| Title: | Plasma proteomics linking primary and secondary diseases: Insights into molecular mediation from UK Biobank data |
| Journal: | Med |
| Published: | 4 Aug 2026 |
| Pubmed: | https://pubmed.ncbi.nlm.nih.gov/42551443/ |
| DOI: | https://doi.org/10.1016/j.medj.2026.101232 |
| URL: | https://www.medrxiv.org/content/medrxiv/early/2025/09/02/2025.08.29.25334726.full.pdf |
Publication 21211
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
Abstract
BACKGROUND: Diabetes, hypertension, and dyslipidemia are major risk factors for cardiovascular, neurological, renal, and pulmonary diseases, yet clinically accessible molecular mediators linking these cardiometabolic conditions to downstream complications remain unclear.</p>
METHODS: We analyzed plasma proteomic data from 53,030 UK Biobank participants with longitudinal follow-up. Mediation analysis evaluated circulating proteins linking three primary cardiometabolic diseases to 18 secondary outcomes. Mendelian randomization assessed potential causal relationships, and machine learning evaluated the predictive value of identified mediators.</p>
FINDINGS: We identified 998 significant mediation pathways involving 337 unique plasma proteins. GDF15 consistently mediated associations between diabetes and cardiovascular diseases, and ACE2 linked poorly controlled diabetes to increased risk of nerve root and plexus disorders. Mediators were enriched in receptor-mediated signaling and molecular-interaction pathways. Mendelian randomization supported potential causal roles for 44 proteins. Incorporating mediator proteins into machine-learning models improved prediction of secondary disease risk beyond traditional clinical factors and other plasma proteins.</p>
CONCLUSIONS: Plasma proteins help mediate progression from cardiometabolic diseases to downstream complications. These findings provide a molecular map of disease mediation pathways, nominate biomarkers and therapeutic targets, and support improved risk stratification and targeted intervention.</p>
FUNDING: This this study was supported by the National Institutes of Health, the American Heart Association, and institutional funding.</p>
18 Keywords
- Biological Specimen Banks
- Biomarkers
- Blood Proteins
- Cardiovascular Diseases
- Diabetes Mellitus
- Dyslipidemias
- Female
- Growth Differentiation Factor 15
- Humans
- Hypertension
- Machine Learning
- Male
- Mendelian Randomization Analysis
- Middle Aged
- Peptidyl-Dipeptidase A
- Proteomics
- UK Biobank
- United Kingdom
5 Authors
- Hanyu Qian
- Chao Wu
- Bo Li
- Anthony Rosenzweig
- Meng Wang
Enabling scientific discoveries that improve human health