A well-characterized mechanistic model for exploring known or hypothesized T cell mediated drug induced liver injury: current capabilities and challenges for future predictivity.

Background: Drug-induced liver injury (DILI) is an adverse event whose emergence can slow or halt drug development programs. Adaptive immune responses have been implicated for several DILI compounds, and drug-specific T cell responses have been characterized, but there are still many unknowns. We describe the extension of a quantitative systems toxicology (QST) model of DILI to include CD8⁺ T cell-mediated DILI.

Research design and methods: To overcome deficits in quantitative data characterizing CD8⁺ T cell-mediated DILI, a translational strategy leveraged a well-defined mouse ovalbumin (OVA) antigen model and adapted it to represent mouse amodiaquine (AQ)-specific CD8⁺ T cell-mediated DILI, with further adaptations to represent human AQ-specific CD8⁺ T cell-mediated DILI.

Results: DILIsym reproduced published data characterizing mouse OVA-specific CD8⁺ T cell-mediated hepatotoxicity, mouse AQ-specific CD8⁺ T cell-mediated DILI, and human AQ-specific CD8⁺ T cell-mediated DILI. Development identified main drivers of the CD8⁺ T cell response, as well as areas where in vitro assay data could inform the simulation of additional compounds.

Conclusions: The DILIsym CD8⁺ T cell sub-model is well-positioned for systematic testing to improve our understanding of CD8⁺ T cell-mediated DILI. It is not yet predictive but indicates a promising direction to reduce DILI events in drug development.