College of American Pathologists
Printable Version





October 2007
Feature Story

bullet ASCO/CAP HER2 guidelines

The following letter and reply have been jointly published by consent in the Archives of Pathology & Laboratory Medicine and CAP TODAY:

Although I agree with the majority of the American Society of Clinical Oncology/College of American Pathologists Guideline Recommendations for Human Epidermal Growth Factor Receptor 2 Testing in Breast Cancer by Wolff, et al,1 I am concerned with the sample exclusion criteria for human epidermal growth factor receptor 2 (HER2) immunohistochemistry (Related Article: SRS: Blood bank information systems, table 5), both as they relate to the issue at hand and for the precedent this list of guidelines sets in an area of pathology certain to see more regulation.

The American Society of Clinical Oncology/College of American Pathologists panel delineated optimal formalin fixation time for HER2 immunohistochemistry performed on excisional biopsies as six to 48 hours. Furthermore, the panel mandates that a negative HER2 result from suboptimal specimens (i.e., those fixed less than six hours or longer than 48 hours) should be qualified in the report. It is intuitive that grossly underfixed breast tissue is not an optimal specimen for HER2 testing, but experience demonstrates that routine weekend fixation (up to 72 hours) allows for good preservation of a wide range of antigens provided appropriate heat-induced epitope retrieval is used. A study by Arber2 analyzing HER2 antigen preservation after prolonged formalin fixation (seven-154 days) found no change in immunoreactivity for up to 20 days in tissues from 33 different infiltrating breast carcinomas. This author has evaluated HER2 antigen preservation in 20 infiltrating breast carcinomas after seven to 14 days of formalin fixation and observed undiminished strong membranous staining after heat-induced epitope retrieval (unpublished data). Boenisch3 has shown convincingly that heat-induced epitope retrieval can restore immunoreactivity after prolonged formalin fixation. Twenty-nine of 30 antigens tested showed equivalent staining intensity after four days of formalin fixation. A recent review by Goldstein, et al,4 titled ''Recommendations for Improved Standardization of Immunohistochemistry,'' suggests a minimum formalin fixation time of eight hours (recommendation one). No maximum fixation time is recommended, and it is stated that ''In general, underfixation is a substantially larger problem than over-fixation'' (page 126).

Reading the Archives article in question I failed to discover any referenced study supporting the sample exclusion criteria. The authors simply allude to previous consensus conferences dealing with HER2 testing (appendix E). Subsequently, Patrick Fitzgibbons, MD, an author of the guideline article1 and chair of the CAP Immunohistochemistry Committee, acknowledged to me in an e-mail (March 2007) that studies confirming the hour-by-hour impact of formalin fixation on HER2 testing over the range of fixation intervals seen in clinical practice have not been completed. It is axiomatic that promulgated guidelines should be evidence based and scientifically validated. In the absence of peer-reviewed study in its support, adoption of the greater than 48-hour formalin fixation exclusion criterion would be counterproductive. This is especially true in light of the potential impact of said criterion on current laboratory practice (i.e., necessitating staffing histology on the weekend to accommodate Friday morning breast cancer surgeries). Based on the evidence cited herein, a 72-hour maximum fixation exclusion criterion would be more appropriate, allowing for routine weekend fixation while enabling accurate and reproducible HER2 testing.

James F. Lombardo, MD
Ogden (Utah) Regional Medical Center


  1. Wolff AC, Hammond MEH, Schwartz JN, et al. American Society of Clinical Oncology/ College of American Pathologists guideline recommendations for human growth factor receptor 2 testing in breast cancer. Arch Pathol Lab Med. 2007;131:18-43.
  2. Arber DA. Effect of prolonged formalin fixation on the immunohistochemical reactivity of breast markers. Appl Immunohistochem Mol Morphol. 2002;10:183-186.
  3. Boenisch T. Effect of heat-induced antigen retrieval following inconsistent formalin fixation. Appl Immunohistochem Mol Morphol. 2005;13:283-286.
  4. Goldstein NS. Recommendations for improved standardization of immunohistochemistry. Appl Immunohistochem Mol Morphol. 2007;15:124-133.

  • M. Elizabeth H. Hammond, MD, of the Department of Pathology, LDS Hospital, Intermountain Healthcare, Salt Lake City, and co-chair of the ASCO/CAP HER2 Guideline Panel, replies (on behalf of the panel): The American Society of Clinical Oncology/College of American Pathologists panel appreciates thoughtful commentary that has been provided to us by personal communication to various members as well as to the CAP or as letters to the editor of Archives of Pathology & Laboratory Medicine or CAP TODAY. These thoughtful efforts help us to review problematic issues so that future revisions can address them.

