The Women's Health Initiative is no longer testing the most widely prescribed estrogen and progestin combination — but the study's job is far from over.

10/01/2002

In July 2002, three years short of its intended conclusion, safety concerns brought an early end to one of the most closely watched and eagerly anticipated trials in the Women's Health Initiative (WHI) — a test of the long-term benefits and risks of combined hormone replacement therapy (HRT). WHI researchers halted this part of the study, which involved 16,000 healthy women, after determining that taking estrogen plus progestin (as Prempro) did more harm than good. An increased risk of breast cancer and cardiovascular events outweighed the benefit of less colorectal cancer and fewer fractures. So researchers told participants to stop taking their study pills.

The story received extensive media coverage and sent women and their doctors scrambling to figure out what the results mean to anyone taking HRT. It also focused attention on how HRT — approved only for treating menopausal symptoms and preventing osteoporosis — had become widely accepted as a means of staving off heart disease. For WHI researchers, the trial's end was an earlier-than-expected opportunity to begin assessing specifically who benefits from taking postmenopausal hormones, who doesn't, and why.

To discuss the implications of the WHI's decision, we talked to Harvard Women's Health Watch advisory board member JoAnn Manson, M.D., Dr. P.H. Dr. Manson is one of the principal investigators with the Women's Health Initiative.


Some women are concerned that cutting short the estrogen-plus-progestin trial means we won't have all the answers we're looking for. Would you provide some perspective?     

The trial was stopped for two reasons: an increased risk of breast cancer and an overall unfavorable benefit/risk ratio. Based on previous research, we expected some increase in breast cancer risk with long-term use. But we also expected that the benefits, in terms of heart disease and fracture, would outweigh those small risks. The cardiovascular benefits never materialized.

There was simply no reason to continue. If anything, breast cancer risk would increase, the longer the duration of treatment. And it was highly unlikely that a clear cardiovascular benefit would emerge.

On the positive side, we have conclusive answers now, three years ahead of schedule, for estrogen plus progestin in the form of Prempro. Women can make use of this information much earlier than expected. Some women are just now hitting the 5-year point on hormone replacement. They can immediately use these findings to decide whether or not to continue.

Physicians were prescribing estrogen plus progestin for women who were in their 60s and 70s because that's when heart disease really increases in women. They were also putting higher-risk women on it, for example, those who had a history of heart disease or heart disease risk factors. The study addressed whether that practice was desirable and gave a conclusive answer: It is not.

Estrogen without a progestin is still being studied. There's not yet any evidence in the WHI hormone trial of an increased risk of breast cancer or an unfavorable benefit/risk ratio. So for women who have had a hysterectomy and are taking estrogen only, there's no urgency to discontinue the medication.

What's the dominant concern now among women and physicians?
    
The main concern is what alternatives women have if they're not going to use hormone therapy. Short-term use is still a viable option. Estrogen is the best treatment available for hot flashes and the disturbed sleep and impaired quality of life that many women experience in early menopause. An increase in breast cancer risk didn't show up until 4–5 years. I don't think women should be frightened about short-term use. But the study does raise questions about longer-term use of estrogen plus progestin for the prevention of chronic disease.

The WHI also found a slight increase in cardiovascular risk in the first year or two. Isn't that a worry for short-term use, too?

Yes, but young, recently menopausal women have a very low risk of heart disease. A small increase in that risk may be a price they're willing to pay for a short-term treatment that's proven to reduce hot flashes and other menopausal symptoms.

It's a different matter when you're treating people for long-term prevention of diseases that haven't occurred yet. In that case, even a small increase in the risk of any disease could be too high a price to pay.

What will you do with the data now?

We're planning to look in detail at the specific characteristics of the women who had (and didn't have) adverse events, as well as at quality-of-life issues, cognitive function, and memory. We'd like to know whether certain groups of women have more or less to gain from hormone therapy. So for heart disease, for instance, we're going to look at age groups, body weight, and other factors including some blood markers of heart disease risk, such as C-reactive protein, clotting factors, even some genetic factors — to see whether those predict who has an early increase in risk. There may even be a subgroup with a decreased risk of cardiovascular disease.

