Two-Stage Adaptive Design for Case-Control Genetic Association Studies

• Activity Number: 123
• Type: Topic Contributed
• Date/Time: Monday, August 2, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Lihan Yan*+
• Companies: FDA
• Keywords: adaptive ; model selection ; genetic model ; MAX ; HWDTT

Abstract

Because of the uncertainties during design stage, an adaptive design becomes a popular solution for adjustment of design parameters based on interim study data. In many genetic studies for rare diseases, the genetic model hence the employment of the optimal test statistic is often uncertain. In present study, we propose two adaptive approaches in a two-stage case-control association study that allow the study to select the genetic model and modify the test statistic based on interim data. The first approach (the MAX method) uses the genetic model that results in the highest test statistic. The second approach selects the genetic model based on the direction of magnitude of the Hardy-Weinberg disequilibrium test statistic (the HWDTT method). Simulations are performed to provide the operating characteristics and compare them among various choices of adaptive and non-adaptive designs.