Abstract

PURPOSE:

The purpose of this study was to develop a new system for computer simulation to predict fraction absorbed (F(a)) of Biopharmaceutical Classification System (BCS) class II (low solubility-high permeability) drugs after oral administration to humans, from a miniscale dissolution test.

METHODS:

Human oral absorption of 12 lipophilic drugs was simulated theoretically by using the dissolution and permeation parameters of the drugs. A miniscale dissolution test and a solubility study were carried out in a conventional buffer and a biorelevant medium (pH 6.5). A dissolution parameter, which can simulate in vivo dissolution, was obtained from the in vitro dissolution curve. Human intestinal permeability was estimated assuming that the permeation was limited by diffusion through the unstirred water layer. The F(a) in humans was predicted and then compared with clinical data.

RESULTS:

The dissolution and solubility of most model drugs were faster and higher in a biorelevant medium than in a conventional buffer. The simulated absorption was limited by the drug dissolution rate and/or solubility. Predicted F(a) was significantly correlated with clinical data (correlation coefficient r2 = 0.82, p < 0.001) when the dissolution profiles in biorelevant medium were used for the simulation.

CONCLUSIONS:

This new system quantitatively simulated human absorption and would be beneficial for the prediction of human F(a) values for BCS class II drugs.