Cancer Genes Don't Tell the Whole Story

Genetic tests for mutations in the so-called breast cancer genes, BRCA1 and BRCA2, may not reveal as much about cancer risk as earlier reports have estimated, according to two studies published in tomorrow's New England Journal of Medicine. Researchers report that women with the mutation are not as likely to develop breast and other cancers as previously thought, and that the mutations may also be less common among women with a family history of breast cancer. The results reinforce the idea that it is too soon to implement widespread testing for the mutations, say experts, who caution against drastic responses--such as mastectomies--for women who test positive. "We don't know enough yet to translate test results into routine patient care," says medical school dean Bernadine Healy of The Ohio State University, former director of the National Institutes of Health.

To find BRCA mutations--which are very rare--most studies have examined families with very high rates of breast and other cancer in young family members. But that sample tends to overestimate the risk of cancer for women at large, says epidemiologist Jeffrey Struewing of the National Cancer Institute. So he and his colleagues recruited a population--more than 5000 people from the Ashkenazi Jewish population in and around Washington, D.C.--without regard to family or personal history of the disease. (Mutations in BRCA1 and BRCA2 are more than 10 times more prevalent in the Ashkenazi Jewish community than in the population at large, allowing Struewing to use fewer participants to obtain a statistically significant pool.)

Only 2.4% of the sample had the mutation, a level consistent with previous studies. These carriers were found to have only a 56% chance of developing breast cancer by age 70, compared with a figure of 87% in previous studies. For ovarian cancer, the group found a 16% risk, compared with previous estimates that ran as high as 64%. And even the new estimates may be too high, says Struewing, because people may have been more likely to volunteer for the trial if they or a family member had breast cancer.

Another study, conducted by Barbara Weber and her colleagues at the University of Pennsylvania, screened 263 breast cancer patients for BRCA1 mutations and asked them about their family history of breast and ovarian cancers. The researchers found that only 16% of women with a family history of breast cancer had BRCA1 mutations, compared with the 45% found in previous studies. However, says Weber, that lower risk could lead to another problem--a false sense of security among family members of cancer patients without the mutation that a negative BRCA1 test represents a clean bill of health.

Healy says the new findings dispel the idea that "if you have mutations in these genes, you're going to get breast cancer." They underscore the need to uncover and sort through many other factors, she says, including more genes and such environmental factors as diet and exposure to estrogen.