Background Although hyperuricemia (HUA) is required for gout development, only a fraction of individuals with elevated serum urate progress to the disease. Metabolomic profiling may help uncover biological mechanisms and improve prediction of gout onset. Methods We conducted a prospective metabolomics analysis among 32,563 HUA individuals in the UK Biobank. Baseline plasma metabolites were measured, and differences between individuals who developed gout (n = 2,856) and those who did not were identified. Cox proportional hazards models with false discovery rate correction assessed metabolite-gout associations. Significant metabolites and clinical factors were selected using LASSO-Cox and stepwise regression to construct a gout risk score (GRS). Model performance was evaluated in a test set using survival analysis and time-dependent receiver operating characteristic (ROC) curves. Sensitivity and subgroup analyses, along with validation in an independent validation set and transcriptomic profiling, tested the robustness and biological relevance of results. Results The final GRS incorporated six metabolites and three clinical variables. It was strongly associated with incident gout (HR = 2.94, 95% CI 2.66–3.23), and individuals in the top quartile had markedly higher risk. The model showed stable predictive ability, with time-dependent AUCs of 0.81–0.79 in the training set and 0.85–0.79 in the test set across 1–10 years. Further validation of the GRS in the independent validation set confirmed the performance. Transcriptomic analyses independently revealed enrichment of inflammatory and lipid-metabolic pathways, consistent with the metabolites included in the GRS. Conclusions We developed a lipid- and inflammation-related GRS that effectively predicts gout onset among HUA individuals, offering a useful tool for early risk stratification and targeted prevention. <Table Float="No" ID="Taba"> <tgroup cols="2"> <colspec align="left" colname="c1" colnum="1" /> <colspec align="left" colname="c2" colnum="2" /> <tbody> <row> <entry align="left" nameend="c2" namest="c1"> Key Points • An NMR-based metabolomics analysis identified several lipid- and inflammation-related metabolites strongly associated with incident gout among individuals with hyperuricemia. • A gout risk score integrating six metabolites and three clinical factors demonstrated robust and stable predictive performance over up to 10 years of follow-up. • Individuals in the highest quartile of the metabolite-based risk score had substantially elevated gout risk, improving early identification of high-risk subgroups. • Transcriptomic profiling revealed enrichment of inflammatory and lipid-metabolic pathways, supporting the biological plausibility of the findings. </entry> </row> </tbody> </tgroup> </Table>

A Lipid- and Inflammation-Related Metabolite Risk Score Predicts Incident Gout Among Individuals With Hyperuricemia: A Prospective Cohort Study

期刊论文
发表日期：

2026年5月8日

查看全文

Keke Ding,
Min Zhu,
Xiaojiao Zheng,
Haoyong Yu,
Fengjing Liu,
Si Chen,
Wei Jia,
Tianlu Chen

摘要

Background

Although hyperuricemia (HUA) is required for gout development, only a fraction of individuals with elevated serum urate progress to the disease. Metabolomic profiling may help uncover biological mechanisms and improve prediction of gout onset.

Methods

We conducted a prospective metabolomics analysis among 32,563 HUA individuals in the UK Biobank. Baseline plasma metabolites were measured, and differences between individuals who developed gout (n = 2,856) and those who did not were identified. Cox proportional hazards models with false discovery rate correction assessed metabolite-gout associations. Significant metabolites and clinical factors were selected using LASSO-Cox and stepwise regression to construct a gout risk score (GRS). Model performance was evaluated in a test set using survival analysis and time-dependent receiver operating characteristic (ROC) curves. Sensitivity and subgroup analyses, along with validation in an independent validation set and transcriptomic profiling, tested the robustness and biological relevance of results.

Results

The final GRS incorporated six metabolites and three clinical variables. It was strongly associated with incident gout (HR = 2.94, 95% CI 2.66–3.23), and individuals in the top quartile had markedly higher risk. The model showed stable predictive ability, with time-dependent AUCs of 0.81–0.79 in the training set and 0.85–0.79 in the test set across 1–10 years. Further validation of the GRS in the independent validation set confirmed the performance. Transcriptomic analyses independently revealed enrichment of inflammatory and lipid-metabolic pathways, consistent with the metabolites included in the GRS.

Conclusions

We developed a lipid- and inflammation-related GRS that effectively predicts gout onset among HUA individuals, offering a useful tool for early risk stratification and targeted prevention.

Key Points

• An NMR-based metabolomics analysis identified several lipid- and inflammation-related metabolites strongly associated with incident gout among individuals with hyperuricemia.

• A gout risk score integrating six metabolites and three clinical factors demonstrated robust and stable predictive performance over up to 10 years of follow-up.

• Individuals in the highest quartile of the metabolite-based risk score had substantially elevated gout risk, improving early identification of high-risk subgroups.

• Transcriptomic profiling revealed enrichment of inflammatory and lipid-metabolic pathways, supporting the biological plausibility of the findings.