The stopping of a major hormone therapy trial in early July 2002 has sent many women to their health care providers for advice on whether to continue hormone use. The multi-center trial, one component of the Women's Health Initiative (WHI) at the National Institutes of Health, examined the long-term benefits of estrogen and progestin in 16,000 healthy postmenopausal women ages 50-79.

The trial was intended to last eight years. But after a little more than five years, the researchers found increased risks of breast cancer, coronary heart disease, stroke, and blood clots in legs and lungs in women taking estrogen plus progestin compared to women taking a placebo. Although there were fewer cases of hip fractures and colon cancer with hormone use, the risks overall outweighed the benefits. A separateWHI study looking at the effect of estrogen alone continues.

To get some advice from experts about the implications of this trial, BenchMarks interviewed several WHI principal investigators (PI) as well as scientists involved in previous hormone trials, the Postmenopausal Estrogen/Progestin Interventions (PEPI) and Heart and Estrogen/progestin Replacement Study (HERS) studies. They include:

• Elizabeth Barrett-Connor, M.D., Professor/Chief of Division of Epidemiology, University of California, San Diego, PI for PEPI and HERS
• Henry Black, M.D., Professor and Chairman, Department of Preventive Medicine, Rush- Presbyterian-St. Luke's Medical Center, Chicago, Ill., PI for WHI
• Leslie Ford, M.D., Associate Director for Clinical Research, Division of Cancer Prevention, National Cancer Institute, Bethesda, Md., WHI Steering Committee
• Curt Furberg, M.D., Ph.D., Professor of Department of Public Health Sciences, Wake Forest University School of Medicine, Winston-Salem, NC, PI for HERS
• Margery L. Gass, M.D., Professor, Department of Obstetrics and Gynecology, University of Cincinnati Medical Center, Cincinnati, Ohio, PI for WHI
• Deborah Grady, M.D., Ph.D., Professor/Director of Department of Epidemiology, University of California, San Francisco, PI for HERS
• Rebecca D. Jackson, M.D., Associate Professor, College of Medicine and Public Health, Ohio State University, Columbus, Vice Chair, WHI Steering Committee
• Dorothy Lane, M.D., School of Medicine, State University of New York, Stony Brook, PI for WHI
• JoAnn E. Manson, M.D., Chief of Preventive Medicine at Brigham and Women's Hospital, Professor of Medicine at Harvard Medical School, Boston, Mass., PI for WHI

From your experience with hormone therapy, would you consider the data collected in the WHI trial the most significant finding in evaluating the risks and benefits of hormone therapy?

Barrett-Connor: Absolutely. The participants were relatively healthy women not sick with other diseases and some of whom were less than five years postmenopausal. The power of the study is quite amazing. It's easy for clinicians to say that they have been treating women with estrogen for 30-40 years and no one has ever had any bad effects. But you can see that it takes many women to show statistically significant differences.

Black: There is no question that this a definitive study.

Ford: Yes. It was a well-done study and had the power that was needed, and the follow-up.

Furberg: Absolutely. It was conducted by a group of excellent investigators, independent of industry with large numbers of participants. I'm exceedingly positive about it. We need more studies of this kind. The sad thing is that we have made a lot of assumptions in the past based on mechanisms of actions and surrogate outcomes. If your cholesterol goes down when you take hormones, or other products, we can't assume that that is good. We need trials like WHI to determine the overall effects.

Gass: Definitely. I think it's incontestably the best data available.

Jackson: Yes, and I say that, in part, because it is a trial of inclusion. The investigators worked very hard to recruit participants that were representative of American women. There is age diversity, ethnic diversity, and geographic diversity, so that it really reflects some of the unique characteristics of American post-menopausal women. What makes the trial significant is being inclusive, being large, being long enough, and looking at multiple outcomes. And over the next year or two, as we continue to analyze a number of other factors that are important to women -- quality of life, depression, and cognition -- we'll have a much better understanding of the total impact of estrogen plus progestin, and similarly, eventually of estrogen alone.

Lane: Definitely. A randomized double-blind controlled trial is the strongest evidence that you can have.

Manson: Yes. It's the most conclusive study to date. It is the first large-scale randomized clinical trial in healthy women. It provides very important information that women can use to make informed choices about their health care. The PREMPRO combination [the hormone used in the stopped trial] is the most commonly used combination therapy, so this will apply to millions of women in the U.S. and worldwide who are making decisions about whether to begin or continue HRT.

What do you think is the most important message for women to hear about the halting of the WHI trial?

Barrett-Connor: The most important message is that the risks exceed the benefits. For every 10,000 women treated for five years, there are 100 undesirable events compared to women not treated.

Black: Women should not accept as dogma things that have not been proven.

