Anti-granulocyte-macrophage colony-stimulating factor antibody otilimab in patients with hand osteoarthritis: a phase 2a randomised trial.

Background: Granulocyte-macrophage colony-stimulating factor (GM-CSF) is a key mediator of signs and symptoms in preclinical models of osteoarthritis. We explored the efficacy, safety, and pharmacokinetics of an anti-GM-CSF antibody, otilimab, in patients with hand osteoarthritis.

Methods: This double-blind, randomised, placebo-controlled phase 2a study was done in 16 centres in the Netherlands, Germany, Poland, the UK, and the USA. Patients aged 18 years or older with inflammatory hand osteoarthritis, who had received at least one course of unsuccessful non-steroidal anti-inflammatory drugs, with two or more swollen and tender interphalangeal joints (on the same hand), signs of inflammation or synovitis identified with MRI in the affected hand, and a self-reported 24 h average hand pain intensity over the past 7 days of 5 or more on a 0-10 numerical rating scale were eligible for inclusion. Patients were randomly assigned (1:1) via interactive response technology (blocked randomisation; block size of four) to receive either subcutaneous otilimab 180 mg or placebo, administered weekly from week 0 to week 4, then every other week until week 10. Patients, investigators, and trial staff were masked to treatment; at least one administrator at each site was unmasked to prepare and administer treatment. The primary endpoint was change from baseline in 24 h average hand pain numeric rating scale averaged over 7 days before the visit at week 6. Secondary endpoints were: change from baseline in 24 h average and worst hand pain intensity at each visit; proportion of patients showing 30% and 50% reductions in 24 h average and worst hand pain intensity at each visit; change from baseline in Australian and Canadian Hand Osteoarthritis Index (AUSCAN) 3·1 numeric rating scale questionnaire components at each visit; change in number of swollen and tender hand joints at each visit; change from baseline in Patient and Physician Global Assessment of disease activity; serum concentration of otilimab; and safety parameters. Efficacy endpoints were assessed in the intention-to-treat population. The safety population included all patients who received at least one dose of study treatment. The study is registered with ClinicalTrials.gov, NCT02683785.

Findings: Between March 17, 2016, and Nov 29, 2017, 44 patients were randomly assigned (22 in the placebo group and 22 in the otilimab group). At week 6, difference in change from baseline in 24 h average hand pain numeric rating scale between the otilimab and placebo groups was -0·36 (95% CI -1·31 to 0·58; p=0·44); at week 12, the difference was -0·89 (-2·06 to 0·28; p=0·13). Patients receiving otilimab showed greater improvement in AUSCAN components versus placebo at each visit. The change from baseline in the 24 h worst hand pain numeric rating scale in the otilimab group at week 6 showed a difference over placebo of -0·33 (95% CI -1·28 to 0·63; p=0·49); at week 12, this difference was -1·01 (95% CI -2·22 to 0·20; p=0·098). The proportion of patients achieving 30% or higher or 50% or higher reduction from baseline in the 24 h average and worst hand pain numeric rating scale scores was consistently greater (although non-significant) with otilimab versus placebo. Similarly, patients receiving otilimab showed greater improvement in AUSCAN pain, functional impairment, and stiffness scores versus placebo at each visit. No differences were observed between otilimab and placebo in the change from baseline in the number of swollen and tender joints across the study. The Patient Global Assessment was consistently lower than placebo at all visits; the Physician Global Assessment showed reductions across the study period, but the changes were similar in both treatment groups. Median otilimab serum concentrations increased during weekly dosing from 1730 ng/mL at week 1 to a maximum of 3670 ng/mL at week 4, but declined after transitioning to dosing every other week (weeks 6-10). In total, 84 adverse events were reported by 24 patients: 54 adverse events in 13 (59%) patients in the otilimab group and 30 adverse events in 11 (50%) patients in the placebo group. The most common adverse events were cough (reported in 4 [9%] patients; 2 in each group), and nasopharyngitis (in 3 [7%] patients; 1 in the placebo group and 2 in the otilimab group). Three serious adverse events occurred in this study (all in the otilimab group) and were deemed not related to the study medication. There were no deaths during the study.

Interpretation: There was no significant difference between otilimab and placebo in the primary endpoint, although we noted a non-significant trend towards a reduction in pain and functional impairment with otilimab, which supports a potential role for GM-CSF in hand osteoarthritis-associated pain. There were no unexpected safety findings in this study, with no treatment-related serious adverse events reported.

Funding: GlaxoSmithKline.