Bayer’s Sunharbo is the first factor XIa inhibitor approved anywhere. The US label has no boxed warning. The trial behind it recorded 134 fewer strokes and 10 more major bleeds. At list price, one prevented stroke takes about $483,000 of drug in the first year.
The asundexian FDA approval was announced on 9 November 2026. Sunharbo (asundexian) is a 50 mg tablet taken once a day with antiplatelet therapy to lower the risk of stroke. It is for adults who have just had a non-cardioembolic ischaemic stroke or a high-risk transient ischaemic attack (TIA), often called a mini-stroke. Bayer’s list price is $748.50 for 30 days.
No other factor XIa inhibitor is approved in any country, and the label is unusually bare for a blood thinner. Under contraindications it says “None”. It has no boxed warning and no warnings and precautions section: the numbering runs from section 4 to section 6. Atrial fibrillation, the use that built Xarelto and Eliquis, is absent, because asundexian lost its trial there.
The label, the trial papers, Bayer’s accounts and the trial registry are all public. Read together they give three things the announcement does not. One is the list-price cost of each stroke prevented. Another is where bleeding rose. A third is the length of Bayer’s lead over the next drug.
01What the asundexian FDA approval covers
Sunharbo is indicated, with antiplatelet therapy, to reduce the risk of stroke in adults with a recent non-cardioembolic ischaemic stroke or a high-risk TIA. A qualifying stroke scores 15 or lower on the NIH Stroke Scale, which takes in mild and moderate strokes. In the US label the word is spelled ischemic. Doctors call this use secondary stroke prevention: stopping a second stroke after the first. The prescribing information lists no contraindications. Bayer announced the approval from Whippany, New Jersey, on 9 November.
Bayer expected shipments to begin in mid-to-late November, with hospitals and pharmacies able to order soon after.
Sunharbo dosage: one 50 mg tablet a day
The label gives one dose: 50 mg once daily, with or without food. A missed tablet can be taken later the same day. Once the day has passed, the label says to skip it and take the next one at the usual time. Kidney function, dialysis included, made no clinically significant difference to drug levels. Use is not recommended in moderate or severe liver impairment.
Two interactions are named. Drugs that strongly induce both P-glycoprotein and CYP3A4 should be avoided, and carbamazepine cut asundexian exposure by about 44%. Sunharbo also raises digoxin levels.
Where the label is wider than the trial
On two points the indication is wider than the trial. OCEANIC-STROKE randomised patients within 72 hours of symptom onset, and it took only those with a non-lacunar infarct, a history of atherosclerosis or plaque on imaging. No time window appears in the indication statement, and no such condition. Nor does the dosing section give a start time, so the 72-hour window appears only in the label’s account of the trial.
02OCEANIC-STROKE results: 134 fewer strokes, 10 more major bleeds
OCEANIC-STROKE is the trial the asundexian FDA approval rests on. It assigned 6,162 patients to asundexian and 6,165 to placebo, all on antiplatelet drugs, and followed them for a median of 567 days. Their mean age was 68, and 34% were women. The results went online at the New England Journal of Medicine on 15 April 2026.
Ischaemic stroke, the primary endpoint, occurred in 384 patients on asundexian and 518 on placebo. Major bleeding by the ISTH definition, the primary safety measure, occurred in 117 and 107. The paper reports the rates. Getting to the differences takes one more step.
518 − 384 = 134 fewer strokes
Major bleeding 117 / 6,124 vs 107 / 6,130 = 1.9% vs 1.7% (reported; HR 1.10, 0.85 to 1.44)
117 − 107 = 10 more major bleeds
Crude risk difference 8.40% − 6.23% = 2.17 points = 1 stroke avoided per 46 treated
Chance could explain the bleeding difference: its confidence interval runs from 15% fewer to 44% more. For strokes the interval runs from 16% to 35% fewer. At one year the absolute reduction was 1.9 points, which is 53 patients treated to prevent one stroke, TCTMD reported when the paper came out.
The wider endpoints moved the same way. Cardiovascular death, heart attack or stroke occurred in 568 patients against 685, a hazard ratio of 0.83. All strokes together ran at 6.6% against 8.8%. Subtract the ischaemic rates and about 0.4 points are left in each arm for every other kind of stroke (computed).
03Sunharbo side effects: where the bleeding sits in the label
The label names diarrhoea as the most common adverse reaction: 3.4% of patients on Sunharbo against 2.8% on placebo. Liver tests moved a little. Transaminases above three times the upper limit of normal, without a rise in bilirubin, were seen in 2.0% against 1.6%. Serious liver injury with jaundice was reported in fewer than 0.02% of treated patients. Overall and serious adverse events were level between the arms, at 69.3% against 70.1% and 19.2% against 19.5%.
