Monoclonal antibodies targeting FXI and/or FXIa represent an important modality in FXI-pathway inhibition and have been evaluated in both preclinical models and early-phase clinical studies. Depending on their epitope, these antibodies may prevent FXI activation, inhibit FXIa activity, or disrupt interactions between FXI/FXIa and other components of the coagulation cascade.

(1) BAY 1213790 (FXIa-targeting antibody): Early clinical evaluation showed dose-dependent aPTT prolongation and reduced FXI activity, with no apparent effects on bleeding time or clinically relevant anti-drug antibody formation, supporting further clinical evaluation^[36]. (2) Abelacimab (dual FXI/FXIa-targeting antibody): Phase 1 and 2 data demonstrate a terminal half-life of 25-30 days, providing sustained FXI suppression and aPTT prolongation for over 4 weeks. Its monthly subcutaneous dosing may offer adherence advantages for chronic anticoagulation^[31,32]. (3) AB023 (FXI-targeting antibody): AB023 selectively inhibits FXIIa-mediated FXI activation while preserving thrombin-mediated feedback activation. In a Phase 2 trial of end-stage kidney disease (ESKD) patients on hemodialysis, AB023 significantly improved circuit patency, reduced occlusive events, and lowered TAT and CRP levels^[25,26]. (4) BJTJ-1837 (FXI-targeting antibody): In preclinical primate models, BJTJ-1837 reduced thrombosis in arteriovenous shunt models without an apparent increase in bleeding^[37]. (5) BAY 1831865 (dual FXI/FXIa-targeting antibody): In early-phase evaluation, BAY 1831865 showed rapid onset after intravenous administration, dose-dependent systemic exposure, and a half-life of up to 208 h. It reduced FXI clotting activity and was not associated with major bleeding signals in the available early-phase data^[38]. (6) SKB336 (dual FXI/FXIa-targeting antibody): An anti-FXI antibody with a prolonged half-life of up to 33 days. Healthy volunteer data showed dose-proportional exposure and sustained FXI inhibition, supporting further evaluation of once-monthly dosing. However, bleeding risk requires confirmation in larger patient studies^[39].

ASOs targeting FXI mRNA reduce hepatic FXI synthesis and have been evaluated in early clinical and selected patient studies. They bind FXI mRNA and promote RNase H-mediated degradation, thereby reducing FXI protein synthesis.

(1) ISIS-FXIRx: This agent achieves FXI suppression by degrading hepatic FXI mRNA. Clinical data showed that repeated dosing could produce marked reductions in FXI activity and FXI antigen levels over several weeks^[29]. (2) Fesomersen: In patients with kidney failure receiving hemodialysis, fesomersen reduced FXI levels in a dose-dependent manner. Lower FXI levels were associated with fewer hemodialysis circuit-clotting events and possibly fewer vascular access thrombotic events^[28].

Small-molecule FXIa inhibitors directly inhibit the active site of FXIa.

(1) Milvexian: An oral FXIa inhibitor with predictable pharmacokinetic/pharmacodynamic profiles and generally favorable tolerability in early studies. Phase 1/2 studies showed dose-dependent aPTT prolongation, with no major bleeding signal reported in the studied populations^[33]. (2) SHR2285: An oral FXIa inhibitor developed in China. It produced dose-dependent FXIa inhibition and aPTT prolongation in early-phase studies. The absence of clinically meaningful pharmacokinetic or pharmacodynamic interactions with aspirin or ticagrelor in early studies supports further evaluation in combination with antiplatelet therapy^[40-42]. (3) Asundexian (BAY 2433334): A once-daily oral FXIa inhibitor. Early studies showed predictable pharmacokinetic/pharmacodynamic profiles and low bleeding rates compared with apixaban in selected settings, but the interpretation of safety must be balanced against later efficacy concerns. However, the Phase 3 OCEANIC-atrial fibrillation (AF) trial was terminated early because asundexian was less effective than apixaban for prevention of stroke or systemic embolism. This finding indicates that reduced bleeding alone does not guarantee non-inferior efficacy and that dose, degree of FXIa inhibition, patient selection, and prior anticoagulant exposure require careful consideration^[34,43,44]. (4) ONO-7684: An oral FXIa inhibitor that showed dose-dependent pharmacodynamic effects, including reduced FXI clotting activity and aPTT prolongation, with a half-life of up to 27.9 h in early-phase evaluation. No major bleeding signal was reported in the available early data^[45]. (5) EP-7041: An investigational parenteral FXIa inhibitor being evaluated for critical care settings such as extracorporeal membrane oxygenation (ECMO), where rapid titration and offset may be clinically useful^[46]. (6) Montelukast (MK): Montelukast has been identified through drug-repurposing approaches as a potential FXIa-inhibitory candidate in preclinical studies. Its possible antithrombotic and anti-inflammatory effects remain exploratory and should not be extrapolated to clinical anticoagulation without further validation^[47]. (7) Frunexian & BMS-962212: Frunexian and BMS-962212 are examples of agents with pharmacokinetic profiles that may be suitable for settings requiring rapid onset or offset. Frunexian is a short-acting intravenous agent, whereas BMS-962212 has shown dose-proportional pharmacokinetics in early studies^[48,49].

