Asundexian
TL;DR
Asundexian (BAY 2433334) is the first-in-class oral direct inhibitor of activated Factor XI (FXIa). In the Phase 3 OCEANIC-STROKE trial (N=12,327), asundexian 50 mg once daily added to antiplatelet therapy reduced recurrent ischemic stroke by 26% vs placebo (HR 0.74, P<0.001) with no statistically significant increase in major bleeding (HR 1.10). It does not work for lacunar strokes or atrial fibrillation. Not yet FDA or EMA approved as of April 2026; regulatory submission expected following NEJM publication.
Why it matters for Vitals
- Secondary stroke prevention is a core cardiovascular coaching domain
- The stroke subtype distinction (lacunar vs non-lacunar) is a critical decision filter — Vitals coaching must never suggest asundexian without confirmed non-lacunar imaging
- The low-bleeding-profile FXIa mechanism is a model for future stack safety reasoning
- First anticoagulant-class agent that can be added to antiplatelet therapy without prohibitive bleeding risk
- Vitals coaching should flag AF screening (ECG/holter) and stroke subtyping (MRI/CT) as prerequisites before any asundexian consideration
- The FXIa inhibitor class failure in atrial fibrillation establishes a clear mechanistic boundary — this is arterial thrombosis, not AF-related fibrin clot
Key facts
| Class | Oral direct FXIa inhibitor (small molecule; first-in-class) |
| Compound ID | BAY 2433334 |
| Dose | 50 mg once daily |
| Phase | Phase 3 complete; not yet FDA/EMA approved |
| Key trial | OCEANIC-STROKE (N=12,327, double-blind, placebo-controlled) |
| Primary efficacy | Recurrent ischemic stroke: HR 0.74 (P<0.001), ARR ~2.2%, NNT ~45 |
| Major bleeding | BARC ≥3: HR 1.10 (95% CI 0.85–1.44), not statistically significant |
| Excluded populations | Lacunar stroke, atrial fibrillation, active bleeding |
| Key competitor | Milvexian (LIBREXIA-Stroke, Phase 3, BID dosing) |
| CYP/DDI profile | Appears minimal (Phase 1/2 data); full profile pending label |
Mechanism summary
Asundexian inhibits Factor XIa (FXIa), the central amplifier of the intrinsic coagulation pathway. The intrinsic loop (FXIIa → FXIa → FIXa → FVIIIa → FXa → FIIa) is disproportionately important in arterial thrombosis — specifically platelet-rich “white clot” formed on atherosclerotic plaque rupture or erosion. It plays a minimal role in primary hemostasis or in fibrin-rich “red clot” formed in atrial fibrillation.
Humans with congenital FXI deficiency have markedly reduced ischemic stroke risk with only mild bleeding diathesis (primarily post-surgical/traumatic, not spontaneous). This separable protection-from-bleeding profile is the biological foundation for FXIa as a therapeutic target.
Mechanism vs other anticoagulants
| Drug Class | Target | Bleeding Risk vs Placebo | Evidence in Non-Cardioembolic Stroke |
|---|---|---|---|
| Asundexian | Factor XIa (direct) | Neutral | Positive (Phase 3) |
| Rivaroxaban | Factor Xa inhibitor | Elevated | Negative (NAVIGATE ESUS) |
| Dabigatran | Factor IIa inhibitor | Elevated | Negative (RE-SPECT ESUS) |
| Aspirin | Antiplatelet | Minimal | Positive (weak) |
| Clopidogrel | Antiplatelet | Minimal | Positive (weak) |
DOACs (Factor Xa and IIa inhibitors) failed in ESUS/cryptogenic stroke. Asundexian succeeded in a similar population, suggesting the FXIa mechanism may address thrombosis pathways not covered by Factor Xa/IIa inhibition or antiplatelet therapy alone.
Why lacunar stroke is excluded
Lacunar infarcts result from lipohyalinosis of small penetrating arteries — a distinct pathology not driven by FXIa-dependent thrombus formation on atherosclerotic plaque. PACIFIC-Stroke confirmed null effect in this subtype (HR ~1.14).
Why atrial fibrillation is excluded
AF generates fibrin-rich clot via the extrinsic (tissue factor) pathway, where FXIa plays a minimal role. OCEANIC-AF was terminated early — asundexian was inferior to apixaban. The Factor XIa mechanism note covers this class limitation in more detail.
What the current evidence suggests
Stroke prevention (non-cardioembolic, non-lacunar)
Confirmed. OCEANIC-STROKE (N=12,327) showed HR 0.74 for recurrent ischemic stroke — robust, pre-specified, statistically significant. ARR ~2.2%; NNT ~45 over the trial period (~1–2 years median follow-up).
Bleeding safety
Supported (no statistically significant excess). Major bleeding HR 1.10 (95% CI 0.85–1.44) — the CI includes a 44% relative increase, which cannot be excluded. The direction is unfavorable; the signal warrants close post-approval monitoring. ICH (intracranial hemorrhage) rate not separately disclosed in the available abstract — the most feared bleeding event in a stroke population is a data transparency gap.
Atrial fibrillation
Confirmed ineffective. OCEANIC-AF terminated early; asundexian inferior to apixaban.
Lacunar stroke
Supported ineffective. PACIFIC-Stroke showed HR ~1.14 in this subtype.
