Abstract
The GARDskin DR assay is a non-animal method that predicts dermal sensitization potency of discrete compounds. This comparative analysis evaluates outputs from a component-based prediction model (CP), GARDskin DR, and historical local lymph node assays (LLNA) exploring the applicability of CP and GARDskin DR for evaluating the sensitization potencies of essential oils. Eight well-studied essential oils were selected including cedarwood, cinnamon bark, clove bud, geranium, lavender, lemongrass, spearmint, and tea tree oils. Each underwent GC-MS and GARDskin DR analysis. CP model outputs, GARDskin DR and reference LLNA data were compared using standard statistical methods. Spearman’s test shows significant correlation between NESILCP vs. NESILGARD (0.881, p = 0.007) and NESILLLNA (0.929, p = 0.007) but not between NESILGARD vs. NESILLLNA (0.607, p = 0.17). Friedman’s test detected a significant difference among methods (χ²(2) = 7.71, p = 0.021), with post hoc tests showing relatively greater NESILCP values. Kendall’s W shows overall strong statistical concordance between methods (0.857, p = 0.017). NESILGARD and NESILCP generally align with NESILLLNA for the essential oils tested, although the CP method yields greater values. With the limited number of materials tested, results should be interpreted descriptively but show promise for GARDskin DR and CP for evaluating dermal sensitization potency of essential oils.

This publication has 53 references indexed in Scilit: