tive control results by pathologists in charge is recom-mended.

2. If positive or negative controls indicate a failed run,documentation of corrective action is required.

3. Electronic documents with secure back-up are rec-ommended.

4. A two-year period is generally recommended for theretention of all laboratory records including therecords of positive and negative control performance.However, shorter periods for retention are also



4 For clinical IHC laboratories, where a large number of different monoclonal antibodies are in use, the most practical solution is to use cell cul-ture medium instead of primary antibody for all negative controls.

5Internal (or specific) negative control is essential for recognizing unexpected crossreactivity of the primary Ab, which can produce false positiveresults. Evaluation of cells that are expected to produce negative results in the tissue section should always be performed. Such expected neg-ative results are more consistent in benign tissues and often unknown or unpredictable in tumours. External controls, if included in the sameslide, may also provide useful information if they contain expected negative tissues.

6Practical approach to recording and retention of control results is to document specific suboptimal and inadequate results. Good results can bereported in aggregate fashion.



acceptable as follows:

a. Class I IHC test records may be kept for only 6 months. Last 10 results should be readily avail-able review.

b. Class II IHC test records should be available for atleast 12 months. Last 10 results should be readilyavailable for review.

c. If the laboratory regularly participates in EQAprograms that support test validation, the recordsshould be kept for the period of the last two chal-lenges for both class I and class II tests.

5. It is recommended that pathologists include theresults of positive and negative controls for all class IIIHC tests in the pathology reports. This is not cur-rently recommended for class I tests, though it isentirely relevant to include this information with alltest results.

6. Reporting of the results of the class I IHC controls isoptional in pathology reports; however, appropriatedocumentation needs to be performed by the IHCclinical laboratory as above.

New Test Validation

The introduction to clinical use of any new test must startwith test verification and validation (see ProposedTerminology for definitions). However, all previously vali-dated IHC assays must be completely revalidated if signif-icant changes are made to the assay procedure. Significantchanges include: change of primary antibody clone orprovider, new detection system, new buffer type or anyother component of antigen retrieval step, new machinefor immunohistochemistry, or switch from pre-diluted toin-house diluted (or the other way around) primary anti-body. For new primary antibody lots, a smaller concor-dance study for lot variation alone is recommended toinclude a 10 case run in parallel with the original lot.

Class II Immunohistochemistry TestPrinciples/Best PracticesScope

There is only one IHC test so far for which testing guide-lines have been issued, namely HER2. In Canada, elevenexpert pathologists issued a document entitled CanadianConsensus for HER2 Testing Guidelines in Breast Cancer,which basically reflects ASCO/CAP Guidelines; however,Canadian modifications are introduced based on theexperience of the 11 consensus group participants. Itdetails pre-analytical, analytical, and post-analytical stepsfor HER2 IHC testing. The impact of the pre-analyticalcomponent was the most challenging to define, since vari-

ous epitopes may have widely different responses to for-malin-fixation and tissue-embedding or other steps in tis-sue processing. The following recommendations for classII tests include most recent published articles and appearto be safe for all class II tests (prognostic and predictiveIHC markers) including HER2, estrogen receptor (ER)and progesterone receptor (PR), and CD117 IHC tests:

Pre-analytical Component/Tissue Processing

Since ER/PR and HER2 tests are generally performed onthe same specimens, at a minimum, compliance withHER2 guidelines may be sufficient for most analytic vari-ables. However, more recently published requirements forER/PR testing may be more informative with respect toappropriate minimum fixation times. The following rec-ommendations are based on published peer-reviewed rec-ommendations for tissue fixation and processing of sam-ples for breast carcinoma markers. At this time, there isno evidence to suggest that the proposed recommenda-tions would not be safely applied to all other class II testsand to all class I tests. The following recommendationsincorporate accumulated scientific knowledge on forma-lin-fixation effects on epitope preservation.

Tissue Handling and Fixation

Tissue processing, in particular the type of the fixativeused and the fixation time prior to loading the tissue ontothe tissue processor needs to be recorded and included inpathology report. Use of decalcified tissues is not recom-mended unless permitted by the specific class II testguidelines. If decalcified tissues are used for testing, fol-lowing times need to be recorded: Fixation time prior todecalcification; type of decalcifying reagent; and time ofdecalcification.

Fixation of tissues should be performed in 10% neutral-pH (pH 7.2–7.6 aqueous), phosphate-buffered formalinfor a minimum of 8 hours (24–72 hours optimal). A fixa-tion time of 8 to 72 hours is generally recommended.Therefore, tissues should be fixed in formalin for at least 8hours before being loaded onto the tissue processor.

Non–formalin-based fixatives and or other fixationmethodologies should not be used for class II IHC tests,because of the lack of published evidence on the perform-ance characteristics of other tissue processing methods.However, if any scientifically supported tissue processingmethod is validated in future, such a method may be usedfor clinical IHC testing. It is of high importance toremember that any such pre-analytical methods must alsobe validated for all other IHC tests (both class I and classII) that may be applied to tissues thus prepared.

The time from surgical excision of the specimen to place-

ment in fixative should be optimized. Epitope degrada-tion due to delayed fixation is a potentially serious prob-lem that generally cannot be modified/corrected by epi-tope retrieval techniques (these mainly adjust for fixative-induced modifications of the epitope, but not for theeffects of delayed fixation, which usually leads to irre-versible tissue deterioration). Both, prolonged prefixationischemic time and fixation of less than 8 hours may irre-versibly modify our ability to detect specific epitopes.Samples should be sliced immediately at 5–10 mm inter-vals after appropriate gross inspection and margin desig-nation, and then placed in sufficient volume (20:1) of10% aqueous neutral buffered formalin.

A longer fixation time of up to 10 days is acceptable andis not an exclusion criterion for most IHC testing, as longas the specimen has been adequately sectioned to allowadequate fixation as described above. Class II IHC testsmay require validation (consult published guidelines foreach test separately) of protocols for fixation timesexceeding 72 hours.

Note the following:

• Under-fixation is more deleterious than over-fixation.

• Fixation requirements are particularly specified forER/PR/HER2 IHC and FISH testing on core biopsiesand surgical excisions. Consensus guidelines for otherclass II tests are not available yet and thus are notspecified in this document. They will be incorporatedinto this document as they become available.

• Long term archival storage of tissue blocks (for even20 or more years) does not preclude HER2 testing aslong as the archived material has not been subject tosignificant temperature fluctuations over time.

Analytical Component

Assay validation and verification, antigen retrieval, selec-tion of positive controls, and use of laboratory methodsas per reference 10.

Assay Validation

As many as 50 – 100 samples may be required when vali-dating a new antibody for a class II test. An assay accuracyof a 95% concordance rate is recommended for test vali-dation for both the positive and negative categories.Ensure adequate validation, preferably by using 50% casesthat are unequivocally positive and 50% cases that are themixture of weakly positive and unequivocally negative.Much smaller number of samples may be sufficient forsome class II test (e.g., CD117). Validation documenta-tion must be kept as long as reported results of these testsare kept. Any significant modifications to the procedure

require additional validation to ensure accurate perform-ance (see re-validation for definitions).

Type of Antigen Retrieval

Stringent compliance with validated Standard OperatingProcedures developed in Assay Validation must beadhered to. QC documentation must be in place indefi-nitely or as long as pathology reports that include theIHC test results are mandated to be retained.

Use of Standardized Control Materials

The controls should include positive and negative cases aswell as a low protein expressor cases. The control tissueshould be fixed and processed in the same manner as thepatient samples. The number of samples is determinedbased on the design of the validation sample as well as onthe power analysis based on the selected level of perform-ance and known characteristics of the test sample. It isrecommended that the design of the in-house validationsample be supported by the recommendations from thestatistician versed in such studies. The material is validat-ed in a prospective clinical trial results or by using theresults of procedures that are validated based on suchprospective clinical trial.

Use of Automated Laboratory Methods

The use of correctly operated automated staining proto-cols and equipment are acceptable and desired for clinicaltesting; however, validated methods must be used.Records of recommended maintenance and servicerecords must be retained indefinitely or as long as thepathology reports that include results of the IHC tests areretained.

Post-analytical Component/Interpretation ofResults

Every test may ultimately have its own guidelines for per-formance and interpretation, although guidelines for classII tests will likely be more rigidly defined than those forclass I tests. HER2/neu is detailed in the CanadianConsensus for HER2/neu Testing Guidelines in Breast Cancerand the ASCO/CAP document. Similar guidelines for theinterpretation of the ER and PR tests are being or havebeen developed. Image analysis may also be suitable forthe interpretation of the results as it was reported to be asgood as expert pathologist’s scoring and some systems arealready approved for such use by the USA Food and DrugAdministration (FDA). It is recommended that imageanalysis scoring results also be validated by either partici-pation in the extra laboratory QC/QA program or by thereference laboratory. The post-analytical components ofthe class I test are generally test specific. However, it couldbe said that for general purposes a cut off value of 10%





positive cells is used to designate a test as positive or neg-ative. Only exceptionally, the test may be positive insmaller numbers of cells if the pattern of staining is suffi-ciently restricted to certain lesions and documented bythe published literature. This is well illustrated using anexample of a CMV immunoreactivity in morphologicallyaltered cells where even one positive cell is sufficient forinterpretation as a “positive test result”. On the otherhand, some class I markers are considered as “positive”only if strongly expressed by a majority of cells as this isthe case in interpretation of some tests, most notably TdTand CD99. Therefore, approved training in pathology isrequired for clinical application and interpretation ofIHC tests’ results.

Proficiency Testing: Monitoring the Quality ofLaboratory PerformanceProficiency Testing and Certification of Class IIIHC Tests versus Laboratory Accreditation7

Laboratory accreditation is currently under the jurisdic-tion of provincial regulatory bodies in several provincesin Canada. However, class II test certification is distinctfrom the current processes for laboratory accreditation.The IHC test certification is defined as successful partici-pation in external quality assurance program for IHCtesting at the pass rate of minimum of 90% with bothpositive and negative results of the reference value usedby the EQA program. This definition assumes the require-ment of sufficient number of test samples for meaningfulstatistical analysis based on power analysis. It appears thatat least 40 samples are required to achieve the desirablepass rates (90% to 95%), or to approach good correlationwith “near perfect” agreement based on kappa-values of0.80. This pass rate may or may not be required by theprovincial or other accrediting bodies. Canadianprovinces or territories that do not have accrediting bod-ies may use the above recommendation for safe clinicalpractice. Therefore, participation in EQA programs thatprovide such test samples would support “certification” ofclinical IHC test. Laboratory accreditation further man-dates/enforces compliance to designated provincial stan-dards, which may be, higher or lower than the above.

The above recommended requirements for class II IHCtest certification may be regularly updated to follow pub-lished literature and developing national and internation-al guidelines.

The CAP National Standards Committee/Immunohisto-

chemistry suggests that implementation of the nationaland international published guidelines for class II testsand their continuous validation could be facilitated bydevelopment of the following:

1. Establishment of a Canadian National Checklist forclinical IHC Laboratory certification. This wouldensure that all Canadian clinical IHC laboratories ful-fill minimum, standard, requirements and elementsof implementation for class I and class II tests, whichare very different and need to be adequate for thetype of tests that are performed by clinical laborato-ries. Class I IHC Test Checklist and class II IHC TestChecklist would form the basis for a step-wiseapproach to appropriate daily QA/QC measures, aswell as for selection of appropriate EQA programsthat would support test validation and verification.

2. Certification for each prognostic and predictive test(class II IHC tests) separately is proposed. Historicalevidence in EQA has indicated that good or evenoptimal performance in one test does not guaranteeequally good performance in another (similar or oth-erwise) IHC test. Therefore, the NSC/IHC recom-mends certification of each class II test separately.

3. The certification of class II tests includes demonstra-tion of at least 90% concordance for both positiveand negative results at least twice annually; the partic-ipation interval is aligned with ASCO/CAP recom-mendations for IHC breast cancer markers. Largetesting centers do not automatically qualify as refer-ence laboratories. Any IHC laboratory that providesevidence of ≥ 95% concordance or kappa-value of> 0.80 with both positive and negative reference val-ues can be considered a “reference laboratory” as thiscut off point correlates with “near perfect agreementwith reference value, which is not the result of chanceagreement” (see above regarding kappa-values inEQA). Also, see above for definitions of IHC test vali-dation.

4. If a laboratory did not pass with a recommendedscore of at least 90% concordance, it is recommendedthat the unsuccessful laboratory send all samples tobe tested, to another certified laboratory or designat-ed reference laboratory until corrective action hasbeen taken and repeat tests passed with an acceptablesuccess rate. Participation in any EQA program thatprovides such samples for test validation is accept-

7Certification for performing class I IHC tests, would continue to be conducted according to current practices (in agglomerate) with recommen-dation to use Canadian IHC QA Checklist for class I IHC tests.

able. Participation in EQA programs that do not pro-vide sufficient number of tissue samples may beinformative and useful, but due to the lack of the pos-sibility to calculate concordance rates, it does not pro-vide sufficient information for class II test certifica-tion.

5. The participation in EQA programs needs to be doc-umented by the clinical IHC laboratories and if sub-optimal results are achieved, corrective actions alsoneed to be documented.

6. All pathologists interpreting class II tests are expectedto provide interpretation that agrees with referenceinterpretations with 90% concordance or a kappa-value of 0.80 in at least one challenge per year.

Education and Training Standards for LaboratoryPersonnel Scope

Minimum standards for clinical IHC staff including labo-ratory directors are not currently set. The NSC/IHC atthis time is recommending that the following would beconsidered as good practice:

1. Clinical IHC laboratories should be adequatelystaffed to meet the volume and complexity of the lab-oratory testing.

2. Clinical IHC laboratories should be directed by afully trained and certified designated pathologist withexperience in immunohistochemistry, who is thedirector of IHC laboratory and is responsible forcomplying with internationally and/or nationally/provincially designated standards for class II tests.

3. The director of the IHC laboratory and the laborato-ry manager and ultimately the director/head of thedepartment of pathology in that institution areresponsible for the results of the IHC tests. (It is reit-erated that the director of the IHC laboratory musthave final authority over the IHC laboratory and thatthe head of the Department of Pathology/Laboratoriesmust have overall responsibility for the work /resultsof that department.)

4. The medical director (pathologist) must have thefinal authority for selecting appropriate tests, anti-body clones, detection systems, or any other signifi-cant component of clinical IHC testing.

5. IHC laboratory director (pathologist) is expected torecommend the implementation of appropriate pre-analytical, analytical, and post-analytical componentsof the IHC testing. Such recommendations should berecorded. (The IHC director cannot be held responsi-

ble for laboratory errors if the above recommenda-tions are not implemented by the laboratory and hos-pital management.)

