There has been an explosion of hormone replacement therapy research over the past few years. The task of wading through the data to determine what is best for a particular patient, however, can be daunting at best and confusing at worst, said Dr. Karen Carlson, Massachusetts General Hospital, Boston, to delegates at the Pri-Med East in Boston. Most clinicians in the United States believe that estrogen therapy is beneficial in terms of reducing the risk of coronary heart disease (CHD) in post-menopausal women. On the physiological level, estrogen has significant rapid and long-term effects on blood vessel walls. The hormone also affects lipid levels, the coagulation and fibrinolytic systems, the antioxidant systems, and production of other vasoactive molecules, such as nitric oxide and prostaglandins. "It's logical to think that with all those benefits, women with heart disease might derive some benefit from HRT," Dr. Carlson told delegates.

The Heart and Estrogen/progestin Replacement Study (HERS) was a randomized trial of daily 0.625 mg conjugated estrogen (Premarin®) plus 2.5 mg medroxyprogesterone (Provera®) that involved 2,763 women with documented CHD. Followed for four years, these women showed no significant difference between hormone replacement therapy (HRT) and placebo in any cardiovascular outcome or all-cause mortality. In the first year of treatment, Dr. Carlson said, cardiovascular events actually increased by 50 percent in the HRT group. Thereafter, there was a significant decline in the trend of events in this group.

The recent Estrogen Replacement and Atherosclerosis (ERA) study found that women followed for three years while taking estrogen or estrogen plus progesterone experienced no difference in CV event rates compared to women taking placebo.

"What is happening in these studies?" Dr. Carlson queried. "Here are some of the hypotheses that have come up."

The Postmenopausal Estrogen/Progestin Intervention (PEPI) trials, which looked at cardiovascular surrogate endpoints (e.g. lipid levels) in women taking estrogen plus progestin or placebo, found that these women tended to have more thromboembolic events on HRT if their fibrinogen levels prior to treatment were significantly lower. Pointedly, all estrogen preparations in this study caused large sustained increases in levels of C-reactive protein, a proven marker of vascular disease.

In light of this finding, HERS investigators went back to their data and examined subgroups with different lipid profiles in an attempt to determine if there were any subgroups that reacted differently. Indeed, they found a correlation with lipoprotein (a) (Lp(a)), an independent risk factor for recurrent CHD in postmenopausal women. Lp(a) is not affected by traditional lipid-lowering drugs, and a high level of the substance is associated with a high risk of coronary disease in all populations. The HERS investigators found that HRT lowered Lp(a) in women whose levels were initially high and actually achieved a reduction in the risk of CV events in this subgroup. "So perhaps Lp(a) may be a way that we can identify women that are more likely to benefit from estrogen replacement," Dr. Carlson said, adding as a caveat, "so should you go out and measure your Lp(a) level? No, but stay tuned for more research in this area."

The above trials were all focused on secondary prevention in that they looked at HRT in women with existing CHD. But what about primary prevention of CHD in women who have no signs of the disease? Animal studies indicate that estrogen may protect arteries not yet burdened by CHD. "The hormone inhibits the development of atherosclerosis in animals, but it does not seem to inhibit the progression of the disease once it is established," Dr. Carlson noted. "It seems that estrogen requires a healthy endothelium in order to confer its protective effects."

Clinical Implications

According to Dr. Carlson, results from HERS do not clarify the role of HRT in the primary prevention of CHD. "As for women who already have CHD and are taking estrogen or thinking of taking it-a good rule of thumb is don't stop if you're already taking it-don't start for the purpose of reducing risk of CHD," she advised. "There are certainly plausible reasons why women with CHD might wish to start or continue taking estrogen (e.g. treatment of osteoporosis), but we don't really know much about how to ameloriate the transient increase in risk of thrombotic activity in the first year of use."

Osteoporosis and HRT

Estrogen is a powerful treatment for osteoporosis, but it does not perform as well as full-dose alendronate sodium (Fosamax®) in terms of preventing and reducing fractures. Dr. Carlson indicated, however, that HRT is still a very good way to treat osteoporosis and increase bone density over time, especially when taken with adequate doses of calcium. Observational studies have shown the benefits of estrogen in preserving spine, and, possibly, hip bone density with 0.3 mg conjugated estrogen plus calcium (spine) and 0.3 mg esterified estrogens (Estratab®) plus calcium (spine and hip). A randomized trial of HRT in older postmenopausal women using 0.3 mg conjugated estrogen and 2.5 mg medroxyprogesterone daily plus 1000 mg calcium and vitamin D daily showed that spine bone density was increased. "So if a patient cannot tolerate the classic 0.625 mg dose of Premarin®, there is a good rationale now for treating her with a lower dose," Dr. Carlson advised.

Colorectal Cancer and Dementia

The Nurse's Health Study, as well as two metaanalyses, found a 20 to 30 percent reduction in colorectal cancer (CRC) risk associated with the current use of estrogen. The proposed mechanism is the reduction of bile acids, which are thought to be cofactors in carcinogenesis.

A potential benefit of estrogen that has not been proven, but is being intensively studied, is the effect it may have on Alzheimer's disease (AD) and other dementias. Dr. Carlson said, however, "there have been several studies looking at estrogen use in women with established AD, and they have all been disappointing. Three of them published in the past year looked at estrogen versus placebo in women with AD followed for three to 12 months. All of the trials showed no particular benefit of estrogen on cognition or functional status. So it may be an indicator that giving estrogen late in the disease may not be very helpful. We still don't know the answer as to how preventive early treatment may be."

