Background/Purpose: Biologic medications have improved outcomes for patients with rheumatoid arthritis (RA), but little is known about factors that predict individual response to therapy. One potential factor is the presence or absence of rheumatoid factor (RF) or anti-cyclic citrullinated peptide antibodies (CCP). Rituximab and some csDMARDs show greater efficacy in patients with RF or CCP than in seronegative patients. Recent studies have shown increased expression of genes and activation of IL-6/STAT3 pathways and transcription factors in seronegative RA patients compared with seropositive patients. Tocilizumab (TCZ), a monoclonal antibody to the IL-6 receptor, targets this pathway. We evaluated whether seronegative RA patients differed in their response to TCZ as compared to seropositive patients.

Methods: Data from the US Corrona RA registry were analyzed. Patients with RA starting on TCZ with a follow up visit 4-8 months following initiation of therapy were included. Serological status was analyzed by three methods: seropositivity was defined as RF positive, CCP positive, and either RF or CCP positive. Univariate tests of change scores for measures of disease activity between baseline and follow-up were made for each definition of seropositivity. General linear models were used to adjust for disease duration and baseline disease activity level.

Results: 482 patients initiating TCZ with information on CCP and/or RF status were included. CCP or RF positive patients were less likely to be white and had longer disease duration than seronegative patients; other demographic and clinical features were similar. Univariate analysis showed significant decrease in most measures of clinical disease activity and ESR and CRP with TCZ treatment in both seropositive and seronegative groups (table 1), but no significant differences in responses between groups were observed. Adjusted analyses accounting for disease duration and baseline activity also showed no difference in improvement by serostatus.

Conclusion: In this cohort of patients with RA, the presence of RF or CCP did not influence initial response to TCZ for most measures of disease activity. Both groups had statistically significant improvements in clinical disease activity measures and inflammatory markers. Table 1: Differences in disease activity from time of Tocilizumab initiation to follow up, by serologic status at baseline, N=482

⁺  p values between baseline and follow up are calculated using sign tests (this test ignores ties when computing p values) * p values between groups use two-sided Wilcoxon two-sample tests and ranked difference scores from each group (this test uses the distribution of ranks when computing p values)