Ir-CPI is a Kunitz-type protein derived from the saliva of the tick Ixodes ricinus. Ir-CPI inhibits several components of the contact pathway, including FXIIa, FXIa, and plasma kallikrein. In preclinical rat and mouse models, Ir-CPI reduced venous and arterial thrombus formation in a dose-dependent manner^[35]. Its clinical relevance remains to be established.

Although FXI/FXIa inhibitors share a common pathway-level rationale, they should not be considered a homogeneous therapeutic class. Monoclonal antibodies generally provide deep and sustained FXI/FXIa inhibition and may improve adherence through infrequent dosing, but their long half-lives may complicate perioperative interruption, urgent reversal, and management of unexpected bleeding. ASOs reduce hepatic FXI synthesis and therefore have a slower onset and offset, making them more suitable for stable chronic-risk settings than for acute anticoagulation. In contrast, oral small-molecule FXIa inhibitors provide more flexible dosing and faster offset, which may be advantageous in perioperative or rapidly changing clinical scenarios, but they require daily adherence and may be more susceptible to renal function, drug-drug interactions, and dose-dependent efficacy limitations. Parenteral short-acting agents may be particularly relevant for critical care or extracorporeal circulation settings, where rapid titration and offset are important. These differences are clinically important and should be considered when interpreting trial results across indications.

In recent years, FXI/FXIa inhibition has become an active investigational strategy for stroke prevention in AF^[66]. The PACIFIC-AF trial (NCT04218266) systematically evaluated asundexian, an oral FXIa inhibitor, in 753 patients with AF. Results showed that asundexian achieved near-complete in vivo FXIa inhibition at doses of 20 and 50 mg once daily and was associated with lower bleeding rates than standard-dose apixaban^[51]. The subsequent Phase III OCEANIC-AF trial (NCT05643573) was terminated early because asundexian was less effective than apixaban for prevention of stroke or systemic embolism. This result challenged the assumption that FXIa inhibition can directly replace DOAC therapy for stroke prevention in unselected AF populations^[34]. Post-hoc analyses suggested that prior anticoagulant exposure may have influenced the observed efficacy difference, but these findings should be considered exploratory and require prospective validation^[67].

In contrast, the Phase II AZALEA-TIMI 71 study (NCT04755283) provided an important bleeding-safety signal for abelacimab, a monoclonal antibody targeting FXI/FXIa. In 1,287 moderate-to-high-risk patients with AF, monthly subcutaneous abelacimab achieved near-complete FXI suppression and reduced major or clinically relevant non-major bleeding compared with rivaroxaban^[53]. Prespecified subgroup analyses suggested that the bleeding reduction was generally consistent in patients receiving concomitant antiplatelet therapy^[68].

Ongoing investigations, including LILAC-TIMI 76 (NCT05712200) and LIBREXIA-AF (NCT05757869), may help clarify whether specific FXI/FXIa inhibitors can provide an acceptable balance between efficacy and bleeding in selected AF populations. Collectively, AF trial results indicate that bleeding reduction and antithrombotic efficacy may diverge. The clinical performance of FXI/FXIa inhibitors appears to depend on drug modality, dose, degree of pathway inhibition, prior anticoagulant exposure, and patient selection.

Orthopedic procedures, particularly total knee and hip arthroplasty (TKA/THA), provide an established clinical model for evaluating novel anticoagulants because of their high postoperative VTE risk^[69]. In a Phase II TKA study (NCT01713361), the FXI-ASO ISIS 416858/IONIS-FXIRx reduced postoperative VTE compared with enoxaparin and was associated with less bleeding, supporting the feasibility of FXI reduction in this setting^[21]. In the FOXTROT trial, osocimab showed non-inferiority to enoxaparin when administered postoperatively, whereas preoperative administration of 1.8 mg/kg achieved clinical superiority (NCT03276143)^[54]. Similarly, the ANT-005 TKA study (EudraCT2019-003756-37) showed dose-dependent reductions in postoperative VTE with a single postoperative infusion of abelacimab compared with enoxaparin, with low bleeding rates in this Phase II setting^[30].

