Roguet, R., H. Beck, E. Boelsma, M. Bracher, C. Faller, I. Harris, C. Lotte, F. Dreher, and M. Ponec. Testing and Improvement of Reconstructed Skin Kits in Order to Elaborate European Standards: First Results. ATLA. 27: 333. 1999.

lauric acid - 00143-07-7; caffeine - 00058-08-2; mannitol - 00087-78-5;

Reconstructed human skin models are one of the most promising approaches to the assessment of the safety and efficacy of cosmetic products and chemicals. Several reconstructed skin models have been developed, and some are sold by European and American companies in the form of kits. The scientific aim of this 3-year project is to establish test protocols and to evaluate the relevance of commercial and experimental (in-house) skin models in the three main areas of cosmetic safety testing: percutaneous absorption, cutaneous metabolism of xenobiotics, and cutaneous irritancy. Test protocols could then form the basis of a proposal for international validation under the auspices of ECVAM. Results of the first phase of the project indicate that all skin models tested (Episkinâ, EpiDermâ from MatTek, SkinEthic’s model, and “in-house” models) reproduced many of the characteristics of human epidermis. In particular, histological examination showed a completely stratified stratum corneum in all models. Low intra-batch variations were observed. Inter-batch variations were low for EpiDerm and moderate for EpiSkin, but considerable variations (thickness of the epidermis, presence of pycnotic cells) were noted in the SkinEthic model. Penetration studies showed the same rank order of penetration in terms of total absorption (lauric acid>caffeine>mannitol) for the three commercial models tested. The rank order was compatible with expected in vitro results on human skin. In any case, intra-batch variation was smaller than inter-batch variation. For all compounds tested, the reproducibility of the results was better with EpiDerm and Episkin than with the SkinEthic model. Metabolic studies showed the presence of NADPH quinone reductase in all commercial models. Episkin and EpiDerm contained equal levels of this enzyme activity. SkinEthic kits contained higher activity than other models tested. Glutathione-S-transferase activity was found in all models. Low intra-batch but higher inter-batch variations were observed for all enzymes and models. Measurement of the enzyme activities in normal epidermis is necessary to select the most representative kit. In vitro assessment of skin irritancy was conducted using cytotoxicity (MTT), release of IL-1a, and cytoplasmic enzymes, as endpoints. Control variability results showed that intra-batch and inter-batch variations are low. After SDS treatment, inter-batch variability of MTT results was lower for EpiDerm, followed by Episkin, the Cosmital model and finally the SkinEthic model. Results of IL-1a, lactate dehydrogenase and glutamate-oxaloacetate transferase release showed a relatively high variability intra-batch or inter-batch.