Clomiphene (Fig. 2), belonging to the triphenylethylene structural series, was the first SERM. Its discovery in 1960 preceded the discovery of tamoxifen, the most widely used SERM, by 2 years. Clomiphene is an ER antagonist and the primary choice for ovulatory dysfunction, including polycystic ovary syndrome (PCOS). Ovulatory dysfunction, such as PCOS, luteal phase deficiency and anovulation are commonly observed reproductive abnormalities in women. As an ER antagonist, clomiphene citrate suppresses the negative feedback inhibition of estradiol and increases secretion of follicle stimulating hormone (FSH), which in turn stimulates follicular development and results in an increase in ovulation rate. Some of the common side effects of clomiphene are hot flashes, abnormal uterine bleeding, nausea and/or vomiting. Although clomiphene increases ovulation by 80–90 % in women with PCOS, it increases pregnancy only by 25–30 %. Endometrial biopsies obtained from women treated with clomiphene indicated that it reduces the thickness of the endometrium, an effect that is more pronounced with clomiphene than other SERMs or aromatase inhibitors.

Clomiphene citrate is also being investigated as a treatment for male hypogonadism. Utilizing the same phenomenon to inhibit the negative feedback loop of gonadotropins, hypo gonadal men treated with clomiphene demonstrated a significant increase in serum testosterone levels and in ADAM scores (Androgen Deficiency in Ageing Male). Interestingly, hypo gonadal men receiving clomiphene for more than 2 years also demonstrated an increase in bone mineral density (BMD) and a decrease in osteoporosis (Moskovic et al. 2012). Data on the changes in BMD in hypo gonadal men and women with PCOS by clomiphene is conflicting. While some studies indicate an increase in BMD with clomiphene, others show that clomiphene might decrease BMD. These seemingly contradictory results could be due to the direct antagonistic effects of clomiphene on ER-α in bone, indirect effects of clomiphene to increase testosterone and estrogens, or differences in the overall hormonal milieu of women and men.

Preclinical studies have indicated that while clomiphene is an antagonist in ovary and endometrium, it is an agonist in the skeletal and cardiovascular systems. Interestingly, while E-clomiphene and Z-chlomiphene both have agonistic activities in bone and cardiovascular tissues, E-chlomiphene is an agonist and Z-chlomiphene is an antagonist in uterus.

Tamoxifen (Fig. 2) was discovered in 1962 by Imperial Chemical Industry (now Astra Zeneca) as a triphenylethylene SERM. Tamoxifen is the most widely used SERM for breast cancer and it functions through its active metabolite, hydroxytamoxifen. Tamoxifen has become the most widely studied SERM from the mechanistic perspective as it functions as an ER antagonist in breast, but agonist in uterus and endometrium. Tamoxifen binds to ER with a high affinity of 2 nM and antagonizes both isoforms of ER.

Tamoxifen is currently in use for node-negative breast cancer in pre- and post-menopausal women and for node-positive breast cancer in post-menopausal women. It is also effective in metastatic breast cancer and has recently been approved for the prevention of breast cancer in high risk women. Multiple clinical trials have demonstrated significant survival benefits in women with breast cancer treated with tamoxifen (Davies et al. 2013; Darby et al. 2011). Being a SERM, prolonged use of tamoxifen provided beneficial anti-cancer effects in the breast and also modestly preserved BMD and decreased the number of fractures (Ryan et al. 1991; Love et al. 1992). Since post-menopausal women undergoing ER-antagonistic treatments have a higher incidence of osteoporosis and fractures, tamoxifen’s small, but beneficial, effects on bone are considered one of the advantages for improved survival of these patients. Recently, tamoxifen also received attention for its potential for favorable effects on cardiovascular diseases (Love et al. 1994). A number of cardiovascular diseases markers such as LDL, HDL, and triglycerides are favorably altered by tamoxifen, making it potentially more valuable for patients with breast cancer.

Despite its many beneficial effects, one major concern with the use of tamoxifen is its effects on the endometrium. Endometrial cancer is increased by 2–4 fold in randomized trials (Chen et al. 2014). Uterine-related symptoms include vaginal bleeding and leucorrhoea. Multiple lines of evidence suggest that while tamoxifen functions as an ER antagonist in breast, it functions as an agonist in the uterus and endometrium (Wood et al. 2010). This warrants careful monitoring of breast cancer patients receiving tamoxifen for uterine cancer.

Raloxifene (Fig. 2) belonging to the benzothiophene structural series was discovered and developed by Eli Lilly for post-menopausal osteoporosis. Raloxifene provides unique SERM properties with one of the best profiles known. While raloxifene functions as an ER agonist in bone and in the regulation of serum lipids, it functions as an antagonist in breast and uterus, making it a highly valuable tool to combat post-menopausal osteoporosis without the disadvantages of activating ER in other tissues (D’Amelio and Isaia 2013). Studies on the effects of raloxifene on bone turnover markers indicated that raloxifene prevented bone loss and fractures at multiple sites with concurrent increase in BMD (D’Amelio and Isaia 2013). Bone turnover markers were significantly reduced and bone formations markers were up-regulated by raloxifene.

