Prevention and intervention such as mammographic screening (MS) and prophylactic surgery (PS) can potentially decrease cancer risks in BRCA1/2 mutation families but their evaluation is generally complicated by the fact that their effects may vary substantially depending upon which of MS or PS occurred first and the age they are applied.
We introduce a competing risks model with either breast cancer, ovarian cancer or death as first event and time-dependent covariates to account for a woman’s history of MS and PS. A frailty model is specified to account for familial residual dependence and the bias induced by the sampling of families through probands is corrected via a prospective likelihood approach.We illutrate our proposed method through an analysis of BRCA1/2 mutation carrier families recruited through the Breast Cancer Family Registries.
The effects of PS and MS on breast and ovarian cancer risks depend largely on the ages of PS and MS performed and the time-varying function.