Abstract
Keywords
One of the fastest growing populations in the world is women over 50 years of age. As the mean age for menopause is 51 years, a rapidly growing number of women will soon be reaching the age of menopause. A total of 75% of women will have menopausal symptoms, of which approximately 25% will seek treatment. Moderate-to-severe vasomotor symptoms are defined by the US FDA as greater than seven hot flashes per day, or 50 per week [1]. Hormone therapy (HT) remains the treatment of choice for menopausal women with moderate-to-severe hot flashes unless significant risk of venous thromboembolism (VTE), cardiovascular disease or breast cancer exists [2].
Menopausal women are also candidates for preventive therapy for increased risk for breast cancer, cardiovascular disease and osteoporosis-associated fractures. HT is approved for the treatment of moderate-to-severe vasomotor symptoms associated with menopause, treatment of vulvar and vaginal atrophy and prevention of osteoporosis [1]. The principal results from the Women's Health Initiative (WHI) revealed an increased risk for breast cancer, cardiovascular disease, stroke and venous thromboembolic events associated with a combination therapy of estrogen and a progestin [3], up to two-thirds of women stopped taking their HT and began to look for alternative treatments [4]. Since then, a reanalysis of the data from the WHI has revealed that women who began estrogen-only therapy within 10 years of menopause or before the age of 60 years actually achieved a cardioprotective effect as compared with those who received combination estrogen and progestin therapy [5]. Nevertheless, there is still concern for an increased risk of breast cancer, stroke, venous thromboembolic events, impaired cognition and cholelithiasis associated with HT. In regards to breast cancer, it appears that the risk is associated with the duration of one's exposure to estrogen and progestins. Results from the WHI revealed a slight yet significant increased risk for breast cancer after the use of estrogen and progestin for 5 years (in women who had been on HT for an average of 2 years prior to the study) [3].
Owing to these concerns, especially with combination estrogen and progestin therapy, investigators have begun to look at new and innovative therapies for the treatment and prevention of menopausal-related health concerns. Estrogen agonists/antagonists, also referred to as selective estrogen-receptor (ER) modulators (SERMs), are one group of compounds that are being used or evaluated for both osteoporosis prevention and treatment and breast cancer prevention in women at risk for breast cancer. Tamoxifen is already approved for use as an adjuvant in chemotherapy in women with breast cancer and as a single-agent for chemoprevention of breast cancer. In regards to chemoprevention, a population at high risk for breast cancer is selected, usually based on the Gail model. The Gail model consists of several variables with known predictive values for breast cancer – age, race, age at menses, age at first live childbirth, first-degree relative with breast cancer, number of breast biopsies and whether any were atypical. If a patient's calculated risk for breast cancer falls above a preselected cut-point, usually 1.67 or the risk at 65 years of age, preventive therapy is recommended. In addition to inhibitory effects on breast tissue proliferation, many estrogen agonists/antagonists have been shown to have estrogen agonistic effects on bone. However, not all class effects of estrogen agonists/antagonists are positive. In fact, most of these compounds cause vasomotor symptoms. Since estrogen is known to be highly effective in suppression of vasomotor symptoms, one may wonder whether the combination of estrogen and an estrogen agonists/antagonists might preserve the positive effects of the agonists/antagonists, and mitigate vasomotor symptoms as well. In this review we will discuss the molecular mechanisms involved with estrogen stimulation, review the estrogen agonist/antagonists currently available and review the data available with regards to the various effects of the estrogen agonists/antagonists, raloxifene and bazedoxifene, both as single-agent therapy and in combination with estrogen. Other estrogen agonists/antagonists are in development, each with their own potential benefits and risks. None so far meet the characteristics of the ideal compound; preventing breast cancer, heart disease, osteoporosis and cognitive decline while not stimulating the endometrium, causing vasomotor symptoms or increasing the risk for VTEs.
Estrogen
Biochemistry
Estrogens are endogenous hormones that produce numerous physiologic actions in several tissues. The principal roles of endogenous estrogens are in pubertal development in girls and preparation of the female reproductive tract for embryo implantation. However, many nonreproductive tissues, including the liver, bone, CNS, heart and the vascular endothelium, express ERs.
The most potent naturally-occurring estrogen in humans is 17 β-estradiol, followed by estrone and estriol. Each of these estrogens is an 18-carbon compound. Steroidal estrogens arise from one of two precursors, androstenedione and testosterone, through aromatization [6]. In premenopausal women, the ovaries are the principal source of circulating estrogen and the primary estrogen synthesized is estradiol. In postmenopausal women, the principal source of circulating estrogen is the stroma of adipose tissue, where estrone is synthesized from dehydroepiandrosterone secreted by the adrenals.
