SUMMARY
- Milvexian dose selection for the LIBREXIA clinical program was informed by phase 2 (AXIOMATIC) dose-ranging studies, pharmacokinetic/pharmacodynamic (PK/PD) analyses, exposure-response modeling, and model-based meta-analysis (MBMA).1,2
- The 25 mg twice-daily (BID) milvexian dose selected for the LIBREXIA ACS trial and LIBREXIA SSP trials was based on findings from the phase II AXIOMATIC-TKR and AXIOMATIC-SSP studies.1,3
- The 100 mg BID milvexian regimen selected for the LIBREXIA AF trial was supported by preclinical PK and PD data, findings from the phase II AXIOMATIC-TKR trial, and a MBMA comparing milvexian with apixaban.1,4
- The table Milvexian Dosing Regimens Across Clinical Trials provides an overview of milvexian dosing regimens evaluated across Phase 2 and Phase 3 clinical trials, along with comparator treatments and trial status/results. Detailed dose-specific efficacy and safety findings are presented in the Clinical Data section.
Milvexian Dosing Regimens Across Clinical Trials
|
|
|
|
|
|---|
Phase 2
| AXIOMATIC-TKR5 (NCT03891524) N=1242
| 25 mg, 50 mg, 100 mg, or 200 mg BID or 25 mg, 50 mg, or 200 mg once-daily (QD).
| Enoxaparin 40 mg QD
| Results available.
|
AXIOMATIC-SSP6 (NCT03766581) N=2366
| 25 mg QD or 25 mg, 50 mg, 100 mg, or 200 mg BID.
| Placebo
| Results available.
|
Phase 3
| LIBREXIA-AF4,7 (NCT05757869) N=20,000
| 100 mg BID
| Apixaban 5 mg or 2.5 mg BID
| Active, not recruiting/Not looking for participants.
|
LIBREXIA-STROKE8 (NCT05702034) N=15,000
| 25 mg BID
| Placebo
| Active, not recruiting/Not looking for participants.
|
LIBREXIA-ACS3,9 (NCT05754957) N=16,000
| 25 mg BID
| Placebo
| Trial discontinued/ Results available.
|
BRISTOL MYERS SQUIBB AND JOHNSON & JOHNSON COLLABORATION
Milvexian is an oral investigational factor XIa (FXIa) inhibitor being studied for the prevention of major thrombotic conditions in a worldwide collaboration program between Bristol Myers Squibb and Johnson & Johnson. Milvexian is not currently approved for any use.
To read more about how Johnson & Johnson processes personal data, please visit https://innovativemedicine.jnj.com, select your country or region (in the upper right corner), and then click on Privacy Policy.
To read more about how Bristol Myers Squibb processes personal data, please visit https://www.bms.com, select your country (in the upper right corner), and then click on Privacy Policy.
PHARMACOLOGY OF MILVEXIAN
Milvexian is an oral, small-molecule FXIa inhibitor that binds selectively and reversibly to the active site of FXIa, thereby inhibiting its activity.1
Pharmacokinetics
Milvexian is rapidly absorbed after oral administration, with a median time to maximum plasma concentration (Cmax) of 2-4 hours and a mean half-life of 11-18 hours following a single dose. The bioavailability of milvexian when administered as 25 mg spray-dried dispersion formulation was approximately 58%.1
Milvexian is eliminated through CYP3A4-mediated hepatic metabolism and direct biliary excretion, with <20% excreted unchanged by the kidneys. It is also a substrate of P-glycoprotein. In drug-drug interaction studies conducted in healthy subjects, itraconazole and diltiazem increased the maximum concentration of milvexian by 1.3-fold and 1.1-fold, respectively; increased the area under the curve by 2.5-fold and 1.4-fold, respectively; and prolonged the half-life by 1.5-fold and 1.1-fold, respectively.1
Pharmacodynamics
Milvexian prolongs aPTT in a dose-dependent manner and inhibits kaolin-initiated thrombin generation in vitro. Milvexian does not affect prothrombin time.1
Phase 2 Studies
AXIOMATIC-TKR
In the AXIOMATIC-TKR trial,1,5 1242 patients undergoing knee arthroplasty were randomized to receive 1 of 7 postoperative milvexian doses (25, 50, 100, or 200 mg BID, or 25, 50, or 200 mg QD) or enoxaparin 40 mg QD for 10-14 days after surgery, until unilateral venography was performed.
