Physiologically based kinetic modelling for species extrapolation of toxicokinetic data between small mammals: A systematic evaluation

Résumé fourni par la source

Environmental risk assessment (ERA) necessitates the evaluation of numerous species that cannot be directly tested due to ethical and resource limitations. Thus, cross-species extrapolation of experimental data is essential for ERA, especially in the context of mechanistically informed (next generation) risk assessments. Physiologically based Kinetic (PBK) models allow for cross-species extrapolation of toxicokinetic (TK) data in ecotoxicology, but a systematic evaluation of performance and data requirements for this application is lacking. This study aimed to assess the data requirements and performance of PBK models when extrapolating TK data among small mammals. We parameterized PBK models for three mammal species (Rattus norvegicus, Mus musculus, Oryctolagus cuniculus) in the PK-Sim software and performed cross-species extrapolations for nine compounds, all six possible reference-target species combinations, while systematically omitting available (in vitro) data. The results indicate a substantial improvement in prediction performance over bodyweight-scaled models, with clearance data contributing most significantly to performance. Notably, a limited in vitro dataset can enable robust extrapolation that approaches the accuracy of a direct fit to the target data. Data from Rattus norvegicus, a common reference species in ecotoxicology, yielded good performance when extrapolating to the other two species. For all three species, prediction accuracy may decline when extrapolating beyond the dose range of the reference dataset or in the presence of saturation effects. The established framework and codebase can be expanded to include additional compounds, species, and administration routes, facilitating a data-efficient ERA with mechanistic models.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.