Risk assessment of paracetamol induced liver toxicity based on human in vitro data

Currently risk assessment is based on animal experiments with limited success. The aim of this study was to explore the feasibility to replace the use of animals in risk assessment for drug-induced liver injury, by hazard identification and kinetic modeling based on human in vitro data for metabolism and toxicity. Paracetamol was used as model compound. Human hepatocytes were used to elucidate the toxic mechanisms by transcriptomics and metabolite formation. Time- and dose-dependent toxicant exposure of the human liver was modeled with a mechanism-based physiology-based toxicokinetic (PBTK) model (Simcyp®) and time and concentration- dependent metabolite production and exposure of the hepatocytes to the toxic reactive metabolite was modeled with an in vitro biokinetic model using experimentally derived and literature data on metabolism. Plasma concentration profiles were adequately modeled and the effect of individual variability in metabolism was assessed. Phase 2 metabolism as observed in human hepatocytes was well predicted by the biokinetic model. However the rate of phase 1 metabolism was over predicted in both models. Pathway analysis learned that oxidative phosphorylation, drug metabolism and immune response were consistently regulated, suggesting that adverse effects of paracetamol are related to oxidative stress and immune dysfunction. The results indicate the feasibility of in vitro-based risk assessment of xenobiotics for the human population without the use of animals taking into account the individual variation in drug metabolism. For future refinement more data on metabolizing enzyme and transporter protein abundance is requ