On the establishment of equivalence acceptance criterion in analytical similarity assessment

For the assessment of biosimilarity of biosimilar products, the United States (US) Food and Drug Administration (FDA) proposed a stepwise approach for providing the totality-of-the-evidence of similarity between a proposed biosimilar product and a US-licensed (reference) product. The stepwise approach starts with the assessment of critical quality attributes (CQAs) that are relevant to clinical outcomes in structural and functional characterization in the manufacturing process of the proposed biosimilar product. FDA suggests that these critical quality relevant attributes be identified and classified into three tiers depending on their criticality or risk ranking. To assist the sponsors, FDA also suggests some statistical approaches for the assessment of analytical similarity for CQAs from different tiers, namely equivalence test for Tier 1, quality range approach for Tier 2, and descriptive raw data and graphical comparison for Tier 3. Analytical similarity assessment for CQAs in Tier 1 is performed based on the equivalence acceptance criterion (EAC), which depends upon the estimate of variability of the reference product. The FDA's recommended approach often underestimates the variability of the reference product because it does not take the worst possible lots into consideration. In this article, we examine the statistical properties of the FDA's recommended approach and proposed alternative methods in establishing an alternative approach under the scenario where multiple samples drew from each lot.