Mechanism-based drug safety testing using innovative <i>in vitro</i> liver models: from DILI prediction to idiosyncratic DILI liability assessment

Idiosyncratic drug-induced liver injury (iDILI) remains unpredictable. As adverse responses arise in a small fraction of patients, drugs often fail in later drug development stages or post-approval, thereby tremendously increasing costs and putting patients at risk, highlighting the need for accurate early identification of iDILI liabilities. Using articles from the last 5 years (PubMed), iDILI risk factors are described, <i>in vitro</i> liver models and mechanism-based readout strategies are evaluated on their potential to enable iDILI liability assessment. Various <i>in vitro</i> liver models are established for disease modeling and DILI prediction. Drawbacks for each of these seem inevitable, making the evaluation of their application domain and iDILI liability assessment potential crucial. A tiered approach could be considered, whereby compounds are initially screened and flagged using simple fit-for-purpose models for DILI prediction, followed by multicellular liver models that integrate the current knowledge of iDILI onset in combination with mechanistic readouts. Multiplexing models within an integrated mechanism-based testing strategy could improve the safety assessment accuracy. Defined <i>in vitro</i> models should integrate critical hepatocyte intrinsic risk factors as well as adaptive immune system components to refine iDILI liability assessment.特异质性药物性肝损伤（Idiosyncratic drug-induced liver injury, iDILI）至今仍难以预测。由于仅小部分患者会出现不良反应，药物常在后期研发阶段或获批后遭遇失败，这不仅大幅推高研发成本，还会使患者面临风险，凸显了精准早期识别iDILI潜在风险特征的迫切需求。本研究依托近5年PubMed收录的文献，梳理了iDILI的风险因素，并评估了体外（in vitro）肝脏模型及基于机制的检测策略在iDILI潜在风险评估中的应用潜力。目前已建立多种体外肝脏模型用于疾病建模与药物性肝损伤（Drug-induced liver injury, DILI）预测，但各类模型均存在难以避免的局限性，因此对其应用范围及iDILI潜在风险评估能力的评估至关重要。可采用分层评估策略：首先利用适配用途的简易模型对化合物进行初筛与标记以预测DILI，随后结合当前对iDILI发病机制的认知，采用整合了机制性检测终点的多细胞肝脏模型开展后续评估。在整合的基于机制的检测策略中引入多重检测模型，可提升安全性评估的精准度。理想的体外模型应整合关键的肝细胞固有风险因素与适应性免疫系统组分，以完善iDILI潜在风险的评估工作。