Abstract

Drug product dissolution for four batches of acalabrutinib 100 mg capsules were analyzed with in vitro dissolution in various pH conditions and in media containing synthetic surfactant micelles or biorelevant micelles. Non-sink conditions, where the drug is unionized, were used to derive a batch specific drug product particle size distribution (P-PSD). The purpose of this P-PSD is to serve as an input in physiological based pharmacokinetic (PBPK) models to calculate in vivo dissolution in various administration conditions. The P-PSD was used to predict dissolution in all other conditions tested, introducing a different Unstirred Water Layer (UWL) thickness for free- and micelle-bound drug and the calculation of surface solubility using a theoretical model. With the proposed P-PSD approach and proposed model inputs, percent dissolved at all time points and for all conditions and batches were adequately anticipated with an 11% overprediction. In contrast, the use of drug substance laser diffraction particle size data with equivalent inputs to the models led to an underprediction of observed percent dissolved by 31% overall. Finally, the use of bulk solubility instead of surface solubility led to an overall 48% overprediction of the dissolution data. Batch specific P-PSD were used to predict in vivo dissolution of acalabrutinib drug products with PBPK models. The current limitations of PBPK models for integration of in vitro dissolution are also discussed and improvements are suggested to improve future predictions.