Efficiency Enhancement in Testing Treatment Efficacy Across Multiple Populations Using Treatment Crossover Data.

Abstract

Recent advances in biotechnology and personalized medicine have driven the development of efficient clinical trial methodologies for assessing treatment efficacy across multiple populations defined by treatment effect modifiers. Within-patient comparison of different treatments is a promising approach for improving study efficiency across multiple populations by eliminating between-patient variability in treatment evaluation. This study provides a framework for evaluating treatment efficacy in multiple populations for <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mn>2</mml:mn> <mml:mo>&#xd7;</mml:mo> <mml:mn>2</mml:mn></mml:mrow> <mml:annotation>$$ 2\times 2 $$</mml:annotation></mml:semantics> </mml:math> crossover trials and evaluates the efficacy gain in comparison with the standard parallel-group analysis. Simulation experiments confirm that the crossover analysis consistently outperforms the parallel-group analysis in statistical power, especially when carryover effects are small. An application to a clinical trial in Type 2 diabetes demonstrates the efficiency advantages of the crossover analysis. These numerical results emphasize the potential of the crossover analysis for enhancing the efficiency of clinical development of personalized medicine.