Predicting the Release Mechanism of Amorphous Solid Dispersions: A Combination of Thermodynamic Modeling and In Silico Molecular Simulation.

Abstract

Background: During the dissolution of amorphous solid dispersion (ASD) formulations, the drug load (DL) often impacts the release mechanism and the occurrence of loss of release (LoR). The ASD/water interfacial gel layer and its specific phase behavior in connection with DL strongly dictate the release mechanism and LoR of ASDs, as reported in the literature. Thermodynamically driven liquid-liquid phase separation (LLPS) and/or drug crystallization at the interface are the key phase transformations that drive LoR.

Methods: In this study, a combination of Perturbed-Chain Statistical Associating Fluid Theory (PC-SAFT) thermodynamic modeling and in silico molecular simulation was applied to investigate the release mechanism and the occurrence LoR of an ASD formulation consisting of ritonavir as the active pharmaceutical ingredient (API) and the polymer, polyvinylpyrrolidone-co-vinyl acetate (PVPVA64). A thermodynamically modeled ternary phase diagram of ritonavir (PVPVA64) and water was applied to predict DL-dependent LLPS in the ASD/water interfacial gel layer. Microscopic Erosion Time Testing (METT) was used to experimentally validate the phase diagram predictions. Additionally, in silico molecular simulation was applied to provide further insights into the phase separation, the release mechanism, and aggregation behavior on a molecular level.

Results: Thermodynamic modeling, molecular simulation, and experimental results were consistent and complementary, providing evidence that ASD/water interactions and phase separation are essential factors driving the dissolution behavior and LoR at 40 wt% DL of the investigated ritonavir/PVPVA64 ASD system, consistent with previous studies.

Conclusions: This study provides insights into the potential of blending thermodynamic modeling, molecular simulation, and experimental research to comprehensively understand ASD formulations. Such a combined approach can be leveraged as a computational framework to gain insights into the ASD dissolution mechanism, thereby facilitating in silico screening, designing, and optimization of formulations with the benefit of significantly reducing the number of experimental tests.