Comparison of in vitro preparations for semi-quantitative prediction of in vivo drug metabolism

Various in vitro preparations were compared with respect to their ability to mimic in vivo metabolism. For this purpose, S9-liver homogenate, microsomes, cryopreserved hepatocytes, cryopreserved liver slices and fresh liver, lung, kidney, and intestinal slices were incubated with three drugs in development, which are metabolized in vivo by a wide range of biotransformation pathways. Metabolites were identified and quantified with liquid chromatography-mass spectometry/UV from the in vitro incubations and compared with metabolite patterns in feces, urine, and bile of dosed rats. In vitro systems with intact liver cells produced the same metabolites as the rat in vivo and are a valuable tool to study drug metabolism. Phase I metabolites were almost all conjugated in intact cells, whereas S9-homogenate only conjugated by sulfation and N-acetylation. Microsomes and S9-homogenate are useful to study phase I metabolism but not for the prediction of in vivo metabolism. Extra-hepatic organ slices did not form any metabolites that were not produced by liver cells, but the relative amounts of the various metabolites differed considerably. Small intestinal slices were more active than liver slices in the formation of the N-glucuronide of compound C, which is the major metabolite in vivo. When the relative contribution of liver and small intestinal slices to the metabolism of this compound was taken into account, it appeared that the in vivo metabolite pattern could be well predicted. Results indicate that for adequate prediction of in vivo metabolism, fresh or cryopreserved liver slices or hepatocytes in combination with slices of the small intestines should be used.