1: THE PREGNANT MARE'S LESSON
In a former British colony, most healers believed the conventional wisdom that a distillation of fluids extracted from the urine of horses, if dried to a powder and fed to aging women, could act as a general tonic, preserve youth, and ward off a variety of diseases. The preparation became enormously popular throughout the culture, and was used widely by older women in all strata of society. Many years later modern scientific studies revealed that long-term ingestion of the horse-urine extract was useless for most of its intended purposes, and that it caused tumors, blood clots, heart disease, and perhaps brain damage.
The former colony is the United States; the time is now; the drug is the family of hormone replacement products that include Prempro and Premarin (manufactured from pregnant mares' urine, hence its name). For decades, estrogen replacement in postmenopausal women was widely believed to have "cardio-protective" properties; other papers in respected medical journals reported that the drugs could treat depression and incontinence, as well as prevent Alzheimer's disease. The first large, well-conducted, controlled clinical trial of this treatment in women was not published until 1998; it found that estrogen replacement actually increased the rate of heart attacks in the patients studied. Another clinical trial published in 2002 presented further evidence that these products increased the risk of heart disease, stroke, and cancer. Further reports a year later found that rather than preventing Alzheimer's disease, the drugs appeared to double the risk of becoming senile. The studies resulted in a reduction, but not an end, to the long-term use of these products.
For decades, these were among the most widely prescribed drugs in the nation. How did we get such an important question so wrong for so long?
Despite the deer-in-the-headlights astonishment with which the nation greeted the 2002 report that hormone replacement caused more harm than good, signs of trouble had been emerging for several years. The estrogen debacle was a case study waiting to happen, and its story can tell us much, both good and bad, about how the health care system evaluates and deploys medications. The fabric of modern medical care is woven of the belief by doctors and patients that the prescription drugs we use have been exhaustively studied and shown to work. The spectacular downfall of estrogen replacement therapy drew attention to the question of just how we determine that a drug actually "works," and why the system broke down in this very high profile case. For years, I've studied how we know what we know about drug benefits and risks, an inquiry I think of as "pharmaco-epistemology." The hormone replacement story is a perfect case study in this domain; it has much to teach us about how fragile our knowledge base can be concerning a drug's ultimate effects.
The word "estrogen" itself comes from the Latin oestrus and the Greek oistros, which mean "gadfly" and, by extension, "frenzy." The roots were chosen by early physiologists to depict the sexual arousal the hormones can cause in animals. They are also related to the semantic lineage of the word "ire," which derives from the Old English words for "haste" and "zeal" as well as from the Greek heiros, or "holy." What better linguistic pedigree for a once-sacred clinical concept that was promoted in haste and defended with zeal, and whose demise precipitated both frenzy and anger? Most important for the present context, the estrogen story has become a gadfly that is provoking a reconsideration of just how we know what we know about a drug’s effectiveness.
The shared delusion about long-term hormone replacement therapy started innocently enough. For centuries, women of a certain age (and their spouses) knew that the end of regular menstruation was often accompanied by the onset of uncomfortable hot flashes, insomnia, and a drying of the internal surface of the vagina. As the new science of physiology developed in the 1900s, these changes were understood to result from a falloff in estrogen production by aging ovaries. If a reduction in natural estrogen was the cause of these problems, then maybe restoring a woman's estrogen to premenopausal levels might ameliorate them. By mid-century it was possible to create a pharmaceutical product that would enable women to replenish their own flagging hormone by ingesting it in pill form. Pregnancy sharply increases the production of estrogens, and they are copiously secreted in the urine. Someone figured out that horses could be used as mass-production hormone factories for this purpose, and pregnant mare’s urine provided Premarin with both its ingredients and its brand name.
By the 1970s, with the support of Ayerst (now Wyeth), the manufacturer of Premarin, Madison Avenue magic redefined the normal age-related reduction in estrogen levels into a new syndrome, "ovarian failure." This novel disease concept was featured prominently in medical journal advertisements for Premarin, which was presented as its logical treatment. And so a normal age-related change took its place alongside kidney failure, congestive heart failure, and liver failure as a newly discovered illness in need of treatment. (I don't intend to belittle the discomfort associated with the hot flashes, vaginal dryness, and other symptoms that accompany menopause, or to suggest that they should not be treated for short periods simply because they are normal. What is at issue is lifelong "replacement therapy.")
