Background: Determination of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) status is standard for predicting prognosis and determining treatment options for patients with breast cancer. In 2010, the American Society of Clinical Oncology (ASCO) and College of American Pathologists (CAP) issued guidelines that tumors with ≥1% positively staining cells should be considered ER positive. Here, we determined how this cutoff relates to molecular subtype.
Methods: Clinicopathological characteristics were compared between ER-negative, ER-positive, and low-ER-staining (1-10%) tumors using chi-square analysis with P<0.05 defining statistical significance. Gene expression data were generated for 26 low-ER-staining tumors, and their intrinsic subtype determined. Immunohistochemistry (IHC)-defined surrogate subtypes, using the threshold of positivity defined by ASCO/CAP guidelines, were compared with molecular subtypes.
Results: Low-ER-staining tumors were clinicopathologically more similar to ER-negative than to ER-positive tumors; 88% of low-staining tumors were basal like or HER2 enriched. Only those tumors expressing 10% ER-positive cells were classified as luminal A subtype.
Conclusions: Under ASCO/CAP guidelines, tumors with 1-10% ER staining would be classified as ER positive, yet most are basal like or HER2 enriched and have pathological features similar to ER-negative tumors. Clinical trials seeking to treat tumors of ER-negative basal-like and/or HER2-enriched subtypes should thus not preclude enrollment based solely on results of ER immunohistochemistry. As ER status is a critical element in the choice of treatments for patients with breast cancer, it is imperative that the most effective method for classifying tumors be developed.