IL-6 modifies Lp(a)-associated coronary risk but not aortic stenosis risk: UK Biobank (JAMA Cardiol 2026)

Original title: Lipoprotein(a), Inflammation, and Risk of Coronary Artery Disease and Aortic Valve Stenosis

JAMA Cardiol · 1 August 2026 · 6

Mohammadnia N, Li L, Ezzat D, Yu Z, Raghu VK, Halford JL, Riksen NP, Natarajan P, Cornel JH, Honigberg MC, El Messaoudi S

In 43,512 UK Biobank participants (16 percent with Lp(a) of 125 nmol/L or more) followed 13.5 years, the hazard ratio for incident CAD with elevated Lp(a) was 1.43 in the top IL-6 quartile and 1.09 in the bottom (interaction P = 0.008); no inflammatory marker modified the Lp(a)-aortic stenosis association. The inflammation interaction returns in a subtler form: not CRP, and not for the valve.

Original abstract

Importance: Clinical heterogeneity exists among individuals with elevated lipoprotein(a) [Lp(a)] levels. Prior studies suggest that low-grade inflammation may modify this risk, but results remain conflicting.

Objective: To test the hypothesis that inflammatory biomarkers may modify Lp(a)-associated risk for coronary artery disease (CAD) and aortic valve stenosis (AS) in a primary prevention population.

Design, Setting, And Participants: Recruitment occurred between March 2006 through October 2010 among a population-based cohort of UK adults. UK Biobank participants without prevalent CAD or AS who underwent plasma proteomic profiling at study entry were eligible for inclusion. Data were analyzed from June 2025 through April 2026.

Exposures: Lp(a) levels (<125 nmol/L vs ≥125 nmol/L) and inflammatory biomarkers interleukin 1β (IL-1β), IL-18, IL-6, and the neutrophil to lymphocyte ratio (NLR).

Main Outcomes And Measures: The primary outcome was the risk of incident CAD. The secondary outcome was the risk of incident AS.

Results: Among 43 512 UK Biobank participants, 6975 (16.0%) had Lp(a) levels of 125 nmol/L or higher, 24 079 (55.3%) were female, and overall mean (SD) age was 56.5 (8.2) years. Median (Q1-Q3) follow-up was 13.5 (12.7-14.3) years for incident CAD and 13.6 (12.9-14.4) years for incident AS. Multivariable-adjusted Cox proportional hazards models tested the interaction between Lp(a) levels and each inflammatory biomarker in relation to the primary and secondary outcomes. Among inflammatory biomarkers tested, IL-6 demonstrated the strongest association with both incident CAD and AS. IL-6 levels modified Lp(a)-associated risk for incident CAD (hazard ratio [HR] for Lp(a) ≥125 nmol/L vs <125 nmol/L in quartile 4 of IL-6: 1.43; 95% CI, 1.25-1.63 vs in quartile 1 of IL-6: HR, 1.09; 95% CI, 0.85-1.38; P for interaction = .008). The interaction between Lp(a) and the NLR (P for interaction = .02) was suggestive but not significant after adjustment for multiple testing. No inflammatory biomarkers modified Lp(a)-associated risk for AS.

Conclusions And Relevance: In this primary prevention cohort study, Lp(a)-associated risk for incident CAD was modified by IL-6, with lower associated risk observed in the setting of lower inflammatory biomarker levels. These findings identify IL-6 as a biomarker of inflammatory risk that may modify Lp(a)-associated CAD risk.

Summary written by lp-a.org from the published abstract; figures as published. Page updated 17 August 2026. Methods.