Use of in vitro to in vivo extrapolation (IVIVE) to support risk assessment
Risk assessment for early life stages: Life stage physiologically based pharmacokinetic (PBPK) modeling for pyrethroids.
Under the new toxicity testing paradigm that is largely based on in vitro and in silico approaches, alternative strategies are needed to address potentially sensitive populations such as early life stages in chemical risk assessment. Here we present an in vitro to in vivo extrapolation (IVIVE) based parameterization approach to build life stage PBPK models to simulate chemical kinetics in different ages of humans. Two pyrethroids, deltamethrin (DLM) and cis-permethrin (CPM) were used as an example to show the utility of IVIVE-based life stage PBPK model to evaluate age-related differences in pharmacokinetics and resulting differences in target tissue exposure. First, an adult human model was developed and then extended to different life stages by incorporating age-specific metabolism and physiological parameters. Biological scaling, i.e., IVIVE, in conjunction with enzyme ontogeny data was used to estimate age-specific total intrinsic clearance (Clint) in the liver based on the in vitro Clint values determined in human expressed enzymes that are responsible for metabolism of DLM and CPM. Enzyme ontogeny curves for the enzymes contributing to these pyrethroids were derived using non-linear regression analysis of the age-specific protein expression data for each enzyme. IVIVE results showed that carboxylesterase (CES) enzymes are largely responsible for DLM and CPM metabolism in humans, contributing for more than 50% of the metabolism of each of these pyrethroids. CES enzymes rapidly mature after birth reaching adult levels at around 6 months. The resulting total Clint of DLM and CPM indicates a rapid metabolism of these compounds in vivo both in the young and the adult. The life stage PBPK model was used to simulate internal exposure in the target tissue (brain) in different ages (0.5, 2, 5, 12, 19 and 25 years old) to represent infants, children, and adults. The maximum concentration in brain in the young after a single or multiple daily exposures was generally lower than that the adult in response to the same level of exposure to a given pyrethroid. In addition, our preliminary work with trans-permethrin indicates that the DLM model can be used as a generic model for other pyrethroids when parameterized with compound specific in vitro metabolism data suggesting a feasibility of conducting clearance-based read-across for developing life stage PBPK models for pyrethroids as a group. The life stage modeling framework supported by IVIVE parameterization demonstrated in this study can be readily applicable to other chemicals with potential concerns for early life sensitivity.