Asundexian edges out milvexian: superior stroke reduction with safer bleeding profile in factor XIa inhibition – a network meta-analysis revolutionizing secondary prevention
Z Khan, S Munir, A Ahmed, R U Haque, M Maity, M N Elshabrawi, K Keswani, E Alupei, D S Rajawat, F Khalid, M Pasha, A Bakhsh, R U Mannam, S Roy, A DasAbstract
Background
Factor XIa (FXIa) inhibitors represent a novel class of anticoagulants targeting secondary stroke prevention in patients with non-cardioembolic ischemic stroke. Asundexian and milvexian have emerged as leading candidates, offering potential efficacy with improved bleeding profiles compared to traditional anticoagulants. However, optimal dosing strategies and comparative effectiveness across different FXIa inhibitors remain uncertain.
Purpose
To conduct a comprehensive network meta-analysis (NMA) comparing the efficacy and safety of asundexian and milvexian at various doses for secondary stroke prevention, employing advanced statistical methods including dose-response analysis, treatment rankings, and clinical impact metrics.
Methods
We performed a systematic NMA and dose response analysis of three randomized controlled trials encompassing 63 treatment arms across nine treatment groups: asundexian (10/20/50mg QD), milvexian (25/50/100/200mg BID, 25mg QD), and placebo. Six outcomes were analyzed: ischemic stroke, all-cause mortality, composite cardiovascular endpoints, major bleeding, intracranial hemorrhage, and covert brain infarction. Advanced analyses included dose-response meta-analysis, meta-regression with dose as moderator, subgroup analyses by drug type, leave-one-out sensitivity analyses, Calculation of Surface Under the Cumulative Ranking (SUCRA) scores, Numbers Needed to Treat/Harm (NNT/NNH), and benefit-risk ratios. Heterogeneity was assessed using Cochran's Q and I² statistics.
Results
Asundexian demonstrated superior stroke reduction (pooled RR=0.762, 95%CI: 0.678-0.857; 24% reduction) compared to milvexian (RR=0.872, 95%CI: 0.673-1.131; 13% reduction). For major bleeding, asundexian showed a modest 15% increase (RR=1.152, 95%CI: 1.000-1.328) versus milvexian's 103% increase (RR=2.033, 95%CI: 1.073-3.850). Treatment-level analysis revealed milvexian 100mg BID had the lowest NNT for stroke prevention (NNT=34), while asundexian 20mg QD demonstrated the highest safety margin (NNH=221 for bleeding). SUCRA rankings favored low-dose milvexian regimens (25mg QD/BID: SUCRA=75.0), though this conflicted with safety profiles. No heterogeneity was detected (I²=0% across outcomes, p>0.43). Dose-response analysis revealed paradoxical findings: milvexian showed significant positive dose-response for composite endpoints (p=0.032), indicating increased risk with higher doses.
Conclusion
Asundexian demonstrates superior benefit-risk balance for secondary stroke prevention, with greater stroke reduction and substantially lower bleeding risk than milvexian. Despite higher SUCRA scores for low-dose milvexian, clinical decision-making should prioritize safety profiles. Dose escalation strategies are not supported by current evidence and may increase harm. Asundexian at moderate doses (20-50mg QD) represents the optimal therapeutic approach, balancing efficacy and safety.visual abstractsucra radar