TRAPS

Published

Pengo V, et al. "Rivaroxaban vs. warfarin in high-risk patients with antiphospholipid syndrome". Blood. 2018. 132(13):1365-1371.
PubMed • Full text

Clinical Question

In high-risk patients with antiphospholipid syndrome, is rivaroxaban non-inferior to warfarin for the prevention of thromboembolic events, major bleeding, and vascular death?

Bottom Line

Use of rivaroxaban in triple-positive, high-risk antiphospholipid syndrome is associated with an increased risk of thromboembolic events and major bleeding as compared to warfarin.

Major Points

Rivaroxaban has been found to be noninferior to warfarin in the treatment of nonvalvular atrial fibrillation (ROCKET AF) and noninferior to warfarin in the prevention of recurrent VTE or bleeding in patients with acute PE (EINSTEIN-PE), but the evidence regarding the safety and efficacy of its use in patients with antiphospholipid syndrome (APLS) had not been rigorously studied. The 2016 RAPS study evaluated rivaroxaban among 116 adults with low-risk APLS and studied a surrogate endpoint of endogenous thrombin potential. There were too few events to determine clinical efficacy of rivaroxaban among this population. Demonstrating equivalence or superiority of rivaroxaban in higher-risk adults would provide clinical evidence for the use of a factor Xa inhibitor in APLS.

Published in 2018, the Trial on Rivaroxaban in AntiPhospholipid Syndrome (TRAPS) randomized 120 adults with APLS with high-risk disease to rivaroxaban or warfarin. High-risk disease was defined as triple-positive laboratory assessment (lupus anticoagulant, cardiolipin antibodies, and beta-2 glycoprotein antibodies) plus prior clinical activity. The primary outcome was thrombosis, major bleeding, or vascular mortality, which were noted to occurred more frequently in the rivaroxaban group during an interim analysis (22% vs. 3%; HR 7.4; 95% CI 1.7-32.9; P=0.008). The trial was stopped early.

TRAPS provides strong evidence against the routine use of factor Xa inhibitors in high-risk APLS. A 2019 non-inferiority Spanish trial by Ordi-Ros compared rivaroxaban to a vitamin K antagonist (VKA) in a population that was 61% triple-positive APLS. The recurrence of thrombosis was 12.6% vs. 6.3% in rivaroxaban vs. VKA. The greatest number of additional arterial events, primarily stroke. Numbers were too small to determine difference in events by triple positivity. Whether a DOAC like rivaroxaban provides clinical benefit among patients with lower risk APLS is unknown.

Guidelines

As of September 2018, no guidelines have been published that reflect the results of this trial.

Design

• Randomized, open-label, parallel assignment, multicenter, non-inferiority study with blinded end-point adjudication
• N=120
  • Rivaroxaban (20 mg or 15 mg once daily) (n=59)
  • Warfarin (INR target 2.5) (n=61)
• Setting: 14 centers in Italy
• Enrollment: 2014-2018
• Mean follow-up: 569 days in the as-treated cohort; 611 days in the intention-to-treat
• Analysis: As-treated and intention-to-treat
• Primary outcome: Thromboembolic events, major bleeding, or death

Population

Inclusion Criteria

• Triple-positive APS, defined by assays demonstrating lupus anticoagulant and same isotype antibodies for cardiolipin and β2-glycoprotein
• History of objectively proven arterial, venous, and/or biopsy-proven micro-thrombosis
• Age 18-75 years

Exclusion Criteria

• Contraindication to rivaroxaban
• CrCl <30 mL/min
• Pregnancy or breastfeeding
• Use of anticoagulants other than warfarin or medications with significant interaction
• High risk for bleeding (recent or planned surgery, GI bleed, CNS bleed, hemorrhagic disease, high-risk intracranial condition, SBP ≥180 mmHg)
• Liver cirrhosis or ALT greater than 3 times the upper limit of normal

Baseline Characteristics

• Demographics: Female 64%, age 46 years
• BMI: 28 kg/m²
• CrCl: 113.1 mL/min
• Autoimmune disease: 41%
• Previous thrombotic events: Arterial 21%, venous events 64%, venous and arterial events 15%, pregnancy morbidity 23%
• PMH: Smoking 50%, HTN 31%, DM 3%, HLD 23%, other hypercoagulable condition 15%
• Medications: Hydroxychloroquine 32%, corticosteroids 21%, ASA 18%, statin 14%

Interventions

• Randomized to a group:
  • Rivaroxaban - 20 mg daily for CrCl >50 mL/min; 15 mg daily for CrCl 30-50 mL/min
  • Warfarin - INR goal 2.5
  • Patients randomized to rivaroxaban who were already taking warfarin started treatment when the INR was below 3
  • Patients randomized to warfarin had INR checked at least every 4 weeks
• All patients had regularly scheduled visits at months 1 and 3 after randomization and every 6 months thereafter

Outcomes

Comparisons are rivaroxaban therapy vs. warfarin therapy.

Primary Outcomes

Thromboembolic events, major bleeding, or death

Intention-to-treat: 22% vs. 3% (HR 7.4; 95% CI 1.7-32.9; P=0.008)

As-treated: 19% vs. 3% (HR 6.7; 95% CI 1.5-30.5; P=0.01)

Secondary Outcomes

Ischemic stroke

Intention-to-treat: 7% vs. 0%

As-treated: 7% vs. 0%

Myocardial infarction

Intention-to-treat: 5% vs. 0%

As-treated: 5% vs. 0%

Venous thromboembolism

Intention-to-treat: 0% vs. 0%

As-treated: 0% vs. 0%

Major bleeding

Intention-to-treat: 7% vs. 3% (HR 2.3; 95% CI 0.4-12.5; P=0.3)

As-treated: 7% vs. 3% (HR 2.5; 95% CI 0.5-13.6; P=0.3)

Death

Intention-to-treat: 0% vs. 0%

As-treated: 2% vs. 0%

Subgroup Analysis

No subgroup analyses were performed.

Adverse Events

No adverse events other than the outcomes listed above were reported.

Criticisms

• Not blinded
• Stopped prematurely
• Unclear if INR was targeted using chromogenic factor X. Traditional INR assays may be unreliable in APLS.

Funding

The study was funded by Bayer, the co-developer of rivaroxaban. No conflicts of interest were disclosed by the study authors.