Mode of action
Estrogens primarily exert their effects through interactions with the two ERs, termed ERα and ERβ. Both ERs are ligand-activated transcription factors that have different tissue distributions and transcriptional regulatory effects on a wide number of target genes [7]. The ER has numerous functional domains including the ligand-binding domain, DNA-binding domain and transcription-activation domain. There are significant differences between the two receptor isoforms in the ligand-binding and transactivation domains and the receptors appear to have different biological functions and respond differently to various estrogenic compounds [8]. ERα is expressed most abundantly in the female reproductive tract, but is also expressed in the mammary gland, hypothalamus, endothelial cells and vascular smooth muscle. ERβ is expressed most abundantly in the prostate and ovaries, but is also expressed in lung, brain, bone and vasculature. Many cells express both ERs.
The concept of ligand-mediated changes in ER conformation is central to understanding the mechanism of action of estrogen agonists and antagonists. After estrogen enters the cell via passive diffusion through the cell membrane, it binds to an ER in the nucleus. Upon binding to estrogen, the ER undergoes a conformational change that causes receptor dimerization, which increases the affinity and rate of receptor binding to the DNA [9]. Both ERα and ERβ homodimers and ERα/ERβ heterodimers can be produced depending on the receptor complement in a given cell. The ER dimer binds to estrogen response elements located in the promoter region of target genes. The ER–DNA complex recruits a cascade of coactivators and other proteins to the promoter region to form a transcription apparatus. The conformational changes at the ligand-binding domain may be similar or different from estrogen, depending on specific properties of the estrogen agonist/antagonist as well as the expression of coactivators and/or corepressors of gene activity within the specific cell or tissue type.
Interactions of ERs with antagonists also promote dimerization and DNA binding. However, an antagonist produces a different conformation change in the ER compared with the agonist-occupied receptor [10]. The antagonist-induced conformation facilitates binding of corepressors and other proteins that together reduce the ability of the transcription apparatus to form initiation complexes.
Other estrogen-associated signaling systems are present in humans. Several studies have suggested that some ERs are located on the plasma membrane of cells. It appears that the same gene that encodes ERα also encodes this form of ER, but it is transported to the plasma membrane [11]. Still, other signaling systems may activate the nuclear ER by ligand-independent mechanisms, and finally, estrogen is known to initiate cell activities through nongenomic actions.
Pharmacokinetics
Estradiol, ethinyl estradiol and other estrogens are extensively bound to plasma proteins. Estradiol is primarily bound to sex hormone binding globulin (SHBG) and to a lesser degree to serum albumin. Ethinyl estradiol, on the other hand, is bound extensively to serum albumin but not SHBG. Unbound estrogens are rapidly distributed and estradiol undergoes rapid hepatic biotransformation and has a plasma half-life of only a few minutes.
Estrogens also undergo enterohepatic recirculation through sulfate and glucuronide conjugation in the liver, biliary secretion of the conjugates in the intestine, and hydrolysis in the gut followed by reabsorption. Ethinyl estradiol is cleared much more slowly than estradiol as a result of decreased hepatic metabolism and its elimination phase half-life ranges from 13 to 27 h. A variety of estrogen preparations are available for oral, transdermal, parenteral, transvaginal and topical administration. When estrogen is taken by mouth, it must first pass from the intestine into the liver where it is conjugated. Nonoral estrogen preparations avoid this ‘first-pass’ effect.
Estrogen agonists/antagonists
Estrogen agonists/antagonists are compounds with tissue-selective actions. The pharmacological goal of these compounds is to produce beneficial estrogenic actions in certain tissues (e.g., bone and brain), but antagonistic activity in tissues such as the breast and endometrium where long-term estrogen actions might be deleterious. Currently approved drugs in the USA in this class include tamoxifen citrate (e.g., Nolvadex®), raloxifene hydrochloride (Evista®) and toremifene (Fareston®), which is chemically related to and has similar actions to tamoxifen. Recently, clinical trials involving bazedoxifene, studied specifically for use in menopausal women, have been completed. However, this compound is not currently approved by the FDA. Tamoxifen and toremifene are approved for the treatment of breast cancer; tamoxifen is also approved for breast cancer prevention. Raloxifene is currently approved for the prevention and treatment of osteoporosis, and for the prevention of invasive breast cancer.
Tamoxifen is a triphenylethylene with the same stilbene nucleus as diethylstilbesterol, it is administered orally and peak plasma levels are reached in 4 to 6 h. It displays two elimination phases with half-lives of 7 to 14 h and 4 to 11 days. Tamoxifen has an antiresorptive effect in bone and stimulates proliferation of endometrial cells. It causes a two to three-fold increase in deep venous thrombosis and pulmonary embolism and a twofold increase in endometrial cancer [12]. Tamoxifen also produces hot flashes, cataracts and nausea. It has been shown to induce ovulation and to be similarly effective as compared to clomiphene citrate.
Raloxifene is a second-generation estrogen agonist/antagonist. It is a polyhydroxylated nonsteroidal compound with a benzothiophene core. It is rapidly absorbed after oral administration and has an absolute bioavailablity of approximately 2%. The drug has a half-life of approximately 28 h. Raloxifene is an estrogen agonist in bone, where it exerts an antiresorption effect. The drug does not cause endometrial stimulation and significantly reduces the risk of ER-positive breast cancer. Adverse effects include hot flashes and leg cramps.
