Liebsch, M., Traue, D., Barrabas, C., Spielmann, H., Gerberick1, F., Cruse2, L., Diembeck2, W., Pfannenbecker2, U., Spieker2, J., Holzhütter3, H. G., Brantom4, P., Aspin4, P., Southee5, J. A. National Center for Documentation and Evaluation of Alternative Methods to Animal Experiments (ZEBET), Federal Institute for Health Protection of Consumers and Veterinary Medicine (BgVV), Diedersdorfer Weg 1, 12277 Berlin, Germany, 1Procter & Gamble, Cincinnati, USA, 2Beiersdorf AG, Hamburg, Germany, 3Humboldt University, Berlin, Germany, 4BIBRA International, Carshalton, UK, 5MA Bioservices, Stirling, UK.

The EpiDerm™ phototoxicity test was chosen for prevalidation within ECVAM’s tiered prevalidation procedure, since production of the full human in vitro skin model Skin2™ had been ceased in October 1996. ZEBET had developed a test protocol of a Skin2 phototoxicity test and had obtained promising results with a refined version of this test in the blind trial of an EU/COLIPA validation project, In Vitro Photoirritation. In phase I, ZEBET (laboratory 1) drafted a Standard Operating Procedure (SOP) and a project plan for conducting the study. ZEBET’s major task in phase I was to adapt the existing Skin2 methodology to the specific needs of the epidermal model EpiDerm. Therefore, phase I was focused on a thorough determination of the UV sensitivity of EpiDerm and on improving chemical exposure conditions. UVA sensitivity experiments revealed 6J/cm2 as the highest non-cytotoxic UVA dose and exposure to test materials for 3 hours, 6 hours and 21 hours before UVA irradiation revealed 21 hours to be the optimum exposure time. Established phototoxins were used to evaluate the method and were correctly identified. In addition, UV filter chemicals (in vivo non-phototoxic) were classified correctly as negative. However, musk ambrette and 6-methylcoumarin (in vivo photoallergens with low acute phototoxic potential) were classified negative. Since data confirmed the results previously obtained with the Skin2 phototoxicity test, neither the test design nor the prediction model had to be changed for the EpiDerm phototoxicity test, and it was decided to proceed to phase II. In phase II, taking into account the patch technique used in vivo, the three participating laboratories (ZEBET, Procter & Gamble and Beiersdorf) amended the application technique and evaluated the usefulness of a patch technique for the EpiDerm test. Results revealed that the patch technique was useful for chemicals applied in oil, but not for chemicals dissolved in water. Since reproducibility of the data within, and between, the three laboratories was excellent, it was decided to proceed to phase III. At the beginning of phase III, ZEBET drafted the final version of the SOP, and BIBRA International selected ten chemicals (5 phototoxic and 5 non-phototoxic) for the blind trial from a list of 16 chemicals, which were distributed to ZEBET, Beiersdorf and Procter & Gamble. Each of the ten chemicals was tested twice independently in each laboratory and the data were submitted to Humboldt University (Berlin) for biometrical analysis, which confirmed the expected predictivity and robustness of the test; only one positive chemical (tetracycline, free base) was not detected as a phototoxin and none of the negative chemicals was overpredicted as a phototoxin.


ECVAM, EpiDerm, Phototoxicity, Pre-validation, Prevalidation, Prevalidation, Reproducibility, Reproducible, Sun screens, UVA, UVB, Validation

Request a copy of this paper, click here.