MUNICH — Direct oral anticoagulant (DOAC) therapy reduced the risk for adverse clinical events by 69% compared with no anticoagulation in patients with atrial fibrillation (AF) who are at intermediate risk for stroke, according to results of the randomized SINGLE-AF trial.

At 24 months, the primary composite endpoint of stroke, systemic embolism, major bleeding, or cardiovascular death occurred in 0.5% of patients receiving a DOAC compared with 1.5% of those receiving no anticoagulation.

“We now have evidence from a randomized trial that patients with AF at intermediate stroke risk benefit from DOAC therapy, without an increase in major bleeding,” said senior author and presenter Boyoung Joung, MD, a professor of cardiology at Yonsei University College of Medicine in Seoul.

“The [prior] level of evidence in this population is actually very low, based on observational studies, and the results vary a lot,” Joung said, noting that evidence for other anticoagulants, particularly vitamin K antagonists, has remained conflicting in patients at intermediate risk. “This is the first randomized controlled trial to support that DOAC therapy can really provide a benefit to this patient population.”

The findings were presented at the European Society of Cardiology (ESC) Congress 2026 and simultaneously published in The New England Journal of Medicine.

Evidence Gap in Intermediate-Risk AF

Current US and European guidelines strongly recommend oral anticoagulation for men with a CHA ₂ DS ₂ -VASc score of 2 or more and women with a score of 3 or more. 

For patients at intermediate risk for stroke, defined as a CHA ₂ DS ₂ -VASc score of 1 in men and 2 in women, anticoagulation should be considered (a class IIa recommendation). However, randomized evidence supporting treatment in this group has been limited and conflicting. 

SINGLE-AF was a multicenter, open-label, randomized superiority trial conducted in South Korea. The trial included 1803 patients randomly assigned 1:1 to DOAC therapy or no anticoagulant therapy. In the DOAC group, 67.4% were assigned apixaban 5 mg twice daily, 14.1% were assigned rivaroxaban 20 mg once daily, and 13.9% were assigned rivaroxaban 15 mg once daily. Mean age of the entire cohort was 60.4 years, 23.7% were women, and 71.8% had paroxysmal AF. The median time from AF diagnosis to randomization was 21 months.

Of those on DOAC therapy, 81.6% received full doses and 18.4% received reduced doses. During the trial, 5.5% of patients assigned to no anticoagulation received a DOAC.

Antiplatelet therapy was used in 36.5% of patients in the no-anticoagulant group compared with 0.4% in the DOAC group. Median follow-up was 730 days, and 94.6% of patients in each group completed 24-month follow-up. The composite primary endpoint was stroke, systemic embolism, major bleeding, or death from cardiovascular causes at 2 years post-randomization.

Fewer Strokes Without an Apparent Major-Bleeding Penalty

Regarding the 24-month primary endpoint, four patients (cumulative incidence, 0.5%) in the DOAC group and 13 (cumulative incidence, 1.5%) in the no-anticoagulant group experienced an event as defined by the composite primary endpoint (difference, −1.0 percentage points; 95% CI, −2.0 to −0.1; P =.03; hazard ratio [HR], 0.31; 95% CI, 0.10-0.94).

Stroke occurred in three patients receiving DOAC therapy compared with 10 patients receiving no anticoagulation (cumulative incidence, 0.3% vs 1.1%; HR, 0.30; 95% CI, 0.08-1.08). Systemic embolism occurred in no patients in the DOAC group and one patient in the no-anticoagulation group, while major bleeding occurred in three patients in the DOAC group (0.3%) and four patients in the no-anticoagulation group (0.5%; HR, 0.75; 95% CI, 0.17-3.35). 

Clinically relevant nonmajor bleeding was numerically more frequent with anticoagulation, occurring in 22 patients (2.5%) receiving a DOAC compared with 14 patients (1.6%) receiving no anticoagulation (HR, 1.58; 95% CI, 0.81-3.08). Intracranial hemorrhage occurred in two patients in the DOAC group, both of whom had hemorrhagic stroke, and one in the no-anticoagulation group, who did not have hemorrhagic stroke.

There were no cardiovascular deaths or myocardial infarctions in either group. Death from any cause occurred in three patients receiving DOAC therapy and five receiving no anticoagulation (HR, 0.60; 95% CI, 0.14-2.51).

Serious adverse events occurred in 8.9% of patients in the DOAC group and 9.3% in the no-anticoagulation group, with the most common being hospitalization related to AF (34 events in the DOAC group and 15 in the no-anticoagulant group).

Low Event Rates Temper Findings

Although the relative reduction was substantial, the investigators cautioned that the small number of events (4 vs 13) may overestimate the benefit and that confirmatory trials or pooled analyses are needed. 

The authors highlighted several limitations, including the open-label design and the lower-than-anticipated amount of primary endpoint events. About 5.4% of patients in each group did not complete the 2-year follow-up.

Generalizability may also be limited because the trial enrolled only East Asian patients from South Korea, and women and patients with nonparoxysmal AF may have been underrepresented.

In an accompanying editorial, Paulus Kirchhof, MD, of University Heart and Vascular Center Hamburg-Eppendorf in Germany and the University of Birmingham in the UK, said the findings provide evidence supporting the efficacy and safety of DOACs in patients with AF and a single stroke risk factor. However, he cautioned that the small number of events makes the size of the apparent benefit uncertain. With only 17 primary endpoint events, the trial “most likely overestimated the true treatment effect of DOACs” and did not detect the expected increase in bleeding associated with anticoagulation, he wrote.

Kirchhof also argued that the results should not necessarily translate into anticoagulation for every patient in this risk category. AF burden is increasingly recognized as an important modifier of stroke risk, he noted, and 72% of SINGLE-AF participants had paroxysmal AF while 31% had previously undergone AF ablation — factors that “probably contributed to the low observed incidence of stroke ,” he wrote. Taken together, he concluded, the findings may encourage anticoagulation in patients with AF and a single stroke risk factor, but “in patients with a low atrial fibrillation burden, no therapy may be preferred.”

During presentation of the data at ESC, Isabelle C. Van Gelder, MD, of University Medical Center Groningen in Netherlands, described SINGLE-AF as “a study we were waiting for,” given the limited and conflicting evidence in this intermediate-risk population. However, she cautioned that the relatively young, predominantly male, entirely South Korean study population may limit the generalizability of the findings to older patients, women, and European populations.

Van Gelder also highlighted the small number of events and lower-than-expected event rates, which limited the precision of the treatment-effect estimate. Nevertheless, she said the trial showed “a clear benefit” of anticoagulant therapy in this very-low-risk population.

The trial was funded by the South Korean Ministry of Health and Welfare, Samjin Pharmaceutical, and Hanmi Pharmaceutical. Joung reports receiving research funds from Samjin, Hanmi, Huinno, Medtronic, Boston Scientific, and Abbott Korea. See the editorial for a full list of Kirchhof’s relevant financial disclosures.