Authors: Ciara Costello1*, Melody Chemaly 2, Aaron Peace 3, Steven Watterson1, Victoria McGilligan1

Affiliations: 1 Personalised Medicine Centre, School of Medicine, Ulster University, Northlands Road, Derry~Londonderry, BT48 7JL. 2 Division of Vascular Surgery, Department of Molecular Medicine and Surgery, Karolinska University Hospital and Karolinska Institutet, Stockholm, Sweden. 3 Department of Cardiology, Western Health and Social Care Trust, Altnagelvin Hospital, Derry~Londonderry, United Kingdom, BT47 6SB

Background/ Introduction: Coronary artery disease (CAD) is a heterogeneous inflammatory disorder in which individual patients exhibit distinct molecular and immune profiles that may influence disease progression and therapeutic response. Although chronic inflammation is a recognised driver of atherosclerosis, the contribution of specific inflammatory pathways to clinically defined CAD phenotypes remains incompletely characterised. We aimed to identify circulating proteomic signatures associated with CAD severity and evaluate the contribution of NLRP3 inflammasome-related pathways as candidates for biomarker-guided patient stratification.

Material & Methods: A cohort of 344 participants from Northern Ireland was stratified by HEART score into low-risk (LR), moderate-risk (MR), high-risk (HR), and very high-risk (VHR) groups. VHR participants were further classified as acute coronary syndrome (ACS; sampled within 20 days of an event) or elective CAD (ELEC; no recent cardiovascular event). Plasma proteins were quantified using the Olink Proximity Extension Assay across 552 cardiometabolic and inflammatory targets. Following quality control, normalisation and missing-value imputation, principal component analysis (PCA) and age- and sex-adjusted differential expression analyses were performed. Statistical significance was determined using false discovery rate (FDR)-adjusted linear models, followed by pathway enrichment analysis using Enrichr.

Results: PCA demonstrated distinct clustering of VHR and LR participants, indicating marked differences in circulating proteomic profiles. Differential expression analysis identified 77 proteins that were consistently altered in VHR, ACS and ELEC compared with LR participants (FDR < 0.05), representing a shared molecular signature of CAD. The ELEC versus LR comparison yielded the greatest number of differentially expressed proteins (94), whereas only three proteins differed between ACS and ELEC, suggesting that the circulating proteome is dominated by persistent disease-associated changes rather than transient acute responses at the time of sampling. Pathway enrichment analysis revealed significant over-representation of immune and inflammatory pathways, with approximately 20% of differentially expressed proteins associated with NLRP3 inflammasome-related signalling.

Conclusion: Comprehensive plasma proteomic profiling identified a reproducible NLRP3 inflammasome-associated molecular signature across clinically defined CAD phenotypes. These findings support inflammasome-related proteins as candidate biomarkers for molecular patient stratification and strengthen the rationale for precision therapeutic strategies targeting the NLRP3 pathway in coronary artery disease.