Abstract 4363130: Multi-Omic Analyses Identifies Novel Candidate Molecular Drivers of Coronary Artery Calcification

Abstract

Background: Coronary artery calcification (CAC) is a strong independent predictor of CAD, yet the underlying molecular mechanisms remain incompletely understood. We aimed to identify novel molecular mediators of CAC using a multi-omics approach, with a focus on those proteins likely to have a direct effect on the vasculature. Methods: We conducted a proteome-wide association study in a pooled cohort of CARDIA, MESA, and FHS participants with Olink plasma proteomic profiling and CT-based CAC scoring (n = 6468). We identified proteins independently associated with CAC after adjusting for established CAC risk factors (age, sex, BMI, hypertension, diabetes mellitus, chronic kidney disease, statin use, ethnicity, and smoking status). These proteins were further evaluated for their ability to predict myocardial infarction in an expanded cohort of MESA, FHS, and UK Biobank participants (n=57,198). To prioritize proteins involved in vascular smooth muscle cell (VSMC)-mediated calcification, we identified proteins with differential gene expression in human VSMCs cultured in osteogenic media versus normal media. Causal effects of these proteins were evaluated using Two-sample Mendelian randomization (using inverse variance weighting (IVW), Egger and maximum likelihood) with plasma proteomics GWAS summary statistics as exposures and the largest existing CAC GWAS meta-analysis summary statistics as the outcome. Results: Of 2,805 proteins measured, 365 were independently associated with CAC scores (FDR p<0.05), and enriched in pathways related to inflammatory response, immune cell signaling, ECM organization, and key vascular cell signaling pathways (Fig. 1). Among these, 216 proteins also independently predicted myocardial infarction. Transcriptomic analysis in human VSMCs identified 43 genes upregulated under osteogenic conditions, of which 8 demonstrated evidence of causal effects on CAC via Mendelian Randomization. These proteins are implicated in pathways related to immune cell signaling, bone signaling, and metabolic regulation. Conclusion: Using a multi-omics strategy integrating proteomic, transcriptomic, and genomic data, we identified several novel candidate molecular drivers of CAC with supporting evidence for their roles in myocardial infarction and VSMC osteogenic phenotypic modulation. These findings provide promising avenues for future mechanistic studies in vascular calcification.