Bazedoxifene is an investigational nonsteroidal indole-based estrogen agonist/antagonist that is being developed as a daily oral drug for the prevention and treatment of postmenopausal osteoporosis. In Phase I clinical trials, the maximum plasma concentration was 1–2 h after dosing with serum, the half-life was approximately 28 h and there was greater than 99% protein binding. Steady-state plasma concentrations are reached by day 7 to 14 [13,14]. Bazedoxifene has shown affinity for both ERα and ERβ, with slightly more affinity for ERα and serves as a competitive inhibitor of 17β-estradiol at either receptor. Bazedoxifene does not promote the proliferation of breast cells, and in the presence of 17β-estradiol-treated cells, inhibits proliferation. This inhibition is dose-dependent, thus bazedoxifene is likely antagonistic in breast tissue, similar to raloxifene [13]. Clinical studies have shown prevention of bone loss in postmenopausal women without osteoporosis and reduction in vertebral fracture risk in postmenopausal women, without stimulation of endometrium or breast. Bazedoxifene combined with conjugated estrogens (CE) is being investigated as a tissue-selective estrogen complex, pairing a selective ER modulator with estrogens. Clinical trials with bazedoxifene/CE in postmenopausal women have shown skeletal benefit with improvement in menopausal vasomotor symptoms and little or no stimulation of endometrial or breast tissue.
Pure estrogen antagonists
Pure estrogen antagonists are distinguished from the estrogen agonists/antagonists in that they produce only estrogen antagonistic effects in all tissues studied. One example of a clinically available pure estrogen antagonist is fulvestrant (Faslodex®). This compound is approved for the treatment of breast cancer in women with disease progression after tamoxifen therapy [15].
Although other pure estrogen antagonists are under investigation, currently there are no agents that have either been approved for the prevention or treatment of the signs and symptoms associated with menopause or that have published data from large clinical trials evaluating their utility in this regard.
Treatment of menopause-associated signs & symptoms
At the molecular level, one can specifically assess the affinity of a given SERM to either ERα or ERβ and the subsequent estrogen–DNA signaling cascade. However, using this information to predict the clinical response of either a given estrogen agonist/antagonist or even further, the combination of estrogen with an estrogen agonist/antagonist in a postmenopausal woman is difficult. Further complicating this problem is the fact that various estrogens are available for administration, including estradiol, ethinyl estradiol and CE. How or if these various estrogens may change the ultimate effects of combination SERM/estrogen therapy is unclear. Finally, estrogen can be administered via several different routes including oral, transdermal, transvaginal and parenteral routes. Therefore, it is incumbent upon clinicians and investigators to rely on the results of clinical trials to determine the relative benefits and risks of these treatment combinations.
In the following sections we will specifically review the available clinical information regarding raloxifene and bazedoxifene, the two estrogen agonists/antagonists that have been studied in combination with estrogen for menopause-associated signs and symptoms. We will also review the data regarding the effects of these two estrogen agonists/antagonists when administered alone. Even though tamoxifen has not been extensively studied to treat the signs and symptoms associated with menopause, it will be briefly reviewed as well.
Tamoxifen
As stated above, tamoxifen has been clinically used for its inhibition of breast cancer cell proliferation in the prevention and treatment of breast cancer. Tamoxifen, however, is stimulatory to the endometrium with reports of both adenocarcinoma and rare sarcoma of the endometrium. It has been studied in combination with estrogen primarily in Europe to relieve hot flashes. In the International Breast Cancer Intervention Study (IBIS), a randomized, double-blind, placebo-controlled, 5-year study of the effects of tamoxifen in women at high risk of developing breast cancer, women were permitted to use HT during the trial at the discretion of their doctor, limited to the duration required to control severe menopausal symptoms. Approximately 70% of women in the tamoxifen arm reported hot flushes compared with 53% of the women in the placebo arm. Severe hot flushes were more strongly related to tamoxifen. In the tamoxifen arm, 60.8% of women taking HT at entry experienced hot flushes in the first 6 months compared with 49.2% for those not taking HT (p = 0.09). Fewer on placebo who took HT at entry experienced hot flushes than for women who stopped HRT (22.9 vs 34.3%, respectively; p = 0.03). For women who first began HT in the first 6 months of the trial compared with women who did not begin HT, HT seemed to be much more effective in controlling hot flushes in months 6 to 12 in the placebo arm (47.9 vs 20.4%, respectively) than in the tamoxifen arm (51.4 vs 39.0%, respectively). Thus, HT use at entry or during the trial was not as effective in alleviating hot flushes for women in the tamoxifen arm [16].
Raloxifene alone & in combination with estrogen
Effects on bone
Raloxifene is available as a once-a-day 60-mg dose that can be administered regardless of the time of day or fasting state. It is the only estrogen agonist/antagonist that is currently approved by the FDA for the treatment and prevention of postmenopausal osteoporosis. Results from the Multiple Outcomes of Raloxifene Evaluation (MORE) trial, a 3-year randomized clinical trial that enrolled 7705 women and was funded by Lilly and Company, were used as evidence of efficacy for the use of raloxifene in the reduction of fracture risk in postmenopausal women [17]. In this trial, treatment with raloxifene demonstrated a relative risk reduction (RRR) of 55% in fractures of the vertebral body in postmenopausal women with osteoporosis who had not had a prior fracture. In women with a prior vertebral fracture, the RRR for a subsequent vertebral fracture was 30%.
