Search

This is the final report of the Management Team for a European Commission/British Home Office (EC/HO) validation study on alternatives to the Draize eye irritation test. The principal goal of the study was to establish whether one or more of nine non-animal tests could be used to replace the Draize test for all severely irritating materials (or those belonging to specific classes) or the animal test completely for chemicals with or without regard to chemical class. Sixty chemicals were independently selected, coded and supplied, then the data obtained in 37 laboratories were analysed independently. The results of comparisons between 27 alternative test index scores and the Modified Maximum Average Scores (MMASs) obtained in the Draize eye test were compared. Tables of results showing Pearson's product moment correlation coefficients and Spearman's rank coefficients for each laboratory are provided, and correlation matrices of alternative test index scores among the different groups of laboratories are shown for each endpoint. Scatterplots are provided, in which the alternative test scores obtained by the lead laboratories for the nine tests are plotted against the MMAS for the full set of chemicals and 12 surfactants. It is concluded that, with the possible exception of predicting the irritancy of surfactants, none of the nine tests met any of the four performance targets. Possible reasons for this outcome are discussed.

The current OECD guideline for the assessment of eye irritation recommends, in initial considerations, the use of data from skin irritation tests as a prescreen to detect the most severely irritating materials, assuming that materials that are severely irritating to the skin are also significantly irritating to the eyes. However, analysis of data for 179 materials tested in this laboratory for both dermal and ocular irritancy, revealed that, at most, only 36% of severe eye irritants were also severe skin irritants. This resulted in a significant number of rabbits developing severe ocular effects that had not been predicted from the dermal responses. This study reports the results of an alternative approach for predicting severe eye irritants. The approach was a two-stage test battery in vitro: the first stage was a cytotoxicity assay utilizing the K562 cell line; the second was the isolated rabbit eye test. In contrast to the use of skin irritation tests, the in vitro battery was significantly more predictive (83% of severe eye irritants were detected). Although the incidence of false positive responses in each of the assays precludes their routine use as a replacement from the in vivo rabbit eye test they provide a powerful aid to reducing animal use and guiding in vivo studies to minimize the severity of effects. The need for an interlaboratory assessment to confirm and extend these findings is discussed.

Seventeen raw materials and chemical formulations were evaluated in the Eytex System to determine the ability of this assay in vitro to predict eye irritation potential in vivo. All the test samples, which represented a wide range of chemical types and eye irritancy potential in vivo, were provided by one of the participating laboratories. Historical data from tests in vivo were available for each of the test samples, so testing in vivo specifically for this study was not necessary. Samples were evaluated by both the membrane partition assay (MPA) and the rapid membrane assay (RMA). The sensitivity, specificity, predictivity and equivalence of the Eytex assay were determined by comparison with the rabbit eye irritation data, using each of the different Eytex Draize Equivalent (EDE) classification schemes. Regardless of the classification scheme used, the correlation between the scores in vivo and in vitro was poor. The Eytex System consistently overclassified materials of low irritancy in vivo and underclassified those test materials of moderate irritancy or above. On the basis of the results from the 17 materials tested in this study, the Eytex System appears unsuitable as a replacement in vitro for ocular irritancy testing of all types of chemical. However, Eytex may have a place as a pre-screening method used as part of a test battery.