Summary

The determination of bioequivalence (BE) for locally acting respiratory products has been a long-standing challenge in the pharmaceutical industry as the lung is typically difficult to analyze directly and there is no guarantee that local drug concentration is at equilibrium with systemic distribution. The rate and extent of drug absorption are usually assessed using in vivo pharmacokinetic (PK) studies in which generic and reference drug plasma concentrations are characterized and compared. A recent study by Sandoz demonstrated batch to batch PK differences between marketed batches of an FDA-approved US drug product – fluticasone propionate/salmeterol xinafoate (FP/SX) dry powder oral inhalation product, Advair Diskus 100/50.

We investigated the performance of test dry powder inhaler (DPI) formulations of tiotropium developed for the Aptar Prohaler and the reference DPI product Spiriva® HandiHaler® DPI formulations. These products were characterised using in an in-silico Regional Deposition Model (RDM) and physiologically-based pharmacokinetic simulation model (PK Simulation). The predicted pharmacokinetic profiles of both the test and reference product batch 3 were similar. Reference products batch 1, 2 and 4 had significantly lower Cmax, which maybe related to their lower predicted regional deposition.

In the present study we have highlighted that significant batch to batch variability exists in the in vitro performance of the reference product Spiriva Handihaler. We have subsequently utilised in silico modelling of the regional deposition and PK simulations to highlight that variability in the peripheral dose deposition will likely impact the PK profile of test and reference products. Once validated with actual PK data these in silico tools will be powerful in ascertaining in vitro and in vivo relationships for dry powder inhalers.