GLP-1 receptor agonists associated with better primary CV prevention vs. DPP-4 inhibitors

Adding glucagon-like peptide-1 (GLP-1) receptor agonists versus dipeptidyl peptidase-4 (DPP-4) inhibitors was associated with lower rates of first major adverse cardiovascular events (MACE) and heart failure (HF) hospitalization, a recent study found.

The retrospective cohort study included adult veterans with diabetes but without established cardiovascular disease receiving care from the Veterans Health Administration from 2001 to 2019 and used data linked to Medicare, Medicaid, and the National Death Index. It included veterans who added GLP-1 receptor agonists, sodium-glucose cotransporter-2 (SGLT-2) inhibitors, or DPP-4 inhibitors onto metformin, sulfonylurea, or insulin treatment alone or in combination. The composite outcome was the time to MACE (acute myocardial infarction, stroke, or cardiovascular death) or HF hospitalization. Results were published May 9 by Annals of Internal Medicine.

The cohort included 28,759 patients taking a GLP-1 receptor agonist paired with 28,628 DPP-4 inhibitor patients and 21,200 SGLT-2 inhibitor patients paired with 21,170 on DPP-4 inhibitors. The median age was 67 years. GLP-1 receptor agonist use was associated with reductions in MACE and HF hospitalizations (adjusted hazard ratio [HR], 0.82; 95% CI, 0.72 to 0.94), yielding an adjusted risk difference of 3.2 events (95% CI, 1.1 to 5.0) per 1,000 person-years. In contrast, use of SGLT-2 inhibitors did not show a significant difference (adjusted HR, 0.91 [95% CI, 0.78 to 1.08]; adjusted risk difference, 1.28 [95% CI, −1.12 to 3.32]). SGLT-2 inhibitor use was associated with numerically fewer HF hospitalizations. In addition, among the complete cohort that included patients both with and without cardiovascular disease, both GLP-1 receptor agonists and SGLT-2 inhibitors were associated with reduced MACE and HF hospitalizations compared with DPP-4 inhibitors. The median follow-up was 0.58 year (7 months) for the comparison of GLP-1 receptor agonists with DPP-4 inhibitors and 0.42 year (5 months) for that of SGLT-2 inhibitors with DPP-4 inhibitors.

Patients were excluded if the initial diabetes therapy did not include metformin, sulfonylurea, or insulin, the study authors noted. Other limitations of the study include its observational design and short duration of follow-up, they added.

Given the short follow-up, the lack of difference between SGLT-2 and DPP-4 inhibitors should not be interpreted as lack of effectiveness in primary prevention, an accompanying editorial noted. In addition, the observed differences in HRs comparing GLP-1 receptor agonists and SGLT-2 inhibitors with DPP-4 inhibitors are too small to provide reliable conclusions, given the limitations of the observational study, the editorial said.

“In summary, large observational studies can be useful and informative, but the choice of outcomes measured and the study method must be carefully considered, and the results interpreted within the context of the study's limitations,” the editorialist wrote. “Caution and skepticism are appropriate when the effects are modest.”