Prediction of hepatic clearance and availability by cryopreserved human hepatocytes: an application of serum incubation method.

A novel and convenient method was established for the prediction of in vivo metabolic clearance in human liver. The present method applied the in vitro-in vivo extrapolation paradigm previously established in rats to the in vitro data obtained from cryopreserved human hepatocytes. Predicted hepatic availability and clearance were compared with the reported oral bioavailability and plasma clearance in humans for 14 clinically used drugs (naloxone, buspirone, verapamil, lidocaine, imipramine, metoprolol, timolol, antipyrine, diazepam, quinidine, caffeine, propranolol, diclofenac, and phenacetin). A large interindividual variation was observed in the intrinsic metabolic clearance among separate cryopreserved preparations from different subjects. The prediction generally resulted in a marked underestimation when the biologically based scaling factor (3.1 x 10(9) cells/kg) was used for the extrapolation of in vitro data (milliliters per minutes per cells) to in vivo value (milliliters per minutes per kilograms). Reasonably good in vitro-in vivo correlations were obtained with empirically calculated scaling factors, 8.5 x 10(9) (cells/kg) from 10 individual preparations and 10.8 x 10(9) (cells/kg) from pooled preparation of two selected lots, which were 3- to 4-fold larger than the biologically based scaling factor. These data suggested that the calibration of inherent interindividual variation of metabolic activities among different cryopreserved preparations of human hepatocytes to obtain the empirical scaling factor, which is applicable only to the preparation used, was an essential step for more reliable and quantitative prediction of in vivo metabolic activity in humans.