Results
Efficacy
- The dose-response relationship between BID milvexian and venous thromboembolism (VTE) was statistically significant. The incidence of VTE across all BID milvexian regimens was 12%, which was lower than the prespecified benchmark of 30%.
- Total daily milvexian doses of ≥100 mg were associated with lower VTE rates than enoxaparin, and total daily doses of ≥200 mg reduced VTE rates by >50% compared with enoxaparin. See Table: Rates of Venous Thromboembolism Associated With Different Milvexian Dose Regimens and Enoxaparin.
- Milvexian treatment resulted in dose-dependent prolongation of aPTT, with BID dosing showing less peak-to-trough variation in activated partial thromboplastin time (aPTT) than QD dosing.
- At total daily milvexian doses of ≥100 mg, the aPTT ratio remained >2 at both peak and trough and was associated with a significant reduction in postoperative VTE compared with enoxaparin.
Rates of Venous Thromboembolism Associated With Different Milvexian Dose Regimens and Enoxaparin1
|
|
|
|---|
Milvexian
| 25 mg BID
| 21
|
50 mg BID
| 11a
|
100 mg BID
| 9b
|
200 mg BID
| 8c
|
25 mg QD
| 25
|
50 mg QD
| 24
|
200 mg QD
| 7ᶜ
|
Enoxaparin
| 40 mg QD
| 21
|
Abbreviations: BID, twice daily; QD, once daily; VTE, venous thromboembolism. aP<0.01 vs enoxaparin. bP<0.002 vs enoxaparin. cP<0.0001 vs enoxaparin.
|
Safety
- Serious adverse events occurred in 2% of patients receiving milvexian and 4% of those receiving enoxaparin.
- Any bleeding event, regardless of severity, was reported in 4% (38/923) of milvexian-treated patients and 4% (12/296) of enoxaparin-treated patients, while clinically relevant bleeding, defined as the composite of major bleeding and clinically relevant nonmajor (CRNM) bleeding, occurred in 1% and 2% of patients, respectively. See Table: Incidence of Any Bleeding and Clinically Relevant Bleeding.
- No dose-dependent increase in clinically relevant bleeding was observed across the
16-fold range of milvexian doses.
Incidence of Any Bleeding and Clinically Relevant Bleeding1
|
|
|
|
|---|
Milvexian
| 25 mg BID
| 1
| 0
|
50 mg BID
| 5
| 1
|
100 mg BID
| 5
| 1
|
200 mg BID
| 3
| 1
|
25 mg QD
| 0
| 0
|
50 mg QD
| 5
| 1
|
200 mg QD
| 6
| 1
|
Enoxaparin
| 40 mg QD
| 4
| 2
|
Abbreviations: BID, twice daily; QD, once daily.
|
AXIOMATIC-SSP
In the AXIOMATIC-SSP trial,1,6 2366 patients aged ≥40 years with acute ischemic stroke (AIS) (<48 hours) or high-risk transient ischemic attack (TIA) were randomized in a 1:1:1:1:1:2 ratio to receive placebo BID for 90 days (n=691), milvexian 25 mg QD (n=328), or milvexian 25 mg (n=318), 50 mg (n=328), 100 mg (n=310), or 200 mg (n=351) BID for 90 days. All participants received clopidogrel 75 mg daily for the first 21 days and aspirin 100 mg daily for 90 days.
Results
Efficacy
- The median age of patients was 71 years (interquartile range [IQR], 62-77), and 859 patients (36%) were female.
- At 90 days, the estimated rates of symptomatic ischemic stroke or covert brain infarction were 16.8% (90.2% CI, 14.5%-19.1%) with placebo, 16.7% (90.2% CI, 14.8%-18.6%) with milvexian 25 mg QD, 16.6% (90.2% CI, 14.8%-18.3%) with milvexian 25 mg BID, 15.6% (90.2% CI, 13.9%-17.5%) with milvexian 50 mg BID, 15.4% (90.2% CI, 13.4%-17.6%) with milvexian 100 mg BID, and 15.3% (90.2% CI, 12.8%-19.7%) with milvexian 200 mg BID.
- No significant dose-response relationship was observed among the 5 milvexian doses for the primary composite efficacy outcome or for major bleeding. See Table: Incidence Rates of Major Bleeding.