6. Any clinical IHC laboratory should have at least onefully trained and qualified technologist for the type ofthe tests that are performed by the clinical laboratory.Support technicians/technologists without specialtraining in IHC techniques may support IHC labora-tory operations (e.g., cutting tissue sections).

7. The training and continuing professional develop-ment of the technologists for the IHC laboratorieshas to be documented by the department manage-ment.

8. Rotation of technologists designated to the clinicalIHC laboratory to oversee and perform clinical IHCtests in addition to other histology duties is appropri-ate if the highly qualified personnel are not replacedby the less experienced technicians/technologist with-out special training in immunohistochemistry tech-niques.

9. There are no recommendations for support techni-cians/technologist at this time; they may rotatebetween IHC duties to other laboratory activities.

BibliographyElias JM. Immunohistochemical methods. In: Immunohistopathology.

A Practical Approach to Diagnosis. Chicago: ASCP Press;2003:1–110.

Goldstein NS, Ferkowicz M, Odish E, Mani A, Hastah F. Minimum for-malin fixation time for consistent estrogen receptor immunohisto-chemical staining of invasive breast carcinoma. Am J Clin Pathol2003 ;120(1):86–92.

Goldstein NS, Hewitt SM, Taylor CR, et al. Recommendations forimproved standardization of immunohistochemistry. ApplImmunohistochem Mol Morphol 2007 Jun; 15(2):124–33.

Guidance for Industry –Guidance for Submission ofImmunohistochemistry Applications to the FDA. U.S. DepartmentOf Health And Human Services, Food and Drug AdministrationCenter for Devices and Radiological Health, Immunology Branch,Division of Clinical Laboratory Devices, Office of DeviceEvaluation. Available at: http://www.fda.gov/cdrh/ode/immuno.html

Hanna W, O’Malley F, et al. Canadian Consensus for HER2/neu TestingGuidelines in Breast Cancer: Technical Information Pamphlet.2007.

Hsu F, Nielsen TO, Alkushi A, et al. Tissue microarrays are an effectivequality assurance tool for diagnostic immunohistochemistry. ModPathol 2002; 15:1374–80.

Nagle RB, Fitzgibbons PL. Consensual interpretive guidelines for diag-nostic immunohistochemistry. Am J Surg Pathol 2002;26(6):816–7.

Maxwell P, McCluggage WG. Audit and internal quality control inimmunohistochemistry. J Clin Pathol 2000; 53(12):929–32.

O’Leary TJ. Standardization in immunohistochemistry. ApplImmunohistochem Mol Morphol 2001 ;9(1):3–8. Review.

O’Leary TJ, Edmonds P, Floyd A, et al. Quality Assurance for





Immunocytochemistry; Approved Guideline. NCCLS documentMM4-AC. December 1999. Vol. 19, No. 26.

Ramos-Vara JA, Kiupel M, Baszler T, et al. Suggested guidelines forimmunohistochemical techniques in veterinary diagnostic labora-tories. J Vet Diagn Invest 2008; 20(4):393–413.

Rasmussen OF. Controls. In: Key M, ed. ImmunohistochemicalStaining Methods. 4th ed pp. Carpinteria, CA: DAKO Corporation;2006: 113–8

Shi SR, Liu C, Perez J, Taylor CR. Protein-embedding technique: apotential approach to standardization of immunohistochemistryfor formalin-fixed, paraffin-embedded tissue sections. JHistochem Cytochem 2005; 53(9):1167–70.

Shi SR, Liu C, Taylor CR. Standardization of immunohistochemistry forformalin-fixed, paraffin-embedded tissue sections based on theantigen-retrieval technique: from experiments to hypothesis. JHistochem Cytochem 2007; 55(2):105–9.

Tafjord S, Behler PJ, Risberg B, Torlakovic E. Estrogen and proges-terone hormone receptor status in breast carcinoma: comparisonof immunocytochemistry and immunohistochemistry. DiagnCytopathol 2002; 26(3):137–41.

Taylor and Cote, eds, Immunomicroscopy: A diagnostic tool for thesurgical pathologists. In: Major problems in pathology. Saunders;2005:26–28.

Taylor CR. FDA issues final rule for classification and reclassificationof immunochemistry reagents and kits. Am J Clin Pathol 1999;111(4):443–4.

Taylor CR. Quantifiable internal reference standards for immunohisto-chemistry: the measurement of quantity by weight. ApplImmunohistochem Mol Morphol 2006; 14(3):253–9.

Taylor CR. The total test approach to standardization of immunohisto-chemistry. Arch Pathol Lab Med 2000; 124(7):945–51. Review.

Taylor CR, Shi S-R, Barr NJ, Wu N. Techniques of immunohistochem-istry: Principles, pitfalls and standardization. In: Dabbs D.Diagnostic Immunohistochemistry. Churchill Livingstone Elsevier;2006;1–42.

Turbin DA, Leung S, Cheang MC, et al. Automated quantitative analy-sis of estrogen receptor expression in breast carcinoma does notdiffer from expert pathologist scoring: a tissue microarray studyof 3,484 cases. Breast Cancer Res Treat 2007 Oct 3.

Wiatrowska BA, Berner A, Torlakovic G, et al. Cultured anaplastic celllines as immunocytochemistry controls: a comparison ofThinPrep-processed smears and conventional air-dried cytospins.Diagn Cytopathol 2001;25:303–8.

Wolff A, Hammond M, et al. American Society of ClinicalOncology/College of American Pathologists GuidelineRecommendations for Human Epidermal Growth Factor Receptor2 Testing in Breast Cancer. J Clin Oncol 2007;23(1):1–28.

Yaziji H, Taylor CR, Goldstein NS, et al. Consensus recommendationson estrogen receptor testing in breast cancer by immunohisto-chemistry. Appl Immunohistochem Mol Morphol 2008 Dec;16(6):513–20.

DISPLAY CLASSIFIEDS

To inquire about rates or to submit material toadvertise in Canadian Journal of Pathology

please contact Brenda Robinson:T: 905.628.4309| E: [email protected]

Subject to availability. Book early!

2-YEAR RESIDENCY TRAINING PROGRAM IN TRANSFUSION

MEDICINE

Applications are invited for the two (2) year ResidencyTraining Program in Transfusion Medicine (sub-speciali-ty with examination), to prepare the candidate for con-sultancy role in Transfusion Medicine. This is a JointProgram with the University of Ottawa, Queen’sUniversity and Canadian Blood Services. At the end ofthe two (2) year program, there is certification byexamination by the Royal College of Physicians andSurgeons of Canada. Salary is commensurate with PGYyear plus research and travel monies provided. Trainingincludes nine (9) months at a blood centre, nine (9)months at a hospital (Transfusion Medicine service),four (4) months Elective, and two (2) months selective.

Eligible applicants will be nearing completion of trainingin, or have a Royal College of Physician and Surgeons ofCanada Certification in one of the following:

Internal Medicine + HematologyPediatrics + Pediatric Hematology Oncology Hematological Pathology Anesthesiology General Pathology

Deadline Date for Applications: July 31, 2009

Date of Commencement of Program: September1, 2009

Applications will be considered until the position isfilled.In accordance with Canadian immigration requirements,this advertisement is directed primarily, but not solely, toCanadian citizens or permanent residents of Canada.The Ottawa Hospital and University of Ottawa are com-mitted to equity in the workplace: women, aboriginalpeoples, members of visible minorities and persons withdisabilities are encouraged to apply.

Please submit curriculum vitae and the names of three referees to:

Dr. Antonio Giulivi, Division HeadHematopathology and Transfusion Medicine

The Ottawa Hospital, General Campus501 Smyth Road, EORLA Lab

3rd Floor, Rm. #3887, Box #115Ottawa, ON K1H 8L6

F EATU R E


In January 2009, the Canadian Association ofPathologists (CAP) conducted a web-based survey of

the general membership to assess the member’s opinionsabout the annual general meeting and other activities ofthe association. By conducting the survey, the CAP exec-utive committee hoped to gain more information aboutwhat the members wanted the meeting to provide fromeducational, social, and scientific perspectives, whereand when meetings should be held, and how membersfelt the meeting could be improved. The survey consist-ed of 17 questions. A maximum of 167 members (about18% of the membership) responded to any of the ques-tions, but most questions were answered by 130–160members. Most of the questions permitted free textcomments to be added.

General FindingsOf the 167 respondents, the majority (55%) attend themeeting less frequently than every other year or not atall. Twenty-six percent stated they attend every year,19% every other year. The most frequent reasons forinfrequent attendance were: the desire/need to attendother meetings (43%) and scheduling conflicts (39%).About one quarter of respondents chose cost of atten-dance and location as reasons for not attending. About20% stated that the meeting did not fit their educationalneeds. There were many specific comments that the tim-ing of the meeting (early summer) was not convenient,mainly because of conflicts with vacation. There wereseveral comments that the meeting should aim to pro-vide higher quality speakers, topics of more practicalvalue to practicing pathologists, and integrate more labmanagement and QA topics. Educational value for dol-lar spent was a concern, and comparisons were madewith other meetings (especially USCAP) and courses.

Of factors that respondents considered important whendeciding whether or not to attend the meeting, morethan 75% considered cost of travel to be very or moder-ately important. Greater than 50% considered price oflodging and local attractions as very/moderately impor-tant. Climate of the location and availability of restau-rants were considered to be very/moderately importantby fewer than 50%.

Fifty-seven percent of respondents felt that there is valuein holding the meeting in conjunction with other soci-

eties’ annual meetings, such as the CSCC or CAMB.However, few respondents (16%) named another specif-ic society with which they would like to meet. Amongothers, the responses included the CANP, Americanpathology organizations (USCAP, CAP, and ASCP), andprovincial associations (e.g., OAP).

Social ActivitiesWhen asked which of the main social activities duringthe meeting they considered valuable, the responseswere as follows: exhibitor’s wine and cheese (87%), pres-ident’s reception (74%), gala banquet (71%), and localtours (57%). Only 37% felt that they would utilize aspousal program if offered.

Poster and Paper SessionsThe members were asked to rank their satisfaction withthe poster sessions from a choice of five categories rang-ing from “very unsatisfied” to “very satisfied.” They wereasked about the overall quality of the posters, the posterviewing time, and the interaction with authors. Therange of responses was similar for each category. Whileonly a low percentage (less than 10%) was unsatisfiedwith any of the three elements, even fewer were “verysatisfied,” with the majority being either “neutral” or“satisfied” (about evenly split). Several respondentscommented that there was little interest in posters andthat they were in inconvenient/poor or remote locations.

There was a greater degree of satisfaction with the oralpaper sessions. Respondents were asked to rank theirsatisfaction with oral paper sessions with respect to the15 minute format, the discussion during the sessions,and the open discussion to complement the session.Very few respondents were unsatisfied. About 60% ofrespondents were “satisfied” or “very satisfied” with the15 minute format and discussion during the session.Half the respondents were “neutral” with respect to anopen discussion, with most of the remainder expressingsome degree of satisfaction.

Workshops and SymposiaIn general, there was a high degree of satisfaction withthe symposia. Members were asked to rank their satis-faction with the symposia topics, format, and opportu-nity for discussion. The percentage of respondents who

Annual Meeting SurveyMartin Bullock, MD, FRCPC

Martin Bullock, MD, FRCPC, is a pathologist, QEII HSC, an associate professor, Faculty of Medicine, and program director, Anatomical Pathology,Dalhousie University, Halifax, Nova Scotia.


were “satisfied” or “very satisfied” with those three com-ponents ranged from 65–69%. Seven percent of respon-dents were “very unsatisfied” or “unsatisfied” with thesymposia topics. The specific comments with respect tosymposia were variable and sometimes conflicting.There were a few comments that the symposia were too“AP-oriented,” that the interest to individuals variedconsiderably from year to year and that the sessionsshould be more practical.

Eighty-two percent of respondents felt that the work-shops are beneficial. When asked when the workshopsshould be held, about half felt they should be during themeeting, with 40% favouring before and 10% after.

Other Members were asked to comment on how CAP member-ship could be increased. As expected, there was a widevariety of responses, but a strong theme was that theCAP should become more relevant by increasing itsfocus on issues of concern to allCanadian pathologists, includingstandards of practice, lab accredita-tion, workload, and compensation.There were multiple suggestions toincrease cooperation with provin-cial pathology associations andorganizations in Europe, Britainand the USA. Several respondentsfelt that the CAP should offer moreCME (online and otherwise) andthat it should encourage the inter-est of general pathologists by offer-ing more clinical pathology topics.While there were a few commentsthat membership fees should bedecreased, there were more com-ments about value for dollar spent,rather than lower fees.

ConclusionThe survey provided importantinformation that should provehelpful in planning future meet-ings. Attendance is limited by com-petition with other meetings thatmembers may feel better fulfilltheir educational needs. The earlysummer timing is inconvenient formany members. Cost of travel andlodging is a concern to many, andtherefore concentration on themost readily accessible Canadiancities should be a consideration.The respondents were generally sat-

B u l l o c k

isfied with symposia and workshops, although many feltthat the workshops should occur during the main meet-ing rather than on the weekend before. The paper andposter sessions received variable ratings. Considerationshould be given to ways of increasing interest in theposters. The expense of organizing local tours andspousal events is probably not worthwhile. The surveywas limited by the fact that of the respondents, only45% attend the meeting at least every other year, socomments may not reflect the current reality. Manymembers felt strongly that the CAP should becomemore involved in providing practical guidance toCanadian pathologists, for example through practicalCME and by increasing emphasis on standards of prac-tice.

The Professional Affairs Committee of the CanadianAssociation of Pathologists performs a periodic sur-

vey of all Canadian laboratories to find the number andspecialty of physicians in active practice. As part of thissurvey, residency program directors are also asked fordetails of physicians in training. The CAP manpowerand resident database is voluntarily maintained andrequires accurate responses to the surveys and requestfor information.

In the past 5 years the number of residency positionsavailable through the Canadian Residency MatchingServices (CaRMS) has increased from 37 in 2004 to 65 in2009 (see Table 1 for a breakdown by discipline for posi-tion in 2009). At the same time, the number of laboratorymedicine positions available for international medicalgraduates (IMGs) has increased from seven in 2004 to 15in 2009. Not all available positions for laboratory medi-cine trainees are filled through the CaRMS match. In thepast 5 years the percentage of positions filled from theprimary match ranged from 59% in 2004 to 84% in 2007(69% of the positions were filled in 2008 after the pri-mary match). Table 2 shows the number of available andfilled positions for the various areas of laboratory medi-cine since 2002.

