Guideline Adherence Is Vital to HER2 Accuracy

M. Elizabeth Hammond, MD

Professor, Pathology

University of Utah School of Medicine, Chair, Department of Pathology, LDS Hospital and Urban Central Region, Intermountain Health Care, Salt Lake City, UT

Oncologists know the power of trastuzumab (Herceptin) in treating patients with breast cancer that overexpresses human epidermal growth factor receptor 2 (HER2). But many have been frustrated by a common barrier to using the drug effectively: test results that incorrectly label tumors as positive or negative for HER2 expression, according to M. Elizabeth Hammond, MD, a pathologist who has helped set the standards for improved assessment of patients’ status.

Although progress has been made on improving the accuracy and understanding of HER2 testing, pitfalls remain, Hammond noted during a presentation at the 11th International Congress on the Future of Breast Cancer, held in Coronado, California, in July 2012. Efforts to improve guidelines continue, not only for HER2 testing but also in evaluating whether patients’ tumors are estrogen receptor (ER)-positive and progesterone receptor (PR)-positive. Hammond said a variety of issues could lead to inaccuracies in specimen handling, analytical testing, interpretation, and reporting of results.

The first guidelines for HER2 testing¹ were published in 2007 by the National Comprehensive Cancer Network in collaboration with the American Society of Clinical Oncology (ASCO)/College of American Pathologists (CAP); guidelines for ER/PR testing followed in 2010.² Due out in 2013, updated HER2 guidelines will offer more information about what to do when a test doesn’t work as expected, or when two labs generate divergent findings for the same patient, with a goal of keeping the nation’s false-negative and false-positive results below 5% each, Hammond said.

While tests are now done by immunohistochemistry or fluorescence in situ hybridization (FISH), the 2013 guidelines will also list bright-field in situ hybridization as an acceptable method, she said.

Inconsistencies Examined

“I have only one motivation, and that is that I want people to improve this testing,” the pathologist told OncologyLive in an interview. “I want the tests to be right. Herceptin is a very effective drug. It’s the most effective drug for breast cancer we have, but the only way it can be applied properly is if the test is done right, and so I want the test to be right every time.”According to three large clinical trials whose results were released around 2002, immunohistochemistry tests for HER2- positive breast cancer were associated with a 20% false-positive rate and a 10% false-negative rate, while FISH testing garnered a 15% rate of false-negative results, Hammond said. Similar pitfalls can occur when testing for ER-positive and PR-positive cancers, she said.

As an example of these inconsistencies in test results, Hammond cited a retrospective study by Intermountain Healthcare that looked at 5044 test results for ER activity conducted between 1997 and 2005.³ The tissue was collected at seven hospitals and then tested in a central lab using standard processes and interpretation.

After controlling for age, stage, and grade, the study found that the rates of ER-negative diagnoses were different for each of the hospitals where tissue had been collected, with 28.6% being the highest rate and 16.5% the lowest.

Investigators proposed that the differences were due to varied specimen-handling practices, and considered whether the day of the week on which tissue was collected made a difference. They found that if a specimen was removed Monday through Thursday, it had a statistically significantly lower chance of being diagnosed as ER-negative than if it was removed on Friday or Saturday, Hammond said.

Research toward understanding such inconsistencies, which led to the establishment of the first guidelines for HER2 testing, revealed that enzymatic processes begin degrading breast tissue as soon as it is removed from a patient, making it extremely important to put samples into a fixative such as formalin within one hour, Hammond said.

OncLive TV Exclusive

Dr. Hammond Discusses HER2 Testing Guideline Revisions

Samples should remain in the fixative solution for at least six hours, but not for longer than 72 hours, and then should be promptly examined, she said. Especially when a sample is going to be analyzed at a facility other than where it was collected, she added, time points should be recorded so that the pathologist who examines the tissue knows it was handled properly.

Simply put, Hammond said, when a sample is ER-positive and fixed properly, “you will see ER activity, but if you fix it in a short[cut] way—which is more likely in the United States, where we want to do everything fast—you’ll get a bad result.”

In fact, she said, sample mishandling is the main reason for false-positives and false-negatives in testing for HER2, ER, and PR overexpression, and accounts for more than half of those erroneous results.

The type of sample used can also have an effect on the accuracy of test results, Hammond said. Despite some risk of missing heterogeneous expression areas, the scientist recommended starting with core specimens and then, if adequate results aren’t achieved, testing resection specimens.

