Anti–Citrullinated Protein Antibody Specificities, Rheumatoid Factor Isotypes, and Incident Cardiovascular Events in Patients With Rheumatoid Arthritis

Abstract

Objective To investigate the relationship between anti‐citrullinated protein antibodies (ACPA), specific sub‐specificities, rheumatoid factor (RF) isotypes, and incident cardiovascular (CV) events in patients with rheumatoid arthritis (RA). Methods We centrally typed baseline sera from Swedish new‐onset RA cases diagnosed 1996‐2009 for anti‐CCP2, 20 ACPA sub‐specificities, and RF isotypes. We followed patients longitudinally in nationwide registers for acute coronary syndrome (ACS), stroke, CV‐related death, and major adverse CV events (MACE). We assessed the association between each serological marker and CV outcome, and the impact of adjustment for baseline DAS28, smoking and income, using Cox proportional hazards model. We also explored associations with all‐cause mortality. Results We included 2814 patients. Median follow‐up was 13 years, during which 375 ACS, stroke, and CV deaths occurred. Occurrence and/or levels of anti‐CCP2 associated with ACS (hazard ratio (HR) 1.46 (95% confidence interval (CI) 1.03‐2.06), stroke (HR=1.47 (95%CI 1.03‐2.10), CV death (levels, p=0.024) and MACE (HR=1.34 (95%CI 1.06‐1.70)). We similarly observed associations with the number of ACPA sub‐specificities, which could not, however, be attributed to any individual or group of ACPA sub‐specificities. IgM RF associated with all CV endpoints except ACS, and IgA RF exclusively associated with CV death. Adjustment for smoking, income and DAS28 decreased most HRs, while IgA RF remained associated to CV death (HR=1.61 (95%1.05‐2.48)). All serological makers associated with all‐cause mortality. Conclusions RF isotypes and ACPA are associated with future CV events. ACPA levels and number of sub‐specificities seem more important than the occurrence of particular sub‐specificities. These associations were not explained by smoking.