Development of Statistical Methods for Analytical Similarity Assessment

Abstract

To evaluate the analytical similarity between the proposed biosimilar product and the US-licensed reference product, U.S. Food and Drug Administration (FDA) statisticians collaborated with Chemistry, Manufacturing and Control (CMC) scientists at FDA in order to develop a three-tier approach. The proposed tiered approach starts with a criticality determination of quality attributes (QAs) based on their potential impact on product quality and the clinical outcomes. Those QAs characterize the biological product in terms of structural, physico-chemical, and functional properties. Then, the QAs are assigned into three tiers based on their criticality ranking. To evaluate the analytical similarity for QAs assigned to different tiers, we recommend different statistical approaches with different statistical rigors. That is, we recommend an equivalence test for the critical quality attributes (CQAs) in Tier 1, a quality range approach for QAs in Tier 2, and a side-by-side graphic comparison approach for QAs in Tier 3. In this article, we mainly focus on the development of the FDA's recommended equivalence test for Tier 1. We also provide some discussion on the statistical challenges of the proposed equivalence test in the context of analytical similarity assessment.

Statistical Primer on Biosimilar Clinical Development

“…4,6 This is done using 2 simultaneous 1-sided tests to analyze the composite null and alternative hypotheses (H 0 and H 1 , respectively), which are: H 0 : μ ≤ θ L or μ ≥ θ U (the treatment difference is outside the equivalence margin) and H 1 : θ L < μ < θ U (the treatment difference is within the margins), where θ L and θ U are the lower and upper margins, respectively, and μ is the analysis criterion (eg, mean ratio or mean difference). 4,9,10 This is done with 2 sets of hypotheses (H a0 : μ ≤ θ L with H a1 : μ > θ L and H b0 : μ ≥ θ U with H b1 : μ < θ U ). 4,9 The H 0 is rejected in favor of H 1 if the CI for μ is contained completely within the range (θ L –θ U ).…”

“…3 Instead, biosimilars are evaluated in a stepwise fashion that includes structural and functional studies, nonclinical assessments of pharmacokinetics (PK) and toxicity, and clinical evaluation of PK, efficacy, and safety (including immunogenicity) to demonstrate similarity of the potential biosimilar to the originator biologic. 1–4 Approval of a potential biosimilar is based on the totality of the evidence in demonstrating similarity of the biosimilar to the reference product. 1–4 Totality of the evidence refers to the extensive comparative structural, functional, nonclinical, and clinical data required to establish biosimilarity.…”

“…1–4 Approval of a potential biosimilar is based on the totality of the evidence in demonstrating similarity of the biosimilar to the reference product. 1–4 Totality of the evidence refers to the extensive comparative structural, functional, nonclinical, and clinical data required to establish biosimilarity. Totality of the evidence is an important concept for biosimilars because of the different paradigm for developing and regulating biosimilar products.…”

Analytical Similarity Assessment in Biosimilar Studies

“…This has raised a number of debatable (controversial) issues in analytical similarity assessment (see also (6,7)). These issues include, but are not limited to, (1) fundamental similarity assumption, (2) primary assumptions for tiered approach, (3) statistical properties of FDA's recommended Tier 1 equivalence test, (4) criticism of fixed approach for margin/range selection, (5) inconsistencies between tired approaches, (6) sample size requirement, (7) heterogeneity within lots and across lots within and between test product and reference product, (8) interpretation of FDA's current thinking on scientific input, (9) relationship between similarity limit and variability, and (10) a proposed unified tiered approach.…”

“…For the analytical studies, FDA suggests that CQAs should be identified and classified into three tiers according to their criticality or risk ranking based on mechanism of action (MOA) or PK using appropriate statistical models or methods. CQAs with CQAs that are most relevant to clinical outcomes will be classified to Tier 1, while CQAs that are less (mild-tomoderate) or least relevant to clinical outcomes will be classified to Tier 2 and Tier 3, respectively (4,5).…”