Ford: I think the most important message goes beyond the data. Now women have good reliable clinical trial data on which to make informed decisions. Up until now, we've been making these decisions based on observational studies and on gut feeling, and we really didn't have the proper level of evidence to back it up.

Furberg: There's no evidence of preventive effect on heart disease. The other message is that these drugs are very potent. I'm also concerned about increases in cancer, breast cancer primarily.

Gass: The most important message is that the combined estrogen and progestin do not seem to have more benefit than risk as a preventive therapy.

Grady: It's not reasonable to be taking hormone therapy for anything other than treatment of very substantial menopausal symptoms.

Lane: Contrary to popular belief, the most commonly used hormone preparation does not prevent heart disease. The long-term use of hormone therapy in postmenopausal women is not recommended for prevention of chronic disease or staying healthy.

Manson: It's important for women to understand that these findings apply to a specific formulation of combined estrogen and progestin. That said, combined estrogen-progestin should not be started or continued for the express purpose of preventing heart disease. The bottom line is that after five years of use, the risks out-weighed the benefits. The study found that these hormones increased risk of both cardiovascular disease and breast cancer, which is too high a price to pay to prevent hip fractures. These are the primary messages.

However, short-term use of estrogen-progestin, four years or less, for treatment of hot flashes or other symptoms of menopause may still have a favorable benefit-risk ratio. HRT may still be a viable option short-term for the indication of treating hot flashes or symptoms of menopause and women should discuss with their physicians whether to initiate or continue hormone therapy.

Is there a particular subset of women that you think should seriously consider stopping use of hormones?

Barrett-Connor: I think every woman should seriously think about stopping, especially women who have a uterus and their ovaries so they are also making some testosterone. These women are not, as a rule, nearly as symptomatic as women in the estrogen-only arm who had their uterus and often one or both ovaries out. It does not make sense that a woman take medicine to prevent something that she may never get, and that has more risk than benefit.

Black: At this point in time, I would have trouble prescribing it for women unless they really want it. There are other ways to deal with menopausal symptoms and other ways to prevent heart disease and osteoporosis.

Ford: I think from the data it's pretty clear that anyone that's been taking them for five or more years should look very seriously at why they continue. Actually, any women taking hormones should evaluate why they're taking them.

Furberg: Those who are taking the drug combination to reduce risk of heart disease should stop. All women who are past menopause by two to four years really should reflect on why they are taking hormones.

Gass: If women have no compelling reason for taking hormones, I think each one of them should consider stopping. Benefits do not outweigh risks for preventive care. The only women who were at higher risk of breast cancer on combined estrogen and progestin were those who had already been using hormones before the study.

Grady: All women should stop except for those that are taking the hormones for treatment of symptoms of menopause.

Jackson: I really don't know all the answers for estrogen alone, but if they're taking estrogen plus progestin for the prevention -- either primary or secondary prevention -- of cardiovascular disease I think the body of the scientific evidence at this point in time suggests that there is no efficacy for preventing cardiovascular disease. Therefore, I think that that subgroup of women who chose to take it particularly for that reason should probably discontinue.

Lane: Based on WHI findings, those who have been using it for more than four years. Those who didn't have any postmenopausal symptoms who are taking as a preventive measure should stop. WHI was not set up to look at short-term use. Hot flushes are very relieved by hormones. The concern is that the negative effects occur very early on, but these effects are small, so you have to balance it.

Manson: Long-term use (four or more years) would not be recommended. There are other ways to prevent osteoporosis and fractures. In particular, women who have a first degree relative with breast cancer, such as mother or sister, should seriously avoid long-term use because they have a higher baseline risk of breast cancer.

Do you think there are enough data to say that the health effects from using estrogen alone are significantly different than the combined therapy?

Black: I can't speculate. That part of the study is continuing.

Furberg: We don't have conclusive data for estrogen alone.

Gass: No, I think it's too soon to say.

Grady: You probably need to go into a particular effect: For preventing osteoporosis, it's clearly an estrogen effect. For coronary heart disease, most of the published results have been with estrogen plus progestin. But, nevertheless, WHI reported increased risk with estrogen alone. WEST [Women's Estrogen for Stroke Trial] reported an increase of coronary events early in their trial, which was with estradiol. For coronary disease, even thought I still want to see WHI results, right now I think the weight of evidence suggests that estrogen will have the same effect. Blood clots are an estrogen effect. That leaves breast cancer. Again, I think there are many lines of evidence that estrogen is involved in breast cancer: estrogen-induced growth of breast tumors in animals; proliferation of breast cancer cell lines if treated in vitro with estrogen; endogenous hormone levels predict breast cancer risk; bone density predicts breast cancer risk; anti-estrogen reduces breast cancer risk. I think that estrogen is clearly a promoter of breast tumor growth.

