Asundexian Outcomes Across Noncardioembolic Stroke Subtypes: Ashkan Shoamanesh, MD, FRCPC

New data from the global phase 3 OCEANIC-STROKE trial are sharpening how clinicians think about antithrombotic therapy across ischemic stroke subtypes. In this video from the International Stroke Conference 2026, Ashkan Shoamanesh, MD, FRCPC, associate professor in the Division of Neurology at McMaster University in Hamilton, Canada, walks through a prespecified secondary analysis evaluating whether the efficacy and safety of the factor XIa inhibitor asundexian vary by underlying stroke etiology.1

Drawing on outcomes from more than 12,000 patients with noncardioembolic ischemic stroke or high-risk transient ischemic attack (TIA) treated on top of standard antiplatelet therapy, Shoamanesh explains how asundexian reduced recurrent ischemic stroke without increasing intracerebral hemorrhage or major bleeding and why the benefit appeared consistent across large artery atherosclerosis, lacunar stroke, and strokes of undetermined etiology. The discussion highlights what these findings mean for individualized decision-making—and why stroke subtype may not need to be a barrier when considering XIa inhibition for secondary prevention.

This transcript was lightly edited; captions were auto-generated.

Transcript

How do the differences observed across ischemic stroke subtypes inform our understanding of underlying thrombotic mechanisms?

So what was quite interesting in the literature to date is that there does seem to be differences in response to antithrombotic treatments according to ischemic stroke subtype. And the main differentiator here is cardioembolic strokes vs noncardioembolic strokes, where patients with cardioembolic stroke, particularly in the context of atrial fibrillation, benefit from anticoagulation, whereas that has not really translated to patients with noncardioembolic stroke, where the standard therapy at present remains antiplatelet therapy, either as dual antiplatelet therapy for a short time period, somewhere between 21 days and 90 days depending on the stroke subtype, and then single antiplatelet therapy thereafter.

How should clinicians interpret these results when evaluating whether a therapy like asundexian may be appropriate for individual patients?

So, OCEANIC-STROKE was a large phase 3 randomized trial that compared asundexian 50 mg daily vs placebo. In 12,327 patients with either noncardioembolic ischemic stroke or high-risk TIA enrolled from over 700 hospitals in 37 countries. As I mentioned, these patients were randomized to either the active treatment, which was asundexian 50 mg, or placebo, and randomization was also stratified according to whether there was planned treatment with dual antiplatelet therapy or single antiplatelet therapy. The primary results of that study were presented today with my colleague Mike Sharma, where we showed that there was an overall 26% reduction in the hazard of the primary efficacy end point that was time to first occurrence of ischemic stroke, and impressively, without any excess in ICH [intracerebral hemorrhage], major bleeding, or actually any bleeding outcome that was prespecified in our analyses.

Now, the focus of this secondary analysis that we'll be presenting here as well is, was there any treatment heterogeneity according to ischemic stroke subtype on the basis of some of the changes that we've seen in the literature with antiplatelet therapy, or differences rather in terms of response to treatment that we've seen in the literature,2 as well as our phase 2 PACIFIC-STROKE study, where there was some suggestion that those that entered the trial with large artery atherosclerosis had a greater treatment effect.

However, across very large subgroups of patients with distinct ischemic stroke subtypes and the umbrella of noncardioembolic ischemic stroke, that was about 5000 patients with large artery atherosclerosis, 3500 patients with stroke of undetermined etiology, and about 2600 patients with lacunar stroke, we found absolutely no heterogeneity in the treatment effect, and all patients who were eligible for this study that entered the study with an ischemic stroke—and in total, that was about 11,100 patients—had the consistent benefit in time to first occurrence of ischemic stroke with asundexian vs placebo, with the cause-specific hazard ratio ranging anywhere from about 0.61 to 0.82.

But overall, the P value for the treatment interaction was negative, meaning that all patients benefited consistently. And importantly, this was again without any excessive major bleeding, including hemorrhagic stroke, which was actually numerically less with asundexian across the board compared to placebo, including in patients with lacunar stroke or small vessel occlusive disease, who are at highest risk of having intracellular hemorrhage as an outcome.

