Abstract

The two-arm equivalence parallel design has been widely utilized to assess the degree of biosimilarity between a biological product and a reference product. The aim is to show that the difference between two population means of a primary end point is less than a prespecified equivalence margin. A well-known sample size formula for the equivalence trial is expressed as follows: H 0 : | μ T − μ R | ≥ δ   versus   H a : | μ T − μ R | < δ