Emil Rydell1, Lennart Jacobsson2 and Carl Turesson3, 1Lund University, Malmö, Sweden, 2Department of Rheumatology and Inflammation Research, Sahlgrenska Academy at Gothenburg University, Gothenburg, Sweden, 3Rheumatology, Department of Clinical Sciences, Malmö, Lund University, Malmö, Sweden
Background/Purpose: Rheumatoid arthritis (RA) is associated with an increased risk of cardiovascular disease (CVD). Systemic inflammation, driven by cytokines such as TNF, IL-6 and IL-17A, has been implicated as a contributing factor to CVD development in RA. The objective of this study was to investigate the impact of 92 inflammatory proteins on the risk of CVD events in patients with early RA.
Methods: Consecutive patients with early RA patients (symptom duration < 12 months), recruited 1995-2005 from a defined area, were followed in an observational study. Stored plasma samples from the baseline visit were analyzed for levels of 92 inflammatory proteins (Inflammation panel by O-link). Data on CVD events (coronary-, cerebrovascular- and peripheral artery disease) were retrieved from the Swedish national inpatient- and cause of death registries from 1969-2019. Statistical analyses were pre-designated as hypothesis-driven (Table 1) or exploratory. For the latter, principal component analysis was used to identify groups of proteins that explain the variance in the proteome. Within components selected based on Eigenvalues, proteins with a factor loading of >0.50 were investigated as potential predictors of CVD events. Protein levels were standardized to facilitate comparison. Patients with CVD events before RA diagnosis were excluded from the analyses. Primary outcome was the first CVD event, and secondary outcome was the first coronary artery disease (CAD) event during follow-up. Cox regression models (crude and age-sex adjusted) were used to assess the relationship between baseline biomarkers and outcomes. Models with significant predictors were further adjusted for ESR, to assess independent inflammatory pathways. For a priori hypothesis-based analyses, multiple testing was taken into account using Holm's correction.
Results: Baseline levels of proteins were available for 163 patients. A first ever CVD event occurred in 47 patients during follow-up, and a first ever CAD event in 37 patients. Among potential biomarkers with an a priori hypothesis, levels of IL-17A were significantly associated with overall CVD events in the crude and age-sex adjusted models (Table 1), and when additionally adjusted for baseline levels of ESR [adjusted hazard ratio 1.34; 95% confidence interval 1.03-1.76), while the associations did not reach significance for CAD events (Table 1). Osteoprotegerin (OPG) levels were significantly associated with both outcomes in crude, but not in adjusted models (Table 1). There were no such associations for IL-6, TNF, Monocyte Chemotactic Protein (MCP) 1 or IL-8 (Table 1). In the exploratory analyses, several proteins, in particular MCP-3, had significant associations with CVD and CAD events, but only in the unadjusted models (Table 2).
Conclusion: Plasma levels of IL-17A at RA diagnosis predicted future CVD events, although we cannot exclude that this finding is due to multiple testing. The observed association was independent of levels of systemic inflammation, measured by ESR, suggesting a particular role for IL-17A in mechanisms leading to CVD. OPG and MCP-3 may also be predictive of CVD in patients with RA and should be further investigated.
Disclosures: E. Rydell, None; L. Jacobsson, Novartis, Eli Lilly, Janssen; C. Turesson, AbbVie/Abbott, Nordic Drugs, Bristol-Myers Squibb(BMS).
.jpg)
.jpg)
