Sci. Aging Knowl. Environ., 24 September 2003
Weathering the HRT Storm
Showers of negative clinical studies have clouded the issue of hormone replacement therapy, but a panel of researchers says the story isn't over
Tabitha M. Powledgehttp://sageke.sciencemag.org/cgi/content/full/sageke;2003/38/nf18
Since the 1980s, a woman of 50 or so could expect to get a prescription for sex hormones to replace the ones her aging body was shutting down. Hormone replacement therapy (HRT) is the only proven way to get immediate relief from the sleeplessness, discomfort, and embarrassment of hot flashes and other symptoms of menopause. But it has also been an article of medical faith, at first based mostly on data from studies of estrogen only, that hormones stave off burdens of aging: the dowager's hump and broken hips of osteoporosis, the epidemic of heart disease that hits women after menopause, and even faltering memory and the horror of Alzheimer's disease (see "More Than a Hot Flash").
No more. A torrent of negative clinical studies published in the last 18 months or so has smashed those comfortable beliefs. Not only does HRT fail to postpone many diseases of aging, the studies reported, it apparently hastens some of them. Interpretations of the work--or, many unhappy scientists argue, irrational misinterpretations--have swept away funding and stranded research on sex hormones and menopause, to say nothing of leaving millions of women and their physicians at sea about the wisdom of hormone replacement.
Last week the North American Menopause Society (NAMS), an organization composed mostly of researchers who study menopause, issued a report that attempts to sort through the wreckage and salvage anything valuable. The report describes what is known about hormone replacement, which turns out to be much less than news reports might have led a person to think. NAMS also presents scientists and policymakers with a long to-do list of tasks that can help answer the multitude of remaining questions about HRT. But in the present climate, researchers might have a tough time tackling some of the projects that NAMS is urging, says NAMS Executive Director Wulf H. Utian, who shepherded the report. Funding sources are shying away from such studies, and many potential subjects are skittish about enrolling.
A Storm's Aftermath
The tempest over HRT hit in July 2002, when the U.S. National Institutes of Health (NIH) terminated the Women's Health Initiative (WHI) 2 years early. A randomized double-blind clinical trial with more than 16,000 postmenopausal study subjects, the WHI revealed small increases in invasive breast cancer, coronary heart disease, stroke, and pulmonary embolism in women taking a drug called Prempro. This HRT pill contains synthetic progestin (an artificial version of the natural hormone progesterone) and a complicated cocktail of so-called conjugated equine estrogens (CEEs), horse compounds treated to make them more physiologically available. It was little consolation that the WHI also reported small benefits, among them a somewhat reduced incidence of hip fracture and colon cancer. Since termination of the study, more negative outcomes have surfaced. In May of this year, the Journal of the American Medical Association published studies on brain function in a subset of about 4500 WHI subjects. The papers documented small increases in dementia and cognitive decline in the Prempro group. And last month The Lancet published the U.K.'s Million Women observational study, which confirmed the link between HRT and breast cancer; the subjects had used a number of HRTs.
The NAMS report is the most detailed and authoritative attempt to disentangle what the data tell us from the many things they don't. More than a dozen experts, some of them investigators in the WHI and some who are dismayed by the way the HRT findings have been interpreted, collaborated to produce it. NAMS panel members agree on some issues, but areas of consensus tend to be narrow and full of qualifications. The experts concurred, for example, that the primary indication for HRT is the treatment of menopausal symptoms such as hot flashes and that hormone use "should be limited to the shortest duration consistent with treatment goals." This symptomatic treatment also got the U.S. government's seal of approval in a 9 September statement from the Food and Drug Administration. But the NAMS panel couldn't reach consensus on how long hormones should be prescribed for symptom relief, or on a definition of short-term and long-term hormone therapy. The NAMS list of tasks includes investigations of different kinds of hormone therapy, such as long-term studies of various hormone preparations other than Prempro, different dosages (especially lower doses), different regimens, and different routes of administration. In addition, although the issue is not specifically addressed in the report, researchers note that studies to date have focused on older women, and much remains unknown about the effects of HRT on the younger--perimenopausal--population.
Waning Hormones, Missing Information
These recommendations are bound to appeal to researchers who are alarmed by what they regard as an unwarranted backlash against hormones. "The problems that we see now are related to the use of combination therapy," says Bruce McEwen, a neuroscientist who heads the Laboratory of Neuroendocrinology at Rockefeller University in New York City. Researchers added progestogens--which include both natural progesterone and its synthetic analogs--to HRT to block uncontrolled growth of uterine cells and cancer after early studies of estrogen-only HRT suggested an increased risk of these disorders. But Provera--the synthetic progestin in Prempro--acts differently from progesterone, according to a number of studies, he says: "It undoes, counteracts, beneficial effects of estrogens, for example on the growth of neurons."
During the WHI design phase, some researchers urged studies of different hormone preparations, especially estrogen alone. "There was a failure in the way decisions were made about how to apply good basic endocrinology to a very important problem and translate it into practice," McEwen says. The WHI selected Prempro because it was what women were using, says women's health researcher Marcia Stefanick, a WHI principal investigator at Stanford University, despite a paucity of data on that combination. "There were 6 million women on that regimen, largely based on data from estrogen only," she says. Physician Marjorie Shuer, who heads the Mood and Menopause Clinic in San Diego, California, offers an explanation, shared by many hormone researchers, for why physicians prescribed Prempro: "That's what doctors were prescribing because Wyeth-Ayerst [Prempro's manufacturer] is incredibly aggressive in their marketing."
