Hormone Therapy: No Excess Mortality in 18-Year WHI Follow-up

Tara Haelle

September 12, 2017

Hormone therapy (HT) use among postmenopausal women did not increase risk for death from cardiovascular, cancer, or all causes nearly 2 decades after the Women's Health Initiative (WHI) trials, a new analysis shows.

The new data were published online today in JAMA.

The research team included many of the same researchers who conducted the original trials: the estrogen with progestin trial published in 2002 and the estrogen-only trial published in 2004.

"To me, the take-home message is very consistent with what we said before," Garnet Anderson, PhD, a coauthor on this new study and co-principal investigator of the original WHI studies, told Medscape Medical News. "When we looked at all-cause mortality at [the conclusion of the first studies], there were no differences. Now years later with more data, we're not seeing a difference."

The researchers conducted a follow-up on participants in both WHI trials of up to 18 years. Those trials included 16,608 women with a uterus who were randomly assigned to receive estrogen plus progestin (conjugated equine estrogens at 0.625 mg/day and medroxyprogesterone acetate at 2.5 mg/day) or a placebo, and 10,739 women with a history of hysterectomy who were randomly assigned to receive estrogen alone or a placebo.

The women participating, ages 50 to 79 years and 80.6% white, were recruited from 40 US clinical centers between 1993 and 1998. The estrogen-only group included 5310 women taking estrogen and 5429 taking placebo for a median 7.2 years. The estrogen+progestin group included 8506 women taking HT and 8102 receiving placebo for a median 5.6 years. After the trials ended, fewer than 4% of the women used any HT afterward.

For the updated mortality analysis, the researchers, led by JoAnn Manson, MD, DrPh, a professor of medicine at Harvard Medical School and chief of the Division of Preventive Medicine at Brigham and Women's Hospital in Boston, had data for more than 98% of the 27,347 women initially enrolled in the trials. The median follow-up time was 18 years and included 7489 deaths through December 31, 2014. Those deaths include 1088 during the intervention and 6401 since the trials ended.

All-cause mortality was similar across all groups: 27.1% in the combined HT groups and 27.6% in the combined placebo groups. No significant difference existed in the estrogen-only group (hazard ratio [HR], 0.94; 95% confidence interval [CI], 0.88 - 1.01) or in the estrogen+progestin group (HR, 1.02; 95% CI, 0.96 - 1.08) vs placebo.

Similarly, pooled cardiovascular mortality (HR, 1.00; 95% CI, 0.92 - 1.08) and total cancer mortality (HR, 1.03; 95% CI, 0.95 - 1.12) showed no statistically significant differences between groups. Absolute risk for combined cardiovascular mortality was 8.9% in the HT groups and 9.0% in the placebo groups; for cancer mortality, it was 8.2% in the HT groups and 8.0% in the placebo groups.

"It is surprising that there's no increase in cancer mortality, given all the concerns about hormones and cancer," Dr Manson told Medscape Medical News. "It appears that [HT] has a very complex relationship with cancer. It increases the risk of some cancers [such as breast cancer] and decreases the risk of other cancers [such as endometrial and uterine], and the net effect on cancer mortality is neutral."

For breast cancer–specific mortality, the estrogen+progestin group showed a trend toward increase in risk relative to placebo (HR, 1.44; 95% CI, 0.97 - 2.15; P = .07), whereas estrogen only showed a significant reduction in risk (HR, 0.55; 95% CI, 0.33 - 0.92; P = .02).

A key point often overlooked in considering the findings from the original WHI trials is that neither trial was intended to assess benefits and risks of HT for management of menopausal symptoms. Rather, both were designed to assess the benefits and risks of HT for prevention of chronic disease, including cardiovascular disease and cancers, and were stopped early when it became evident that HT did not improve cardiovascular outcomes.

"If you're going to be starting a medication for the purpose of prevention and potentially very long-term use, you would like to see evidence that there's some reduction in mortality," Dr Manson told Medscape Medical News. "However, our study does not provide support for the use of hormone therapies for prevention of cardiovascular disease or other chronic diseases."

That conclusion, however, is distinct from the decision about using HT to manage menopause symptoms.

"The problem is that the original WHI findings were misunderstood and misinterpreted as applying to women who were seeking treatment for menopausal symptoms," Dr Manson continued. "The results of the WHI were never intended to be extrapolated to women in their 40s and 50s who were seeking hormone therapy for management of distressing hot flashes and other symptoms that disrupted sleep and impaired quality of life."

These mortality findings now offer reassurance to women and their clinicians that HT for menopause symptoms management is appropriate, with a positive risk–benefit profile, said Dr Manson, who is also former president of the North American Menopause Society.

Even that finding does not necessarily provide simple answers for all women, however. Women with certain risk factors, such as a history of or high-risk profile for breast cancer or a history of blood clots, may still find that the risks of HT, even for menopausal symptoms, may outweigh the benefits for them individually. And HT still increases risk for stroke and breast cancer at the same time it decreases risk for endometrial and uterine cancer and hip fractures.

"No study is going to be able to answer that question for every person, but this study brings the best data to bear on those discussions," Dr Anderson told Medscape Medical News. "Yes, mortality is a really compelling reason to change something, but these other events are important too. Breast cancer can be very disfiguring and affect mortality, and having a stroke when you're 60 can have major quality-of-life effects even if it doesn't affect your mortality."

Still, knowing that mortality from any cause is neither increased or decreased with HT use is helpful information for women and clinicians, Dr Manson said. "The problem is, you can't look at a single outcome in isolation. I think that's the really key point. Hormone therapy has such a complex effect on health."

In an accompanying editorial, Melissa McNeil, MD, MPH, from the University of Pittsburgh in Pennsylvania, highlighted that complexity.

"Although the long-term data on total and cause-specific cumulative mortality of pooled data for hormone users vs nonusers is both compelling and reassuring, several questions remain," Dr McNeil writes. "Perhaps the most challenging question involves the issue of whether there is a difference in overall mortality by age and menopausal status at the time of initiation of hormone therapy."

Dr Manson's team calculated mortality risk stratified by the age when women began taking HT and had initially found, during the intervention, that women who began HT between ages 50 and 59 years had a 29% lower risk for death than women who began it between ages 70 and 79 years (HR, 0.61; 95% CI, 0.43 - 0.87). Long-term follow-up, however, blunted the reduction enough that it lost statistical significance (HR, 0.87; 95% CI, 0.76 - 1.00).

"This reduction in mortality...thus remains suggestive but not definitive," Dr McNeil writes. "Other questions that remain include the optimal duration of hormone therapy and if an even earlier initiation of hormone therapy, such as within 2 years of the menopausal transition, would provide additional benefits."

The study was funded by the National Heart, Lung, and Blood Institute, National Institutes of Health, and the US Department of Health and Human Services, with the drugs donated by Wyeth Ayerst. Dr LaCroix reported fees from Sermonix for serving on a scientific advisory board and consulting fees from Pfizer. Dr Chlebowski has received consulting fees or honoraria from Novartis, Genentech, Amgen, Astra-Zeneca, Novo Nordisk, and Genomic Health, as well as reviewing activity fees from Pfizer, lecture fees from Novartis, and educational activity fees from Educational Concepts Group. The authors have disclosed no relevant financial relationships.

JAMA. 2017;318:911-913, 927-938. Article

For more news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: