Freshly isolated hepatocytes are currently regarded as the most superior in vitro model for use in prediction studies, in particular to provide estimates of in vivo intrinsic clearance (CL(int)). However, due to their loss of viability within 4 h and a decrease in cytochrome P450-dependent metabolism upon culture, newer cellular models are being developed. Cryopreserved hepatocytes have several potential advantages, but to date evaluation of the utility of this model for estimating in vitro CL(int) has been limited to the substrate depletion approach. We have incubated eight compounds with suspensions of freshly isolated and cryopreserved rat hepatocytes and obtained in vitro CL(int) via metabolite formation kinetics (for 14 pathways). A substantial range of in vitro CL(int) values (0.1-98 microl/min/10(6)cells) was obtained in both models, and the freshly isolated suspension data were in good agreement with the literature. Cryopreserved suspensions were able to give a comparable estimation (within 2-fold) of in vitro CL(int) to fresh cells for six pathways, namely tolbutamide, three diazepam metabolites, propranolol, and 7-hydroxylation of warfarin. A higher estimation of in vitro CL(int) was obtained for the three other metabolites of warfarin due to a decrease in the K(M) values. Lower estimations of in vitro CL(int) were observed for four compounds (six pathways), and this was particularly pronounced (4-16%) for pathways showing atypical Michaelis-Menten kinetic profiles (dextromethorphan, nordiazepam) but less so (25-45%) for pathways showing biphasic Michaelis-Menten kinetics (7-ethoxycoumarin and phenytoin).