Background
Phase 3 open-label, non-inferiority RCT (Hokusai-VTE Cancer). 1,050 patients with cancer and acute VTE (DVT or PE). Randomized to oral edoxaban versus subcutaneous dalteparin (the prior standard per CLOT trial). Edoxaban given after ≥5 days of initial LMWH. Duration 6–12 months at investigator discretion. First major DOAC vs LMWH trial in cancer-associated VTE. Key secondary concern: risk of GI bleeding with DOACs in GI malignancies.
Interventions and follow up
Arm A: Edoxaban 60 mg once daily (or 30 mg if dose-reduction criteria met) for 6–12 months, following ≥5 days of initial heparin/LMWH
Arm B: Dalteparin 200 IU/kg once daily × 1 month, then 150 IU/kg once daily for 5–11 months (6–12 months total)
Primary endpoint: Composite of recurrent VTE or major bleeding — non-inferiority margin HR ≤1.50
mFollow up: 12 month
Arm B: Dalteparin 200 IU/kg once daily × 1 month, then 150 IU/kg once daily for 5–11 months (6–12 months total)
Primary endpoint: Composite of recurrent VTE or major bleeding — non-inferiority margin HR ≤1.50
mFollow up: 12 month
Results
Composite VTE recurrence or major bleeding: 12.8% (edoxaban) vs 13.5% (dalteparin); HR 0.97, 95% CI 0.70–1.36 — non-inferior (P=.006 for non-inferiority)
Recurrent VTE: 7.9% vs 11.3%; HR 0.71, 95% CI 0.48–1.06 (P=.09 for superiority)
Major bleeding: 6.9% vs 4.0%; HR 1.77, 95% CI 1.03–3.04 — P=.04 (higher with edoxaban)
GI major bleeding (edoxaban): 3.8% vs 0.9% — concentrated in GI cancer subgroup
Recurrent VTE: 7.9% vs 11.3%; HR 0.71, 95% CI 0.48–1.06 (P=.09 for superiority)
Major bleeding: 6.9% vs 4.0%; HR 1.77, 95% CI 1.03–3.04 — P=.04 (higher with edoxaban)
GI major bleeding (edoxaban): 3.8% vs 0.9% — concentrated in GI cancer subgroup
Adverse events
Main adverse events: Major bleeding significantly higher with edoxaban overall (6.9% vs 4.0%; P=.04), driven almost entirely by GI and GU cancers (upper GI malignancy: edoxaban major bleed 12.5% vs 3.8% dalteparin). Non-GI cancer patients had similar bleeding. Fatal bleeding: 0.4% each arm. GI major bleeding most concerning subset — GI/GU cancer patients had markedly elevated risk with edoxaban.
Conclusions
Edoxaban was non-inferior to dalteparin for the composite endpoint in cancer VTE but caused significantly more major bleeding, particularly in patients with GI malignancies, suggesting DOACs should be used cautiously in luminal GI cancers.
Key Limitations
Key Limitations: Open-label design. The composite primary endpoint (efficacy + safety combined) is unusual and makes clinical interpretation less intuitive. Edoxaban reduced recurrent VTE (numerically) but increased major bleeding — net benefit depends on individual cancer type and bleeding risk. The GI cancer bleeding signal is prominent but the subgroup was small. Variable treatment duration (6–12 months, not randomized). GI cancer patients may need to be identified prospectively to avoid DOAC harm.
Clinical Context
Hokusai-VTE Cancer (2018) was pivotal in shifting cancer-VTE guidelines from LMWH to DOACs, but with the critical caveat of GI cancer bleeding. CARAVAGGIO (apixaban) later showed no significant increase in major bleeding vs LMWH even in GI cancers, making apixaban preferred over edoxaban or rivaroxaban in most cancer-VTE settings. Current ASCO/NCCN guidelines recommend apixaban as preferred DOAC for cancer-VTE; edoxaban/rivaroxaban are alternatives for non-GI/GU cancers.
References