OA05 Harnessing high throughput multiplex immunoassays and protein association networks to identify diagnostic protein biomarkers of rheumatoid arthritis-cardiovascular disease (RA-CVD) multimorbid axis

Abstract

Abstract Background/Aims People with rheumatoid arthritis (RA) have higher mortality rates compared to the general population, primarily due to excess cardiovascular (CV) disease (CVD). The earliest stages of RA are associated with CV abnormalities but biomarkers to identify CVD in early RA are lacking. We aimed to identify focused protein biomarkers that associate with cardiac magnetic resonance imaging (CMR) abnormalities in a treatment-naïve, new-onset RA trial cohort and the predominant inflammatory and metabolic pathways involved in RA-CVD. Methods CADERA (Coronary Artery Disease in Early RA) was a bolt-on study to a parent RCT in an early RA inception cohort. 81 treatment-naïve early RA patients underwent CMR at baseline as part of CADERA. Proximity extension immunoassay (OIink) was used to measure normalised protein expression across inflammation, cardiovascular II/III and cardiometabolic panels in CADERA patients. Bayesian mixed effects linear regression was used to identify significant proteins associated with CMR parameters of myocardial oedema/fibrosis and vascular stiffness at baseline. Then, an expanded protein-protein interaction (PPI) network using above proteins was created using an induced network approach (String-DB) and k-means clustering applied to identify protein clusters that were subsequently subjected to Gene ontology (GO) and KEGG enrichment analysis. Results Of 340 proteins analysed using Olink, 108 proteins were associated with CMR markers of myocardial fibrosis/oedema (64 proteins) and vascular stiffness (44 proteins) Table 1 highlights the top 5 proteins for each CMR parameter. K-means clustering of the expanded PPI network identified 4 clusters with roles in vascular endothelial growth factor receptor binding and JAK-STAT signalling pathway (cluster 1); IL27 receptor binding and NF-kappa B signalling pathway (cluster 2); Macrophage CSF receptor binding and ErbB signalling pathway (cluster 3); and IL10 receptor activity and complement and coagulation cascades (cluster 4). Conclusion This is the first study to identify protein biomarkers of RA-CVD, an area of unmet clinical need, using a combination of targeted high throughput immunoassays, Bayesian and network analyses. Targeted techniques such as Olink may help identify potential diagnostic biomarkers of RA-CVD. Pathway analysis provides insight into the inter-relationship between inflammation and metabolic mediators of RA-CVD axis. Disclosure R. Shukla: None. N. Black: None. D. Plant: None. C. Miller: None. S. Plein: None. M.H. Buch: None.