F EATU R E


Laboratory Medicine ResidencyTraining 2007–2008

Aaron Pollett, MD, FRCPC

Aaron Pollett, MD, FRCPC, is a member of the Professional Affairs Committee, Mount Sinai Hospital, Toronto, Ontario.

Table 1. Number of Laboratory Medicine SpecialtyPositions (Allocated Quota) in the 2009 CaRMS

Discipline Canadian IMGGraduates Positions

Anatomic pathology 24 1

General pathology 1 1

Hematopathology 3

Laboratory medicine 19 13

Medical biochemistry 5

Medical microbiology 9

Neuropathology 4

Total 65 15

CaRMS = Canadian Residency Matching Services; IMG = internationalmedical graduate.

Table 2. Adjusted Allocated Positions (quota)/Accepted Positions for Canadian Graduates in Primary CaRMSMatch, 2002–2008, for Laboratory Medicine Specialties

Discipline 2002 2003 2004 2005 2006 2007 2008

Anatomic pathology 8/3 10/5 11/3 8/5 20/14 19/14 15/11

General pathology 3/0 3/1 3/0 1/0 1/0 1/0 1/0

Hematopathology 1/0 1/0 1/0 1/1 2/1 1/1

Laboratory medicine 13/11 17/15 20/19 18/14 20/14 26/24 23/15

Medical biochemistry 1/1 3/3 4/4 3/1

Medical microbiology 1/1 1/1 1/0 1/0 7/6 7/6 8/8

Neuropathology 1/0 1/0 1/0

CaRMS = Canadian Residency Matching Services.

P o l l e t t


If we assume that all positions filled through the primaryCaRMS match stay within the program, and that allaccepted IMGs are in the program for 5 years, the num-ber of laboratory medicine trainees estimated in 2008 was222. This is close to the 240 residents currently listed inthe CAP manpower database. There has been an approxi-mate year-over-year growth of 29% in 2007 and 12% in2008 and a 70% increase in the number of laboratorymedicine residents compared to 5 years ago (140 patholo-gy residents were in the CAP database in 2003).Table 3 shows the breakdown of current pathology resi-dents by discipline and training level (PGY year). Of the240 residents in the database, 87 (36%) are indicated ashaving entered through the IMG program. Approximately63 residents are expected to be graduating from residencypositions in July 2008.There is rising concern that there will not be sufficientpositions for the laboratory medicine residents when theycomplete their training. While the number of availablepositions with the residency programs has increased by70% in the past 5 years, the number of full-time practic-ing laboratory medicine professionals has increased byapproximately 5%. The majority of this increase in theworkforce is the result of filling vacant positions. Therehave been very few new positions created for laboratorymedicine professionals in the past 5 years. The majorityof full-time practicing pathologists in Canada are nowover 50 years of age, with an average age of 53.5. This haslead to the concern that there will be a shortage of pathol-

ogists within the next 5–10 years as the number of full-time laboratory medicine practitioners retire. This retire-ment replacement need is very difficult to estimate. Thereis no mandatory age of retirement and approximately12% of the current full-time pathology manpower is overthe age of 60.The incidence of cancer has increased by approximately4% per year for the past 5 years. This increase is largelydue to the growing and aging population within Canada.Screening programs and cancer wait-time improvementhas become a focus of the provincial cancer care organi-zations. The rising incidence of cancer and other cancercare initiatives have increased the workload withinpathology laboratories with essentially no increase in thepathology workforce. Alberta has been successful in thecreation of several positions in the past few years, but thishas been in response to a growing population and a sys-tem strained by large cutbacks which occurred in the1990s. In Ontario, there have been less than five new posi-tions in the past 5 years, despite large institutions demon-strating an increase in workload by 20–25%. Our trainingprograms have been very successful in growing the labo-ratory medicine residency programs, and we are in a posi-tion to increase the pathology workforce in Canada withhighly trained laboratory medicine professionals. Thereis a need to increase the number of laboratory medicinepositions across the country to handle the increasingworkload and ensure that all Canadians have access tohigh-quality laboratory medicine.

Table 3. Number of Laboratory Medicine Residents Currently Being Trained in Canada, by Discipline and Program Year

Discipline PGY-1 PGY-2 PGY-3 PGY-4 PGY-5 Total

Anatomic pathology 29 40 33 27 46 175

General pathology 5 6 5 10 7 33

Hematopathology 0 3 2 3 4 12

Medical biochemistry 2 0 0 3 0 5

Medical microbiology 2 3 2 1 5 13

Neuropathology 1 0 0 0 1 2

Totals 39 52 42 44 63 240

FEATURE


The origins of surgical pathology as a specialty were

thought to be born in the immediate aftermath of

the Renaissance. At the same that chiaroscuro emerged

with its interpretations of light and dark embodied in

the art of tenebrism, so too the appreciation of the sub-

tleties and nuances of microscopy were stirring and tak-

ing a meaningful shape in pathology. Through the post-

Renaissance period, surgical pathology grew and was

adorned by beautifully crafted descriptions of diseases

and tumours. The art of surgical pathology expanded

and refinements of description devoted to morphology

were the initial landmark changes. From its romantic

Renaissance origins, surgical pathology in the mid-20th

century had evolved into a pivotal specialty on which

treatment options were based. The surgical pathologist

was a fully fledged member of the management team,

and intra-operative consultations became the daily

norm, especially in North America. Surgical pathology

remained somewhat static into the latter parts of the

20th century with electron microscopy and then

immunohistochemistry taking centre stage as the new

development that would impact diagnostic practice.

The past 15–20 years have seen pathology undergo its

own form of the industrial revolution, with the emer-

gence of molecular biology and pathology as an irre-

sistible force that has changed the way we think about

and practice pathology. Luddite thinking has been shed

slowly and painfully at first, but the inexorable march,

hearteningly, is being embraced more readily. With this

have come more challenges and difficulties for the surgi-

cal pathologist. The milieu and landscape in which we

now practice has changed, and with this change the surgi-

cal pathologist and indeed pathologists of all persuasion,

are confronted by a series of new questions and applica-

tions.

The ChallengesAs if technological advances were not enough, surgicalpathology has found itself in the midst of a maelstromof public mistrust, especially here in Canada. Public per-ception of medicine in general has changed and with anincreasingly sophisticated and informed general publicthe ambiance is now shrouded by litigation and secondguessing. Somewhere along the line, we as a professionhave become inured to the opinion of the public, abouthow they are treated, are they getting the best treatment,and what is in their best interests. While this has notbeen a wilful neglect of the very people we care for, ithas been an imperceptible reality.

As a profession we need to stand back, be introspective,and institute the necessary checks and balances thatensure optimal diagnostic accuracy and hence the bestpatient care. While it is true that errors have been madeand these are regrettable, it is appropriate to keep inmind that errors are unavoidable. However, as a special-ty we have to ensure that the best possible steps are inplace to minimize errors and harm to patients. We mustbe mindful of public perception at all times and respon-sible, rational behaviour is mandatory at all times fromall parties concerned. Whistle blowing is an acceptableform of drawing attention to a problem if done properlyand not sensationalized. It is an unfortunate dictum oflife that we dwell on the negatives, forgetting the posi-tives that far outweigh the former. While we must strivefor perfection, it must be borne in mind that surgicalpathology is a subjective field with personal bias like theinterpretation of the shades of light and dark ofchiaroscuro. Problem and/or difficult areas abound insurgical pathology, and herein lies the challenge: to min-imize the difficult areas and replace subjectivity withobjectivity.

Surgical Pathology: Still the Jewel in the Crown?

Runjan Chetty, MB BCH, FRCPath, FRCPC, DPhil

Runjan Chetty, MB BCH, FRCPath, FRCPC, DPhil, is with the Department of Pathology, Laboratory Medicine Programme, University HealthNetwork and University of Toronto, Toronto, Canada.

C h e t t y


Maintenance of Standards and Safe PracticeThe licensing and examining bodies have a set of stan-dards that have to be met before anyone is allowed topractice pathology. These standards should constantly bemodified and adapted so that the modern surgical pathol-ogist is in tune with the demands of modern day medi-cine. Regular maintenance of knowledge is required andmaintenance of certification is to be supplemented by re-certification or re-validation, a more rigorous process thatwill ensure fitness to practice. In the interest of goodpractice and transparency it is a development that mustbe welcomed by all medical practitioners. The vexingissue currently is the mechanism and implementation ofthis goal. It is incumbent upon us to ensure that a fair,just, and efficacious method is found as it is in the bestinterests of everyone.

In the meantime, several other strategies are in place toensure quality. Quality assurance is at the forefront of alllaboratories, both at internal and external levels.1 Stepsare in place to monitor all aspects of quality from slidegeneration, to the final end product which is the surgicalpathology report. Internal consultation between col-leagues on all types of cases is routine and is encouraged.In an attempt to ensure accuracy and as part of internalconsultation, several centers also ensure that two patholo-gists view and concur on every cancer diagnosis. This hasobvious benefits but may sometimes be impractical. Apotential negative impact is to take away an individualpathologist’s ability to diagnose independently, somethingthat the entire training mechanism is geared towards.Perhaps a variation on this would be that a pre-deter-mined percentage of cancer cases have two pathologistsexamine them. In addition, one must also rely on pathol-ogists having the necessary insight to realize when theyare confronted by a difficult case. Allied to this is randomaudit within a department. A pre-determined percentageof cases (usually about 10%) are randomly vetted by aquality assurance team within a department.Discrepancies are discussed and those with clinical signif-icance result in an amended report and the responsibleclinician being contacted. Clinicopathological meetings(rounds) or multi-disciplinary management meetings arean extremely important facet of quality assurance. Therole of the clinicians and ancillary diagnostic modalitiesin this setting provides essential feedback to the patholo-gist.

Everything that can be done to ensure accuracy and time-ly output should be done within a laboratory.

Turnaround times are monitored scrupulously to ensurethat effective treatment can be instituted anddifficult/problem cases referred to appropriate centres.Other innovations that are now integrated into routinepractice in some institutions are standardized synopticreporting which ensures that key information is transmit-ted to treating physicians in a uniform manner.Laboratory information systems should be fully comput-erized allowing for swift transfer of results and reports,and information to be placed in cancer repositories forplanning and epidemiological studies. External qualityassurance is also fairly routine and many laboratoriesundergo strenuous external review of all facets of theiroperation before achieving accreditation. It will be obliga-tory to participate in some form of external quality assur-ance across the board in due course. The College ofAmerican Pathologists and the Royal College ofPathologists in the United Kingdom have establishedguidelines regarding requirements for accreditation, bestpractice and multidisciplinary team meetings.2 In addi-tion, the Association of Directors of Anatomic andSurgical Pathology have also published recently on qualityassurance and improvement in surgical pathology.3

Sub-specializationThe expansion of knowledge has meant that the need forsub-specialization is greater than ever before. Not only arenew therapeutic molecules that impact on patient man-agement discovered regularly, but smaller lesions arebeing identified by sophisticated radiological techniqueswhich means surgical pathologists have to be acutelyaware of early and/or extremely subtle lesions that mayhave clinical consequence. This brought a new kind ofpressure for the general surgical pathologist: how doesone maintain sufficient expertise in all fields? There iscompelling evidence to support sub-specialty expertise sothat every patient has access to the best opinions. Thebalkanization of surgical pathology is inevitable.However, it must be borne in mind that only a small, yetsignificant, proportion of cases need expert opinion:remember, common things are common. Under no cir-cumstances should a so-called “triage pathologist” emergeout of fear or a headlong dash for expert opinions. Aboveall, the individual surgical pathologist needs to haveinsight: when someone is out of their depth and when tosolicit another opinion cannot be mandated. However,advances in telepathology have meant that time and dis-tance are no longer rate limiting factors to obtaining asecond opinion. Regionalization of specialized cancer

services has also meant that quality is ensured but distinctreferral pathways will also aid in the acquisition of secondopinions. It is idealistic to think that all surgical patholo-gists will be au fait with all specialties and the nuanceswithin each. Training should be geared to preparingpathologists with an overall understanding of what is need-ed for a modern surgical pathologist. The specialty is wellcatered for in terms of meetings, conferences, and updatesthat provide an excellent opportunity to upgrade knowl-edge. Electronic means of acquiring knowledge is at handwith numerous educational resources from dedicated web-sites to journal being available at the click of an icon.

We have to ensure that colleagues are not isolated becauseof geographical factors in a large country such as Canada.It is not unrealistic to suggest that remote solo practicesare not to be encouraged. If unavoidable, then measuresneed to be in place to allow ready access to colleagues,and telepathology is the ideal mechanism to make thispossible.

The FutureThe future of surgical pathology is bright, enticing, andfull of promise. It attracts the brightest minds as it stillhas the mystique that only a blend of art and science can

conjure. The ingredients have been made more intoxicat-

ing as we stand squarely at the interface of applied and

basic molecular biology (“translational pathology”). The

importance of morphology should not be underscored in

the frantic pursuit of a molecular solution. Instead, it is

becoming abundantly clear that the two are inextricably

linked with both having an impact on each other. Both

should be harnessed in a meaningful manner that will

allow for the progress that we all demand. There is much

to be proud of in terms of what has already been achieved

in pathology. As a specialty we do have some daunting

challenges that are not insurmountable but which can be

addressed with the openness, honesty, and élan that befits

the jewel in the crown of modern medicine.

References1. Parham DM. Are external quality assessment (EQA) slides

schemes a valid tool for the performance assessment ofhistopathologists? Pathol Res Pract 2005; 201:117–21.

2. Association of Directors of Anatomic and Surgical Pathology.Recommendations for quality assurance and improvement in sur-gical and autopsy pathology. Am J Clin Pathol 2006; 126:337–40.

3. College of American Pathologists. Accreditation and laboratoryimprovement. Northfield (IL): Author.http://www.cap.org/apps/cap.portal?_nfpb=true&_pageLabel=accreditation.

S u r g i c a l P a t h o l o g y : S t i l l t h e J e w e l i n t h e C r o w n ?


The Incoming Editor-in-Chief

Dr. Godfrey Heathcote,incoming editor-in-chief, ishead of the Department ofPathology at DalhousieUniversity (2004–) and chiefof pathology and laboratorymedicine for the CapitalDistrict Health Authority inHalifax, Nova Scotia.

Dr. Heathcote has a wealth of academic and adminis-trative experience as well as many publications to

his credit related to ophthalmic (and head and neck)pathology, his area of special interest. He is chair of theCanadian Ophthalmic Pathology Society and sectioneditor of general and ophthalmic pathology for theCanadian Journal of Ophthalmology.

Please join me in welcoming Dr. Heathcote. We look for-ward to his stewardship of the journal.

