SERMs in practice: An idea whose time has come

Vol. 16, No. 1 / March 2008

Menopausal Medicine: For clinicians who provide care for women

SERMs in practice: An idea whose time has come

Brian M. Clark, MD

Associate Professor, Department of Obstetrics and Gynecology, Director, Division of Reproductive Endocrinology and Infertility, UMass Memorial Medical Center, University of Massachusetts Medical School, Worcester, Massachusetts

When the results of large, randomized trials cast a shadow on the use of hormone therapy (HT), particularly for secondary prevention of cardiovascular disease,^1-3 some research shifted to finding a selective estrogen receptor modulator (SERM) that might be a better alternative. Indeed, a good profile for SERMs would include a positive effect on bone and climacteric symptoms, without negative effects on the breast, endometrium, cardiovascular system, and central nervous system (CNS) function. An ideal agent would have a positive effect on all of those tissues and functions.

This article reviews the data on the effects of various SERMs on bone and cardiovascular health, uterine and breast stimulation, climacteric symptoms, and CNS function. The SERMS considered here include the 2 approved agents, raloxifene and tamoxifen, and 2 promising newer SERMs, ospemifene and bazedoxifene (TABLE 1).

The SERMs that are not discussed here—idoxifene, droloxifene, levomeloxifene, arzoxifene, acolbifene, and lasofoxifene—have less clinical impact or increased side effects (or both) and are reviewed elsewhere.⁴

TABLE 1

Clinical Use of SERMs

	Raloxifene	Tamoxifen	Ospemifene	Bazedoxifene
Approved indication	Osteoporosis	Breast cancer	None currently	None currently
Probable or possible future use	Breast cancer	Unknown	Vaginal atrophy Osteoporosis	Osteoporosis Breast cancer

Bone health

Various SERMs have positive influences on bone density and fracture rates.^5,6 The efficacy of SERMs in this regard is evaluated using markers of bone turnover as a measure of bone health and density. This implies better results for fractures, but most of the evaluations discussed here do not include data on fracture rates.

The formation of bone is evaluated with tests for alkaline phosphatase levels (and bone-specific levels), osteocalcin levels, and type I procollagen propeptide levels involving the C-terminal and N-terminal ends (called PICP and PINP, respectively); serum markers are generally used.⁶

Bone resorption is evaluated with urine and serum tests. Urine tests include hydroxyproline, deoxypyridinoline, pyridinoline, and N- and C-telopeptide (NTX and CTX) levels. Serum tests include levels of NTX, CTX, matrix metalloproteinase (MMP), carboxyterminal telopeptide of type 1 collagen (ICTP), and the vesicle enzyme tartrate-resistant acid phosphatase (TRAP).⁶

Raloxifene. As the only approved SERM for the treatment of osteoporosis in postmenopausal women, raloxifene is the best-studied SERM for this indication. It binds well to both estrogen receptors (ERα and ERβ), does not significantly stimulate the endometrium, and appears to be active against breast cancer.^4-8 Climacteric symptoms are not alleviated by the use of raloxifene, and the risk of thrombosis is similar to the risk with estrogen.^4-8

In several studies, use of raloxifene significantly increased bone mineral density (BMD).^9-12 Overall, these studies showed that vertebral fractures are reduced by approximately 40%. No studies have documented decreased fracture rates at nonvertebral sites. However, structural assessment of bone at the hip using biopsy has shown that BMD and bone strength do increase with raloxifene use, indicating that there may be a decrease in fractures that has not yet been documented.¹²

Additionally, raloxifene may decrease bone turnover, compared with placebo, in postmenopausal women.^11,13,14 Several authors have reported that the bone turnover markers alkaline phosphatase, osteocalcin, pyridinoline, NTX, CTX, and PINP are all altered with raloxifene use.