Lastly, on the heels are 4 randomized trials to date that have unfortunately been unsuccessful in expanding antithrombotic therapy from just antiplatelet monotherapy for long-term stroke prevention and patients with embolic stroke of undetermined source, which is our subgroup of patients with stroke of undetermined etiology. We perform sensitivity analysis in the patients that entered the study with ESUS, or embolic strokes of undetermined source. And here, we actually saw a robust 47% risk reduction in time to first occurrence of ischemic stroke, meaning that this new paradigm of factor XI inhibition on top of background therapy is also generalizable to those with ESUS. So in total, the trial has really enrolled a large global generalizable population to the spectrum of noncardiomyotic schematic stroke that we generally treat, and there's consistency overall, without any suggestion that one etiology should be favored over another for individualized treatments, and that if they were eligible for the study, they benefited.

What additional analyses or data would help clarify whether certain ischemic stroke subtypes truly derive differential benefit from asundexian?

So we also looked at some other parameters that can speak to underlying etiology, and in particular, as it pertained to atherosclerosis. So what I've spoken to thus far is about the categorization of large artery atherosclerosis, but on vascular imaging, there are people who have atherosclerosis, very minor plaques, or plaques that don't meet the threshold of 50% and proximal to the infarct to be categorized as large atherosclerosis [risk], and we looked at whether patients had more than 50% stenosis of any blood vessel within the arch, neck, or brain; if they had less than 50%, no difference. Both groups benefited. We looked at if they had any atherosclerosis vs no atherosclerosis at all—so not even a speck of atherosclerosis in their arch, neck, or brain. It didn’t matter whether it was present or not; both groups benefited from it. So overall, we're actually quite pleased to see that the results are really generalizable to the spectrum of noncardioembolic ischemic stroke that was eligible for study entry.

Do these results change how you think about combining asundexian with other antiplatelet or anticoagulant strategies across stroke subtypes?

Well, already the design of OCEANIC-STROKE was one where we're adding asundexian on top of standard of care antiplatelet therapy. Two-thirds of patients received DAPT [dual antiplatelet therapy] in the study; one-third received aspirin. So already this is an add-on to antiplatelet therapy, and we've shown that this is completely safe compared to placebo, right? You don't have an active comparator here. Compared to placebo, we're not even seeing an excess in minor bleeding, which is really impressive, in my view, and really historic if you think about the development of antithrombotic therapy over time. But on the other hand as well, there’s the fact that we're seeing absolutely no excess in bleeding. And indirect comparisons would suggest that this may be even safer than aspirin. It does open up the opportunity of adding it to perhaps existing anticoagulants for other indications like cardioembolic stroke.

So I think some of the robust safety that's been now well validated, suspected, hypothesized, and suggested in phase 2 studies, including in the phase 3 OCEANTIC-AF trial where asundexian had reduced bleedings by large proportions relative to a safe anticoagulant, apixaban, is really validated now that even against placebo, you're not seeing excess bleeding events, and it really allows us to think a bit more creatively about how we can combine a factor XI inhibitor with a spectrum of antithrombotics, including anticoagulants for other indications.

From your perspective, what is the most important takeaway clinicians should remember about asundexian outcomes by stroke subtype?

I think the number 1 take-home point is you don't need to think about it. If they were eligible for this study, they benefited from treatment, irrespective of underlying stroke etiology.

References

1. Shoamanesh A, Joundi RA, Dong Q, et al. Qualifying ischemic stroke subtypes and response to asundexian: prespecified secondary analysis of the OCEANIC-STROKE trial. Presented at: International Stroke Conference 2026. February 3-6, 2025; New Orleans, LA. Abstract LB020.
2. Sharma M, Joundi RA, Dong Q, et al. Factor XIa inhibition with Asundexian in acute non-cardioembolic stroke or high-risk transient ischemic attack: primary results of the OCEANIC-STROKE trial. Presented at: International Stroke Conference 2026. February 3-6, 2025; New Orleans, LA. Abstract LB009.