Plant extracts, polymers and new approach methods: Practical experience with skin sensitization assessment

Area of Science:

• Toxicology
• Dermatology
• Chemical Safety

Background:

• Non-animal methods for skin sensitization hazard identification are increasingly adopted as OECD test guidelines.
• Predictive accuracy of these methods beyond their initial validation chemical domains is often limited.
• Evaluating new methodologies is crucial for reliable chemical safety assessments.

Purpose of the Study:

• To compare the predictive accuracy of in vitro and in chemico methods against in vivo skin sensitization data for plant extracts and polymeric materials.
• To assess the performance of the Direct Peptide Reactivity Assay (DPRA), KeratinoSens (h-CLAT), and SENS-IS methods.
• To evaluate the '2 out of 3' approach for predicting skin sensitization.

Main Methods:

• Tested 12 plant extracts and 15 polymeric materials using DPRA, h-CLAT, and SENS-IS.
• Compared results from in vitro/in chemico methods with existing in vivo skin sensitization data.
• Utilized the '2 out of 3' approach and assessed SENS-IS performance separately.

Main Results:

• The '2 out of 3' approach yielded 50% balanced accuracy for 8 plant extracts using DPRA and h-CLAT.
• SENS-IS achieved 88% balanced accuracy for 11 plant extracts.
• For polymers, the '2 out of 3' approach showed 63% balanced accuracy (excluding inconclusive results), while SENS-IS achieved 68% balanced accuracy (excluding one inapplicable material).

Conclusions:

• The SENS-IS method demonstrated higher predictive accuracy for plant extracts and polymers compared to the DPRA/h-CLAT combination.
• Results suggest that certain substance subgroups might fall outside the applicability domains of specific methods.
• Careful analysis is necessary to accept positive or negative skin sensitization predictions from these non-animal methods.