Question: How accurate are the predictions the doctors give me for my risk of getting breast cancer?
Answer: The way that I like to think about predicting risk is to divide women into two main groups: those who are at hereditary risk for breast cancer and those who are not at hereditary risk. And by hereditary risk, I mean women who are in families where there is clearly a predisposition to breast cancer.
What does that mean? It means numerous cases of breast cancer, often ovarian cancer in the family. The cancers are diagnosed at young ages -- women in their 20s and 30s. Clearly there's a gene being transmitted in such a family from generation to generation that's putting women at significantly increased risk.
So that's hereditary risk and there we do have some fairly sophisticated and specific ways to predict risk. In some of those families we can do gene testing. Now that makes up only a minority of breast cancer; maybe 5-10 percent of all breast cancer is hereditary. What about in the nonhereditary case? And this would be women who have no family history of breast cancer or I would also include women who have some family history, but it's very mild or moderate in nature. In other words, maybe a mother had breast cancer when she was 70 and that's the only family history that there is. That's clearly very different from the hereditary case that I mentioned. In these women we have a tool called the Gail Model -- G-A-I-L -- named after the statistician who developed it. And that model uses several factors, all of which we know increase risk of breast cancer -- the women's age, her reproductive history, her menstrual history, whether she's had breast biopsies or not, and yes, there's a piece there too about family history. And that's a fairly good predictor for women in the nonhereditary category. So for both groups, hereditary, and then, others, I think we have fairly good risk prediction capabilities today.