Dr. Susan Love Research Foundation | Breast Cancer

Reports from two large studies of postmenopausal women indicate that short-term use of estrogen appears to be safe, while long-term use appears to increase breast cancer risk.

The first study, an updated analysis of the Women's Health Initiative (WHI) Estrogen-Alone Trial, published in the April 12, 2006, issue of the Journal of the American Medical Association, found that estrogen-alone hormone therapy does not increase the risk of breast cancer in postmenopausal women.

The second study, an analysis of data from the Nurses' Health Study, published in the May 8, 2002, issue of the Archives of Internal Medicine, found that short-term use of estrogen-alone hormone therapy was safe, but that long-term use of 15 years or more markedly increased breast cancer risk.

Estrogen-alone hormone therapy can only be taken by women who have had a hysterectomy. Women with a uterus who take estrogen have an increased risk of uterine cancer, requiring them to take estrogen combined with progestin.

The WHI Estrogen-Alone Trial included 10,739 postmenopausal women between the ages of 50 and 70 who had previously had a hysterectomy. The study randomized half of the women enrolled to 0.625mg of estrogen (Premarin) and half to placebo.

In 2004, after the women had been followed for about seven years, the researcher stopped the trial after an interim analysis found that estrogen alone appeared to increase the risk of stroke. At the time the study was stopped, the effect that estrogen had on invasive breast cancer was unclear, although it appeared that women taking estrogen had a lower risk of developing breast cancer than the women taking the placebo. As a result, researchers decided to examine the data more closely to better assess the relationship between estrogen and breast cancer.

The new analysis confirmed that study participants taking estrogen had fewer breast cancer tumors than those in the placebo group. Among the 4,763 women in the placebo group, there were 163 breast cancers (133 invasive cancers, 30 cases of ductal carcinoma in situ [DCIS] or lobular carcinoma in situ [LCIS]). In comparison, among the 4,718 women taking estrogen, there were 129 breast cancers (104 invasive cancers, 25 cases of DCIS or LCIS). However, the difference between the two groups was not statistically significant. This means the difference could have happened by chance.

Although the women in the estrogen group were less likely to develop breast cancer, those who did develop the disease tended to have larger tumors than the women in the placebo group. The average size tumor in the estrogen group was 1.8cm, compared with 1.5cm in the placebo group. In addition, the women in the estrogen group were more likely to have tumors that had already spread to the lymph nodes by the time they were diagnosed than were the women in the placebo group (35.5 percent in the estrogen group had positive nodes compared with 23.3 percent in the placebo group).

The researchers also found that the women taking estrogen were more likely to have an abnormal mammogram that required follow-up than were the women in the placebo group. Specifically, 747 of the 4,718 women in the estrogen group needed follow-up compared with 549 of the 4,763 women in the placebo group.

Based on these findings the researchers concluded "initiation of [estrogen] alone in women after hysterectomy should continue to be based on careful consideration of potential risks and benefits for a given individual."

The Nurses' Health Study, established in 1976, is one of the largest prospective studies ever conducted on risk factors for major chronic diseases in women. Every two years the 121,700 nurses in the study complete a health questionnaire.

For this analysis, the researchers assessed the relationship between estrogen use and breast cancer risk in 28,835 of the nurses who were postmenopausal and had had a hysterectomy. Of these women, 3,288 had never used estrogen; 3,255 had used estrogen for less than five years; 5,006 had used estrogen between five and 10 years; and 4,492 had used estrogen for more than 10 years.

The researchers found that 934 of the 28,835 women studied developed invasive breast cancer: 226 cases developed in women who had never used estrogen and 708 cases occurred in women who were using estrogen. Breast cancer risk increased the longer the women were on estrogen. Women who used estrogen for 10 years or less had no increased risk. But women who used estrogen for 15 years or more did, with women using estrogen for 20 years or more having the greatest increased risk.

Based on these findings the researchers concluded "although current use of estrogen therapy for less than 10 years was not associated with a statistically significant increase in breast cancer risk, the WHI has shown an increased risk of stroke and [blood clots] in the same time period. Women who take estrogen therapy for prevention or treatment of osteoporosis typically require longer-term treatment and should thus explore other options, given the increased risk of breast cancer with longer-term use."

Susan says:

There are a few aspects of these studies we need to look at in more detail:

The WHI found that the women taking estrogen had fewer breast cancers than the women taking the placebo. However, the difference between the two groups was not statistically significant. This means the difference could have happened by chance. What does this mean? It is not enough to compare the total number of cancer occurrences. In any large study such as this, there is always going to be some difference that is due to chance alone. (If you flip a coin three times, you could get heads three times. That doesn't mean that when you flip a coin you are more likely to get heads. It's just a random occurrence that has happened by chance.) For this reason, researchers use statistical tests that can help to determine the likelihood that a difference between two groups is by chance alone or is in fact a real difference.

The fact that there was no statistically significant difference between the two groups means that this study does not provide evidence that taking estrogen reduces breast cancer risk. We know that women who undergo oophorectomy (removal of the ovaries) have a lower risk of breast cancer—their risk is decreased by 70 percent. So, it would be important to look back at the women in this study and assess the difference in breast cancer risk between women who had ovaries and women who did not. I suspect that we would find that the women without ovaries who took estrogen had a lower risk of breast cancer than did the women with ovaries who took estrogen.

The WHI analysis found that the women taking estrogen had larger tumors than the women in the placebo group and that these tumors were more likely to have spread to the lymph nodes before they were discovered. This suggests that estrogen may have an effect on some types of tumors but not on others. It's also possible that the tumors were diagnosed later in the women taking estrogen because the estrogen made their breasts more dense, which makes it harder to detect cancer on a mammogram.

The findings from this WHI study contrast with the previously reported WHI Estrogen plus Progestin Trial, which found an increase in breast cancer over about five years among the women taking combined hormone therapy. This arm of the trial was stopped in 2002 after an interim analysis found that hormone replacement therapy increased the risk for breast cancer, heart attack, stroke, and blood clots. Based on these findings we can conclude that while adding progestin is necessary to protect the uterus from cancer, it is not good for the rest of the body.

The Nurses' Health Study found that risk increased in women who had been on estrogen for more than 10 years. This should not be interpreted as finding that estrogen use is safe for 10 years. It would also be interesting to know what impact oophorectomy had in this study. I would suspect that among these women as well, those without ovaries who took estrogen had a lower risk of breast cancer than did the women with ovaries who took estrogen.

The bottom line: These new findings support the current mantra about hormones—they should only be used to treat menopausal symptoms, they should only be used for as short a time as possible, and they should only be used at the smallest effective dose.

Note: The WHI is sponsored by the National Heart, Lung, and Blood Institute (NHLBI) of the National Institutes of Health. The National Institutes of Health (NIH)—the nation's medical research agency—includes 27 Institutes and Centers and is a component of the US Department of Health and Human Services. It is the primary federal agency for conducting and supporting basic, clinical, and translational medical research, and it investigates the causes, treatments, and cures for both common and rare diseases.

The Nurses' Health Study, one of the largest prospective studies ever conducted, is following 122,000 nurses throughout the United States. The study is led by a team of researchers at Harvard University's Channing Laboratory and the surrounding medical community.