Background Rheumatoid arthritis (RA) typically requires lifelong treatment. However, some RA patients achieve sustained disease-modifying antirheumatic drug (DMARD)-free remission, which is a proxy for cure of RA that has become increasingly achievable, as reported previously. DMARD-free sustained remission has been reported mostly in autoantibody-negative RA, yet the underlying mechanism is unknown. The multi-biomarker disease activity (MBDA) score combines 12 serum biomarkers and is developed to measure RA disease activity. We hypothesise that the subgroup of RA patients that is most likely to achieve DMARD-free sustained remission is identifiable at disease presentation by cytokines such as those combined in the MBDA score.

Objectives To evaluate whether the MBDA score or its component cytokines at the presentation of RA are associated with ability to later achieve DMARD-free sustained remission.

Methods 300 patients with RA (by the 1987 and/or 2010 criteria) who had been consecutively enrolled in the Leiden Early Arthritis Clinic between 2010 and 2015 were studied. At time of diagnosis, before DMARD treatment was started, the MBDA score, with a scale of 1–100, was determined from the serum concentrations of 12 biomarkers (VCAM-1, EGF, VEGF-A, IL-6, TNF-RI, MMP-1, MMP-3, YKL-40, Leptin, Resistin, CRP, SAA) with a pre-specified, validated algorithm. Patients were categorised as having a low (<30), moderate^30–44 or high (>44) MBDA score. DMARD-free sustained remission was defined as the absence of synovitis (by physical examination) that sustained after discontinuation of all DMARD therapy (including biologics and systemic and intra-articular corticosteroids) for the entire follow-up period, but had to extend to at least one year after DMARD withdrawal. Analyses were stratified for ACPA and restricted to 5 years follow-up as thereafter the number of patients became small. The median follow-up duration of all patients was 4.3 years.

Results A total of 54 RA patients (18%) had achieved DMARD-free sustained remission. For the total group of RA patients, baseline MBDA category (p=0.03) and ACPA-negativity (p<0.001) were associated with achieving DMARD-free sustained remission. For ACPA-positive RA patients, the MBDA category at baseline was not associated with achieving DMARD-free sustained remission (p=0.89, figure 1). By contrast, among ACPA-negative RA patients, none of those with low MBDA score achieved DMARD-free sustained remission during 5 years follow-up, whereas the estimated rate of remission was 50% for those with moderate or high MBDA scores (p=0.009, figure 1). Of the 12 biomarkers in the MBDA test, only SAA showed a significant difference between ACPA-negative patients with and without DMARD-free sustained remission (p=0.01).

Conclusions ACPA-negative RA patients who achieved DMARD-free sustained remission were characterised by moderate to high MBDA scores at disease presentation. This is the first evidence that a cytokine profile at disease onset can identify a subgroup of ACPA-negative RA patients with a high likelihood of maintaining clinical remission after treatment withdrawal.

Disclosure of Interest None declared