Estrogen replacement is considered the first-line approach for the prevention and treatment of multiple conditions affecting women's health. It has been widely recommended for the prevention and treatment of osteoporosis, reduction of the risk of mortality from cardiovascular disease, improvement of lipid profiles, amelioration of the signs and symptoms of menopause and possible protection against the development of Alzheimer's disease.^1–3

Although estrogen replacement therapy is generally considered effective, its long-term use is not without significant risks. A recent reanalysis of 90 percent of worldwide observational data on the relationship between hormone replacement therapy and breast cancer showed that the risk of this cancer may be increased in women who are taking estrogen.⁴ The risk of endometrial cancer is also increased in women with an intact uterus who take unopposed estrogen for long periods.^1,5 In addition, estrogen replacement therapy has been associated with an increased risk of thromboembolic events.^6,7 Other significant side effects of estrogen replacement are vaginal bleeding and breast swelling and tenderness.⁸

Because of the potential side effects and risks, women often decline to be treated with estrogens, and physicians may be reluctant to prescribe them. Furthermore, the overall rate of compliance among women on estrogen therapy is relatively poor.⁸

In an effort to broaden the treatment options available to postmenopausal women, research efforts have been directed at the development of compounds that maintain the vasomotor, skeletal and cardiovascular benefits of estrogen replacement therapy but have little to no significant adverse effect on reproductive organs and the clotting processes. The search for this “ideal” compound has led to the development of a class of drugs termed “selective estrogen receptor modulators” (SERMs). These agents may provide the beneficial effects of estrogen replacement therapy without some of its bothersome or potentially serious side effects.

Selective Estrogen Receptor Modulators

Raloxifene (Evista) has the ability to bind to and activate the estrogen receptor while exhibiting tissue-specific effects distinct from estradiol.⁹ As a result, raloxifene is the first of a benzothiophene series of antiestrogens to be labeled a SERM. (Droloxifine, idoxifene and toremifene are similar SERM agents, but they are still considered experimental.) Raloxifene was specifically developed to maintain beneficial estrogenic activity on bone and lipids and antiestrogenic activity on endometrial and breast tissue. In December 1997, the U.S. Food and Drug Administration (FDA) labeled raloxifene for the prevention of osteoporosis.

Although the exact mechanism of action of raloxifene and other similar compounds has not yet been determined, it has been hypothesized that these agents work by inducing conformational changes in the estrogen receptor, resulting in differential expression of specific estrogen-regulated genes in different tissues.¹⁰ Activation of the estrogen receptor by these compounds may involve multiple molecular pathways that may result in gene expression of ligand-, tissue- and/or gene-specific receptors.¹¹

Because SERMs are capable of inducing specific changes in the estrogen receptor, it is not surprising that they may mediate specific pharmacologic activity through their unique agonist or antagonist properties. For example, the agonistic properties of raloxifene on bone tissue were recently demonstrated by the specific activation of the human transforming growth factor-β3 gene, which is an important regulator of bone remodeling.¹²

Pharmacokinetics

The pharmacokinetic parameters of raloxifene have been evaluated in conventional clinical pharmacology studies and selected clinical trials involving 1,570 postmenopausal women.¹¹ Raloxifene is rapidly absorbed after oral administration, and its absolute bioavailability is 2 percent. Although the drug can be administered without regard to foods, consumption of high-fat meals may increase its systemic bioavailability; however, this effect is not considered to result in clinically significant changes in systemic exposure.

Raloxifene undergoes extensive systemic biotransformation, but it does not appear to be metabolized by the cytochrome P450 pathway. Clinically significant interactions are unlikely to occur with drugs typically eliminated by this route. Raloxifene has a plasma elimination half-life of approximately 27 hours. This prolonged elimination half-life has been attributed to the drug's reversible systemic metabolism and significant enterohepatic cycling.¹¹

Although raloxifene and its metabolites are highly bound to plasma proteins (more than 95 percent), in vitro studies have yielded no evidence that raloxifene interacts with the binding of other highly protein-bound drugs. Nonetheless, caution should be exercised when raloxifene is combined with highly protein-bound drugs. Single-dose studies have shown that the coadministration of raloxifene and warfarin (Coumadin) may cause a 10 percent decrease in prothrombin times. Thus, patients receiving these two drugs should have their prothrombin times checked frequently until an adequate level of anticoagulation has been achieved.¹¹

With the exception of significant binding that can occur with cholestyramine (Questran), in which raloxifene's absorption can be decreased by 60 percent, little to no information is available on important drug interactions with this SERM. Dosage adjustments are not required in patients with renal insufficiency because raloxifene is eliminated primarily in the feces and only negligible amounts appear in the urine. The safety and efficacy of raloxifene have not been adequately evaluated in patients with hepatic failure.

Effects on Bone

The FDA labeling of raloxifene for the prevention of osteoporosis in postmenopausal women was originally based on the results of three large randomized, placebo-controlled trials conducted over a 24-month period.¹¹ Only the results of the European trial have been published.¹³ In this trial, 601 women were randomly assigned to receive raloxifene, in a dosage of 30, 60 or 150 mg per day, or placebo. Eligible candidates had to be between 45 and 60 years of age and within two to eight years of menopause. All participants also received 400 to 600 mg per day of elemental calcium supplementation.

