Interlaboratory assessment of the bovine corneal opacity and permeability (BCOP) assay.

A multinational interlaboratory study to investigate the bovine corneal opacity and permeability (BCOP) assay is presented. The aim of this work was to determine the capability and possible limitations of this method to predict ocular irritancy of a large set of chemicals. The assays were carried out in 12 European laboratories with different types of activity. In each of these laboratories 52 substances, with a wide range of structure, physical form and irritant properties, were tested and in vitro scores were compared with those obtained from concurrent rabbit eye (Draize) tests. The technique was easily learned by workers in the participating laboratories, as shown by the fact that there were consistent responses between treated corneas within an individual laboratory. Interlaboratory variability was also very good. It was found that a given laboratory had a 96% chance of classifying irritants or non-irritants similarly to the other laboratories. In addition, it was observed that corneas preserved overnight responded similarly to freshly prepared tissues, thus allowing flexibility for those laboratories where the availability of corneas is limited. Comparisons between in vivo and in vitro data showed that the BCOP data correctly predicted whether a compound would be irritating or non-irritating for 44 of the 52 compounds (84.6%). Specificity and sensitivity were also greater than 84%, and the same number of substances were overestimated as were underestimated (four out of 52). All of the false negatives were solids whereas most of false positives were liquids, indicating that some adjustment in the protocol may be required depending on the physical state of the substance to be tested. All of the substances selected could be evaluated, with no limitation such as colour, insolubility, low or high pH. Given the number of products evaluated and the reproducibility within and among the laboratories involved, the overall results are quite satisfactory and therefore confirm the usefulness of the assay for screening chemicals for ocular irritation.