CLINICAL AND LABORATORY INVESTIGATIONS BJD British Journal of Dermatology Prediction of chemical absorption into and through the skin from cosmetic and dermatological formulations S. Gre ´goire C. Ribaud F. Benech J.R. Meunier A. Garrigues-Mazert and R.H. Guy* L’Ore ´al Recherche, 1 avenue Euge `ne Schueller, 93601 Aulnay-sous Bois, France *Department of Pharmacy and Pharmacology, University of Bath, Bath, U.K. Correspondence Se ´bastien Gre ´goire. E-mail: sgregoire@rd.loreal.com Accepted for publication 14 July 2008 Key words chemical properties, cosmetics, partition coefficient, QSPR modelling, skin absorption, skin penetration relationships Conflicts of interest None declared. DOI 10.1111/j.1365-2133.2008.08866.x Summary Background To date, risk assessment following topical exposure to cosmetic ⁄ derma- tological formulations cannot be precisely evaluated. Objectives To provide a tool for optimization of active permeation into ⁄ through skin and for risk assessment. Methods A predictive model was developed for estimating the cumulative mass of a chemical absorbed into and across the skin from a cosmetic ⁄ dermatological formulation. Account was taken of (i) the ionization state of the chemical, to correct the skin ⁄ vehicle partition coefficient; and (ii) the nature of the cosmetic ⁄ dermatological formulation. Three specific assumptions were made: firstly, steady-state transport across the skin was achieved despite application of a finite dose of chemical; secondly, vehicle effects were small relative to the precision of the prediction; and, thirdly, each formulation could be treated as an oil-in-water emulsion, in which only that fraction of the chemical in the aqueous phase was available to partition into the stratum corneum. A database of 101 ex vivo human skin experiments involving 36 chemicals was analysed. Results For 91% of the data, the difference between predicted and experimental values was less than a factor 5; when the aforementioned corrections were not used, on the other hand, only 26% of the data was well predicted. The model was successfully applied to predict skin absorption of two compounds not included in the database, for which in vitro percutaneous penetration from cos- metic vehicles have been measured. Conclusion A model has been developed to predict the mass of a chemical absorbed into and through the skin from a cosmetic or dermatological formulation. The accurate prediction of dermal uptake and exposure of topically applied chemicals is relevant to both formulation development and risk assessment. Modelling this process has the ultimate goal of predicting the extent to which a molecule may penetrate the skin, without recourse to an experimental measurement. However, any predictive model is only as good as the information on which it is based. Hence, because in vitro percutaneous absorption experiments typically lead to the assessment of permeability coefficients, maximum fluxes and other ‘steady-state’ endpoints, the majority of quantitative structure–penetration relationships (QSPRs) have been devel- oped to predict this situation. 1,2 For example, the model of Potts and Guy 3 predicts the permeability coefficient (K p ) of a compound which contacts the skin in an aqueous solution applied under infinite dose conditions. While K p provides information on a chemical’s ability to pass through the skin, the model has to be adapted 4 to calculate the total amount absorbed (i.e. the quantity completely transported and that which is in the tissue). The objective of dermal absorption studies of cosmetic active ingredients, then, is to obtain quantitative information about skin uptake and transport in typical ‘in-use’ conditions. 5 Cosmetic and dermatological products designed for topical application are usually complex systems containing many ingredients in addition to the ‘active’ component. 6 Such for- mulations are rarely simple aqueous solutions, so predictions based on results from experiments, in which exposure occurs via such vehicles, must be viewed with care. Alternative mod- els have therefore been developed to improve this point. For example, a semi-empirical method, 7 based on a physicochem- ical descriptor, the Relative Polarity Index, has been proposed. The approach enables suitable emollient to be selected Ó 2008 L’Ore ´al 80 Journal Compilation Ó 2008 British Association of Dermatologists • British Journal of Dermatology 2009 160, pp80–91