The development of a tool that can provide early-stage predictive toxicology data may accelerate the identification of safer drug candidates, and thereby improve the clinical progression of drug candidates to pharmaceuticals. Such a system would require an accurate and reliable technique that is amenable to the large number of drug candidates that must be screened in the lead discovery and optimization stages of drug development. A key component of predictive toxicology is the ability to harness the metabolite-generating capacity of human cytochromes P450, which are involved in first-pass drug metabolism function of the liver. We have miniaturized P450 catalysis into a microarray format consisting of up to 11,200 isolated P450 reactions, each in 5 nL sol-gel spots, on a single functionalized glass microscope-size biochip. This dramatic scale down from more conventional 96 and 384-well plate scales (at least a 1000-fold reduction in volume) did not adversely affect P450 catalytic activity. Based on the functionality of the P450-containing microarray, we developed the metabolizing enzyme toxicology assay Chip (MetaChip), which combines high-throughput P450 catalysis with cell-based screening on a microscale platform. Proof of concept was demonstrated using anticancer prodrugs cyclophosphamide and Tegafur, as well as the analgesic acetaminophen. The MetaChip may provide a high-throughput microscale alternative to currently used in vitro methods for human metabolism and toxicology screening.