The study revealed women with the highest levels of either serum estradiol (an endogenous estrogen) or testosterone were three times more likely to develop breast cancer.

The multi-center study involved participants from the Study of Osteoporotic Fractures and included 97 white women with breast cancer who were at least 65 years of age and not taking estrogen.

According to Dr. Cauley and her colleagues, the average incidence of breast cancer in white women aged 65 years old in the US is 4.6 out of 1,000 people. The study indicated, "the incidence of breast cancer in women with the highest concentrations of bioavailable estradiol and free testosterone is about 40 percent higher than this expected rate."

Sources of testosterone in postmenopausal women include direct secretion from the ovary and indirect secretion of precursor hormones. Testosterone could influence the risk of breast cancer directly or indirectly (as a source of estradiol).

Hormone Concentrations
While many women in America are put on hormone replacement therapy that increases their levels of estrogen, Dr. Cauley said her study did not necessarily indicate these women should stop their therapy.

"It [the study results] suggests if you give them hormone replacement therapy and their estrogen levels are raised, you would predict that you?d find a much higher level of breast cancer," Dr. Cauley said. But she stressed this is not necessarily the case. Previous studies have indicated women on long-term hormone replacement therapy have an average incidence of breast cancer of 1.4 to 1.6 out of 1,000 people, she said. This indicates a high risk of breast cancer may be linked with how estrogen is metabolized in the body.

But there are natural ways to lower serum hormone concentrations, such as a low-fat diet, weight reduction, and a vegetarian diet. For instance, Dr. Cauley?s study referred to the "Women?s Health Trial," a 10- to 22-week low-fat diet intervention, that resulted in a 17 percent reduction in estradiol concentrations in healthy postmenopausal women.

More Studies Needed
Phyllis Wingo, PhD, director of surveillance research for the American Cancer Society, said while Dr. Cauley?s study"fits" with information that epidemiologists have predicted, she would like to see more studies conducted.

"We need to see studies with different racial and ethnic groups and studies that have larger [numbers of] participants. We need to see studies conducted by different investigators and in different countries. When you see similarities [from the results of many studies] you can then say we have evidence of a relationship," Dr. Wingo said. Dr. Cauley said her next study will test the hypothesis that Raloxifene, an anti-estrogen drug, can reduce breast cancer incidence in those women who have higher levels of natural estrogen.

The Women?s Health Initiative is a $625 million, fourteen-year study involving approximately 160,000 women at 445 clinical centers, as well as thousands of others in community studies across the US. They have offices at Emory University in Atlanta and the University of Washington in Seattle, and are currently conducting similar tests with women of various ethnic groups, according to Dr. Cauley.

• Breast Cancer Resource Center