Development of Generic Equivalence Margins for Early-Stage Implementation of Equivalence Testing for Potency Assays

Abstract

Relative potency assays for biological therapeutics require statistical evaluation to demonstrate similarity between the dose-response curves of a reference standard and the test samples. We developed an equivalence testing approach that can be utilized for the complete potency assay life cycle, from early development until commercialization. This approach was based on the use of generic equivalence margins to enable equivalence testing at the beginning of assay development, when the body of assay-specific data is still very limited. Generic equivalence margins for equivalence testing of four-parameter logistic curve fits were established for bioassays and binding assays spanning a variety of designs, formats, and read-outs. We also established that equivalence testing using ratios of the reference standard and test sample was superior to equivalence testing using absolute differences. Based on a large body of historical data, generic equivalence margins were determined for the curve upper asymptote, slope, and dynamic range. Furthermore, we developed a road map to guide the implementation of generic or assay-specific margins to ensure the appropriate data analysis approach is being applied during the assay life cycle.