Mixed Effects Historical Varying Coefficient Model for Evaluating Dose Response in Flexible Dose Trials

• Activity Number: 428
• Type: Contributed
• Date/Time: Tuesday, August 6, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Toshihiro Misumi*+ and Sadanori Konishi
• Companies: Astellas Pharma Inc. and Chuo University
• Keywords: Varying coefficient model ; Dose response ; Flexible dose trial ; Longitudinal data ; Basis expansion ; Mixed effects model

Abstract

Varying coefficient models provide a useful tool to explore the dynamic pattern in various fields of science, such as epidemiology, medical research, ecology, and so on. Crucial issues arise in assessing dose response from flexible dose clinical trials; a current response is affected by not only the current dose level but also past dose levels, that is, there is a time lag in the effectiveness of treatment, and also there is considerable variability between subjects. In order to overcome these issues, we propose a novel nonlinear varying coefficient model for estimating dose response curves in longitudinal flexible dose trial, called a mixed effects historical varying coefficient model (ME-HVCM). It enables us to describe the historical effectiveness curves and subject specific curves. Unknown parameters included in the ME-HVCM are estimated by the regularization methods along with the EM algorithm. Monte Carlo simulations are conducted to investigate the effectiveness of ME-HVCM for evaluating dose response in flexible dose trials.