On sample size requirement for analytical similarity assessment.

For approval of biosimilar products, the U.S. Food and Drug Administration (FDA) recommends a stepwise approach for obtaining the totality-of-the-evidence for assessing biosimilarirty between a proposed biosimilar product and its corresponding innovative (reference) biologic product. The stepwise approach starts with the assessment of analytical similarity of critical quality attributes (CQAs) for structural/physicochemical and functional properties in the manufacturing process of biosimilar products. For Tier 1 CQAs which are most relevant to clinical outcomes, the FDA recommends an equivalence test be performed for similarity assessment based on an equivalence acceptance criterion (EAC). While performing the equivalence test, sample size is a critical component of the equivalence test. This article focuses on the discussion of the FDA's proposal: select an appropriate sample size by adjusting EAC margin and variability ([Formula: see text]). The article provides a thorough discussion on the FDA's proposal; sample size requirement under different scenarios are briefly described and a numerical study which compares sample size requirement under various combinations of study parameters is conducted.

Duke Scholars

Author Shein-Chung Chow Biostatistics & Bioinformatics, Division of Biostatistics