Things began to go astray when the temporary management of menopausal symptoms became transformed into a belief that ovarian failure was itself a treatable risk factor for other dread conditions, just as elevated cholesterol or blood pressure was. If that were true, then lifelong drug management would be necessary to tame this risk and prevent a host of horrible clinical outcomes. The concept was fed by public fascination with the idea of a pharmacological fountain of youth for women. The Dupont company had already introduced the motto "Better Living Through Chemistry" to promote its line of household products; a few years later my own generation would co-opt that phrase more ironically in defense of psychedelics. In the same spirit, the 1966 best seller Feminine Forever popularized the notion that a woman's aging (and quite explicitly, her loss of sexual appeal) was now preventable, thanks to pharmaceutical research. Its author, Dr. Robert Wilson, was a gynecologist who took on the curing of menopause as a personal crusade, to save the millions of women who "suffered sweeping metabolic disturbances that literally put them in mortal danger." His views were a twisted precursor of the argument that anatomy isn't destiny; but he seemed to warn that without proper medical attention, physiology could become tragedy:
Though the physical suffering from menopausal effects can be truly dreadful, what impressed me most tragically is the destruction of personality. Some women, when they realize that they are no longer women, subside into a stupor of indifference.... I rest my case on the simple contention that castration is a bad thing and that every woman has the right--indeed, the duty--to counteract the chemical castration that befalls her during her middle years.
A new magic bullet could replace, molecule for molecule, the female hormones that failing ovaries could no longer secrete. Feminine Forever brought this message to women and their doctors all over the country, supplemented by well-placed articles in women's magazines and news releases describing this bold, modern treatment. The company could not legally advertise the drug for long-term use, since it had not presented clinical data to FDA demonstrating the promised benefits. But it could support numerous educational programs put on by hospitals, medical schools, and medical communications companies, with the understanding that they would promote lifetime-use messages. Premarin's manufacturer heavily but quietly subsidized the writing and distribution of Dr. Wilson's book as well as these prolific "public information" campaigns. Authors friendly to Wyeth and the gospel of estrogen wrote articles that appeared by the dozens in medical journals and women's magazines from the 1970s to the late 1990s. (We will return to the role of pharmaceutical companies in shaping beliefs about drugs in a later chapter.)
Prolonged use of pharmaceutical estrogen came to be known as hormone replacement therapy, or HRT; the very name gave legitimacy to the treatment by connoting the restoration of a vital bodily ingredient that went pathologically missing at menopause. The logic was appealing, but appealing logic alone has resulted in some of the very worst drug treatments in human history. After all, the second-century a.d. texts of Galen helped keep medicine in the Dark Ages for centuries with their supremely logical but completely incorrect theories of how derangement of the four humors caused most human disease. HOW WE WENT WRONG
Some of the apparent evidence favoring HRT came from measurements of surrogate outcomes--laboratory markers used as stand-ins for real clinical events. It takes years of study to show that a treatment eventually lowers the rate of heart attack or stroke, but you can demonstrate changes in patients' blood tests in just a few weeks or months, using far fewer subjects at a fraction of the cost of a full-scale, long-term clinical study. Here is how the logic of the surrogate outcome worked. It was known that patients with high levels of the "bad" cholesterol LDL and low levels of the "good" cholesterol HDL were more likely to suffer from heart disease and stroke. Estrogens were shown to lower LDL and raise HDL in the blood, so it seemed reasonable that they would eventually reduce the risk of cardiovascular disease. Hormone therapy was also found to produce apparently beneficial short-term effects on several other more arcane measures of vascular function.
The logic of estrogen replacement was bolstered even more compellingly by voluminous epidemiological evidence, only some of it correct. The argument began with the observation that until late middle age, women have dramatically lower rates of heart disease and cardiac death than men. But once menopause occurs, a woman's risk of cardiovascular disease begins to catch up to that in men of the same age. Next came a series of observations that were even more seductive. Several papers reported that postmenopausal women who took estrogen had less heart disease than women of the same age who did not. Replacing the estrogen a woman could no longer make for herself didn’t just prevent hot flashes and keep the vaginal tissue supple; it also seemed to maintain the heart and its arteries as healthy as those of younger women.
Additional observational studies seemed to uncover other connections as well. Long-term estrogen users were reported to have less incontinence, Alzheimer's disease, and depression than women of the same age who remained bereft of estrogen replacement. Here as well there were plausible biological mechanisms to explain the observed associations. In many older women, incontinence results from the thinning and shrinkage of vaginal tissue that occur when natural estrogen levels drop; adding pharmaceutical estrogen, either in pill form or as a cream applied directly to the affected areas, perhaps reversed these changes. The explanations for Alzheimer's disease and depression were more tenuous, since we still have little understanding of the basic causes underlying these conditions. Nonetheless, the papers reporting reduced rates of each illness usually included fascinating sections proposing possible mechanisms of such effects, related to the presumed effects of estrogen on derangements in brain chemistry.
