Background/Purpose: Cardiovascular disease (CVD) is the leading cause of death in rheumatoid arthritis (RA). While chronic inflammation contributes to CVD pathogenesis, the role of specific circulating inflammatory mediators with CVD risk in RA is unclear. Because existing risk calculators underestimate CVD risk in RA patients, additional biomarkers to prognosticate CVD risk are needed. We evaluated the associations of serum cytokines and chemokines with incident major adverse cardiovascular events (MACE) and whether subclinical inflammation as assessed by these mediators predicts future CVD.

Methods: We included patients from a prospective, multicenter cohort of US Veterans with RA between 2003-2020 and followed them from enrollment to incident MACE, death, or end of follow-up. MACE was defined as a composite of myocardial infarction, coronary revascularization, stroke, or CVD-related death and identified with previously validated algorithms using ICD codes and the National Death Index. A multiplex assay was used to measure cytokines and chemokines (n=33) using serum banked at study enrollment. Values were log-transformed, standardized, and considered elevated if >2 SD above the mean. Covariates included demographics, smoking status, BMI, comorbidities (hypertension, diabetes, lung disease, prior MACE), LDL cholesterol, DAS28, and medications (methotrexate, TNFi, prednisone) at enrollment. Associations between individual analytes with incident MACE were estimated using multivariable Cox regression. In secondary analyses, we additionally adjusted for DAS28 and restricted the cohort to patients in remission or low disease activity (LDA).

Results: We studied 2,712 RA patients (mean age 72 years, 90% male, mean DAS28 of 3.7). We identified 406 MACE outcomes over 22,216 person-years of follow-up (19.3 events per 1,000 PY, mean time to event 4.9 years). After multivariable adjustment including traditional CVD risk factors, 12 of 33 analytes were significantly associated with an increased risk of MACE (HR per 1 SD range 1.11-1.22). Associations between cytokines / chemokines and incident MACE persisted after further adjusting for baseline DAS28 (Figure 1). Among 683 patients in remission/LDA at enrollment, 97 MACE outcomes occurred in follow-up. Five analytes (IL-6, IL-15, IL-17A, IFN-g, and macrophage inflammatory protein-3-ɑ) remained associated with incident MACE (HR range 1.19-1.50) in these patients. The number of elevated cytokines was associated with an increased risk of MACE in the overall (HR 1.06 [95% CI 1.01-1.10]) and LDA / remission cohort (HR 1.11 [1.06-1.11]).

Conclusion: In a prospective, multicenter RA cohort, several circulating cytokines and chemokines were associated with a heighted risk of incident MACE, independent of traditional CVD risk factors and clinical RA disease activity. Elevated concentrations of these pro-inflammatory mediators similarly predicted MACE even in those in LDA or remission. Together, these findings suggest that measuring pro-inflammatory mediators in RA patients may aid in CVD risk stratification and prevention beyond existing traditional and RA-related CVD risk factors.

ACR Meeting Abstracts - https://acrabstracts.org/abstract/circulating-cytokines-and-chemokines-are-associated-with-the-risk-of-incident-cardiovascular-disease-in-rheumatoid-arthritis-independent-of-conventional-disease-activity-measures/