Translating in vitro mechanistic findings to in vivo toxicity outcomes: A case study of Usnic acid hepatotoxicity

New Approach Methodologies (NAMs) utilizing cellular models provide mechanistic insight into chemical hazards, but their utility in human health risk assessment depends on demonstrating how effectively in vitro findings translate to real-world outcomes. Quantitative in-vitro-to-in-vivo extrapolation (QIVIVE), supported by physiologically based pharmacokinetic (PBPK) modeling, enables this translation by linking in vitro concentration-response data to in vivo dose-response relationships. Here, we evaluated QIVIVE performance using usnic acid as a model compound. Usnic acid was selected because in vitro data and clinical case reports are available, enabling direct assessment of translational accuracy. Bayesian Benchmark Doses (BBMDs) were derived from in vitro toxicity data spanning thirteen endpoints in three hepatic cell models. Freely dissolved concentrations were estimated using a mass balance distribution model and incorporated into QIVIVE to predict equivalent administered doses (EADs), the in vivo doses expected to reproduce observed in vitro effects. Three PBPK platforms (GastroPlus, Berkeley Madonna, and httk R package) were applied. Predicted EADs aligned with reported exposure levels associated with hepatotoxicity, supporting the translational relevance of QIVIVE. Results indicated activation of multiple hepatotoxicity pathways, including redox disturbance, stress response, and DNA damage, across 1-384 mg oral doses. At human relevant intake levels, these mechanisms may explain the progression to liver injury in consumers of usnic acid. This study demonstrates that QIVIVE has the potential to translate in vitro mechanisms into possible human health outcomes, offering more mechanistically informed evidence for safety assessments of botanical constituents.