Application of Bayesian Joint Modeling of Time-to-Event and Patient-Reported Outcomes in an Oncology Clinical Trial

• Activity Number: 283
• Type: Topic Contributed
• Date/Time: Tuesday, August 3, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Luping Zhao*+ and Wei Shen and Haoda Fu and Michelle Denise Hackshaw and Mark E. Boye
• Companies: Eli Lilly and Company and Eli Lilly and Company and Eli Lilly and Company and Eli Lilly and Company and Eli Lilly and Company
• Keywords: Bayesian approach ; Joint modeling ; Oncology clinical trial ; Patient reported outcomes ; Time-to-event data

Abstract

The measurement of patient-reported outcomes (PROs) has gained substantial attention in recent oncology clinical research. Although several joint models of longitudinal and time-to-event data have been developed in recent years, few research studies have been conducted to show the association between PROs and time-to-event endpoints, and to evaluate and interpret differences between treatment arms. We developed a Bayesian joint model where latent variables link the repeatedly measured PROs and time-to-event outcomes. We applied the model to a randomized oncology clinical trial, and compared our results to those obtained from the standard methods, which analyze PROs and time-to-event endpoints separately. Our results provided better understanding of the treatment effects through their associations with both PROs and time-to-event outcomes.