ReviewUse of SERMs for treatment in postmenopausal women
Introduction
Selective estrogen receptor modulators (SERMs) are synthetic non-steroidal agents that have varying estrogen agonist and antagonist activities in different tissues, most likely due to the receptor conformation structural changes associated with that SERMs binding and the subsequent effect on transcription. Most estrogenic responses in tissues are mediated by estrogen receptors (ERs), ERα or ERβ, which involve ligand-dependent transcription factors. SERMs are ER ligands that in some tissues act like estrogens, but block estrogen action in others through competitive inhibition of estrogen binding to ERs. Each SERM has the ability to induce distinct structural changes in the receptor that influence the receptors ability to interact with coactivators (CoA) or corepressors (CoR), which are involved in the regulation of target gene transcription. The resulting biologic action can vary according to the specific type of ER, cofactors, responses and ligands leading to tissue specific agonist and antagonist activity. Different ligands can induce distinct receptor conformations in ERα and ERβ, leading to structures that are different than that seen with an unliganded receptor [1], [2], [3].
The history and development of SERMs was recently reviewed by Jordan and his co-authors [4]. We will discuss the mechanism of action briefly in a subsequent section. SERM are compounds with tissue-selective actions. The pharmacologic aim in development of SERMs has been to elicit specific positive effects on certain targeted tissues such as bone, heart and brain with neutral or antagonist effects on other tissues such as the breast and endometrium, where long-term estrogen stimulation may be harmful. SERMs show mixed agonist and antagonist activities depending on target tissue as the shape of the ligands that bind to the ERs, ERα or ERβ, causes the complex to elicit an estrogenic or anti-estrogenic signal [4], thus allowing SERMS to have variable tissue activity [4], [5]. Preclinical and clinical testing suggest that each SERM has unique tissue-selective effects that can allow for individualization of care depending on specific estrogen agonist or antagonist effects desired. Currently SERMs are approved by FDA to prevent and treat osteoporosis (raloxifene), to prevent or treat breast cancer (tamoxifen, toremifene, raloxifene) and for the relief of dyspareunia associated with postmenopausal vaginal atrophy (ospemifene). The first tissue-selective estrogen receptor complex (TSEC) approved by the FDA pairs the SERM, bazedoxifene, with conjugated equine estrogens, for treatment of menopausal hot flashes and prevention of bone loss.
Many SERMs have been discontinued after making it to phase 3 clinical testing and we will review in a later section the key clinical trial findings that led to either discontinuation (droloxifene, idoxifene, ormeloxifene, arzoxifene, and levormeloxifene) or development being placed on hold (lasofoxifene). The ideal SERM would provide agonist effects on the bone to prevent bone loss and the brain to treat hot flashes, while providing neutral or antagonist effects on the breast and endometrium to reduce cancer risks. To date, no SERM has been discovered that provides the tissue specific actions desired to be ideal. Different SERMs provide different tissue specific actions allowing for individualization depending on the medical needs of the postmenopausal women.
Clomiphene citrate is used for premenopausal women with infertility [6], but will not be discussed as there is no recommendation for its use in postmenopausal women. This review of literature will present randomized clinical trial (RCT) information about SERMs that have clinical utility in postmenopausal women's health. Relevant English-language articles published between 1980 and 2013 were identified through PubMed database (search string: “[Selective Estrogen Receptor Modulator OR SERM] AND [bone, cardiovascular or lipid, neuroprotection, breast, vulvar, vaginal, vasomotor symptoms, endometrium, or venous thrombosis], by using article reference lists, and by searches for individual SERMs and tissue specific effects.
It is not clearly established how different SERMs act as an estrogen agonist in some tissues and estrogen antagonist in other tissues, but the mechanisms involved will be briefly reviewed. There is a change in the conformation of the ER following binding by a SERM ligand to an ER [4], [5]. However, the mechanistic model [4] described by Jordan et al. suggests that the external shape of the receptor–ligand complex is the catalyst for determining the expression of the specific estrogen agonist or antagonist, and that this programing is a dynamic and variable process affected by interactions of the receptor complex with cofactors determining specific gene regulation at the cellular level [4], [5]. About 200 CoA have been found to be involved in promoting gene expression [7]. Through opposing actions, a balance is found between CoA and CoR that shape the responses to nuclear receptor ligands [7]. This balance allows individual target tissues to express different responses varying from a full estrogenic response to an antiestrogenic response based on shape of ligand and tissue modulation [8]. Variations in the expression of CoA among different individuals are believed to be associated with phenotypic differences among humans and lead to variability in tissue response to therapies [7].
