The End of the Age of Estrogen

By Geoffrey Cowley and Karen Springen
NEWSWEEK, July, 22-29, 2002

LUCKILY THERE WAS good news for the menopausal woman, “if only she be a clever member of her sex.” Lorand had discovered that extracts from pigs’ ovaries could “put off old age for a score of years,” or at least “mitigate its effects when it has asserted itself with all its terrors.” By the early 1940s, drugmakers were mass-producing estrogen from pregnant mares’ urine (hence the brand name Premarin). And by 1960 the august New England Journal of Medicine was recommending the stuff for “everyone with evidence of an estrogen lack”—which is to say virtually every woman over 50. Last year U.S. pharmacists filled some 45 million prescriptions for Premarin and an additional 22 million for Prempro, which consists of the same drug with a progestin chaser.

Small wonder, then, that news last week about hormone-replacement therapy caused such gasps. This wasn’t just another isolated study contradicting the last one to make headlines. Federal health officials announced last Tuesday that the jury was finally in—and that Prempro does significantly more harm than good when taken for long periods. Women had been told for decades that estrogen taken with progestin would not only ease hot flashes and insomnia but help preserve bone strength, mental acuity and, most important, heart health. There’s no question that HRT can ease the acute symptoms of menopause, and the claim about bone strength has held up to scrutiny. But after observing more than 16,000 women for roughly five years, researchers found conclusively that the hormones in Prempro raise the risk of heart attack, stroke, blood clots and breast cancer. The federally sponsored study was supposed to run for eight years, but the five-year results were so decisive that researchers cut it short and urged the participants to stop taking their pills. Dr. JoAnn Manson, a women’s health expert at Harvard Medical School, calls it “the most dramatic sea change I’ve seen in clinical medicine.”

NOT RULED OUT
She’s not alone in that sentiment. The announcement caused a near panic among the more than 13 million American women now on hormone-replacement therapy. Doctors’ offices were overwhelmed by calls, and Wyeth Pharmaceuticals, the maker of Prempro, saw its stock price plunge by 25 percent overnight. Fortunately, the study didn’t uncover huge, unimagined dangers to patients. “This is not like an urgent medical alert or the withdrawal of a drug from the market,” says Dr. Jacques Rossouw, acting director of the federal Women’s Health Initiative, which organized the trial. The findings don’t rule out Prempro as a short-term remedy for menopausal symptoms. And they don’t apply to women who take estrogen without progestin following hysterectomy. But for millions of women juggling the pros and cons of long-term HRT, the new findings offer something virtually unprecedented, which is clarity. The evidence is in, and an era is ending.

Why has it taken us so long to assess the risks of estrogen? How could a practice that does more harm than good become such a medical mainstay? “The incestuous relationship between the pharmaceutical companies and the medical establishment is part of it,” says Dr. Susan Love, a prominent women’s health specialist and longtime critic of HRT. “A lot of the information doctors get is channeled through the companies. I would bet the average doctor thought this was already proven.”

Does the medical establishment devote adequate resources to studying women's health?

No. Researchers have always focused on men and they always will

No. But women are getting much more attention than they used to

Yes. It seems as if heatlh news these days is all about women

But lazy dupes were not the only ones prescribing Prempro. Until recently, the available research suggested that long-term HRT was a boon to women’s health. When scientists observed large groups of women through their later years, the hormone users always seemed to fare best.

Those on long-term treatment suffered more than their share of breast cancer, but they experienced far fewer heart attacks (a seemingly predictable finding, since estrogen helps modulate cholesterol). Since heart disease dwarfs breast cancer as a health threat, the overall advantages seemed clear. “We thought most postmenopausal women should take HRT unless they were already at high risk of breast cancer,” says Dr. Deborah Grady, the head of a panel that wrote the American College of Physicians’ once authoritative 1992 guidelines. There was just one problem: nobody knew whether hormones made women healthy—or whether healthy women were more likely to take hormones.

ETHICAL QUESTIONS
The study cut short last week was designed specifically to tackle that question. Instead of passively monitoring women who had made their own choices about HRT, researchers at 40 clinics across the country recruited 16,608 healthy volunteers, ages 50 to 79, and randomly assigned them to take either Prempro or an identical dummy pill once a day for eight years. When the study began in 1997, critics questioned the ethics of giving people placebo when HRT had such huge ostensible benefits. But like any large clinical trial, this one had an independent monitoring board that would analyze the data as they accumulated, and halt the study early if decisive risks or benefits emerged. To the dismay of women like Janice Marabella, that’s exactly what happened. Marabella, a 62-year-old office worker in San Jose, Calif., joined the trial five years ago and felt fine until last week, when she followed the directive to stop taking her pills. It turned out she had been on hormones, not placebo, and she found herself teary and volatile within days of quitting the regimen. She now plans to resume treatment as a private patient, if only to wean herself more slowly. “The hot flashes and night sweats I could deal with,” she says. “The mood swings are something else.”
Marcia Stefanick, Ph.D., of Stanford University School of Medicine

Why should she give it up at all? What dangers has this huge clinical trial brought to light? It was clear during the first year of the study that Prempro was creating hazards as well as benefits. And as the researchers reported last week in The Journal of the American Medical Association, the ratio grew less favorable over time. Over the course of the study, women on HRT reduced their hip-fracture rate by a third (compared with women on placebo) and suffered significantly less colon cancer as well. Their chances of suffering a heart attack or pulmonary blood clot were elevated throughout the study, and their stroke risk was high from the second year on. None of those hazards caused a decisive imbalance. But five years into the study, it became clear that women on Prempro were developing invasive breast cancer at a higher rate than the control group. “That alone wouldn’t have stopped the trial,” says Marcia Stefanick of Stanford, who chairs the Women’s Health Initiative’s steering committee. “But when we combined it with the cardiovascular outcomes, it became very clear that the risks of treatment exceeded the benefit.”

That doesn’t mean HRT poses grave risks to individual women. During the study, the heart-attack rate was 29 percent higher in the treatment group than it was in the control group. The stroke rate was 41 percent higher, and the risk of invasive breast cancer was 26 percent higher. The odds sound alarming when you frame them that way, less so when you consider the rates at which these conditions actually occur. In a typical year, heart attacks struck 37 of every 10,000 Prempro users in the study, versus 30 out of 10,000 placebo users. In other words, taking Prempro boosted a woman’s annual heart-attack risk from 0.3 percent up to 0.37 percent. The annual stroke rate rose from 21 to 29 per 10,000, and the breast-cancer rate increased from 30 to 38 per 10,000.

ASSESSING RISKS
Unfortunately, even a small risk to individuals can have big consequences when applied to a large population. The study suggests that, on balance, a group of 10,000 long-term Prempro users would suffer 31 excess health crises each year (strokes, heart attacks, blood clots, breast cancers), while avoiding only 11 bone fractures and colon cancers. That’s a net increase of 20. If 100,000 people take up long-term HRT, they’ll suffer 200 of these needless events each year. A million long-term Prempro users will experience 2,000 of them annually—which means 20,000 over the course of a decade. This is no way to prevent hip fractures.