Despite more than a decade of research in the field of endocrine active compounds targeting the androgen receptor (AR), and although suitable cell lines can be obtained, no validated human stably transfected androgen sensitive transactivation assay is available. Bayer Schering Pharma (BSP) and the Flemish Institute for Technological Research (VITO), partners within the EU-sponsored 6th framework project ReProTect, made first steps towards such a validation. A standard operation protocol (SOP) developed at BSP based on the androgen sensitive PALM cell line was transferred to VITO and its performance and transferability were thoroughly studied. The investigation followed a generic protocol prepared for all reporter gene assays evaluated within ReProTect, and in both laboratories at least three independent experiments were performed. The highest concentration to be tested was limited to 10 microM, if needed. A few compounds, 17alpha-methyltestosterone (17alpha-MT), vinclozolin and linuron, were studied using a real world scenario, i.e., assuming that their interaction with the AR was not known: A prescreening for agonism and true, competitive antagonism was used to select conditions such as the appropriate mode of action, and the working range excluding cytotoxicity for the final screening. All other compounds were tested according to the generic protocol: Compounds screened for agonism were the reference androgen 17alpha-methyldihydrotestosterone (MDHT), levonorgestrel, norethynodrel, progesterone, o,p'-DDT, and dibutylphthalate (DBP), while compounds screened for antagonism were the reference anti-androgen flutamide, prochloraz, o,p'-DDT, progesterone, norethynodrel, and DBP. Cytotoxicity was assessed in parallel as lactate dehydrogenase release. The prescreen classified 17alpha-MT as androgenic, vinclozolin and linuron as anti-androgenic and compounds were tested accordingly. In the absence of cytotoxicity, appropriate androgenic properties of reference and test compounds were detected by both laboratories, o,p'-DDT and DBP had no androgenic activity. Across the two laboratories EC(50)-values for MDHT, 17alpha-MT, and levonorgestrel varied by not more than a factor of 3.4, for norethynodrel by a factor of 9.7. Progesterone effects could not fully be evaluated, as frequently concentration response curves were incomplete. In the absence of cytotoxicity anti-androgenic properties of reference and test compounds were also detected in both laboratories. DBP, the putative negative reference compound, was inactive, norethynodrel rather showed agonistic properties. Progesterone was an antagonist at low concentrations, but agonistic properties were observed in one laboratory at high concentrations. Since the highest test concentration was limited to 10 microM, for some compounds no complete concentration response curves were obtained and estimation of EC(50)-values was less robust. Our data demonstrated that the SOP was transferable, and that the assay was able to rank compounds with strong, weak, and without affinity for the AR and to discriminate agonists and antagonists. The sensitivity of the assay could be improved further, if the limit of solubility or beginning cytotoxicity was chosen as the highest test concentration. The assay avoids the use of tissues from laboratory animals, and thus contributes to the 3R concept. Furthermore, it could be adjusted to an intermediate/high throughput format. On the whole, this PALM assay is a promising candidate for further validation.