Eye Irritation Testing
The Draize Rabbit Eye Test (DRET) is the regulatory ‘Gold Standard’ for assessing and categorizing ocular irritation/corrosivity. The DRET focuses attention on damage to a number of different ocular structures, which are scored and weighted based on toxicological importance. The structures are as follows: Cornea (CO) – 80 points, Conjunctiva (Conj) – 20 points, and Iris (IR) – 10 points.
The heaviest weighting is on corneal damage, which is 80 points out of the total 110-point scale devised by Draize. Corneal irritation scoring is based on assessment of opacities on the cornea, Conj irritation is assessed by increased vascularization, and iridial damage is assessed by function of the iris (ability to constrict or dilate pupil)and deepening of the rugae.
PorCORA – Corneal Irritation Focus
Since the CO scores have the heaviest weight, and most often are the drivers of eye irritation, we developed the Porcine Cornea Opacity Reversibility Assay (PorCORA), an ex vivo corneal model. The PorCORA can distinguish between a material’s potential to cause severe (reversible) versus corrosive (irreversible) damage. Excised cultured porcine corneas are topically dosed with a test material and kept in culture for 21 Days. Corneal damage is visually assessed and scored. If CO reverse and are not visible by Day 21, the test material is categorized as a non-corrosive material and deemed to not cause permanent ocular damage.
In internal validation studies, we tested 56 chemicals and dilutions of chemicals ranging from corrosive (GHS category 1) to non-irritating (GHS not categorized). Using Cooper Statistics, we arrived at an accuracy of 89% with a positive and negative predictivity of 85% (cat. 1) and 93% (not cat.1), respectively.
To determine if these Cooper Statistics could be improved, we used the drivers of classification concept based on Barroso et al., 2016*. Upon re-examination of our data based on this published database and methodology, we found that four chemicals (1,2,4-Triazole, N-Butanol, 2,5-Dimethyl-2,5-Hexanediol, and Potassium Cyanate) had invalid tests (animals euthanized prior to day 21) or produced irritation not driven by corneal opacities. These chemicals were removed from our dataset. Without these four chemicals the accuracy improved to 92%. Moreover, the major change was in our positive predictivity, which increased to 91%. The negative predictivity for this subset of chemicals remained the same.
Lastly, PorCORA’s predictivity was assessed based on CO persistence; i.e., materials that produced low CO scores (from Draize Rabbit Eye Tests). Of the 56 chemicals tested, 40 had low severity (i.e., CO mean scores < 3). The accuracy for persistence of low CO scores was 88% with positive and negative predictivity of 79%, and 92%, respectively.
* Barroso, J., Pfannenbecker, U., Adriaens, E., Alépée, N., Cluzel, M., De Smedt, A., Hibatallah, J., Klaric, M., Mewes, K., Millet, M., Templier, M., McNamee, P. 2016. Cosmetics Europe compilation of historical serious eye damage/irritation in vivo data analysed by drivers of classification to support selection of chemicals for development and evaluation of alternative methods/strategies: the Draize eye test Reference Database (DRD). Arch. Toxicol. 91:521-547.
Based on these data, the PorCORA was proven to be a valid test to assess a material’s potential to distinguish reversible versus irreversible eye damage.
Appropriate choices of validation chemicals, based on the drivers of classification, have proven valuable and should be applied to replacement ocular assays that previously failed to meet acceptance criteria.
PorCORA testing results are encouraging and future studies are planned that will pursue mixtures and chemical substances of unknown or variable composition, complex reaction products and biological materials. PorCORA
If you would like to find out more about PorCORA and other in vitro/alternative toxicology testing methods, please click here.