A predictive algorithm for skin permeability: the effects of molecular size and hydrogen bond activity.

Abstract

To develop a predictive algorithm of nonelectrolyte transport through skin based upon a partitioning-diffusion model. Drug permeability is described by a partitioning-diffusion equation. Through free-energy relationships, partitioning is related to the drug's molecular volume (MV), and hydrogen bond donor (Hd) and acceptor (Ha) activity. Diffusion is related to the drug's MV using a theory of diffusion through lipid lamellae based on free-volume fluctuations within the lipid domain. These two explicit descriptions are combined to give an equation describing permeability in terms of the permeant's physical properties. The aqueous permeability coefficients of 37 nonelectrolytes through human epidermis were evaluated as a function of these physical properties using a multiple regression analysis. The results of the regression analysis show that 94% of the variability in the data can be explained by a model which includes only the permeant's MV, Hd and Ha. These results further provide an algorithm to predict skin permeability based upon the values of these parameters. In addition, the relative contribution of various chemical functional groups (e.g., -COOH) is derived, and can be used to predict skin transport from drug structure alone. A biophysically relevant model of drug transport through human skin is derived based solely on the physical properties of the drug. The model provides an algorithm to predict permeability from the drug's structure and/or physical properties. Moreover, the model is applicable to a number of lipid barrier membranes, suggesting a common transport mechanism in all.