2021 Volume 55 Issue 3 Pages 288-297
The risk of eye irritation due to drugs is determined primarily by the in vivo rabbit eye irritation test (Draize test) described in the OECD TG 405 and generally labeled by the Globally Harmonized System of Classification and Labeling of Chemicals (GHS). In the interest of animal protection, the development of in vitro alternatives to animal testing not involving the Draize test are being pursued. In this study, we examined whether the GHS eye-irritation classification could be predicted using the Hansen solubility parameter (HSP). Hansen spheres were created based on the safety information of the test substance and its HSP values, and the potential parameters were obtained. By comparing these potential parameters with the HSP value of each test substance, the GHS classification was predicted. For substances that cause serious damage (GHS category 1), the prediction accuracy was 87.7% (64/73), false-negative rate 26.7% (4/15), and false-positive rate 8.6% (5/58). For substances not classified as eye irritants (GHS not classified), the accuracy was 87.7% (64/73), false-negative rate 9.5% (4/42), and false-positive rate 16.1% (5/31). These data suggest the possibility that eye-irritation classified by GHS can be predicated with high accuracy by comparing the potential parameters to the HSP of test compounds.