Bayer AG has secured United States Food and Drug Administration approval for Sunharbo, or asundexian, creating the first commercially approved Factor XIa inhibitor and introducing a new antithrombotic mechanism into secondary stroke prevention. The once-daily 50 mg tablet is approved in combination with antiplatelet therapy to reduce the risk of stroke in adults following a recent non-cardioembolic ischaemic stroke with a National Institutes of Health Stroke Scale score of 15 or lower, or a high-risk transient ischaemic attack.
The November 9 decision is supported by the global Phase 3 OCEANIC-STROKE trial, which randomised 12,327 patients and showed recurrent ischaemic stroke in 6.2% of patients receiving asundexian compared with 8.4% receiving placebo on top of antiplatelet therapy. That translated into a 26% relative reduction in risk, with a hazard ratio of 0.74, while International Society on Thrombosis and Haemostasis major bleeding occurred in 1.9% and 1.7% of patients, respectively.
Those figures matter because the long-standing ambition behind Factor XI and XIa inhibition has been to separate prevention of pathological clot formation from the normal haemostatic response needed to stop bleeding. Sunharbo does not eliminate bleeding risk, and gastrointestinal bleeding occurred more frequently with asundexian, but OCEANIC-STROKE delivered the first Phase 3 evidence strong enough to convert that scientific proposition into an approved medicine.
The approval is also a striking reversal in Bayer’s own development story. OCEANIC-AF, another major Phase 3 programme evaluating asundexian, was stopped early in 2023 because the drug was less effective than apixaban at preventing thromboembolic events in atrial fibrillation. Three years later, the same molecule has become the first approved member of its class because the secondary stroke setting produced a fundamentally different efficacy result.
Why did Factor XIa inhibition work after stroke when asundexian failed in atrial fibrillation?
The answer begins with the fact that arterial thrombosis is not one disease process.
Atrial fibrillation can produce blood stasis inside the atria, allowing clots to develop and subsequently travel to the brain. Established direct oral anticoagulants such as apixaban provide strong protection in this setting by inhibiting components of the coagulation cascade at levels sufficient to prevent cardioembolic clot formation.
OCEANIC-AF tested whether Factor XIa inhibition could provide enough anticoagulant effect to replace apixaban in patients with atrial fibrillation. It could not. An independent monitoring committee concluded that asundexian had inferior efficacy, forcing Bayer to terminate the study early.
OCEANIC-STROKE addressed a different problem. Patients had already experienced a non-cardioembolic ischaemic stroke or high-risk TIA, meaning their index event was not attributed to a clot originating from atrial fibrillation or another cardiac source. Standard management therefore centres heavily on antiplatelet treatment rather than full-dose oral anticoagulation.
Asundexian was added to that antiplatelet foundation rather than used as a replacement for an established anticoagulant. In this context, suppressing Factor XIa appears to have provided additional protection from pathological thrombus formation without producing a statistically significant increase in major bleeding.
The contrast makes indication selection central to the Factor XIa story. OCEANIC-AF showed that the mechanism should not be assumed to provide enough anticoagulation for every thrombotic disease. OCEANIC-STROKE showed that it can provide clinically useful incremental protection in a setting where the balance between recurrent ischaemia and bleeding has historically limited more intensive antithrombotic therapy.
How large was the absolute benefit behind the 26% relative stroke-risk reduction?
The relative reduction is clinically important, but the absolute numbers provide a clearer picture of treatment magnitude.
Ischaemic stroke occurred in 6.2% of patients receiving asundexian and 8.4% receiving placebo during the trial, producing an absolute difference of 2.2 percentage points. Published analyses have estimated that approximately 53 patients would need to be treated for one year to prevent one additional ischaemic stroke, falling to roughly 43 patients over two years as cumulative benefit increases.
Those numbers need to be interpreted in the context of stroke severity. Avoiding a recurrent ischaemic stroke can prevent major disability, loss of independence or death, so even a modest absolute event reduction can carry substantial clinical value in a high-risk population.
