Discontinuation versus continuation of maintenance treatment with tumor necrosis factor inhibitors in patients with rheumatoid arthritis with low disease activity or remission: A randomized double-blind placebo-controlled trial

Objective

Patients with rheumatoid arthritis (RA) in sustained remission/low disease activity with tumor necrosis factor inhibitor (TNFi) treatment may be candidates for treatment de-escalation. We compared the frequency of relapses and flares of RA after TNFi discontinuation with TNFi continuation among such patients. We also investigated predictors of relapse.

Methods

This randomized, double-blind, placebo controlled noninferiority trial enrolled adults with RA who were treated with either adalimumab, etanercept, or infliximab, and who had a Disease Activity Score 28-C-reactive protein (DAS-CRP) < 2.6 for at least six months. Patients were randomly assigned 2:1 to either placebo injections/infusions (i.e. discontinuation) or active TNFi. Patients, investigators, and study staff were masked to treatment assignment. The primary endpoint was relapse over 48 weeks, defined as either DAS29-CRP ≥ 2.6 or treatment escalation. Secondary endpoints included flare, defined as an increase in DAS28-CRP ≥ 1.2 from baseline.

Results

The study was stopped early based on an interim analysis that rejected noninferiority. At study interruption, 102 patients (69 placebo; 33 active TNFi) were randomized. Relapses occurred in 59.4 % of the placebo group and 18.1 % of the active TNFi group (hazard ratio 4.88; 95 % confidence interval 2.05, 11.61). Flares occurred in 43.5 % and 15.1 % of these groups, respectively. Over 48 weeks, successful discontinuation was achieved in 55 % of patients who discontinued adalimumab and 26 % of patients who discontinued etanercept.

Conclusion

Discontinuation of maintenance TNFi resulted in higher rates of relapse of RA than continuation.

Trial Registration

Clinicaltrials.gov NCT01793519