The OECD adopted a tiered approach for dermal testing described in the revised TG 404. Data from structurally related chemicals (e.g. pH extremes), human data and experience (weight of evidence) are sometimes sufficient for classifying the skin irritation/corrosion potential of chemicals. If this information is not sufficient, validated in vitro methods can then be used e.g. in vitro EpiSkin methods accepted in both OECD TG 431 (skin corrosion) and TG439 (skin irritation).
The aim of the study is to present the Top-Down and Bottom-Up approaches to the use of tiered in vitro testing strategies for predicting skin corrosion/irritation potential without the use of animals.
For such purpose, 98 chemicals (partially from EURL-ECVAM validation studies and OECD TG431 & TG439) were evaluated in both skin corrosion and irritation EpiSkin methods.
Amongst the 35 in vivo non irritants, 85% might be correctly classified in vitro as irritants (I). Therefore those 35 chemicals were also defined as non-corrosives (NC). Thus, when applying the bottom-up approach the GHS classification of those substances was accurately defined by performing the skin irritation evaluation. Considering the 40 known in vivo corrosives (C), all chemicals were correctly predicted as C and also identified as I. The substances are well classified by using the top-down approach which consists of conducting primarily skin corrosion test method.
Amongst the 23 in vivo I but NC chemicals, 20 out of 23 were correctly classified as I. The 3 others were therefore not classified as C (true false negatives). By applying the top-down approach, 6/20 irritants might be misclassified as C (false positives C) whereas they were well classified as irritants only using the bottom-up approach. The bottom-up approach is more accurate for those chemicals.
The use of existing data and a tiered in vitro approach (in conjunction with an intelligent selection of tests) should be sufficient and consider in the context of the revision of OECD TG404. This comprehensive evaluation should also facilitate regulatory acceptance decision on the methods, and it proposed uses.