The trial also found a lower incidence of the composite of cardiovascular death, myocardial infarction or stroke with asundexian. The primary regulatory case, however, remains the reduction in ischaemic stroke itself rather than extrapolation to a broader cardiovascular benefit beyond the endpoints studied.
The study enrolled patients within 72 hours of their index event, placing treatment near the period when recurrent stroke risk is particularly high. Approximately 63% of participants initially received dual antiplatelet therapy, while the remainder received single antiplatelet therapy according to the planned treatment strategy.
That background therapy makes the result especially relevant because Sunharbo is not replacing aspirin, clopidogrel or another recommended antiplatelet medicine. It is adding another antithrombotic mechanism to existing care.
Does similar major bleeding mean Sunharbo avoids the bleeding trade-off of anticoagulation?
Not completely.
ISTH major bleeding occurred in 1.9% of asundexian-treated patients and 1.7% of placebo-treated patients. The hazard ratio was 1.10, with a 95% confidence interval of 0.85 to 1.44, and the difference was not statistically significant.
Major or clinically relevant non-major bleeding occurred in 5.5% of patients receiving asundexian and 5.0% receiving placebo. Any bleeding occurred in 12.5% and 12.3%, respectively. These findings support the central argument that adding asundexian did not produce the broad increase in major bleeding that has limited previous attempts to intensify long-term antithrombotic treatment after stroke.
The safety signal is not uniformly neutral, however. Gastrointestinal bleeding occurred more frequently with asundexian. A detailed analysis reported any gastrointestinal bleeding in 3.0% of patients receiving asundexian compared with 2.4% receiving placebo, while lower gastrointestinal bleeding occurred in 1.7% and 1.0%, respectively.
The FDA also highlighted greater major bleeding with Sunharbo among patients older than 75 years and during the first 90 days of treatment. Those findings mean clinicians will still need to consider age, gastrointestinal history, concurrent medicines and overall bleeding vulnerability rather than treating Factor XIa inhibition as risk-free anticoagulation.
The distinction is important commercially as well. Bayer can credibly position Sunharbo around the absence of a statistically significant overall increase in ISTH major bleeding. It cannot claim that the medicine eliminates bleeding or offers identical safety across every patient subgroup.
Why could Factor XIa be different from Factor Xa and thrombin as an antithrombotic target?
Factor XI occupies a different position in the coagulation cascade from targets used by established oral anticoagulants. Factor Xa and thrombin are central to clot formation in both pathological thrombosis and normal haemostasis, which helps explain why highly effective inhibition can increase clinically important bleeding.
Factor XI appears to play a larger role in amplification and propagation of pathological clotting while contributing less critically to the initial haemostatic plug formed after vascular injury. Individuals with inherited Factor XI deficiency often have substantially less spontaneous bleeding than patients with severe deficiencies of some other coagulation factors, helping create the biological rationale for developing drugs against XI or XIa.
Asundexian selectively inhibits activated Factor XI. OCEANIC-STROKE provides the first approved clinical demonstration that this strategy can reduce thrombotic events in a large population without a statistically significant increase in the study’s primary major-bleeding endpoint.
That should not be interpreted as validation of every Factor XI programme or every cardiovascular indication. The contrasting OCEANIC-AF outcome demonstrates that the amount of anticoagulant effect provided by this mechanism may be appropriate for some thrombotic settings but insufficient for others.
The class is therefore likely to develop according to disease-specific evidence rather than a simple assumption that Factor XIa inhibitors can replace existing anticoagulants across cardiovascular medicine.
Why is the Sunharbo label narrower than the phrase “stroke prevention” might suggest?
Sunharbo is approved for secondary prevention after a recent non-cardioembolic ischaemic stroke or high-risk TIA. It is not a general stroke-prevention medicine.
