Abstract Body (Do not enter title and authors here): Background: Persons with chronic kidney disease (CKD) and type 2 diabetes (T2D) face high but modifiable risks of incident heart failure (HF). The HFpEF-ABA algorithm, originally developed for HFpEF diagnosis, may additionally predict adverse outcomes. Yet, whether the HFpEF-ABA model can stratify risk of incident HF remains uncertain.

Purpose: To ascertain whether the HFpEF-ABA model 1) distinguishes between patients with and without HF 2) is associated with adverse cardiovascular(CV) outcomes in patients without HF and 3) modifies the efficacy of finerenone.

Methods: In this participant-level pooled analysis from FINE-HEART including 3 phase-3 trials (FINEARTS-HF and FIDELIO-DKD/FIGARO-DKD (FIDELITY)) HFpEF-ABA score (based on age, body mass index, and atrial fibrillation status) was calculated at baseline. First, the performance for HF identification in the overall cohort was assessed. Second, clinical outcomes and treatment effects of finerenone versus placebo were evaluated according to continuous and categorical (low/intermediate<75%; high≥75%) HFpEF-ABA score among FIDELITY participants without HF.

Results: Among 18,991 FINE-HEART participants, 18,943(99.7%) had a calculable HFpEF-ABA score at baseline (mean age 67±10years; 35% female; 37% with HF). The HFpEF-ABA model had moderate performance for discriminating participants with vs. without HF (Cstatistic 0.73). Among FIDELITY patients without HF (n=11,950), those with a high (28%) vs. low/intermediate (72%) HFpEF-ABA score had a higher rate of CV death or incident HF hospitalization (HR 2.20; 95%CI, 1.89-2.57;P<0.001) and incident HF hospitalization (HR 2.58; 95%CI, 2.12-3.14;P<0.001). However, the CV event rates were lower compared with established HF, even among participants with HFpEF-ABA scores >90%(Figure 1). Finerenone reduced CV death or incident HF hospitalization irrespective of baseline HFpEF-ABA score(Figure 2), with greater absolute benefits among those with a high (absolute rate reduction[ARR] 0.8 per 100 person-years (py)) vs. low/intermediate (ARR 0.2 per 100 py) HFpEF-ABA score. Serious adverse events were less common with finerenone vs. placebo in both HFpEF-ABA score categories.

Conclusions: The HFpEF-ABA model identified patients with CKD and T2D with higher risk of incident HF, but only select individuals with the highest scores experienced clinical events commensurate with those with established HF. Finerenone consistently reduced CV events across a broad HFpEF-ABA score spectrum.