Recent evidence suggests that raloxifene significantly prevented breast cancer incidence in post-menopausal women, with results comparable to that of tamoxifen (Vogel et al. 2006). Women treated with raloxifene demonstrated a breast cancer incidence of only 1.7 per 1000, whereas placebo-treated women had a breast cancer incidence of 3.7 per 1000. The outcome of the Study of Tamoxifen and Raloxifene (STAR) trial indicated that raloxifene was as effective as tamoxifen in reducing invasive breast cancer without any thromboembolic side effects (Freedman et al. 2011). While raloxifene had a better benefit/risk ratio than tamoxifen in post-menopausal women with a uterus, raloxifene and tamoxifen had comparable benefit/risk ratios in post-menopausal women who had undergone hysterectomy. Overall, the clinical data indicates that raloxifene has a better benefit/risk ratio than tamoxifen.

Raloxifene lowered LDL cholesterol without affecting HDL cholesterol, suggesting that it might also have beneficial effects on these markers of cardiovascular disease (Francucci et al. 2005). Raloxifene also showed favorable effects on CNS function, including effects on climacteric symptoms such as hot flashes. Among all the commercially available first-generation SERMs, raloxifene is considered by many to have a near-perfect SERM profile, with favorable benefits in desired organs and a lack of unwarranted side effects.

Toremifene (Fig. 2) differs from tamoxifen by only one chlorine atom and possess properties comparable to that of tamoxifen (Buzdar and Hortobagyi 1998). Toremifene is also used for the treatment of advanced breast cancer. While toremifene functions as an ER antagonist in breast, it functions as an agonist with regard to bone, lipids, and uterus. Toremifene was also tested as a treatment for osteoporosis in men undergoing androgen deprivation therapy (ADT) (Smith et al. 2011). Although it protected bone loss in men undergoing ADT, the studies necessary to establish its efficacy and safety in this population were never completed. Similar to tamoxifen, toremifene’s active N-desmethyl and 4-hydroxylated metabolites bind to and antagonize ER more potently than the parent molecule.

Several SERMS such as lasofoxifene, arzoxifene, ospemifene, afimoxifene, arzoxifene, and bazodoxifene (Fig. 2) have been approved since 2000 (Komm and Chines 2012). All of these SERMs offer distinct tissue- selective properties and an unique profile compared to the first generation SERMs described above. While lasofoxifene has ER agonistic activity in bone, it has antagonistic activity in breast and uterus making it suitable for the treatment of osteoporosis, breast cancer, and vaginal atrophy. Phase III trials with lasofoxifene showed that it increased BMD significantly without any associated endometrial or thromboembolic effects (Malozowski 2010). In other trials, lasofoxifene improved vaginal atrophy, reduced ER-positive breast cancer growth and coronary disease risk (LaCroix et al. 2010; Ensrud et al. 2010; Goldstein et al. 2011). Similarly, bazodoxifene reduced the incidence of osteoporosis in post-menopausal women without causing an increase in endometrial hyperplasia (Miller et al. 2008). As second generation SERMs, these compounds have better safety profiles and exhibit near ideal SERM properties such as ER antagonistic effects in breast and uterus and agonistic properties in bone, blood and CNS, making them more broadly applicable to many women’s health issues.

The effects of estrogens are mediated by two isoforms of ER, ER-α and ER-β. Although the LBD of the two isoforms is only 59 % homologous, the ligand binding pocket (LBP) differs only by two amino acids, making them highly similar (Katzenellenbogen 2011). Despite this striking similarity, academic investigators and pharmaceutical companies have successfully developed ligands that bind preferentially to one or the other isoforms in order to take advantage of the beneficial effects of the individual isoforms without impinging on the function of the other isoform. Several structurally diverse molecules have been identified that bind to ER-β at low nanomolar to picomolar affinity, without cross reacting with ER-α (Fig. 3).