Estrogen is known as an antiresorptive agent in bone and results from the WHI revealed a significant reduction in the risk for fractures. To date, there are no published data to assess the effects of raloxifene in combination with estrogen on markers of bone turnover, changes in bone mineral density (BMD) or fracture risk reduction. As both raloxifene and estrogen are known antiresorptive agents, one might speculate that combination therapy with these two compounds might produce a positive impact on fracture risk reduction. However, whether this effect is equivalent to that of raloxifene or estrogen alone, reduced, synergistic or additive is unknown.
Effects on breast tissue
In 2007, raloxifene was approved by the FDA for the prevention of invasive breast cancer. Initially, results from the Continuing Outcomes Relative to Evista (CORE) trial, a report of 5213 women treated for 8 years on raloxifene or placebo funded by Lilly and Company, demonstrated a 66% reduction in the risk of invasive breast cancer as compared with placebo [18]. Since that time, the Study of Tamoxifen and Raloxifene (STAR) trial, which enrolled 19,747 women sponsored by the National Cancer Institute, and followed them on therapy for an average of 3.9 years, demonstrated no significant difference between tamoxifen and raloxifene with regard to the rate of invasive breast cancers [19]. Furthermore, results from the Raloxifene Use for the Heart (RUTH) study, which included prevention of breast cancer as a second primary outcome, demonstrated a significant reduction in the risk for invasive breast cancer [20]. The results from the STAR and RUTH trials were submitted to the FDA for its invasive breast cancer reduction indication. In each of these two trials, the Gail model, which utilizes variables such as age at menarche, age at menopause, family history, age and results from prior breast biopsy to calculate a 5 year and life-time risk for breast cancer, was used. Patients were considered to be candidates for inclusion into the study if their 5-year risk for breast cancer was greater than or equal to 1.67%.
In the WHI, estrogen-only therapy, using conjugated equine estrogens over a 5-year period, did not significantly increase the risk for breast cancer [21]. Currently, there are no data available to assess the effects of combination raloxifene and estrogen therapy on the risk of developing breast cancer. This assessment would require either a large cohort of women to be followed for 3 or more years or a randomized clinical trial with sufficient statistical power.
Effects on vasomotor symptoms
One rather consistent ‘class effect’ of estrogen agonists/antagonists has been that they tend to cause vasomotor symptoms. In this regard, raloxifene is no exception. The effect appears to vary depending upon the clinical scenario, including the prior use of HT and the age of the patient. In the MORE trial, raloxifene produced an absolute percentage increase in vasomotor symptoms in 15% in women under 65 years of age and 6% in women over 65 years of age [17]. Therefore, it appears that older postmenopausal women tend to have less hot flashes when taking raloxifene as compared with a younger cohort. In the MORE trial, most of the study participants had not transitioned directly off HT and onto raloxifene. In a study in which 83 postmenopausal women had transitioned off either estrogen only or combination estrogen and progestin therapy, the transition to raloxifene did not change the frequency of vasomotor symptoms, hot flashes or night sweats [22]. This finding is either coincidental or related to receptor and comodulator protein dynamics. The impact that vasomotor symptoms may have on compliance on raloxifene is likely to vary depending not only on the age of the patient and their transition off HT, but on the indication for therapy. In other words, it is likely that vasomotor symptoms will have a less negative impact on compliance if the indication of the therapy is for breast cancer prevention as opposed to prevention of osteoporosis.
Estrogen therapy alone or in combination with progestins has long been accepted as the gold-standard therapy for vasomotor symptoms. One randomized trial compared the effects of taking raloxifene alone to a combination of oral estradiol 1 mg/day and raloxifene 60 mg/day over a 52-week period in postmenopausal women [22]. This trial found that although there were no significant changes in vasomotor symptoms from baseline in the raloxifene group, the combination of raloxifene and oral estradiol significantly reduced the frequency of vasomotor symptoms as compared with baseline and as compared with the raloxifene-only treatment group. Therefore, it appears that in postmenopausal women, adding oral estrogen to raloxifene therapy will reduce vasomotor symptoms. The safety of this combination is still unproven as discussed in the endometrial and venous thromboembolism sections below.
Effects on the endometrium
Treatment with estrogen therapy results in stimulation and proliferation of the endometrial tissue. This response is not lost after menopause. Therefore, one must consider the potential endometrial effects of any estrogen agonist/antagonist administered to postmenopausal women. There is a significant body of data available to assess the effects of raloxifene on the endometrial lining. Probably the best evidence comes from the CORE trial that followed over 5000 women who received either raloxifene 60 mg or placebo for 8 years [18]. The CORE trial found no increase in the incidence of uterine cancer, endometrial hyperplasia or postmenopausal bleeding in women taking raloxifene as compared with placebo. The STAR trial compared the effects of raloxifene to tamoxifen on the reduction of invasive breast cancer. The STAR trial found that the women who were randomized to raloxifene had 36% fewer uterine cancers as compared with the women treated with tamoxifen.
