Mechanistic bottom-up estimation of passive drug absorption from the gastrointestinal tract: Comparison among primary cultured human intestinal cells, Caco-2 cells, artificial membrane, and animal scale-up

Objective: The fraction of drug absorbed (Fa) from the intestine is an important parameter to characterize the pharmacokinetics of a drug. We aimed to search for an experimental system that provides the best parameters for estimating the effective permeability (P_eff) used for the bottom-up prediction of Fa. Materials and methods: The absorption kinetics of 12 passively absorbed drugs were simulated by a compartment absorption transit (CAT) model using absorption parameters from four different experimental systems: human intestinal epithelial cell (HIEC) monolayer, Caco-2 monolayer, parallel artificial membrane permeability assay (PAMPA), and in situ rat intestinal perfusion. All absorption parameters were obtained from the literature. The in vitro apparent permeability coefficient (P_app) and rat in situ P_eff were converted to human P_eff using a bottom-up approach for each region, based on the morphological features of the human intestine. The simulated Fa values were compared to the respective observed values. Furthermore, plasma concentration profiles of the drugs were simulated by convolution using the time-course of the absorption rate simulated using the P_eff values calculated from the HIEC P_app. Results: The Fa values were best predicted by using the P_eff values calculated from HEIC, within a 1.3-fold range of observed Fa in 11 out of 12 drugs. The simulated C_max values of pharmacokinetic simulation using HIEC P_app fell within a 1.5-fold range of observed values for all the drugs examined. Conclusion: The HIEC monolayer was identified as the most suitable permeation parameter for estimating Fa and C_max using a morphological feature-based bottom-up approach.