As rheumatoid arthritis research continues, biologics for different targets are being used in the clinic. a multicenter study published in the BMJ in March 2015 showed that combination intensive synthetic disease-modifying drug replacement strategies are non-inferior to tumor necrosis factor inhibitors for treatment of patients with active rheumatoid arthritis, but at substantially lower cost. OBJECTIVE: To determine whether similar clinical outcomes can be achieved with low-cost intensive combination therapy with synthetic disease-modifying agents versus high-cost biologics, such as tumor necrosis factor inhibitors, in patients with active rheumatoid arthritis who are refractory to initial methotrexate and other synthetic disease-modifying agents. DESIGN: Open pragmatic randomised, multicentre, two-arm non-inferiority trial over 12 months. PLACE: 24 rheumatology clinics in the UK. Subjects: Patients with rheumatoid arthritis suitable for tumour necrosis factor inhibitors according to current UK guidelines were randomly assigned to either tumour necrosis factor inhibitor strategy or a disease-modifying drug combination treatment strategy. Intervention: Biologic strategy: start tumor necrosis factor; non-responders are reintroduced to biologics within 6 months. Alternative strategy: start with disease-modifying drug combination therapy; non-responders use tumor necrosis factor inhibitors after 6 months. MAIN OUTCOME MEASURES: Primary outcome: reduction in disability at month 12 as measured by the Patient Recorded Health Assessment Questionnaire with a non-inferiority margin of 0.22 (range 0.00 to 3.00) for combination therapy compared to the biological strategy. Secondary outcomes: quality of life, joint damage, disease activity, adverse events, and costs. Missing data from the intention-to-treat analysis were analyzed using multiple interpolation. RESULTS: 432 patients were screened: 107 patients were randomly assigned to the tumor necrosis factor inhibitor group and 101 started the drug; 107 were randomized to the combination drug strategy group and 104 started the drug. Initial assessments were similar; 16 patients were lost to follow-up (7 were patients in the tumor necrosis factor strategy group and 9 in the combination drug strategy); 42 discontinued the intervention but remained on follow-up (19 and 23, respectively). The primary outcome showed a mean decrease in health assessment questionnaire scores of -0.30 in the tumor necrosis factor strategy group and -0.45 in the alternative drug combination strategy group. differences between groups in unadjusted linear regression analysis favored the alternative drug combination strategy. The mean difference was -0.14, with a 95% confidence interval (-0.29 to 0.01) below a prespecified non-inferiority boundary of 0.22. Secondary endpoints at 12 months of improvement, including quality of life and disease course, were similar for both strategies. The initial reduction in disease activity was greater in the biological strategy group, but these differences did not persist beyond 6 months. 72 remissions (44 biological strategy; 36 alternative strategy); 28 patients experienced serious adverse events (18 and 10, respectively); and 6 and 10 of the two groups discontinued treatment due to toxicity, respectively.