B0261
Title: Bayesian semiparametric model for sequential AML treatment decisions with informative timing
Authors: Arman Oganisian - Brown University (United States) [presenting]
Jason Roy - Rutgers University (United States)
Abstract: A Bayesian semiparametric model is developed for the impact of dynamic treatment rules on survival among patients diagnosed with pediatric acute myeloid leukemia (AML). The data are from a phase III clinical trial in which patients move through a sequence of four treatment courses where they are treated with either anthracycline-based chemotherapy (ACT) agents or non-anthracycline-based agents only. While ACT is thought to suppress AML aggressively, it is also cardiotoxic, so that treating overzealously with either may reduce survival. Our task is to estimate the potential survival probability under hypothetical treatment rules, but there are several impediments. First, since ACT is not randomized, its effect on survival is confounded over time. Second, subjects initiate the next course at varying times depending on when they recover from the previous course - making timing potentially informative of subsequent ACT decisions and survival. Third, patients may die or drop out before completing the full treatment sequence. To address these issues, we develop a generative Bayesian semi-parametric model based on Gamma Process priors that capture subjects' transition to subsequent treatment or death in continuous time. A g-computation procedure is used to compute posterior potential survival probabilities while adjusting for time-varying confounding.