At 24 months, the women who were taking raloxifene had significant increases in bone mineral density (BMD) in the lumbar spine, total hip, femoral neck and total body, compared with the women who received placebo (P < 0.03 for all dosages of raloxifene). For example, in the women who received 60 mg of raloxifene per day, the mean increase in BMD from baseline was 1.6 percent for the lumbar spine, 1.6 percent for the total hip, 1.2 percent for the femoral neck and 1.4 percent for the total body.¹³

The two unpublished trials, which included a total of 1,263 women (619 who had undergone hysterectomy and 544 who were post-menopausal) reported findings relating to raloxifene's effect on BMD that were consistent with those of the published trial.¹¹

Historical comparison suggests that raloxifene has less favorable effects on BMD than hormone replacement therapy and alendronate (Fosamax).^14–16 For example, the FDA-labeled dosage of 60 mg per day of raloxifene may actually be half as effective as estrogen replacement therapy in increasing lumbar spine BMD in postmenopausal women.

The short-term effects of raloxifene on bone turnover markers have also been evaluated. In an eight-week study, 251 healthy post-menopausal women were randomly assigned to receive 200 or 600 mg per day of raloxifene, 0.625 mg per day of conjugated estrogen or placebo.¹⁷ Bone turnover markers, such as serum alkaline phosphatase level, serum osteocalcin level and urinary calcium excretion, were found to be similarly decreased in both the raloxifene and the estrogen groups. These findings seem to be supported by results of the recently published European trial,¹³ in which raloxifene decreased bone turnover markers to levels similar to those found in premenopausal women.

Bone remodeling studies have compared raloxifene (60 mg per day) with hormone replacement therapy (0.625 mg of conjugated estrogen per day and 5 mg of medroxyprogesterone for the first two weeks of each month) given over a 31-week period to postmenopausal women. Each therapy produced a significant positive shift in the calcium balance at four weeks and again at 31 weeks.¹⁸

Effects on Lipids

Raloxifene also appears to have a favorable effect on lipid parameters in postmenopausal women. In the published European trial,¹³ treatment with raloxifene in a dosage of 30, 60 or 150 mg per day resulted in significant decreases in the serum concentrations of total and low-density lipoprotein (LDL) cholesterol over a 24-month period (P < 0.05 versus placebo). These decreases were evident during the first three months of therapy and were maintained thereafter. Notably, none of the treatment groups showed any changes in serum concentrations of high-density lipoprotein (HDL) cholesterol and triglycerides.

The effect of raloxifene on serum lipid levels was also assessed in 390 healthy post-menopausal women.¹⁹ In this study, raloxifene (60 and 120 mg per day) was compared with hormone replacement therapy (0.625 mg per day of conjugated estrogen and 2.5 mg per day of medroxyprogesterone given continuously) and placebo. Assessments were made at baseline, three months and six months. Over the six-month study period, both dosages of raloxifene lowered serum LDL cholesterol levels by about 12 percent compared with placebo (P < 0.001). This finding was similar to the 14 percent reduction that occurred with continuous hormone replacement therapy.¹³

As with the previous trial,¹³ serum levels of HDL cholesterol remained unaffected by the use of raloxifene but increased by 10 percent with the use of hormone replacement therapy.¹⁹ Triglyceride levels were also unaffected by the use of raloxifene but increased by 20 percent with the use of hormone replacement therapy (P < 0.001).¹⁹

The HDL cholesterol level is considered a strong inverse predictor of cardiovascular disease in women.²⁰ Therefore, the absence of an increase in serum HDL cholesterol levels raises concern that raloxifene may not be as effective as estrogen replacement in preventing cardiovascular disease. Although the findings of animal studies are difficult to generalize to humans, recent animal data have also raised concerns that raloxifene may not prevent the progression of coronary artery disease.²¹ Because no long-term trials have been conducted, it is impossible to determine whether the small lipid effects produced by raloxifene correlate with a smaller degree of cardioprotective activity compared with hormone replacement therapy.

Effects on the Uterus

One particular advantage of raloxifene over hormone replacement therapy is its apparent lack of proliferative effects on endometrial tissue. Data from both animal and human studies demonstrated that raloxifene has minimal effects on the uterus and causes no significant changes in the histologic appearance of the endometrium. Two six-month studies involving a total of 969 postmenopausal women showed that endometrial thickness did not differ between women receiving raloxifene (30 to 150 mg per day) and those receiving placebo.²²

Another short-term study of healthy post-menopausal women found that raloxifene (200 to 600 mg per day given over eight weeks) did not induce endometrial proliferation as measured by endometrial biopsies.²³ By comparison, 77 percent of the women who received unopposed estrogen (0.625 mg per day of conjugated estrogen) had moderate to marked estrogenic proliferation of endometrial tissue. Women who received conjugated estrogen were also noted to have a much higher incidence of vaginitis than those who received raloxifene or placebo.

