Estrogen receptor alpha/beta ratio and estrogen receptor beta as predictors of endocrine therapy responsiveness-a randomized neoadjuvant trial comparison between anastrozole and tamoxifen for the treatment of postmenopausal breast cancer

Background: The role of estrogen receptor beta (ER-beta) in breast cancer (BC) remains unclear. Some studies have suggested that ER-beta may oppose the actions of estrogen receptor alpha (ER-alpha), and clinical evidence has indicated that the loss of ER-beta expression is associated with a poor prognosis and resistance to endocrine therapy. The objective of the present study was to determine the role of ER-beta and the ER-alpha/ER-beta ratio in predicting the response to endocrine therapy and whether different regimens have any effect on ER-beta expression levels. Methods: Ninety postmenopausal patients with primary BC were recruited for a short-term double-blinded randomized prospective controlled study. To determine tumor cell proliferation, we measured the expression of Ki67 in tumor biopsy samples taken before and after 26 days of treatment with anastrozole 1 mg/day (N = 25), tamoxifen 20 mg/day (N = 24) or placebo (N = 29) of 78 participants. The pre-and post-samples were placed in tissue microarray blocks and submitted for immunohistochemical assay. Biomarker statuses (ER-beta, ER-alpha and Ki67) were obtained by comparing each immunohistochemical evaluation of the pre- and post-surgery samples using the semi-quantitative Allred's method. Statistical analyses were performed using an ANOVA and Spearman's correlation coefficient tests, with significance at p <= 0.05. Results: The frequency of ER-beta expression did not change after treatment (p = 0.33). There were no significant changes in Ki67 levels in ER-beta-negative cases (p = 0.45), but in the ER-beta-positive cases, the anastrozole (p = 0.01) and tamoxifen groups (p = 0.04) presented a significant reduction in post-treatment Ki67 scores. There was a weak but positive correlation between the ER-alpha and ER-beta expression levels. Only patients with an ER-alpha/ER-beta expression ratio between 1 and 1.5 demonstrated significant differences in Ki67 levels after treatment with anastrozole (p = 0.005) and tamoxifen (p = 0.026). Conclusions: Our results provide additional data that indicate that the measurement of ER-beta in BC patients may help predict tamoxifen and anastrozole responsiveness in the neoadjuvant setting. These effects of hormonal treatment appear to be dependent on the ratio of ER-alpha/ER-beta expression.