Virtual Bioequivalence Assessment of Long-acting Injectable Suspensions Using PBPK Modeling: Part 2. Type 1 Error and Safe space Analyses

We present an analysis of statistical type 1 error and safe space calculations in virtual bioequivalence (VBE) assessments using a previously published PBPK model (1) for 3-month long-acting paliperidone palmitate injectable suspensions. The type 1 error for the two-one-sided t test (TOST) applied to virtual parallel design bioequivalence (BE) trials was estimated through calculation of the true BE region for mean drug particle radius (the formulation critical quality attribute (CQA)). Monte Carlo simulations were then used to combine these limits with power calculations to display estimates of the safe space for BE from a predefined particle radius. Type 1 error for detecting formulation difference in the model was controlled at 5% for pharmacokinetic endpoints. The true bioequivalence region for 3-month PP LAI suspension mean particle radius extended over 5 micrometers, but acceptable statistical power (≥80%) was obtained only when the mean particle radii was within 1 micrometer of the reference formulation. For PBPK models, type 1 error calculations are notably more complex than power calculations because the true bioequivalence limits for CQAs need to be determined before error assessment. This study appears to be the first to discuss the intersection of type Irror control and safe space estimation in PBPK modeling for BE assessment. Our case study shows that the VBE assessment controlled type 1 error well. Safe space is shown to depend on both formulation characteristics and the statistical power afforded by BE studies, offering valuable insights for formulation design considerations.