Eye irritation hazard of chemicals and formulations assessed by methods in vitro.

BACKGROUND The aim of this study was to compare human and animal skin irritation data with results of selected in vitro methods, including HET-CAM test, Neutral Red Release Assay, Neutral Red Uptake Assay and EpiOcular eye irritation test and with already existing data of eye irritation obtained from animal experiments. METHODS Chemicals employed in previous skin irritation validation studies and commercially available cosmetic formulations were subjected to further testing using in vitro methods Neutral Red Release (NRR) assay, Neutral Red Uptake (NRU) assay, HET-CAM test and EpiOcular assay. RESULTS The study revealed that skin irritants are not necessarily eye irritants; specifically volatile or solid materials may be misclassified. NRR assay provided false negative results in case of substances with fixative effect or not removable under standard washing procedure, emphasizing the role of microscopical evaluation as a crucial additional endpoint. Although overpredictive, HET-CAM test provided the lowest false negative rate. The most aggressive cosmetic formulation was correctly identified by EpiOcular assay, in accordance with NRU and NRR assays results, while HET-CAM test correctly identified the mildest formulation. CONCLUSIONS Each of the in vitro methods is related to a specific endpoint of ocular irritation and provides only partial information on the mode of action of the tested material. Despite good reproducibility of individual in vitro assays, only the weight-of-evidence approach and results of multiple selected in vitro tests can allow for estimation of eye irritation hazard in vivo.