Use of a two‐sided tolerance interval in the design and evaluation of biosimilarity in clinical studies

Abstract

SummaryIn assessing biosimilarity between two products, the question to ask is always “How similar is similar?” Traditionally, the equivalence of the means between products is the primary consideration in a clinical trial. This study suggests an alternative assessment for testing a certain percentage of the population of differences lying within a prespecified interval. In doing so, the accuracy and precision are assessed simultaneously by judging whether a two‐sided tolerance interval falls within a prespecified acceptance range. We further derive an asymptotic distribution of the tolerance limits to determine the sample size for achieving a targeted level of power. Our numerical study shows that the proposed two‐sided tolerance interval test controls the type I error rate and provides sufficient power. A real example is presented to illustrate our proposed approach.

Tolerance interval testing for assessing accuracy and precision simultaneously

“…In this study, a two-sided tolerance interval test is considered. As pointed out by Chiang et al [4], there must be two unknown parameters θ L and θ U leading to P(θ L <X<θ U )�γ. Therefore, when being linked to the prespecified acceptable interval, one of the following four situations must be true: (i) c L �θ L and θ U �c U , which is what we expect; (ii) θ L �c L and θ U �c U , which must indicate that the expectation of X has a negative bias from our expectation; (iii) c L �θ L and c U �θ U , which must indicate that the expectation of X has a positive bias from our expectation; and (iv) θ L <c L and c U <θ U , which must indicate that the variability of X exceeds what we expected.…”