Photodynamic therapy (PDT) in dermatology has been proven to be a successful noninvasive therapeutic modality for treating skin cancer. To facilitate its clinical introduction, the development of topical photosensitizers is necessary to avoid generalized, cutaneous photosensitivity. Therefore the penetration of synthetic chemically pure 9-acetoxy-2,7,12,17-tetrakis-(beta-methoxyethyl)-porphycene (ATMPn) into human skin was studied. Single specimens of freshly excised perilesional skin (n = 70) and basal cell carcinomas (n = 28) were evaluated after topical application of ethanolic ATMPn solutions (0.1% and 0.05%) for various times (2, 6, 16, 20 h). The penetration depth of ATMPn, recognized as red fluorescence in cryostat sections, was determined qualitatively by fluorescence imaging using a system of scoring related to the morphological structure of human skin (0 no fluorescence, 5 fluorescence deeper than basement membrane). Perilesional skin incubated for 2 or 6 h revealed fluorescence restricted to the upper parts of the epidermis, while after 16 or 20 h of incubation fluorescence was detected down to the basement membrane resulting in a significantly higher score (mean sum of scores : 2 h 2.6 +/- 0.4; 6 h 3.2 +/- 0.1; 16 h 3.8 +/- 0.1; 20 h 3.6 +/- 0.1). Quantitative evaluation by digital image analysis confirmed the qualitative results. Fluorescence was limited to the epidermis and the fluorescence intensity of the epidermis was higher after 16 h (4.9% of the fluorescence standard) than after 6 h (4.1%) incubation. Basal cell carcinomas showed fluorescence in the deep dermis as early as after 6 h incubation, but restricted to tumour cell nests. These results suggest that penetration of ATMPn into tumour tissue after topical application might be sufficient for topical PDT and that poor penetration into surrounding tissue might prevent scar formation following irradiation for PDT. The penetration characteristics of ATMPn now have to be proven in an in vivo setting.