Interlaboratory validation of in vitro eye irritation tests for cosmetic ingredients. (2) Chorioallantoic membrane (CAM) test

A chorioallantoic membrane (CAM) assay evaluates the blood vessel reaction and damage to the CAM of a fertilized hen's egg. Two types of CAM assays, t he hen's egg test-chorioallantoic membrane (HET-CAM) method and the chorioa llantoic membrane-trypan blue staining (CAM-TB) method, were evaluated as a lternative methods to the Draize eye irritation test (Draize test). The val idation project was composed of three test phases in which 10, 15 and 14 te st chemicals, respectively, were evaluated. The test procedure of the five independent laboratories was controlled under the same standard operating p rocedure (SOP). The interlaboratory variation was relatively high for both methods. However, the rank correlation was relatively high among the values obtained by the five laboratories. The variation associated with the CAM-T B method was smaller than that of the MET-CAM method, which requires macros copic observation, suggesting that the objectivity and quantitativeness dif fers between the assay systems. The average values using these two methods were compared with the maximum average Draize total score (MAS). The correl ation coefficient (r) between the MET-CAM scores and the MAS was 0.688. Thi s suggests that a simple linear regression may not be appropriate for MET-C AM. However, the Spearman's rank correlation coefficient (rs) was relativel y high (rs = 0.802). In contrast, the CAM-TB test results showed a good cor relation with the MAS when the test chemicals were classified according to their physical properties (r = 0.801, liquid and r = 0.926, powder). These results suggest that both the MET-CAM and CAM-TB methods may present altern ative method of evaluation of eye irritation despite problems of interlabor atory reproducibility. (C) 1999 Elsevier Science Ltd. All rights reserved.