Advantage of the Dissolution/Permeation System for Estimating Oral Absorption of Drug Candidates in the Drug Discovery Stage

In order to increase the success rate in the development of oral drugs, an in vitro method, which can accurately estimate human oral absorption of a large variety of compounds from solid formulations, is required in the drug discovery stage. A dissolution/permeation (D/P) system is an in vitro system that simultaneously evaluates dissolution and permeation processes of drugs administered orally. In this study, we have investigated the advantages of a D/P system for use in the provisional estimation of human oral absorption of a drug (absorbed fraction, Fa) by applying it in its solid state. The D/P system mounted with a Madin-Darby canine kidney (MDCK) II cell monolayer was used to simultaneously evaluate the dissolved and the permeated amounts (% of dose) of 48 marketed drugs. Slightly modified, fasted-state simulated intestinal fluid (FaSSIFmod, 8 mL) was used as the apical medium of the D/P system. Each test drug was applied to the apical side of the D/P system as a suspension at one-hundredth of the clinical dose. The apparent permeability coefficient across the MDCK II cell monolayer was estimated in a buffer solution (pH 6.5). The octanol/water distribution coefficient (Log D) was measured at pH 6.5 by a flask shaking method. Transport medium (TM, pH 6.5), a buffer solution removing lecithin and taurocholate from FaSSIFmod, was used to determine the solubility at 24 h after applying drugs. The solubility in TM was used as a free drug concentration in FaSSIFmod. A good correlation was obtained between observed human Fa and the permeated amount in the D/P system. When the sigmoidal curve was obtained by the curve fitting to the data, the determination coefficient was R(2) = 0.79 and the 95% interval of the predicted Fa values was about ±24% for all drugs tested in the present study. For comparison, the permeated amount was calculated by multiplying the permeability of each drug (in vitro Papp) by the solubility in FaSSIFmod. However, the calculated permeated amount showed a lower correlation with the observed human Fa compared to the observed permeated amount in the D/P system. The ratio of the observed permeated amount to the calculated permeated amount was in inverse proportion to the ratio of solubility in FaSSIFmod to solubility in TM. This finding suggests the necessity of determining the free fraction of the dissolved drug in the gastrointestinal (GI) tract when calculating the human Fa. In the case of the D/P system, since applied drugs dissolved in FaSSIFmod first, and then only the free fraction permeated the membrane, an accurate estimation of the human Fa was possible from only the observed permeated amount. This in vitro system is expected to contribute to the selection of better compounds for oral use during the lead- and formulation-optimization processes in the drug discovery stage.