Joint poster with dōTERRA: Case study on Essential Oils
Comparative analysis of skin sensitization thresholds for Essential Oils: Human, murine, and GARD®skin Dose-Response
Presented at 2025 Eurotox
Conclusion
- NESIL predictions from GARD® were consistent with other skin sensitization data, particularly when compared to reference human data (HRIPT) and constituent predictions.
- The assay supports quantitative skin sensitization potency assessment of complex mixtures, offering a reliable non-animal alternative to traditional testing methods.
Abstract
Essential oils (EOs), widely used in consumer products, require robust skin sensitization hazard and potency assessment. However, dose-response thresholds for EOs remain understudied, and current classifications under CLP mixture criteria are often overly conservative. Traditional methods, such as animal testing and human patch tests, face ethical concerns, regulatory restrictions, and reliability issues. While New Approach Methodologies (NAMs) address some of these challenges, most of them are validated only for hazard identification and lack quantitative potency assessment capabilities.
GARD®skin (OECD TG 442E) is an in vitro assay that identifies chemical skin sensitizers based on the transcriptional profiling of a 196-gene biomarker signature in the dendritic-like SenzaCell® cell line. Predictions are made using a machine-learning algorithm, which classify test chemicals as sensitizers or non-sensitizers based on the assay’s readout, Decision Values (DVs). GARD®skin Dose-Response (OECD TGP 4.106) extends this approach by evaluating test chemicals across a concentration range to establish a dose-response relationship between DVs and test chemical concentration. Sensitizing potency is quantified using cDV0, the lowest dose required to elicit a positive response in GARD®skin. Depending on the need, the readout can be used to predict LLNA EC3 values, No Expected Sensitization Induction Level (NESIL), and UN GHS/CLP classification (1A or 1B), all with high statistical significance.
This study evaluates the utility of GARD®skin Dose-Response to predict NESILs for EOs, bridging the gap between hazard identification and quantitative risk assessment. Three EOs (A, B, C) were analysed using GARD®skin Dose-Response. Their skin sensitization potency threshold levels were estimated by the assay readouts (cDV0) and NESIL predictions. Results were compared to existing murine (LLNA EC3) and human (HRIPT/HMT) data through weight-of-evidence analysis.
For test item A, the predicted NESILs spanned were 5700 (GARD®), 6900 (constituent prediction), 1800 (LLNA), 2000 (HRIPT), and 3500 µg/cm² (HMT). Test item B showed NESILs of 24000 (GARD®), 29000 (constituent prediction), 6700 (LLNA), 44000 (HRIPT), and 11000 µg/cm² (HMT). Test item C was classified as a non-sensitizer by GARD®skin Dose-Response, with constituent prediction, LLNA, HRIPT, and HMT NESILs at 20000, 3900, 20000, and 690 µg/cm², respectively.
NESIL predictions from GARD®skin Dose-Response aligned with other skin sensitization test results, supporting its utility in quantitative sensitization potency assessment of complex natural extracts like EOs. While confirmatory human studies (e.g., CNIH protocols) are recommended, this assay reduces reliance on animal and human testing and advances the use of NAMs in safety assessment.
