OBJECTIVES: To review randomized controlled trials (RCTs) evaluating first-line infliximab treatment for rheumatoid arthritis (RA), with a focus on long-term outcomes.

METHODS: The study was performed in accordance with PRISMA guidelines. A systematic search of MEDLINE, CENTRAL, and EMBASE was conducted up to May 25, 2018 for infliximab RCTs in RA using the Cochrane Highly Sensitive search filter. Manuscript eligibility criteria were: (1) full-text manuscript of a RCT, including post-hoc analyses; (2) infliximab treatment arm; and (3) study population of disease-modifying anti-rheumatic drug (DMARD)-naïve RA patients. Efficacy (clinical, patient-reported outcomes, and radiological progression) and key safety endpoints were extracted.

RESULTS: The search result contained 2,679 citations; 1,830 after duplicates were removed. After excluding conference abstracts, reviews, pooled analyses, and non-relevant studies, 7 RCTs (35 publications) were identified: ASPIRE (7 publications), BeSt (16), Durez et al. 2007(1), IDEA (2), NEO-RACo (5), Quinn et al. 2005 (3), and Tam et al. 2012 (1). Sample sizes in the 7 RCTs ranged from 20 to 1,049 patients with follow-up durations between 3 months and 10 years. Results for 64 relevant endpoints at 19 timepoints were reported, resulting in 200 timepoint-outcome combinations. Out of these, 183 (92%) were reported from 1 of the studies, 17 from ≥2 studies, and 11 from ≥3 studies. Results for American College of Rheumatology response, Health Assessment Questionnaire, and 1-year remission rate were reported from 4, 5, and 4 studies, respectively. Beyond 1 year of follow-up, results for these 3 major efficacy endpoints were only reported from 3 studies.

CONCLUSIONS: Reports on first-line infliximab treatment for patients with DMARD-naïve RA are abundant. However, diversity of study design and the reporting of long-term outcomes hampers meta-analyses of the results. These findings support the need for developing guidelines for reporting from long-term studies, in line with cohorts and registries in RA.