A recent Canadian trial in 136 healthy post-menopausal women compared the stimulatory effects on the uterus of raloxifene (150 mg per day) and continuous hormone replacement therapy (0.625 mg per day of conjugated estrogen with 2.5 mg per day of medroxyprogesterone).²⁴ After a period of 12 months, the women who received estrogen replacement therapy experienced significant changes in endometrial thickness and uterine volume. In contrast, the women who were treated with raloxifene exhibited no changes in either parameter.²⁵

Additional short-term trials appear to support the view that raloxifene may not produce endometrial stimulation.^13,17 Although these short-term observations seem promising, it is unclear whether raloxifene provides long-term protection against endometrial cancer.

Adverse Effects

Because of the recent labeling and limited widespread use of raloxifene, little information is available on its side effects. Even though most data on adverse events are derived from a limited number of clinical trials, it appears that raloxifene is well tolerated. Most reported adverse effects have been mild. A pooled analysis of data showed that raloxifene use had to be discontinued in 11.4 percent of women (n = 581) compared with 12.2 percent of the women who received placebo (n = 584).¹¹

Commonly reported adverse events experienced by women treated with raloxifene have included hot flushes (24.6 percent versus 18.3 percent for placebo) and leg cramps (5.9 percent versus 1.9 percent for placebo).¹¹ A comparison of the adverse events reported with raloxifene (60 mg per day) and estrogen (conjugated estrogen and cyclic or continuous medroxyprogesterone) showed that hot flushes occurred more frequently in women treated with raloxifene than in those treated with estrogen (28.7 percent for raloxifene, 5.9 percent for cyclic hormone replacement therapy and 3.1 percent for continuous combined hormone replacement therapy). However, compared with either cyclical or continuous hormone replacement therapy, raloxifene did not cause significant uterine bleeding or breast tenderness.¹¹ Other trials appear to support the high incidence of hot flushes and the lack of significant breast tenderness produced by raloxifene.^13,17

Thromboembolic events (relative risk not stated) such as deep venous thrombosis, pulmonary embolism and retinal vein thrombosis have been reported with raloxifene therapy and are more likely to occur during the first four months of treatment.¹¹ Because of this potential problem, raloxifene should not be used in women with active venous thromboembolic disease or a history of such disease. Raloxifene should also be avoided in pregnant women and women with a history of hypersensitivity reactions to the drug.¹¹

Data from a short-term (three-year) trial showed that raloxifene may not increase the risk of breast cancer.²⁶ Additional trials are currently being conducted to examine the apparent protective effects of raloxifene against breast cancer.

Therapeutic Options

Several agents for the prevention and treatment of osteoporosis are now available. The challenge is to select an appropriate regimen to maximize therapeutic benefits while reducing potential risks. The advantages and disadvantages of currently available agents are summarized in Table 1. Clinical benefits and potential limitations of selected agents are compared in Table 2, and the monthly costs of various regimens for osteoporosis are presented in Table 3.

Estrogen replacement therapy remains the treatment of choice for most postmenopausal women because of its proven benefits in preventing bone loss and counteracting the development of heart disease. It should be emphasized, however, that most “proven benefits” of this treatment are based on data derived from observational studies rather than randomized blinded trials and that the outcomes are based on intermediate end points such as lipid changes rather than actual mortality rates.²⁷

Although estrogen replacement therapy offers many therapeutic advantages, the benefits are offset by potential undesirable side effects, including the increased risk of endometrial cancer and, possibly, breast cancer. Because of these side effects and the risk of poor compliance, committing postmenopausal women to long-term estrogen replacement therapy may prove difficult.

Raloxifene may be considered an effective alternative for the prevention of postmenopausal bone loss in women who are unwilling or unable to take estrogen. The drug appears to have a better side effect profile but less effect on bone loss than traditional estrogen replacement therapy. Furthermore, its long-term benefits in cardiovascular disease prevention and its apparently protective effects against breast cancer are still unknown. Additional factors of concern may be the tendency of raloxifene to exacerbate hot flushes and to increase the risk of thromboembolic events.

In some women, tolerance may be a more important issue than efficacy in determining the need for long-term therapy for osteoporosis prevention. In these situations, alendronate should be considered a useful alternative to both raloxifene and estrogen replacement therapy. Although alendronate is effective in the prevention and treatment of osteoporosis, it lacks the important cardiovascular benefits often attributed to estrogen replacement therapy. Its use may also be limited by a complicated administration regimen and the significant associated risk of gastroesophageal side effects.

Despite the availability of a new therapeutic option for women with osteoporosis, important questions about raloxifene remain to be answered. The long-term clinical benefits of raloxifene therapy in preventing bone loss must be firmly established. In addition to demonstrating safety and efficacy, it will be important to establish the real benefits of raloxifene relative to estrogen replacement therapy, especially with regard to cardiovascular, central nervous system and other diseases. Although long-term studies of raloxifene are lacking, postmenopausal women now have an additional therapeutic option to consider when deciding on therapy for osteoporosis.