Do hormones cause breast cancer?
Over the past decade, this is by far the most common question I’ve heard from women trying to decide whether to take hormone replacements to reduce the symptoms of menopause and generally help stay healthy and vital.
No doubt it’s also the most common question most ob-gyn’s have gotten from their patients hitting menopause since 2002. That’s when the federal government’s big Women’s Health Initiative (WHI) study reported the first results of its large-scale study of the benefits and risks of hormone replacement therapy in post-menopausal women.
Unfortunately, American women have largely gotten the wrong answer to that question. It’s only just recently that the myth of the WHI’s hormone study has finally started to come undone. New Views of Hormone Therapy in Menopause, The Wall Street Journal announced recently. Not quite as attention-grabbing as all the “Hormone Therapy Raises Risk of Breast Cancer” headlines, but a start.
The story that took hold in 2002 and didn’t let go was that women who took estrogen and progestin had a higher risk of breast cancer, as well as heart disease and stroke. The results were considered so striking and unexpected that the study was abruptly halted three years early so participants could stop taking these supposedly dangerous hormones. Naturally, millions of women across the country followed suit, and millions more never started.
And what a terrible shame. Despite the headlines, the WHI study didn’t demonstrate a link between hormone replacement and a higher risk of breast cancer for most women—particularly not for those just entering menopause, the most relevant group. But it’s only been in recent months that the medical establishment and the media are finally coming to this realization—after nearly a decade of misinformation, misinterpretation and misguided advice by people who should have known better.
led to misguided advice
As recently as a year ago, the skewed warnings were still coming from the most respected medical journals and experts and spread by the major news media. In October 2010, the Journal of the American Medical Association reported that a follow-up of the initial WHI study found that participants on hormone replacement had died of breast cancer at a higher rate than those on placebo. “Another round of bad news today for hormone replacement therapy,” National Public Radio reported that day last October. “The first results from the WHI showed hormone replacement doesn’t prevent heart disease. And it does raise the risk of breast cancer. But only now can the WHI answer one of the big questions about hormone replacement therapy: Does it cause more women to die from breast cancer? And that answer is yes.”
Not quite. The real answer—the more rational, scientifically and intellectually honest one—isn’t so simple. It depends on which women, when they start taking it, and what types of estrogen and progestin they use. When the data is re-examined keeping these factors in mind, the answer is actually much more no than yes for the vast majority of women considering starting HRT.
Here’s what’s happened: For years, researchers, physicians and the media have taken grossly generalized data and—no surprise—come to grossly generalized conclusions. And they’ve used those conclusions to make a single recommendation to every single post-menopausal woman: Don’t go on hormone replacement. What’s changed? Not the data. Just how they’re being interpreted after a closer look behind the numbers. Finally, they’re coming to be seen as far more nuanced than the headlines (all but written by HRT opponents) have had us believe.
Problem #1: Age matters. The 16,608 post-menopausal women in the WHI study who took hormone therapy had a broad age range—from 50 to 79. Moreover, on average they were much older than those in the study who were just entering menopause. And many of these older participants were started on hormone therapy for the study—an average of 12 years after menopause.
But the results, and the resulting advice, took none of this into account. The data were reported and interpreted for the entire group of 16,608—as if a 51-year-old woman going on hormones at the start of menopause was the same as a 64-year-old taking them for the first time, and whose age put them at higher risk of the cancer, heart disease and stroke later reported.
A closer look reveals a game-changing difference that seems nothing if not intuitive: Yes, women in the study who were taking hormones had a higher incidence of breast cancer and cardiovascular disease—but they were largely women in their 60s and 70s who had only started on the therapy for the study years after menopause. But it was a whole different story for women in their 50s, taking hormones early in menopause. They had no statistically significant increase in breast cancer. And they actually had less heart disease and had an overall lower mortality rate, from any cause, than participants who got placebos.
The difference between older and younger women in the study was so significant that it skewed the overall results—and consequently the judgment of thousands of physicians and millions of women deciding whether hormone replacement is a good thing or a bad thing.
Problem #2: Statistics don’t lie, but they can deceive. The October 2010 JAMA paper reported on an 11-year follow-up of the 12,788 women in the WHI study who had not had hysterectomies. They had taken estrogen and progestin for an average of about five and a half years. Even ignoring the age factor, the entire group had a surprisingly small increased risk of breast cancer—surprising given the earth-shaking headlines.
The difference in breast cancer incidence was 0.42 percent for those who had hormone replacement against 0.34 percent for those who took placebo. That works out to a difference of less than one case per 1,000 women. Meanwhile, the difference in deaths attributable to breast cancer was 0.026 percent versus 0.013 percent. That might seem like a doubled risk, but the fractions are so minute that it’s a difference of a little over one per 10,000 women.