Jagdish Butany

FEATURE


Go to where the puck is going, not where it has been.– Walter Gretzky

With these famous words, Walter Gretzky helped to

shape the remarkable “anticipatory” skills used so

successfully by his son, Wayne, one of the greatest hock-

ey players of all time.1 In the various disciplines of

pathology and laboratory medicine, we are living in an

era of rapid advances in knowledge and unprecedented

opportunities for transforming health care. As a former

residency program director and department head at the

University of British Columbia (UBC) and now depart-

ment chair at the University of Toronto, I have been

asked to comment on the state of post-graduate training

in pathology and laboratory medicine disciplines, focus-

ing on some current trends that might shape our future

as a sustainable and relevant part of the health care sys-

tem. Unlike Wayne Gretzky, I have not developed any

special aptitude for anticipating what lies ahead (despite

having a crystal ball on my office desk); however, I have

had the privilege of working in two outstanding institu-

tions with many highly talented and accomplished col-

leagues, and have learned from those for whom innova-

tion, creativity, renewal, and sustainability are hard-

wired. Moreover, there are a number of trends affecting

areas outside of pathology and laboratory medicine, and

there may be useful analogies that could be applicable to

our profession. In this presentation, I will briefly review

the role of the university in post-graduate training in

the disciplines of pathology and laboratory medicine,

identify some current developments for which there is

rationale for devoting thought and attention, and specu-

late on what the future might look like if we answer the

call to action for ongoing development and improve-

ment to what we do and who we are.

Role of the UniversityA major conceptual framework that emerged some yearsago was for the accreditation of residency training pro-grams of the Royal College of Physicians and Surgeons ofCanada to run through universities, as opposed to indi-vidual hospitals, institutes, or agencies. I am not aware ofthe original basis for creating this particular framework,nonetheless it has had profound implications in shapingthe relationship between clinical practice (a responsibilityof bodies affiliated with the university) and the academicmission of a university (creating and conveying knowl-edge through research and education). An easy way to

Richard G. Hegele, MD, FRCPC, PhD, is with the Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario.

Post-graduate Training in Pathologyand Laboratory Medicine:

Time for a Paradigm Shift?

Richard G. Hegele, MD, FRCPC, PhD

ABSTRACT

Post-graduate training in the various disciplines of pathology and laboratory med-icine must be relevant to serve current and future needs of society. We are livingin an age of unprecedented technological advancement, data generation, andincreasing accountability. This article examines the basis by which universities, aspart of their academic mandate, participate in post-graduate education, highlightssome current trends that are impacting the profession, and proposes an alterna-tive approach of “consilience” as a new paradigm for us to position ourselves as acrucial part of a complex health care system.

think about this is that the university is not in the busi-ness of patient care, and the roles and responsibilities ofeach of the parties are articulated in affiliation agree-ments signed between the university and its partners.While clear for Royal College residency training pro-grams, the role of the university has not entirely beendefined for the various post-PhD diploma programs inclinical laboratory science, some of which are runthrough the university and some of which are not. In myopinion, this is an area that requires revisiting and discus-sion with government, for harmonization and clarifica-tion of expectations and performance. In addition,despite universities being charged with accreditation ofresidency training programs, there have been episodes ofmisunderstanding on the part of universities in recogniz-ing post-graduate teaching by faculty, for example in thecontext of academic promotion. In this area, recent trendsare encouraging, as universities seem to be becoming bet-ter “educated” about post-graduate teaching, and academ-ic departments are doing a better job of “packaging” theseactivities in a manner understandable by the greater uni-versity.

Some Current Trends Facing Our ProfessionPersonalized Medicine and TranslationalResearch

Through the late part of the 20th century, the prospective,double-blind randomized clinical trial was considered tobe the “gold standard” for evaluating the effectiveness ofnew therapies in comparison to usual treatment.2 Thismodel provided a robust scientific and statistical founda-tion to the evaluation of innovative therapies and repre-sented a marked advance over previous activities such asanecdotal case reports or uncontrolled case series. By con-trast, current indications are that we are reaching (or per-haps have already reached) a plateau for the utility of therandomized clinical trial, manifested by lower expecta-tions by which the “success” of new pharmaceuticals isconsidered more in terms of lack of inferiority, ratherthan superiority, when compared to standard treatment.Pharmaceutical companies are experiencing difficulties indeveloping and bringing to market so-called “block-buster” compounds that served the industry and its share-holders so well for many years, further indicating thatnew approaches are needed. In 2003, a worldwide vicepresident at GlaxoSmithKline stated, “The vast majorityof drugs – more than 90% – only work in 30 or 50% ofthe people.”3 This underscores the need for better predict-ing who is likely to respond to a given therapy, as well as

who is likely to experience side effects. With the comple-tion of sequencing of the human genome, we are now inthe “post-genomic era” in which gene-based and other “-omics” (i.e., proteomics, metabolomics, etc.) approachesare being aggressively developed and pursued to fulfillthe promise of personalized medicine: “giving the rightperson, the right treatment, at the right time, in the rightamount.”4 In addition to predicting effects of therapy, thediscovery, and validation of biomarkers of disease haveprofound implications for risk stratification (“pre-diag-nosis”) and more refined diagnosis and prognosis of dis-ease (“theranostics”). Our profession is particularly wellpositioned to accelerate “translational research,” definedhere as the process by which discoveries are validated,qualified, and implemented into practice for improvedhealth outcomes. In addition to the intellectual propertyand commercialization aspects that tend to dominatediscussions surrounding translational research, we arepoised to contribute to the growing “evidence base” ofnewly developed and validated biomarkers, to larger pur-poses such as cost-benefit analyses and informing healthcare policy and legislation. This represents a markedadvance over the previous tendency where new discover-ies were promoted loudly, with lesser regard to potentialdownstream implications. For example, if one considersthe implementation of molecular laboratory testing forthe SARS-associated coronavirus, the assay itself is nodoubt robust and satisfactorily validated, yet going for-ward, how many clinical specimens would have to betested, at what cost, before there is a single true positiveresult? Our profession has the requisite expertise in cru-cial areas of pre-test and post-test likelihoods, andinsight into pre-analytical, analytical and post-analyticconsiderations regarding laboratory data. This gives uslegitimacy and relevance for which our post-graduatetrainees can become expert, and continue to build andrefine.

Biobanking

Exquisitely phenotyped and annotated biospecimens arethe fuel of transitional research. Pathology and laboratorymedicine is the obvious steward of human biospecimens,and initiatives such as the BC BioLibrary5 provide anopportunity for our profession to exhibit innovation andleadership in the design, standards, and public engage-ment related to biobanking. We are living in a time ofincreasing accountability to our funders and to the pub-lic, and it behooves us to be visible front and centre withrespect to biobanking: the risk is that we will becomemarginalized or forfeit our legitimate and highly value-

P o s t - g r a d u a t e T r a i n i n g i n P a t h o l o g y a n d L a b o r a t o r y M e d i c i n e : T i m e f o r a P a r a d i g m S h i f t ?


H e g e l e

35Canadian Journal of PathologyJuly 2009 35Canadian Journal of Pathology 35

added role if we choose not to lead. Post-graduatetrainees need to appreciate the gravity of these considera-tions and themselves become pro-active and assume lead-ership as we progress toward personalized medicine.

Disruptive Technologies

Pathology and laboratory medicine has a rich traditionof applying new technologies to examine diseases inunprecedented ways. In anatomical pathology, the hema-toxylin and eosin stain done on sections prepared fromformalin-fixed, paraffin-embedded tissues has changedlittle since the 19th century; this technique remains thecornerstone for rapid and inexpensive screening anddiagnosis of disease. Over the years, advances in specialstains, histochemistry, electron microscopy, immunohis-tochemistry, and confocal microscopy have been effec-tively implemented to refine and improve the capabilitiesof conventional light microscopy. In recent years, nucleic-acid based technologies such as in situ hybridization andthe polymerase chain reaction have transitioned from theresearch to the clinical laboratory setting. For example, inthe diagnostic virology laboratory, viral culture isbecoming a technique of the past. Technologicaladvances have also provided increasing automation andrange of applications that often blur the boundariesbetween traditional laboratory medicine disciplines. Amass spectrometer for proteomics can be used for suchdiverse purposes as cancer diagnosis, microbiology,metabolic disease and organ transplantation. In addition,newer imaging modalities such as micro CT,6 in whichthe resolution rivals that of light microscopy, are begin-ning to blur the boundaries between radiology andpathology/laboratory medicine. What such technologicalconvergence means for the training of the next genera-tion of pathologists, laboratory physicians, and scientistsis that acquisition of sound fundamentals in the underly-ing assumptions and factors (pre-analytical, analytical,and post-analytical) that go into the production of a“result” are paramount. Having such a knowledge base isimportant to distinguish the domain expertise from thatof our other colleagues or the lay public, who may alsohave access to data and information, but not the addi-tional background and perspective that goes into mean-ingful interpretation. Our training programs have tomeet the challenge of ensuring that graduates understandthese considerations and can continue to do so with theadvent of future disruptive technologies. This approachis analogous to the major shift in emphasis that occurredin legal education toward the end of the 19th century,when law school curricula changed from having students

memorize every statute on the books to learning how to“think like lawyers think” (today this is known as prob-lem-based learning).7 It took many decades for a similarchange in emphasis to occur in undergraduate medicaleducation, which involved a transformation from rotememorization of medical facts, to studying pathogenesisas a means for rationally approaching diagnosis andtherapy, to having the ability to access and interpret amyriad of data and information (see below). For post-graduate education, emphasizing solid fundamentals andbeing able to understand new advances are crucial toensuring graduates will be well prepared for long-termcompetence and relevance.

Electronic Networking

It is self-evident that post-graduate training has to be rel-evant to “real-world” practice. Within Canada, there iscurrently a wide diversity of practice models, and what isrelevant to one geographic region (and its associated cul-ture and other characteristics of disease prevalence, inci-dence, referral patterns and the like) may have little or norelevance elsewhere. While the setting of any individualpractice may have unique considerations, any practicemodel must enable acceptable performance that isaccountable to the standards of the profession. Withtechnological advances in videoconferencing andtelepathology, the world is becoming increasingly linkedand networked, and this has potential for more effective-ly transmitting new knowledge for continuous improve-ment. Examples showing the capability of high-speed,broadband connectivity include the UBC medical under-graduate curriculum, which, in the face of expansion todistributed sites on Vancouver Island and in NorthernBritish Columbia, necessitated implementation of inno-vative technologies so that students at each site wouldhave a comparable educational experience, in fulfillmentof accreditation standards. The electronic videoconfer-encing and distance education approaches in the UBCmedical undergraduate curriculum have performed verywell in terms of overall functionality and reliability. Withthe upcoming expansion to the Okanagan, the look andfeel of the technology will remain consistent. In Toronto,at the University Health Network (UHN), approximately17 hospitals in Ontario have high-speed, broadbandtelepathology connection to the UHN department. Thereis considerable interest in evaluating the impact of thistelepathology system on indices of patient care, withexploration of education and research potential still inthe early days.

Consilience: A Path to Wisdom?

The systems theorist and professor of organizationalchange, Russell Ackoff, has posited that the human mindcan be classified into five categories: data (raw material,symbols); information (processed data that answers“who,” “what,”“where,”“when”-type questions); knowledge(application of data and information to answer “how”questions); understanding (appreciation of “why”), andwisdom (evaluated understanding).8 Of these categories,wisdom is the pathway toward creating the future, andgiven what is at stake for our profession, the question is,how do get there? In the academy, the time-honoured wayto attain wisdom has been through increased specializa-tion (i.e., being “the big fish in a small intellectualpond”), but there are problems with this approach. Pondstend to shrink and dry up due to irrelevance (“knowmore and more about less and less”), fragmentationincreases and overall this compromises the ability to tack-le the really big, important issues. As a response to thistrend, funding agencies have encouraged increased net-working and cross-talk between disciplines, with forma-tion of multi-disciplinary, inter-disciplinary, and trans-disciplinary teams.9 Consilience, from the Latin “jumpingtogether,” is an approach for the unity and integration ofknowledge through synthesis and is a direct contradis-tinction to the paradigm of increased specialization.10

With technological advances and increased automation,data are being obtained faster than they can be interpret-ed. Systems biology focuses on understanding the rela-tionships between the components of a biological system,requiring a new skill set that goes above and beyondacquiring data and trying to make sense of information.11

Pathology informatics12 offers a compelling approach tothis situation, for which training will result in the nextgeneration of our laboratory physicians and scientistsbecoming more adept at the type of abstract reasoningneeded to derive meaning from the mounds of data beinggenerated by ever-increasing sophisticated technologies.

Concluding RemarksIn summary, I have attempted to anticipate “where thepuck is going” based on several trends that promise to

profoundly impact the future of pathology and laboratorymedicine and which have implications for both the struc-ture and content of post-graduate training. Our profes-sion has an admirable track record of being able to adaptto change: the challenge is to deal with a world in whichknowledge is rising exponentially. I propose that con-silience provides a possible means for us to achieve wis-dom and be vibrant leaders who shape the future in a rel-evant and compelling manner.

AcknowledgementsSpecial thanks for Dr. Malcolm Silver, who encouragedmy career development at crucial stages; Dr. JamesCullen, for his inspirational example; Dr. James Hogg, anoutstanding role model and mentor for embracing inno-vation and change; and Dr. David Hardwick, for helpingme learn to see the world with curiosity and discipline.

References1. Rosenfeld J. CDU to Gretzky: The puck stops here. Fast Company,

Issue 36, 2000; http://www.fastcompany.com/magazine/36/cdu.html?page=0%2C1. Accessed May 17, 2009.

2. Wikipedia. http://en.wikipedia.org/wiki/Randomized_con-trolled_trial. Accessed May 17, 2009.

3. Connor S. Glaxo chief: Our drugs do not work on most patients.The Independent 2003 Dec 8; http://www.independent.co.uk/news/science/glaxo-chief-our-drugs-do-not-work-on-most-patients-575942.html. Accessed May 17, 2009.

4. Proof Centre of Excellence Home Page. www.proofcentre.ca.Accessed May 17, 2009.

5. BC Biolibrary Home Page. www.bcbiolibrary.ca. Accessed May 17,2009.

6. Wikipedia. http://en.wikipedia.org/wiki/Microtomography.Accessed May 17, 2009.

7. Bok D. Our Underachieving Colleges: A Candid Look at HowStudents Learn and Why They Should Be Learning More.Princeton, NJ: Princeton University Press; 2006:109–45.