An estrogen or a SERM can bind to ER, ERα or ERβ or both, which results in a specific 3-D conformation of the ligand–receptor complex, which then allows various cofactors to bind to the receptor [8], [9], [10]. The agonist or antagonist effect of a specific SERM in a given target tissue is based on many factors including the differential expression of ERα and ERβ, conformation of the ER upon ligand binding (when the SERM and nuclear receptor interact), coregulators (CoA and CoR) and the tissue specific cellular environment [8], [9], [10]. Target tissue-specific effects of SERMs are thought to be related to different effects on gene transcription with distinct genes regulated through either ERα or ERβ when challenged with different SERMs [4], [11], [12]. For example, bone has both types of ERs while uterus has predominately ERα [11].
McDonnell et al. [1] present their model of ER pharmacology in which the conformation of the ER is influenced by the nature of the ligand to which it is bound. This action enables a differential interaction of the receptor with functionally distinct coregulators. In this theory, pure ER agonists will facilitate the interaction of the receptor with CoA. At the cellular level, the response to an ER agonist is determined by the presence of various coregulators in different target tissues. In contrast, antagonists allow ER receptors to interact with CoR. Thus SERMs act differentially in different tissues by permitting bound ERs to interact with different subsets of CoA and CoR. The ER modulator is influenced by its binding affinity for ERα and ERβ with differing expressions of estrogenic or antiestrogenic activity in target cells [13], [14].
Differential gene regulation with different SERMs ultimately contributes to the varying tissue-specific activities with most SERMs having estrogen agonist activity in bone and antagonist activity in the breast, while activity in the endometrium and the vagina differentiates many of the SERMs [4].
An ideal SERM (Table 1) would provide desirable tissue-selective estrogenic agonist activity in bone, brain, cardiovascular system, vagina, urogenital system and skin with ER neutral or estrogenic antagonistic activity in breast, endometrium, pelvic floor, and on risk of venous thrombosis [4], [9]. Based on pharmacology of SERMs identified to date, it seems unlikely that a single molecule will be developed that will provide the beneficial effects of classical estrogens without increasing the risk of breast and uterine cancers, blood clots or stroke. Instead, research is examining how to optimize a particular biological response (i.e. antiresorptive activity in bone) and screen preclinically against negative effects in other target tissues (i.e. uterine stimulation) to evaluate tissue-selective SERMs with improved efficiency in ER-modulation at target tissues (Table 2).
A new class of drug combination is in development, the TSECs, which combine a SERM and an estrogen which, when paired together, will hopefully provide a clinically effective and safe therapy for use in postmenopausal women. Different conformational changes in the ER receptors may occur with different estrogen/SERM combinations which may lead to either estrogen agonist or antagonist effects in targeted tissues [15], [16]. Only one TSEC has made it to advanced clinical development with a safe and effective profile and this is the SERM, bazedoxifene, combined with conjugated equine estrogens. Clinical testing of this combination has shown relief of menopausal hot flashes and vaginal atrophy and prevention of bone loss without stimulation of the breast or uterus. Bazedoxifene is significantly antiproliferative on the uterus, which obviates the need for concomitant progestogen use [17]. Both components of the TSEC compete for the same ligand binding on the ER [18].
SERMs have been approved for the management and treatment of varying postmenopausal women's health concerns, including prevention and treatment of estrogen sensitive breast cancer, the prevention and treatment of osteoporosis, and recently, for menopausal vaginal changes and associated dyspareunia.