- Lower point estimates of recurrent symptomatic ischemic stroke were reported with all milvexian doses except the 200 mg BID dose.
- Rates of symptomatic intracranial hemorrhage (ICH) were comparable across all milvexian doses vs placebo. See Table: Incidence Rates of Recurrent Symptomatic Ischemic Stroke and Intracranial Hemorrhage With Milvexian vs Placebo.
- An increased rate of ischemic stroke events was observed in the milvexian 200 mg BID group compared with placebo.
Incidence Rates of Major Bleeding1
|
|
|
|
|
|
|---|
Placebo
| -
| 4/682
| 0.6
| Reference
| -
|
Milvexian
| 25 mg QD
| 2/325
| 0.6
| 1.05
| 0.14-5.89
|
25 mg BID
| 2/313
| 0.6
| 1.09
| 0.15-6.11
|
50 mg BID
| 5/325
| 1.5
| 2.62
| 0.64-11.75
|
100 mg BID
| 5/306
| 1.6
| 2.79
| 0.68-12.47
|
200 mg BID
| 5/344
| 1.5
| 2.48
| 0.60-11.09
|
Abbreviations: BID, twice daily; CI, confidence interval; QD, once daily; RR, relative risk.
|
Incidence Rates of Recurrent Symptomatic Ischemic Stroke and Intracranial Hemorrhage With Milvexian vs Placebo1
|
|
|
|
|
|
|---|
Placebo
| -
| 38/691
| 5.5
| 2/682
| 0.3
|
Milvexian
| 25 mg QD
| 15/328
| 4.6
| 0/325
| 0
|
25 mg BID
| 12/318
| 3.8
| 0/313
| 0
|
50 mg BID
| 13/328
| 4
| 3/325
| 0.9
|
100 mg BID
| 11/310
| 3.5
| 0/306
| 0
|
200 mg BID
| 27/351
| 7.7
| 1/344
| 0.3
|
Abbreviations: BID, twice daily; QD, once daily.
|
Phase 3 Studies
LIBREXIA ACS
LIBREXIA ACS (NCT05754957)3 was a phase 3, multicenter, double-blind, randomized, placebo-controlled, parallel-group trial evaluating whether milvexian, when added to standard-of-care therapy, can reduce the risk of major adverse cardiovascular events (MACEs) in patients with a recent acute coronary syndrome (ACS) without increasing significant bleeding. Following a planned interim analysis of 556 adjudicated efficacy end points, the trial was terminated for futility.10
Dosing Rationale
- A 25 mg BID oral dose of milvexian was selected based on findings from two phase 2 studies.
- In the AXIOMATIC-TKR trial, this dose demonstrated efficacy comparable to enoxaparin for the prevention of VTE and showed a comparable safety profile.
- In the AXIOMATIC-SSP trial, the 25 mg BID dose administered in combination with 21 days of dual antiplatelet therapy was associated with numerically lower rates of recurrent ischemic stroke and similar bleeding rates at 90 days compared with placebo. This dose also demonstrated measurable pharmacologic effects on aPTT and FXIa inhibition. Higher doses did not improve efficacy and were associated with numerically higher rates of major bleeding.
LIBREXIA STROKE
LIBREXIA-STROKE (NCT05702034)8 is a Phase 3, randomized, double-blind, parallel-group, placebo-controlled study evaluating the efficacy and safety of milvexian compared with placebo for reducing the risk of recurrent ischemic stroke following an acute ischemic stroke or high-risk transient ischemic attack.
Dosing Rationale
- In the LIBREXIA Stroke program, milvexian 25 mg BID is administered in addition to standard antiplatelet therapy, including dual or single antiplatelet treatment.1
- The 25 mg BID regimen, the lowest BID dose evaluated in the phase II AXIOMATIC-SSP trial, was selected to assess its efficacy and safety in patients receiving concurrent antiplatelet therapy. The dose selection was supported by prior dual pathway inhibition studies in which anticoagulants were combined with antiplatelet agents to target coagulation and platelet activation through complementary mechanisms.1
LIBREXIA AF
LIBREXIA AF (NCT05757869)4 is a phase III, randomized, double-blind, double-dummy, parallel-group, multicenter, event-driven, active-controlled trial evaluating milvexian vs apixaban in patients with AF.
