Article Text

SAT0109 The Number of Different Autoantibodies is Associated with Phenotypic Characteristics at Presentation in Patients with Rheumatoid Arthritis
  1. V. Derksen,
  2. A. van der Helm-van Mil,
  3. T. Huizinga,
  4. R. Toes,
  5. D. van der Woude
  1. Rheumatology, Leiden University Medical Center, Leiden, Netherlands


Background Patients with rheumatoid arthritis (RA) are frequently positive for one or several autoantibodies like rheumatoid factor (RF), anti-citrullinated protein antibodies (ACPA) and anti-carbamylated protein antibodies (Anti-CarP). Seropositive and seronegative patients differ in respect to underlying risk factors and disease outcome.1,2 However, conflicting data have been reported about whether seropositive and seronegative patients are also phenotypically different at presentation.3,4 Furthermore, it is unclear whether patients with different numbers of autoantibodies have different characteristics at the time of presentation.

Objectives To investigate whether the number of RA-associated autoantibodies is correlated with phenotypic characteristics at presentation in RA patients.

Methods 843 RA patients fulfilling the 1987 ACR criteria from the Leiden Early Arthritis Cohort (EAC) of whom RF, ACPA (positive for at least one of second, third generation anti-cyclic citrullinated peptide or anti-modified citrullinated vimentin antibodies) and anti-CarP (positive for Anti-CarP fetal calf serum or fibrinogen antibodies or both) status was known, were evaluated. Baseline characteristics were compared between groups of patients with 0, 1, 2 or 3 autoantibodies and p-values were calculated with either one-way ANOVA tests for normally distributed, or Kruskal-Wallis tests for not normally distributed numerical variables and chi-square tests for categorical data. To investigate if the observed differences are caused solely by presence of ACPA, the analysis was repeated after stratifying for ACPA. In case only two groups were compared, p-values were calculated either using unpaired t-tests for normally distributed, or Wilcoxon rank sum tests for not normally distributed numerical data and chi-square tests for categorical data.

Results The number of autoantibodies at baseline is significantly associated with differences in several baseline characteristics. Patients with several (2 or 3) autoantibodies were younger, had a lower BMI, more often a positive family history of RA, a longer symptom duration and a different distribution of affected joints compared to patients with 0 or 1 autoantibodies (see table). After stratification for ACPA, these differences were also seen between patients with ACPA only, both ACPA + 1 and ACPA + 2 other autoantibodies, but they were no longer statistically significant, except for symptom duration (ACPA only; median 3.00 months, IQR:(1.00-8.00), ACPA + 2; 4.00 (2.00-8.00), p=0.017).

Conclusions Seropositive and seronegative patients differ with regard to initial clinical presentation and an association exists between the number of autoantibodies and several baseline characteristics. Because different kinds of autoantibodies are often present together, it cannot be discerned whether the presence of one specific antibody or the summative effect of separate autoantibodies as a proxy of the underlying autoimmune response is at the base of these differences.


  1. Aletaha D. et al. 2010 Arthritis Rheum. 62(9): 2569-81.

  2. Shi J. et al. 2011 PNAS. 108(42): 17372-7.

  3. Van der Helm-van Mil A.H. et al. 2005 Arthritis Res Ther. 7(5): R949-58.

  4. Mouterde G. et al. 2014 J Rheumatol. 41(8): 1614-22.

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.