Raloxifene adds to the bone-sparing effects of calcium and vitamin D₃ and monofluorophosphate therapy.^15,16 Raloxifene also protects against bone loss associated with exercise-induced weight loss.¹⁷ The effects of raloxifene appear to be modulated through estrogen receptors that influence osteoblast function.¹⁸

Taken together, these data support significant improvement in BMD, with reduced vertebral fracture risk, and modifications in bone turnover favoring increased BMD (including nonvertebral sites) with raloxifene use in osteoporotic postmenopausal women.

Tamoxifen. Tamoxifen has been approved for the treatment of breast cancer; therefore, fewer data exist regarding the effect of tamoxifen on bone health. However, increased BMD and decreased fractures have been documented in breast cancer patients treated with tamoxifen.^19,20

Ospemifene. Ospemifene is under investigation and has not obtained approval for any indication, although it shows promise for treatment of vaginal atrophy.²¹ Its bone turnover effects are similar to those of raloxifene and, similar to raloxifene, its mechanism seems to be through an estrogen-mediated influence on osteoblastic functions.^14,18 Further testing is needed to analyze ospemifene’s influence on bone health and to see if the effect is significant enough to consider it for treatment for osteoporosis.

Bazedoxifene. Bazedoxifene is also under investigation and shows promise for treatment of osteoporosis. In studies, bone turnover decreased by approximately 25%, with decreases noted in CTX, NTX, PINP, PICP, and alkaline phosphatase levels.^21-23 This decrease in bone turnover is similar to but slightly less than that seen with raloxifene. Increases in bone density have been shown in animal studies but must be fully evaluated in humans.²²

Cardiovascular health

Several end points are used to evaluate the cardiovascular effects of HT or SERMs, including lipid parameters, clinical events, inflammatory or other markers of cardiovascular disease, and vascular functional changes.^1-4,9 Actual cardiovascular events are relatively rare among newly postmenopausal women, so using these clinical end points poses challenges in designing studies of adequate power within logistic and financial constraints.

Recent reports show that HT has less cardiovascular impact when initiated close to menopause, unlike the early reported results.^3,24 This suggests that a temporal relationship is of great importance. Finally, the effects of different dosages, compounds, and routes of administration are also factors to consider.

Given this large set of parameters, it is difficult to assess and effectively summarize the impact of a treatment on cardiovascular function in postmenopausal women. Despite these challenges, we will try to unravel the data available for these SERMs. In all cases, further evaluation is needed.

Raloxifene. In various studies, use of raloxifene has consistently shown to be associated with a modest decrease in total cholesterol, low-density lipoprotein cholesterol (LDL-C), lipoprotein(a) (Lp[a]), fibrinogen, and homocysteine levels, as well as a significant increase in high-density lipoprotein cholesterol (HDL-C) levels.^9,13,25,26 These results suggest a decrease in cardiovascular risk, but importantly, no studies have shown a decrease in clinical end points. Raloxifene has not been shown to have a negative influence on echocardiographic evaluations or on plasminogen activator inhibitor-1 or C-reactive protein (CRP) levels.^9,13,25-27

My own evaluation showed that raloxifene has a positive effect on vasodilation similar to that of estrogen through a direct endothelial mechanism (unpublished data). Additionally, raloxifene is associated with an increased risk of venous thrombotic disease and stroke, similar to that seen with estrogen.²⁸

Overall, these data suggest a probable neutral influence of raloxifene on cardiovascular health, but further study is needed to support this statement more concretely.

Tamoxifen. Observations that women in breast cancer trials who were treated with tamoxifen had decreased rates of myocardial infarction (MI) fueled studies of its effect on various cardiovascular end points.^29,30

Further evaluation has shown that tamoxifen is associated with a decrease in total cholesterol, LDL-C, CRP, homocysteine, Lp(a), and fibrinogen levels. No consistent pattern has been seen with HDL-C and triglyceride levels.^31,32 Taken together, these data support an improved or neutral cardiovascular profile and decreased rate of MI.

