Abstract 17139: Angiopoietin 2 Independently Predicts Heart Failure Hospitalization and is Beneficially Modified by GLP1RA: Results From the EXSCEL Trial
Harrington, J; Nguyen, M; Peters, A; Felker, M; Karra, R; Sattar, N; Sourij, H; Hernandez, AF; Green, J; Mentz, RJ; Holman, R; Shah, SH
Published in: Circulation
Higher plasma concentrations of angiopoietin 2 (ANGPT2) are associated with higher risk for hospitalization for heart failure (HF) and death in univariate analyses. However, the independent association of this ANGPT2 after adjustment for clinical factors or clinically available biomarkers is not known.
High throughput proteomic profiling was performed using the SomaLogic SomaScan platform in 5473 patients with type 2 diabetes enrolled in the EXSCEL clinical trial of the GLP1RA exenatide vs. placebo who had available biospecimens at baseline. The association between ANGPT2 with time-to-HF hospitalization, MACE, and death was assessed using univariable and multivariable Cox modeling, with adjustment for age, sex, HbA1c, NT-proBNP, hs-TnT, left ventricular ejection fraction, HF etiology and study treatment. These analyses were also performed in the subset of participants living with obesity (BMI >30 kg/m
), and those with CKD (CrCl <60ml/min).
In 5473 EXSCEL trial participants included in this biomarker sub-study, 215 (3.9%) experienced a hospitalization for heart failure, 813 (14.9%) a MACE event and 528 (9.7%) died during a median follow-up of 3.2 years. Relative ANGPT2 concentration was significantly associated with HF hospitalization (aHR 1.70 [95% CI 1.39 to 2.08]), MACE (1.20 [1.07 to 1.35]), and death (aHR 1.21 [1.05 to 1.39]) in univariable and multivariable analyses, including after adjustment for NT-proBNP and TnT (Figure). Exenatide use resulted in significant changes from baseline in ANGPT2 concentrations (p interaction<0.001) (Figure).
ANGPT2 is predictive of hospitalization for HF, MACE, and death independent of clinical factors including traditional biomarkers of NT-proBNP and TnT. ANGPT2 was beneficially modified by exenatide, a GLP1RA, suggesting a potential role for this biomarker in optimizing HF therapy.
