Metabolic signatures and machine learning identify gut-liver-heart axis dysfunction as a potential link to major adverse cardiovascular events in coronary artery disease

Q: What are the key findings of this study?

An XGBoost machine learning model utilizing circulating metabolic signatures robustly discriminated patients with CAD and MACE from non-CAD controls with an AUC of 0.984.

Q: What is the clinical evidence from this study?

Study design: Case-Control. Population: Coronary artery disease (CAD) and major adverse cardiovascular events (MACE) (n=800). Intervention: Circulating metabolic signatures (gut-liver-heart axis dysfunction) vs. Non-CAD controls and CAD without MACE. Primary outcome: Discrimination of CAD with MACE from non-CAD controls using an XGBoost machine learning model (AUC 0.984).

Q: What does this research mean for the field?

Machine learning models utilizing circulating metabolic signatures of gut-liver-heart axis dysfunction—specifically elevated oxidized linoleic acid metabolites and depleted primary bile acids with elevated FGF19—accurately discriminate coronary artery disease patients with major adverse cardiovascular events. Novelty: ClaimNovelty.INCREMENTAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

Min-Ge Lin; Chun-Ka Wong; Ka‐Wing Au; Chloe Yu-Yan Cheung; Yee-Man Lau; Song-Yan Liao; Karen Siu-Ling Lam; Aimin Xu; Hung-Fat Tse

doi:10.1186/s40842-026-00329-w

August 11, 2026Cardiovascular Diabetology – Endocrinology ReportsOpen Access

Metabolic signatures and machine learning identify gut-liver-heart axis dysfunction as a potential link to major adverse cardiovascular events in coronary artery disease

View Full Paper

Ask AI

Bookmark

Share

Key result

XGBoost model using metabolic signatures discriminates CAD with MACE from controls with ~0.98 AUC.

AUC 0.984
n=800

Figures

Why the study?

Traditional risk factors do not fully account for residual cardiometabolic risk of MACE in CAD, motivating identification of circulating metabolic signatures and gut-liver-heart axis mechanisms.

Can untargeted metabolomics and machine learning identify circulating metabolic signatures associated with MACE susceptibility in patients with CAD?

Key Points

To identify circulating metabolic signatures linked to major adverse cardiovascular events (MACE) in coronary artery disease and evaluate gut-liver-heart axis disruption.
Conducted untargeted HPLC-MS metabolomics on fasting serum from 200 patients with CAD and MACE, 200 with CAD without MACE, and 400 matched non-CAD controls (N=800).
Applied univariate analysis, metabolite set enrichment, and an XGBoost machine learning model to profile metabolic pathways and risk predictors.
Quantified serum fibroblast growth factor 19 (FGF19) levels using enzyme-linked immunosorbent assay (ELISA) to assess enterohepatic endocrine disruption.
The XGBoost machine learning model discriminated CAD patients with MACE from non-CAD controls (AUC = 0.984) and from CAD patients without MACE (AUC = 0.932).
Pathway enrichment revealed marked dysregulation of linoleic acid metabolism and PPAR signaling (p < 0.05), with significant elevations of pro-inflammatory 9- and 13-HODE in the MACE cohort.
Atheroprotective primary bile acids were significantly depleted in CAD patients (p < 0.001) alongside elevated circulating FGF19 levels (p = 0.003).

Study Design

Type

Case-Control (n=800)

Multicenter

No

Structured PICO

Can untargeted metabolomics and machine learning identify circulating metabolic signatures associated with MACE susceptibility in patients with CAD?

P

Population

800 participants including 200 patients with CAD and MACE, 200 with CAD without MACE, and 400 matched non-CAD controls, with a mean age of 62 years.

E

Exposure

Untargeted high-performance liquid chromatography-mass spectrometry (HPLC-MS) metabolomic profiling combined with eXtreme Gradient Boosting (XGBoost) machine learning

C

Comparator

Patients with CAD without MACE and matched non-CAD controls

O

Outcome

Discrimination of patients with CAD and MACE from non-CAD controls and CAD without MACE using an XGBoost machine learning model based on metabolic signaturessurrogate

Main Result

Effect estimate: AUC 0.984

Machine learning applied to untargeted metabolomics identifies gut-liver-heart axis dysfunction, specifically linoleic acid and PPAR signaling dysregulation, as strongly associated with MACE susceptibility in CAD patients.

