Predictors of recurrent venous thromboembolism and bleeding in patients with cancer: a meta-analysis

BACKGROUND AND AIMS: Patients with cancer and venous thromboembolism (VTE) have a high risk of recurrent VTE and anticoagulant-related bleeding. This study aimed to identify prognostic factors for these complications. METHODS: A systematic review was performed for randomized trials and cohort studies evaluating prognostic factors for recurrent VTE or anticoagulant-related bleeding in adult patients with cancer and VTE. Adjusted hazard ratios (aHRs) for factors were pooled using random-effects meta-analysis. The certainty of evidence was assessed using the Grading of Recommendations, Assessment, Development and Evaluation approach. RESULTS: Thirty-three studies (n = 96 753) were included in the meta-analyses. Factors with high certainty of association with increased risk of recurrent VTE included a previous history of VTE aHR 1.50 (95% CI 1.08-2.09), Eastern Cooperative Oncology Group (ECOG) performance status >0 1.81 (1.34-2.46) or >1 2.44 (1.55-3.84), advanced cancer 1.38 (1.15-1.65), and specific cancer sites including lung 1.78 (1.29-2.46), hepatobiliary 2.37 (1.70-3.30), pancreas 3.20 (2.06-4.96), and genitourinary 1.38 (1.14-1.67). Conversely, recent surgery aHR 0.56 (95% CI 0.40-0.76) and breast cancer 0.43 (0.23-0.81) had a high certainty of association with a decreased risk. Factors with a high certainty of association with an increased risk of anticoagulant-related bleeding included a history of bleeding aHR 2.41 (95% CI 1.50-3.88), ECOG performance status ≥2 2.10 (1.48-2.99), advanced cancer 1.60 (1.29-1.97), and cancers of the brain 2.25 (1.64-3.09), gastrointestinal system 1.74 (1.44-2.11), genitourinary system 1.90 (1.48-2.45), and prostate 1.72 (1.26-2.34). CONCLUSIONS: The prognostic factors identified in this meta-analysis should be considered as part of risk stratification frameworks for anticoagulation management in patients with cancer and VTE.