Physical stability of hydroxypropyl methylcellulose-based amorphous solid dispersions: Experimental and computational study

The preparation of an amorphous solid dispersion (ASD) is a promising strategy for improving the poor oral bioavailability of many active pharmaceutical ingredients (APIs). However, poor predictability of ASD long-term physical stability remains a prevalent problem. The purpose of this study was to evaluate and compare the predictive performance of selected models concerning solid–liquid equilibrium (SLE) curve and glass-transition temperature (Tg) line modeling of ibuprofen (IBU) in cellulosic polymers (i.e., hydroxypropyl methylcellulose (HPMC) and hydroxypropyl methylcellulose acetate succinate (HPMCAS)). For SLE curve modeling, an empirical analytical approach (Kyeremateng et al., 2014) and the Perturbed-Chain Statistical Associating Fluid Theory (PC-SAFT) equation of state (EOS) were chosen. Due to the unavailability of PC-SAFT parameters for both polymers, an approximation procedure for parametrization was applied. The Gordon–Taylor equation and Kwei equation were considered for Tg line determination. The impact of various computational set-ups (e.g., model parametrization or extrapolation length) on IBU solubility prediction at storage conditions was thoroughly investigated, assessed and confronted with the results from an 18-month physical stability study. IBU developed stable 20 wt% API content ASDs with both HPMC and HPMCAS. The extrapolation behavior and subsequent ASD thermodynamic stability prediction at storage conditions deduced from the aforementioned models were significantly different. Overall, the PC-SAFT EOS predicted higher IBU solubility in both polymers and, thus, a lower recrystallization tendency when compared to the empirical analytical approach. At higher IBU concentrations, liquid-liquid demixing in IBU–polymer systems was predicted by the PC-SAFT EOS, which was in qualitative disagreement with experimental observation.