Jan. 15, 2003

Comments, Suggestions?

The HRT dilemma

By Ruth SoRelle, MPH

At the beginning of the Women's Health Initiative's study of the effects of hormone replacement therapy (HRT) on the health of women in menopause, several physicians criticized it as unethical. They said it denied half the study's women the heart protective benefits of estrogen.

Jennifer Hays, PhD

Their criticisms gave the study's investigators pause, but they pushed on because the criticism centered on the critical question they wanted answered: "Does taking estrogen reduce your risk of cardiovascular disease, osteoporosis-induced fracture and colorectal cancer, and does it raise your risk of developing breast cancer?"

Last July, they had their answer - at least for the 16,000 women in the study who were taking a combination of estrogen and progestin. Hormone therapy in this group actually raised the risk of heart attack, stroke and breast cancer while reducing broken bones and colorectal cancer.

In fact, they stopped the study three years early because the evidence was clear that there was more harm than benefit to the treatment.

Jennifer Hays, PhD, director of the Center for Women's Health at Baylor College of Medicine and the principal investigator of the Baylor arm of the study, has been dealing with the fallout ever since. At least 6 million women take the form of estrogen-progestin (Prempro) that was used in the study. In many cases, they did so, in part, because they thought it conferred some protection to their hearts.

In fact, as the results of the Women's Health Initiative (WHI) show, the combination not only did not protect women's hearts, the risk of heart attack and strokes increased, 29 percent and 41 percent, respectively. The combination also increased the risk of breast cancer by 26 percent.

On the positive side, the drug reduced hip fractures by 34 percent, a definite proof of its effect on osteoporosis. It also reduced the risk of colorectal cancer by 37 percent after three years of hormone use.

Hays knows that the message was shocking. As her co-investigator at Baylor, Ronald Young, MD, told her "It's like finding out that insulin causes diabetes. You have been prescribing this to decrease heart disease and stroke, only to find that in fact it increases those problems in some women."

If nothing else, Hays said, she would like to get across two points. "The increases are small, but real and need to be acknowledged. A 26 percent increase is a 26 percent increase."

But a 26 percent increase in risk does not translate into an absolute 26 percent increase in disease. For example, the 26 percent increase in risk for breast cancer translates into eight more breast cancers per year for every 10,000 women in the population. (An estimated 30 women of 10,000 who do not take the drug would be expected to get breast cancer and 38 women of 10,000 who do take the drug would develop the disease.) Likewise, for heart disease, the 29 percent increase translates into seven more heart attacks each year for every 10,000 women and the stroke increase means that there will be eight more strokes annually for every 10,000 women.

As Hays sees it, the whole story of the Women's Health Initiative and hormone replacement therapy is a narrative that began in the 1980s when an analysis of the Nurses Health Study and other longitudinal studies indicated that there was a lower risk of cardiovascular disease in menopausal women who took hormones. The Nurses Health Study involved more than 116,000 women who were nurses who filled out questionnaires about their lives and their health on a regular basis.

"The problem was that these were an observational studies," said Hays. There was no placebo group against which to compare the information and no way to insure that the hormones were actually the reason for the difference. As Gina Kolata of the New York Times asked in a story on July 10, 2002, "The question is, does estrogen make women healthy, or do healthy women take estrogen?"

The Nurses Health Study was good for generating a hypothesis, but only a clinical trial that compared the medical histories of women who took hormone replacement therapy to the histories of women who did not would determine if the drugs had an effect on heart disease. In the absence of that, however, doctors began to prescribe the drug in increased numbers in hopes of warding off heart disease in their patients. It demonstrated the need for a good study, although it took time to get it off the ground.

Recruitment for the hormone replacement therapy portion of the Women's Health Initiative, the largest study of its kind ever done, began in 1993 and was completed in 1998. It was expected to last from the eight to 10 years. However, when it crossed the point where harm seemed to outweigh benefit, an independent study monitoring board called a halt.

Even before that, however, said Hays, the Heart and Estrogen Replacement Study (HERS) designed to evaluate the effect of hormone replacement therapy on women with heart disease was completed and found that hormone replacement therapy had no beneficial effect. "We thought that was going to be a bombshell," said Hays. "No one even noticed it."

Add the WHI data to it, and "now we know that estrogen and progestin in this form (the pill Prempro that combines the two) is not effective for heart disease prevention," said Hays. They sent out letters informing the 16,000 women in the combination HRT trial of the decision. One woman called the office in tears. She told clinic staff, "I can't stop taking these." When they determined that she had been on placebo or an inactive drug for the entire time, she laughed.

"The mind is very powerful," said Hays.

Another 11,000 women who have had hysterectomies are continuing to take estrogen because the data from the trial are not yet conclusive. They do not need progestin, which is only taken to prevent malignant change in the lining of the uterus.

Now Hays and her colleagues are frequently faced with the question of what women should do. She is analyzing the data about quality of life, hoping to find some clues.

"We want to help women identify alternatives," she said. "The problem is that we have less information on the alternatives than we do on the hormones themselves. Prempro is the only one on which we have good data."

She knows that advertisements advise women to take alternatives because they are natural or plant-based, "but we don't know if those act differently in the body in terms of triggering breast cancer or clotting factors that can lead to stroke."

Good clinical trials that compare the alternatives to placebo will be the "gold standard" in determining their value to women, she said.

Today, she is glad that the study was done and the knowledge in the public domain. She feels privileged to have worked with the women who took part in the study.

"These women are so courageous. I am so grateful to them. I think every woman in this country should say a prayer of thanks to these women. Can you imagine taking a pill once a day every day for 8-10 years, not knowing what it was whether it was helping or hurting you? That's a gift."

© Copyright 2003 Baylor College of Medicine. All Rights Reserved. MLE.