ABSTRACT
This chapter outlines the method of pairwise comparisons for analytical similarity assessment with multiple references. For critical quality attributes in Tier 1, Food and Drug Administration recommends that an equivalent test can be performed to assess analytical similarity. The statistical analysis methods proposed by S. H. Kang and Shein-Chung Chow consider the relative distance based on the absolute mean differences. One of the most commonly used designs for assessing biosimilarity between a biosimilar product and an innovative biological product is probably a two-arm balanced design, and biosimilarity is assessed by using the difference between two population means. The chapter describes the simultaneous confidence interval approach as suggested by the Oncologic Drugs Advisory Committee. It includes a simulation study for evaluation of relative performances between the method of pairwise comparisons and the simultaneous confidence interval approach for various scenarios of different reference products.