We have a very detailed study in the WHI that's looking at cognitive function and hormone therapy. The estrogen-alone trial, which continues, will shed further light on the role of estrogen in preventing memory loss and cognitive decline. We'll also learn more about estrogen's effects, longer term, on breast cancer and heart disease risk.

So this isn't over?
    
Absolutely not. In terms of understanding the results, it's just the beginning.

It's not even over for the women in the study who took combined estrogen and progestin. They've stopped the study pills, but we're continuing to follow them very closely. Their continued participation is extremely important, because we need to look at outcomes over the next three years or more. What happens when you stop after an average of 5.2 years of treatment?

Have any study participants chosen to stay on estrogen plus progestin?
       
A few have decided with their own doctors to continue and they've gotten their own prescriptions. But the large majority seems to have stopped with no plans to restart.

The women in the study took one particular dose of one preparation — Prempro. How generalizable are the results to other estrogen-progestin preparations?

These results do apply to a specific dose and formulation of estrogen plus progestin; other preparations are less well studied. But with almost any estrogen-progestin combination, you can assume that, given long-term use, there would be an increased risk of breast cancer.

Long-term estrogen use has also been strongly implicated in breast cancer, but recent evidence — from the Women's Health Initiative and other studies — suggests that the combination of estrogen and progestin may have a more adverse effect.

So it looks like the progestin may be the problem?

The addition of a progestin seems to be implicated in a greater magnitude of breast cancer risk than estrogen alone. In the estrogen-only trial, for example, we haven't seen a similar increase in risk. Also, it's possible that progestin contributes to the increased risk of cardiovascular disease.

What is the message about HRT and cardiovascular disease?

These hormones definitely should not be started or continued solely to prevent cardiovascular disease. Should you take them to prevent fractures? Well, an increased risk of both breast cancer and cardiovascular disease is a high price to pay for preventing fractures, especially when there are many other options available.

Should women switch to something else, to some other formulation or hormone preparation?

There aren't many indications for taking these hormones beyond the treatment of hot flashes and other menopausal symptoms. So why just switch to something else, especially when you don't know its benefits and risks? No other hormone preparations have been as well studied as the conjugated equine estrogens and progestin in Prempro. So the answer isn't just to switch to something else and assume that you'll be all right. I think the burden is on other preparations to be proven safe and effective.

Animal research and observational studies suggested that hormone replacement therapy might prevent heart disease. The WHI's randomized trial found otherwise. Is this kind of discrepancy unusual?

It's not unusual to have some differences, but this case is particularly surprising because the other outcomes are so similar. There was remarkable agreement between the observational studies and the randomized trials with regard to stroke, breast cancer, pulmonary embolism, colorectal cancer, and fractures — but not coronary heart disease. It's really important to understand why.

There's also the question of whether the way to derive heart benefits is to start these hormones at menopause. For instance, would a woman who started taking them at menopause and continued until age 60 or 65 have a lower risk of heart disease than a woman who didn't begin until age 60 or 65?

We have the data in the WHI to look at women according to their prior use of hormone therapy, before their enrollment in the study, so we can begin to address that question. Although they contain more estrogen than hormone replacement therapy, oral contraceptives don't increase breast cancer risk, but HRT does. Why not?

It's an entirely different situation. A postmenopausal woman's natural level of estrogen is very low; it's virtually undetectable. Taking hormones pushes that level higher than what she would normally have. A premenopausal woman, on the other hand, has relatively high estrogen and progesterone levels. Oral contraceptives, although they suppress ovulation, simply replace her natural production of these hormones with a similar amount. Her hormone level is not appreciably above what it normally would be.

Also, postmenopausal women are at a higher baseline risk of breast cancer than premenopausal women and may be more likely to have early tumors whose growth could be promoted by hormone therapy.