There was still the problem of cancer. In the 1970s, it became clear that estrogen replacement increased a woman's risk of developing cancer of the endometrium, the lining of the uterus. Bathing this tissue in extra estrogen for so long provided hormonal stimulation far beyond what those cells expected in their retirement years, and led to a higher rate of malignancy. HRT advocates responded that this problem could be handled with regular endometrial biopsies; at the first sign that cells were becoming cancerous, they could be removed. Or the drug regimen could be altered to a more natural format by combining the estrogen with another female hormone, progestin, to mimic the normal premenopausal cycle more closely. The addition of progestin would help protect the uterus from estrogen overdrive and help it to catch its cellular breath, preventing it from becoming overstimulated and then cancerous.
Alternatively, a few zealots (all men, as far as I can tell) even argued that given the obvious benefits of long-term estrogen replacement, the simplest way to prevent endometrial cancer would be just to cut out that pesky uterus. After all, once it had finished its purpose of childbearing it was a useless organ whose only final act could be malignant transformation. The United States already had the world's highest rates of hysterectomy, often performed for indications now considered questionable, such as heavy periods or abdominal pain of unclear origin. But cooler heads prevailed, and for a woman who kept her uterus the estrogen-progestin combination worked well to eliminate the excess risk of uterine cancer; the most widely used product was Wyeth-Ayerst's Prempro. Women who had already undergone hysterectomy could take the straight stuff, marketed primarily by Wyeth as Premarin.
The use of Premarin and related products soared--not just for management of symptoms related to menopause itself, but also for all those other age-defying preventive purposes as well. It was conventional wisdom during my early years of practice that enlightened physicians put their older female patients on estrogen replacement for life as soon as they reached menopause. Besides the apparent protective effects of HRT on the heart, estrogen replacement also caused a reduction in bone-thinning osteoporosis, reducing the risk of fractures that can be so devastating in older women. Yes, there was also an increase in the frequency of breast cancer, but elaborate calculations based on the expected reduction in heart disease and fractures showed that those benefits would far outweigh the breast cancer, blood clots, and other problems the drugs were known to cause.
The Food and Drug Administration allowed Wyeth to advertise its products only for the management of menopausal symptoms such as hot flashes and insomnia, as well as for preventing age-related loss of bone mass--the only outcomes for which the company had submitted clinical outcome data. But beyond their text the Premarin ads conveyed a more comprehensive vision of feminine youthfulness. The words may have stuck to the letter of the law, but the larger message was carried in color photos of robust middle-aged women, their hair barely flecked with gray, romping zestily along a beach with a large dog or virile-looking husband. This notion was reinforced by virtually unanimous assertions of the benefits of HRT that could be found in nearly all textbooks. Hundreds of continuing education programs for doctors, many of them supported by Wyeth, were offered at professional society meetings of gynecologists or primary-care physicians. The groups that arranged these sessions were permitted to bring in experts of their choosing who could, in turn, say whatever they wished about the drug; this was beyond the pale of the direct company promotion that the FDA controlled.
THE BIRTH OF DOUBT
In the early 1990s, some voices of skepticism began to be noticed. Granted, numerous studies seemed to show that estrogen users had lower rates of several diseases when compared to age-matched nonusers. But nearly all these insights had come from epidemiologic analyses. Such studies follow users of a specific drug as well as nonusers and track their fates, adjusting the results for differences such as age and preexisting illnesses. In the studies of HRT use and heart disease, numerous adjustments were also made for cardiac risk factors such as smoking, diabetes, high blood pressure, and even family history.
But purists argued that the findings came solely from observational studies of people who did or didn't happen to be taking the drug. That's not the kind of evidence FDA requires: a clinical trial in which patients are put on Treatment A or Treatment B and then followed forward in time. With HRT so widespread, some began to ask whether such an experimental study should be conducted to demonstrate convincingly the preventive effects everyone expected. To accomplish this, instead of simply observing rates of disease in users and nonusers, researchers would have to randomly assign women to take estrogen or a dummy pill, and follow them for years; neither the women nor their doctors would be told who was getting which treatment.From the Hardcover edition.
Excerpted from Powerful Medicines by Dr. Jerome Avorn. Copyright © 2004 by Jerry Avorn, M.D.. Excerpted by permission of Vintage, a division of Random House, Inc. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.