EINSTEIN Jr
Rivaroxaban compared with standard anticoagulants for the treatment of acute venous thromboembolism in children: a randomised, controlled, phase 3 trial
Clinical Question
Is rivaroxaban as effective and safe as standard anticoagulants for the treatment of acute venous thromboembolism (VTE) in children?
Bottom Line
In 500 children (0-17yr) with acute VTE, bodyweight-adjusted rivaroxaban resulted in similar recurrent VTE (1% vs 3%; HR 0.40; 0.11-1.41) vs standard anticoagulation, with no increase in major bleeding (0% vs 1%). Rivaroxaban produced significantly greater clot resolution (complete 38% vs 26%; OR 1.70; P=0.012). First completed phase 3 DOAC trial in children. Oral suspension developed for young children.
Major Points
- Recurrent VTE: 1% (4/335) rivaroxaban vs 3% (5/165) standard (HR 0.40; 0.11-1.41).
- No major bleeding in rivaroxaban arm (0%) vs 2 in comparator (1%). CRNM bleeding 3% vs 1%.
- Significantly greater clot resolution: complete 38% vs 26% (OR 1.70; 95% CI 1.11-2.58; P=0.012).
- Net clinical benefit (recurrent VTE + major bleed): 1% vs 4% (HR 0.30; 0.08-0.93).
- First completed phase 3 DOAC trial in children — largest pediatric anticoagulation trial (500 children, 107 hospitals, 28 countries).
- Novel oral suspension (1 mg/mL) for children <6yr eliminates parenteral anticoagulation/lab monitoring.
- Bodyweight-adjusted dosing targeting adult 20mg-equivalent exposure: once/twice/thrice daily based on weight.
- 97% first-episode VTE; 51% non-catheter-related, 27% catheter-related, 22% cerebral vein/sinus thrombosis.
- Not formally powered for non-inferiority (upper CI 1.41 exceeds standard margins).
- Bayer/Janssen funded. Led to regulatory approval of rivaroxaban for pediatric VTE.
Design
Study Type: Randomized, open-label, multicenter, active-controlled, phase 3 trial
Randomization: 1
Blinding: Open-label with blinded outcome adjudication
Enrollment Period: July 2014 – Feb 2018
Follow-up Duration: 3 months (6 months in children <2y with catheter-related VTE)
Centers: 107
Countries: Europe, North America, South America, Asia
Sample Size: 335
Analysis: Intention-to-treat and per-protocol analysis; time-to-event using Cox regression
Inclusion Criteria
- Age 0–17 years with documented acute VTE (DVT, PE, or cerebral sinus thrombosis)
- Stable after initial parenteral anticoagulation (5–9 days)
- Weight-adjusted rivaroxaban dosing feasible
Exclusion Criteria
- Active bleeding or bleeding risk
- Platelet count <50 x 10⁹/L
- Severe renal or hepatic impairment
- Known contraindication to anticoagulation
- Planned thrombolysis or surgery
- Participation in other trials
Arms
| Field | Rivaroxaban | Control |
|---|---|---|
| Intervention | Oral rivaroxaban (tablets or suspension), dosed by body weight to match adult exposure | Unfractionated heparin, LMWH, or fondaparinux followed by VKAs (e.g., warfarin) |
| Duration | 3 months (or 6 months for children <2y with catheter-related thrombosis) | 3 or 6 months |
Outcomes
| Outcome | Type | Control | Intervention | HR / OR / RR | P-value |
|---|---|---|---|---|---|
| Symptomatic recurrent VTE | Primary | 3/121 (2.5%) | 4/335 (1.2%) | 0.33 | NS |
| Change in thrombotic burden (imaging) | Secondary | 80% | 81% | NS | |
| Major Bleeding | Adverse | 1/335 (0.3%) vs 1/121 (0.8%) | |||
| Clinically relevant non-major bleeding | Adverse | 9 (2.7%) vs 2 (1.7%) | |||
| Any bleeding | Adverse | 72/335 (21%) vs 21/121 (17%) |
Subgroup Analysis
Treatment effect consistent across age groups (0–2y, 2–12y, 12–17y), VTE type, and regions [oai_citation:1‡EINSTEIN Jr.pdf](file-service://file-9bRknGZZ88FShsh5xVQ8x9)
Criticisms
- Open-label design could introduce bias despite blinded adjudication
- Relatively short treatment/follow-up duration
- Event rates were low, limiting statistical power for rare outcomes
- No power to assess superiority
Funding
Bayer and Janssen
Based on: EINSTEIN Jr (The Lancet Haematology, 2019)
Authors: Christoph Male, Anthonie Lensing, Joseph Palumbo, ..., Marcela Torres
Citation: Lancet Haematol. 2019 Jan;6(1):e89–e98
Content summarized and formatted by NeuroTrials.ai.