Most SERMs in clinical testing have been shown to have antagonist or neutral effects in the breast [19], [20], [21], [22], [23], [24], [25], [26], [27] while having positive agonist effects on bone [19], [28], [29], [30], [31]. In general, SERMs provide protection against menopausal bone loss, although with less robust effects on bone mineral density (BMD) compared with standard-dose estrogen. The estrogen agonist activity found with most SERMs allows them to prevent bone density loss and sometimes reduce fracture risk [31], [32]. Fracture protection appears to be greater and out of proportion to the SERM's modest effect on preventing or improving bone density [33], suggesting benefits on bone microarchitecture beyond bone density improvement. SERMs have not been found to increase or decrease cardiovascular events, overall mortality or cardiovascular mortality despite their positive effect on lowering total and low-density lipoprotein (LDL)-cholesterol concentrations [34], [35]. No clear cardioprotection or cardiovascular risk has been found [31], [33], [34], [36], [37], [38], [39], [40], beyond an increase in fatal stroke for raloxifene, although not an increase in overall stroke incidence [37], [38]. Until large RCTs with cardiovascular primary endpoints are performed, potential cardioprotective benefits of SERMs will remain unclear [41]. SERMs as a class appear to have an increased risk of venous thromboembolism (VTE) similar to estrogens [19], [21], [22], [23], [24], [25], [42], [43], [44], [45].
Central nervous system effects (CNS) are variable and not well defined. There is some evidence of decreased effect on proinflammatory markers in women at neurodegenerative risk [46], [47], [48]. SERMs act with tissue and cellular selectivity within the brain, which may lead to differential modulating of microglia, astroglia cells, and motor neurons [48]. Research with SERMs has evaluated possible neuroprotection and even reduction of neural damage in situations such as neural trauma, brain inflammation, cognitive impairment, neurodegenerative disorders and mood disorders [48]. SERMs may promote the interaction of ERs with the neuroprotective signaling of growth factors [47]. In vitro studies [46], [47], [48] have shown that tamoxifen, raloxifene, ospemifene, and bazedoxifene reduced mRNA levels of proinflammatory molecules (such as interleukin-6 [IL-6] and interferon-gamma-inducible protein-10) released by astrocytes, with raloxifene and ospemifene being more effective than tamoxifen and bazedoxifene in reducing protein levels in lipopolysaccharide-treated cultures. Thus, it is possible that SERMs will be found to have estrogen agonist properties on proinflammatory molecules produced by astrocytes, which could counteract brain inflammation in neurodegenerative disease [46], [47], [48]. Effects of SERMs on cognition are not clear.
A key differentiator amongst SERMs has been their variable estrogen effect at the level of the endometrium, varying from an agonist effect seen with tamoxifen, to neutral with raloxifene, to antagonist with bazedoxefine [17], [24], [42], [43], [44], [49]. Concern about endometrial safety continues to be an important consideration in the development of SERMs [17]. Determining whether a SERM can be paired with a systemic or vaginal estrogen depends on how strong the estrogen antagonist effect is on the uterus and potentially the particular type of estrogen (estradiol, estrone or conjugated estrogens) [15]. SERMs with stronger estrogen antagonist effects on the endometrium would be expected to have less risk of stimulating estrogen sensitive endometrial cancer if used either alone or in combination with vaginal or systemic estrogens. This theory led to the development of combining bazedoxefine, a strong estrogen antagonist in the uterus, with conjugated equine estrogens, a combination of estrogens that are known to have both estrogen agonist and SERM-like activity [4].
In general, SERMs have inconsistent effects on the vagina [15], [40], [44], [49], [50], but a few SERMs which specifically target the pathophysiology underlying vulvovaginal atrophy (VVA) may provide an alternative to vaginal or systemic estrogen therapy for symptomatic postmenopausal women. Tamoxifen, raloxifene and bazedoxefine have no direct positive effects on the vagina [15]. Ospemifene and lasofoxifene have demonstrated improvements in the physiological changes seen with VVA [15], [19], [39], [40]. Additional RCT data are needed to establish the effect of various SERMs on pelvic organ prolapse [15].
SERMs as a class have shown an estrogen antagonist effect with a mild increase in hot flashes, generally not significant enough to discontinue therapy [16], [51], [52]. One SERM in development, RAD-1901, has shown improvement in hot flashes in a morphine-dependent ovariectomized rat model, but clinical trials will be needed to see if it is effective for hot flash relief in humans [53]. The combination of a SERM and estrogen might relieve hot flashes, but the SERM must be significantly antagonistic in the endometrium to prevent proliferation. A new development is the pairing of a SERM with an estrogen to form a TSEC to provide effective therapy with hopefully an improved safety profile over the combination of estrogen and a progestogen. The first TSEC, bazedoxifene combined with conjugated equine estrogens, has been approved for treatment of menopausal hot flashes and prevention of osteoporosis. To date, only bazedoxifene combined with conjugated estrogens appears to have an adequate safety profile; raloxifene with oral estradiol was associated with a reduction in hot flashes but an increase in endometrial hyperplasia [54].