Decreased intimal thickness may also play a positive role in cardiovascular outcomes. Tamoxifen use has been shown to decrease intimal thickness as well as increase flow-mediated dilation.³³ My own data suggest that tamoxifen has a minimal effect directly on the vascular endothelium, similar to that of placebo.³⁴ Unfortunately, a consistent increase in thrombotic disease has also been documented, with a risk similar to that seen with estrogens.^35,36

Similar to raloxifene, these data support a positive or neutral effect overall on cardiovascular risk for patients taking this compound, but further evaluation is needed.

Ospemifene. Cardiovascular evaluations of ospemifene have been limited to nonclinical end points at this time. Decreases have been documented in total cholesterol, LDL-C, and fibrinogen levels, and increases in HDL-C and triglycerides have been noted. Neutral effects on homocysteine levels were also noted.^21,37-39

These data are too preliminary to draw conclusions about ospemifene’s risk profile for postmenopausal women, but they do suggest that further evaluation may indeed yield a positive cardiovascular effect, and they do not suggest a negative effect.

Bazedoxifene. A very preliminary study of bazedoxifene noted a decrease in both LDL-C and fibrinogen levels.⁴⁰ Although these initial data suggest a positive influence, the current data set is far too small to reliably assess the impact that this SERM may have on cardiovascular function and risk.

Uterine and breast stimulation

Raloxifene. Raloxifene minimally stimulates the endometrial lining and has a negative influence on breast tissue growth. Trials of its use against breast cancer have shown positive results thus far.^4,7,8

Tamoxifen. Tamoxifen is one of the cornerstone treatments for breast cancer and is approved for this indication. However, tamoxifen stimulates the uterine lining, causing an increased incidence of uterine cancer and polyps, which has somewhat limited the drug’s utility.^{4,7,29,30,35,36}

Ospemifene. Few data exist in the literature on breast stimulation with ospemifene. Minimal endometrial stimulation has been documented, similar to that of raloxifene.^37,39

Bazedoxifene. Evaluation in MCF-7 breast cancer cell lines suggests a negative effect of bazedoxifene on the growth of breast cancer tissue; human trials have not been performed. In both animal and human modulastudies, the influence of bazedoxifene on the endometrium seems negligible and it may cause beneficial thinning of the lining.²²

Climacteric symptoms

Raloxifene and tamoxifene may actually worsen hot flushing.^4,41 Mixed data have been reported on ospemifene’s effect on vasomotor symptoms, with neutral to beneficial effects seen.³⁹ Little has been reported on climacteric symptoms with bazedoxifene use.

CNS influence

Raloxifene has not been evaluated for clinical outcomes such as cognitive function, but an increase in stroke has been reported with the use of this compound.²⁸ Little has been reported on the incidence of CNS problems with the use of tamoxifen, ospemifene, or bazedoxifene.

Summary

The variety of responses is a hallmark of SERMs (TABLE 2). The differential tissue responses show that their mechanism of action is variable, relating to the interaction with the receptor and subsequent modifications of transcription ability within different tissues.

Currently, the perfect SERM does not exist, but continued work in this area will undoubtedly yield improved compounds. With a promising future, the perfect SERM may someday be available to optimize postmenopausal hormone therapy.

TABLE 2

Summary of SERM Effects

	Raloxifene	Tamoxifen	Ospemifene
Bone health	• Increased BMD • Improved bone turnover • Decreased vertebral fractures	In breast cancer patients^19,20: • Increased BMD • Decreased fractures	• Improved bone turnover^14,18
CV health	• Improved lipid profile • Decreased fibrinogen, antithrombin III • No clinical outcomes data^9,13,25-28	• Improved lipid profile • Possibly decreased myocardial infarction rate^30-36	• Improved markers of cardiovascular health?^21,37-39
Other	• Poor treatment of climacteric symptoms (may worsen) • Increased thromboembolic events • Increased rates of stroke • Little effect on endometrium^4,7,8	• Increased rates of endometrial cancer • Increased incidence of polyps • Increased thromboembolic events^{4,7,29,30,35,36}	• Few data • Little effect on endometrium • May improve climacteric symptoms^21,37-39
BMD, bone mineral density; CV, cardiovascular; SERMs, selective estrogen receptor modulators.