Milvexian also showed a dose-response relationship for postoperative VTE reduction in the AXIOMATIC-TKR trial (NCT03891524)^[56]. More recently, the ROXI-VTE (NCT05618808; NCT06454630) program further suggested that epitope and domain selection may matter: REGN7508, which targets the catalytic domain, performed better than enoxaparin for the trial-defined endpoint, whereas REGN9933, which binds the Apple 2 domain, did not demonstrate superiority^[70].

Ongoing studies, including KN060, may provide additional information on alternative antibody-based approaches (NCT06180889). Collectively, orthopedic surgery provides some of the most supportive Phase II evidence for FXI/FXIa inhibition, suggesting effective postoperative VTE reduction with low bleeding rates in selected trials. However, these findings still require confirmation in larger outcome studies and should not be extrapolated uncritically to all surgical populations.

CAT is clinically challenging because patients with cancer often have both increased thrombotic risk and increased bleeding risk^[71-74]. Although low-molecular-weight heparins (LMWHs) and DOACs remain standard options, their use may be complicated by bleeding, particularly in patients with gastrointestinal or genitourinary malignancies. This unmet need has made FXI/FXIa inhibitors an attractive investigational strategy in CAT. Two Phase III trials, MAGNOLIA (NCT05171075) and ASTER (NCT05171049), are evaluating whether abelacimab can improve the safety profile of anticoagulation in cancer patients at increased bleeding risk.

Central venous catheter-associated thrombosis is another important complication in patients with cancer. In a Phase II trial, gruticibart was associated with a lower incidence of catheter-related thrombosis than placebo, without an apparent increase in bleeding in the studied population^[57] .

These findings suggest that FXI/FXIa inhibition may address an important unmet need in selected CAT populations. If ongoing Phase III trials confirm favorable efficacy and safety, FXI/FXIa inhibitors may become a useful option for selected cancer patients at high bleeding risk.

Patients with ESKD undergoing maintenance hemodialysis face both extracorporeal circuit thrombosis and increased bleeding risk. This clinical context makes anticoagulation difficult and often requires individualized risk assessment^[75].

AB023 (Xisomab 3G3) provided early clinical evidence supporting FXI-pathway inhibition in hemodialysis. In a Phase II trial (NCT03612856), a single pre-dialysis bolus enhanced circuit patency and achieved a 42% reduction in TAT complexes without any drug-related bleeding events^[25]. Similarly, fesomersen, an ASO that reduces hepatic FXI synthesis, also produced dose-dependent FXI reduction in dialysis patients (NCT04534114)^[28]. A Phase II randomized, double-blind, placebo-controlled study of IONIS-FXIRx/fesomersen showed consistent pharmacokinetic exposure in relation to dialysis timing and no apparent accumulation after repeated dosing (NCT02553889)^[59]. A Phase IIb evaluation of osocimab reported bleeding rates comparable to placebo and acceptable tolerability, but larger studies are needed to confirm clinical safety (NCT04523220)^[58]. Among oral agents, asundexian showed a predictable pharmacokinetic profile across different degrees of renal impairment, with exposure largely unaffected by dialysis (NCT04510987; EudraCT 2020-000626-25/022-000196-38)^[60].

However, the evidence remains mixed. A Phase II trial of the monoclonal antibody MK-2060 achieved substantial FXI inhibition but did not significantly improve arteriovenous graft patency and was associated with increased access-related bleeding (NCT05027074)^[76].

Ongoing investigations, including the EMERALD study (NCT03358030) evaluating ISIS 416858 and a Phase I/II trial of milvexian (NCT03000673), may provide additional information on dosing, pharmacokinetics, and safety in this high-risk population. Overall, hemodialysis is a biologically plausible setting for FXI/FXIa inhibition because of contact activation within extracorporeal circuits. However, mixed clinical results indicate that efficacy and bleeding risk may differ by molecule, endpoint, and patient population.

Patients with ACS continue to harbor a significant residual thrombotic risk despite standard-of-care dual antiplatelet therapy (DAPT)^[77]. Because conventional anticoagulation added to DAPT increases bleeding risk, FXIa inhibitors have been explored as a potential adjunctive strategy. The PACIFIC-AMI trial (NCT04304534) evaluated the addition of asundexian (10 mg, 20 mg, or 50 mg daily) to DAPT in patients following acute myocardial infarction. The study showed dose-dependent pharmacodynamic inhibition of FXIa, with the 50 mg dose achieving more than 90% inhibition. All three doses achieved pharmacodynamic inhibition without an apparent increase in bleeding compared with placebo^[61]. However, the favorable bleeding signal did not translate into a reduction in ischemic events, suggesting that efficacy in ACS may depend not only on the degree of FXIa inhibition but also on the platelet-driven and high-shear nature of arterial thrombosis. This uncertainty was further reinforced by the early termination of LIBREXIA-ACS for futility. The trial evaluated whether milvexian added to standard antiplatelet therapy could reduce major adverse cardiovascular events (MACE) (NCT05754957)^[52]. An independent interim analysis concluded that the trial was unlikely to meet its primary efficacy endpoint, leading to its early cessation.

