Risk factors for recurrence and major bleeding in patients with cancer-associated venous thromboembolism.

Risks of recurrence and treatment-emergent bleeding are high in patients with cancer-associated venous thromboembolism (VTE) but factors associated with these risks remain substantially undefined. The aim of this analysis in patients with cancer-associated VTE included in the Caravaggio study was to identify risk factors for recurrent VTE and major bleeding. Variables potentially predictive for recurrent VTE or major bleeding were evaluated in a Cox proportional hazard multivariable analysis with backward variable selection. Recurrent VTE occurred in 78 patients (6.8%) and major bleeding in 45 (3.9%). Independent risk factors for recurrent VTE were deep vein thrombosis (DVT) as index event (Hazard ratio (HR) 1.84, 95% CI 1.17-2.88), ECOG status of 1 or more (HR 1.95, 95% CI 1.11-3.43), pancreatic or hepatobiliary cancer site (HR 2.20, 95% CI 1.19-4.06), concomitant anti-cancer treatment (HR 1.98, 95% CI 1.03-3.81) and creatinine clearance (HR 1.10, 95% CI 1.00-1.20 for every 10 ml/min absolute increase). Independent risk factors for major bleeding were ECOG status of 2 (HR 2.31, 95% CI 1.24-4.29), genitourinary cancer site (HR 2.72, 95% CI 1.28-5.77), upper gastrointestinal cancer site (HR 3.17, 95% CI 1.22-8.23), and non-resected luminal gastrointestinal cancer (HR 2.77, 95% CI 1.38-5.56). This analysis of the Caravaggio study in patients with cancer-associated VTE who were on standardized anticoagulant treatment identified five independent predictors for recurrent VTE and four independent predictors of treatment-emergent major bleeding. Considering these risks could help clinicians to optimize the anticoagulant treatment in patients with cancer-associated VTE.