← Back
NeuroTrials.ai
Neurology Clinical Trial Database

ATIS-NVAF

Optimal Antithrombotics for Ischemic Stroke and Concurrent Atrial Fibrillation and Atherosclerosis: A Randomized Clinical Trial (Antithrombotic Therapy in Ischemic Stroke Patients with Non-Valvular Atrial Fibrillation and Atherothrombosis)

Share Share Copied! Compare

Year of Publication: 2025

Authors: Shuhei Okazaki, Kanta Tanaka, Yukako Yazawa, ..., for the ATIS-NVAF Trial Investigators

Journal: JAMA Neurology

Citation: JAMA Neurol. doi:10.1001/jamaneurol.2025.3662

Link: https://doi.org/10.1001/jamaneurol.2025.3662

Clinical Question

Does adding an antiplatelet agent to anticoagulant therapy provide net clinical benefit over anticoagulant monotherapy in patients with ischemic stroke or TIA and concurrent nonvalvular atrial fibrillation and atherosclerotic cardiovascular disease?

Bottom Line

In patients with ischemic stroke or TIA and concurrent nonvalvular atrial fibrillation and atherosclerotic cardiovascular disease, adding an antiplatelet agent to anticoagulant therapy provided no net clinical benefit over anticoagulant monotherapy and was associated with significantly higher bleeding risk. Anticoagulant monotherapy appears to be a safer option for secondary stroke prevention in this high-risk population.

Major Points

  • First randomized trial evaluating secondary prevention in patients with concurrent NVAF and atherosclerotic disease
  • 316 patients randomized to combination therapy (n=159) or monotherapy (n=157) at 41 Japanese sites
  • Trial terminated early for futility after interim analysis on July 18, 2023
  • Primary outcome (composite of ischemic cardiovascular events and major bleeding): 17.8% combination vs 19.6% monotherapy (HR 0.91, 95% CI 0.53-1.55, P=0.64)
  • Ischemic cardiovascular events: 11.1% combination vs 14.2% monotherapy (HR 0.76, 95% CI 0.39-1.48, P=0.41)
  • Major and clinically relevant nonmajor bleeding: 19.5% combination vs 8.6% monotherapy (HR 2.42, 95% CI 1.23-4.76, P=0.008)
  • Most common atherosclerotic disease was atherothrombotic or lacunar stroke (55%), followed by intracranial artery stenosis (33%)
  • 94% of patients received direct oral anticoagulants; only 6% received warfarin
  • In combination group: aspirin 52%, clopidogrel 31%, cilostazol 17%
  • Gastrointestinal bleeding was most common hemorrhagic event (6% combination vs 3% monotherapy)
  • Results consistent across sensitivity analyses and subgroups

Design

Study Type: Multicenter, open-label, parallel-group, phase IV randomized clinical trial

Randomization: 1

Blinding: Open-label for treatment allocation; blinded independent central review committee for clinical event adjudication. No placebo used due to funding constraints.

Enrollment Period: November 2016 to March 2025 (randomization April 6, 2017 to April 30, 2022)

Follow-up Duration: 2 years (median 707 days, IQR 588-735)

Centers: 41

Countries: Japan

Sample Size: 316

Analysis: Intention-to-treat analysis using Kaplan-Meier method for survival curves, log-rank test for comparison (1-sided for primary outcome, 2-sided for secondary outcomes), proportional hazard model for hazard ratios with 95% CI. Per-protocol analysis performed excluding poor adherence, incomplete registration, and protocol deviations. Subgroup analyses for homogeneity of treatment effects. Pocock-Simon minimization method for randomization with adjustments for sex, history of ischemic heart disease, and history of acute ischemic stroke. No imputation for missing data as all intention-to-treat patients had time-to-event data with censoring. Statistical analyses conducted using SAS version 9.4 and R version 4.2.0.

Inclusion Criteria

  • Age ≥20 years
  • Ischemic stroke or transient ischemic attack (TIA) between 8 and 360 days prior to randomization
  • Chronic or paroxysmal nonvalvular atrial fibrillation
  • Started or continued oral anticoagulant therapy
  • At least 1 of the following atherosclerotic conditions: symptomatic or asymptomatic carotid artery stenosis ≥50% diameter
  • Symptomatic or asymptomatic intracranial artery stenosis ≥50% diameter
  • History of carotid artery stenting or carotid endarterectomy
  • History of symptomatic atherothrombotic or lacunar stroke
  • History of ischemic heart disease
  • History of peripheral artery disease

Exclusion Criteria

  • Myocardial infarction or acute coronary syndrome within past 12 months
  • Percutaneous coronary intervention with drug-eluting stents within past 12 months
  • Percutaneous coronary intervention with bare-metal stents within past 3 months
  • Carotid, intracranial, or lower extremity stent placement within past 3 months
  • Symptomatic intracranial hemorrhage within past 6 months
  • Gastrointestinal bleeding within past 6 months
  • Hemorrhagic diathesis or blood coagulation disorders
  • Active cancer
  • Severe disability (modified Rankin Scale score of 5)
  • Anticipated discontinuation of anticoagulants or antiplatelets for more than 4 weeks
  • Planned revascularization procedures

