Background
Noninferiority phase III RCT, N=190, antiphospholipid syndrome (61% high-risk/triple positive) with prior arterial or venous thrombosis. Exclusions: pregnancy, CrCl <30 mL/min, Child-Pugh B/C cirrhosis or LFT >3×ULN, high bleeding risk, CYP3A4 inducers.
Interventions and follow up
Arm A: rivaroxaban 20 mg daily (15 mg if CrCl 30-50 mL/min)
Arm B: warfarin, INR 2.0-3.0
Primary endpoint: proportion of patients with a new thrombotic event during the study
Arm B: warfarin, INR 2.0-3.0
Primary endpoint: proportion of patients with a new thrombotic event during the study
Results
Per-protocol, all recurrent thrombosis (A vs B): 11.6% vs 6.3%, RR 1.83, 95%CI 0.71-4.76, noninferiority P=.29; superiority of B P=.20
Per-protocol arterial thrombosis: 10.5% vs 3.2%, RR 3.33, 95%CI 0.95-11.73, P=.060
Per-protocol venous thrombosis: 2.1% vs 3.2%, RR 0.67, 95%CI 0.11-3.90, P=.065
ITT all recurrent thrombosis: 12.6% vs 6.3%, RR 2.0, 95%CI 0.78-5.11, noninferiority P=.57; superiority of B P=.13
ITT arterial thrombosis: 11.6% vs 3.2%, RR 3.67, 95%CI 1.06-12.73, P=.040
ITT venous thrombosis: 2.1% vs 3.2%, RR 0.67, 95%CI 0.11-3.90, P=.065
Major bleeding: 6.3% vs 7.4%, RR 0.86, 95%CI 0.30-2.46, P=.77
Per-protocol arterial thrombosis: 10.5% vs 3.2%, RR 3.33, 95%CI 0.95-11.73, P=.060
Per-protocol venous thrombosis: 2.1% vs 3.2%, RR 0.67, 95%CI 0.11-3.90, P=.065
ITT all recurrent thrombosis: 12.6% vs 6.3%, RR 2.0, 95%CI 0.78-5.11, noninferiority P=.57; superiority of B P=.13
ITT arterial thrombosis: 11.6% vs 3.2%, RR 3.67, 95%CI 1.06-12.73, P=.040
ITT venous thrombosis: 2.1% vs 3.2%, RR 0.67, 95%CI 0.11-3.90, P=.065
Major bleeding: 6.3% vs 7.4%, RR 0.86, 95%CI 0.30-2.46, P=.77
Adverse events
Major bleeding (n): 6 (11.3%) rivaroxaban vs 7 (12.5%) warfarin, RR 0.90
Thrombotic events: more frequent with rivaroxaban (11.6% vs 6.3%, RR 1.83)
Stroke: numerically higher with rivaroxaban (9 vs 0)
Thrombotic events: more frequent with rivaroxaban (11.6% vs 6.3%, RR 1.83)
Stroke: numerically higher with rivaroxaban (9 vs 0)
Conclusions
Rivaroxaban did not demonstrate noninferiority to warfarin in thrombotic APS and was associated with more arterial thrombosis, particularly stroke. Warfarin remains preferred.
Key Limitations
Open-label; modest N with wide confidence intervals; mixed-risk APS population (61% triple positive); noninferiority not met, precluding firm equivalence conclusions.
Clinical Context
Together with TRAPS, this trial supports vitamin K antagonists over DOACs for thrombotic APS. ISTH and EULAR guidance recommends against rivaroxaban in APS, especially triple-positive disease and prior arterial events.