ABSTRACT
Recent guidance from the US Food and Drug Administration (FDA) for the evaluation of new therapies in the treatment of Type 2 diabetes mellitus (T2DM) calls for a program-wide meta-analysis of cardiovascular (CV) outcomes. In this context, we consider a Bayesian meta-analysis approach using survival regression models to assess whether the size of a clinical development program is adequate to evaluate a particular safety endpoint. The Bayesian meta-analysis trial design is general and exible, which allows us to control the type I error and power. In addition, within the Bayesian framework, historical meta-survival data can be easily incorporated into the statistical design, which leads to a substantial reduction in the sample size
CONTENTS
2.1 Introduction .................................................................................................. 14 2.2 Federal Guidelines for Bayesian Design and CV Risk Assessments ... 15 2.3 Motivating Case Study ................................................................................ 16 2.4 Meta-Regression Models for Survival Data ............................................. 18 2.5 General Methodology for Bayesian Meta-Design ................................... 20 2.6 Specication of Prior Distributions ...........................................................22