8. Systems Thinking Homepage. http://www.systems-thinking.org/dikw/dikw.htm. Accessed May 17, 2009.

9. Wikipedia. http://en.wikipedia.org/wiki/Interdisciplinary.Accessed May 17, 2009.

10. Brean J. The anti-specialists. National Post 2009 Apr 18; p. A6.

11. Wikipedia. http://en.wikipedia.org/wiki/Systems_biology.Accessed May 17, 2009.

12. Association for Pathology Informatics Homepage.http://www.pathologyinformatics.org/. Accessed May 17, 2009.

P o s t - g r a d u a t e T r a i n i n g i n P a t h o l o g y a n d L a b o r a t o r y M e d i c i n e : T i m e f o r a P a r a d i g m S h i f t ?


FEATURE


Research in pathology has been and continues to be acornerstone of pathology. Over 150 years ago,

Rudolf Virchow carried out observational research onthe clinical tissues he and his colleagues obtained frompatients with disease. He applied histology to pathologyand came to the conclusion that the cell regulates thephysiology and pathology of the human being.1 Withsimple tools and very keen observational skills, he wasable to begin the modern era of research into the patho-genesis of disease. Over the years, research has alwaysbeen a component of the practice of pathology andpathologists developed and embraced state-of-the-arttechnologies to further their studies into pathogenesis.As medicine advanced, organized pathology developedresearch, education and clinical platforms which favouredthe strong presence of scholarship and knowledge cre-ation. These platforms became embedded in the emerg-ing practice of anatomic pathology as it became anessential patient care speciality in medicine

Drivers of Pathology ResearchToday, research in pathology takes on many formsextending from basic to translational to clinical investiga-tions. The common thread in all these is that the inspira-tion and ideas to drive the research originates from thecells, tissues, and body fluids pathologists examine intheir clinical practice.2 This careful in-depth study of clin-ical material to describe and characterize the pathology isthe first step along the research path. This has been themainstay of research in pathology and has been invalu-able to medical progress in general. Today pathologistsinvestigate tissue, cells, and molecules to provide an inte-grated understanding of the pathobiology of disease. Thegoal of the pathologist then is to ask innovative questionsand to pursue answers through research. This is donethrough independent research programs organized andled by individual pathologists as principle investigators(PIs) and/or by collaborative group or team initiatives in

which pathologists are the active members. Pathologistswith high quality subspeciality training make excellentleaders of research programs. The research programsthemselves must be well funded to carry out the expen-sive state-of-the-art research. However observationalstudies and clinical-pathologic studies can be done withvery little funding and have proven valuable. The aim ofresearch in pathology is to transform pathology practicefrom purely diagnostic to predictive and curative.

Subspecialization Enhances ResearchA focus on subspecialty pathology practice allows pathol-ogists to create research platforms around their clinicalpractice. As consultants, pathologists need to have in-depth, state-of-the-art knowledge about health and dis-ease. In the modern medicine practiced in Canada today,it is impossible to keep up with all areas. To continue tobe truly effective, subspecialization has become a neces-sary way to strengthen the quality of practice. The forma-tion of site groups and subspecialty groups not only pro-vides excellent ways to deliver consultations to clinicalcolleagues, but also educational and research platforms.Subspecialization provides high volume case material,time for in-depth study and excellent opportunities tocarry out meaningful studies on mechanisms of disease

Translational Research and PersonalizedMedicineTo be at the forefront of translational research patholo-gists need to be well trained in this area of research. Sincepathologists sit at the crossroads of basic science and clin-ical medicine, they will need to be agents of change. Theyhandle human biological materials – tissues, cells, serums,other fluids – and as custodians of this material, patholo-gists need to ensure that this valuable research resource isutilized it in an efficient and effective way to promotehigh quality health care. This requires training programsin translational approaches at the graduate and fellowship

Research in PathologyAvrum I. Gotlieb, BSc, FRCPC, MD

Avrum I. Gotlieb, BSc, FRCPC, MD, is with the Department of Laboratory Medicine and Pathobiology, University of Toronto, and the Departmentof Pathology, Toronto General Research Institute, University Health Network, Toronto, Ontario. Correspondence can be directed [email protected].

levels as well as investments in institutional infrastructureto support this research and develop new technologiesthat will lead to new basic discoveries that can be movedto clinical practice.

As we discover more about the abnormal biology of dis-ease, we are beginning to understand the heterogeneity ofdisease in individual patients in order to develop cus-tomized treatment. This will require sound research at alllevels of investigations. Laboratory physicians have theopportunity to be leaders in the use of both cell andmolecular biology and imaging to investigate disease.3

There have been successes although progress is slow fornew discoveries to move into mainstream clinical prac-tice. Pathologists are at the forefront of personalized med-icine and are making the investment in time, energy, andmoney to pursue this goal with vision and vigour. Toaccomplish this, laboratory physicians will require highlyspecialized knowledge in the cell and molecular biologyof human disease and be willing to engage in interdisci-plinary research. This requires adjustments to the curric-ula of speciality and subspecialty training programs. Toachieve robust programs in research in pathology, excel-lent well-funded research fellowships need to be madeavailable in Canada to train pathology and laboratorymedicine subspecialists at our Canadian training centres.Any time we send our trainees out of the country to train,we run a serious risk of losing them to other jurisdictions.We have the capacity to train fellows in Canada and thisshould be exploited by provincial and federal health careproviders. These training platforms should enable resi-dents and fellows the opportunities to become welltrained and make new discoveries in molecular laboratorymedicine even at early stages in their careers.

Maintaining a Research AgendaTo maintain Research in Pathology programs, stablefunding sources are required that keep up with inflation.Funding is needed in five main categories: (1) operating

grants, (2) equipment, (3) training grants, (4) personnel(principle investigator/faculty) awards, and (5) overhead.The Canadian Foundation for Innovation has been effec-tive in providing state-of-the-art, expensive equipmentfor big science. The Canada Research Chair program hasprovided much needed salary support for Canada’s bestbiomedical scientists. Canadian Institutes of HealthResearch provide overhead for operating research grantsbut at an average of 20% which is not nearly enough tofund the ongoing costs of maintaining the infrastructureof research facilities. We need to provide the intellectualand the financial infrastructure to support an importantgroup of biomedical investigators, physician-scientistsand physician-investigators. Pathologists are in a positionto explain to the lay community, to politicians and tocommunity leaders that there is much value in biomed-ical research. Pathologists need to look for partners ingovernment, academia, industry, and patient advocacygroups to move the research agenda forward. Pathologistsshould accept the advocacy challenge and act to make thecase of biomedical research.

Research in pathology is broadening its domain. The gapbetween Academic Health Sciences Centers (AHSC) andthe community setting is narrowing and research inpathology is extending beyond AHSC. Community hospi-tals are now providing services that were once the exclu-sive domain of fully affiliated teaching hospitals. Withthese changes, one can envision subspecialty researchteams working side-by-side with community laboratoryphysicians outside the domain of the university setting toinvestigate pathobiology of disease.

References1. Virchow RLK. Cellular pathology as based upon physiological and

pathological histology, 7th edition. New York: R.M. DeWitt; 1913.

2. Liu AC, Joag VR, Gotlieb AI. The emerging role of valve interstitialcell phenotypes in regulating heart valve pathobiology. Am JPathol 2007; 171:1407–18.

3. Coleman WR, Tsongalis GJ. Molecular pathology: the molecularbasis of human disease. Amsterdam: Academic Press; 2009.

R e s e a r c h i n P a t h o l o g y


would both be improved by 2–3 months of mandatoryintroductory pathology rotations giving a broad-basedexposure to autopsy, grossing, microscopy, and the func-tioning of a histology laboratory.

Many junior residents are worried about manpower andresource allocation in pathology, especially the large num-ber of pathology trainees in Canada. Many believe thesupply of pathologists now greatly exceeds demand, animbalance exacerbated by the current recession. Althoughrising cancer rates in an aging and growing populationshould translate to an increase in pathology jobs, thereseem to be absolutely no policy initiatives outside ofNewfoundland to create such positions. In this context, itis no wonder that sigh-filled anecdotes from graduatingresidents about their fruitless job searches have alreadybegan to fray the nerves of junior residents. In a paradox-ical trend, pathology programs across Canada continue toaccept a large number of residents, with Toronto’s pro-gram alone admitting around 13 annually in recent years.Many current residents would prefer to see the number ofresidency positions significantly decreased until marketconditions improve and the demand for pathologistsincreases.

Moving to the subject of important topics in residencytraining, quality assurance/improvement (QA/QI), andlaboratory management seem a logical place to start.QA/QI is crucial to pathology, with recent media coverageof so-called “scandals” only reinforcing this point.Surprisingly, QA/QI training is either minimal or absentin many residency programs. The recent addition of aquality assurance elective at a Toronto teaching hospital isencouraging, but it would be beneficial to our specialty if

FEATURE


As residents in the earliest stage of training inanatomical pathology (AP), our perspectives on res-

idency training are based on our limited personal expe-rience and that of our junior colleagues. Nevertheless,we will attempt to voice junior resident concerns andidentify areas where pathology residency programs mayneed improvement. We will comment on the structureand logistics of pathology residency programs and advo-cate for increased training in a few areas that we feel areimportant.

The Royal College has recently discussed changing thestructure of pathology training, and while the current 5-year AP training framework generally works well, there iscontention among junior residents regarding the rotatingclinical internship year (PGY-1). Some feel it is minimallyrelevant to pathology training and should be eliminated,bringing us into line with American programs, while oth-ers argue that it provides invaluable exposure to clinicalspecialties that have frequent contact with pathologists.Most agree that if the clinical year exists, it should consistof more pathology-relevant rotations such as medical andsurgical oncology and less of areas like general internalmedicine, obstetrics, or general pediatrics. Another con-cern is the frequent absence of pathology exposure inPGY-1, such that a resident can begin PGY-2 unclear onsuch fundamental topics as autopsy procedure, grossing,and the logistics of a histology lab. Without early expo-sure to pathology, it may also be difficult for residents todecide whether they have chosen the right specialty. Ourdiscipline does not need unhappy residents, nor do wewish to lose those truly suited to pathology who just needmore on-service time to realize it. The above situations

Zhongchuan Will Chen, MDCM, is with the Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario.Zhongchuan Will Chen can be contacted at [email protected]. Eric K. Morgen, MD, is with the Department of Laboratory Medicine andPathobiology, University of Toronto, Toronto, Ontario.

Anatomical Pathology Training in Canada:

Perspectives from Junior Residents

Zhongchuan Will Chen, MDCM, Eric K. Morgen, MD

all residency programs incorporated a well-structured,mandatory, 1- to 2-month or longitudinal rotation deal-ing with QA/QI topics. A closely related area is laboratorymanagement, which is clearly relevant to practicingpathologists, but in which residents also receive little for-mal training. Fortunately, a QA rotation could easilyinclude a laboratory management component, such asobservation/participation in the duties of departmentleadership in assuring quality, managing resources, main-taining accreditation, etc. Since most residents will needto fulfill one or more leadership roles at some point intheir careers, such exposure would prove invaluable.

In recent years, technological expansion in pathology hasbegun to accelerate, and since today’s residents embodyour specialty’s future, it is crucial that growth areas arenot neglected in Canadian training. One area showingtremendous promise is molecular diagnostics. With ori-gins in the 1940s, the field is by no means new, but recentyears have seen a proliferation of exciting clinical applica-tions. This has led to expectations of better diagnostictests and disease-tailored therapies, extending to calls for“personalized medicine” from all fronts. The progress ofmolecular diagnostics and its role in pathology practicehave gathered unstoppable momentum and, encouraging-ly, some residency programs have responded, with at leastfour Canadian universities routinely training residents inmolecular pathology. Nonetheless, with 14 pathology pro-grams in Canada, there is ample room for increased focusin this area.

Two other fields that promise to shape the future ofpathology lie at the intersection of pathology with com-puters. The first is digital pathology, the application topathology of digital imaging techniques such as whole-slide scanning, telepathology (already successfully imple-mented in Toronto1) and multispectral imaging2 with thepotential to bring greater efficiencies, instant glasslessconsults, improved QA/QC, and the ability to extractmore relevant information from the same slide. As moreslides become digitized, there will be an increasing rolefor computer-assisted diagnosis, with software programs

accomplishing tedious tasks prone to human error, allow-ing pathologists to focus on higher-level information.3

The second field is pathology informatics, whose impor-tance has been recognized for decades4 and which useselectronic information systems to facilitate various labo-ratory operations, from accessioning to reporting toretrieval of information for research and teaching.Internationally, there are a growing number of dedicateddivisions and training programs in the area.5 This is incontrast to the Canadian situation, where residents haveno options for in-country training and only minimalresearch opportunities in pathology informatics or digitalpathology.

In an era of accelerating change in medical knowledgeand technologies, it is not sufficient for universities tomaintain training programs at a level that previously pro-vided an exemplary education. To sustain the vitality ofour specialty and build it a better future, we must accom-plish a “technological revolution” similar to what hasoccurred in radiology. We must confidently embrace newtechnologies and use them to provide the best clinicallyrelevant, patient-centred diagnostic information in a col-laborative and highly integrated environment with radiol-ogists and clinicians. This vision of true 21st century lab-oratory medicine can only be realized if we have the fore-sight to train our residents today in the technologies andmethods that will be the mainstay of tomorrow.

References1. Evans AJ. Making the transition to digital pathology. Healthcare

Information Management and Communications Canada 2008;22:8–10.

2. Levenson R. Putting the “More” Back in morphology: Spectralimaging and image analysis in the service of pathology. ArchPathol Lab Med 2008; 132:748–57.

3. Feldman MD. Beyond morphology: Whole slide imaging, comput-er-aided detection, and other techniques. Arch Pathol Lab Med2008; 132:758–63.

4. Buffone GJ, Beck JR. Informatics. A subspecialty in pathology. Am JClin Pathol 1993; 100:75–81

5. Association for Pathology Informatics. Training opportunities inpathology informatics. Association for Pathology Informatics,2008; http://www.pathologyinformatics.org/InformaticsTrainingPrograms.htm. Accessed May 2, 2009.

C h e n a n d M o r g e n


FEATURE


Often, when talking with one of my staff patholo-gists, the conversation sometimes veers into, “When

I was a resident ... we would come in at 6 am, gross1,000 specimens, then do 12 autopsy cases, and wouldn’tleave the hospital until midnight.” OK, I’m being face-tious; but often as a resident today, I’m always made tofeel like when the staff pathologists were residents,things were harder and busier. Well, I’m writing thisessay to illustrate that residency today is just as challeng-ing as it was before – just different.

