Sample Size Determination for Equivalence Assessment

Abstract

Equivalence assessment between reference and test treatment is often conducted by two one-sided tests (TOST). The nature of two one-sided tests leads to the issue of sample size determination with multiple comparison. If the drug product is formulated through concentration in blood system, a bioavailability study is required to show bioequivalence in two parameters, AUC (Area under the curve) and Cmax (maximum concentration). It further leads to a multiple comparison with two correlated endpoints. The corresponding power function and sample size determination can be derived from a joint distribution of the sample means and sample variances. This is the first part of the presentation based on a new article to appear in JBS 2015 (written by Sun, Dong and Tsong). For a non-systemic drug product, a three-arm (placebo, test and reference) three-stage parallel trial needs to be conducted to show assay validation of reference product, efficacy of test product and equivalence of test and reference products. The sample size and power is therefore derived based on the joint distribution of sample means and variances of the three tests. Chang et al (to appear in JBS 2015, by Chang, Tsong, Dong and Zhao) investigate the problem with the assumption that when the reference product has assay sensitivity established before designing the trial. The second part of the presentation is of the results of Chang, Tsong, Dong & Zhao research.