In silico model to predict dermal absorption of chemicals in finite dose conditions.

Abstract

The development of in silico approaches that can estimate the dermal absorption of chemicals exposed in practical conditions is highly anticipated. In the present study, an in silico model to estimate both the dermal absorption rate and dermal permeation profile was developed for the application of chemicals in finite dose conditions. Forty-three chemicals with molecular weights in the range 116-362 and logK_o/w in the range 1.1-4.5 were used to develop an in silico model. A gradient boosting tree approach was applied to estimate permeation parameters for diffusion and partition coefficients of the chemicals in skin using physicochemical parameters of the chemicals such as molecular weight, lipophilicity, and the highest and lowest occupied molecular orbitals as the descriptor. In addition, 11 chemicals with different molecular weights and lipophilicities were applied on excised human skin in a finite dose condition, and dermal absorption profiles were obtained. Consideration of donor-solvent evaporation time, saturated concentrations of the chemicals, and donor-solvent coverage area on the skin surface, in addition to estimated skin permeation parameters of the chemicals, showed comparatively good dermal absorption profiles, although some cases of underestimation of dermal absorption were identified. It will be necessary to verify the accuracy of this model through experiments using more chemicals. However, the obtained results suggested that the established model may be valid to estimate the dermal absorption of chemicals in practical conditions.