Idiosyncratic Drug-Induced Liver Injury (IDILI) is an adverse drug reaction manifesting exclusively in a limited subset of individuals[1]. The unpredictability and diagnostic difficulty of IDILI hinder both drug development and clinical medication safety. Although the precise pathogenesis of IDILI remains incompletely understood, current evidence indicates that this phenomenon stems from a complex interplay between pharmacological properties and host-specific factors—including genetic susceptibility, underlying diseases, and immune status—which collectively determine individual susceptibility and clinical manifestations[2], [3].

Polygonum multiflorum (PM), derived from the dried tuberous root of Polygonum multiflorum Thunb., is a traditional Chinese medicine used to nourish blood, and alleviate constipation. However, its expanded clinical use has led to increasing reports of PM-induced liver injury in the literature.[4], [5], [6]. Domestic and international studies reveal that while high-dose, long-term PM treatment can induce liver injury in normal animals, PM more frequently causes characteristic idiosyncratic liver injury[7], [8]. Previous work by our team employing multi-cytokine profiling in prospective clinical serum samples revealed that immune dysregulation—particularly elevated inflammatory cytokine levels—strongly correlates with PM-induced liver injury. Tumor necrosis factor-α (TNF-α) emerged as a potential biomarker highly associated with susceptibility to PM-induced liver injury[9]. We subsequently established a TNF-α-mediated susceptibility model that successfully evaluated the hepatotoxic effects of PM[10]. However, the molecular mechanisms underlying the interaction between PM and this susceptibility factor remain elusive, and the precise genes/targets modulated by this interaction remain to be fully characterized.

Network pharmacology, grounded in systems biology theory, elucidates the complexity of compounds, diseases, and biological systems through a network-based approach to predict their interaction mechanisms[11], [12]. Lipidomics studies explores the role of lipids in physiology and disease by profiling metabolic changes. This approach is now central to studying liver diseases, aiding in diagnosis, monitoring, and toxicity assessment.[13], [14]. Transcriptomics serves as a functional bridge between genes and proteins by comprehensively profiling gene expression. As a powerful method for acquiring organism-wide gene expression data, it enables robust detection of differentially expressed genes and systematic analysis of metabolic pathway regulation[15], [16]. Transcriptomics powers the elucidation of molecular mechanisms and biomarker prediction, especially in complex diseases[17]. Integrated multi-omics analysis provides a scientific foundation for more effective discovery of biomarkers related to disease mechanisms and therapeutic responses. For example, Zhang et al[18]. employed multi-omics techniques to elucidate the underlying mechanism of PS-NP-induced hepatotoxicity. Teka et al[19]. used a combination of multi-omics and Western blotting to demonstrate that cis-TSG reduces liver injury via lipid metabolism modulation. Ge et al[20]. elucidated polystyrene nanoplastics-induced liver injury through lipidomic and transcriptomic analyses.

Therefore, this study employs integrated network pharmacology, lipidomics, and transcriptomics to investigate PM-induced liver injury mechanisms under the influence of susceptibility factors, aiming to establish a scientific basis for rational clinical use of PM and attenuate associated liver injury risks.