Of course, we would like to avoid each of these deaths, and nobody wants to be that one per 10,000, but a good look at the actual numbers makes a strong argument for the individualized approach to personal health: In this case, looking beyond the headlines to see if they exaggerate the reality, and weighing the risks against the benefits to make the decision that’s right for you.
Problem #3. Not all hormones are created equal—because some are created by man. A more misleading aspect of the JAMA report may lie not in the statistics, but in its title: “Estrogen Plus Progestin and Breast Cancer Incidence and Mortality in Postmenopausal Women.”
It would be more accurate and less deceiving if it specified Premarin and Provera, the brand names for the particular versions of synthetic estrogen and progestin used in the study. That’s because there is much evidence that not all progestins are alike in their effect on breast cancer risk.
Provera is medroxyprogesterone acetate (MPA). It is one of several versions of what are classified as progestins—synthetically altered forms of the molecule called progesterone that occurs naturally in a woman’s body. Premarin, meanwhile, is the brand name for a hormone therapy that contains a combination of estrogens derived from pregnant mare urine. It made perfect sense to use Premarin and Provera in the WHI study because at the time they were the Coke and Pepsi of hormone replacement—90 percent of American women on HRT took them.
But these synthetic hormone therapies are not molecularly identical to others. Nor are they the same as the natural versions–progesterone and estradiol, the name for naturally occurring estrogen. Menopausal therapy with these natural hormones is called bio-identical hormone replacement therapy, or BHRT. Because the various natural and synthetic versions are not identical, they don’t have the same effects. This is akin to say, statins, a class of drugs used to lower cholesterol. Some brand-name statins have troublesome side effects that others don’t.
Yet, astonishingly, the authors of the Women’s Health Initiative papers still engage in this “class effect” thinking about the results of their investigations into just one type of estrogen and progestin — Premarin and Provera.
In Europe, where Premarin and Provera have not dominated the market, the medical thinking is different. A very large study French study, called E3N, followed 80,377 women on various combinations of estrogen and progestin for 10 years and found no increased risk of breast cancer. (1) This study focused on women going through menopause in real life conditions as opposed to older women being placed on Premarin and Provera more than a decade after. In vitro evidence also supports the difference between various progestins. In a study that looked at the effect of various progestins on the proliferation of breast cancer cells found that progesterone had an inhibitory effect whereas Provera did not. (2) The bottom line: Not all progestins are alike; results from a study with one progestin should not be generalized to all progestins.
Another part of the re-evaluation of the data involved separating women who took combined estrogen and progestin from those who took estrogen alone. The combination group had eight more breast cancers per 10,000 women than the placebo group. But the estrogen-only group had 10 fewer breast cancers per 10,000 women than the placebo group. Again, that was without separating by age, a factor that might strengthen the argument for estrogen alone as a protector against breast cancer for women in their 50’s.
It’s hard to know what led to the reconsideration of the data in recent months. Maybe it was the October 2010 JAMA report and the scare headlines that followed that got some leading experts to take a more sophisticated look at the numbers. And they have finally begun to end the decade-long mythology.
Wulf Utian, founder and former president of the North American Menopause Society (NAMS), recently told the Wall Street Journal: “Considering all the current date, it’s actually quite safe to take hormones for five to 10 years after menopause. If you minimize a woman’s exposure to progesterone, you minimize her slight risk of breast cancer. Meanwhile, the estrogen will have a beneficial effect on her brain, her skin, her bones and her heart.”
Considering what Utian and so many others had been saying for nearly 10 years, it was a stunning if tacit admission that they’d gotten it all wrong. Still, despite the 180-degree flip in the recommendations, they haven’t yet grasped the problem of the class-effect concept. I hope we don’t have to wait another 10 years for that.
I can’t think of a better illustration of the problem of broadly applying the results of a large clinical study and making recommendations that assume that everyone is just like the average of the people in that study. And if you’re a health consumer, it’s the best argument for looking beyond the latest headline and questioning what, if anything, it really means for you. JMR
1. Fournier A, Berrino F, Riboli E, Avenel V, Clavel-Chapelon F. Breast cancer risk in relation to different types of hormone replacement therapy in the E3N-EPIC cohort. Int J Cancer. Apr 10 2005;114(3):448-454.
2. Seeger H, Wallwiener D, Mueck AO. The effect of progesterone and synthetic progestins on serum- and estradiol-stimulated proliferation of human breast cancer cells. Horm Metab Res. Feb 2003;35(2):76-80.