As it appears that raloxifene tends to be an estrogen antagonist in the uterus, it would be interesting to determine if raloxifene could block the proliferative effects of estrogen on the endometrium. In a randomized trial of 123 women treated for 52 weeks with either raloxifene alone or in combination with oral estradiol, the raloxifene-estradiol treated group demonstrated an increased endometrial thickness per transvaginal ultrasound, of 0.74 mm as compared with the raloxifene-only treated group [22]. Although this small difference may not be clinically significant, two of the women in the combination-therapy group developed endometrial hyperplasia, one with atypia. This was in comparison to no endometrial hyperplasia seen in the raloxifene-only treatment group. Therefore, currently the combination of oral estradiol and raloxifene cannot be recommended unless subsequent data reveal alternative findings regarding endometrial safety.
Another symptom commonly associated with menopause is vaginal dryness. In this regard, clinicians have successfully used locally administered estrogen to treat vaginal atrophy. One trial evaluated the influence of raloxifene on the efficacy of vaginally administered estrogen to treat vaginal atrophy [23]. In this trial, at 6 months, women who received both raloxifene and used the 17β-estradiol-releasing vaginal ring had similar efficacy with regard to resolution of their vaginal atrophy symptoms as compared to women who were treated with the vaginal ring alone in signs and symptoms of vaginal atrophy at 6 months, with no significant differences in the degree of improvement between groups. There were no signs of endometrial proliferation in either group.
Effects on venous thromboembolism, heart attack & stroke
For postmenopausal women not taking HT, the risk of having a VTE is one to two per 1000 women per year. Estrogen therapy has been associated with an increase in the risk of VTEs. This effect appears to be due to the stimulation of clotting factors produced in the liver and is not mitigated by combination progestin therapy, as demonstrated in the WHI. Based on the data from the WHI, the RR of a VTE is increased by two-to three-fold, particularly in the first 2 years of therapy. Furthermore, the increase in VTE risk with estrogen therapy was specifically seen in women who did not begin HT until they were 10–15 years beyond menopause. Treatment with raloxifene has also been shown to increase the risk for VTEs. In the CORE trial, the RR of VTE over an 8-year period of time was 2.17 as compared with placebo. In the STAR trial, treatment with raloxifene resulted in 29% less VTEs as compared with tamoxifen. Therefore, it appears that tamoxifen may have a more potent estrogenic effect on VTEs. As this effect is mediated through stimulation of the liver, nonoral routes of administration of estrogen, raloxifene or tamoxifen may reduce this risk, however, insufficient data exist to make any definitive conclusions in this regard. Furthermore, there are no data regarding the effect of combination estrogen and raloxifene therapy on VTEs. As both therapies are associated with an increased risk for VTE, one would speculate that combination therapy would result in either an equivalent or increased risk for VTE as compared with either agent alone.
The effect of estrogen on the incidence of cardiovascular events appears to vary with regard to the population being treated and whether estrogen is given alone or in combination with a progestin. Although data from the WHI initially demonstrated a small but significant increase in cardiovascular events, a reanalysis of the data reveals that this effect is primarily seen in women over 70 years of age and that estrogen therapy alone, when administered to a population of women near the age of menopause, may actually result in a cardioprotective effect [5]. With regard to raloxifene, results from the CORE trial with 8 years of raloxifene treatment revealed no increase in the risk of coronary events or strokes compared to placebo.
Effects on quality of life
No matter how efficacious a therapeutic regimen or agent might be, if it imparts a significant detriment to ones quality of life, that therapy will likely result in poor compliance and reduced effectiveness. Therefore, when one must take any pharmaceutical agent over a significant number of years, if not indefinitely, the quality of life associated with that therapy is of particular importance. One study used the Women's Health Questionnaire, a validated instrument that covers nine quality-of-life domains, to assess the effects of raloxifene, estrogen and placebo on the quality of life in healthy, asymptomatic, postmenopausal women over a 12-month period [24]. This study found that estrogen and raloxifene were similar with regard to six domains: depressed mood, somatic symptoms, memory/concentration, sexual behavior, sleep problems and perceived attractiveness. However, while mean scores for vasomotor symptoms were unchanged in the raloxifene-treated group; these symptoms significantly improved in the estrogen-treated group. Menstrual symptoms worsened in the women who received estrogen while anxiety/fears scores were improved in the women who received raloxifene. There was no clear rationale for the changes in the anxiety/fears scores demonstrated in this placebo-controlled, double-masked study.
Bazedoxifene alone & in combination with estrogen
Bazedoxifene is a third-generation SERM currently in Phase III clinical trials that has been shown to bind to both ERs with high affinity. It has been found to act as an agonist on skeletal tissue, with bone turnover reduced by 20–25% with doses of 20 or 40 mg daily. In addition, bazedoxifene has been found to be an antagonist on breast and uterine tissue, demonstrating potent dose-dependent inhibition of breast-tissue proliferation in human breast adenocarcinoma cell line (MCF 7) cells similar to raloxifene, and decreased endometrial stimulation as the dose was increased [14,25].