While analytic pitfalls are much less common, pathologists and labs need to prevent them by checking results against external control samples, placed on the test slide for side-by-side comparison, Hammond said. Controls should include weakly positive samples and, when possible, samples of normal breast tissue from the patient in question, she said.

Experts should consider a specimen ER- or PR-positive if 1% of its cells are positive.

Algorithm for HER2 Testing

Three categories of test results for each test

1 Positive

• IHC HER2 protein expression of 3+ OR
• FISH HER2 gene/CEP17 ratio ≥2.2 OR
• FISH HER2 gene copy number of >6.0

2 Equivocal

• IHC HER2 protein expression of 2+ OR
• FISH HER2 gene/CEP17 ratio of 1.8-2.2 OR
• FISH HER2 gene copy number of 4.0-6.0

3 Negative

• IHC HER2 protein expression of 1+ or 0 OR
• FISH HER2 gene/CEP17 <1.8 OR
• FISH HER2 gene copy number of <4.0

"CEP indicates chromosome enumeration probe; FISH, fluorescence in situ hybridization; HER2, human epidermal growth factor receptor 2; IHC, immunohistochemistry.

Source

Hammond, ME. Pitfalls and solutions in HER2/ER/PR testing. Presented at: 11th International Congress on the Future of Breast Cancer; July 26-28, 2012; Coronado, CA.

A HER2 test is positive if membrane staining is thick and uniform and the cells form a chicken wire-like pattern. An IHC test should show protein expression of 3+; alternatively, a FISH test should show a HER2 gene/CEP17 ratio ≥2.2 or a HER2 gene copy number of >6.0, Hammond said. A cancer should be diagnosed as HER2-negative, she said, when IHC shows protein expression of 1+ or 0, or when FISH reveals a HER2 gene/CEP17 ratio of <1.8 or a HER2 gene copy number of <4.0.

Interpreting Results

When HER2 results are equivocal, testing by a second method should be conducted for more definitive results, Hammond said. In IHC testing, equivocal results are defined as HER2 protein expression of 2+. FISH results are equivocal if they show a HER2 gene/CEP17 ratio of 1.8-2.2, or a gene copy number of 4.0 to 6.0, she said.In interpreting results, it’s vital for pathologists to consider whether findings fit with a patient’s clinical profile, Hammond added. Tests should correlate with grade and histology, with ER- and PR-positive tumors typically being low-grade and HER2 tumors usually high-grade. “If it doesn’t fit the clinical profile, it’s very important that you conduct a mandatory retest of the information,” she said.

The first ASCO/CAP guidelines include standards for determining whether a test is positive or negative; which samples should be used; exclusion criteria; how labs should be monitored and what quality assurance they should do; and education that should be available to help pathologists succeed.

Since the publication of the guidelines, “the number of labs submitting for inspection, and doing proficiency testing, has dramatically gone up,” Hammond said. “Surveys show marked improvement in those labs, so these methods appear to be working.”

To increase their knowledge, Hammond said, pathologists could take a 28-hour continuing medical education program, the “Advanced Practice Program in Breast Predictive Factors,” sponsored by the College of American Pathologists, or an equivalent course.

Oncologists in community practice, meanwhile, should communicate with pathologists in person to help ensure that testing is done correctly, Hammond told attendees.

“Have meaningful communication with the pathologists that serve you,” she said. “Ask them if they’re using the ASCO/CAP guideline, and if they understand what the recommendations are. Are they following the specimen-handling concerns? Can you help them make sure the times are actually recorded? Is their lab being inspected, and is it passing? If not, are the pathologists taking training to make sure they understand what they should be doing?”

“Remarkably, there are still a lot of pathologists in the United States who don’t understand what these guideline recommendations are,” Hammond said. “You, as oncologists, can have an important role in changing that.”

References

• Wolff AC, Hammond EH, Schwartz JN, et al. American Society of Clinical Oncology-College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer [published online ahead of print December 11, 2006]. J Clin Oncol. 2007;25(1): 118-145. doi: 10.1200/JCO.2006.09.2775
• Hammond MEH, Hayes DF, Dowsett M, et al. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer [published online ahead of print April 19, 2010]. J Clin Oncol. 2010:28(16);2784-2795. doi: 10.1200/ JCO.2009.25.6529.
• Nkoy FL, Hammond ME, Rees W, et al. Variable specimen handling affects hormone receptor test results in women with breast cancer: a large multihospital retrospective study. Arch Pathol Lab Med. 2010;134(4):606-612.