BackgroundAcute fish toxicity test (AFT) is one of the cornerstones of environmental risk assessment (ERA) of chemicals for the aquatic environment. Despite many efforts to find an alternative able to fully replace the test, there is still lasting pressure from stakeholders for AFT results.ResultsHere, we present the results of a case study with eight pharmaceuticals from various pharmaceutical groups with different levels of expected toxicity to fish. Selected compounds were tested in two validated alternative tests-fish embryo toxicity test with zebrafish (Danio rerio) (zFET) and in vitro RTgill-W1 assay according to their corresponding OECD guidelines TG 236 and TG 249, respectively. Data for AFT were collected from PubMed and ECOTOX knowledgebase databases, and acute toxicity to fish was further predicted in silico by the ECOSAR program. Predicted environmental risks (risk quotients, RQ, calculated using the exposure data from NORMAN) from both zFET and RTgill-W1 well correlated with the average RQs based on AFT LC50s. The strongest and most significant correlation was observed while comparing the AFT results with the median of combined alternative methods (zFET, RTgill-W1, ECOSAR).ConclusionsThis proposed approach combining experimental data with modeling could serve as a reliable tool for predictions of environmental risks promoting the 3R alternatives to acute fish toxicity testing.