Chapter 37 - In vitro to in vivo extrapolation methods in chemical hazard identification and risk assessment

Abstract

Humans are continually exposed to a vast number of chemicals that may enter the environment through various pathways. Nevertheless, information on exposure and hazards of many chemicals remains elusive. Relying on conventional in vivo toxicity testing cannot timely and effectively obtain these profiles for the chemicals. High-throughput screening assays have generated large amounts of reliable in vitro bioactivity data. However, the nominal in vitro test concentrations cannot directly represent equivalent in vivo external doses. In vitro to in vivo extrapolation (IVIVE) methods can qualitatively or quantitatively convert in vitro data to in vivo ones, which have been used in chemicals risk assessment and validated with some chemicals. In this chapter, two primary concepts and workflows of IVIVE were introduced, including IVIVE for ADME parameters (IVIVE-ADME) and IVIVE for exposure doses (IVIVE-Dose). Application status of IVIVE in chemical risk assessment was summarized, including IVIVE-ADME in extrapolating clearance rate and bioconcentration factors and IVIVE-Dose in extrapolating toxicity of chemicals. Data-driven IVIVE models constructed from machine learning that predict in vivo toxicity endpoints from in vitro toxicity data were summarized. Challenges and perspectives of IVIVE were put forward.