Dosing Rationale
- Selection of the 100 mg BID milvexian dose was based on preclinical PK and PD data, results from the phase II AXIOMATIC-TKR trial in venous thromboembolic disease, and a MBMA comparing milvexian with apixaban.
- Assessment of antithrombotic efficacy and dose selection in venous thromboembolic disease has traditionally been used to inform dosing strategies in AF. The dose-selection approach used for milvexian in LIBREXIA AF was consistent with the development strategy previously applied to anticoagulants, including apixaban.
Zhou et al (2026)2 conducted a study to support phase III dose selection for LIBREXIA AF. The following 2 approaches were applied:
- Population PK and exposure-response analyses using data from phase I studies and the phase II AXIOMATIC-TKR trial to evaluate relationships among milvexian exposure, pharmacokinetics, clinical efficacy, safety outcomes, and coagulation biomarkers
- An MBMA incorporating published clinical trial data from comparator anticoagulants in patients undergoing THR or TKR.
Milvexian plasma concentration-time data from phase I studies in healthy participants (18,449 PK samples; 317 participants) and the AXIOMATIC-TKR trial (922 patients) were used to develop the population PK model. Clinical efficacy, safety, and anticoagulation biomarker data from the AXIOMATIC-TKR trial were included in the exposure-response and PK-PD analyses.
For the MBMA, data on VTE and bleeding events were collected from international, randomized, placebo- or active-controlled, parallel-design studies of antithrombotic agents in patients undergoing total knee replacement (TKR) or total hip replacement (THR), published between 1986 and 2021. Data from the AXIOMATIC-TKR trial were also incorporated into the MBMA dataset.
Results
Milvexian Population PK Modelling
- A dose-dependent increase in modeled median plasma milvexian concentrations was observed.
- The population PK model was used to simulate milvexian exposure in patients with AF, assuming that apparent clearance would be similar to that observed in patients undergoing TKR.
- For the 100 mg BID milvexian regimen, the model estimated a steady-state Cmax to steady-state minimum plasma concentration (Ctrough) ratio of 1.35 in patients with AF, indicating limited fluctuation in plasma concentrations across the 12-hour dosing interval.
Exposure-Response Analysis and PK-PD Modelling
- Increasing milvexian systemic exposure was associated with lower probability of total VTE. At the 100 mg BID regimen, the simulated median VTE rate was 9.4% (5th-95th percentiles, 7.1%-12.8%), compared with an observed enoxaparin VTE rate of 21.4% (5th-95th percentiles, 16.4%-26.5%).
- Across the evaluated dose range (25 mg QD to 200 mg BID), no meaningful relationship was observed between milvexian exposure and the incidence of any bleeding or CRNM bleeding. Major bleeding was not assessed because no major bleeding events occurred in the milvexian-treated group.
- Milvexian exposure was positively associated with aPTT prolongation in the AXIOMATIC-TKR trial. Increases in plasma milvexian concentrations were accompanied by increases in aPTT, with the response appearing to plateau at concentrations >9000 ng/mL.
- For the clinically relevant 100 mg BID regimen, the modeled mean aPTT ratio to baseline at 4 hours and 12 hours after dosing was 79 seconds (5th-95th percentiles, 72-88 seconds) and 73 seconds (5th-95th percentiles, 66-83 seconds), respectively.
- No change in prothrombin time (PT) was observed with increasing milvexian exposure. Additional analyses did not identify a relationship between changes in aPTT (measured as the maximum aPTT ratio to baseline) and the probability of any bleeding event.
Model-based Meta-analysis (MBMA)
- A total of 45 clinical studies evaluating anticoagulant therapy in patients undergoing TKR or THR were included in the MBMA. Data were extracted from 163 treatment groups representing 56,392 participants.
- Among the included studies, 43 reported any bleeding events (53,009 participants), 32 reported CRNM bleeding events (46,925 participants), 45 reported major bleeding events (56,392 participants), and 45 reported VTE outcomes (41,582 participants). Twenty-four studies included patients undergoing THR and 22 included patients undergoing TKR. The dataset comprised 9 active treatments and placebo, including 11 placebo-controlled trials.
- The MBMA joint nonlinear mixed-effects model incorporated data from the phase II AXIOMATIC-TKR trial, in which enoxaparin served as the comparator. Model diagnostics showed agreement between modeled probabilities and reported event rates, with weighted residual analyses indicating no apparent trend by study publication year or dose. Random effects were generally centered around zero and were consistent with a normal distribution.
