For analytical similarity assessment of a given critical quality attribute between a proposed biosimilar (test) product and an innovative (reference) biological product, FDA recommended an equivalence test with a margin of 1.5 σR (standard deviation of the reference product) be performed. The FDA recommended similarity margin has been criticized by many authors in the literature due to its inflexibility. In this article, we proposed an equivalence test with flexible margin for controlling possible inflation/deflation of the variability associated with the response. The performance of the proposed equivalence test with flexible margin is evaluated both theoretically and by means of simulation. The results indicate that flexible margin can be selected within the range of (1.575 σR, 2.025 σR) for achieving reasonable statistical assurance, for example, controlling type I error at the α=5% level of significance and 90% power for analytical similarity assessment.
Author(s): Lee SJ, Oh M, Chow SC