Incorporating a dermal absorption route into high throughput toxicokinetic modeling

Dermal absorption of chemicals represents an important route of exposure in pharmaceutical, occupational, and environmental settings. Thousands of chemicals with little toxicity or toxicokinetic (TK) data are in use. However, in many cases, in vitro bioactivity data are available. While it is not feasible to collect in vivo TK data, TK may be estimated using high-throughput methods.

This study developed a generalized physiologically-based TK (PBTK) dermal exposure model for in vitro-in vivo extrapolation (IVIVE). This model estimates dermal exposures that result in systemic concentrations comparable to those associated with in vitro bioactivity.

The PBTK model simulated dermal exposures for 22 unique exposure scenarios across 12 chemicals with published in vivo concentration time course TK data. Two different methods for estimating chemical- and vehicle-specific skin permeability were evaluated: Potts-Guy [1] and Surrey [2]. Root mean squared log₁₀ errors (RMSLE) were calculated on a per-chemical and method basis.

Given only 12 chemicals with in vivo TK data to permit evaluation, a single, optimal method for predicting dermal permeability could not be identified. IVIVE was performed separately using both permeability methods to calculate administered equivalent doses (AEDs) relevant to potential occupational exposure for 561 chemicals with in vitro bioactivity data. AEDs were defined here as parts per million (ppm) solution concentrations that would result in bioactive plasma concentrations after eight hours of submerged hands. The Potts-Guy method indicated that AED concentrations were not achievable for many chemicals.

The new dermal PBTK model works with a pre-existing database of more than one thousand compounds including industrial chemicals and pesticides. Regardless of method, IVIVE indicated that dermal exposures can lead to bioactive plasma concentrations for only a small fraction of the chemicals with in vitro bioactivity data examined. Gloves might be recommended for handling those chemicals.

Rapid estimation of risk posed by chemicals through dermal contact is an important need for occupational settings. A generic PBTK model was developed to characterize dermal absorption using chemical-specific in vitro data for metabolism and protein binding and physico-chemical properties. The model was evaluated using in vivo toxicokinetic data for multiple chemicals. The data were equivocal with respect to different dermal absorption assumptions. The new model allows for conversion of in vitro chemical bioactivity data from high-throughput toxicity screening to solution concentrations that would result in bioactive plasma concentrations in occupationally relevant conditions.