To date, most concerns relate to specimen handling and its effect on human epidermal growth factor receptor 2 (HER2) test results. Understandably, there is consternation expressed about the lack of scientific evidence related to fixation types and fixation times. We share that consternation. The panel reviewed existing scant information and depended also on the extensive experience of many of the panel members and experts from large commercial laboratories who provided us with their input. All guidelines created by the American Society of Clinical Oncology or the College of American Pathologists in the past have depended on both scientific evidence and expert opinion, especially when existing evidence was scant or nonexistent.

It was the considered and unanimous opinion of panel experts that defining the guideline fixative should be based on the requirements of the existing Food and Drug Administration-approved testing methods for HER2. All of those methods specify that tissue must be fixed in buffered formalin, although the length of fixation time is not described. The panel and these experts also felt strongly that the guideline needed to define standardized fixation intervals and then modify those requirements if necessary when and if data about optimal intervals became available. The expert panel members felt that intervals of six to 48 hours were most commonly recommended for breast cancer predictive tests. I presented published data from Intermountain Healthcare, which clearly demonstrated the adverse impact of prolonged fixation on estrogen receptor results in our health care system.1 Because breast specimens are submitted for estrogen receptor assays as well as for HER2 assays, these intervals seemed the most appropriate recommendation at this time. The adverse effects of underfixation are well known, as is pointed out by the letter writer and documented in the literature.2-4 It is likely that the recommendation for fixation of core biopsies will also be revised in the future to be a minimum of six hours, based on recent published work.2 We look forward to other data validating or suggesting modification of fixation intervals during the next year. The panel is committed to modifying the document as needed based on input from the pathology and oncology community.

Our recommendation to provide information in any negative HER2 report about the potential impact of prolonged or foreshortened fixation time was included to guide oncologists in clinical decisions. Oncologists usually assume that any test result is accurate. Providing information about mitigating circumstances will help them to understand the complexities of these test interpretations so that they can make better informed decisions for their patients. The intent of all pathologists and oncologists is to provide the most appropriate treatment to patients based on accurate testing data. It is fervently hoped that the current guideline will provide us with strategies to make such accurate testing more widespread. The panel is committed to changing the guideline when and if data are provided to recommend those changes.


  1. Nkoy F, Hammond EH, Rees WL, et al. Day of surgery affects estrogen receptor test results in women with breast cancer. The 28th San Antonio Breast Cancer Symposium; December 2005. Abstract 5107.
  2. Goldstein NS, Ferkowicz M, Odish E, et al. Minimum formalin fixation time for consistent estrogen receptor immunohistochemical staining of invasive breast carcinoma. Am J Clin Pathol. 2003;120:86-92.
  3. Nadji M, Gomez-Fernandez C, Ganjei-Azar P, et al. Immunohistochemistry of estrogen and progesterone receptors reconsidered: experience with 5,993 breast cancers. Am J Clin Pathol. 2005;123:21-27.
  4. Yaziji H, Goldstein LC, Barry TS, et al. HER-2 testing in breast cancer using parallel tissue-based methods. JAMA. 2004;29:1972-1977.

bullet Unacceptable specimens

I concur with Gregory Tetrault, MD (Letter, August 2007, page 6). The adage "garbage in, garbage out" certainly applies to results reported out on unacceptable specimens.

I have been a medical technologist for almost 40 years, the last 17 in management but I still work the bench. Though preanalytical factors are now buzzwords, they have been important to most laboratorians for years. Standard laboratory practice includes not running tests on hemolyzed specimens, unfilled coagulation tubes, specimens drawn above an IV, clotted CBCs, and so on. In my opinion, if you run tests on these specimens, it is comparable to running tests on the wrong patient.

We spend a lot of time in the laboratory ensuring the integrity of the patient samples. We then run quality control, look at Levey-Jennings charts, follow Westgard rules, etc., to give clinicians the most accurate laboratory test results possible so they can best treat their patients.

If you establish a reputation as a quality-driven, reliable laboratory, clinicians will call you when they question a result, and this is the best relationship you can establish for the care of patients. On the other hand, if you report out inaccurate results, or physiologically impossible results, even with a disclaimer, without trying to obtain a new specimen, you are not only doing a disservice to the laboratory but also to the clinician and, most important, to the patient.

Most laboratorians I have come across over the years consider themselves professionals, and, much like the medical doctor oath, we strive to "do no harm."

Marilyn C. Kenyon, MT(ASCP)
Director of Laboratory Services

St. Joseph Hospital

Bangor, Me.