AHA Scientific Sessions 2026 Abstract · Nov 2, 2026

Cardiovascular Outcomes of Glucagon-Like Peptide-1 Receptor Agonists Compared with Dipeptidyl Peptidase-4 Inhibitors in Adults with Type2 Diabetes Mellitus Receiving Anthracycline Based Chemotherapy: A Propensity Matched TriNetX Multicenter Cohort Study

By Komolafe B, Vasquez S, Okorigba EM et al.

Background: Anthracycline-based chemotherapy causes cardiotoxicity through oxidative stress, mitochondrial dysfunction, and direct myocardial injury. Type 2 diabetes mellitus (T2DM) increases this risk through impaired myocardial energetics and systemic inflammation. Glucagon-like peptide-1 receptor agonists (GLP-1 RA) may counter these pathways, whereas dipeptidyl peptidase-4 inhibitors (DPP-4) are cardiovascularly neutral comparators. Real-world evidence remains lacking in oncology patients with T2DM. Hypothesis: We sought to determine whether GLP-1 RA use, compared with DPP-4 inhibitor use, was associated with a lower risk of incident heart failure with reduced ejection fraction (HFrEF) or dilated cardiomyopathy (DCM) in adults with T2DM receiving anthracycline-based chemotherapy. Methods: We used the TriNetX United States Collaborative Network, a federated EHR database spanning 67 health care organizations, to identify adults with T2DM who received anthracycline-based chemotherapy from 2015 through 2024 and later initiated a GLP-1 RA (including tirzepatide, a dual glucose-dependent insulinotropic polypeptide receptor and GLP-1 RA) or DPP-4 inhibitor. Cohorts were mutually exclusive. After one-to-one propensity score matching, 878 matched pairs were analyzed over 365 days. The primary outcome was incident HFrEF or DCM. Secondary outcomes included all-cause mortality (ACM), three-point major adverse cardiovascular events (MACE), composite arrhythmia, and incident atrial fibrillation (AF). Cox proportional hazards models were used, with results reported as hazard ratios (HR) with 95% confidence intervals (CI). Results: After matching, most baseline characteristics were balanced; however, there were imbalances in body mass index (32.9 versus 30.3 kg/m ² ) and left ventricular ejection fraction (58.1% versus 61.3%). The primary outcome did not differ significantly (2.5% versus 2.7%; HR 0.90, p=0.724). GLP-1 RA use was associated with lower ACM (9.8% versus 17.7%; HR 0.53, CI 0.41-0.69; p<0.001), lower MACE (10.8% versus 19.2%; HR 0.54, CI 0.41-0.71; p<0.001), lower AF (2.6% versus 4.8%; HR 0.52, CI 0.30-0.90; p=0.018), and lower composite arrhythmia (3.5% versus 6.1%; HR 0.55, CI 0.34-0.88; p=0.012). Conclusions: GLP-1 RA use was not associated with lower one-year risk of HFrEF or DCM, compared with DPP-4 inhibitor use. The null primary endpoint likely reflects insufficient follow-up, as anthracycline-induced cardiomyopathy typically manifests over years. Nevertheless, GLP-1 RAs were associated with significantly lower ACM, MACE, and incident AF. These findings are limited by the observational design, post-matching imbalances, unmeasured confounding from cancer severity and performance status, and follow-up insufficient to capture anthracycline-induced cardiomyopathy; prospective trials with extended follow-up are warranted.

doi.org/10.1161/circ.154.suppl_1.16588Interventions

AgingResearch.News