Skin sensitisation testing in practice: Applying a stacking meta model to cosmetic ingredients

• Published in: Toxicology in vitro 2020-08, Vol.66, p.104831, Article 104831
• Main Authors: Tourneix, Fleur, Alépée, Nathalie, Detroyer, Ann, Eilstein, Joan, Ez-Zoubir, Mehdi, Teissier, Silvia Martinozzi, Noçairi, Hicham, Piroird, Cécile, Basketter, David, Del Bufalo, Aurélia
• Format: Article
• Language: English
• Subjects: Animals; Biomedical materials; Computer Simulation; Consumer safety; Cosmetic ingredients; Cosmetics; Cosmetics - toxicity; Dermatitis, Allergic Contact; Fragrances; Haptens - toxicity; Humans; In vitro methods and tests; In vivo methods and tests; Integrated testing strategy; Local lymph node assay; Lymph nodes; Mathematical models; Models, Biological; New alternatives methodologies; Parameters; Preservatives; Raw materials; Risk assessment; Skin sensitisation; Skin Tests; Stacking; Volatility
• ISSN: 0887-2333; 1879-3177
• DOI: 10.1016/j.tiv.2020.104831

Recently, several non-animal approaches contributing to the identification of skin sensitisation hazard have been introduced. Their validation and acceptance has largely been directed towards regulatory classification. Considering the driving force for replacement of in vivo tests centred on cosmetics, it is reasonable to ask how well the new approaches perform in this respect. In the present study, 219 substances, largely cosmetic raw materials (including dyes, preservatives and fragrances), have been evaluated in our Defined Approach integrating a stacking meta model (version 5), incorporating the individual outcomes of 3 in vitro validated methods (Direct Peptide Reactivity Assay, Keratinosens™, U-SENS™), 2 in silico tools (TIMES SS, TOXTREE) and physicochemical parameters (volatility, pH). Stacking meta model outcomes were compared with existing local lymph node assay (LLNA) data. Non-sensitisers comprised 68/219; 86 were weak/moderate and 65 were stronger sensitisers. The model version revision demonstrate the gain to discriminate sensitizers to non-sensitiser when the in silico TIMES model is incorporated as input parameter. The 85% to 91% accuracy for the cosmetics categories, indicates the stacking meta model offers value for the next generation risk assessment framework. These results pinpoint the power of the stacking meta model relying on a confidence based on the probability given in any individual prediction. •Multiple non-animal approaches for identifying skin sensitisation hazard exist.•Distilling information from many inputs can refine and add probability to this.•A Stacking Meta-model offers such a system and is now in version 5.•The model has been applied successfully to a range of cosmetic ingredients.