OP0101 RHEUMATOID ARTHRITIS DISEASE ACTIVITY OVER TIME AND SUBSEQUENT CARDIOVASCULAR RISKS.

Abstract

Background:Rheumatoid arthritis (RA) patients have an increased risk of cardiovascular (CV) events not fully explained by traditional CV risk factors. The relationship between fluctuating inflammation due to RA disease activity and CV events is of interest.Objectives:To examine the influence of time-varying disease activity on the subsequent risks of CV disease.Methods:We followed patients from a large US registry of clinically diagnosed RA patients, starting at their first visit with a Clinical Disease Activity Index (CDAI) through the end of follow-up or first CV event. Exposure of interest was disease activity measured by categorical CDAI (high, moderate, low, and remission) averaged within each 6-month window. The outcome of interest was major adverse CV events (MACE) defined as non-fatal myocardial infarction, non-fatal stroke (excluding transient ischemic attacks), and CV death. For baseline confounders we considered age, gender, race, disease duration, Health Assessment Questionnaire, hypertension, diabetes, hyperlipidemia, family history of premature (age<50) CV events, and RF/ACPA seropositivity. For time-varying variables we considered tumor necrosis factor inhibitor (TNFi), non-TNFi biologic, methotrexate, oral glucocorticoid, non-steroidal anti-inflammatory drugs, statin, and aspirin use. We used the marginal structural model (MSM) framework to examine the impact of CDAI at each 6-month interval on MACE. We estimated time-varying hazard ratios (HRs) comparing high CDAI during follow-up to CDAI remission. Several predicted survival curves were constructed under different hypothetical CDAI scenarios, such as early and late transition to CDAI remission.Results:40,721 patients were eligible for our analyses. 77% were female and 84% were Caucasian. The mean age was 58 (SD 13) years with mean disease duration of 8.8 (median 5) years. Mean CDAI at their first registry visit was 14 (SD 13; remission 19%, low 31%, moderate 28%, and high 22%). Other baseline characteristics include: 41% current/former smokers, 31.5% with hypertension, 8.6% with diabetes, 18% with hyperlipidemia, and 52% seropositive. The average follow-up duration after baseline was 4.4 (median 3.3; max. 17.6) years. The crude event count of MACE was 1,050 events / 180,402 person-years.In the MSM analysis, the average HRs, assuming a constant HR, were 1.31 [0.90, 1.90] for low, 1.46 [1.01, 2.10] for moderate, and 1.43 [0.89, 2.31] for high CDAI disease activity categories during each 6-month interval. When approximating time-varying HR with linear trends, the highest estimates during the first 6 months of follow up were 1.61 [0.93, 2.77] for low CDAI, 1.97 [1.13, 3.43] for moderate CDAI, and 2.11 [1.13, 3.96] for high CDAI. These HRs gradually diminished during the follow up (Table). When we constructed hypothetical survival curves with transition to CDAI remission at different time points, earlier transition to CDAI remission was related to better event-free survival (Figure).Table 1.Time-varying hazard ratio estimates [95% confidence intervals] by duration in studyDisease activity measured by CDAIYearRemissionLowModerateHigh0.51.00 [ref]1.61 [0.93, 2.77]1.97 [1.13, 3.43]2.11 [1.13, 3.96]11.00 [ref]1.54 [0.97, 2.44]1.85 [1.17, 2.93]1.94 [1.13, 3.31]21.00 [ref]1.42 [1.00, 2.01]1.63 [1.16, 2.29]1.62 [1.04, 2.54]31.00 [ref]1.31 [0.90, 1.90]1.43 [0.97, 2.11]1.36 [0.80, 2.31]41.00 [ref]1.20 [0.72, 2.02]1.26 [0.72, 2.21]1.14 [0.55, 2.36]51.00 [ref]1.11 [0.54, 2.26]1.11 [0.51, 2.42]0.96 [0.36, 2.53]Figure 1.MACE-free survival curves under hypothetical CDAI scenariosConclusion:High and moderate CDAI were associated with higher hazard of MACE during the earlier period of follow-up, but the increased hazard diminished over time. In hypothetical senarios, earlier transition to CDAI remission would improve MACE free-survival.Acknowledgements:This study was sponsored by Corrona, LLC. Corrona is supported through contracted subscriptions with multiple pharmaceutical companies. The analysis was financially supported by Amgen Inc.Disclosure of Interests:Kazuki Yoshida Consultant of: OM1, Inc., Grant/research support from: Corrona, LLC., Hongshu Guan: None declared, Scott Stryker Shareholder of: Amgen, Inc., Employee of: Amgen, Inc., Elaine Karis Shareholder of: Amgen, Inc., Employee of: Amgen, Inc., Leslie Harrold Consultant of: AbbVie, Bristol-Myers Squibb, Genentech/Roche, Grant/research support from: Pfizer, Daniel Solomon Grant/research support from: DHS receives salary support from research contracts through Brigham and Women’s Hospital with Abbvie, Amgen, Corrona, Genentech and Janssen.