The approved stroke population must have an NIHSS score of 15 or lower, corresponding to the mild-to-moderate population studied in OCEANIC-STROKE. Patients receive the medicine in combination with antiplatelet therapy rather than as monotherapy.
The FDA explicitly states that Sunharbo is not approved for stroke prevention in patients with atrial fibrillation. That limitation is especially important because Bayer’s unsuccessful OCEANIC-AF study could otherwise be obscured by broad descriptions of asundexian as a new anticoagulant.
The label also contains practical restrictions. Concomitant treatment with combined P-glycoprotein and strong CYP3A4 inducers is not recommended because these medicines can reduce asundexian exposure and potentially reduce effectiveness. The medicine is not recommended in patients with moderate or severe hepatic impairment.
Sunharbo should also be withheld before elective surgery or invasive procedures and after thrombolytic treatment according to the approved prescribing requirements. These considerations mean implementation will still require education across stroke units, neurologists, primary-care teams and pharmacists even though the drug itself is a simple once-daily oral tablet.
Could a $748.50 monthly list price slow adoption after hospital discharge?
Bayer has set the U.S. list price at $748.50 for a 30-day supply, equivalent to roughly $9,000 annually before rebates, insurance negotiations or patient assistance.
That creates an immediate reimbursement question because antiplatelet medicines used after stroke are generally inexpensive and widely generic. Sunharbo therefore enters not as a replacement for a costly incumbent but as an additional branded medicine layered onto low-cost existing therapy.
Payers will need to assess the economic value of the absolute reduction in recurrent stroke against the drug’s acquisition cost. Recurrent strokes are expensive events that can generate hospitalisation, rehabilitation, long-term disability care and substantial indirect costs, giving Bayer a potentially strong health-economic argument if real-world effectiveness mirrors OCEANIC-STROKE.
Initial access could still be slowed by prior authorization, formulary reviews or prescriber uncertainty around patient selection. Treatment must also transition smoothly from the acute hospital setting into outpatient care because recurrent-stroke prevention is a long-term strategy rather than a short inpatient intervention.
Bayer expects Sunharbo shipments to begin in mid-to-late November, with availability to hospitals and pharmacies shortly afterwards. The speed of formulary uptake may therefore become one of the first measurable commercial indicators after launch.
Why is Sunharbo strategically important as Xarelto declines?
The timing is important for Bayer’s pharmaceutical business.
Third-quarter 2026 Xarelto sales fell 30.7% on a currency and portfolio-adjusted basis to €377 million as generic competition intensified, while sales for the first nine months declined 38.2% to €1.12 billion. Xarelto was once one of Bayer’s most important pharmaceutical growth products, making replacement revenue a strategic priority as exclusivity erodes.
Sunharbo gives Bayer another cardiovascular franchise, but it should not be viewed simply as a next-generation Xarelto. The drugs inhibit different coagulation targets, have different approved indications and currently address different patient populations.
Bayer’s pharmaceutical division is already being supported by rapid growth from Nubeqa and Kerendia. Third-quarter pharmaceutical sales rose 6.4% on a currency and portfolio-adjusted basis to €4.68 billion, helping the group argue that it is moving beyond its most difficult patent losses.
Asundexian could strengthen that transition if the first approved indication generates substantial adoption and additional indications eventually succeed. Bayer has described the medicine as a potential blockbuster, but that remains a commercial expectation rather than an established outcome.
The approval nevertheless removes one of the most important uncertainties facing the programme. After the OCEANIC-AF failure, investors could reasonably question whether Bayer would ever commercialise asundexian. It now has approval in both the United States and China, while applications are under review in Europe and Japan.
How much of a first-mover advantage does Bayer have over milvexian?
Bayer has won a meaningful regulatory lead, but the Factor XI field remains competitive.
Bristol Myers Squibb and Johnson & Johnson are developing the oral Factor XIa inhibitor milvexian in a broad Phase 3 programme. The LIBREXIA-STROKE trial has completed enrolment and is evaluating milvexian against placebo, on top of standard therapy, after acute ischaemic stroke or high-risk TIA.
