Abstract

The majority of drug candidates exhibit weakly basic characteristics with high lipophilicity. The risk of intraluminal compound precipitation has been studied in vivo and extensively in vitro using advanced dissolution transfer setups mimicking drug transfer from the stomach to the small intestine. The present investigation aims to evaluate the usefulness of the recently introduced Artificial Stomach-Duodenum in silico tool in the DDDPlusTM platform (ASD-D+) to simulate intraluminal drug behavior. The weakly basic drugs ketoconazole and dipyridamole were used as model drugs within the ASD-D+ model at two dose levels. The simulated amounts per volume were compared to intraluminal data collected from fasted healthy adults. Four different in silico transfer models running on a continuous or a stepwise mode were utilized for the simulations. Each transfer model exhibited different capabilities to simulate observed intraluminal drug presence. Three out of the four in silico models overestimated the total drug amount measured in vivo (dissolved and precipitated drug), while only two of the four models matched the intraluminal drug concentrations. The stepwise model enabled adequate simulations of both drug concentration and total drug amount. The present investigation highlighted the importance of simulating drug transfer appropriately within the applied methodology prior to estimating precipitation kinetics. As a future step, optimization of ASD-D+ model would enable evaluations of solid/semi-solid dosage form simulations. Lastly, prediction of drug precipitation kinetics following simulation of gastrointestinal transfer may provide mechanistic understanding of drug absorption and appropriate justification of drug-formulated parameters within physiologically based pharmacokinetic models.

By Marina Statelova, Maria Vertzoni & Alexandros Kourentas