ORIGINAL RESEARCH article
Front. Pharmacol.
Sec. Gastrointestinal and Hepatic Pharmacology
Machine learning-assisted analysis of serum metabolomics for Identifying Biomarkers in Intrinsic and Idiosyncratic Drug-Induced Liver Injury
Provisionally accepted
- Xiamen Haicang Hospital, Xiamen, China
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Objective: This study aimed to investigate metabolic mechanisms differentiating intrinsic and idiosyncratic drug-induced liver injury (DILI) using high-performance chemical isotope labeling liquid chromatography-mass spectrometry (HP-CIL LC-MS) and to identify predictive biomarkers through machine learning. Methods: According to EASL guidelines, 44 DILI patients were classified into intrinsic (n=17) and idiosyncratic (n=27) types. Serum metabolomic profiling was performed using HP-CIL LC-MS. Differentially expressed metabolites identified through both univariate and multivariate analyses were selected for receiver operating characteristic (ROC) analysis. Machine learning models were developed to distinguish DILI subtypes and evaluate biomarker performance. Results: Four metabolites—Alanyl-Glycine, N2-Acetyl-L-Cystathionine, and two isomers of 5-Hydroxyindoleacetic acid—were identified as potential biomarkers. All machine learning models achieved AUC values >0.8, with the multiple regression model showing exceptional performance (AUC=0.983 in cross-validation, 0.935 in holdout validation). Pathway analysis revealed significant alterations in amino acid metabolism pathways, including tryptophan, tyrosine, and cysteine-methionine metabolism. Conclusion: Machine learning-assisted serum metabolomics effectively characterizes metabolic differences between intrinsic and idiosyncratic DILI and identifies promising biomarkers. These findings provide insights for mechanistic subtyping of DILI induced by various hepatotoxic drugs.
Keywords: Metabolomics, Drug-Induced Liver Injury, intrinsic, Idiosyncratic, biomarkers, machine learning
Received: 17 Oct 2025; Accepted: 21 Nov 2025.
Copyright: © 2025 Wei, Wei, Huang and Lian. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Xianni Wei, 511457572@qq.com
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.