By Charles Bankhead, Staff Writer, MedPage Today
A claim that thousands of women have died because they avoided estrogen therapy represents an oversimplification of the issues and reinforces the risks of subgroup analyses, according to investigators in the Women's Health Initiative (WHI).
Hormone therapy, whether hormone replacement therapy (HRT) or unopposed estrogen, confers risks and benefits that vary according to a woman's age, time since menopause, and numerous other factors. The WHI showed that postmenopausal hormone has variable effects and emphasized the need for individualized therapy, said JoAnn E. Manson, MD, DrPH, of Brigham and Women's Hospital in Boston.
"I don't think the WHI estrogen-alone trial can be interpreted as simply leading to many excess deaths among younger women," Manson told MedPage Today. "In fact, it was saving many lives among the older age groups, and across all age groups it was helping clinicians identify the most appropriate candidates for therapy and minimizing some of the adverse events."
"The WHI results were never intended to be generalized to younger women who have early surgical menopause and early removal of the ovaries prior to the onset of menopause," she added.
The analysis, published last week, was not so much a study as it was a math exercise, said WHI investigator Andrea LaCroix, MD, of the Fred Hutchinson Cancer Research Center in Seattle. She was the lead author of a 2011 report that provided the genesis for the new analysis of hysterectomized women 50 to 59.
"They [authors of the new analysis] took one statistic from the JAMA paper and put it into a formula and calculated the expected number deaths under a set of assumptions that may or may not be reasonable," said LaCroix. "That's what they published. Is that true?"
"These data from the WHI were first reported in 2004, and there were declines in estrogen use in 2002," she added. "My first question is, where is the real evidence that death rates have increased in women 50 to 59 in the U.S. in those communities where the greatest decreases in hormone therapy have occurred?"
The reaction centered on a study suggesting that as many as 91,000 women 50 to 59 might have died prematurely as a consequence of avoiding unopposed estrogen therapy after hysterectomy. At a minimum, the authors estimated, 18,000 women died prematurely over the past 10 years.
The findings and conclusions came from extrapolation of an excess mortality observed among younger hysterectomized women in WHI randomized to placebo. The observation was first reported in 2004 and confirmed in the 2011 report that LaCroix co-authored.
"As a result (of the methodology), the findings are unreliable projections, which are, at best, hypothesis-generating and should not inform clinical practice," WHI investigator Rowan Chlebowski, MD, of the University of California Los Angeles, said via email.
"We don't highlight or make any clinical recommendations based on speculative projections based on subgroup analyses of breast cancer adjuvant or advanced disease trials. The same caution should be given in this case, where headline prominence in media reports seems to suggest more importance than such a speculative hypothesis can support."
Authors of last week's report took the media to task for oversimplification of WHI findings, often with the complicity of researchers and clinicians. Chlebowski did the same.
"I recognize the general interest in menopausal hormone therapy findings, but it seems the media should weigh carefully the strength of any evidence presented before endorsing an author's conclusions," he said.
Beyond cautions about the validity of subgroup analyses, Manson said "estrogen therapy should not be used routinely in women with hysterectomy after age 50. It's also important to understand that the WHI was studying women who were postmenopausal.
"The findings do not necessarily apply to women who have hysterectomy and surgical menopause at a younger age, 40 or 45, where it's clear, much clearer, that the benefits of estrogen therapy will outweigh the risk for nearly all women who undergo early surgical menopause."
In the 2004 and 2011 reports, WHI investigators emphasized the need for counseling about the appropriateness of postmenopausal hormonal treatment, whether estrogen alone or estrogen plus progestin. In the same vein, LaCroix took issue with any implied promises about the benefits of estrogen therapy.
"There is no pill available that will reduce the risk of death in postmenopausal women," she said. "Anyone who implies otherwise is overreaching."