Successful bioequivalence prediction of immediate-release vortioxetine tablets through prospective virtual crossover...

We prospectively conducted virtual bioequivalence (VBE) trials under fasted conditions to guide a clinical trial design and candidate selection for a model drug with a long elimination half-life. We combined biopredictive dissolution methods that mimic gastric motility with innovative biopharmaceutical modeling, enabling suitability checks of crossover and parallel study layouts. Three batches of immediate-release tablets containing 20 mg of vortioxetine were tested against a reference product. Subsequent VBE trials combined a semi-mechanistic biopharmaceutics model with an empirical population approach to describe drug disposition. Parallel VBE included between-subject variability, while the crossover design added interoccasion variability (IOV) at 5% to 15% for the primary PK parameters. Ten trial series were simulated across different sample sizes (12-48), with and without unexplained residual error. Crossover VBE predicted 100% bioequivalence probability for all candidate batches, regardless of the sample size or IOV. In parallel VBE, a maximum bioequivalence probability of 80% was achieved, with at least 60 subjects (30 per arm) required to attain a success probability rate of ≥50%. The selected formulation was bioequivalent with the reference product (90% CI: 97.77-106.05% for C_max and 96.52-104.32% for AUC_0-72h, n = 36). Proposed workflow provides a realistic approximation of bioequivalence study outcome.