Strategy for Similarity Margin Selection in Comparative Clinical Biosimilar Studies

• Activity Number: 612 - Statistics for Clinical Trials and Medical Research
• Type: Contributed
• Date/Time: Thursday, August 1, 2019 : 8:30 AM to 10:20 AM
• Sponsor: ENAR
• Author(s): Mengdie Yuan* and Yabo Niu and Lei Nie and Thomas Gwise and Gregory Levin and Shein-Chung Chow
• Companies: Food and Drug Administration and Texas A&M University and FDA and FDA and FDA and FDA
• Keywords: clinical similarity study; similarity margin; non-inferiority margin; type I error; power; two one-sided tests

Abstract

The selection of a clinical similarity margin has always been challenging and critical in testing similarity between a proposed biosimilar product and its reference drug product. For comparative biosimilar clinical studies, a scientific justification based on clinical knowledge about the reference product, usually obtained from historical studies, is commonly considered to establish an appropriate similarity margin. In practice, different similarity margins are sometimes proposed by FDA and the sponsors. In this case, the sponsors are usually asked to provide scientific rationale or justification for a wider margin proposed. In this article, we propose conducting a benefit and risk analysis to facilitate the communication between the sponsor and the FDA in making final decision for margin selection. We define a new "risk" factor which serves as a bridge between the sponsor's proposed margin and the expected drug approval rate under FDA recommended margin. The performance of the proposed strategy is evaluated via extensive clinical trial simulation for various scenarios. Concluding remarks are given following the simulation results.