Arms

FieldCombination TherapyControl
InterventionOral anticoagulant (warfarin, dabigatran, rivaroxaban, apixaban, or edoxaban) plus antiplatelet agent (aspirin, clopidogrel, or cilostazol). Specific drug selection and dosing at physician discretion according to predefined protocol and Japanese standards. Treatment initiated within 3 days of randomization and continued for 2 years. Changes in agents permitted if clinically indicated but limited to predefined drugs. Temporary interruptions <4 weeks allowed for procedures.Oral anticoagulant alone (warfarin, dabigatran, rivaroxaban, apixaban, or edoxaban). Specific drug selection and dosing at physician discretion according to predefined protocol and Japanese standards. Treatment initiated within 3 days of randomization and continued for 2 years. Changes in anticoagulant permitted if clinically indicated but limited to predefined drugs. Temporary interruptions <4 weeks allowed for procedures.
Duration2 years2 years

Outcomes

OutcomeTypeControlInterventionHR / OR / RRP-value
Composite of ischemic cardiovascular events (cardiovascular death, ischemic stroke, myocardial infarction, systemic embolism, and ischemic events requiring urgent revascularization) and major bleeding (ISTH criteria) occurring within 2 years of randomizationPrimary28 of 157 patients, cumulative incidence 19.6% (95% CI 12.7-26.0%)25 of 159 patients, cumulative incidence 17.8% (95% CI 11.0-24.0%)0.910.64
Ischemic cardiovascular eventsSecondary20 of 157 (14.2%, 95% CI 8.1-19.9%)15 of 159 (11.1%, 95% CI 5.4-16.4%)0.760.41
All ischemic cardiovascular events (including TIA, unstable angina, symptomatic PAD progression)Secondary20 of 157 (14.2%, 95% CI 8.1-19.9%)16 of 159 (11.8%, 95% CI 6.0-17.3%)0.810.53
Ischemic strokeSecondary18 of 157 (13.0%, 95% CI 7.1-18.5%)12 of 159 (9.2%, 95% CI 3.9-14.2%)0.670.28
All-cause mortalitySecondary10 of 157 (7.3%, 95% CI 2.8-11.7%)8 of 159 (5.6%, 95% CI 1.7-9.3%)0.810.65
Myocardial infarction or cardiovascular deathSecondary2 of 157 (1.8%, 95% CI 0.0-4.3%)2 of 159 (1.3%, 95% CI 0.0-3.1%)0.980.99
ISTH major bleedingAdverse8 of 156 (5.6%, 95% CI 1.8-9.4%)13 of 159 (9.4%, 95% CI 4.3-14.2%)1.630.27
Major and clinically relevant nonmajor bleedingAdverse12 of 156 (8.6%, 95% CI 3.8-13.2%)28 of 159 (19.5%, 95% CI 12.6-25.9%)2.420.008
Intracranial hemorrhageAdverse5 of 156 (3.8%, 95% CI 0.4-6.9%)7 of 159 (5.2%, 95% CI 1.2-8.9%)1.390.57
Gastrointestinal bleedingAdverse5 of 156 (3%)10 of 159 (6%)
Nasal bleedingAdverse0 of 156 (0%)6 of 159 (4%)
HematuriaAdverse0 of 156 (0%)3 of 159 (2%)

Subgroup Analysis

Forest plot analysis showed no significant differences in primary outcome across predefined subgroups including age (<80 vs ≥80 years), sex, history of ischemic stroke and major bleeding, carotid or intracranial stenosis, hypertension, diabetes, dyslipidemia, congestive heart failure, CHADS2 score, anticoagulant type (warfarin vs DOAC), body weight (<60 vs ≥60 kg), and creatinine clearance categories. Sensitivity analyses excluding asymptomatic stenosis or lacunar stroke yielded consistent results with main findings.

Criticisms

  • Trial terminated early for futility with only 316 of planned 400 patients enrolled, limiting statistical power
  • Open-label design may have introduced bias in patient adherence, outcome reporting, and clinical decision-making
  • Heterogeneity in choice of specific anticoagulants and antiplatelet agents and their dosages left to physician discretion, creating variability in interventions
  • No CYP2C19 genotyping performed to screen for clopidogrel resistance
  • Stroke subtypes did not strictly follow TOAST criteria due to coexisting atrial fibrillation, potentially introducing classification bias
  • No data collected on whether AF was diagnosed before index stroke or whether patients had stroke despite therapeutic anticoagulation
  • Patients at particularly high bleeding risk may have been excluded due to prior bleeding events during acute stroke phase
  • Only Japanese patients enrolled, limiting generalizability to other populations with different genetic, environmental, or healthcare factors
  • No placebo used due to funding constraints
  • Different enrollment window (8-360 days) than initially planned (15-180 days), expanded during trial based on similar event rates
  • Small sample size of subgroups limits definitive conclusions about treatment effects in specific populations
  • Median follow-up of 707 days slightly less than planned 2 years
  • Higher than expected recurrence rate may reflect enrollment of patients within first year after stroke when risk is highest

Funding

Partially supported by the Japan Thrombosis Investigator-Initiated Research Program, funded by Bristol Myers Squibb and Pfizer

Based on: ATIS-NVAF (JAMA Neurology, 2025)

Authors: Shuhei Okazaki, Kanta Tanaka, Yukako Yazawa, ..., for the ATIS-NVAF Trial Investigators

Citation: JAMA Neurol. doi:10.1001/jamaneurol.2025.3662

Content summarized and formatted by NeuroTrials.ai.