Detection and Potency Classification of Pre- and Pro-Haptens Using the SENS-IS Assay: The Role of Metabolically Competent Reconstructed Skin

Skin sensitisation assessment increasingly relies on non-animal methods, yet the detection and characterisation of substances requiring metabolic activation, such as pre- and pro-haptens, remain challenging for several existing new approach methodologies (NAMs). The present study evaluated the performance of the SENS-IS assay using blinded testing of 50 substances reported in the literature as putative pre-haptens or pro-haptens on the historical EpiSkin™ model, in comparison with available LLNA, human and other NAM data. Additional experiments evaluated transferability to the current IS-SKIN model and the impact of a full-thickness skin model for selected metabolically demanding compounds. SENS-IS identified sensitisation as a hazard for the majority of substances in this pre- and pro-hapten-enriched dataset, including compounds requiring abiotic or metabolic activation, and provided potency-related categorisation broadly consistent with LLNA EC3 trends. Comparative analyses further supported comparable hazard identification and potency categorisation when SENS-IS was performed using the IS-SKIN and EpiskinTM reconstructed epidermis models for the tested subset of pro-haptens, supporting functional equivalence between tissues within the SENS-IS framework. For a limited subset of aromatic compounds reported to require extensive metabolic activation, sensitisation potency appeared to be underestimated in epidermal-only models. To address this limitation, the assay was applied to a full-thickness skin model. This approach improved potency characterisation of these metabolically complex compounds, while not increasing potency estimates for substances already adequately detected using epidermal models. Enhanced potency detection was associated with increased induction of phase I and phase II biotransformation genes included in the SENS-IS signature. Together, these results demonstrate that SENS-IS enables robust hazard identification and potency characterisation for a broad range of pre- and pro-haptens, and that the optional use of full-thickness skin models may extend its applicability to metabolically demanding substances while maintaining interpretability and avoiding non-specific signal amplification.