Bleeding is the question any blood thinner has to answer, and here the headline and the fine print differ. Across the whole trial, major bleeding ran at 1.6 events per 100 patient-years on Sunharbo and 1.5 on placebo. The label then breaks that down.
ISTH major bleeding per 100 patient-years: Sunharbo vs placebo
For the first 90 days the label prints a hazard ratio: 1.46, with an interval of 0.99 to 2.16 that just takes in no difference. No intervals are given for the subgroup rows, so the ratios beside them are ours and should be read as rough. They still mark where a prescriber will look: the first three months, dual antiplatelet therapy, patients over 75, and those whose index event was a TIA. Most of the gastrointestinal bleeding, 74%, was minor or clinically relevant non-major, and the commonest site was rectal.
Xarelto and Eliquis both carry boxed warnings, for clots after early stopping and for bleeding around the spine. Sunharbo carries none. The label has no data in pregnancy, and Bayer advises against breastfeeding during treatment and for five days after the last dose.
04Why asundexian is not approved for atrial fibrillation
Bayer tested the same 50 mg dose against apixaban in 14,810 patients with atrial fibrillation. The trial, OCEANIC-AF, was stopped early on the advice of its data monitoring committee. Stroke or systemic embolism occurred in 98 patients on asundexian and 26 on apixaban, a hazard ratio of 3.79 (95% CI 2.46 to 5.83). Major bleeding went the other way: 17 patients against 53, a hazard ratio of 0.32. The paper is in the New England Journal of Medicine.
One molecule at one dose gave a hazard ratio of 0.74 in one setting and 3.79 in the other. The comparators explain part of it: placebo in stroke, a proven anticoagulant in atrial fibrillation. Whether a larger dose would have held up is not settled. Milvexian’s atrial fibrillation trial uses 100 mg twice daily, and it will be the test.
The cost to Bayer is on record. In January 2023 it projected peak sales of more than €5 billion for asundexian across thrombosis and stroke prevention. It now speaks of blockbuster potential, meaning more than €1 billion. Floor to floor, that is 80% lower (computed). We saw the same thing with ravulizumab: a mechanism that works in one disease does not always travel to the next.
05How asundexian works: factor XIa inhibition
Asundexian is an oral small molecule that blocks activated factor XI, written factor XIa or FXIa. Factor XIa sits in the contact pathway of coagulation. Once a clot has started, it feeds a loop that makes more thrombin and lets the clot grow.
Sealing a cut depends far less on that loop. It runs mainly through tissue factor and factor Xa, the enzyme that rivaroxaban and apixaban block. So the theory behind the class is that a factor XIa inhibitor should slow clot growth in an artery and leave ordinary wound sealing closer to normal.
Human biology pointed the same way before any trial did. Low factor XI levels are associated with a lower risk of ischaemic stroke, the trial’s authors note. OCEANIC-STROKE is the first phase 3 trial of an oral factor XIa inhibitor to show what the theory predicts: fewer strokes, with no significant rise in major bleeding. The label puts the drug’s half-life at about 24 hours, which fits a once-daily tablet.
06Asundexian vs milvexian: a 357-day head start
Milvexian, from Bristol Myers Squibb and Johnson & Johnson, is the other oral factor XIa inhibitor in phase 3. No trial compares the two drugs. The registry shows how close the race was at the start and how far apart it finished.
The two stroke trials began 20 days apart and reached primary completion 357 days apart (computed from registry dates). Bayer made its lead in the running of the trial.
When the rival result arrives is less clear than that date suggests. The registry still lists the milvexian stroke trial as active, not completed. Bristol Myers Squibb still expected the stroke findings by the end of 2026 when it reported second-quarter results, BioSpace wrote on 30 July. The atrial fibrillation readout moved to the first quarter of 2027, Fierce Pharma reported. A month later Philippe Gabriel Steg, principal investigator of the coronary trial, told HCPLive that both remaining trials would report in 2027.
The class record so far is one win in three. Three phase 3 trials of oral factor XIa inhibitors have reported, covering 41,351 enrolled patients on the registry’s counts. Two more, with 32,807 patients, are due. It is the kind of class-wide count we ran on OX40 in atopic dermatitis, with a better ending for one drug.
A third candidate takes another route. Novartis has abelacimab, an antibody against factor XI, in late-stage trials. Its 2,500-patient LILAC-TIMI 76 study lists primary completion on 30 December 2027.
