A Mechanistic Model for Predicting the Physical Stability of Amorphous Solid Dispersions

In this paper, we establish a mechanistic model for the prediction of amorphous solid dispersion (ASD) stability. The novel approach incorporates fundamental physical parameters, principally supersaturation, diffusivity, and interfacial energy, to model crystallization in ASDs accounting for both kinetic and thermodynamic drivers. API dependent decoupling coefficients were also considered which allowed dynamic mechanical analysis to probe molecular mobility, with viscosity measurements, across an exceptionally broad range of temperatures to support ASD stability simulations. ASDs are multicomponent systems in which the amorphous form of active pharmaceutical ingredients (APIs) are molecularly dispersed within a carrier. This gives rise to a transiently supersaturated API solution upon dissolution which increases the driving force for oral absorption and results in increased bioavailability as compared to that of the crystalline API. A major shortcoming of ASDs, however, is that there is the potential for amorphous APIs to revert to their more stable crystalline form during storage, despite the use of polymer carriers to stabilize formulations and limit recrystallization. Hot melt extrusion (HME) has been employed as the preparation method for ASDs used in this study as it is well-suited for the formation of uniform dispersions. The ASDs were stored under controlled temperature conditions, in the absence of humidity, to determine recrystallization kinetics. Our mechanistic model, considering both crystal nucleation and growth processes, describes temporal ASD stability through a system of coupled differential equations that connect the physiochemical properties of the ASD system to drug recrystallization. The model and prolonged time scale of crystallization observed highlight the importance of considering both thermodynamic and kinetic factors in the preparation of stable ASDs. Experimental observations were found to be in good agreement with predictions of the model confirming its utility in predicting the temporal physical stability of amorphous solid dispersions through a mechanistic lens.