Breast Cancer

Epidemiological evidence indicates that after five years of use, there is a modest increase in the risk of breast cancer. This risk increases with the duration of use of HRT. However, some studies show lower breast cancer mortality rates among hormone users. The Iowa Women's Health Study examined the association of HRT use with histological types of invasive breast carcinoma in a population-based sample of over 37,000 women. The tumors associated with estrogen were histologically favorable, which, Dr. Carlson indicated, may explain findings that women who get breast cancer on HRT tend to have a better prognosis overall.

According to Dr. Carlson, the results of recent studies of progestin tend to be rather confusing. In a follow-up study of a cohort of 46,000 women enrolled in the Breast Cancer Diagnosis and Detection Program, estrogen replacement therapy (ERT) alone was associated with an approximate 20 percent increase in breast cancer risk. "This is on par with what we see in other studies-about a 30 percent increased risk," Dr. Carlson noted. The finding, however, did not reach statistical significance. Further, with combined HRT the relative risk was 1.4 (40 percent increased risk), which did reach statistical significance.

In the National Cancer Institute case-control study, ERT alone and combined HRT had about the same level of risk for breast cancer. "Progestin probably does not have a good effect on the breast; it may have a negative effect on the breast, but I think it is far too early, based on these very equivocal studies, to assert that we should be using less progestin, thereby subjecting the endometrium to risk in order to protect against breast cancer," Dr. Carlson commented. "Certainly the lowest effective dose of progestin is best, but I think it is too early to drastically change our progestin prescribing practices."

Endometrial Cancer

Most observational studies have shown lower rates of endometrial cancer in women on combined HRT. However, some studies have shown increased rates, despite the use of cyclic progestins. The rates of hyperplasia with short-term use of both continuous and cyclic progestins have been similarly low (one to two percent), but the effects of continuous progestin on long-term cancer risk are unknown. A Swedish population-based case-control study found a 60 percent increased cancer risk after five or more years of combined estrogen and progestin use. Dr. Carlson commented, "I think it is too early to make the assertion that we need to reduce progestin cycling to every three months in order to protect our patients from breast cancer, at the risk of, perhaps, really increasing the risk of endometrial cancer."

Selective Estrogen Receptor Modifiers

The selective estrogen receptor modifier (SERM) raloxifene hydrochloride (Evista®) has been shown to increase bone density in postmenopausal women by two to three percent over three years, but the effect on fracture risk is unknown. The Multiple Outcomes of Raloxifene Evaluation (MORE), involving 7,705 postmenopausal women with osteoporosis, showed that raloxifene hydrochloride reduces the rate of vertebral fractures in this group by 30 percent. No effect was seen on hip fracture rates, but this may be related to short duration of follow-up, Dr. Carlson indicated. A 60 mg dosage of raloxifene hydrochloride daily has since been approved by the FDA for the treatment and prevention of osteoporosis in postmenopausal women, but its effect on CHD risk has not been established. In clinical trials, raloxifene hydrochloride had a beneficial effect on total cholesterol, LDL, Lp(a), and fibrinogen. There was no worsening of triglycerides, as is often seen with estrogen.

Androgen Replacement

A woman's androgen levels begin to decline in her early 40s, and by menopause, they are 50 percent less than pre-menopausal levels. Androgen effects are mediated not only by androgen receptors but also by aromatization of testosterone to estradiol, Dr. Carlson explained. "For that reason, it is important to keep in mind that women treated with androgens are getting some extra androgen effects on the bone, which is good-but also on the endometrium, which is not so good." There are many approaches to treating decreased libido in postmenopausal women. The first is to try an empiric trial of 0.625 mg esterified estrogens daily with 1.25 mg methyltestosterone (Estratest® HS), with progestin cycling, if appropriate. If there is no response to that, Dr. Carlson recommends checking the free testosterone level, "not because they are very predictive of sexual function but more as a baseline for working in the dark when prescribing supplemental androgens. Occasionally, I will use supplemental testosterone."

Phytoestrogens

There is fairly good evidence for the benefits of phytoestrogen on lipoproteins and bone, but the effects on the breast are less clear. Evidence from a randomized trial showed a 45 percent reduction in hot flashes with 60 grams of soy protein (76 mg phytoestrogens) per day. One cup of soy milk contains 40 mg of phytoestrogens; one gram of soy protein powder contains one to two milligrams of phytoestrogens. Dr. Carlson advised, "It is important to tell patients to read labels carefully when you recommend phytoestrogens because the content of isoflavones varies dramatically from one preparation to the next."

Individualizing HRT at Menopause

Patients are not statistics, and they should be treated individually. For example, for the healthy woman at low risk for CHD, breast cancer, and hip fracture, all treatments would provide a modest gain in life expectancy compared to no therapy. Raloxifene hydrochloride would add an estimated three months to life expectancy, HRT two months, and alendronate one month. "These advantages may seem small, but they are comparable to other proven interventions, such as mammography for breast cancer and fecal occult blood test for colorectal cancer," Dr. Carlson explained.

For the woman at high risk for CHD > 2 risk factors), HRT is preferred. For the woman at high risk for breast cancer > 2 first-degree relatives), raloxifene hydrochloride is preferred. The woman at high risk for hip fracture > 2 risk factors) should also be given raloxifene hydrochloride.

"We are still a bit in the dark about the effects of estrogen on heart disease, and it is vitally important to treat coronary risk factors, such as elevated lipids, with known effective interventions, such as statin drugs," Dr. Carlson concluded. "The risk for breast cancer is often overblown in the media, but it remains a concern to our patients. It is very important as new evidence comes along to re-evaluate the question of whether estrogen is right for an individual patient."