Connection: What is the difference between the H score and the Allred score? Which is better? What do you prefer? Why are you not using the Allred scoring system? Both methods combine assessment of percentage of tumour cell nuclei staining and the intensity of staining. The H score is the sum of the percentage of weakly stained cells, the percentage of moderately stained cells multiplied by two, and the percentage of strongly staining cells multiplied by three. This gives a range of scores from 0 to 300. The Allred score is the sum of a score for percentage of cells staining (no staining = 0, staining in <1% of cells = 1, 1 to 10% = 2, 10 to 33% = 3, 33 to 67% = 4 and 67% to 100% = 5) and a score for intensity (absent = 0, weak = 1, moderate = 2, strong = 3). The possible scores are 0 and 2 to 8. I prefer the H score as it gives a more detailed assessment. The Allred score gives too much emphasis to the weakly positive group: Most tumours score 0, 7 or 8 (ref 2). There are other scoring methods. The Quick score is the sum of a score for percentage of cells staining (no staining = 0, staining in 1 to 25% of cells = 1, 26 to 50% = 2, 51 to 75% = 3 and 76% to 100% = 4) and a score for intensity (absent = 0, weak = 1, moderate = 2, strong = 3). Other alternatives are to use a cutoff of 1% or 10% staining of any intensity. Connection: Is the H score, the universal system for the assessment of the intensity and distribution of positivity on sections stained for oestrogen receptor (ER)? There is no universal system for scoring. Connection: What cutoff should be used for scoring oestrogen receptors? There is no consensus on what cutoffs should be used for scoring hormone receptors. The chance of response to endocrine therapy has been studied in patients with metastatic breast cancer. Tumours with no expression of oestrogen receptor rarely respond to endocrine treatment, and tumours with convincing expression of oestrogen receptor have a good chance of responding to such treatment. It is possible to divide tumours into negative and positive with significantly different chances of response as long as the cutoff is somewhere between the group with no staining and the bulk of tumours with a convincingly positive staining. Barnes et al (7) assessed eight different methods of categorising tumours with between 31 and 81% of tumours classified as positive; all showed a significant relationship with response to endocrine treatment. This negative/positive categorisation is an oversimplification. There is evidence that the level of expression relates to the chance of response (7, 8). There are far fewer data on the choice of cutoff in the adjuvant setting. Sadly, a large study by Harvey et al (9) reused frozen samples referred from other centres that were pulverised and later fixed with less control over specimen quality. The high proportion of weakly positive tumours probably reflect the sub-optimal methods used Tumours that are positive for both oestrogen and progesterone receptor are more likely to respond to endocrine therapy than tumours that are oestrogen receptor positive and progesterone receptor negative. However, the levels of oestrogen receptor tend to be higher in the double positive group. In conclusion, one cannot be dogmatic about exactly where the cutoff or cutoffs should be. Large studies in both adjuvant and metastatic settings are needed. With large studies it may be possible to give reliable estimates of the chance of response according to different degrees of hormone receptor expression. Connection: How do laboratories report their results? Do they just say positive or negative or report on the actual score (and the cutoff used)? With no consensus on the cutoffs that should be used it is scarcely surprising that there is wide variation in the way that hormone receptors are reported. We report the H score and percentage of positive nuclei and categorise tumours as negative (H score 0 to 9), weakly positive (H score 10 to 49) and positive (H score 50 to 300). It is useful to report the score as different cutoffs are useful in different clinical circumstances. We tend to use the lower cutoff for decisions about adjuvant endocrine treatment and the higher cutoff when considering primary endocrine treatment without surgery. Connection: Is it true that three to six separate core needle insertions are needed to obtain a sufficient sample of breast tissue? Of these, how many cores do laboratorie stain for ER/PR? How do they report on the results, if more than one core is stained? Do they report each core separately or just report a positive result, if one of those cores are stained? The number of samples varies between