The expectations of a senior resident are high and varied.Let’s first start with acquiring the expected knowledge baseto pass the Royal College exam, become a “competentpathologist,” and gain those beautiful five letters after ourname. Learning all of anatomical pathology is no smalltask and it seems like on a daily basis a new diagnosticentity is being discovered, renamed, or reclassified. It’s nolonger basal cell carcinoma with squamous differentiation,but rather metatypical BCC. Also, for such a small organ,the thyroid certainly has many types, sub-types, and vari-ants of carcinoma! Then, add into the mix, molecularpathology. Seriously, how many chromosome deletions,translocations, and inversions can one person remember?When we are practising, are we not allowed to Google?

Next on our “to do” list is applying for those all-encom-passing, coveted fellowship positions. To be a competi-tive applicant and score that staff position at the aca-demic centre of our choice, we need to do a fellowship.Sure, we can work in a community centre, with generalsign-out, but is that the way pathology is going? Let’snot forget that pathology has been in the media quite afew times. Things have gone awry in Newfoundland,New Brunswick, Ontario, and British Columbia. Whichprovince is next? It seems to me that pathology is mov-ing towards sub-specialty sign-out, with centralizationof services. So, now that we as residents are absolutelyscared of signing out on our own, of course we want todo a fellowship. In fact, we need to do a fellowship.

Another year of supervision is good for me!

Of course, to get that fellowship position, to get thatstaff position, we need to do research. When, pray tellshall we fit in our numerous research projects, posters,case reports, and abstract submissions? More and more,time management is becoming a necessary skill.Residents have to find the time to do research whilegrossing, signing out, and writing up autopsy reports.No worries, we’ll do it!

Next, we move onto the expectation of teaching. I per-sonally love teaching. It means helping others, which is abig reason I chose medicine to begin with. There is a lotof responsibility with teaching. As my program directortold me, it can be quite stressful, depending on the levelof education of your audience. So, when we have highschool students come through the department, we canpretty much tell them anything! However, when it comesto our peers, we better be prepared for tough questions,even more so for grand rounds and the questions fromstaff. So, preparation for grand rounds, gross rounds,autopsy rounds, journal club, and so on takes time. Wedo need those reference letters, for those fellowshippositions, for that job!

So, last of the expectations are the administrative dutiesthat come along with being a senior/chief resident: callschedule and vacations, stipends, teaching, and prepar-ing junior residents. In short, we need to make sure allis good in the world of pathology residents. Sometimesthat is a tall order, but we do it and we learn from itskills that we will take with us when we do becomestaff, we hope!

All in all, the expectations of a senior resident inanatomical pathology are realistic and “doable.”However, when I become a staff person, I hope that Iwill remember that being a resident is not easy andevery generation deals with its own challenges. Ours arenot harder than those past – just different.

Expectations of a Senior Anatomical Pathology Resident

Marie S. Abi Daoud, MD, MHSc

Marie S. Abi Daoud, MD, MHSc, is with the Department of Pathology and Molecular Medicine at Queen’s University, Kingston, Ontario. She canbe reached at [email protected].

DISPLAY CLASSIFIEDS


Applicants are invited to forward their resume to

Dr. David KogonChief of Staff/Medical Director,

fax (506)857-5545

or email at [email protected].

Permanent position available for a successful candi-

date to join a congenial group of laboratory physicians,

consisting of five anatomical and general pathologists,

a medical biochemist, a haematopathologist and a

microbiologist. We have a full spectrum of anatomic

pathology, cytology and haematopathology in addition

to a very active service in chemistry and microbiology

in a dynamic laboratory accredited by the College of

American Pathologists since 1987. The laboratory will

move to a new facility in Summer of 2009, located near

the surgical suites and intensive care units, and with

much new equipment. There is a potential for a

departmental directorship on rotational basis for an

individual with administrative experience or interest.

Haematopathology or cytopathology experience would

be an asset.

The Moncton Hospital is part of a large network of hos-

pitals, health centres, services and programs under the

direction of Regional Health Authority B which pro-

vides health services to the southern, central, and east-

ern regions of New Brunswick. The 365-bed critical

care and major referral hospital serves communities

throughout New Brunswick, Prince Edward Island and

northern Nova Scotia. It offers services in Trauma,

Neurosurgery, Burns, Oncology, Infectious Diseases,

Neonatal Care, Provincial Child and Adolescent

Psychiatry and Vascular Surgery plus extensive services

in elevated secondary and primary care. The Moncton

Hospital is one of the largest single employers within

metro Moncton, employing 3,000 staff and physicians.

The Department of Pathology and Laboratory Medicine

requires one of the following qualifications: a) certifi-

cation or Fellowship in Pathology of the Royal College

of Physicians and Surgeons of Canada, or eligibility to

write the examination for Fellowship in Pathology; b)

certification by the American Board of Pathology; c)

certification by le Collège des médecins du Québec; d)

membership of the Royal College of Pathologists of

England and Ireland. The salary range is determined by

the New Brunswick Medical Pay Plan, depending on

qualification and experience.

Zone 1 Moncton New Brunswick

GENERAL/ANATOMICAL AND HAEMATOLOGICAL PATHOLOGISTS

FEATURE


We are fortunate to be residents in a pathologytraining program in Canada. Residency programs

undergo stringent review on a regular basis to assure that allprograms meet accreditation requirements. A pathologistgraduating today must not only be a great morphologist butmust also possess skills in immunohistochemistry interpre-tation, molecular and cytogenetic testing, and quality sys-tems in the practice of pathology.

In recent years, we have seen pathology merge closely withthe fields of molecular biology and cytogenetics. In somecases, diagnosis relies, to some extent, on immunohisto-chemistry and molecular and cytogenetic findings, whichhave important implications for prognosis and treatment.Residency training in some centres has responded to thisevolution by increasing our exposure to these fields.Furthermore, opportunities for involvement in research infields including molecular and cytogenetics, are readilyavailable and encouraged. Similarly, recent news headlineshave brought the importance of quality systems in the prac-tice of pathology into focus and training programs arebeginning to respond. Quality assurance and quality controlhas recently become a “hot topic” and has been incorporat-ed into the curriculum through lectures and elective rota-tions. In these and other areas CAP workshops providevaluable learning tools, but could be expanded to includemore of the new technologies as well.

Despite the challenges that the field has faced in the pastcouple years, we believe that the state of pathology trainingin Canada is strong. Residency training is evolving torespond to the changing role of the pathologist and webelieve that most residents generally feel adequately pre-pared for practice. As we approach the end of training welook forward to applying our knowledge.

Unfortunately, as we observe the cohort ahead, it is becom-ing increasingly apparent that the number of job opportu-nities does not seem to be keeping pace with the perceived

demand, number of graduates and residents in training.How can this be when only a few years ago we were toldabout looming retirements and a dire shortage of patholo-gists? Our training program responded logically to this andgreatly increased the number of residents in training for thelast two years. But now we are told that the situation maybecome somewhat analogous to the 1990s when many qual-ified pathologists could not find work and resorted to mov-ing to the United States, pursued further training, orchanged careers. This is a difficult pill to swallow for manyof us with 15 or more years already invested in training andseveral hundred thousand dollars of debt. Why does thesupply and demand for pathologists go through so manypeaks and troughs? Who is doing human resource plan-ning? As a profession should we not show accountability totaxpayers to ensure the money they invested in our educa-tion does not sit as an idle resource at the end of our train-ing? Clearly, it seems that there is not sufficient dialoguebetween the government, universities, and our governingbodies. As we have learned recently in pathology, miscom-munication is the source of many errors!

We are assured that the government plans to fund somepathologist positions but this may still fall well below thenumber of graduates per year from pathology at theUniversity of Toronto alone! Furthermore, fewer patholo-gists are retiring and entry requirements have recently beeneased for American-trained physicians. As we look at thenumber of junior residents in our program, we cannot helpbut wonder if some of us will be without jobs, at the end offive years of rigorous training!

This is a very unfortunate situation. Individuals, universi-ties, and government have invested so much. New pathol-ogy graduates in Canada are highly trained, ready, andeager to enter the work force. We hope that the recent his-tory of the 1990s does not soon repeat itself and weencourage open communication and human resourceplanning to address pathology needs in Canada.

Pathology Training in Canada: A Perspective from Two Senior Residents

Michael D’Agostino, MD, MBA, Andrea Grin, MD

Michael D’Agostino, MD, MBA, PGY-4, general pathology, and Andrea Grin, MD, PGY-4, anatomical pathology, are with the Department ofLaboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada. Michael D’Agostino can be reached [email protected].

Colorectal cancer (CRC) is the third most commoncause of cancer death in the United States1 with

148,810 new cases and 49,960 deaths in 2008.2

Worldwide, CRC is the third most common cancerdeath, after lung and breast.3

CRC with distant metastases is challenging to treat andthe 5-year survival rate is less than 10%.4 The existing,non-targeted chemotherapies are only partially effective.These include the combination of oxaliplatin (a platinumcontaining agent) or irinotecan (a topoisomerase Iinhibitor) with an antimetabolite, 5-fluorouracil (5-FU)and leucovorin, also known as folinic acid (FA).5,6 Sincethe introduction of these non-targeted therapies, the

mortality rate of CRC has decreased by 20–25%4 and themedian survival has increased from 6 months with notreatment, to 14–16 months.7 Many patients, however, donot respond, or become resistant to, combinationchemotherapies administered without additional treat-ment.

Recently, the development of targeted therapy has provid-ed more options for CRC treatment. These include beva-cizumab, a monoclonal antibody against the vascularendothelial growth factor (VEGF), as well as cetuximaband panitumumab, monoclonal antibodies against theepidermal growth factor receptor (EGFR). These drugsare currently used to treat patients with advanced CRC,

FEATURE


Molecular Pathology of Colorectal Cancer:

Identifying the Status of the KRAS Gene for EGFR-Targeted Therapy

Dana Shor, Serge Jothy, MD, PhD, FRCPC

ABSTRACT

The monoclonal antibodies cetuximab and panitumumab against the extracellulardomain of the epithelial growth factor receptor (EGFR) have recently been intro-duced in the targeted therapy of metastatic colorectal cancer (CRC). KRAS is animportant intermediate in the EGFR signalling cascade and is mutated in 30–45%of colorectal cancers. Wild-type KRAS in tumour cells is associated with afavourable response to therapeutic anti-EGFR monoclonal antibodies whereastreatment resistance correlates with KRAS mutations. Therefore, molecularpathology laboratories are increasingly solicited to perform molecular tests toidentify the status of the KRAS gene in CRC tumours. Although KRAS genesequencing remains the reference technique and is part of quality assurance, new tests based on variants of quantitative real-time polymerase chain reaction(qRT-PCR), have been developed which can be performed on formalin-fixed paraf-fin embedded tumour tissues. This review discusses the role of EGFR and KRAS inthe pathogenesis and the targeted therapy of colorectal cancer, and the testing ofKRAS mutations in pathology specimens.

Dana Shor and Serge Jothy, MD, PhD, FRCPC, are with the Department of Laboratory Medicine and Keenan Research Centre of the Li Ka ShingKnowledge Institute, St. Michael’s Hospital, and the Department of Laboratory Medicine and Pathobiology, University of Toronto. Serge Jothycan be contacted at [email protected]

either in combination with chemotherapy or as singleagent for patients resistant to chemotherapy. CRC sur-vival has increased by more than 20 months following theintroduction of targeted therapies.7

A subset of patients responds to the addition of anti-EGFR monoclonal antibodies to standard chemotherapy.The molecular mechanisms involved in resistance to thesedrugs are still incompletely known. In order to reduce thecosts ($30,000 to $160,000 per year, per patient) and sparepatients from unnecessary toxicities (acne-like rash andhypomagnesemia) of anti-EGFR treatments it is essentialto discover molecular markers that will help identifypatients that can benefit from these drugs.8

This article reviews the structure and function of theKRAS gene, how the maintenance of its normal (wildtype) structure in CRC tumours is a reliable molecularmarker for identifying patients sensitive to anti-EGFRmonoclonal antibody therapy, in contrast to patients withtumour harbouring mutated KRAS. It also discusses thecurrent methods to test for the KRAS genotype in a clini-co-pathological setting.

EGFR Activation and RAS SignallingEGFR is a 170 kD member of the HER tyrosine kinasegrowth factor receptor family, and is also known as HER1or Erb-B1.9–11 It is a transmembrane glycoprotein consist-ing of an extracellular, ligand binding domain, a trans-membrane region and an intracellular, ATP binding, tyro-sine kinase domain. Ten different ligands can selectivelybind to the extracellular domain of EGFR. These includethe epidermal growth factor (EGF), transforming growthfactor receptor alpha (TGF alpha), amphiregulin, betacel-lulin, heparin binding EGF and epiregulin.

The EGFR gene is not commonly amplified or mutated inCRC.12–14 It is, however, over-expressed in 65–70% of col-orectal tumours.15–17 Higher EGFR expression correlateswith tumour progression and hepatic metastasis.18

The binding of EGF ligand to EGF receptor activates acomplex intracellular signalling cascade. During the sig-nalling cascade subsequent to ligand binding, EGFRforms a homodimer with another EGFR, or a het-erodimer with another member of the EGFR family.Dimerization activates several tyrosine kinase basedmotifs that initiate transphosphorylation of receptorpairs. Autophosphorylation promotes the binding ofadaptor proteins (e.g., Grb2) that recruit SOS1 a guaninenucleotide exchange factor. SOS1 facilitates the conver-sion of RAS protein from its inactive guanoside diphos-

phate (GDP) binding state to its active, guanosinetriphosphate (GTP) state19 (Figure 1).

Figure 1. RAS mediated signalling pathway: a growth factor binds toEGFR and activates it. This leads to activation of SOS1, which convertsthe inactive RAS-GDP to the active RAS-GTP. RAS-GTP activates sever-al signalling cascades that lead to cellular survival and proliferation.

In its GTP bound state, RAS activates several downstreameffector pathways including RAF/MEK/ERK and PI3K(phosphoinositide 3 kinase)/Akt (Figure 2).20 Thesedownstream effector pathways act together to induce theexpression of cyclin D1 which regulates the progression ofcells through the G1 to the S phase of the cell cycle, pro-moting cellular proliferation, cell survival and apoptosisinhibition.21,22

RAS activation is terminated by GTP hydrolysis viaintrinsic RAS-GTPase activity, which is dramaticallyaccelerated by GTPase Activating Proteins (GAP).23 Oncesignal transduction is complete, EGFR receptors are inter-nalized and down-regulated, or regenerated on the cellsurface.24

KRAS Genes and Their Mutation in ColorectalCancerThe human RAS proto-oncogenes coding for RAS proteinconsists of three principal members: KRAS, HRAS, andNRAS. The transforming activities of RAS oncogeneswere first identified in the Harvey (HRAS) and Kirsten

S h o r a n d J o t h y


Growth Factor

Receptor

Tyrosinekinase

RAFERKMEK

Ras GDP

Ras GTP

Sos 1

P13K

CELL SURVIVAL, PROLIFERATION

(KRAS) murine sarcoma viruses during the mid 1960s.25

NRAS was identified in 1983.