- Apixaban 5 mg BID was selected as the comparator of interest for the phase III LIBREXIA AF program. In TKR/THR populations, the modeled VTE odds ratio improved with increasing milvexian doses, with doses ≥75 mg BID showing lower modeled VTE odds relative to apixaban 5 mg BID. For the milvexian 100 mg BID regimen, the modeled VTE odds ratio vs apixaban 5 mg BID was 0.84 (95% CI, 0.65-1.12).
- Although bleeding endpoints were included in the MBMA, comparative bleeding assessments across treatments were limited because of the low incidence of major or CRNM bleeding observed in the AXIOMATIC-TKR trial, where the combined rate across milvexian doses ranging from 25 mg QD to 200 mg BID was 0.8%.
Literature Search
A literature search of MEDLINE®, EMBASE®, BIOSIS Previews®, Derwent® (and/or other resources, including internal/external databases) was conducted on 09 October 2026.
| 1 | Weitz JI, Harrington RA, LIBREXIA Program Executive Council. Rationale for the milvexian dosing in the phase 3 LIBREXIA program. J Thromb Haemost. 2026;24(6):2158-2169. |
| 2 | Zhou W, Bozenhardt E, Alexander GE, et al. Quantitative model‐informed dose selection for a milvexian phase III study in patients with atrial fibrillation. Clin Pharmacol Ther. 2026;119(1):139-146. |
| 3 | Gibson CM, Bahit MC, Mehran R, et al. Oral factor XIa inhibitor milvexian after a recent acute coronary syndrome: rationale and design of the phase 3 (Librexia ACS). Am Heart J. 2025;285:21-28. |
| 4 | Jain SS, Mahaffey KW, Pieper KS, et al. Milvexian vs apixaban for stroke prevention in atrial fibrillation: the LIBREXIA atrial fibrillation trial rationale and design. Am Heart J. 2024;277:145-158. |
| 5 | Janssen Research & Development, LLC. A study of JNJ-70033093 (BMS-986177) versus subcutaneous enoxaparin in participants undergoing elective total knee replacement surgery (AXIOMATIC-TKR). In: ClinicalTrials.gov [Internet]. Bethesda (MD): National Library of Medicine (US). 2000- [cited 2026 October 09]. Available from: https://clinicaltrials.gov/ct2/show/study/NCT03891524 NLM Identifier: NCT03891524. |
| 6 | Janssen Research & Development, LLC. A study on BMS-986177 for the prevention of a stroke in patients receiving aspirin and clopidogrel (AXIOMATIC-SSP). In: ClinicalTrials.gov [Internet]. Bethesda (MD): National Library of Medicine (US). 2000- [cited 2026 October 09]. Available from: https://clinicaltrials.gov/ct2/show/study/NCT03766581 NLM Identifier: NCT03766581. |
| 7 | Janssen Research & Development, LLC. A study of milvexian versus apixaban in participants with atrial fibrillation (LIBREXIA-AF). In: ClinicalTrials.gov [Internet]. Bethesda (MD): National Library of Medicine (US). 2000- [cited 2026 October 09]. Available from: https://clinicaltrials.gov/ct2/show/NCT05757869 NLM Identifier: NCT05757869. |
| 8 | Janssen Research & Development, LLC. A study of milvexian in participants after an acute ischemic stroke or high-risk transient ischemic attack - LIBREXIA-STROKE (LIBREXIA-STROK). In: ClinicalTrials.gov [Internet]. Bethesda (MD): National Library of Medicine (US). 2000- [cited 2026 October 09]. Available from: https://www.clinicaltrials.gov/ct2/show/NCT05702034 NLM Identifier: NCT05702034. |
| 9 | Janssen Research & Development, LLC. A study of milvexian in participants after a recent acute coronary syndrome (LIBREXIA-ACS). In: ClinicalTrials.gov [Internet]. Bethesda (MD): National Library of Medicine (US). 2000- [cited 2026 October 09]. Available from: https://clinicaltrials.gov/ct2/show/study/NCT05754957 NLM Identifier: NCT05754957. |
| 10 | Gibson CM, Steg PG, Bahit MC, et al. Milvexian with antiplatelet therapy after acute coronary syndrome event. [published online ahead of print August 29, 2026]. N Engl J Med. 2026. doi:10.1056/nejmoa2608717. |