Diverse pharmacophores ranging from a simplistic linear diphenolic compound that achieves an inter-phenolic distance similar to estradiol to highly complex structures have been identified (Minutolo et al. 2011). The most common ER-β agonist template configuration is a (hetero) bicyclic compound substituted with a phenyl side chain such as genistein. Other templates include carbaldehyde oxime derivatives (Wyeth), hydroxyl-biphenyl carbaldehyde oximes (Wyeth), diaryl substituted salicylaldoximes (UIUC), alkyl linked (UIUC), cycloalkanes-linked biphenyls (Acadia), spiro indene-indenes (Merck) and many more. Academic and industrial groups have patented close to 50 different structural templates as ER-β ligands. These templates represent a heterogeneous mixture with regard to their level of characterization and degree of design success. Although several groups have been successful in discovering and developing high affinity ER-β-selective templates, Wyeth (now Pfizer) stands out as the leader in the field in terms of attaining extremely high affinity ligands and exploring a wide variety of templates. Further, Wyeth is the only pharmaceutical company to advance a synthetic ER-β agonist, prinaberel (ERB-041), to clinical trials. They explored a variety of clinical indications, including rheumatoid arthritis, Crohn’s disease, and endometriosis. However, the data were neither convincing nor published in entirety. The only other clinical candidate ER-β agonist, MF-101 (Bionovo) is an herbal extract containing a combination of ER-β agonists of plant origin. However their recently reported clinical trial results were only marginally successful in suppressing post-menopausal hot flashes. Diarylpropionitrile (DPN), which was initially developed by Katzenellenbogen et al., has been used extensively for academic exploration of the role of ER-β (Meyers et al. 2001).

ER-β agonists have been highly sought agents following the discovery of ER-β in 1996 (Kuiper et al. 1996). The distinct tissue-distribution alone seemed to suggest that tissue-selective estrogenicity was possible, which is further supported by the diverse and often opposing functional roles of ER-β, compared to ER-α. The theoretical problem of designing ER-β-selective agents was fully defined following the first crystal structure in 1999 of genistein bound to the ER-β LBD. Combined with the earlier ER-β LBD structures, this work revealed that the polar interactions available within the ER-α and ER-β ligand binding pockets are identical. Importantly, the ER-α and ER-β ligand binding pockets only varied by two amino acids, M336/L384 and I373/M421, producing topological and pocket size differences that necessarily must be the basis for subtype selectivity.

Multiple ER-β selective chemotypes with promising preclinical activity exist across a very broad chemical space. Many of these molecules have demonstrated promise as potential treatments for inflammation (Wyeth), cardiovascular diseases, obesity, and metabolic diseases (GTx), hormone therapy (Wyeth and Bionovo for endometriosis and post-menopausal symptoms) and prostatic disease (Eli Lilly). However, only a small fraction of the theoretically useful indications have been explored clinically, leaving room for exploration of many clinical indications including anxiety or depression, colon cancer, hepatic fibrosis, memory, neuroprotection, neuropathic or inflammatory pain, and wound healing, to name just a few.

Further clinical testing is needed to better understand the beneficial effects and liabilities of ER-β agonists in man. Nonetheless, the breadth and depth of ER-β agonists suggest that the preclinical pipeline is robust. Even in the absence of a clear clinical proof-of-principle, the prospects for ER-β agonist development seem promising.

GTx, Inc. has published extensive results on the preclinical pharmacology and potential therapeutic applications of aryl propionamide SARMs. Enobosarm (Fig. 4; also referred to as S-22, GTx-024 and ostarine in the literature), their lead aryl propionamide SARM, is the most advanced clinical candidate. Enobosarm demonstrated exciting data in proof-of-concept Phase II clinical trials. A phase II double blind, randomized, placebo-controlled trial in elderly men and postmenopausal women and a phase II, double blind, randomized, placebo-controlled trial in men and women suffering from cancer cachexia have been reported (Dalton et al. 2013; Dobs et al. 2013). Without a prescribed diet or exercise regimen, all subjects treated with enobosarm had a dose-dependent increase in total lean body mass (LBM), with the 3 mg/day cohort achieving an increase of more than 1 kg compared to baseline and placebo after 3 months of treatment. Treatment with enobosarm also resulted in a dose-dependent improvement in functional performance measured by a stair climb test, with the 3 mg/day cohort achieving clinically significant improvement in speed and power. The SARM also demonstrated a favorable safety profile with no serious adverse events reported and no clinically significant changes in measurements for serum PSA (prostate), sebum production (skin and hair), or serum LH (pituitary) compared to placebo. Interestingly, subjects treated with 3 mg/day of enobosarm had a decline in fasting blood glucose, reduction in insulin levels, and reduction in insulin resistance (homeostasis model assessment) as compared to baseline, suggesting that SARMs might have therapeutic potential in diabetics or people at risk for diabetes. Phase I clinical studies with enobosarm showed that it was rapidly absorbed after oral administration with a half-life of about 1 day. Phase III trials to examine the safety and efficacy of enobosarm to prevent and treat muscle wasting in patients with stage III or IV non-small cell lung cancer were completed in 2013, and will be the subject of forthcoming reports by the investigators. Preliminary reports suggest that enobosarm was very well tolerated and demonstrated promising efficacy on LBM, but produced mixed results on stair climb power depending on the type of chemotherapy that the patients received.