Quantitative structure-activity relationships (QSARs) for the prediction of skin permeation of exogenous chemicals

Quantitative structure-activity relationships (QSARs) for the skin permeability coefficients of 158 compounds through excised human skin in vitro have been developed. A number of compounds, including hydrocortisone derivatives, were removed from the dataset as reported permeability data for these compounds was considered to be in error. QSARs developed for the dataset with the outliers removed included terms for hydrophobicity, molecular size, and hydrogen bonding. These descriptors provided an excellent fit to the data (r2 = 0.90), are easily calculated from molecular structure, and are mechanistically interpretable. Further analyses of the dataset indicated that good QSARs could be developed utilising hydrophobicity and molecular size alone, with molecular volume and molecular weight providing good quantification of molecular size.