Statistical Analysis of Sparse Bioequivalence Study with Pharmacokinetics Endpoints

• Activity Number: 30
• Type: Contributed
• Date/Time: Sunday, August 9, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Government Statistics Section
• Author(s): Guoying Sun* and Huaixiang Li and Fairouz Makhlouf and Donald Schuirmann
• Companies: FDA and FDA and FDA and FDA
• Keywords: bioequivalence ; confidence interval ; sparse design ; pharmacokinetic ; bias corrected and accelerated

Abstract

A sparse sample design is used for some pharmacokinetic (PK) bioequivalence studies with ophthalmic products or other areas. The concentration sample can only be collected at one time point from each subject. Therefore, we can only obtain one value (averaged over subjects) for each of the PK parameters - Area under the Curve (AUC) at time t, Cmax, and Tmax from the study. The 90% confidence intervals for the ratios of mean PK parameters for these sparse sample designs are constructed differently from traditional PK studies. Several different bootstrap methods, including the bias corrected and accelerated (BCa) method, and different stratified bootstrap strategies are used to re-sample subjects for crossover sparse design to calculate the confidence intervals. A simulation study will be conducted to compare different 90% CI methods for bioequivalence tests for these sparse sampling PK studies.