The Fred Hutch Biostatistics Program hosts seminars featuring presentations by Hutch and outside scientists to share their latest developments and recent research. Each seminar includes an hour-long presentation and discussion during which speakers showcase their work and findings.

This seminar will be held on Zoom due to the COVID-19 pandemic.

Abstract:

Heterogeneous treatment effects (HTE) based on patients’ genetic or clinical covariates are of rising interest to precision medicine. Simultaneously modeling HTE and associated main effects for high-dimensional markers is challenging. Motivated by the modified covariate approach, we propose a two-stage statistical learning procedure for estimating HTE with efficiency augmentation, exploiting powerful gradient boosting trees (GBT). Target estimands for HTE are defined as the mean difference for quantitative outcomes, or the risk ratio for binary outcomes, solving specialized loss functions. The first stage is to estimate the main-effect equivalency of baseline covariates on the outcome, which is then used as an augmentation term in the second stage estimation for HTE. The proposed two-stage procedure is robust to model mis-specification of main effects and achieve optimal efficiency for estimating HTE through consistent nonparametric function estimation. A permutation test is proposed for the global assessment of high-dimensional HTE via GBT. Simulation experiments and an analysis of a genetic study in Prostate Cancer Prevention Trial are conducted to showcase the properties and the utilities of the two-stage method.