Section snippets
SERMs in clinical use for postmenopausal women
SERMs in clinical use are categorized based on their chemical structures (Fig. 1). Tamoxifen, toremifene, and ospemifene (a toremifene derivative) are triphenylethylenes; raloxifene is a benzothiophene; lasofoxifene is a naphthalenol; and bazedoxifene is an indole derivative [28], [55].
Triphenylethylene SERMs, considered first generation SERMs, include tamoxifen and toremifene which are approved to prevent and treat breast cancer. Benzothiophene SERMs are considered second-generation SERMs. In
Discontinued SERMs
In the past 20 years, many SERMS in development have failed to make it through clinical trials due to lack of efficacy or significant adverse events including gynecological adverse events.
Tissue selective estrogen complex (TSEC): bazedoxifene and conjugated estrogens
The goal of a TSEC is to blend the desired tissue-selective properties of the SERM (neutral/antagonist activity in breast and endometrium) with the desired activities of the estrogen(s) (reduced vasomotor symptoms and improved lipid profiles, BMD, and vaginal atrophy without significant adverse events. In preclinical trials, bazedoxifene had a neutral effect on mammary gland and endometrial tissue, and an antagonist effect when combined with conjugated equine estrogens. The pairing of
Future perspective
Postmenopausal women are at risk for medical problems related to changes in hormone levels after menopause. Many women chose not to take or continue hormone therapy, or are not good candidates for hormone therapy due to breast cancer risk or elevated risk for thrombosis. These women may be good candidates for SERMs, which provide estrogen agonist and antagonistic actions in targeted tissues such as breast, bone, heart, endometrium, brain, vagina, and genitourinary system. More research is
Conclusion
Clinical differences in ER agonist and antagonist responses differentiate SERMs. To date, SERMs are approved to provide prevention and treatment of breast cancer (tamoxifen and toremifene), prevention and treatment of osteoporosis (raloxifene) and dyspareunia related to vaginal atrophy (ospemifene). The first TSEC (bazedoxifene and conjugated equine estrogens) is approved for the treatment of moderate-to-severe vasomotor symptoms associated with menopause and the prevention of postmenopausal
Disclosure statement
In the past 12 months, Dr. Pinkerton has served as a consultant (fees to the University of Virginia) for Pfizer Inc., Noven Pharmaceuticals, DepoMed, and Shionogi; received grants/research support (fees to the University of Virginia) from DepoMed, Bionova, and Endoceutics; and received travel funds from Pfizer Inc., Noven Pharmaceuticals, Shionogi, and DepoMed. Dr. Thomas has no disclosures.
Acknowledgements
The authors wish to thank the reviewer, Bhagu R. Bhavnani, PhD, for his thoughtful suggestions and editing on our behalf and to Frank Z. Stanczyk, PhD for his support. Final editorial assistance and proofing were provided by Kathleen Ohleth, PhD, at Precise Publications, funded by the authors.
References (109)
- et al.
The structural basis of estrogen receptor/coactivator recognition and the antagonism of this interaction by tamoxifen
Cell
(1998) - et al.
The expanding cosmos of nuclear receptor coactivators
Cell
(2006) - et al.
The molecular mechanisms underlying the pharmacological actions of estrogens, SERMs and oxysterols: implications for the treatment and prevention of osteoporosis
Bone
(2013) - et al.
In vitro and in vivo biologic effects of Ospemifene (FC-1271a) in breast cancer
J. Steroid Biochem. Mol. Biol.
(2001) - et al.
SERMs: progress and future perspectives
Maturitas
(2010) - et al.
Prevention and treatment of osteoporosis in women with breast cancer
Mayo Clin. Proc.
(2000) - et al.
Selective estrogen receptor modulators (SERMs) for the brain: current status and remaining challenges for developing NeuroSERMs
Brain Res. Brain Res. Rev.
(2005) - et al.
Neuroprotective actions of selective estrogen receptor modulators
Psychoneuroendocrinology
(2009) - et al.
Characterization of hot flashes reported by healthy postmenopausal women receiving raloxifene or placebo during osteoporosis prevention trials
Maturitas
(2000) - et al.
Overview of the main outcomes in breast-cancer prevention trials
Lancet
(2003)