A Competing Risks Model with Time Varying Covariates for Estimation of Breast Cancer Risks in BRCA1 Families

Mammographic screening and prophylactic surgery can potentially reduce breast cancer risks among mutation carriers of BRCA families. The evaluation of these interventions is usually complicated by the fact that their effects may change over time and by the presence of competing risks. We propose a competing risks model that accounts for time-varying interventions and provide cause-specific penetrance estimates for breast and ovarian cancers in BRCA1 families. A shared frailty model is specified to account for familial residual dependence with an ascertainment correction through affected probands, which accounts for competing risks and time-varying covariates (TVCs). Via simulation studies we demonstrate the good performances of our proposed approach in terms of bias and precision of the estimators of model parameters and cause-specific penetrances over different levels of familial correlations. We apply our new approach to 498 BRCA1 mutation carrier families recruited through the Breast Cancer Family Registry and illustrate the importance of our approach accounted for both competing risks and TVCs when estimating cause-specific penetrance of breast cancer.