07Sunharbo price: $748.50 a month, and the cost of one prevented stroke
The list price of Sunharbo is $748.50 for a 30-day supply, a Bayer spokesperson told BioPharma Dive. The label lists a bottle of 30 tablets, so that is one bottle. Bayer says the list price is not what most health plans or patients will pay. Eligible patients can get the drug free through its patient assistance foundation. List price is still where payers start.
748.50 / 30 = $24.95 a day
24.95 × 365 = $9,107 a year
Patients treated for one year to prevent one stroke 53 (trial estimate)
That figure is crude. It ignores rebates, patients who stop, and everything a stroke costs afterwards. A formulary committee can still check it in a minute.
Payers also have an anchor. Medicare’s negotiated 2026 prices for the two leading anticoagulants are far lower, though neither drug is approved for this use.
Cheryl Bushnell of Wake Forest University School of Medicine told TCTMD that cost is “probably going to be the main limiting factor”. The same test faced the first oral PCSK9 inhibitor, which we priced out in LIPFENDRA against Repatha.
08What Sunharbo means for Bayer after Xarelto
The asundexian FDA approval lands as Bayer’s old anticoagulant shrinks. Xarelto sales were €1,115 million in the first nine months of 2026, down from €1,823 million a year earlier, as patents expired. Both figures are from Bayer’s third-quarter report of 3 November.
The US arrangement is different this time. Johnson & Johnson sells Xarelto in the United States under licence. Sunharbo is Bayer’s own product there.
Bayer’s release counts about 795,000 strokes a year in the United States, 87% of them ischaemic. Jefferies reckons Sunharbo suits about 40% of stroke patients. On Bayer’s count that is 318,000 people a year (computed).
The revenue arithmetic is less generous. At list price, $1 billion of gross sales takes about 110,000 patient-years of treatment (computed). Net prices will be lower, so the real number is higher. Bayer’s own figure of more than €1 billion sits below both analyst estimates quoted here.
09Sunharbo in India: 12 trial sites, Rule 101 and no filing yet
Stroke is common in India. A systematic review of nine population studies put the crude incidence at 108 to 172 per 100,000 people a year. One-month case fatality ran from 18% to 42% (Jones et al., 2022, via PubMed).
Indian hospitals took part. The registry record for OCEANIC-STROKE lists 741 sites in 37 countries. Twelve are Indian hospitals in 10 cities, AIIMS New Delhi and CMC Vellore among them. That is 1.6% of sites (computed). China has 59 and the United States 108. P. N. Sylaja of the Sree Chitra Tirunal Institute in Thiruvananthapuram is an author of the New England Journal paper.
India’s clearance trailed the global start. The neurology committee that advises India’s drug regulator cleared Bayer’s protocol 20604, the sponsor’s code for this trial, with one condition on its entry criteria. It did so at its meeting of 12 and 13 December 2023, Medical Dialogues reported. By then the trial had been open for 320 days, since 26 January 2023 (computed).
Milvexian’s sponsors used India far more. All five phase 3 trials of oral factor XIa inhibitors list Indian sites on the registry, 127 of 4,745 in total. Their stroke trial has 54 of them, 4.5 times Bayer’s 12 (computed). Neither drug is approved in India.
Indian sites in the five phase 3 trials
A route to approval exists. In August 2024 India’s regulator named six jurisdictions whose approvals let it consider waiving a local trial, and the United States is one of them. China, which approved Sunharbo 14 days earlier, is not. So the asundexian FDA approval, and not the Chinese one, is what counts under Rule 101. The order limits the waiver to five categories of drug, one of them a significant therapeutic advance over standard care, and it leaves the decision with the regulator.
We found no Indian filing on record. Indian sites have now helped test a first-in-class drug that India cannot yet prescribe, the pattern we described in India’s next CRO boom.
10Witfire Risk Score: 4.3 / 10
This score covers Bayer’s Sunharbo franchise after the asundexian FDA approval.
11What to watch
- By 31 December 2026. The LIBREXIA-STROKE headline result (BMY, JNJ), on the timing Bristol Myers Squibb gave in July. A win makes this a two-drug class in the same indication. A miss leaves Bayer alone and puts the 25 mg dose in question.
- First quarter of 2027. The LIBREXIA-AF readout, after its move from 2026. It decides whether the class reaches atrial fibrillation at all.
- 24 February 2027. Bayer’s full-year results (BAYN in Frankfurt, BAYRY in New York), the first to carry US Sunharbo sales. Shipments were due to begin in mid-to-late November, so the quarter will hold about six weeks at most.
- Europe and Japan. Both regulators are reviewing. Either approval would also count under India’s Rule 101.
- Guidelines and reimbursement. Ashkan Shoamanesh, co-principal investigator of the trial, named guideline inclusion, physician education and reimbursement as the next steps. Until they arrive, the list price does the talking.
