Background Several clinical studies on tumor necrosis factor inhibitors in the treatment of rheumatoid arthritis (RA) have revealed that some RA patients could achieve long-term remission and subsequent biologic-free remission. However, a considerable number of RA patients experience flare after withdrawal, and progressive joint destruction is known to occur in these patients. Thus, it can be difficult to decide to withdraw a biologic agent in the clinical setting.

Objectives It is believed to be more difficult to maintain biologic-free remission in patients treated with the receptor construct etanercept (ETN), which was approved in 2005 as the second biologic agent in Japan, than in those treated with an antibody agent. Here, we investigated the possibility of biologic-free remission after ETN therapy by examining the characteristics of RA patients who had withdrawn from ETN therapy and successfully maintained biologic-free remission without flare or joint destruction.

Methods Of the 110 patients treated with ETN (25 mg or 50 mg weekly started before December 2011) for at least 1 year, steroids were tapered and ultimately discontinued, and then ETN dosage was decreased to 25 mg every 2 weeks. ETN was withdrawn in patients whose disease activity score 28-erythrocyte sedimentation rate (DAS28-ESR) was maintained below 2.6 for at least 6 months (including the period with reduced ETN dosage). At the end of the study period, which was at least 1 year after ETN withdrawal, patients who had withdrawn from ETN therapy were classified into those who maintained biologic-free remission (BFR group; absence of both flare (DAS28-ESR >3.2) and joint destruction) and those who did not (non-BFR group). The characteristics of each group were statistically compared using Mann-Whitney's U test.

Results Twenty-six patients (mean age, 54.8 years) withdrew from ETN therapy. Their mean disease duration was 7.8 years and mean DAS28-ESR was 1.78. Fourteen patients successfully maintained biologic-free remission for one year, and 5 developed flare and/or progression of joint destruction by the end of the study period. The 1-year and 2-year survival rates after ETN withdrawal, estimated by the Kaplan-Meier method, were 53.8% and 34.6%, respectively. At the time of ETN withdrawal, there were no significant differences between the BFR and non-BFR groups in terms of patient characteristics, methotrexate (MTX) dosage, disease duration etc. All patients in both groups received MTX (4-10 mg weekly) in addition to ETN. All 17 patients in the non-BFR group reached a state of low disease activity or remission after restarting ETN therapy. The mean DAS28-ESR determined by the last-observation-carried-forward method at the end of the study did not differ significantly between the BFR (2.02) and non-BFR (2.07) groups. However, DAS28-ESR was significantly higher at the end of the study than at the time of ETN withdrawal in the non-BFR group (P=0.026).

Conclusions No sufficient data are available to demonstrate subsequent biologic-free remission. Our findings suggest that some RA patients who had been on ETN-MTX combination therapy and maintained clinical remission for at least 6 months (including the period with ETN dosage reduced to 25 mg every 2 weeks) can successfully maintain clinical and structural remission for at least 1 year after ETN withdrawal, regardless of disease duration.

Disclosure of Interest None declared

DOI 10.1136/annrheumdis-2014-eular.1546