DAPT combination
The DAPT (dual antiplatelet) subgroup showed no excess major bleeding vs placebo + DAPT — the most clinically meaningful safety finding for patients on maximum antiplatelet therapy.
Likely wearable / Vitals relevance
Currently no validated wearable signal. No specific wearable biomarker is supported by current evidence for asundexian.
What Vitals should not do without clinician sign-off:
- Use wearable data to guide asundexian dosing, duration, or safety monitoring
- Recommend asundexian for any stroke subtype without confirmed imaging
- Recommend asundexian where AF has not been excluded
- Use HRV, HR, sleep, or activity as surrogate for asundexian efficacy or bleeding risk
Potential future relevance (speculative):
- Bleeding risk monitoring — HRV, activity, or other biometric proxies for subclinical bleeding have not been validated for asundexian
- Post-stroke autonomic function — FXIa inhibition does not have a known HRV or sleep architecture signature
Risks and uncertainty
| Risk | Status |
|---|---|
| Major bleeding — upper bound CI 44% increase | Cannot be excluded |
| ICH rate — not disclosed in available abstract | Unknown; most feared bleeding event in stroke population |
| Long-term safety (>1 year) | Unknown |
| Optimal treatment duration | Unknown |
| PK in severe hepatic impairment | Not established |
| PK in severe renal impairment | Not established |
| Regulatory approval | Not yet approved; submission expected |
| Head-to-head vs milvexian | None |
| No reversal agent | Confirmed; unlike DOACs (idarucizumab, andexanet alfa) |
Vitals coaching protocol
APPLICABLE WHEN:
user.has_prior_stroke_or_tia == true
AND user.is_under_neurology_or_cardiology_care == true
AND user.prescribed_asundexian == true
COACHING PILLARS (complementary to asundexian):
1. BP optimization: <130/80 mmHg target for secondary stroke
2. LDL-C: <70 mg/dL for atherosclerotic cerebrovascular disease
3. HbA1c: <7% for diabetic stroke patients
4. HRV assessment: track recovery, flag persistent autonomic dysfunction
5. Physical activity: 150 min/week moderate intensity minimum
6. Smoking cessation: absolute priority for cerebrovascular patients
DO NOT USE:
- HRV, HR, sleep, or activity as surrogate for asundexian efficacy
- Vitals recommendations to adjust or discontinue asundexian
- Any biometric signal as proxy for coagulation status
FLAG FOR PHYSICIAN REVIEW:
- Any bleeding event (GI, genitourinary, epistaxis requiring intervention)
- New neurological symptoms
- Planned invasive procedures requiring anticoagulation hold
Best stack context
Asundexian sits within the secondary stroke prevention stack — an add-on to existing antiplatelet therapy, not a standalone intervention.
Appropriate stack anchors:
- Antiplatelet therapy (aspirin, clopidogrel, or aspirin + dipyridamole) — required background
- Statin (high-intensity) — standard of care post-ischemic stroke
- BP control (target <130/80 per current guidelines) — standard of care
- Lifestyle (smoking cessation, physical activity, Mediterranean diet) — standard of care
Do not combine:
- With anticoagulants (rivaroxaban, apixaban, dabigatran, edoxaban) — excess bleeding risk
- With fibrinolytics — excess bleeding risk
- In lacunar stroke — ineffective per PACIFIC-Stroke
- In atrial fibrillation — inferior to apixaban per OCEANIC-AF
Caution:
- GI bleeding history
- Abnormal uterine bleeding (limited data; flagged in systematic review PMID 41275204)
- Concomitant NSAIDs
Evidence summary
| Claim | Evidence Grade | Source |
|---|---|---|
| 26% RRR in ischemic stroke (HR 0.74, P<0.001) | Confirmed | OCEANIC-STROKE, PMID 41985132 |
| Major bleeding identical to placebo (1.9% vs 1.7%, HR 1.10) | Confirmed | OCEANIC-STROKE, PMID 41985132 |
| First-in-class oral FXIa inhibitor | Confirmed | PACIFIC-Stroke, PMID 36063821 |
| Dose-dependent FXIa inhibition at 10/20/50 mg | Confirmed | PACIFIC-Stroke, PMID 36063821 |
| Phase 2b primary MRI endpoint was null | Confirmed | PACIFIC-Stroke, PMID 36063821 |
| FXIa inhibitor failed in AF stroke prevention | Confirmed (class) | Milestone-AF, NCT04272814 |
| No DOAC comparator exists | Gap | — |
| No reversal agent | Confirmed | — |
| Lacunar stroke excluded | Confirmed | Trial design |
| Regulatory approval pending | Confirmed | NCT06142383 |
What stays inside this hub
- Full OCEANIC-STROKE trial data (Phase 3 primary results)
- PACIFIC-Stroke Phase 2 mechanistic context
- Patient selection algorithm
- Safety profile (bleeding, contraindications, cautions)
- Pharmacology summary (dose, DDI profile, PK gaps)
- Comparative landscape (milvexian, FXa inhibitors, antiplatelets)
- Vitals coaching boundaries
- Implementation algorithm hooks
Related notes
- Factor XIa — reusable mechanism note; intrinsic coagulation amplification; biological rationale for FXIa as therapeutic target; class limitations in AF
- Milvexian — competing FXIa inhibitor; Phase 3 ongoing (LIBREXIA-Stroke, BID dosing)
- Cardiovascular risk — broader cardiovascular risk management; links to stroke prevention domain