Overall, PACIFIC-AMI suggests that FXIa inhibition can be added to DAPT with acceptable bleeding in a Phase II setting, but it did not demonstrate clinical efficacy. Together with the futility termination of LIBREXIA-ACS, these findings raise the possibility that FXI/FXIa inhibition may be less effective in platelet-rich, high-shear arterial thrombosis than in venous or contact pathway-mediated thrombosis.

Non-cardioembolic ischemic stroke and TIA are associated with substantial risks of recurrence and long-term disability. Antithrombotic therapy is central to secondary prevention, but recurrent ischemic risk must be balanced against intracranial bleeding risk, particularly in older or high-risk patients^[78].

The PACIFIC-Stroke trial (NCT04304508) evaluated the oral FXIa inhibitor asundexian in patients with NCEIS already receiving background antiplatelet therapy. At 26 weeks, asundexian did not significantly reduce the composite primary endpoint across all dose cohorts compared with placebo^[63]. Similarly, the AXIOMATIC-SSP trial (NCT03766581) for milvexian did not establish a clear dose-response relationship for its primary endpoint; however, a directional trend toward reduced recurrence was observed in specific dose groups without a significant increase in major bleeding^[64]. These Phase II findings suggest low bleeding rates but uncertain efficacy. Future studies need to determine whether patient selection, dose intensity, or the biological contribution of FXI/FXIa to cerebrovascular thrombosis explains these mixed results. Several Phase III trials are ongoing to further evaluate this question. The OCEANIC-STROKE trial (NCT05686070) is evaluating whether asundexian added to antiplatelet therapy can reduce recurrent stroke. In parallel, the LIBREXIA-STROKE trial (NCT05702034) is evaluating milvexian added to antiplatelet therapy for secondary stroke prevention. The results of these ongoing trials may help clarify whether FXIa inhibition has a role in secondary prevention after NCEIS or TIA.

FXI/FXIa inhibition has also been investigated in high-acuity settings, although evidence remains very limited. EP-7041/frunexian is a short-acting parenteral FXIa inhibitor developed for settings requiring rapid titration. However, the Phase II trial (NCT05040776) in critically ill COVID-19 patients was withdrawn, and no clinical efficacy data are available from this program^[46].

In mechanical circulatory support (MCS) settings such as ECMO and left ventricular assist device (LVAD), FXI/FXIa inhibition is biologically plausible because contact activation and artificial surfaces contribute to thrombosis. Preclinical baboon models suggest that FXI targeting can reduce fibrin deposition and circuit thrombosis^[65]. Nevertheless, translating these findings to the ICU remains challenging, as device-related thrombosis is a multifactorial process involving high shear stress, platelet activation, and complement responses beyond the contact pathway. Carefully designed early-phase trials are needed to define feasibility, dosing, safety, and efficacy signals in these complex high-risk settings.

Recent failed or neutral trials highlight that FXI/FXIa inhibition should not be considered a homogeneous substitute for established anticoagulants. First, a reduction in bleeding does not necessarily translate into non-inferior antithrombotic efficacy. This is particularly evident in AF, where OCEANIC-AF was terminated early because asundexian was less effective than apixaban for prevention of stroke or systemic embolism. Second, the role of FXI/FXIa may differ across thrombotic mechanisms. In venous thrombosis or extracorporeal circuit-related thrombosis, contact activation and thrombin amplification may be more relevant, whereas ACS and non-cardioembolic stroke are more platelet-driven and occur under high-shear arterial conditions. Third, drug-specific factors, including molecular modality, dose, degree of FXIa inhibition, onset and offset, renal clearance, and prior anticoagulant exposure, may strongly influence trial outcomes. Finally, patient selection remains unresolved. In addition to exploratory biomarkers, clinical high-bleeding-risk frameworks may help guide future trial design and patient stratification, particularly in populations in whom the intensity or duration of antithrombotic therapy must be balanced against bleeding complications. Recent reviews of antithrombotic therapy in high-bleeding-risk patients undergoing cardiac and noncardiac percutaneous interventions further emphasize the importance of individualized bleeding-risk assessment and treatment modulation^[79,80]. Potential biomarkers such as FXI activity, FXI antigen levels, thrombin generation assays, contact activation markers, FXIa-antithrombin complexes, D-dimer, and NET-related markers may help identify responsive subgroups, but these remain exploratory and are not yet established for routine clinical decision-making.