Limitations

Retrospective case-control design precludes causal inference and introduces imbalances in baseline characteristics.
Employed the full metabolomics dataset as input without stepwise reduction of input variables.
Absence of an external validation dataset limits the evaluation of the XGBoost model's performance.
Did not validate the levels of downstream metabolites identified in the pathway analysis by targeted metabolomics.
Lack of matched fecal microbiome sequencing prevents linking circulating bile acid shifts to specific microbial disturbances.
Retrospective case-control design precludes causal inference and introduces imbalances in baseline characteristics
Lack of detailed medication data
Full metabolomics dataset used as input without stepwise reduction of variables
Absence of an external validation dataset
Lack of targeted metabolomics validation for downstream metabolites
Lack of matched fecal microbiome sequencing to link circulating bile acid shifts to specific microbial disturbances

Ask about this paper

Answers are grounded in this paper and cite their sources.

Implication

Metabolomic signatures may aid MACE risk stratification in CAD; leaves open whether targeting gut-liver-heart pathways alters outcomes.

Abstract

BACKGROUND: Traditional risk factors do not fully account for the residual cardiometabolic risk of major adverse cardiovascular events (MACE) in coronary artery disease (CAD). We aimed to identify circulating metabolic signatures associated with MACE susceptibility and uncover potential pathobiological mechanisms underlying the gut-liver-heart axis. METHODS: In this retrospective case-control study, untargeted high-performance liquid chromatography-mass spectrometry (HPLC-MS) was performed on fasting serum from 200 patients with CAD and MACE, 200 with CAD without MACE, and 400 matched non-CAD controls. Metabolomics data were processed using univariate analysis, multivariate analysis, and eXtreme Gradient Boosting (XGBoost) machine learning. Pathway enrichment was conducted using metabolite set enrichment analysis. Circulating fibroblast growth factor 19 (FGF19) was quantified via enzyme-linked immunosorbent assay to validate enterohepatic endocrine disruption. RESULTS: The MACE cohort exhibited a pronounced cardiometabolic phenotype, characterized by significantly highest rates of diabetes, hypertension, and dyslipidemia (p < 0.01). The XGBoost model robustly discriminated patients with CAD and MACE from non-CAD controls (area under the curve [AUC] = 0.984) and from patients with CAD without MACE (AUC = 0.932). Pathway analysis revealed marked dysregulation of linoleic acid metabolism and peroxisome proliferator-activated receptor (PPAR) signaling (p < 0.05). Specifically, pro-inflammatory oxidized linoleic acid metabolites, including 9- and 13-hydroxyoctadecadienoic acid (HODE)-which drive plaque instability-were significantly elevated in the MACE cohort. Furthermore, atheroprotective primary bile acids were significantly depleted in patients with CAD (p < 0.001). This depletion was accompanied by an elevated serum FGF19 level (p = 0.003), reflecting a potential disruption of the gut-liver-heart endocrine axis. CONCLUSIONS: In conclusion, dysregulated linoleic acid oxidation, altered PPAR signaling, and disturbed primary bile acid-FGF19 metabolism may represent key metabolic pathways associated with MACE susceptibility. Integrating these gut-liver-heart axis signatures into machine learning models holds significant promise for refining cardiovascular risk stratification and guiding targeted preventive interventions.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Lin et al. (2026) conducted a case-control in Coronary artery disease (CAD) and major adverse cardiovascular events (MACE) (n=800). Circulating metabolic signatures (gut-liver-heart axis dysfunction) vs. Non-CAD controls and CAD without MACE was evaluated on Discrimination of CAD with MACE from non-CAD controls using an XGBoost machine learning model (AUC 0.984). An XGBoost machine learning model utilizing circulating metabolic signatures robustly discriminated patients with CAD and MACE from non-CAD controls with an AUC of 0.984.

synapsesocial.com/papers/6a7ec6eab70b84ec8b913077 https://doi.org/10.1186/s40842-026-00329-w

View Full Paper

Ask AI

Bookmark

Share

Metabolic signatures and machine learning identify gut-liver-heart axis dysfunction as a potential link to major adverse cardiovascular events in coronary artery disease

Figures

Key Points

Study Design

Structured PICO

Main Result

Limitations

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Metabolic signatures and machine learning identify gut-liver-heart axis dysfunction as a potential link to major adverse cardiovascular events in coronary artery disease

Figures

Key Points

Study Design

Structured PICO

Main Result

Limitations

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading