Quantitative Prediction of ln Vivo Drug-drug Interactions from ln Vitro Data: Effects of Active Transport of Inhibitors into the Liver

Shin-ichi Kanamitsu1, Yutaka Shinozaki2, Kiyomi Ito3, Takafumi Iwatsubo4, Hiroshi Suzuki and Yuichi Sugiyama

Graduate School of Pharmaceutical Sciences, University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan.
1Current address: Section of Drug Metabolism Research, Naruto Research Institute, Otsuka Pharmaceutical Factory, Inc.
2Current address: Central Research Laboratories, Zeria Phannaceutical Co., Ltd.
3Current address: School of Pharmaceutical Sciences, Kitasato University
4Current address: Drug Metabolism Laboratories, Institute for Drug Development Research, Yamanouchi Pharmaceutical Co., Ltd.

Correspondence:Yuichi Sugiyama,Graduate School of Pharmaceutical Sciences, University of Tokyo,7-3-1 Hongo,Bunkyo-ku,Tokyo l13-0033,Japan.
Tel:+81-3-5689-9094 Fax:+81-3-5800-6949

Original paper :AATEX 7(1):23-29,2000

The degree of in vivo drug-drug interactions caused by competitive or non- competitive inhibition of drug metabolism can be predicted using the in vitro inhibition constant (Ki) and the unbound concentration of inhibitor in the liver (Iu). Although it can be assumed for most inhibitors that the value of Iu is equal to the unbound concentration in the liver capillary (sinusoid) (Iout,u), Iu is larger than Iout,u if the inhibitor is actively taken up by the liver. In the present study, the possibility of active transport of inhibitors into the liver was evaluated using isolated rat hepatocytes. An uptake study using FCCP, an 4TP-depletor, showed that unbound quinidine, erythromycin, sulfaphenazole, ketoconazole, and omeprazole are concentrated 2.2-, 1.4-, 1.2-, 1.2-, and 1.0- fold. respectively, in hepatocytes due to active transport. This value of the unbound concentration ratio of each inhibitor was multiplied by the unbound concentration at the inlet to the liver (Iin,u) to estimate the maximum value of Iu, and the in vivo increase in the AUC of the corresponding substrates(sparteine, cyclosporine, tolbutamide, terfenadine, and diazepam, respectively) was predicted based on the Iu/Ki ratio. Because none of the investigated inhibitors was found to be highly concentrated in the liver, the predicted in vivo interaction was not greatly affected by taking account of active transport of the inhibitor.

Keywords: drug interaction, active transport, isolated rat hepatocyte