Article Text

Download PDFPDF

SAT0060 Response to conventional synthetic dmards differs depending on rheumatoid factor levels in anti-citrullinated positive patients with early rheumatoid arthritis
  1. S Bugatti,
  2. A Manzo,
  3. G Zanframundo,
  4. F Benaglio,
  5. G Sakellariou,
  6. C Montecucco,
  7. R Caporali
  1. Rheumatology and Translational Immunology Research Laboratories (LaRIT) and Early Arthritis Clinic, Division of Rheumatology, University of Pavia, IRCCS Policlinico San Matteo Foundation, Pavia, Italy


Background In rheumatoid arthritis (RA), several autoantibody characteristics, including specifities, levels and isotypes, may influence their pathogenic properties and impact on clinical presentation and outcomes of the disease (1, 2). In particular, classical IgM rheumatoid factor (RF) may boost inflammation triggered by anti-citrullinated protein antibodies (ACPA) (3).

Objectives To investigate whether RF impacts on disease characteristics and response to therapy in ACPA-positive early RA patients treated with conventional synthetic disease modifying drugs (csDMARDs).

Methods 574 early RA patients consecutively enrolled in our Early Arthritis Clinic between 2005 and 2014 were included. Patients had symptoms' duration <12 months, were glucocorticoid- and DMARD-naïve, and fulfilled RA criteria at inclusion. IgM RF and IgG ACPA were determined in baseline sera by immunonephelometry and a second-generation EliA CCP assay respectively. Autoantibody levels were considered high when >3 times the upper limit of normal (ULN). Patients were treated with incremental doses of methotrexate according to a treat-to-target strategy aiming at low disease activity (LDA, DAS28 ≤3.2). The associations between autoantibody specifities and levels and the achievement of LDA and disease remission (DAS28 <2.6) over 6 months were investigated by Cox regression.

Results 360 patients tested RF and/or ACPA positive (31.9% single RF-positive, 10.3% single ACPA-positive, 57.8% ACPA and RF double-positive) and were selected for outcome analyses. Among seropositive patients, both LDA and remission were less frequently achieved in case of RF-positivity. Relevantly, in ACPA-positive patients (n=273), the co-occurrence of RF dose-dependently influenced clinical outcomes. After 6 months of treatment, LDA and disease remission were achieved respectively by 69.6% and 47.8% of single ACPA-positive, 68.8% and 37.5% of ACPA-positive RF-low, and 57% and 27.6% of ACPA-positive RF-high patients (chi-square for trend p=0.18 for LDA, p=0.05 for remission). After adjusting for confounders (age, gender, symptoms' duration, baseline disease activity, use of prednisone, recruitment period), high levels of RF independently predicted failure to achieve LDA with an HR (95% CI) of 0.61 (0.39 to 0.95) and failure to achieve remission with an HR (95% CI) of 0.63 (0.35 to 0.99) (Figure 1). In contrast, ACPA levels did not show any significant predictive value, neither for thresholds of >3 ULN nor of >100 U.

Conclusions Among ACPA-positive RA patients, disease characteristics may vary in association with the extent of overall humoral autoimmunity. In particular, the concomitant presence of high levels of IgM RF seems associated with lower response rates to csDMARDs. Collectively, these findings highlight the importance of further subclassifying patients with autoantibody-positive RA in order get deeper insights into disease mechanisms and clinical outcomes.


  1. Aletaha D et al. Arthritis Res Ther 2015;17:229.

  2. Derksen VFAM et al. Ann Rheum Dis 2017;doi:10.1136/annrheumdis-2016–209794.

  3. Sokolove J et al. Arthritis Rheumatol 2014;66:813–21.


Disclosure of Interest None declared

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.