Testing Framework for Prediction of Ocular Irritation Using the Bovine Corneal Opacity and Permeability (BCOP) Assay and Chorioallantoic Membrane Vascular Assay (CAMVA)
In vitro ocular irritation assays, such as the Chorioallantoic Membrane Vascular Assay (CAMVA) and Bovine Corneal Opacity and Permeability (BCOP) test, are routinely used by personal care products companies because they are rapid and economical to conduct, do not require the use of live animals, and provide reliable predictive data. Previous research using an extensive CAMVA and BCOP database at Kao USA Inc. has shown that ocular irritation potential for new hair shampoos, ethanol-based hair stylers, skin cleansers, and skin lotions can be reliably predicted using a decision tree that systematically compares the ingredient composition, particularly ethanol and surfactant content, of the new formulation to previously tested formulations. Because the studies comprising this original database were conducted at a single contract laboratory, a follow-up study using a second contract laboratory was conducted to demonstrate inter-laboratory reliability of the CAMVA/BCOP data-derived decision tree for prediction of ocular irritation potential. Thirty-five personal care products were tested using the CAMVA and/or BCOP assays. The ethanol and surfactant content of each test material was evaluated, and the results of the assays were compared to the decision tree-based predictions of ocular irritation potential. Our data confirmed the ocular irritation predictions made using the decision tree model for 33 of 37 test samples (89% correlation rate) and verified the inter-laboratory reliability of the CAMVA and BCOP assays when conducted using appropriate controls. Our results also strengthened the ocular irritation decision tree model by confirming that deodorants are consistently predicted not to be ocular irritants based on composition.