Prospective cohort study of effects of infliximab on rheumatoid factor, anti-cyclic citrullinated peptide antibodies and antinuclear antibodies in patients with long-standing rheumatoid arthritis

Abstract

Background

Antibodies to cyclic citrullinated peptide (anti-CCP) and IgM rheumatoid factor (IgM-RF) are well-established serological markers for rheumatoid arthritis (RA). Lupus-like disease with antinuclear antibodies (ANA) has been reported during TNFα antagonist therapy. Our objectives were to investigate the effect of infliximab therapy on these three autoantibodies in patients with established RA and to look for correlations linking IgM-RF and anti-CCP titres to a treatment response (defined as a good or moderate EULAR response) after 48 weeks of infliximab therapy.

Methods

Thirty-six patients with long-standing RA not responding to disease-modifying anti-rheumatic drugs (DMARDs) received intravenous infliximab (starting dose: 3 mg/kg) at 0, 2, and 6 weeks then at 8-week intervals, in combination with a DMARD. At baseline, week 24, and week 48, C-reactive protein (CRP) and the erythrocyte sedimentation rate (ESR) were determined and the disease activity score (DAS28) was calculated. Serum samples collected at the same time points were used to measure anti-CCP (commercial second-generation ELISA), IgM-RF (quantitative nephelometric assay), and ANA (indirect immunofluorescence in HEp2 cells). Correlations linking baseline autoantibody titres to changes in autoantibody levels were examined.

Results

At baseline, tests were positive for anti-CCP in 31/36 (94.6%) patients, IgM-RF in 29/36 (80.5%) patients, and ANA in 16/36 (44%) patients. IgM-RF titres decreased significantly (p < 0.001), whereas anti-CCP showed little change (p = 0.053). ANA titres increased significantly (p < 0.001). The treatment response was not associated with changes in anti-CCP or IgM-RF titres during infliximab therapy (OR for a response in patients with a 50% anti-CCP decrease, 0.77 [95%CI, 0.16–3.58]; OR for a response in patients with a 50% IgM-RF decrease, 0.82 [95%CI, 0.16–4.13]).

Conclusions

During infliximab therapy used to treat established RA, IgM-RF titres showed larger decreases than anti-CCP titres. Changes in IgM-RF and anti-CCP failed to correlate with the 48-week treatment response.

Introduction

Rheumatoid arthritis (RA) is a systemic autoimmune disease characterised by synovial pannus formation leading to cartilage destruction, bone erosion, and disability. The disease course is highly variable, unpredictable, and progressive. Disease-modifying anti-rheumatic drugs (DMARDs) may slow the progression of the disease. However, not all patients with RA can tolerate or respond to conventional DMARDs [1]. In recent years, research into the pathogenic mechanisms driving synovial inflammation and tissue damage in RA established a key role for pro-inflammatory cytokines such as tumour necrosis factor α (TNFα) and interleukin 1 (IL-1), leading to the development and clinical use of biological agents that bind to and inactivate these cytokines. Among these agents, the anti-TNFα monoclonal antibody infliximab is effective in relieving the signs and symptoms of RA, slowing the progression of radiological joint damage, decreasing serum C-reactive protein (CRP) levels, and downregulating inflammatory cytokines stimulated by TNFα [2], [3], [4], [5].

However, despite the remarkable overall clinical effectiveness of TNFα antagonists, more than one-quarter of patients have a poor clinical and radiological response [6]. Until now, no reliable predictors of the treatment response have been identified, although recent reports suggest that decreases in rheumatoid factor (RF) and antibodies to cyclic citrullinated peptides (anti-CCP) may help to assess treatment efficacy [7], [8], [9], [10]. In addition, up to 40% of patients with RA experience early progression to erosive disease [11], and early identification of these patients is crucial to enable aggressive treatment aimed at slowing joint damage and preserving joint function. Together with the classical clinical features of RA, the presence of RF and anti-CCP help to establish the diagnosis, most notably in the early stages of the disease. Both markers also assist in predicting disease severity and radiographic joint damage [12], [13], [14], [15]. Anti-CCP correlates with RF but is better than RF for predicting an aggressive disease course [14]. With the currently available second-generation anti-CCP test (anti-CCP2), the presence of anti-CCP is highly specific (95%; 95%CI, 94%–97%) and reasonably sensitive (67%; 95%CI, 62%–72%) for diagnosing RA [16], [17], suggesting a key role for this antibody in the pathogenesis of RA. Side effects of TNFα antagonist therapy include an increased risk for infection and induction of autoantibodies such as antinuclear antibodies (ANA), anti-double-stranded DNA antibodies (anti-dsDNA) and antiphospholipid antibodies [18], [19]. Anti-dsDNA antibodies were found in 5%–20% of RA patients treated with infliximab or etanercept (human soluble TNFα receptor p75 fusion protein), although the development of lupus-like illness was uncommon [20], [21]. The mechanism responsible for autoantibody production during TNFα antagonist therapy remains unknown. Nevertheless, inhibition of TNFα, a pivotal T-helper-1 cytokine, may induce a shift toward a predominantly T-helper-2 response, leading to increased (auto)antibody production.

In the present study, we evaluated IgM-RF, anti-CCP, and ANA in a cohort of RA patients followed prospectively for 48 weeks, during which they received infliximab therapy.

Our primary objectives were to determine whether baseline IgM-RF and anti-CCP titres predicted a 48-week treatment response and whether decreases in these autoantibodies during treatment were associated with a 48-week treatment response. Our secondary objective was to monitor ANA levels, given the reports of infliximab-induced lupus.