Background M-ficolin is a soluble pattern recognition molecule that activates the complement system. We recently reported a factor 300 difference in the synovial fluid M-ficolin concentration between rheumatoid arthritis (RA) and osteoarthritis, suggesting a pathogenic role of M-ficolin¹.

Objectives To assess M-ficolin in DMARD naïve early RA patients, investigate correlations between M-ficolin and disease activity markers, and analyze the predictive value of M-ficolin.

Methods 180 DMARD naïve RA patients with disease duration <6 months were included in a randomized double blind placebo-controlled trial (OPERA) of methotrexate, intraarticular glucucorticoids + either adalimumab (ADA) or placebo (PLA). A sandwich-type time-resolved immunofluorometric assay using monoclonal antibodies was used for quantification of M-ficolin in the OPERA cohort, 101 healthy adults and 51 chronic RA patients in remission. Concentrations are reported as medians. Spearman test, Students T-test and one-way analysis of variance were used. Logistic regression analyses were performed and adjusted for CRP, treatment group (ADA/PLA), x-ray erosions, anti-CCP, IgM-RF, neutrophils and monocytes.

Results The highest M-ficolin levels were measured in the OPERA cohort at baseline (2.84μg/l, CI 2.63-3.09), and during treatment the level decreased 24% (CI 18-31%, p<0.001) at year one (2.31μg/l, CI 2.14-2.50). The healthy adults had significantly lower concentrations (1.88μg/l, CI 1.72-2.06) than the OPERA cohort at baseline (p<0.001) and year one (p<0.001), and the chronic RA patients (2.17μg/l, CI 1.94-2.42) (p=0.03).

At baseline M-ficolin correlated to DAS28 (rho=0.29, p<0.001) and the four variables constituting DAS28 (0.001< p≤0.04). At year one M-ficolin correlated to DAS28 (rho=0.36, p<0.001) and 3 of the 4 variables constituting DAS28 (0.001< p≤0.05), except tender joint count 28 (p=0.23). Furthermore M-ficolin correlated to HAQ at debut (rho=0.25, p=0.003) and year one (rho=0.23, p<0.001).

Logistic regression analysis determined M-ficolin as the strongest predictor of DAS28<2.6 at year one (coefficient=1.42, p=0.0001) followed by treatment group (PLA/ADA) (coefficient=1.23, p=0.001) and erosions on baseline x-ray (coefficient=-1.19, p=0.02), but M-ficolin was the only variable that predicted DAS28<3.2 at year one (coefficient=0.89, p=0.02).

Low baseline M-ficolin level was the only variable associated with low disease activity at year one, and this was further analyzed by comparing the group with the 25% lowest baseline M-ficolin levels with the remaining 75% of patients (cutoff 2.00 μg/l). This resulted in a sensitivity of 29%, a specificity of 96%, and a positive predictive value of 98% in determining a DAS28 score <3.2 at year one.

Conclusions The elevated baseline M-ficolin levels in early RA correlated consistently to disease activity markers, most notably DAS28 and HAQ, thus reflecting essential parts of the disease activity. Low M-ficolin levels at baseline predicts a positive predictive value of 98% of a DAS28<3.2 after one year irrespective of treatment and well-known prognostic factors.

1. Ammitzbøll CG, et al. Rheumatol Int 2011.

Disclosure of Interest None Declared