Background/Purpose: 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 of risk factors and disease outcome.^1,2 However, it is
unclear whether these patients differ at baseline. Therefore, this study aimed
to investigate whether the number
of autoantibodies is associated with phenotypic characteristics at presentation
in RA patients.   

Methods: Two independent cohorts of early RA patients were analysed: the Leiden
Early Arthritis Cohort (EAC) (845 patients) and the
Swedish BARFOT (Better Anti‐rheumatic Farmaco-therapy) study (805 patients). All patients
fulfilled the 1987 ACR criteria for RA and had a symptom duration of less than
24 months. Autoantibody status was determined at baseline. RF
was measured with commercial enzyme linked immunosorbent assays (ELISAs) in the EAC and with commercial agglutination tests
in the BARFOT. Commercial ELISAs were also used to
determine anti-CCP2 status in both cohorts. Antibodies
against carbamylated fetal
calf serum (Anti-CarP) were measured for both cohorts
in Leiden using in-house ELISAs. Baseline characteristics between patients with
few (0 or 1) and many (2 or 3) autoantibodies were compared using logistic
regression. Variables with a univariate p <0.10
were included in a multivariate model. The multivariate analysis was repeated after
exclusion of highly correlated variables and variables with >15% missing
values.

Results: The
distribution of the autoantibodies in both cohorts was similar. In both
cohorts, patients with many RA-associated antibodies were younger, more often
smokers, had a longer symptom duration, more often an intermittent onset of
symptoms and a higher erythrocyte sedimentation rate (ESR)
compared to patients with few autoantibodies (Table 1). Furthermore, patients
having many autoantibodies differed with respect to distribution of affected
joints and swollen joint count in the EAC (Table 1). Other variables which were
significantly associated with the number of autoantibodies in univariate
analysis, were BMI, family history of RA and HAQ in
the EAC, and C-reactive protein (CRP), SHS and erosions at baseline in the BARFOT
study. When these variables were included in the multivariate model, only
erosions was significant: OR 2.62 (1.63-4.21).

Conclusion: Patients
carrying few and many autoantibodies differ with regard to initial clinical
presentation. Patients with many RA-associated autoantibodies are younger, more
often smokers, have a longer symptom duration, more often an intermittent onset
of symptoms and a higher ESR compared to patients
with few autoantibodies.

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

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

Table 1