NATURE MEDICINE • VOLUME 8 • NUMBER 12 • DECEMBER 2002 1341 COMMENTARY Tamoxifen binds to the estrogen receptor and competitively blocks the prolifera- tive effects of estrogen on breast cancer through direct interaction with estrogen receptor–positive cells and paracrine influences on neighbor- ing estrogen receptor–negative cells (Fig. 1). Approximately two-thirds of women with estrogen receptor–containing metastatic tumors respond to treatment, and tamoxifen re- duces the annual odds of death by 25% when given for five years after surgery to patients with early disease, whether or not they are also treated with chemo- or radiotherapy. In this latter setting, the drug has become established over the last two decades as the bedrock of medical therapy for breast cancer and has extended or saved the lives of thousands of patients. Tamoxifen also reduces the incidence of new tumors in the contralateral breast by half, an observation that led to trials of its use for prevention in women at high risk of the disease. The NSABP-P1 tamoxifen prevention trial 1 found a 49% reduction in the early incidence of breast cancer in women at increased risk, and this effect has recently been confirmed by the International Breast Cancer Intervention Study (IBIS) report of a 33% reduction 2 . The long-term hegemony of tamoxifen is being challenged, however, by the introduction of new and potent inhibitors of aromatase, the enzyme that converts androgens to estrogens (Fig. 1). Aromatase is present at high concentrations in the ovaries of premenopausal women. The feedback control of es- trogen production renders this group of breast cancer patients unsuitable for aromatase inhibition. In postmenopausal women, the ovary becomes devoid of aromatase but residual estrogens are synthesized by aromatase present elsewhere, pre- dominantly in subcutaneous fat and in other peripheral tissues including normal breast and some breast cancers. Aromatase inhibitors were first used to treat metastatic breast cancer in 1973, but the early compounds, such as amino- glutethimide, were subject to problems with toxicity and po- tency. New compounds—the triazoles anastrozole and letrozole and the steroid exemestane—are relatively nontoxic and virtually eliminate aromatase activity in postmenopausal women. All three of the new inhibitors show some superiority over tamoxifen in the treatment of metastatic breast cancer 3 , and we now have a first report of the effectiveness of one, anastrozole, in preventing relapse after surgery. The so-called Anastrozoic, Tamoxifen Alone or in Combination (ATAC) trial, which com- pared anastrozole with tamoxifen and with both treatments combined, is the largest therapeutic cancer trial ever per- formed, with over 9,000 patients recruited 4 . Over a median fol- low-up period of 33 months, anastrozole enhanced disease-free survival by 19% and decreased the incidence of new contralat- eral tumors by 58% above the effect of tamoxifen. Longer fol- low-up is required to be certain of the magnitude of the increased benefit for patients, to determine whether it trans- lates to a survival advantage and to as- sess the significance of the different side-effect profiles of tamoxifen and anastrozole. Nonetheless, considering all results to date, it seems that aromatase inhibitors are likely to topple tamoxifen from its current position as the standard endocrine therapy in postmenopausal women. A further notable result from the ATAC trial was that patients in the combination arm performed less well than those receiv- ing aromatase inhibitor alone, and nearly the same as those re- ceiving tamoxifen alone, in regard to both anti–breast cancer effects and all of the numerous tolerability end-points. Thus the presence of tamoxifen seemed to negate any beneficial ef- fects of anastrozole. Both tamoxifen and aromatase inhibitors act by depriving estrogen receptor–positive cells of stimulatory estrogen signals, and both treatments act through the estrogen receptor (Fig. 1). There are known to be two distinct estrogen receptors, ER-α and ER-β, but at present no discrete role for ER-β has been es- tablished in breast cancer. Why should blocking the estrogen receptor while using tamoxifen lead to inferior therapeutic ef- fects? And why should using tamoxifen in addition to with- drawing estrogen be inferior to withdrawing estrogen alone? The short answer is that we are not sure, but there are substan- tial clues, primarily related to the molecular pharmacology of the drugs. Binding of estradiol to the monomeric 65-kD estrogen recep- tor protein leads to a cascade of events: displacement of heat- shock proteins, dimerization, phosphorylation of activating function 1 (AF1), conformational change in the ligand-binding domain or activation function 2 (AF2), and binding of the li- gand–receptor dimer to estrogen response elements upstream of estrogen-responsive genes (Fig. 2). The conformational change promotes binding of co-activators that stabilize the transcriptional complex, possess histone acetylase activity and induce enhanced transcription. The near-complete withdrawal of estrogen achieved by contemporary aromatase inhibitors markedly attenuates this signal transduction, decreases bind- ing of estrogen receptor to estrogen response elements and leads to decreased transcription. Binding of tamoxifen or its active metabolites leads to a sim- ilar cascade of events, but the resulting conformational change in the estrogen receptor is different. Activation of AF1 occurs but activation of AF2 does not, such that incomplete antago- nism results: whereas genes dependent on AF2 signaling are antagonized, those dependent on AF1 are promoted. Enhanced binding of co-repressors also occurs, and the relative concen- tration of co-activators and co-repressors seems to influence the degree of agonism or antagonism. This differs between tis- Breast cancer is the most common malignancy in females in most western countries, with about 1 in 10 women at risk of developing the disease in the course of their lifetimes. Since its introduction over 30 years ago, tamoxifen has been the mainstay of the endocrine treatment of breast cancer. It has become the most widely used anticancer drug, and may be thought of as among the first targeted therapies. Yet the results of a major new trial of endocrine therapy after surgery for breast cancer using aromatase inhibitors cast into question tamoxifen’s future role. Breast cancer: Aromatase inhibitors take on tamoxifen MITCH DOWSETT 1 & ANTHONY HOWELL 2 © 2002 Nature Publishing Group http://www.nature.com/naturemedicine