Computational Fluid Particle Dynamics (CFPD)-Based Virtual Next Generation Impactor (vNGI) to Predict the Aerodynamic Particle Size Distribution (APSD) of Respiratory Drug Delivery Products: Toward New Approach Methodologies (NAMs) in Inhaler Performance Evaluation

Abstract

The Next Generation Impactor (NGI) is one of the regulatory gold standards for characterizing aerodynamic particle size distributions (APSDs) of orally inhaled drug products (OIDPs); however, its reliance on complex, resource-intensive in vitro testing under tightly controlled environmental conditions limits experimental flexibility and introduces variability. In alignment with the growing regulatory emphasis on New Approach Methodologies (NAMs) for drug development, this study presents a rigorously validated computational fluid particle dynamics (CFPD)-based virtual NGI (vNGI) as an in silico method complementary to conventional testing. The vNGI replicates a significant portion of the NGI geometry and airflow physics, enabling high-resolution spatiotemporal analysis of aerosol transport and deposition mechanisms that are otherwise inaccessible experimentally. A comprehensive verification and validation framework was implemented, including mesh and particle independence studies, turbulence model assessment, and comparison of stage-wise deposition efficiencies with available in vitro data at 30 L/min. The models capabilities were further extended to low and high flow rates, and two bio-relevant mouth-throat models and polydisperse particle-laden aerosol were added. The model demonstrates strong predictive capability for a few stages and provides mechanistic insight into discrepancies in other stages, depending on the type of analysis. Importantly, this work establishes the vNGI as a fit-for-purpose according to NAM by (i) defining a clear context of use for APSD prediction and inhaler performance evaluation, (ii) capturing physically and biologically relevant air-particle interactions, and (iii) demonstrating technical robustness and reproducibility through systematic validation. The platform can potentially further enable simulation of environmental and physiological conditions, such as humidity effects, that are difficult to control experimentally, thereby improving human relevance and reducing reliance on costly and time-consuming in vitro testing. This study positions the vNGI as a scalable, regulatory-aligned NAM capable of supporting early-stage drug-device combination product development, device optimization, and an alternative bioequivalence assessment, contributing to ongoing efforts to enhance predictive performance, reduce experimental burden, and transition toward human-centric, inhalation product evaluation.