Jackson: No. We don't have enough data. I think that the Women's Health Initiative will answer that question.

Lane: I don't think they are the same. Otherwise, they wouldn't have continued the estrogen trial. It's not clear about risk versus benefit for the estrogen alone. It may be that the estrogen is better.

Manson: It's too early to say for sure, but there's likely to be a difference between estrogen alone and combined therapy. There is increasing evidence that the combination increases the risk of breast cancer more than estrogen alone. And there is a theoretical reason to believe that the addition of progestin could adversely affect heart disease risk.

What are the unresolved questions in hormone therapy?

Barrett-Connor: Whether lower doses or a different route is safer or more effective. I frankly don't think such trials will happen. It's hugely expensive. Companies prescribing drugs will not be eager to show that the risk is greater than benefit.

Another question that remains is whether combined therapy is worse than estrogen alone. They didn't stop the other arm. That may mean that estrogen is less dangerous. On the other hand, the estrogen-only arm isn't as large as the combined therapy, so maybe it's too soon to see the adverse effects.

Ford: Well, there are unanswered questions that we will eventually know the answer to. First, what happens when you stop taking hormones? We will get the answers over time because we will continue to follow the women in the stopped trial. Another question is the effect of estrogen alone, and we will know the answer to that when the other half of the study is completed or stopped. The cardiac effects have been reported to be pretty much the same.

If the estrogen alone has a better risk/benefit profile, then the question is what can we give aside from MPA (medroxyprogesterone acetate) to decrease the risk of endometrial cancer. The only reason we're giving the combination is to prevent endometrial cancer; everything else about the combination seems bad. There's a progesterone ring inserted vaginally that might decrease risk of endometrial cancer and keep systemic levels low. These are the kinds of experiments that need to be done if estrogen alone proves to be better.

We also don't know about the safety of short-term hormone use. We know that hormone therapy relieves menopausal symptoms, but we don't know how long women can safely take it. And finally, we need to evaluate alternatives to hormone replacement.

Furberg: Other hormone formulations should be carefully tested. We don't know whether they will be harmful. But in my mind, I think we have enough information to say that these drugs are "under a dark cloud." You can't say the other hormone formulations are okay until proven safe. I don't think we need to do more hormone studies until we can explain the findings. For example, let's find out if some women are more susceptible to hormone therapy than others. Then I would be more interested in knowing more about other formulations.

Gass: The effect of hormones on cognitive function is the biggest unanswered question to be resolved. WHI includes a study to evaluate the effect of hormones on memory and Alzheimer's disease.

Grady: The results for estrogen alone remain to be seen. If you look at the collaborative group data, they found no difference with estrogen alone compared with the combination, and neither did the biggest of the cohort studies, the Nurses' Health Study. There have been a number of studies published -- notably there's one from NCI that showed risk is somewhat higher with progestin. The best information will come from WHI. It could be that all of the risks are the same. Or it may be that estrogens are not that bad. I'd be surprised if it's good. They have already said after two to three years with estrogen alone there's an increased risk of cardiovascular events - coronary heart disease, stroke, and blood clots. But we don't know estrogen's effect on breast cancer.

Jackson: Well, there are a lot of unresolved research questions. First, what were the mechanisms behind the adverse effects? Can we identify specific factors in women that might potentially put them at greater or lesser risk? Can we find information that will help clinicians individualize this information.

I think the second unresolved question is whether estrogen alone is different from estrogen plus progestin. And of course, the million- dollar question that everyone wants to know is whether this is true for all estrogen plus progestin formulations, or is this unique to this combination, or this progestin, or this estrogen? Can we come up with a regimen that would give us all the benefits and take away all the risks?

The colorectal cancer is fascinating data. Perhaps we can figure out the molecular mechanisms responsible for this effect that would enable us to design the potential next generation of chemoprevention for colorectal cancer.

Lane: The effect on dementia is one that women are very interested in. There is an ancillary study attached to the WHI that's going to be looking at that. Obviously, we stopped the trial so we'll only know the effect of the hormones for the five-year period of treatment. We'll also want to see if the reduced risk of colorectal cancer can be duplicated in other studies. This trial was the first time we've seen that. And, I hope there will be studies on reformulations of the hormones.

Manson: It's clear that after five years of use the risks outweigh the benefits, but we need to examine whether certain subgroups of women have more or less to gain from HRT. The study has already provided some conclusive answers, but other questions about effects on cognition and quality of life haven't yet been answered. Also, these findings may not apply to estrogen use alone for women who have had a hysterectomy. And they may not apply to other doses or formulations of hormone therapy.

###