In silico models to predict dermal absorption from complex agrochemical formulations.

Area of Science:

• Toxicology and Environmental Science
• Computational Chemistry
• Dermal Absorption Studies

Background:

• Dermal absorption is crucial for risk assessment of agrochemical formulations.
• Current methods involve extensive in vivo or in vitro testing.
• Mixture-specific effects on absorption are complex to predict.

Purpose of the Study:

• To develop an in silico prediction model for dermal absorption.
• To account for mixture-related effects in agrochemical formulations.
• To reduce the number of required animal and non-animal absorption experiments.

Main Methods:

• Utilized an experimental dataset from regulatory in vitro studies with human and rat skin.
• Generated descriptors for 56 individual substances and over 150 mixtures.
• Employed multiple linear regression analysis to build the prediction model.

Main Results:

• The developed in silico model achieved a good fit (r² = 0.75).
• The model demonstrated strong predictive ability for new formulations (Q²Ext = 0.73).
• It accurately estimates the influence of mixture composition on dermal absorption.

Conclusions:

• The in silico model effectively predicts dermal absorption of complex mixtures.
• Its application can significantly reduce animal and non-animal testing in formulation development.
• This approach can streamline regulatory processes through read-across acceptance.