Background/Purpose: Tapering of tumor necrosis factor inhibitor (TNFi) treatment in patients who have reached sustained remission is debated in current guidelines, and further data are needed regarding the long-term consequences of such strategies. Our aim was to assess the 3-year effect of tapering to withdrawal of TNFi versus continued stable TNFi on disease activity flare and remission status.

Methods: ARCTIC REWIND was a randomized, multicenter, open-label, non-inferiority trial including RA patients in sustained remission for ³12 months on stable TNFi therapy, with no swollen joints at inclusion. Patients were randomized 1:1 to taper to withdrawal of TNFi (four months half-dose, thereafter withdrawal) or continue stable TNFi therapy, with scheduled visits every four months for 3 years. Full-dose TNFi therapy was reinstated if a flare occurred. The primary endpoint of the current study was disease activity flare over 3 years. A flare was defined as a combination of DAS >1.6 (loss of remission status), an increase in DAS ≥0.6 units (change above minimal detectable change) and ≥2 swollen joints, or if the physician and patient agreed that a clinically significant flare had occurred. Secondary endpoints included remission status (ACR/EULAR Boolean 2.0 and DAS), medication use and adverse events (AE). Flare-free survival was analyzed by Kaplan-Meier, flare by center adjusted Cox regression, and remission status by logistic regression adjusted for baseline status, analyzed in a per protocol population.

Results: Of 99 randomized patients, 92 received the allocated therapy, and 80 (87%) completed 3-year follow-up. Mean baseline DAS was 0.8 in the tapering TNFi group, and 0.9 in the stable TNFi group. csDMARD co-medication was used by 89% in the tapering group and 91% in the stable group. After 3 years, 25% (95% CI: 13 to 38) remained flare-free in the tapering TNFi group compared to 85% (70 to 93) in the stable TNFi group (Fig 1), with corresponding hazard ratio for flare of 9.4 (3.9 to 22.8). Most patients regained DAS remission within the next visit after a flare (81% in tapering group, 67% in stable group). We observed significantly lower Boolean 2.0 remission rates in the tapering TNFi group than the stable TNFi group throughout the study period (Fig 2), adjusted risk difference 0-36 months -25% (-33 to -16), p< 0.001 (Boolean 1.0 revealed similar results). Systemic glucocorticoids (≥1 treatment period during the study) were used by 23% in the tapering TNFi group and 13% in the stable TNFi group during the study, and 10% switched to other types of TNFi or JAK inhibitor treatment in the tapering group, and 11% in the stable group (Table). AE/serious AE occurred in 81%/21% of the patients in the tapering group, and 89%/11% of the patients in the stable group.

Conclusion: A large majority of RA patients in remission tapering TNFi to withdrawal experienced a flare within 3 years, while a minority of patients receiving stable TNFi treatment flared over the same time period. Even though most patients regained remission within the next visit after a flare, TNFi tapering was associated with significantly lower Boolean 2.0 remission rates throughout the study. These findings do not support tapering of TNFi treatment among RA patients in sustained remission.

« Back to ACR Convergence 2023

ACR Meeting Abstracts - https://acrabstracts.org/abstract/3-year-results-of-tapering-tnfi-to-withdrawal-compared-to-stable-tnfi-among-rheumatoid-arthritis-patients-in-sustained-remission-a-multicenter-randomized-trial/