Time-dependent effect of GLP-1 receptor agonists on cardiovascular benefits: a real-world study

Background: Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) have shown cardiovascular benefits in cardiovascular outcome trials in type 2 diabetes mellitus. However, the most convincing evidence was obtained in subjects with established cardiovascular (CV) disease. We analyzed the determinants of GLP-1 RA-mediated CV protection in a real-world population of persons with type 2 diabetes with and without a history of CV events with long-term follow-up. Methods: Retrospective cohort study of 550 individuals with type 2 diabetes (395 in primary CV prevention, 155 in secondary CV prevention), followed at a single center after the first prescription of a GLP-1 RA between 2009 and 2019. CV and metabolic outcomes were assessed. Results: Median duration of follow-up was 5.0 years (0.25-10.8) in primary prevention and 3.6 years (0-10.3) in secondary prevention. Median duration of treatment was 3.0 years (0-10.8). In the multivariate model, in primary prevention, duration of GLP-1 RA treatment >3 years (HR 0.19, 95% CI 0.04-0.96, P=0.044) and use of pioglitazone (HR 1.01x10-6, 95% CI 2.74x10-12-0.37, P=0.035) were associated to the risk reduction of composite outcome (MACE) which included non-fatal myocardial infarction or unstable angina, non-fatal stroke and all-cause death, while discontinuation of GLP-1 RA treatment (HR 5.58, 95% CI 1.84-16.87, P=0.034) and age (HR 1.06, 95% CI 1.01-1.11, P=0.022) were associated to higher risk. Similarly, in secondary prevention, duration of GLP-1 RA treatment >3 years (HR 0.07, 95% CI 0.01-0.30, P<0.001) turned out to be protective and conversely GLP-1 RA withdrawal (HR 4.93, 95% CI 1.37-17.69, P=0.014), was significantly associated to an increased risk of MACE. When adding hospitalizations for heart failure to the composite outcome, duration of GLP-1 RA treatment >3 years remained significant in the multivariate model in both groups. With respect to those who withdrew treatment, subjects who continued the GLP-1 RA had significantly greater weight loss and lower glycated hemoglobin levels during follow-up. Conclusions: In this real-world type 2 diabetes population, longer duration of GLP-1 RA treatment was associated to a reduced risk of major cardiovascular events, while medication withdrawal increased MACE risk in both subjects with and without a history of CV events.