Evaluate anti‐interleukin‐1β (IL‐1β) antibody, canakinumab, in patients with type 2 diabetes and impaired glucose tolerance (IGT) in whom hyperglycaemia may trigger IL‐1β‐associated inflammation leading to suppressed insulin secretion and β‐cell dysfunction.

This 4‐week, parallel‐group study randomized 190 patients with type 2 diabetes 2 : 1, canakinumab versus placebo, into the following treatment arms: metformin monotherapy, metformin + sulfonylurea, metformin + sulfonylurea + thiazolidinedione or insulin ± metformin. IGT population (n = 54) was randomized 1 : 1, canakinumab versus placebo. Primary efficacy assessment was change from baseline in insulin secretion rate (ISR) relative to glucose 0–2 h.

Mean changes from baseline to week 4 in ISR relative to glucose at 0–2 h or other time points were not statistically significant for canakinumab versus placebo across groups. ISR (relative to glucose) at 0–0.5 h (first‐phase insulin secretion) numerically favoured canakinumab versus placebo in insulin‐treated patients {difference in mean change from baseline [point estimate (PE)] 3.81 pmol/min/m²/mmol/l; p = 0.0525} and in the IGT group (PE 3.92 pmol/min/m²/mmol/l; p = 0.1729). Mean change from baseline in fasting plasma glucose favoured canakinumab in the type 2 diabetes/metformin group and the IGT group; however, differences were not statistically significant. Mean change from baseline in peak insulin level and insulin AUC 0–4 h were statistically significantly higher in the canakinumab group in IGT patients. Canakinumab was well tolerated and consistent with known safety experience.

The trend towards improving ISR relative to glucose 0–0.5 h in patients treated with insulin supports the hypothesis that insulin secretion can be improved by blocking IL‐1β.