Data about the effects of estrogen alone--or at least CEEs alone--will become available sometime between now and 2005, the original WHI endpoint. 2005 is still the probable endpoint for the only part of the WHI that wasn't shut down in 2002: a study of CEEs in women without a uterus. The fact that the estrogen study continues might be a sign that the single hormone is more benign than the combination. However, Stefanick cautions, "you can't assume benefit or discard risks because a study hasn't stopped."
NAMS's Utian is a tad more optimistic. Observational data on estrogen's effects, he says, hint that estrogen alone will turn out to be less risky than Prempro. The NAMS panel wants researchers to investigate whether the frequency at which progestin is given or the type of progestin that is prescribed creates problems, Utian notes.
Researchers are also complaining about the age of the WHI study subjects. About a third of the women were in their 50s at enrollment, but the average age was 63, well past menopause. Subjects in the WHI memory study averaged approximately 70 years of age, says neuroendocrinologist Barbara Sherwin of McGill University in Montreal, Canada. "And of course the WHI found that hormones didn't protect, which is no surprise to those of us who are working in this area." Animal studies have shown that females must replace estrogen soon after ovaries are removed in order to maintain mental performance, and some human studies suggest the same thing, Sherwin says. "The work that needs to be done is to look at effects of hormones on women who are much younger, who are perimenopausal, who are actually the women who end up getting treated in the real world."
Frederick Naftolin, a physician who heads the Reproductive Neuroscience Unit at Yale University, also emphasizes the age of the subjects when considering the cardiovascular data. Necrotic plaques, a sign that cardiovascular disease is developing, start turning up in women's blood vessels in their 50s, he says, which means that to forestall a progressive ailment such as heart disease, women must begin treatment early.
HRT's Uncertain Forecast
The NAMS panel shares the belief that to understand the long-term effects of HRT on perimenopausal women, studies must focus on that population. But the barriers to doing clinical hormone studies on perimenopausal women are currently immense, due to the negative press. Sherwin, for example, reports difficulty recruiting women for her study of estrogen and memory, and she says that some subjects have dropped out, although they have only a 50% chance of being on the hormone and even then, for only 3 months. By the time NIH shut down the WHI, the subjects had been taking Prempro for years. "There's no reason to believe, certainly not from the WHI study, that anything's going to happen to them in 3 months, and yet they feel they've heard from the media that you don't do this anymore," she says.
Neuroscientist Neil MacLusky, scientific director of the Center for Reproductive Science at Columbia University College of Physicians and Surgeons in New York City, has a much grimmer tale to tell than that of recruitment problems. MacLusky and his colleague Mary Tierney, head of geriatrics research at Sunnybrook Hospital in Toronto, Canada, wanted to know whether hormones could prevent or stall development of Alzheimer's disease in women at risk for the disease. The researchers elected to use low-dose estradiol , the form of estrogen made in women's ovaries, along with a small amount of progestin to prevent endometrial cancer. "If you're going to look at potential effects of hormone replacement, you should start from what a woman has in her own body, not what the pharmaceutical companies are trying to push," MacLusky says. Although the study was eventually funded through the Canadian Institutes of Health Research (CIHR), NIH rejected the protocol because it departed from the WHI Prempro regimen, MacLusky says.
Recruitment involved years of screening thousands of women to find around 200 suitable subjects. About 100 women had already begun treatment when, last May, the WHI cognitive studies appeared. "As soon as they were published, CIHR abruptly terminated the trial on grounds that it was unethical to continue," says MacLusky. "To extrapolate from [Prempro] to all forms of HRT ... was really not justified." He and Tierney appealed, MacLusky says, even convening an independent data safety monitoring board to break the code and analyze the data collected so far for any indication of adverse effects. The safety board reported that it saw no medical reason to discontinue the trial, he says, but "the damage had been done." The CIHR money ran out this month, but the study will continue temporarily with emergency funds from a private donor, MacLusky says. The researchers are seeking foundation grants to finish it. "Without at least a little bit more money to complete the trial, unfortunately we're not going to get any conclusive data out from it."
MacLusky is understandably gloomy about prospects for much-needed HRT research in general. NAMS's Utian says that studies on cells and animals--and even some limited human studies--probably will proceed. In particular, funding agencies are likely to approve experiments that probe mechanisms of hormone action and of still-mysterious menopause symptoms such as the hot flash, he says. But Utian isn't hopeful about studies examining, for instance, whether lower hormone doses will reduce the risk of stroke or curb the slight increase in breast cancer that previous work suggests. "A study that will look at enough individuals of the appropriate age and profile that would match what is normally seen in clinical practice, I just don't think will ever get done," he says. "So I don't know that we'll ever get that final answer."
Despite the clinical trial data showing small risk increases in several disorders, especially with regular use of combination HRT, a great deal of preclinical data from both animal and human studies suggest that sex hormones--especially natural ones--have beneficial effects. "This really should be telling us that we need to do more work, not to reject hormone replacement, but just to improve it, to make it so it works more like natural hormones," MacLusky says. Reaching that goal could mean sunnier skies for millions of women.
September 24, 2003
Tabitha M. Powledge is a science writer and editor based near Washington, D.C. Thanks to Hurricane Isabel, she needs electricity replacement therapy.
Suggested ReadingBack to Top
Citation: T. M. Powledge, Weathering the HRT Storm. Sci. SAGE KE 2003 (38), nf18 (2003).
Science of Aging Knowledge Environment. ISSN 1539-6150