On paper the asundexian FDA approval is clean: a first-in-class label with no boxed warning, built on 12,327 patients and 134 fewer strokes.
The risk sits in the calendar and the price. Bayer started 20 days ahead of milvexian and finished 357 days ahead. It has that lead to spend while a rival of the same class waits to report. At $748.50 a month, about $483,000 of drug prevents one stroke in the first year at list.
Bayer has already told the market what the atrial fibrillation failure cost: four-fifths of the forecast, floor to floor. What remains is a real drug for a defined group of stroke patients, and a short window in which to make it the standard before there are two.
FAQSunharbo (asundexian): questions and answers
What is Sunharbo (asundexian) approved for?
The asundexian FDA approval, announced on 9 November 2026, covers one use. Sunharbo is taken with antiplatelet therapy to reduce the risk of stroke in adults who have had a recent non-cardioembolic ischaemic stroke or a high-risk transient ischaemic attack. A qualifying stroke scores 15 or lower on the NIH Stroke Scale. Sunharbo is the first factor XIa inhibitor approved in the United States. Bayer expected shipments to begin in mid-to-late November 2026.
How much does Sunharbo cost?
The US list price is $748.50 for a 30-day supply, which is $24.95 a day and about $9,107 a year. Bayer says the list price is not what most health plans or patients will pay. Eligible patients can receive the drug free through its patient assistance foundation. At list price, preventing one stroke in the first year takes about $483,000 of drug (computed).
How is Sunharbo taken?
The label gives one dose: a 50 mg tablet once a day, with or without food, alongside antiplatelet therapy. A missed dose can be taken later the same day. After that the label says to skip it and take the next dose at the usual time. Use is not recommended in moderate or severe liver impairment.
What are the side effects of Sunharbo?
The label names diarrhoea as the most common adverse reaction, at 3.4% against 2.8% on placebo. Major bleeding ran at 1.6 events per 100 patient-years against 1.5 on placebo, and at 4.5 against 3.1 in the first 90 days. Gastrointestinal bleeding was more frequent with Sunharbo, at 2.7 against 2.1 per 100 patient-years. The label has no boxed warning and lists no contraindications.
Is asundexian approved for atrial fibrillation?
No. The OCEANIC-AF trial was stopped early after asundexian 50 mg proved inferior to apixaban in 14,810 patients with atrial fibrillation. Stroke or systemic embolism occurred in 98 patients on asundexian and 26 on apixaban, a hazard ratio of 3.79. The registry lists no asundexian trial as recruiting in atrial fibrillation.
How does asundexian work?
Asundexian is an oral inhibitor of activated factor XI (factor XIa), an enzyme that amplifies thrombin production once a clot has begun to form. Ordinary sealing of a wound relies mainly on tissue factor and factor Xa. So blocking factor XIa is expected to slow clot growth in an artery with less effect on normal bleeding control.
Asundexian vs milvexian: what is the difference?
Both are oral factor XIa inhibitors. Asundexian (Bayer) is approved in the United States and China for secondary stroke prevention at 50 mg once daily. Milvexian (Bristol Myers Squibb and Johnson & Johnson) is still in phase 3. It is dosed at 25 mg twice daily in the 12,523-patient LIBREXIA-STROKE trial and at 100 mg twice daily against apixaban in LIBREXIA-AF. No trial compares the two drugs directly.
Is Sunharbo available in India?
Not yet. Bayer has announced no Indian filing or price. Twelve Indian hospitals were sites in OCEANIC-STROKE. Under Rule 101, India’s regulator can consider waiving a local trial for a drug approved in the United States, the UK, Japan, Australia, Canada or the EU. So the asundexian FDA approval can support a filing. The waiver is discretionary.
Author. Dr. Akhilesh Vats is Editor-in-Chief of Witfire Elite and founder of ACME Research Solutions, a pharmaceutical contract research organisation.
Disclosure: Editorial analysis, not investment or medical advice. The Witfire Risk Score is our own framework, weighting Regulatory Exposure (30%), Competitive Displacement (25%), Capital Position (20%), Evidence Integrity (15%) and Execution & Credibility (10%) on a 1–10 scale. Scores are judgments, not forecasts.
Put this framework on your own file.
The same method used in these briefs, applied privately to a decision you are actually facing — a target, a partner, a facility, a filing.
- CMC & manufacturing risk assessment
- Regulatory / CRL readiness review
- Due diligence on a target or partner
- Investor brief or market analysis
- Commissioned editorial
- Something else
The signal before the market catches up.
A low-noise briefing on the pharma events that change strategy, risk and value. No more than one a week.