Effects on bone
Several Phase III trials regarding bazedoxifene are either completed or on-going and have yet to be published. One completed Phase III, 2-year, multicenter, double-blind trial evaluated the safety and efficacy of bazedoxifene monotherapy (10, 20 and 40 mg) compared with raloxifene 60 mg and placebo for the treatment of osteoporosis in 1583 women postmenopausal for at least l year [14,26]. All treatment groups had a significant sustained treatment effect compared with a significant loss of bone in those on placebo. The changes in lumbar spine for women receiving bazedoxifene were similar to those receiving raloxifene at 6 months and end of treatment. All active groups had preserved BMD in total hip at 24 months while placebo had significant declines from baseline in total hip BMD (p < 0.001). Again, in the total hip, the bazedoxifene treatment effect was similar to raloxifene at all time points. No concerns were raised related to the cardiovascular system. A low incidence of VTEs was seen, similar to raloxifene. The incidence of leg cramps was similar among all treatment groups. The number of breast cancers was low, and distributed among all the groups. No cases of endometrial hyperplasia or cancer were found in any of the bazedoxifene groups; one case of endometrial cancer was found in the placebo group. The most common adverse events included headache, infection, arthralgia, pain, hot flush and back pain. Hot flashes were reported in 19.9% on bazedoxifene 20 mg, 22.6% on bazedoxifene 40 mg, 18.3% on raloxifene 60 mg and 13.2% for placebo (p < 0.05).
A recently completed 3-year, multisite, double-blind trial presented at the American Society for Bone and Mineral Research in October 2007, evaluated the safety and efficacy of bazedoxifene 10, 20 and 40 mg in preventing osteoporosis in 7492 postmenopausal women aged 55–85 years (mean age 66.4 years) who were at least 2 years postmenopausal and had moderate-to-severe osteoporosis [27]. Inclusion included BMD less than 2.5 standard deviations (SDs) at femoral neck or lumbar spine without prevalent vertebral fractures. Women with a prevalent vertebral fracture needed a T-score of no worse than −4.0 SD at femoral neck and lumbar spine. Bazedoxifene 20 and 40 mg significantly reduced the relative risk of new vertebral fractures by 42 and 37%, respectively, compared with placebo. A post hoc subgroup analysis of women at higher fracture risk showed that bazedoxifene 20 mg significantly reduced the incidence of nonvertebral fracture by 52% relative to placebo [14]. At 36 months, the incidence of new vertebral fractures was 2.3, 2.5, 2.3 and 4.1% for bazedoxifene 20 mg, bazedoxifene 40 mg, raloxifene 60 mg and placebo groups, respectively. Post hoc subgroup analysis of 1782 women at higher risk for fracture (femoral neck T-score of −3.0 or less, at least one moderate vertebral fracture or multiple vertebral fractures) showed that bazedoxifene significantly reduced the incidence of nonvertebral fracture with fracture incidence of 3.0, 3.8, 5.9 and 6.3% in bazedoxifene 20 mg, bazedoxifene 40 mg, raloxifene 60 mg and placebo groups, respectively. When both bazedoxifene doses were combined, a risk reduction of 46% compared with placebo was found. With regard to the safety and tolerability observed in this clinical trial, a nonstatistically significant increase in the incidence of venous thromboembolic events was observed in all active treatment groups compared with the placebo group. No safety concerns related to the reproductive systems, including breast, were observed in the bazedoxifene treatment groups.
Effects on vasomotor symptoms
In the published 2-year osteoporosis prevention trial, the percentage of women who reported either worsening of hot flashes or developed new hot flushes was higher in the bazedoxifene group than in the placebo, similar to that observed in the raloxifene group. However, the number of subjects withdrawing for hot flashes was low [26].
Clinical trials are currently ongoing to evaluate the use of bazedoxifene in combination with CE for menopausal symptoms. The pairing of bazedoxifene with CE offers a hormone replacement option that does not require the need for a progestin in women with an intact uterus. The rationale is to relieve hot flashes and prevent bone loss without stimulating either the endometrium or the breast. Early results are promising.
The Selective Estrogen Menopause and Response to Therapy (SMART) 2 study, a 12-week, double-blind, randomized, controlled trial of bazedoxifene 20 mg/CE 0.625 mg and bazedoxifene 20 mg/CE 0.45 mg was presented at the North American Menopause Society in October 2007 [28]. After 12 weeks, those on bazedoxifene 20 mg/CE 0.625 mg had 80% fewer hot flashes and those on bazedoxifene 20 mg/CE 0.45 mg had 74% fewer hot flashes compared with the placebo response of 51%. Median time for 50% reduction in hot flashes for bazedoxifene and either dose of CE was 14–15 days compared with 30 days for placebo. A greater number of subjects reported at least a 75% reduction in the number of hot flushes at week 12 in the bazedoxifene 20 mg/CE 0.45 mg and the bazedoxifene 20 mg/CE 0.625 mg treatments groups (61 and 73%, respectively) compared with the placebo group (27%). A similar 5% incidence was found for uterine bleeding and breast pain within the bazedoxifene/CE groups and placebo.
