GLP-1s Provide Benefit in HF Patients: SOUL Subanalysis

An oral GLP-1 medication may lower the risk of hospitalization or urgent visits for heart failure (HF), as well as CV death, in HF patients with diabetes and atherosclerotic cardiovascular disease (ASCVD) and/or chronic kidney disease, according to a subanalysis of the SOUL trial.

Compared with placebo, the risk of the composite outcome of time to first occurrence of HF hospitalization, urgent HF visit, or CV death was 22% lower over approximately 4 years of follow-up in those who had HF at baseline and took once-daily oral semaglutide (Rybelsus; Novo Nordisk). There was no reduction in this HF-oriented endpoint in those without HF at baseline.

Rodica Pop-Busui, MD, PhD (Oregon Health & Science University, Portland, OR), who led the analysis, said the findings provide clinicians with additional options when considering how to treat this population of patients.  

“Another important observation is that the efficacy of the oral formulation of semaglutide has been now demonstrated. There were some concerns that with a daily oral medication, absorption or compliance issues may affect [its] efficacy on important outcomes,” she noted. “That’s important because having an opportunity to give another medication, particularly to people who are more frail or who may not have the capability to access [and store] injections, offers much broader access to these drugs.”

Moreover, “there were some concerns about the potential effect of GLP-1s in people with heart failure regarding safety and that has obviously not been the case,” said Pop-Busui.

In the main SOUL analysis, oral semaglutide taken once daily cut the risk of cardiovascular mortality, nonfatal MI, or nonfatal stroke by 14% at a mean follow-up of 47.5 months, a benefit driven by the reduction in nonfatal MI. In October 2025, the US Food and Drug Administration gave the drug an expanded indication to reduce major adverse cardiovascular events in patients with type 2 diabetes and high cardiovascular risk on the basis of the SOUL trial results.

Barry Borlaug, MD (Mayo Clinic, Rochester, MN), who led an analysis of the SUMMIT trial that found that another GLP-1 agent, tirzepatide (Zepbound; Eli Lilly), improved outcomes in patients who had heart failure with preserved ejection fraction (HFpEF) and obesity, said the subanalysis adds to a growing body of evidence supporting GLP-1 receptor agonists in this patient population.

Compared with placebo, those with HFpEF who were on oral semaglutide in the SOUL trial had a 41% reduction in risk of time to first composite HF outcome event, whereas those with heart failure with reduced ejection fraction (HFrEF) had no significant reduction compared with placebo.

“If you really want to individualize treatment, you want to identify those patients that are best positioned to respond,” Borlaug said.  ”It looks like it’s a more favorable effect in HFpEF. It’s also interesting that in patients without heart failure history there was no evidence—not even really a suggestion—of a benefit, which would suggest that there doesn’t appear to be any kind of a role [for GLP-1s] in terms of preventing heart failure events.”

He further added that the accumulating data make the case that non-obese patients can benefit from the GLP-1s with no evidence that they are losing weight unnecessarily.

SOUL Analysis Results

The SOUL trial randomized 9,650 patients from 444 centers in 33 countries. The subanalysis focused on the 2,229 patients (mean age 66 years; 30% female) who had a diagnosis of HF at baseline: HFpEF in approximately 44%, HFrEF in 26%, and unknown subtype in the remainder.  

The rate of HF hospitalization or urgent HF visit among patients with HF was 6.5% in the oral semaglutide group and 8.1% in the placebo group, whereas it was 2.0% versus 2.1%, respectively, when no HF was present at baseline. Rates of CV death were 10.1% with oral semaglutide and 11.7% in the placebo group in those with HF at baseline, and 5.1% in both groups in those with no HF.

An analysis of HF composite outcome events through 3 years showed an absolute risk reduction of 3.5% with semaglutide versus placebo (P = 0.01), with a number needed to treat of 29 to prevent one event.

For MACE, the overall population saw a 14% reduction with oral semaglutide compared with placebo. For those with HFpEF at baseline, the risk of MACE was 32% lower with semaglutide versus placebo, while in those with HFrEF it was 7% lower and in those with unknown HF subtype it was 11% lower.

With all this knowledge and all this evidence that we are generating, we are now in a position to try to apply a more personalized care to our patients. Rodica Pop-Busui

Pop-Busui and colleagues theorize that the larger treatment effect in HFpEF may be related to  mechanistic differences by HF subtypes. Specifically, HFpEF is driven by coronary microvascular dysfunction and compromised nitric oxide. While SOUL wasn’t powered to detect treatment differences by HF phenotype, Pop-Busui said about one-quarter of all patients in the trial—whether they had HF or not, or were on semaglutide or placebo—were taking sodium-glucose co-transporter-2 (SGLT2) inhibitors at baseline.

“Being able to see the benefits regardless of the SGLT2 use demonstrates the potential benefits of adding oral semaglutide [because] even in people who have been treated with another agent [you] still see a reduction in these hard outcomes,” she added.

No meaningful interactions were seen between the effects of oral semaglutide on HF outcomes in those on loop diuretics, mineralocorticoid receptor antagonists, or SGLT2 inhibitors. Adverse safety events were numerically lower in the HFpEF patients randomized to oral semaglutide compared with placebo.

Pop-Busui said that while cost considerations always enter the conversation when considering GLP-1 agents, the other part of the conversation is the potential to cut costs down the road by not only reducing the risk for events in this population, but also scaling back on some of their medications.

“With all this knowledge and all this evidence that we are generating, we are now in a position to try to apply a more personalized care to our patients . . . and there are lots of opportunities to integrate tools in the electronic medical record that can give clinicians a more comprehensive view of what’s happening with a given patient at a given time [to help] make treatment choices,” she added.