Background
Phase 3, double-blind, placebo-controlled RCT (AMPLIFY-EXT). 2,486 patients with proximal DVT or PE who had completed 6–12 months of anticoagulation therapy and for whom there was clinical equipoise regarding continuation or cessation. Patients were randomized 1:1:1 to apixaban 2.5 mg BID, apixaban 5 mg BID, or placebo for 12 months. The trial addressed the critical clinical question of extended VTE treatment, where indefinite anticoagulation reduces recurrence but full-dose therapy carries continued bleeding risk; the 2.5 mg dose was hypothesized to provide sufficient antithrombotic effect with lower bleeding than the 5 mg treatment dose.
Interventions and follow up
Arm A: Apixaban 2.5 mg orally twice daily × 12 month
Arm B: Apixaban 5 mg orally twice daily × 12 month
Arm C: Placebo orally twice daily × 12 month
Primary endpoint: Symptomatic recurrent VTE or death from VTE
mFollow up: 12 month
Arm B: Apixaban 5 mg orally twice daily × 12 month
Arm C: Placebo orally twice daily × 12 month
Primary endpoint: Symptomatic recurrent VTE or death from VTE
mFollow up: 12 month
Results
Recurrent VTE or VTE death (apixaban 2.5 mg vs placebo): 1.7% vs 8.8%, difference 7.2 pp (95% CI 5.0–9.3), P<.001
Recurrent VTE or VTE death (apixaban 5 mg vs placebo): 1.7% vs 8.8%, difference 7.0 pp (95% CI 4.9–9.1), P<.001
Major bleeding: 0.2% (2.5 mg), 0.1% (5 mg) vs 0.5% (placebo) — not increased
Clinically relevant nonmajor bleeding: 3.0% (2.5 mg), 4.2% (5 mg) vs 2.3% (placebo)
All-cause mortality: 0.8% (2.5 mg), 0.5% (5 mg) vs 1.7% (placebo)
Recurrent VTE or VTE death (apixaban 5 mg vs placebo): 1.7% vs 8.8%, difference 7.0 pp (95% CI 4.9–9.1), P<.001
Major bleeding: 0.2% (2.5 mg), 0.1% (5 mg) vs 0.5% (placebo) — not increased
Clinically relevant nonmajor bleeding: 3.0% (2.5 mg), 4.2% (5 mg) vs 2.3% (placebo)
All-cause mortality: 0.8% (2.5 mg), 0.5% (5 mg) vs 1.7% (placebo)
Adverse events
Main adverse events: Major bleeding was not increased above placebo at either dose — a key finding. Grade ≥3 AEs were similar across arms. No significant excess in hepatotoxicity or thrombocytopenia. Clinically relevant nonmajor bleeding was slightly higher with both active doses, primarily mucocutaneous (bruising, epistaxis). Discontinuation rates similar across arms.
Conclusions
Extended anticoagulation with apixaban at both a treatment dose (5 mg BID) and a thromboprophylactic dose (2.5 mg BID) reduced recurrent VTE by ~80% versus placebo without increasing major bleeding, establishing both doses as effective and safe options for extended VTE treatment.
Key Limitations
Key Limitations: Placebo-controlled rather than active-comparator — does not directly compare apixaban extended therapy to extended warfarin or other DOACs. The 2.5 mg BID dose was not separately compared against 5 mg BID in a powered head-to-head analysis; both showed equivalent primary endpoint rates (both 1.7%). Generalizability: patients had already completed 6–12 months of standard therapy and were in clinical equipoise — enriched for those with uncertain long-term bleeding risk. Duration of 12 months is longer than most clinical practice extended-phase prescribing but shorter than indefinite anticoagulation. Post-trial recurrence after stopping extended therapy was not captured.
Clinical Context
AMPLIFY-EXT established apixaban 2.5 mg BID as the preferred extended-treatment dose for VTE after completing initial 6-month therapy — offering significant VTE reduction without excess major bleeding. This lower dose is now broadly used for extended-phase VTE treatment in patients with moderate recurrence risk (first unprovoked VTE, no ongoing strong risk factor). The comparison with EINSTEIN-CHOICE (rivaroxaban 10 mg once daily) and WARFASA/ASPIRE (aspirin) informs stratified decision-making: anticoagulation at reduced dose outperforms aspirin; direct comparisons between DOACs at reduced doses have not been done. Guidelines (ACCP, ASH, ISTH) recommend extended-phase anticoagulation for unprovoked VTE if bleeding risk is acceptable, with DOAC preferred over warfarin.
References
References: Agnelli G et al, NEJM 2013 (AMPLIFY-EXT primary)