Most endocrine therapies for breast cancer inhibit tumor growth by depriving the cell of estrogen or by blocking its receptor. However, some drugs, such as tamoxifen, can bind to the estrogen receptor (ER) and have both estrogenic and antiestrogenic effects, depending on the tissue, cell, or promoter context. These mixed properties may be explained by new information on ER function at the molecular level. Whether a synthetic drug acts as an estrogen or antiestrogen on a specific gene may be dictated by the particular ensemble of ER subtype, receptor interacting proteins, other transcription factors, or specific elements within the promoter of estrogen-regulated genes. Alterations in these other factors may also play a role in resistance to hormonal therapies. Aromatase inhibitors, like ovarian ablation, inhibit growth by lowering the estrogen concentration in blood or in the tumor tissue itself. Aromatase inhibitors are effective even in postmenopausal women with low estrogen concentrations - probably because of the ability of the tumor to become hypersensitive to estrogen after prolonged estrogen deprivation. Given that the ER itself is the prime target for endocrine manipulation, the ideal endocrine therapy may be one that reduces or eliminates ER from the tumor cell. Pure steroidal antiestrogens are then of great interest because, not only do they inhibit ER-induced transactivation of estrogen regulated genes, they also induce ER degradation. Additional clinical trials are necessary to identify the optimal endocrine therapy and optimal sequence of available therapies.