Major adverse cardiovascular events with basal insulin peglispro versus comparator insulins in patients with type 1 or type 2 diabetes: a meta-analysis

BACKGROUND: To identify possible differences in cardiovascular (CV) risk among different insulin therapies, we performed pre-specified meta-analyses across the clinical program for basal insulin peglispro (BIL), in patients randomized to treatment with BIL or comparator insulin [glargine (IG) or NPH]. METHODS: One phase 2 (12-week) and 6 phase 3 (26 to 78-week) randomized studies of BIL compared to IG or NPH, in patients with type 1 or type 2 diabetes, were included. The participants were diverse with respect to demographics, baseline glycemic control, and concomitant disease or medications, but treatment groups were comparable in each study. For any potential CV or neurovascular event, relevant medical information was provided to a blinded external clinical events committee (C5Research, Cleveland Clinic, Cleveland, OH, USA) for adjudication. Cox regression analysis was used to compare treatment groups. The primary endpoint was a composite of adjudicated MACE+ [CV death, myocardial infarction (MI), stroke, or hospitalization for unstable angina]. RESULTS: The pooled population included 5862 patients in the safety evaluation, with randomization to BIL:IG:NPH of 3578:2072:212. Mean age was 54.1 years, 27 % had type 1 diabetes, 56 % were male, and 88 % were white. Baseline demographic and clinical characteristics, including use of statins or other lipid-lowering drugs, were comparable between BIL and comparators. A total of 83 patients experienced at least 1 MACE+ and 70 patients experienced at least 1 MACE (CV death, MI, or stroke). Overall, there were no treatment-associated differences in time to MACE+ [hazard ratio (HR) for BIL versus comparator insulin (95 % CI): 0.82 (0.53–1.27)] or MACE [0.83 (0.51–1.33)]. In 4297 patients with type 2 diabetes, there were 71 MACE+ events [HR: 1.02 (95 % CI: 0.63–1.65), p = 0.94]. In 1565 patients with type 1 diabetes, there were only 12 MACE+ [0.24 (0.07–0.85), p = 0.027]. There were no differences in all-cause death between BIL and comparators. Sub-group analyses did not identify any sub-population with increased risk with BIL versus comparator insulins. CONCLUSIONS: Treatment with BIL versus comparator insulin in patients with type 1 diabetes or type 2 diabetes was not associated with increased risk for major CV events in the studies analyzed. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12933-016-0393-6) contains supplementary material, which is available to authorized users.