Advances in antituberculosis drug-induced liver injury: Molecular mechanisms, genetic insights, and emerging animal-free prediction models
Yang R, Jin Z, Jin X, Li Y, Wang D, Chao J
Toxicology · 2025-11
Abstract
Approximately 10 million tuberculosis patients worldwide require drug therapies annually. However, prolonged use of tuberculocidal drugs in combination regimens may induce hepatotoxicity, ranging from asymptomatic serum transaminase elevation to acute liver failure, which often necessitates treatment discontinuation. Antituberculosis drug-induced liver injury (AT-DILI) represents a predominant cause of clinical DILI in humans globally. The multifactorial and incompletely uncovered mechanisms underlying AT-DILI necessitate human-relevant models to predict liver injury progression and guide timely interventions. This review synthesizes current knowledge of mechanisms implicated in AT-DILI, focusing on metabolic activation of first-line drugs, immune-mediated hepatocyte damage, and cholestasis via transporter inhibition. Genetic polymorphisms in N-acetyltransferase 2, Cytochrome P450 2E1 and glutathione S-transferase genes are highlighted as critical determinants of interindividual susceptibility. Meanwhile, we critically evaluate the evolution of animal-free prediction models, spanning in vitro systems (2D hepatocytes and 3D culture system including hepatic spheroid, organoids, microfluidic systerms, and 3D bioprinting) to computational approaches such as quantitative structure-activity relationship modeling and machine learning. These models address ethical concerns, interspecies variabilities, and translatability limitations inherent to traditional animal studies while enabling high-throughput toxicity profiling. Despite advancements, challenges persist in standardizing experimental architectures, reducing operational costs, establishing physiological fidelity, and validating computational predictions. Robust characterization of emerging animal-free platforms including spanning the physiological fidelity, multi-omics integration capacity, and clinical translatability is therefore indispensable for advancing regulatory-ready frameworks toward safer tuberculosis therapy.
MeSH terms
- Animals
- Humans
- Antitubercular Agents
- Chemical and Drug Induced Liver Injury