KRAS mutations occur in 30–45% of colorectal carcino-mas.26,27 These mutations develop early in the progressionfrom adenoma to colorectal carcinoma28,29 and aredetectable in colon cancer precursor lesion known as theaberrant crypt foci.30,31

KRAS mutations are localized to exons 1 and 2 in fourcodons (12, 13, 59, and 61). In colon cancer, 98% ofmutations occur in codon 12 and 13, both coding forglycine amino acid in the wild-type gene.32,33 Codon 12 ismore commonly mutated than codon 13 (77% versus21% respectively). Codon 61 is mutated in 2% of thecases but the effect of this mutation on therapeuticresponses is currently unknown.33 Occasional case reportsof CRC have described mutations in other codons, suchas codons 117 and 146, but these are rare and their signif-icance is unknown.23,34 Andreyev et al report that muta-tions in codons 12 and 13 have a negative effect on sur-vival and increase the risk of cancer recurrence.32,35

Mutations target the guanosine triphosphate (GTP) bind-ing domain of KRAS when located in codons 12 and 13,and the GTPase domain when in codon 61. Mutationsinhibit the effects of GAPs, giving rise to a permanentlyactivated mutant RAS-GTP complex that no longerdepends on EGFR regulation.33 This results in abnormal

differentiation and unregulated colon cell growth as wellas enhanced tumour angiogenesis.36,37

KRAS Mutation Detection Immunohistochemistry is inadequate to detect KRASmutations because presently available antibodies cannotreliably distinguish wild-type versus mutant KRAS pro-teins. Molecular biology techniques are currently used todemonstrate KRAS mutations. The use of highly sensitivemethods, some of them requiring microdissection, arenecessary because non-neoplastic cells are frequentlypresent in tumour specimens. KRAS mutations were pre-viously detected by direct sequencing; however thismethod has a sensitivity of 25% which is insufficient.Polymerase chain reaction (PCR)-restriction fragmentlength polymorphisms, PCR-single strand conformationpolymorphism with denaturing gradient gel electrophore-sis (SSCP/DGGE), and mutant-allele-specific amplifica-tion were the other methods being used to detect KRASmutations. More recently, he amplification refractorymutation system (ARMS) variant of real-time PCR hasproven to be a sensitive method for KRAS mutationdetection in fixed tissue specimens containing as low as1% of tumour cells.38,39

A highly sensitive KRAS mutation assay has been recentlydescribed, adapting the smart amplification process ver-sion 2 (SMAP-2) with the addition of a peptide nucleicacid clamp with exact homology to the wild-type allele.40

Mutant DNA could be identified even when it was presentin 0.1% of the DNA in a specimen. With this technique, asingle assay in a single tube can be used to identify allpossible mutations in codon 12 of KRAS. Whatever tech-nique is being used, it requires adequate validation byDNA sequencing.

Subtypes of Colon Cancer Associated with KRASMutationsCRC occurs as both sporadic and hereditary conditions.Hereditary CRC, which includes hereditary non polyposiscolorectal cancer (HNPCC) and the rarer familial adeno-matous polyposis (FAP), accounts for approximately5–10% of all cases.41 HNPCC is an autosomal dominantcondition caused by germline mutations in mismatchrepair (MMR) genes such as hMLH1, hMSH2, hMSH6that result in deficient DNA repair and the developmentof tumours with high levels of microsatellite instability inshort tandem repeat DNA sequences. FAP-associatedCRC, also an autosomal dominant hereditary cancer,accounts for about 1% of all CRCs and is caused by a

M o l e c u l a r P a t h o l o g y o f C o l o r e c t a l C a n c e r


R A S-G T P

AktCell Motility PI3K

RAF

MEK

ERK

Cell-Cycle Progression

Transcription

Figure 2. RAS-GTP complex activates two pathways, the PI3K/Aktpathway to initiate cell motility and the RAF/MEK/ERK pathway to ini-tiate cell-cycle progression and gene transcription.

germline mutation in the adenomatous polyposis (APC)gene.42 The specific causes of sporadic CRC are currentlyunknown, but are probably multifactorial and have beenrelated to environmental factors, including diet andsmoking, the effect of which on the KRAS gene areincompletely understood.

KRAS mutations have been found in both sporadic andhereditary CRCs, but there is conflicting evidence on theirfrequency. Several studies report a similar frequency ofKRAS mutations in sporadic tumours and in patientswith FAP and HNPCC.35,43,44 Losi,45 however, found alower frequency of KRAS mutations in HNPCC than insporadic cancer. Oliveira et al. published the mostdetailed analysis of KRAS mutation frequency in differentsubtypes of colon cancer.33 They report that KRAS muta-tions are more frequent in HNPCC than in sporadic can-cers with microsatellite instability (40% and 22% respec-tively). However, HNPCC and sporadic CRC cases withstable microsatellite DNA do not have a significantly dif-ferent incidence of KRS mutation. In patients withHNPCC, those with germline mutations in hMSH2 andhMSH6 mismatch repair genes, have a higher frequencyof KRAS mutations compared to those with hMLH1mutations. The lowest frequency of KRAS mutations isfound in sporadic cancers with microsatellite instabilitythat also have hMLH1 promoter hypermethylation.33

The presence of KRAS mutations in primary CRCtumours versus in metastatic sites is an area that requiresfurther investigations. Most studies screening for KRASmutations evaluate the presence of this molecular markerin the primary colorectal tumour, while the response ofanti-EGFR antibodies is assessed in the setting ofmetastatic disease. The absence of KRAS mutations in theprimary tumour does not necessarily mean that KRASmutation is absent in metastases.46,47 The study by DiFiore et al. reports an identical KRAS mutational status inprimary and metastatic tumours; however, their samplesize is too small to show any significant results. Furtherresearch is required, but it might become important toscreen metastases directly for KRAS mutations.

Targeted Therapies of CRC with Anti-EGFRAntibodiesCetuximab is a chimeric, mouse-human immunoglobulin(Ig) G1 against EGFR. The US Food and DrugAdministration (FDA) and the European MedicinesAgency (EMEA) approved its use in 2004. Panitumumabis a fully humanized IgG2 antibody against EGFR. It wasapproved by the FDA in 2006 and by the EMEA in 2007.

Cetuximab and panitumumab bind to the extracellulardomain of EGFR with high specificity and affinity, andcompetitively inhibit the natural ligand from binding toEGFR. The receptor becomes unable to dimerize and theintrinsic tyrosine kinase is not activated, leading to EGFRinternalisation and degradation.47 Binding of anti-EGFRmonoclonal antibodies to EGFR prevents the activationof EGFR dependent RAS/RAF/MEK and PI3K down-stream signalling pathways. Cellular proliferation andangiogenesis are inhibited and apoptosis is stimulated.48–50

Cetuximab is indicated for use, in combination withirinotecan or on its own, for the treatment of EGFR-expressing stage IV colorectal carcinoma in patients whoare refractory to treatment with irinotecan alone or incombination with other chemotherapy drugs.51 The indi-cation for panitumumab is comparable to cetuximab, butpatients must also carry a wild-type KRAS gene.39,52

Cetuximab is given intravenously at weekly intervals withdoses of 250 mg/m2 until disease progression or toxicitydevelopment.51–54 Panitumumab is administered intra-venously at a 6 mg/kg dose once every 2 weeks until dis-ease progression or toxicity development. 39,52,55

Severe hypersensitive reactions are infrequent, occurringin 3% of patients.7 The most common side effect is anacne-like rash, experienced by 88–90% of patients.Pruritus, dry skin, skin fissures, and desquamation arealso possible but uncommon. Conjunctivitis and hypo-magnesemia have been reported in a small number ofpatients. Infusion related side effects also include nausea,vomiting, headache, dizziness and chills.55 Response toanti-EGFR treatment is documented by medical imaging,including chest radiography, magnetic resonance imaging,and computed tomography.

Nimotuzumab is another humanized anti-EGFR mono-clonal antibody used to treat various carcinomas. Its usein the treatment of CRC with different KRAS genotypeshas not been reported in a large series of patients.

Evidence for the Association of the TherapeuticResponse to Cetuximab and Panitumumab withKRAS Status of the TumourPatient response to cetuximab and panitumumab ismeasured by assessing complete response defined as thecomplete disappearance of all measurable lesions, withoutthe appearance of any new lesions, and partial responsedefined as reduction in bi-dimensionally measurablelesions by at least 50% of the sum of the products of theirlargest perpendicular diameters and an absence of pro-



gression in other lesions, without the appearance of anynew lesions.51

As is customary in adjuvant treatment studies, the overallresponse to anti-EGFR drugs is measured by overall sur-vival (OS), and progression-free survival (PFS). The latteris calculated from the date of randomization to the firstobservation of disease progression, or to death from anycause within 60 days after randomization or the mostrecent tumour assessment.

The overall response rate to cetuximab and panitumumabin patients expressing EGFR, but whose tumours have notbeen pre-screened for KRAS mutation, remains poor.Twenty three percent of patients respond (by tumourshrinkage) when cetuximab is combined with irinotecanand only 11% respond when cetuximab is administeredon its own. The OS and PFS are longer in patients receiv-ing both cetuximab and Irinotecan than cetuximab on itsown (OS: 8.6 months versus 6.9 months; PFS: 4.1 monthsversus 1.5 months respectively).51

Using panitumumab, 10% of patients had an objectiveresponse. The mean PFS is longer for patients receivingpanitumumab than patients receiving basic standard ofcare (13.8 weeks versus 8.5 weeks respectively) but themean PFS is similar (8 weeks for panitumumab and 7.3weeks for basic standard of care). Seventy-six percent ofpatients in BSC group crossed over to receive panitu-mumab after disease progression. There are no observeddifferences between the two groups for OS.52 Patients thatdo not respond to cetuximab and panitumumab remainresistant to these drugs.

The discovery that the OS in tumours with wild-typeKRAS is longer than in patients having KRAS mutationswas a major advance in the documentation of the thera-peutic benefits of cetuximab.7,8 Disease progression fol-lowing cetuximab therapy is faster in metastatic CRCpatients bearing KRAS mutations than in patients withwild-type KRAS.46,56 The presence of wild-type KRAS cor-relates with greater disease control.57 After cetuximabtreatment, wild-type KRAS expressing tumours demon-strate a greater (> 9.66%) reduction in tumour size thanin tumours expressing mutant KRAS.58

Overall, approximately 50% of patients with wild-typeKRAS tumours receiving cetuximab respond to it, but themechanism(s) by which the remaining 50% resist thistreatment is unknown.59 One possibility is that unlike thetumour cells in the primary tumour, those present inmetastatic sites have mutated KRAS and have clonallyexpanded. Another mechanism would be that metastatic

tumour cells might have evolved a transduction step thatblocks signalling from EGFR.

Patients with wild-type KRAS tumours who receive pani-tumumab monotherapy have higher OS and PFS thanthose with mutant KRAS tumours. Tumour reduction fol-lowing panitumumab administration is observed in thewild-type KRAS group but not in the KRAS mutantgroup.39

Summary and Future DirectionsFollowing the introduction of anti-EGFR therapy theprognostic of patients with metastatic CRC has improved.However, treatment of a patient with anti-EGFR therapyis expensive and side effects, although minor, can beunpleasant. The treatment is successful in some patientsbut others remain resistant. In order to spare patientsfrom high costs and unnecessary toxicity of these drugs, itwas necessary to identify good markers that could sepa-rate responders from non-responders. The mutationalstatus of KRAS gene is the best possible marker availabletoday for this purpose. KRAS mutation is present in CRCtumours of most non-responding patients. The majorityof patients with a wild-type copy of KRAS have a betterresponse to anti-EGFR treatment, as shown by longeroverall survival and progression free survival. KRASmutations occur frequently in CRC with the highest fre-quency occurring in the hereditary cancer HNPCC withmicrosatellite instabilities. The most common KRASmutations are localized on codons 12 and 13, but muta-tions in other codons, 59, 61, 117 and 146 are also report-ed. Mutation on codon 12, changing glycine to valine isassociated with the worst outcome for CRC.

The necessity to test colorectal tumours for EGFR expres-sion by immunohistochemistry is controversial. InAmerican and European countries both cetuximab andpanitumumab are approved for use in metastatic CRCpatients who express EGFR protein in their tumour asdemonstrated by immunohistochemistry (EGFR posi-tive). There is evidence, however, that a subset of patientswho do not express EGFR (EGFR negative) can alsorespond to anti-EGFR treatments. Patients with EGFRnegative metastatic CRC have a 20–25% response rate tocetuximab plus irinotecan therapy,8,60 which is similar tothe 23% response rate to cetuximab given on its own inEGFR positive tumours.51 Based on these results, theimmunohistochemical detection of EGFR expression mayno longer be suitable for selecting patients for anti-EGFRtreatments, as it may exclude patients likely to benefitfrom such therapies. More research in this area is required



but it appears that identifying the KRAS gene status is asuperior screening marker.

The second challenge concerns the validity of KRAS sta-tus as an exclusive marker or the need to search for newmarkers predictive of responses to anti-EGFR targetedtherapies. Some researchers propose that KRAS statusserves as the only basis for selecting metastatic CRCpatients as candidates for treatment with anti-EGFRdrugs. There are, however, reports that may contradictthis proposal. Two studies show that KRAS mutants withadvanced CRC experienced a significant response to anti-EGFR therapy and had improved OS and long-term dis-ease control.14,56 This suggests that the inhibitory signalexerted on EGFR can be transduced by another pathway,independent of KRAS in some tumour cells. Among themost likely signalling candidates, there might be a need toevaluate the role of RAF, MEK, and ERK in relation toresponse or resistance to anti-EGFR drugs.

Increased EGFR gene copy number has also been pro-posed to be a candidate as a predictive marker. Differentgroups have reported varying percentages of CRCpatients with increased EGFR gene copy number, somereporting its rarity (10–15%),60–64 while others havereported a much higher percentage (31%).15 A number ofresearch studies conclude that there is a positive associa-tion between increased EGFR copy number in patientswith metastatic CRC and improved response to anti-EGFR monoclonal antibodies.15,65,66 However, patientswithout EGFR copy number amplification can stillrespond to cetuximab.59,67–69 This marker therefore playsonly a minor role compared to KRAS when measuringresponse to anti-EGFR therapy.

References1. Centers for Disease Control and Prevention (CDC) Website.

www.cdc.gov/cancer/colorectal/basic_info/.

