Two Factor XI Antibodies For the Prevention of Post- operative VTE: Results from ROXI-VTE-I and ROXI-VTE-II Trials

Key Points:

• ROXI-VTE-I and ROXI-VTE-II evaluated two distinct factor XI monoclonal antibodies REGN9933^A2 and REGN7508^Cat, for prevention of venous thromboembolism (VTE) after elective knee replacement.
• The Apple II domain-binding antibody (REGN9933^A2) blocks factor XI activation by factor XIIa, while the catalytic domain-binding antibody (REGN7508 ^Cat) inhibits activation by both factor XIIa and thrombin.
• REGN7508Cat was superior to enoxaparin for venous thromboembolism prevention; REGN9933A2 was not superior to enoxaparin. The findings with REGN7508Cat confirm the role of FXI in postoperative venous thromboembolism, and the findings with REGN9933A2 suggest that factor XIIa-driven factor XI activation contributes to this process.

While factor XI has recently emerged as a promising target for safer anticoagulation, its precise contribution to postoperative thrombosis remains unclear. Historically, new anticoagulants are first tested in orthopedic surgery patients, who are at high risk for postoperative deep vein thrombosis (DVT) detectable by routine venography. At the American Heart Association 2025 Scientific Sessions, Dr. Jeffrey Weitz (McMaster University) presented the ROXI-VTE-I and ROXI-VTE-II phase 2 trials, with simultaneous publication in Lancet, which further illuminate the mechanistic role of factor XI activation in the postoperative setting.

ROXI-VTE-I (NCT05618808) and ROXI-VTE-II (NCT06454630) were consecutive, open-label, randomized phase 2 trials in patients undergoing elective knee arthroplasty. In ROXI-VTE-I, 373 patients were randomized 1:1:1 to receive REGN9933^A2, enoxaparin, or exploratory comparator apixaban. In ROXI-VTE-II, 170 patients were randomized 2:1 to receive REGN7508^Cat or enoxaparin. The primary efficacy endpoint in both trials was objectively confirmed VTE (symptomatic or venographically detected DVT or pulmonary embolism) by day 12. The primary safety endpoint was major or clinically relevant non-major bleeding.

In ROXI-VTE-I, VTE occurred in 17% of patients treated with REGN9933^A2 compared with 22% receiving enoxaparin and 12% with apixaban. Although the difference did not meet prespecified superiority criteria, the lower point estimate versus enoxaparin suggested biologic activity even when only factor XIIa–mediated activation of XI was inhibited. In ROXI-VTE-II, VTE occurred in 7% of patients receiving REGN7508^Cat compared with 17% of those on enoxaparin, meeting superiority criteria. No major or clinically relevant non-major bleeds occurred in either study. A pooled analysis of enoxaparin arms (n≈21%) confirmed the superior efficacy of 75-08 and showed that 99-33 achieved non-inferior protection with a potential safety margin.

According to Dr. Weitz, “These two antibodies allow us to tease apart whether postoperative clotting is driven by thrombin feedback or factor XIIa activation. We’ve learned that factor XIIa–mediated activation of XI does contribute to postoperative VTE, and the catalytic-domain antibody provides superior efficacy with no bleeding signal.” He also noted that ongoing trials—including ROXI-CATH (central venous catheter thrombosis) and ROXI-APEX (phase 3 orthopedic surgery trial versus aspirin)—will clarify the clinical scope and optimal application of these distinct mechanisms.

In conclusion, the ROXI-VTE program demonstrates that factor XI inhibition, particularly via catalytic-domain blockade, significantly reduces postoperative VTE without excess bleeding. These findings not only validate factor XI as a therapeutic target but also deepen mechanistic understanding of thrombus formation after surgery.