Background Remission is achievable in many patients receiving biologic therapies. Only limited information on predictors of discontinuation of biologic therapy in patients with rheumatoid arthritis (RA) is available.

Objectives To evaluate how often biologic DMARDs are discontinued due to remission and to identify predictors of discontinuation of biologic treatment in patients from a Spanish registry.

Methods Retrospective, observational cohort study based on data from a national registry. RA patients receiving a first biologic DMARD for ≥3 consecutive months between April 1998 and December 2013 were included. Patients receiving rituximab were excluded. The study endpoint was defined as discontinuation of biologic DMARD due to remission, defined by treating physician. Censoring occurred administratively (end of registry data), when patients ceased treatment due to other causes (side effect, lack of efficacy, pregnancy, etc.) or due to loss to follow up. Multivariate proportional sub-distribution hazards (SHR) models were used to evaluate associations between predictor factors of discontinuation due to remission, discontinuation due to lack of efficacy, side effects, loss to follow-up or other causes as competing events.

Results 3,516 patients were included, of which 3,161 patients had received ≥3 months of biologic DMARD: 753 patients discontinued treatment due to side effects, 867 patients due to lack of efficacy, 101 were lost to follow-up, 143 for other reasons, 48 due to pregnancy, and 15 patients due to unknown reasons. 1175 patients were receiving biological DMARD at the study conclusion. Only 59 (1.8%) patients discontinued biologic therapy due to remission. The Table shows baseline characteristics of patients at initiation of biologic DMARDs by discontinuation for remission.

After multivariate SHR analysis, sex (female) (SHR 2.81, 95% CI 1.01-7.83), age at onset (SHR 1.04, 95% CI 1.01-1.07) and disease duration (HR 0.94, 95% CI 0.90-0.98) were significant predictors of discontinuation due to remission adjusting by methotrexate and steroids use.

Conclusions Less than 2% of RA patients discontinued biologic DMARD therapy due to disease remission. Sex, age at onset and disease duration were predictors of this discontinuation. The prognosis of biologic-free patients after remission is still unknown and further studies are required to determine their clinical outcomes.

Disclosure of Interest None declared