Just 10 years ago doctors routinely recommended estrogen for women at menopause, because they believed it would lower the risk of heart disease as well as relieve symptoms, such as hot flashes. Women who had not had hysterectomies (in which the uterus is removed) also took progesterone to prevent an estrogen-fueled thickening of the uterine lining, which increases the risk of uterine cancer. (Women who have had their uteruses and ovaries removed, as a result of cancer or fibroids or other health issues, take only estrogen.) But in 2002 a major federal study, the Women's Health Initiative (WHI), was halted early because researchers found that the combination of estrogen and progesterone actually increased the risk of heart attack, stroke and blood clots. Even scarier for many women was the increased risk of breast cancer after taking estrogen and progesterone, especially after five years. Within weeks of the widely publicized announcement of these findings, millions of women threw out their pills. Since then many scientists have been looking more closely at the WHI results and conducting their own research on the effects of estrogen therapy. This week researchers from the Fred Hutchinson Cancer Research Center in Seattle are reporting that taking estrogen and progesterone for three or more years during or after menopause may result in a fourfold increase in the risk of getting lobular breast cancer, the second-most common type of the disease. In the study, published in the January issue of Cancer Epidemiology, Biomarkers and Prevention, researchers looked at 1,500 postmenopausal women; 1,044 had breast cancer, while 469 without cancer were a control group used for comparison. NEWSWEEK's Barbara Kantrowitz talked to the lead author of the study, Dr. Christopher Li. Excerpts:
NEWSWEEK: Many doctors currently tell patients that taking estrogen therapy for fewer than five years to relieve menopausal symptoms is reasonably safe. Do your results mean there's no safe length of time to take it?
Dr. Christopher Li: I think you have to look at the findings in context. There are different types of breast cancer. If we look at it by histology [the microscopic structure of the tissue], the most common type is ductal carcinoma. That accounts for about 70 percent of all cancers. The second-most common type is the lobular type, and that's now about 20 percent of all cancers. In this study we found that women who used the combined therapy, the estrogen plus progesterone, for three to five years had a fourfold increased risk of the lobular type. But they had no elevation in their risk of the most common type, the ductal type. That short duration of use may impact risk of the rare type of breast cancer but probably is not having any real impact on the most common type of breast cancer. In the study we also looked at use of unopposed estrogen, which is also still commonly used. There was no effect, even among women who used [only] estrogen for 10 years or longer, for any type of breast cancer, whether it was ductal or lobular. So I think women who are using that regimen could actually feel somewhat reassured. The WHI trials essentially found the same thing, that there was no elevation in risk of breast cancer for women using unopposed therapy. It is women who are using the combined therapy who really have to worry about breast cancer risk.
How concerned should women be who use the combined therapy?
The lobular cancer is challenging clinically in that it's harder to detect and treat. But at the same time it actually has a somewhat better survival rate than the ductal carcinoma. That is really because the lobular cancers are more hormonally responsive. They're more likely to be estrogen-receptor positive, so therefore they are much more amenable to some of the targeted therapies we have available, like hormonal therapies and aromatase inhibitors. It is, in a sense, a better type of breast cancer to get than the ductal type because it has a better survival rate and it is treatable. But of course it is still going to cause the same amount of anxiety and worry that any breast cancer would induce in women.
There's no way a woman starting estrogen therapy can know whether she is more likely to get one or the other kind of breast cancer.
Right. We don't know that.
So can we say who would be a good candidate for estrogen therapy and who would not be, in light of your study?
It ultimately is an individual decision. We can't really stratify and say one group of women is going to be more likely to develop breast cancer than another. We're not at the point where we can do that. One of the real problems with breast cancer is that we have a lot of risk factors that we have identified, but none of them really seem to add together in any simple mathematical way. We're stuck with all these different risk factors. A lot of them are not even modifiable, such as family history or reproductive history. Those are things women can't really change once they are postmenopausal. So I think it's a tough decision, and women have to weigh the risks and benefits. Breast cancer is certainly one of the risks. But with the WHI trials there were all sorts of other risks associated with combined therapy, and I don't think our study really changes the overall situation, which is that women should obviously talk to their doctors about this and they should use the lowest dose of hormones possible and use them for the shortest time possible. Quality-of-life issues are certainly going to be important and maybe even more important to some women than these risks of disease. It ultimately comes down to a woman's individual choice.
Since it is only the combined therapy that appears to increase risk, could progesterone be the problem?
I certainly think that progesterone is playing a very important role because of the total lack of effect on women who use unopposed estrogen. That is something that we are also trying to understand better. What is it about the progesterone that creates risk, particularly of the lobular subtype? On this study we collected all of the tumor tissues, so we're running different tests on the tumors to try and see what is correlated with hormone therapy, what markers may be associated with that. That's what we're actively working on, but we don't have the answers quite yet.
Women who use unopposed estrogen generally have had their ovaries removed or have other health issues. How does that affect the difference in risk?
Women who have had a hysterectomy, particularly an oophorectomy [in which the ovaries and sometimes the fallopian tubes are removed], will have a lower risk of breast cancer regardless of whether or not they take hormones. That is because breast cancer risk is driven by hormones produced by the ovaries. The sooner that ovarian production of hormones stops—and it would typically occur earlier in a woman who had her ovaries removed—the lower the lifetime risk of breast cancer.
Why are epidemiological studies like yours an important tool for following this and understanding risk?
The advantage of this type of study is really that we can collect very detailed information from women and we can also get tumor tissue blocks. The WHI was a big study, but it had certain limitations because it was so big. It was more expensive. They didn't collect any tumor tissue blocks in that study, so this is the first study to confirm what the actual histology was by a centralized histology review.