GLP-1 Agonists Reduce Recurrent Atrial Fibrillation

SAN DIEGO — Patients with obesity and diabetes are at a substantially reduced risk of having a recurrence of atrial fibrillation (AF) after ablation if they are taking a glucagon-like peptide 1 (GLP-1) receptor agonist vs another diabetes drug, according to an analysis of matched cohorts with 3 years of follow-up.

The study was retrospective, but it included more than 2500 patients taking a GLP-1 or one of the two comparators, reported Varun Sundaram, MD, PhD, section chief, Advanced Heart Failure, Louis Stokes Cleveland Veterans Affairs (VA) Medical Center, Cleveland.

After a median of 3.1 years of follow-up, the odds ratio (OR) of the primary composite outcome of time to first AF hospitalization, an AF-related procedure, or all-cause mortality was reduced by 13% (OR, 0.87; P = .03) compared with treatment with a dipeptidyl peptidase 4 (DPP-4) inhibitor or sulfonylurea.

Weight Loss Might Not Explain Effect

While weight loss on the GLP-1 did occur, Sundaram speculates that it did not fully explain the benefit of the drug, as it was only modestly greater than in the control arms (14.1% vs 10.9%). He instead believes GLP-1s have “pleiotropic” effects, resulting in a favorable impact on the metabolic abnormalities that provide a substrate for AF recurrence.

An antidiabetic drug with a favorable impact on the long-term risk for AF recurrence is an important unmet need, according to Sundaram. While ablation offers high rates of acute AF control, he pointed to the fact that one third of patients have a recurrence within 1 year and that rates continue to rise with longer follow-ups.

“We believe this is likely related to progressive atrial substrate remodeling,” said Sundaram, who said this study supports the ability of GLP-1s to reduce the residual risk for AF recurrence after a primary ablation.

In this trial, called TRANSFORM-AF, which was a late-breaking clinical trial presented at Heart Rhythm 2025, patients with type 2 diabetes and a body mass index (BMI) > 30 who were starting a new prescription of a GLP-1, DPP-4 inhibitor, or sulfonylurea were drawn from a pool of more than 70,000 patients in the VA system.

After propensity matching on the basis of 31 covariates, 1226 patients starting a GLP-1 were compared with 1284 of those taking either a DPP-4 inhibitor or sulfonylurea. The standardized between-group difference for any one of the covariates was less than 10%.

In a secondary analysis that assessed repeating AF events with mortality as a competing risk, the risk reduction of AF recurrence over the course of follow-up was 15% (OR, 0.85), which suggested a trend (95% CI, 0.61-1.03).

The TRANSFORM-AF study was limited to patients with AF and diabetes only because this is the population in the VA system taking a GLP-1. Sundaram said it is unknown if a similar reduction in recurrence would be observed in patients without diabetes with or without obesity taking a GLP-1.

The major limitation of this study was that it was not randomized, Sundaram acknowledged. Despite propensity matching, he cautioned that the risk for residual confounding cannot be excluded.

Randomized Trial Needed

A randomized trial is needed to confirm these findings, but Sundaram said such a trial will be “challenging” for a number of reasons. In patients with obesity and diabetes, a study design that includes randomizing patients to drugs other than GLP-1 might impair recruitment. The long-term follow-up required to accrue sufficient events might also make the study cost unattractive.

The association between GLP-1 and a reduced risk for AF recurrence in TRANSFORM-AF supported a hypothesis based on experimental studies. In a 2023 editorial accompanying a published study showing a reduction in AF in an animal model of diabetes, the authors of the editorial cited other effects — such as improvement in left ventricular function that they saw as support for effects beyond weight loss and a basis for conducting clinical trials.

The level of interest in the concept modifying the substrate of AF recurrence is “high,” according to Sanjeev Saksena, MD, an electrophysiologist and a clinical professor of medicine at the Robert Wood Johnson Medical School, Rutgers University, New Brunswick, New Jersey. He said this study is an early step in exploring the hypothesis.

Saksena considers supportive evidence from a “robust pilot” study essential to move toward a pivotal trial. He said that the data from this study, although provocative, are insufficient for projecting benefit.

On the basis of this study alone, “there are several reasons to question the impact on AF,” Saksena said. For one, there was a large proportion of patients with obesity, diabetes, and AF who were excluded from the analysis for various reasons. For another, AF was not measured quantitatively. He also pointed to potential problems with the matching technique to optimize two comparable groups.

As the Heart Rhythm Society–invited discussant on this study, Saksena called for a randomized trial to prove the hypothesis. He did not agree that such a controlled trial would be impossible, noting that that GLP-1 drugs such as semaglutide are not the first-line treatment for obesity, AF, or diabetes, and that more evidence of a favorable benefit-risk ratio is needed.

He pointed out that TRANSFORM-AF had many strengths. For one, the evidence of even greater benefit in those with the highest baseline BMI was reassuring. He also agreed that that new options for preventing AF recurrence are certainly needed. Saksena said he looks forward to further evidence that the evidence of benefit can be confirmed, providing a new option for clinical practice.

Sundaram reported having no conflicts of interest relevant to this study. Saksena reported having financial relationships with Abbott and Sanofi.