centres. Our routine practice for mass lesions visible on ultrasound lesions (including most invasive carcinomas) is to perform one or two ultrasound-guided core biopsies. These are usually embedded in one block. Connection: Which is better: Core or excision biopsy? Is there a trend towards greater accuracy with larger gauge samples in the CNB procedure? Core biopsies have more reliable fixation. The main reason that we assess hormone receptor status on core biopsy is that the result is available earlier. This is particularly useful, if primary hormonal therapy is being considered. Good results can also be achieved in excision specimens, if attention is paid to good fixation - we routinely receive all specimens fresh and incise the tumour. One advantage of surgical specimens is that there is almost always an internal control present. Another advantage is that very occasionally a tumour shows marked variation in expression of hormone receptors in different areas. Such heterogeneity of expression may not be apparent in the core biopsy (1). It is possible to achieve excellent results with standard 14 gauge needle core biopsy. I am not aware of any data on larger gauge core biopsies, but it is unlikely to offer a significant advantage over standard gauge core biopsy. Connection: When laboratories get a falsepositive or false-negative result, the impact for the patient is enormous. What is the reason for the slow progress towards ER/PR standardisation? The methods used in many laboratories have arisen in an ad hoc way. A possible reason for lack of standardisation is the work required. Connection 2008 | 32
April 2008, VOL 11 Connection IHC STANDARDIZATION AROUND THE WORLD In this issue Q&A with Patho
Contents APRIL 2008, VOL 11 2 3 4 Editorial George L. Kumar, PhD Featured Laboratory The Osamura
Editorial Immunohistochemistry (IHC) Standardization Around the World George L. Kumar, PhD Managing
Featured Laboratory The Osamura Laboratory Laboratory headed by Prof. Robert Yoshiyuki Osamura, MD,
Q&A Ask the Experts: On Immunohistochemistry (IHC) Standardization Professor Leong is Medical Di
time interval between removal of tissue and immersion in fixative, the temperature of tissue storage
Connection: How would you like to standardize IHC in Australia? As discussed above, standardization
Connection: Antigen retrieval is essential in immunohistochemistry (IHC) in order to restore epitope
Connection: Is a universal image analysis system feasible? No, not until we standardize the all impo
Q&A Fernando Soares, MD, PhD University of São Paulo, São Paulo, Brazil Dr. Fernando Soares is F
Connection: In your opinion, what is the biggest hurdle for standardization? Probably education in l
There are no standards in Latin America. We always follow the manufacturers protocol during our firs
Q&A Bryan R. Hewlett, ART, MLT Consultant Technologist to the Anatomic Pathology EQA program of
There is no standardization of AR steps circumstances, it would be impossible toand, undera the pre
Connection: How would you rate European (UK NEQAS, NordiQC) and US IHC standards to Canadian IHC sta
Q&A Prof. Chen Jie Prof. Cui Quancai Peking Union Mediacl College Hospital, Beijing, China Prof.
Connection: According to Goldstein et al. Appl Immunohistochem Mol Morphol 2007;15:124133 Immunohis
Connection: Is it true that a particular histology feature may be better demonstrated by other fixat
Connection: What constitutes standardization of image analysis as applied to immunohistochemistry (I
Q&A Dr. Tanuja Manjanath Shet Dr. Vani Parmar Tata Memorial Hospital, Mumbai, India Tanuja Manj
... the biggest hurdle in India is suboptimal fixation and processing of tissues. Though I agree th
Connection: Is it true that a particular histology feature may be better demonstrated by other fixat
Connection: Can you comment on the internal and external quality control (EQC) procedures followed i
Q&A Prof. Robert Yoshiyuki Osamura Department of Pathology, Tokai University School of Medicine
Connection: In your opinion, what is the biggest hurdle for standardization? The biggest hurdle for
the standardization of image ... appropriate for pre-screeninganalysis is of the staining quality.
Connection: Why is standardization of image analysis in diagnostic pathology important? Because the
Q&A James F. Happel, DLM (ASCP), HTL Technical Director of Surgical Pathology, Research and Deve
Connection: United Kingdom National External Quality Assessment Service (UK NEQAS) helps to ensure t
Connection: The American Society of Clinical Oncology (ASCO) and the College of American Pathologist
would recommend that standardization Ibegin with identifying a reliable and trustworthy source ...
