Correlating percutaneous absorption with physicochemical parameters in vivo in man: agricultural, steroid, and other organic compounds.

In vitro data are currently used to predict cutaneous chemical exposure based on physicochemical parameters. However, this in vitro data may not sufficiently account for what occurs in vivo. Previously, we modeled (via multivariate analysis) percutaneous absorption with physicochemical parameters using in vivo human transdermal patch-based data. In our current study, we correlated absorption data from three human in vivo data sets to physicochemical parameters. Most univariate and multivariate analyses did not provide satisfactory fits, and only steroids demonstrated significant relationships, where: (1) total percentge absorption inversely correlated with molecular weight and number of hydrogen bond acceptor groups on the molecule; and (2) maximal absorption rate inversely correlated with molecular weight, and number of hydrogen bond donor and acceptor groups on the molecule. For the most part, however, disparities exist between our previous results with transdermal patches and our current results with acetone as the solvent. Reasons for this discordance may include: drug-vehicle interactions, compound variability and endpoint differences. With such variability between in vivo human data, current in vitro predictive models should be critically scrutinized.