“TKPlate”: A web-based platform for toxicokinetic modelling in food safety

Abstract

CYP3A4 is the major human cytochrome P450 isoform responsible for the metabolism of more than 50% of known xenobiotics. Here, inter-ethnic differences in CYP3A4 metabolism have been investigated through a systematic review of pharmacokinetic data for 15 CYP3A4 probe substrates and parameters reflecting acute (Cmax, oral route) and chronic exposure (clearance and area under the plasma concentration-time curve, oral and intravenous route). All data were extracted in a structured database and meta-analyses were performed using a hierarchical Bayesian model in the R freeware to derive parameter, route and population-specific variability distributions for CYP3A4 metabolism. Two different approaches were applied. 1) Inter-individual differences were quantified using North American healthy adults as a reference group to compare with European, Asian, Middle East, and South-American healthy adults and with ethnicity, elderly, children and neonates. 2) Intra-ethnic–specific variability distributions were derived without comparing to a reference group. Overall, subgroup-specific distributions for CYP3A4-variability provided the basis to derive CYP3A4-related uncertainty factors (UF) to cover 95th or 97.5th centiles of the population and were compared with the human default toxicokinetic UF (3.16). The results indicate that CYP3A4-related UFs in healthy adults were higher for chronic oral exposures (2.5–3.0, UF95 and UF97.5, 10 compounds) than for intravenous exposures (1.7–1.8, 2 compounds). All UFs were within the default TK UF. These distributions allow for: 1) the application of CYP3A4-related UFs in the risk assessment of compounds for which in vitro CYP3A4 metabolism evidence is available without the need for animal data; 2) the integration of CYP3A4-related variability distributions with in vitro metabolism data into physiologically based kinetic (PBK) models for quantitative in vitro to in vivo extrapolation (QIVIVE) and 3) the estimation of UFs in chemical risk assessment using variability distributions of metabolism.