Effects on the endometrium
Endometrial stimulation is a concern when ER-based therapies are evaluated. Several previous ER agonists/antagonists, such as idoxifene, droloxifene and levomeloxifene, have been halted in clinical development, partly on the basis of uterine concerns. Preclinical work involving histologic examination of an in vivo rodent model did not reveal any apparent stimulation of the endometrium. Studies in the immature rat uterine model demonstrated no significant increases in uterine wet weight, endometrial stimulation or hyperplasia or myometrial hypertrophy [25,29].
Phase III trials with bazedoxifene with and without CE are either ongoing or recently completed. The early results of clinical studies of bazedoxifene that have been presented support the endometrial safety profile observed during preclinical development.
Ronkin et al. reported on a subsequent 6-month, double-blind, randomized, active- and placebo-controlled study that evaluated the endometrial effects of bazedoxifene at doses of 2.5–40 mg/day in healthy postmenopausal women [30]. CE 0.625 mg/medroxyprogesterone acetate 2.5 mg served as the active control. Transvaginal ultrasonography and endometrial biopsy were used to assess endometrial thickness and histology at baseline and after 6 months of treatment. At doses of 2.5–20 mg/day, the mean changes in endometrial thickness with bazedoxifene were similar to those seen with placebo; at the 30- and 40-mg doses, the mean changes in endometrial thickness were significantly smaller with bazedoxifene than with placebo. These results indicated a significant inverse relationship (p $ 0.001) between the bazedoxifene dose and change in endometrial thickness. Moreover, endometrial biopsy results demonstrated no evidence of endometrial hyperplasia.
The endometrial safety of bazedoxifene has been evaluated in large, randomized, double-blind, Phase III studies. The results from SMART 1 were presented at the 2007 Endocrine Society Annual Meeting in Toronto [14,27]. This was a 2-year, double-blind trial of 3397 women aged 40–75 years, at least 1 year postmenopausal who were randomized to receive either bazedoxifene 10, 20 or 40 mg combined with CE 0.45 or 0.625 mg, raloxifene 60 mg or placebo. The primary end point was the incidence of endometrial hyperplasia. Hyperplasia was seen with the lower doses of bazedoxifene but 0.1% endometrial hyperplasia was found with bazedoxifene 40 mg at 2 years, with either 0.45 or 0.625 mg of CE. No significant differences in breast pain were reported, comparable to placebo.
In a 2-year study of healthy postmenopausal women at risk for osteoporosis-related fracture, subjects received daily therapy with bazedoxifene 10, 20 or 40 mg, raloxifene 60 mg or placebo [30–31]. Of 1583 subjects (mean age 57.6 years) who received at least one dose of study drug, 1116 were evaluable for endometrial safety analyses and 1387 were evaluable for ovarian safety analyses. Endometrial thickness with bazedoxifene 10, 20 and 40 mg remained stable over the 2-year treatment period and was not different from baseline, placebo or raloxifene 60 mg. There were no reports of endometrial hyperplasia or malignancy with bazedoxifene or raloxifene.
Archer presented data from a 3-year, Phase III study of postmenopausal women with osteoporosis (n = 7492), comparing bazedoxifene 20 or 40 mg daily, raloxifene 60 mg daily or placebo [32]. Among subjects who were evaluable for endometrial safety analyses, data on endometrial thickness, ovarian volume, and the incidence of endometrial polyps and ovarian cysts were similar among the bazedoxifene and placebo treatment groups. Endometrial hyperplasia was diagnosed in one subject (0.1%) in each treatment group, and the rates of endometrial carcinoma with bazedoxifene 20 or 40 mg, raloxifene 60 mg and placebo were 0.0, 0.1, 0.1 and 0.2%, respectively. Overall, bazedoxifene was associated with a neutral effect on the endometrium similar to that of placebo.
Effects on venous thromboembolism
Favorable changes have been found with bazedoxifene with lowering of total cholesterol, low density lipoprotein cholesterol and high density lipoprotein cholesterol relative to placebo, but the clinical relevance is unknown. Myocardial infarction and hemorrhagic or ischemic strokes were reported in less than 1% of subjects in the Miller paper. The number of venous thromboembolic events was low and appeared to be similar to the risk seen with raloxifene in younger postmenopausal women. Results for VTE risk from the combination trial of bazedoxifene and CE have not been reported.
