A robust structure-activity relationship (SAR) model for esters that causeskin irritation in humans

A structure-activity relationship (SAR) model has been developed to discrim inate skin irritant from nonirritant esters. The model is based on the phys icochemical properties of 42 esters that were tested in humans for skin irr itation. Nineteen physicochemical parameters that represent transport, elec tronic, and steric properties were calculated for each chemical. Best subse ts regression analysis indicated candidate models for further analysis. Reg ression analyses identified significant models (p < 0.05) that had variable s that were also significant (p < 0.05), These candidate models were evalua ted using linear discriminant analysis to determine if the irritant esters could be discriminated from nonirritant esters. The stability of the model was evident from the consistency of parameters among ten submodels generate d using multiple random sampling of the database. The sensitivity of the te n models, evaluated by "leave-one-out" cross-validation, ranged from 0.846 to 0.923, with a mean of 0.885 +/- 0.025 (95% CI). The specificity ranged f rom 0.615 to 0.923, with a mean of 0.738 +/- 0.06 (CT). Compared with nonir ritant esters, irritant esters had lower density, lower water solubility, l ower sum of partial positive charges, higher Hansen hydrogen bonding parame ter, and higher Hansen dispersion parameter. The results indicate that phys icochemical features of esters contribute to their ability to cause skin ir ritation in humans, and that chemical partitioning into the epidermis and i ntermolecular reactions are likely important components of the response. Th is model is applicable for prediction of human irritation of esters yet unt ested.