No Phase 3 stroke result from LIBREXIA-STROKE had been publicly reported as of November 11. The programme therefore remains capable of creating future competition in substantially the same secondary-stroke market if efficacy and safety results support regulatory filings.
Milvexian has already produced a reminder that class biology does not guarantee efficacy. The LIBREXIA ACS study in acute coronary syndrome was stopped for futility and later showed no reduction in major cardiovascular events despite pharmacological evidence that the drug inhibited Factor XIa.
Novartis is also developing abelacimab, an antibody targeting Factor XI and XIa, although its pivotal strategy currently includes atrial fibrillation patients considered unsuitable for standard oral anticoagulation rather than the same approved secondary-stroke population as Sunharbo.
These programmes reinforce the lesson from Bayer’s own trials: Factor XI inhibition is not one uniform market. Each disease setting must establish whether the degree of anticoagulation created by the mechanism produces enough efficacy while preserving the hoped-for bleeding advantage.
What does Bayer’s share-price reaction suggest about investor expectations?
Bayer shares rose by as much as approximately 2% on November 10 and reached a 2026 high around €57.78 following the U.S. approval. Several analysts responded positively to the label and the regulatory outcome, although Bayer’s share price remains influenced by far more than its pharmaceutical pipeline, including crop-science performance, debt and longstanding litigation exposure.
The relatively measured reaction is understandable because approval was increasingly anticipated after the positive Phase 3 result, Priority Review and earlier Chinese authorization. The more important valuation question is now commercial execution.
Sunharbo needs to move from a successful clinical programme into guidelines, payer coverage and routine post-stroke prescribing. A new antithrombotic class can face inertia because physicians have decades of experience with established antiplatelet and anticoagulant strategies and may initially adopt an additional agent selectively.
Large OCEANIC-STROKE enrolment, publication in the New England Journal of Medicine and the straightforward once-daily dosing regimen give Bayer a substantial evidence and education base. Price and bleeding management in older patients may create the most immediate friction.
What will determine whether Sunharbo establishes Factor XIa as a major drug class?
The first answer will come from real-world secondary stroke prevention.
OCEANIC-STROKE demonstrated a statistically robust 26% relative reduction in recurrent ischaemic stroke and a 2.2-percentage-point absolute difference, with no statistically significant increase in the overall primary major-bleeding endpoint. That is enough to establish efficacy and support approval, but commercial success will depend on whether hospitals, neurologists, payers and guideline committees consider that benefit worth adding another long-term medication.
Longer-term safety will also matter. Median follow-up in OCEANIC-STROKE was approximately 19 months, whereas patients may remain at recurrent stroke risk for years. Gastrointestinal bleeding and the age-related bleeding signal will need monitoring as exposure expands far beyond the controlled trial population.
The next major scientific test comes from competitors, particularly LIBREXIA-STROKE. If another Factor XIa inhibitor reproduces the combination of lower recurrent stroke risk and limited major-bleeding impact, the mechanism could evolve from a Bayer breakthrough into a genuine therapeutic class. Failure would instead strengthen the possibility that asundexian’s success depends on characteristics specific to the molecule, dose or trial population.
Sunharbo’s approval is therefore more consequential than an ordinary first-in-class launch. Asundexian went from a high-profile Phase 3 failure in atrial fibrillation to the first regulatory validation of Factor XIa inhibition because Bayer continued testing the molecule in a disease setting with a different antithrombotic requirement.
That history may ultimately be the most important lesson from OCEANIC-STROKE. The trial did not prove that Factor XIa inhibition is a safer substitute for every anticoagulant. It showed that adding precisely the right amount of coagulation inhibition to antiplatelet therapy after a non-cardioembolic stroke can prevent meaningful recurrent events without recreating the major-bleeding penalty that has repeatedly undermined more aggressive long-term antithrombotic strategies.
