Chicago Breast Health Project (CBHP)

Concerns about breast cancer, the most commonly diagnosed cancer among women in the United States, cause some women to avoid postmenopausal hormone replacement therapy (HRT). Indeed, there appears to be an approximately 30-40% higher risk of breast cancer for women who used HRT for 5 years or more compared to never users. It has been proposed that exposure to HRT promotes the growth of less aggressive, slow growing tumors. In a large prospective study of postmenopausal women, we recently observed no association between ever use of HRT and the incidence of ductal carcinoma in situ, or with invasive ductal/lobular carcinomas, which account for 85-90% of invasive tumors. However, duration of ever HRT use was significantly and positively associated with risk of invasive carcinomas with a favorable histology (i.e., medullary, papillary, mucinous and tubular), which are less commonly occurring, have a lower risk of axillary node metastases, better prognosis and are more likely to be estrogen receptor (ER) positive. A significant increase in risk of invasive breast cancers with a favorable histology among HRT users provides one possible explanation for the lower breast cancer mortality rates (or improved survival) among hormone users observed in some studies.

There are several biologically plausible reasons for a specific risk association of HRT with favorable histologic types of breast cancers. Some studies found that breast tumors among estrogen users appear to be well-differentiated, with fewer cells in mitosis compared to breast tumors among women who never used HRT. Moreover, results of in vitro studies indicate that treatment of ER positive breast cancer cells with estrogen results in down regulation of Her2/neu, a tyrosine kinase receptor expressed in approximately 25-30% of all breast tumors that is a marker of poor prognosis. Conversely, the anti-estrogen tamoxifen appears to increase Her2/neu expression in vitro. We are aware of only one study that compared Her2/neu expression in HRT users compared to non-users. In that study, there was a weak, non-statistically significant, positive association between HRT use and Her2/neu expression. However, some of the limitations of that study were that analyses were not conducted separately for current vs. past HRT users, or for ER+ vs. ER- breast cancers.

Given the wide spread use of HRT among postmenopausal women in the United States, further research is warranted to better understand differences in Her2/neu expression among current HRT users, past users and non-users taking into account other prognostic tumor characteristics such as ER status. In this study, we will determine the independent associations of current, past and never use of HRT with Her2/neu expression in breast tumors of postmenopausal women. These results could have important implications for better understanding the etiologic association of HRT use with breast cancer risk and prognosis.