Truace
Subscribe

Use of an AI metabolomics score to predict cardiovascular risk among patients with Psoriasis in the MGB and UK Biobanks

Background: Systemic inflammation in psoriasis accelerates atherosclerosis and confers excess cardiovascular risk that is not fully captured by traditional clinical predictors. Metabolomic profiling may identify disease-specific metabolic signatures associated with cardiovascular events and enhance risk prediction beyond conventional factors. Objectives: To determine whether plasma metabolomic profiling improves cardiovascular risk prediction in patients with psoriasis and to develop and externally validate an a…

By the Truvace desk1 min read1 sourcePeer-reviewedTRV-2026-2036· cite & verify
Use of an AI metabolomics score to predict cardiovascular risk among patients with Psoriasis in the MGB and UK Biobanks
Untargeted metabolomics profiles delineate metabolic alterations in mouse plasma during lung carcinoma development using UPLC-QTOF-MS in MSE mode by Wu, Huan; Chen, Yang; Li, Zegeng; Liu, Xianhua. CC BY 4.0 · https://creativecommons.org/licenses/by/4.0

In brief

Researchers used an extreme gradient boosting algorithm on nuclear magnetic resonance measurements of circulating metabolites to build MetaPsoAI. It was developed in 1086 Mass General Brigham Biobank participants with psoriasis and no prior coronary artery disease, then tested in the UK Biobank.

Psoriasis carries cardiovascular risk that standard predictors do not fully capture, so better stratification could guide earlier prevention. The reported results come from biobank cohorts, and the authors describe the benefit as potential rather than demonstrated in clinical care.

Main points

  1. The development cohort had 1086 people with psoriasis and no prior coronary artery disease, with 168 metabolites quantified.
  2. HDL subfractions and amino acid metabolites were among the metabolites associated with cardiovascular outcomes.
  3. In the MGB cohort the top MetaPsoAI quartile had a 3.6-fold higher risk of myocardial infarction or stroke.
  4. In the UK Biobank the top quartile had a 1.9-fold higher risk among psoriasis patients, with stronger effects than in non-psoriasis controls.

The gain

A machine-learning score built from 168 blood metabolites separated psoriasis patients by risk of myocardial infarction or stroke, which may support more targeted cardiovascular prevention.

The rundown

The study starts from the premise that systemic inflammation in psoriasis accelerates atherosclerosis and creates excess cardiovascular risk. Metabolomic profiling was tested as a way to find disease-specific signatures that improve on conventional factors.

Analyses first linked individual metabolites to incident all-cause death or myocardial infarction in the MGB Biobank. The model was then trained to predict myocardial infarction or stroke.

External validation used the UK Biobank, with about 274,000 participants stratified by psoriasis status. Risk gradients were larger in psoriasis patients than in controls, though lower in magnitude than in the MGB cohort.

Sources

  1. Peer-reviewedAmerican Journal of Preventive Cardiology2026-06-06

The debate