2. Ries LAG, Melbert D, Krapcho M, et al. SEER Cancer Stat Rev,1975-2005, Bethesda, MD: National Cancer Institute.

3. Parkin DM, Bray F, Ferlay J, and Pisani P. Global cancer statistics,2002. CA Cancer J Clin. 2006; (55):74–108.

4. Abeloff MD, Armitage JO, Niederhuber JE, et al. (2008) Abeloff’sClinical Oncology, 4th edition. Philadelphia: Churchill Livingstone,imprint of Elsevier Inc; 2008.

5. de Gramont A, Figer A, Seymour M, et al. Leucovorin and fluo-rouracil with or without oxaliplatin as first-line treatment inadvanced colorectal cancer. J Clin Oncol 2000; 18:2938–47.

6. Saltz LB, Cox JV, Blanke C, et al. Irinotecan plus fluorouracil andleucovorin for metastatic colorectal cancer. N Engl J Med 2000;343: 905–14.

7. Meyerhardt JA and Mayer RJ. Systemic therapy for colorectal can-cer. N Eng J Med 2005; 352:476–87.

8. Hebbar M, Wacrenier A, Desauw C, et al. (2006) Lack of useful-ness of epidermal growth factor receptor expression determina-tion for cetuximab therapy in patients with colorectal cancer.

Anticancer Drugs 2006; 17: 855–57.

9. Citri A. and Yarden Y. EGF-ERBB signalling: towards the systemslevel. Nature 2006; 7:505–16.

10. Ng K and Zhu AX. Targeting the epidermal growth factor receptorin metastatic colorectal cancer. Crit Rev Oncol Hematol 2008;65:8–20.

11. Hynes NE and Lane HA. (2005) ERBB receptors and cancer: Thecomplexity of targeted inhibitors. Nat Rev Cancer 2005; 5:341–54.

12. Nagahara H, Mimori K, Ohta M, et al. Somatic Mutations ofEpidermal Growth Factor Receptor in Colorectal Carcinoma. ClinCancer Res 2005; 11:1368–71.

13. Tsuchihashi Z, Khambata-Ford S, Hanna N, and Janne PA (2005)Responsiveness to Cetuximab without mutations in EGFR. N Engl JMed 2005; 353:208–9.

14. Moroni M, Veronese S, Benvenuti S, et al. Gene copy number forepidermal growth factor receptor (EGFR) and clinical response toantiEGFR treatment in colorectal cancer: a cohort study. LancetOncol 2005; 6: 279–86.

15. Salomon DS, Brandt R, Ciardiello F, Normanno N. Epidermalgrowth factor-related peptides and their receptors in humanmalignancies. Crit Rev Oncol Hematol 1995; 9:183–232.

16. Messa C, Russo F, Caruso MG, and Di Leo A. EGF, TGF-alpha, andEGF-R in human colorectal adenocarcinoma. Acta Oncol 1998;37:285–9.

17. Porebska I, Harloziƒska A, and Bojarowski T. Expression of thetyrosine kinase activity growth factor receptors (EGFR, ERB B2,ERB B3) in colorectal adenocarcinomas and adenomas. TumourBiol 2006; 21:105–15.

18. Radinsky R, Risin S, Fan D, et al. Level and function of epidermalgrowth-factor receptor predict the metastatic potential of humancolon-carcinoma cells. Clin Cancer Res 1995; 1:19–31.

19. Buday L, and Downward J. Epidermal Growth-Factor regulatesP21(RAS) through the formation of A complex of receptor, GRB2Adaptor protein, and SOS nucleotide exchange factor. Cell 1993;73:611–620.

20. Gille H, and Downward J. Multiple Ras effector pathways con-tribute to G(1) cell cycle progression. J Biol Chem 1999;274:22033–40.

21. Lavoie JN, L’Allemain G, Brunet A, et al. Cyclin D1 expression isregulated positively by the p42/p44(MAPK) and negatively by thep38/HOG(MAPK) pathway. J Biol Chem 1996; 271: 20608–16.

22. Kennedy SG, Wagner AJ, Conzen SD, et al. The PI 3-kinase/Aktsignaling pathway delivers an anti-apoptotic signal. Genes Dev1997; 11:701–13.

23. Donovan S, Shannon KM, and Bollag G. GTPase activating pro-teins: critical regulators of intracellular signaling. BiochimBiophys Acta. Reviews on Cancer 2002; 1602: 23–45.

24. Waksal HW. Role of an anti-epidermal growth factor receptor intreating cancer. Cancer Metastasis Rev 1999; 18:427–36.

25. Malumbres M, and Barbacid M. RAS oncogenes: the first 30years. Nature Rev Cancer 2004; 3(6):459–65.

26. Bos JL, Fearon ER, Hamilton SR, et al. Prevalence of Ras gene-mutations in human colorectal cancers. Nature 1987; 327:293–7.

27. Forrester K, Almoguera C, Han KY, et al. Detection of high-inci-dence of K-ras oncogens during human-colon tumorigenesis.Nature 1987; 327:298–303.

28. Burmer GC, and Loeb LA. Mutations in the KRas2 oncogene dur-ing progressive stages of human colon carcinoma. Proc Natl AcadSci USA 1989; 86:2403–407.

29. Fearon ER, and Vogelstein B. A genetic model for colorectaltumorigenesis. Cell 1990; 61:759–67.

30. Vivona AA, Shpitz B, Medline A, et al. K-Ras mutations in AberrantCrypt Foci, adenomas and adenocarcinomas duringazoxymethane-induced colon carcinogenesis. Carcinogenesis1993; 14:1777–781.


49Canadian Journal of PathologyJuly 2009 4949

31. Smith AJ, Hartley SS, Penner M, et al. (1994) Somatic APC and K-ras codon 12 mutations in aberrant crypt foci from human colons.Cancer Res 1994; 54:5527–530.

32. Andreyev HJN, Norman AR, Cunningham D, et al. Kirsten ras mutations in patients with colorectal cancer: the ‘RASCAL II’ study. Br JCancer 2001; 86: 692–96.

33. Oliveira C, Westra JL, Arango D, et al. Distinct patterns of KRASmutations in colorectal carcinomas according to germline mis-match repair defects and hMLH1 methylation status. Hum MolGenet 2004; 13:2303–311.

34. Wójcik P, Kulig J, Okoƒ K, et al. (2008) KRAS mutation profile incolorectal carcinoma and novel mutation—internal tandem dupli-cation in KRAS. Pol J Pathol 2008; 59:93–6.

35. Andreyev HJN, Norman AR, Cunningham D, et al. Kirsten ras muta-tions in patients with colorectal cancer: the multicenter ‘‘RASCAL’’study. J Natl Cancer Inst 1998; 90:675–84.

36. Shirasawa S, Furuse M, Yokoyama N, and Sasazuki T. AlteredGrowth of Human Colon Cancer Cell Lines Disrupted at activatedKi-ras. Science 1993; 260:85–8.

37. Rak J, Mitsuhashi Y, Bayko L, et al. Mutant ras OncogenesUpregulate VEGFIVPF Expression: Implications for Induction andInhibition of Tumor Angiogenesis. Cancer Res 1995; 55:4575–80.

38. Newton CR, Graham A, Heptinstall LE, et al. Analysis of any pointmutation in DNA. The amplification refractory mutation system(ARMS). Nucleic Acids Res 1989; 17:2503–16.

39. Amado RG, Wolf M, Peeters M, et al. Wild type KRAS is requiredfor panitumumab efficacy in patients with metastatic colorectalcancer. J Clin Oncol 2008 26: 1626–34.

40. Tatsumi K, Mitani Y, Watanabe J, et al. Rapid screening assay forKRAS mutations by the modified smart amplification process. JMol Diagn 2008; 10:520–6.

41. Lynch HT, and de la Chapelle A. Hereditary colorectal cancer. NEngl J Med 2003; 348:919–32.

42. Cao Y, Pieretti M, Marshall J, et al. (2002) Challenge in the differ-entiation between attenuated familial adenomatous polyposisand hereditary non polyposis colorectal cancer: case report withreview of the literature. Am J Gastroenterol 2002; 97:1822–27.

43. Fujiwara T, Stolker JM, Watanabe T, et al. Accumulated clonalgenetic alterations in familial and sporadic colorectal carcinomaswith widespread Instability in microsatellite sequences. Am JPathol 1998; 153:1063–78.

44. Sánchez-de-Abajo A, de la Hoya M, van Puijenbroek M, et al.Molecular analysis of colorectal cancer tumors from patients withmismatch repair proficient hereditary nonpolyposis colorectalcancer suggests novel carcinogenic pathways. Clin Cancer Res2007; 13: 5729–35.

45. Losi L, DeLeon MP, Jiricny J, et al. K-ras and p53 mutations inhereditary non-polyposis colorectal cancers. Int J Cancer 1997;74:94–6.

46. Di Fiore F, Blanchard F, Charbonnier F, et al. Clinical relevance ofKRAS mutation detection in metastatic colorectal cancer treatedby Cetuximab plus chemotherapy. Br J Cancer 2007; 96:1166–69.

47. de Castro-Carpeno J, Belda-Iniesta C, Saenz EC, et al. EGFR andcolon cancer: a clinical view. Clin Transl Oncol 2008; 10:6–13.

48. Wu XP, Rubin M, Fan Z, et al. Involvement of p27KIP1 in G1 arrestmediated by an anti-epidermal growth factor receptor monoclonalantibody. Oncogene 1996; 12:1397–403.

49. Liu B, Fang M, Schmidt M, et al. Induction of apoptosis and activa-tion of the caspase cascade by anti-EGF receptor monoclonal anti-bodies in DiFi human colon cancer cells do not involve the c-junN-terminal kinase activity. Br J Cancer 2000; 82:1991–99.

50. Shaheen RM, Ahmad SA, Liu W, et al. Inhibited growth of coloncancer carcinomatosis by antibodies to vascular endothelial andepidermal growth factor receptors. Br J Cancer 2001; 85:584–89.

51. Cunningham D, Humblet Y, Siena S, et al. Cetuximab monothera-

py and cetuximab plus irinotecan in irinotecan-refractory metasta-tic colorectal cancer. N Engl J Med 2004; 351:337–45.

52. Van Cutsem E, Peeters M, Siena S, et al. Open-label phase III trialof panitumumab plus best supportive care compared with bestsupportive care alone in patients with chemotherapy-refractorymetastatic colorectal cancer. J Clin Oncol 2007; 25:1658–64.

53. Jonker DJ, O’Callaghan CJ, Karapetis CS, et al. Cetuximab for thetreatment of colorectal cancer. N Engl J Med 2007; 357:2040–48.

54. FDA website. www.fda.gov/medwatch/SAFETY/2005/Sep_PI/Erbitux_PI.pdf.

55. FDA website. www.fda.gov/medwatch/safety/2007/Nov_PI/Vectibix_PI.pdf.

56. Benvenuti S, Sartore-Bianchi A, Di Nicolantonio F, et al.Oncogenic activation of the RAS/RAF signaling pathway impairsthe response of metastatic colorectal cancers to anti-epidermalgrowth factor receptor antibody therapies. Cancer Res 2007;67:2643–48.

57. Khambata-Ford S, Garrett CR, Meropol NJ, et al. Expression ofepiregulin and amphiregulin and K-ras mutation status predictdisease control in metastatic colorectal cancer patients treatedwith cetuximab. J Clin Oncol 2007; 25:3230–37.

58. De Roock W, Piessevaux H, De Schutter J, et al. KRAS wild-typestate predicts survival and is associated to early radiologicalresponse in metastatic colorectal cancer treated with cetuximab.Ann Oncol 2008; 19:508–15.

59. Lievre A, Bachet J, Boige V, et al. KRAS mutation status is predic-tive of response to cetuximab therapy in colorectal cancer. CancerRes 2006; 66:3992–95.

60. Chung KY, Shia J, Kemeny NE, et al. Cetuximab shows activity incolorectal cancer patients with tumors that do not express theepidermal growth factor receptor by immunohistochemistry. J ClinOncol 2005; 23:1803–10.

61. Lenz HJ, Van Cutsem E, Khambata-Ford S, et al. Multicenter phaseII and translational study of cetuximab in metastatic colorectalcarcinoma refractory to irinotecan, oxaliplatin, and fluoropyrim-idines. J Clin Oncol 2006; 24:4914–21.

62. Lievre A, Bachet JB, Boige V, et al. KRAS mutations as an inde-pendent prognostic factor in patients with advanced colorectalcancer treated with cetuximab. J Clin Oncol 2008; 26:374–79.

63. Shia J, Klimstra DS, Li AR, et al. Epidermal growth factor receptorexpression and gene amplification in colorectal carcinoma: animmunohistochemical and chromogenic in situ hybridizationstudy. Mod Pathol 2005; 18:1350–56.

64. Sauer T,Guren MG, Noren T, and Dueland S. Demonstration ofEGFR gene copy loss in colorectal carcinomas by fluorescence insitu hybridization (FISH): a surrogate marker for sensitivity tospecific anti-EGFR therapy? Histopathology 2005; 47:560–64

65. Sartore-Bianchi A, Moroni M, Veronese S, et al. Epidermal growthfactor receptor gene copy number and clinical Outcome ofmetastatic colorectal cancer treated with panitumumab. J ClinOncol 2007; 25:3238–45.

66. Cappuzzo F, Finocchiaro G, Rossi E, et al. EGFR FISH assay pre-dicts for response to cetuximab in chemotherapy refractory col-orectal cancer patients. Ann Oncol 2008;19:717–23.

67. Di Fiore F, Van Cutsem E, Laurent-Puig P, et al. Role of KRAS muta-tion in predicting response, progression-free survival, and overallsurvival in irinotecan-refractory patients treated with cetuximabplus irinotecan for a metastatic colorectal cancer: Analysis of 281individual data from published series. J Clin Oncol 2008; 26.

68. Perrotte P, Matsumoto T, Inoue K, et al. Anti-epidermal growthfactor receptor antibody C225 inhibits angiogenesis in humantransitional cell Carcinoma growing orthotopically in nude mice.Clin Cancer Res 1999; 5:257–64.

69. Schrag D. The price tag on progress—chemotherapy for colorectalcancer. N Engl J Med 2004; 351:317–9.



CARING FOR PEOPLEGamma-Dynacare is one of Canada'slargest and most respected providers oflaboratory services and solutions.

A proven and innovative industry leader,Gamma-Dynacare is proud to work withgovernments, regional health authorities,hospitals, physicians and companies tocontribute to the health and wellness ofpeople across Canada.

Toll-Free: 800.668.2714

www.gamma-dynacare.com

Canadian Journal ofPathology

Documents

Transcript of Canadian Journal ofPathology