Family history and risk of breast cancer: an analysis accounting for family structure

PURPOSE: Family history is an important risk factor for breast cancer incidence, but the parameters conventionally used to categorize it are based solely on numbers and/or ages of breast cancer cases in the family and take no account of the size and age-structure of the woman’s family. METHODS: Using data from the Generations Study, a cohort of over 113,000 women from the general UK population, we analyzed breast cancer risk in relation to first-degree family history using a family history score (FHS) that takes account of the expected number of family cases based on the family’s age-structure and national cancer incidence rates. RESULTS: Breast cancer risk increased significantly (P (trend) < 0.0001) with greater FHS. There was a 3.5-fold (95% CI 2.56–4.79) range of risk between the lowest and highest FHS groups, whereas women who had two or more relatives with breast cancer, the strongest conventional familial risk factor, had a 2.5-fold (95% CI 1.83–3.47) increase in risk. Using likelihood ratio tests, the best model for determining breast cancer risk due to family history was that combining FHS and age of relative at diagnosis. CONCLUSIONS: A family history score based on expected as well as observed breast cancers in a family can give greater risk discrimination on breast cancer incidence than conventional parameters based solely on cases in affected relatives. Our modeling suggests that a yet stronger predictor of risk might be a combination of this score and age at diagnosis in relatives. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1007/s10549-017-4325-2) contains supplementary material, which is available to authorized users.