Vol. 19 • Issue 5 • Page 30
“I have your test results. It’s cancer.”
Nearly 1.5 million people in the United States will hear those words this year.1Approximately 200,000 of those people will be women diagnosed with breast cancer.2And while advances in early detection have led to better survival rates, cancer still has the upper hand in the industry’s ongoing fight to improve diagnostics, ðtreatment and the search for a cure.
A steady stream of diagnostic tools, new understanding, treatment methods and classification techniques has led to an ongoing evolution in how breast cancer is identified, classified and treated. Assessing certain predictive factors can help the team of caregivers understand a patient’s specific case-including the likelihood of cancer metastasizing, responding to treatment and, if a patient is treated and becomes cancer-free, the probability of recurrence.
Recent years have seen inherent changes in the identification, classification and handling of breast cancer with a near-constant improvement in understanding genotypes. In addition to the traditional morphologic classification, breast cancer cells are now being identified based on their genotype, explains Elizabeth Hammond, MD, pathologist, Intermountain Healthcare and director of the Department of Pathology at the University of Utah School of Medicine in Salt Lake City. “It turns out that the distinction between lobular or ductal is less important than how differentiated the cells are,” she explains, which was the focus of previous classification methods.
William Smith, MD, chairman of the Department of Pathology, Suburban Hospital/Johns Hopkins Medicine in Bethesda, MD, adds, “We’re seeing more and more acceptance of the concept that the biology of the tumor is what drives the outcomes and the responses to therapy. We’re moving away from traditional predictive factors such as tumor size and nodal status because the biology of the tumor can differ widely in any of those groups.”
“The new classification divides the cells into luminal and basal. The basal type is so primitive that they don’t express any of the usual markers for breast cancer,” Dr. ðHammond explains. Luminal cancers express estrogen receptor (ER), progesterone receptor (PR) or human epidermal growth factor receptor 2 (HER2). But the change in classification is still in flux, she adds, and there are still people who are more comfortable with the old method.
Basal breast cancers also are referred to as triple-negative, as they lack receptors to estrogen, progesterone and HER2. Within each of these groups, however, there are many more subsets that require different treatments and have varying prognostic implications for the patient. Of course, Dr. Hammond notes, “cancers don’t have to obey any rules, so they can change in any way they want into any of the other types.”
Still, establishing whether any of these three most significant biomarkers are expressed helps identify treatment options and predicts response to those treatments. Basal tumors that don’t express any of these tend to be high-grade and hard to treat.
A New Gold Standard
Thus, ER/PR immunohistochemical (IHC) testing is critical for determining treatment and assessing prognosis for breast cancer patients. However, lack of standardization for the test has led to serious reliability concerns. The American Society of Clinical Oncology/College of American Pathologists Guideline Recommendations for Immunohistochemical Testing of Estrogen and Progesterone Receptors in Breast Cancer3is a new recommendation published in the online April issue of Archives of Pathology and Laboratory Medicine that aims to fix those concerns. Prolonged specimen handling times, manual IHC techniques and use of lab-specific staining methods have all been shown to contribute to breast cancers being falsely reported as ðnegative for ER/PR, as cited in the study. Standardizing the pre-analytic sample handling procedures could greatly improve the accuracy of ER/PR testing.
Of the new guidelines, Dr. Hammond-one of the study’s authors-tells ADVANCE that the goal is to get laboratories to use an exact, standardized method to test for ER and PR to predict a patient’s response. She adds that while not specified in the guidelines, “It’s valuable to use an automated platform rather than a manual platform, because manual platforms are much more likely to have variations. Using image analysis for the interpretation is also a good idea. What is required,” she stresses, “is that the samples be handled in a very specific way to ensure the antigen you’re looking for is actually there.
“The new guideline specifies, very clearly, exactly what you have to do to a specimen to make sure a test is done accurately,” Dr. Hammond adds. “We need to know the time the tissue came out of the patient, what time it went into the fixative, that it was fixed for six to 72 hours, and that the fixative was neutral-buffered formalin.”
Dr. Hammond likens the guideline to creating a detailed recipe. “Instead of just saying bake a cake with some flour, sugar, eggs and milk, then cook it for a while, we’re specifying exactly what must be done.”
A major problem that led to poor results in the past was unclear reporting requirements-there was no clear definition of what to call positive. Now, if you look at a whole slide, 1% of those tumor cells should be positive for ER or PR to call that result positive. The internal elements of normal breast tissue in the sample should also express some estrogen or progesterone activity. This goes a long way in eliminating variation between pathologists interpreting results.
Other Predictive Factors
Predictive factors are predicated on the idea that they can identify which patients are more likely to benefit from adjuvant treatment and whether or not the tumor is likely to return. While ER, PR and HER2 status are the most important determinations because they help establish therapy requirements, other biomarkers also come into play. EGFR, for example, and cytokeratin 5/6, 8 and 18 are used to help determine the subgroup of a patient’s cancer.
Other prognostic tools include gene expression profiles (Oncotype DX®from Genomic Health®and MammaPrint®by Agendia Inc.® that predict outcomes based on different sets of genes. Says Dr. Hammond, “These are predicated on the idea that they can identify which patients will benefit from treatment of any kind and whether or not the tumor is likely to come back.”
Dr. Smith adds that these gene expression profiles are useful for patients that aren’t as clear-cut as some ER+, PR+ or HER2+ patients are. Such profiles help measure and improve information that can be given to the oncologist to make prognostic and predictive evaluations.
Predictive assays will likely continue to come to market as more biomarkers are discovered, gene interactions are understood, and targeted therapies are developed. Staying abreast of guidelines and ensuring standardization and careful results reporting are crucial.
Kelly J. Graham is associate editor.
1. Cancer Facts and Figures 2009. American Cancer Society. Accessed at http://www.cancer.org/downloads/STT/500809web.pdf. Last accessed April 22, 2010.
2. Breast Cancer Statistics. Susan G. Komen for the Cure. Accessed at http://ww5.komen.org/BreastCancer/Statistics.html. Last accessed April 22, 2010.
3. Hammond MEH, Hayes DF, et al. American Society of Clinical Oncology/College of American Pathologists Guideline Recommendations for Immunohistochemical Testing of Estrogen and Progesterone Receptors in Breast Cancer. Arch Pathol Lab Med 2010;134:E1-E16.