(88g) Integrating Amorphous Solid Dispersions (ASDs) in a PBPK Software

Physiologically based pharmacokinetic (PBPK) software for oral drug delivery typically revolves around the intricate interplay between drug permeation, distribution, and clearance dynamics. While dissolution kinetics for crystalline active pharmaceutical ingredients (APIs) are well understood and can be captured by a reduced set of parameters, such as dissolution rate and the challenge amplifies significantly for amorphous solid dispersions (ASDs). Unlike crystalline forms, ASDs do not face limitations imposed by crystalline solubility, leading to supersaturation in the gastrointestinal tract and is often followed by a subsequent sudden or progressive crystallization of the API (“spring-and-parachute” effect), causing a reduction in the dissolved API amount. Most PBPK software have challenges handling this phenomenon accurately, mis predicting API fraction absorbed.

The distinctive dissolution behavior among different ASDs poses challenges in formulating for optimal in vivo performance. PBPK modeling becomes crucial to avoid inaccurate decisions based solely on dissolution metrics like area under the curve or maximum concentration. Permeation affects ASD dissolution profiles by reducing API concentration in the bulk intestinal medium and reducing the driving force for nucleation and crystallization. Consequently, in vitro dissolution profiles may differ from in vivo dissolution and are not enough to rank and select optimal formulations. Thus, PBPK software for ASDs must grasp these complexities for an accurate formulation ranking in vivo.

Following up on this need, Open Systems Pharmacology (OPS) software suite [1], an open-source multi-compartment whole-body physiologically based pharmacokinetic model, composed of PK-Sim and Mobi tools, was used as a starting point. This software allows the customization of individual models (e.g., dissolution) and seamless integration with other software and programming languages (using Mobi). This groundwork enables comprehensive modeling of in vivo behavior, covering permeation, clearance, distribution, and excretion models for diverse subjects, with the sole absence of an accurate dissolution model for ASD’s.

To circumvent this limitation, a model was developed to captures the specific dissolution characteristics of ASDs using a diffusion-based approach to avoid limiting API release based on crystalline solubility. This strategy required considering two distinct population balances: one accounting for initial ASD powder particle size distribution and its decrease as API and excipients dissolve into the medium, and another for crystallized particle populations formed due to supersaturation-induced nucleation and growth. By resolving the intra particle diffusion of each component the impact of the drug load and excipients solubility can also be captured.

Integrating both models was simplified by leveraging Mobi's compatibility with the R programming language, eliminating the need for software recompilation or significant limitations in dissolution model implementation. A coupled approach was adopted, with the dissolution profile pre-calibrated for dissolution, nucleation, and crystallization parameters based on in vitro data being iteratively updated based on permeated drug levels, while PK-Sim handled permeation and clearance. The model bypasses additional emphasis on drug distribution among remaining body compartments, as the ASD formulation's primary effects are already addressed, enabling comprehensive understanding of formulation ranking (by exposure) and exposure.

[1] Lippert J, Burghaus R, Edginton A, et al. Open Systems Pharmacology Community-An Open Access, Open Source, Open Science Approach to Modeling and Simulation in Pharmaceutical Sciences. CPT Pharmacometrics Syst Pharmacol. 2019;8(12):878-882. doi:10.1002/psp4.12473