Cagrilintide-semaglutide (CagriSema) as an add-on to basal insulin in adults with type 2 diabetes (REIMAGINE 3): a randomised, double-blind, placebo-controlled, multicentre, phase 3 study
Abstract
Background: Basal insulin treatment for type 2 diabetes often results in inadequate glycaemic control and is associated with weight gain and increased risk of hypoglycaemia. We aimed to compare the efficacy and safety of a once per week combination of cagrilintide with semaglutide (CagriSema) versus placebo as an add-on to basal insulin in individuals with type 2 diabetes.
Methods: This double-blind, parallel-group, randomised, controlled, phase 3a study (REIMAGINE 3) was done at 46 centres (university hospitals, health-care centres, research centres, and other clinical trial sites) in six countries (the USA, China, Japan, Serbia, Slovakia, and South Africa). Adults with type 2 diabetes (glycated haemoglobin [HbA1c] 7·0-10·5%) receiving stable once per day basal insulin with or without metformin were randomly assigned (2:2:1:1) to once per week subcutaneous cagrilintide 2·4 mg plus semaglutide 2·4 mg (hereafter cagrilintide-semaglutide [2·4 mg each]) or cagrilintide 1·0 mg plus semaglutide 1·0 mg (hereafter cagrilintide-semaglutide [1·0 mg each]) or dose-matched placebo (2·4 mg plus 2·4 mg or 1·0 mg plus 1·0 mg) for 40 weeks. Randomisation was stratified by HbA1c of less than 8·5% at screening (yes or no) and country (Japan; yes or no). Participants, care providers, investigators, and outcome assessors were masked within each dose level to treatment allocation, with active treatments visually identical to placebo. The primary endpoint was mean HbA1c change (percentage points) from baseline to week 40 in all participants randomly assigned to treatment; secondary endpoints included change in bodyweight and safety, including hypoglycaemia. This trial was registered with ClinicalTrials.gov (NCT06323161) and is complete.
Findings: Between March 26 and Nov 29, 2024, we screened 340 individuals and 274 were included and randomly assigned to cagrilintide-semaglutide (2·4 mg each; n=90), cagrilintide-semaglutide (1·0 mg each; n=93), or placebo (pooled; n=91). Mean baseline HbA1c was 8·8% (SD 1·0), 159 (58%) participants were male, and 115 (42%) were female. Mean HbA1c reductions were significantly greater with cagrilintide-semaglutide (2·4 mg each -2·33% [SE 0·08] and 1·0 mg each -2·10% [0·08]) versus placebo (-0·66% [0·11]) at week 40, using the efficacy estimand (estimated treatment difference for cagrilintide-semaglutide [2·4 mg each] vs placebo -1·68 percentage points [95% CI -1·95 to -1·41], p<0·0001; for cagrilintide-semaglutide [1·0 mg each] vs placebo -1·44 percentage points [95% CI -1·71 to -1·17], p<0·0001). Cagrilintide-semaglutide provided bodyweight reductions of 10-12%. Adverse events were reported by 72 (80%) of 90 participants receiving cagrilintide-semaglutide (2·4 mg each), 66 (71%) of 93 receiving cagrilintide-semaglutide (1·0 mg each), and 65 (71%) of 91 receiving placebo, and were mostly mild or moderate gastrointestinal disorders. No severe hypoglycaemia was reported. There was one death in the cagrilintide-semaglutide (1·0 mg each) group not related to treatment (due to malignancy).
Interpretation: Cagrilintide-semaglutide at doses of 2·4 mg each and 1·0 mg each met the primary endpoint, with statistically significant and clinically relevant HbA1c reductions versus placebo when added to basal insulin-treated type 2 diabetes. These reductions were accompanied by robust bodyweight reduction and no additional risk of hypoglycaemia. The safety profile was consistent with that of the GLP-1 receptor agonist class and previous safety data for cagrilintide. Findings support the use of cagrilintide-semaglutide as an add-on to once per day basal insulin to significantly improve glycaemic control.
Funding: Novo Nordisk.
Authors: Julio Rosenstock, Liana K Billings, Reena Gajria, Francesco Giorgino, Nanna Borup Johansen, Klara R Klein, Usha K Thamattoor, John B Buse, Akshay B Jain
Journal: Lancet (London, England)