Print  Close

Design and Analysis of Clinical Trials with Survival Outcome by Incorporating Pre-Randomization Longitudinal Biomarkers

Haolin Li Speaker
University of North Carolina at Chapel Hill
 
Tuesday, Aug 5: 8:55 AM - 9:15 AM
Topic-Contributed Paper Session 
Music City Center 
In contemporary cancer research, there is an increasing need to incorporate longitudinal biomarkers into randomized clinical trials to develop personalized treatment strategies that adapt to biomarkers measured prior to treatment. However, existing statistical methods for sample size/power calculations in clinical trials with survival outcomes often overlook the integration of longitudinal biomarker data. This article presents a sample size/power calculation formula with a robust inference method for estimating treatment effects without relying on distributional assumptions regarding random effects from longitudinal biomarkers. The proposed formula only requires easily accessible quantities from existing literature or pilot studies, avoiding any distributional constraints on survival or censoring times. Extensive simulation studies demonstrate that the proposed inference method and sample size/power calculation formula exhibit strong finite sample performances. The practical application of this method is illustrated through the design of a lung cancer prevention trial, utilizing data from the National Lung Screening Trial (NLST).

Keywords

Clinical trial design

Longitudinal-survival joint modeling

Personalized Treatment Strategies

Power Calculation