Interview Immunohistochemistry for Oestrogen and Progesterone Receptors Dr. Andrew Lee Consultant H
Connection: What is the difference between the H score and the Allred score? Which is better? What d
Connection: Can you comment on the burden in the laboratory, if one changes from a current ER/PR ass
Opinion & Interview IHC Standardization: A Dako Perspective Dr. Ole Rasmussen R&D Director,
Connection: Dako has developed Readyto-Use Antibodies for in vitro diagnostic applications. How is t
Articles UK NEQAS-ICC & ISH: Historical perspective, current role, future directions Andy Dodso
UK NEQAS-ICC in the 1990s In his first year as Scheme Organiser, Keith oversaw the transition to sub
The application could be argued to represent a field change in terms of the rigour with which the an
Assessment teams consist of four assessors, who view slides around a multi-headed microscope and sco
The archive which UK NEQAS holds, both in terms of stained slides and methodological data, must sure
For Immunocytochemistry and FISH RESULT: RUN 80L SLIDE: NEQAS Laboratory No: XXX Mr. A. Scientist De
Figure 6. Feedback on results has always been given high priority, and for many years this has been
a b c d Figure 7. The antigen chosen by Gerry Reynolds for his very first assessment run was kap
Bibliography Selected UK NEQAS-ICC & ISH papers. Ibrahim M, Dodson AR, Barnett S, Fish D, Jasani
Articles Nordic Immunohistochemical Quality Control (NordiQC) An Organization for External Quality A
parameters (i.e. results interpretation and reporting) (4, 5). In an EQA setting, by circulating ser
CD79a (Fig. 2) Among 112 laboratories submitting stains in the latest run, most used mAb clone JCB11
References 1. Rhodes A, Jasani B, Barnes DM, Bobrow LG, Miller KD. Reliability of immuno-histochemic
Fig. 2. CD79a A. Optimal CD79a staining of the tonsil using the monoclonal antibody (mAb) clone JCB
Standardization of Ki-67 Immunohistochemical Staining for Diagnosing Grade of Gastrointestinal Strom
was conducted in 49 GIST cases. The concordance rate for the evaluation results at three laboratorie
CB pH6 a b c d TE pH9 e f g h Autoclave 121° C/10 min Water bath 95° C/40 min Microwave
50 Table 1. Correlation between NCC and STD methods R2=0.9483 Categories of proportion NCC Method 3
Opinions Importance of Standardization for Predictive Prognosis David J. Dabbs, MD Chief of Pathol
is documented and serves as a surrogate marker for the initial exposure to formalin. Since the first
Opinions The New Era for ER and PRIts time to Standardize! Dr. Ian Ellis, B.Med.Sci. BM, MS, FRCpa
et al 2001). The main reason for false-negative results is due to inefficient heat-induced epitope (
Standardization of HER2 TestingInconsistency Raises Questions Opinions Sunil S. Badve, MD, FRCPath
rence seen in these trials is in the order of 50%. This is the major reason for all the excitement a
which now requires expression of HER2 by at least 30% of tumor cells (instead of 10%). It has also r
IHC CONSENSUS MEETING, JANUARY 27 2008, SANTA BARBARA, CA, USA , IHC CONSENSUS MEETING, JANUARY 27,
Richard Cartun, PhD, Sunil Badve, MD Jon Askaa, PhD Søren Nielsen, HT, CT, Mogens Vyberg, MD Elizab
Dako Abstracts Abstracts presented at the 30th Annual San Antonio Breast Cancer Symposium December
Dako Abstracts Amplification of ESR1 may predict resistance to adjuvant tamoxifen in postmenopausal
Dako Abstracts Abstracts presented at the United States and Canadian Academy of Pathology (USCAP) A
Dako Abstracts Metastatic Pancreatic Endocrine Tumors in the Liver Express KOC Briones AJ, Bourne P
Dako Abstracts Merkel Cell Carcinomas Express K Homology Domain Containing Protein Overexpressed in
Dako Abstracts Immunohistochemical Analysis of KOC/IMP3 in Malignant Pleural Mesothelioma Xu H, Sim
Dako Abstracts KOC, TTF-1 and CDX2 Discriminate Small-Cell Carcinoma from Carcinoid and Pancreatic
Dako Publications Publications Co-authored by Dako: In Press Li L, Xu H, Spaulding B, Cheng L, Si
Dako Meetings 2007 - National Society for Histotechnology Meeting. Denver, CO NSH workshop attend
New Premium Quality Concentrated Antibodies New Products Your current and future needs drive the c
Targeting Colorectal Adenocarcinomas, Anti-TIMP-1 and Anti-Villin Monoclonal Mouse Anti-Human Tissue
Dakos FLEX Ready-to-Use Concept Enhance performance of your laboratory by running FLEX Ready-to-Use
2008 Product Catalog Available Now! The catalog features more than 170 new products, including the A