A multidimensional proteomic framework of circulating inflammation identifies outcome-specific cardiovascular risk axes in 39,494 UK biobank participants
Abstract
Inflammation is central to cardiovascular disease, yet conventional biomarkers do not resolve the underlying organisation of circulating inflammatory biology. Among 44,787 UK Biobank participants passing proteomic quality control, 736 inflammation-related proteins were used to derive orthogonal proteomic dimensions in a White Development sample (n = 20,830), with definitions fixed and applied unchanged in White Validation (n = 20,831). A CVD-free multimodal cohort of 39,494 participants provided the source population for downstream analyses; primary prospective associations were evaluated in White Development complete cases (n = 17,334) and replicated in White Validation (n = 17,190). Eight dimensions showed high structural stability and distinct protein, pathway, metabolic, imaging, and cardiovascular profiles. In the primary analysis, 23 of 48 RC outcome associations were FDR significant; RC2 was positively associated with all six outcomes, with the strongest association for heart failure (HR 1.395, 95% CI 1.273–1.527), while RC8 showed a more moderate but similarly broad pattern. RC4 was mainly associated with coronary outcomes, RC6 showed inverse associations with several endpoints, and RC7 was most clearly related to heart failure and atrial fibrillation. Major patterns reproduced in White Validation. Adding RC1–RC8 improved discrimination across all six outcomes (ΔC-index 0.0055–0.0136), with incremental gains retained beyond CRP, GlycA, and NLR. These findings support a reproducible multidimensional proteomic architecture of inflammation with heterogeneous cardiovascular relevance.
Related
- Machine learning clustering of multidimensional clinical features identifies four prognostically distinct sepsis subtypes
- Prevalence of Cardiovascular Risk Factors and Established Cardiovascular Disease Among All of Us Participants: Benchmarking Against National Estimates
- Association Between Plasma microRNA Levels and Major Adverse Cardiovascular Events in Participants from an Epidemiological Cohort
- Is adherence to the 24-hour movement guidelines related to a lower risk of cardiovascular events? A prospective cohort study including 13 897 participants from the SUN project.
- Large-scale proteomics identifies circulating markers of pulmonary pressure and pulmonary hypertension risk
BibTeX
@article{song2026multidimensional,
title = {{A multidimensional proteomic framework of circulating inflammation identifies outcome-specific cardiovascular risk axes in 39,494 UK biobank participants}},
author = {Xiangning Song and Haipeng Zhang and Guangcheng Liu and Zhujun Shen},
journal = {Scientific Reports},
year = {2026},
month = oct,
doi = {10.1038/s41598-026-73289-x},
url = {https://www.nature.com/articles/s41598-026-73289-x},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - A multidimensional proteomic framework of circulating inflammation identifies outcome-specific cardiovascular risk axes in 39,494 UK biobank participants
AU - Xiangning Song
AU - Haipeng Zhang
AU - Guangcheng Liu
AU - Zhujun Shen
PY - 2026
DA - 2026/10/13
JO - Scientific Reports
DO - 10.1038/s41598-026-73289-x
UR - https://www.nature.com/articles/s41598-026-73289-x
AB - Inflammation is central to cardiovascular disease, yet conventional biomarkers do not resolve the underlying organisation of circulating inflammatory biology. Among 44,787 UK Biobank participants passing proteomic quality control, 736 inflammation-related proteins were used to derive orthogonal proteomic dimensions in a White Development sample (n = 20,830), with definitions fixed and applied unchanged in White Validation (n = 20,831). A CVD-free multimodal cohort of 39,494 participants provided the source population for downstream analyses; primary prospective associations were evaluated in White Development complete cases (n = 17,334) and replicated in White Validation (n = 17,190). Eight dimensions showed high structural stability and distinct protein, pathway, metabolic, imaging, and cardiovascular profiles. In the primary analysis, 23 of 48 RC outcome associations were FDR significant; RC2 was positively associated with all six outcomes, with the strongest association for heart failure (HR 1.395, 95% CI 1.273–1.527), while RC8 showed a more moderate but similarly broad pattern. RC4 was mainly associated with coronary outcomes, RC6 showed inverse associations with several endpoints, and RC7 was most clearly related to heart failure and atrial fibrillation. Major patterns reproduced in White Validation. Adding RC1–RC8 improved discrimination across all six outcomes (ΔC-index 0.0055–0.0136), with incremental gains retained beyond CRP, GlycA, and NLR. These findings support a reproducible multidimensional proteomic architecture of inflammation with heterogeneous cardiovascular relevance.
SN - 2045-2322
ER -