The essential role estrogen plays in reproductive endocrinology has been deciphered during the 20th century. Estrogen is also important for supporting physiologic homeostasis in a woman's body as evidenced by the progressive changes that occur at menopause when ovarian estrogen synthesis stops around the age of 50. Knowledge of the biological actions of sex steroids (estrogens and progestins) resulted in the development of oral contraceptives to prevent pregnancy and the application of estrogen replacement therapy (ERT) as a supplement to alleviate symptoms and urogential atrophy at the time of menopause. These are landmarks in drug development because, for the first time, the target populations were well women and a disease state was not being treated or prevented. In other words, the quality of life for the individual is improved by either planned parenthood or a short course of hormone replacement. However, the strong inverse relationship between age and bone mineral density in the decade following menopause suggested that hormone replacement could elevate bone density and reduce the risk of fractures in elderly women. The protective effect of hormone replacement therapy (HRT), i.e., a combination of estrogen and progestin, on increasing bone mineral density is clearly demonstrated in the postmenopausal/progestin intervention (PEPI) trial. Women taking placebo lost an average of 1.8% and 1.7% in bone mineral density of the spine and hip, respectively, over the 3 years of the study. In contrast, women taking HRT gained an average of between 3.5% and 5% in the spine and 1.7% in the hip. It is estimated that current users of HRT have a 50% and 25% reduced risk of vertebral and hip fractures, respectively. Hormone replacement therapy is used widely by postmenopausal women for the treatment and prevention of osteoporosis. Additionally, the epidemiologic link between the use of HRT and a reduction of coronary heart disease (CHD) and Alzheimer's disease provoked a wider use of HRT by well women. Unfortunately, the benefits of HRT based on prospective clinical trials has not, as yet, been demonstrated for Alzheimer's or CHD and the cardiovascular system. Indeed, the recent results from the Women's Health Initiative demonstrate, in a placebo controlled trial of 160 000 postmenopausal women, an increase in heart disease (23%), stroke (38%), and blood clots (100%). Although the absolute changes per 1000 women were only increased from 3.0 (placebo) to 3.7 (HRT) for heart disease, 2.1 (placebo) to 2.9 (HRT) for stroke, and 1.3 (placebo) to 2.6 (HRT) for blood clots, the trend is all in the wrong direction. Despite the unproven benefits of HRT in Alzheimer's and CHD, estrogen does encourage the development and growth of cancer in the breast and uterus. The Women's Health Initiative noted an increase in breast cancer of 26% with an absolute change per 1000 women of 3.0 (placebo) to 3.8 (HRT). The link between ovarian hormones and breast cancer has been known throughout the 20th century. Oophorectomy of premenopausal patients with metastatic breast cancer caused tumor regression in approximately one-third. The reason for this apparently arbitrary responsiveness to estrogen withdrawal was unknown until the discovery of the estrogen receptor (now referred to as ERα) and the application of the knowledge to predict the hormone responsiveness of breast cancer. In the late 1950s, Jensen and Jacobson synthesized the first high specific activity tritium labeled estradiol-17β. They showed that radiolabeled estradiol was bound to, and was retained by, estrogen target tissues (uterus, vagina, pituitary gland) but was not retained by nontarget tissues such as muscle or lung. They hypothesized that there must be a receptor molecule for estrogen in target tissues that initiates the cascade of biochemical events associated with estrogen action at that site. Subsequently, the ER was isolated as an extractable protein from rat uterus and subcellular models of estrogen action were designed and refined. However, Jensen took these concepts one step further by suggesting that if the ER were present in a breast tumor, then this would increase the probability of a response to endocrine ablative therapy (oophorectomy, adrenalectomy, hypophysectomy). This was shown to be true and is the basis for the steroid receptor assay used routinely in the prediction of endocrine sensitivity of breast cancer. Overall, the discovery of the ER rationalized the target site-specific effect of estrogen around a woman's body. The finding of ER in breast tumors also provided the rationale for the eventual development of antiestrogens as a safe and simple alternative to ablative surgery for the treatment of breast cancer. However, the development of a simple, nonsteroidal antiestrogen tamoxifen that blocks the estrogen-stimulated growth of breast cancer was to become the key to discovering selective estrogen receptor modulation. The nonsteroidal antiestrogens tamoxifen and raloxifene prevent the development of carcinogen-induced rat mammary carcinoma but were tested, in 1986, to determine whether they would have a detrimental effect on bone density. These studies were conducted with a view to using nonsteroidal antiestrogens for chemoprevention. Indeed, at that time, a pilot study of tamoxifen as a chemopreventive in high-risk women was initiated at the Royal Marsden Hospital in London. Both tamoxifen and raloxifene maintained bone density in ovariectomized rats. These data were subsequently translated to the clinic where tamoxifen was found to preserve bone density in postmenopausal breast cancer patients. Although tamoxifen was not considered for use as a drug to prevent osteoporosis, the safety of the drug in bone provided important assurances to advance with the testing of tamoxifen as a chemopreventive in high-risk women. However, not all women who develop breast cancer have risk factors other than age. The following question therefore arose: Could breast cancer be prevented in postmenopausal women without identifiable risk factors? In 1990, on the basis of existing laboratory data, a paradigm shift was proposed to develop tamoxifen analogues that would prevent osteoporosis and atherosclerosis and would prevent breast cancer as a beneficial side effect. The result was raloxifene, which is now the first selective estrogen receptor modulator (SERM) to be available for the treatment and prevention of osteoporosis that is also being tested as a preventive for breast cancer and CHD. Part 1 of this review will describe the development of the scientific ideas about antiestrogen action and the evolution of our understanding of their molecular mechanisms of action. As in most areas of pharmacology, the use of antagonists and partial agonists has provided enormous insight into the mechanics of receptor function and the associated molecules that must be recruited to complete a signal transduction pathway. The targets for SERM action are ERs, and developing molecular knowledge is now being utilized to dissect the multiple mechanisms of estrogen action. Through the application of this knowledge, new strategies of drug discovery can be exploited either to develop the ideal SERM, as a multifunctional medicine, or to target SERMs to specific organ sites. The clinical considerations and novel agents will be addressed in part 2.