Objectives BUILDER-1 and BUILDER-2 aimed to assess the efficacy and safety of tocilizumab (TCZ) in patients with ankylosing spondylitis (AS).

Methods BUILDER-1 was a two part, phase II–III parallel-group trial in patients with AS naive to antitumour necrosis factor (aTNF) treatment. Patients in part 1 received TCZ 8 mg/kg or placebo for 12 weeks. In part 2 (beginning after part 1 enrolment ended), newly enrolled patients received TCZ 4 or 8 mg/kg or placebo for 24 weeks. The same treatment arms were used in BUILDER-2, a phase III study in aTNF-inadequate responders. The primary endpoint for both studies was the proportion of patients achieving 20% improvement in the Assessments in Axial SpondyloArthritis international Society (ASAS). Secondary and exploratory endpoints included ASAS40 response rates, Bath Ankylosing Spondylitis Disease Activity Index improvement, changes in joint counts, enthesitis score and C reactive protein (CRP).

Results 102 patients were randomised in BUILDER-1 part 1; 99 (48 TCZ, 51 placebo) completed 12 weeks. Week 12 ASAS20 response rates were 37.3% and 27.5% in the TCZ and placebo arms, respectively (p=0.2823). Secondary and exploratory endpoints did not differ between treatment arms. CRP levels declined with TCZ treatment, suggesting adequate IL-6 receptor blockade. As a result, BUILDER-1 part 2 and BUILDER-2 were terminated. TCZ safety results were consistent with previous observations in rheumatoid arthritis, except for a cluster of anaphylactic and hypersensitivity events at Bulgarian study sites. No apparent explanation for this clustering could be found.

Conclusions BUILDER-1 failed to demonstrate TCZ efficacy in treating aTNF-naive patients with AS.

Clinical trial registration numbers: NCT01209702 and NCT01209689 (www.clinicaltrials.gov).