Effects on quality of life
Phase III trials are either ongoing or not yet published. The SMART 2 trial found that bazodoxifene 20 mg/CE 0.45 mg and the bazodoxifene 20 mg/CE 0.625 mg doses were effective in reducing the number and severity of hot flushes, achieving a 74 and 80% reduction, respectively, in the number of moderate-to-severe hot flushes at week 12. More subjects reported at least a 75% reduction in the number of hot flushes at week 12 in the bazodoxifene 20 mg/CE 0.45 mg and the bazodoxifene 20 mg/CE 0.625 mg treatment groups (61 and 73%, respectively) compared with the placebo group (27%). Additional benefits noted by bazodoxifene/CE-treated subjects included improvement in the time to fall asleep, improvement in sleep adequacy, a reduction in sleep disturbance, and a positive impact on menopausal quality of life (reduction in vasomotor symptoms). Satisfaction was high in both bazodoxifene/CE groups (overall satisfaction > 73% and tolerability to side effects > 81%). Incidence of breast pain was also not increased with bazodoxifene/CE. In the SMART 3 trial, also presented at North American Menopause Soceity, October 2007, a 12-week prospective trial of bazedoxifene 20 mg with CE 0.625 and 0.45 mg was compared with placebo; vaginal dryness and sexual function were improved over placebo [33]. The percentage of superficial cells significantly increased with a significant decrease in parabasal cells compared with placebo (p < 0.05). Lubrication score, menopause quality-of-life and sexual function scores were significantly improved (p < 0.05).
Conclusion
An ideal estrogen agonist/antagonist would relieve menopausal symptoms (e.g., hot flashes and vaginal atrophy), prevent bone loss and fracture, improve lipid profiles and cardiovascular risk, have a positive benefit on cognition, improve sexual function and have either an inhibitory or neutral effect on the endometrium and breast tissue. The ideal safety profile of such a compound would be comparable to placebo, including a low risk for VTEs. That pharmaceutical agent has not yet been definitively identified. In the interim, while awaiting the perfect compound, it is important for clinicians to understand the unique benefits and risks of available estrogen agonists/antagonists and whether or not they can be safely combined with either systemic or local estrogen. Tamoxifen is approved as adjunctive therapy for breast cancer and for the prevention of breast cancer in women at high risk for breast cancer. Tamoxifen causes vasomotor symptoms and stimulates the endometrium in some women. There are no data on the use of tamoxifen in combination with estrogen specifically for treatment of menopause-associated diseases. Raloxifene has recently been approved for the prevention of invasive breast cancer in women at high risk for breast cancer, is approved for the treatment and prevention of postmenopausal osteoporosis and does not stimulate the endometrium. It appears that raloxifene has a favorable quality-of-life effect and can be safely paired with local vaginal estrogen but not systemic estrogen. Bazedoxifene is still undergoing clinical trials. Current data suggest that it mildly increased vasomotor symptoms, similar to raloxifene, prevents bone loss and does not stimulate the endometrium. Early, small, randomized, controlled trials suggest that bazedoxifene paired with CE relieves hot flashes without stimulating the endometrium. In these trials, breast pain and bleeding profiles were similar to placebo. These are no fracture prevention data or breast cancer data yet available and large multicenter trials are ongoing. Thus, there is a clear need for continued investigation in this area and room for newer estrogen agonists/antagonists with improved safety and tolerability profiles that can be administered either as single-agent therapy or combined with estrogen.
Executive summary
One of the fastest growing groups in our population is women of menopausal age and their healthcare needs are of significant societal importance.
Results of the Women's Health Initiative have changed the way both clinicians and their patients think about hormone therapy.
Estrogen agonists/antagonists produce their various effects through binding with the two estrogen receptors.
Currently, the only method available to determine the effect of combing an estrogen agonist/antagonist with estrogen on a given clinical outcome is through a randomized, controlled trial.
The only estrogen agonists/antagonists that are currently US FDA approved for use in postmenopausal women are tamoxifen, toremifene and raloxifene.
Tamoxifene and toremifene are FDA approved for the treatment of breast cancer and tamoxifen is also approved for the prevention of breast cancer.
Raloxifene is FDA approved for the prevention and treatment of postmenopausal osteoporosis and for the prevention of invasive breast cancer in postmenopausal women at high risk for breast cancer.
It appears to be safe to use raloxifene with estrogen if the estrogen is administered vaginally, but until proven otherwise, it does not appear safe to combine oral estrogen with raloxifene.
Bazedoxifene alone and in combination with oral estrogen is currently being studied for the prevention and/or treatment of a wide variety of menopause-associated problems, including vasomotor symptoms, osteoporosis and breast cancer. Current data look promising.
Future perspective
Over the next 5–10 years, estrogen agonists/antagonists will be widely used both as single agent therapy and in combination with estrogen for prevention of breast cancer, postmenopausal osteoporosis and vasomotor symptoms. Patients at high risk for VTEs will not be candidates for therapy with estrogen agonists/antagonists with or without estrogen supplementation. With the advent of these new treatment modalities and their new prevention indications, primary care clinicians will have to become more diligent in their screening for patients at high risk for both breast cancer and osteoporosis. These screening endeavors will continue to push the need for comprehensive guidelines that will define the most cost-effective method to administer therapy.
Footnotes
Dale W Stovall has received funding for clinical trials from Wyeth and Lilly and is on the speaker's bureau for Wyeth, and Lilly. JoAnn V Pinkerton has received funding for research and is on the board of advisors and speaker's bureau for Wyeth. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed. No writing assistance was utilized in the production of this manuscript.