B1521
Title: Complex assurance considerations when designing biosimilar trials
Authors: Arne Ring - Hexal (Germany) [presenting]
Din Chen - University of Pretoria (South Africa)
Rachid El-Galta - Hexal-Sandoz (Germany)
Abstract: To demonstrate biosimilarity of a new test (T) product with an existing biologic reference (R) drug, it is generally required to perform a phase I trial in healthy volunteers to compare the pharmacokinetic properties and a phase III trial in patients to compare efficacy. These trials will be successful if the T/R ratio of their (co-) primary endpoints is statistically demonstrated to be within pre-defined margins. In the phase I trial, one test product is typically compared against two reference formulations. The co-primary endpoints are Cmax and the Area under the Curve (AUC). The pharmacokinetics of the reference product have been investigated, and it is assumed that an estimate of the between-subject coefficient of variation (CV) of the endpoints is known. Hence, using these estimates to define a prior distribution to investigate the assurance for each equivalence test is possible. The total assurance for all comparisons and all primary endpoints also needs to consider the respective correlations. The design considerations are demonstrated using an RShiny App that combines the available information to provide total assurance of the trial.
