POS0515 THE ASSOCIATION BETWEEN AUTOANTIBODIES AND RISK FOR VENOUS THROMBOEMBOLIC EVENTS AMONG PATIENTS WITH RHEUMATOID ARTHRITIS

Abstract

BackgroundPatients with rheumatoid arthritis (RA) have an increased risk for cardiovascular disease, including venous thromboembolic events (VTE)1. The reason behind the increased VTE risk is incompletely understood, but inherent features of RA, such as RA specific autoantibodies, could potentially play a role. For example, studies have linked occurrence and levels of rheumatoid factor (RF) in the general population to increased VTE risk2. We and others have demonstrated an association between ACPA and risk of later ischemic cardiovascular events3. There are also potential mechanistic links; citrullinated fibrinogen (cFib) has been associated to clot stability4.ObjectivesWe aimed to examine the association between anti-modified protein antibodies (AMPAs) and risk of VTE in RA.MethodsWe included 2809 individuals newly diagnosed with RA and included in the Swedish EIRA study 1996-2009. Through linkage to nationwide health care registers we identified past and incident events of VTE based on validated ICD code algorithms. We centrally typed baseline sera for anti-CCP2, 20 different ACPA sub-specificities, RF isotypes, carbamylated antibodies and 10 additional post-translational modifications. We followed all individuals from RA diagnosis up until their first ever VTE event, migration, death or end of study (2020-12-31) whichever occurred first. We used a Cox regression to estimate hazard ratios (HR) with 95% confidence intervals (CI). Individuals with a history of a VTE event (n=27) at RA diagnosis were excluded.ResultsWe included 2782 individuals; 72% were women, median age at RA diagnosis was 54 years (inter quartile range (IQR) 18 years) and median follow-up time was 15.5 (IQR 6.8) years. During follow-up 177 incident VTE events were observed corresponding to an incidence of 5.0 per 1,000 person years.1797 (64.6%) patients were positive for IgG anti-CCP2 and the HR for VTE (vs. being negative for anti-CCP2) was 1.33 (95%CI 1.00-1.78). The risk of VTE increased with the level of anti-CCP2, with an HR of 1.49 (95%CI 0.99-2.22) for the group with extreme levels compared to those negative for anti-CCP2 (p-value for trend 0.048). For IgA anti-CCP2 the HR was 1.35 (95% CI 0.99-1.84) when comparing those expressing IgA anti-CCP2 against those who did not.Of 20 ACPA fine-specificities studied, 18 occurred with a frequency > 10% in our sample. The median number of fine-specificities expressed was 6 (IQR 11). The risk of VTE increased with the number of ACPA fine-specificities expressed (p-value for trend 0.033). At the 0.05 significance level, two fine-specificities were each associated with VTE; cPept Z1 [HR=1.40 (95%CI 1.06-84)] and cPept-1 [HR=1.47 (95%CI 1.12-1.93)]. None of the six antibodies against cFib assessed were statistically significantly associated with VTE risk. No associations were observed for other AMPAs. Among the three RF isotypes, only IgM RF was statistically associated with VTE [HR=1.38 (95%CI 1.04-1.83)].ConclusionRA-related antibodies analysed in clinical practice (anti-CCP2 IgG, RF) are associated not only with risk of myocardial infarction, stroke and cardiovascular death as previously demonstrated but also with VTE. There were no clear specific signals with ACPA fine-specificities, other AMPAs, or IgA RA autoantibodies.