Grant Number:	4R37CA256810-05 Interpret this number
Primary Investigator:	Jiang, Shu
Organization:	Washington University
Project Title:	Dynamic Prediction Incorporating Time-Varying Covariates for the Onset of Breast Cancer
Fiscal Year:	2025

PROJECT SUMMARY Accurate assessment of risk is a top priority in oncology due to the population burden of cancer. Breast cancer is the leading cancer diagnosis among women worldwide and accordingly has the longest and broadest focus on risk prediction. Most traditional prediction models only utilize baseline factors known to be associated with breast cancer risk. The intrinsic heterogeneity between and within patients over time are reflected in part, by the time-varying covariate trajectories, which may provide important information for the prediction of breast cancer risk. The accumulation of cancer risk over life, well documented for breast cancer, is ideally suited to methods that incorporate time-varying covariates. The objective of this proposal is to provide novel statistical models that can incorporate patient heterogeneity in a personalized, dynamic manner leading to a more accurate risk prediction scheme. The proposed algorithms encompass innovative functional approaches to comprehensively characterize the changing pattern of the longitudinal trajectories by a set of outcome- independent/unsupervised and outcome-dependent/supervised features. The set of individual-specific features will contain information on the observed time-varying ‘pattern’ rather than one-time exposure in existing methods, leading to a higher predictive power. The dynamic prediction models will be built in a stepwise fashion, starting with a single time-varying covariate, and extended to the multivariate settings, to accommodate multiple time-varying covariates. In addition to contributions in prediction, the temporal change/trajectory of risk factors can add insights to pathways operating on risk of breast cancer. To develop viable preventive strategies, understanding the causal mechanisms whereby an exposure affects such dynamic trajectories (or mediators), to then in turn produce the breast cancer outcome is crucial, as these will provide insights into pathways that can better target breast cancer prevention and intervention trials. Given our ability to characterize the dynamic trajectories, we are positioned to fill in this gap and assess their intermediate role on the breast cancer pathway. In the two-year extension, we propose to develop a computationally efficient causal mediation framework to quantify the extent to which the effect of risk factors on breast cancer risk is mediated through BMI, hormone type and duration, mammographic density, and the whole mammogram image trajectory and the extent it is through other pathways. Successful completion of the proposed project will provide a transparent, robust, and reproducible statistical basis for inferences with the potential to shift the current paradigm leading to new pathways that can be targeted in breast cancer prevention and intervention trials to expand capacity for precision prevention.