The risk of recurrent venous thromboembolism after discontinuation of anticoagulant therapy in patients with cancer-associated thrombosis: a systematic review and meta-analysisResearch in context

Background: The optimal duration of anticoagulation in patients with active cancer and venous thromboembolism (VTE) is unknown. Current clinical guidelines advocate anticoagulant therapy for 3–6 months and to continue anticoagulant therapy for as long as the cancer is active. However, an adequate systematic review on the rate of recurrent VTE after discontinuation of anticoagulant therapy has not been performed. Methods: For this systemic review and meta-analysis, we searched Embase.com, Medline (Ovid), Web of Science, Cochrane Library, and Google Scholar, from database inception to February 16, 2023, for studies on anticoagulant therapy in patients with cancer and the recurrence of venous thromboembolism after discontinuation of this therapy. We included randomised controlled trials and cohort studies published in English that reported on patients who met the following: cancer and a first VTE, completed at least 3 months of anticoagulant therapy, were followed after discontinuation of anticoagulant therapy, and with symptomatic recurrent VTE as an outcome during follow-up. Study-level data were requested from study authors. The primary outcome was the rate of recurrent VTE after discontinuation of anticoagulant therapy. A Bayesian random-effects meta-analysis was used to estimate the rate of recurrent VTE per 100 person-years for the pooled studies at different time intervals after discontinuation of anticoagulation therapy. We also calculated the cumulative VTE recurrence rate at different time intervals. Forest plots were mapped and the results were summarized by the median and 95% credible interval (CIs). This study was registered with PROSPERO, CRD42021249060. Findings: Of 3856 studies identified in our search, 33 studies were identified for inclusion. After requesting study-level data, 14 studies involving 1922 patients with cancer-associated thrombosis were included. The pooled rate of recurrent VTE per 100 person-years after discontinuation of anticoagulant therapy was 14.6 events (95% credible interval 6.5–22.8) in the first three months, decreasing to 1.1 events (95% CI 0.3–2.1) in year 2–3, and 2.2 events (95% CI 0.0–4.4) in year 3–5 after discontinuation of anticoagulant therapy. The cumulative VTE recurrence rate was 28.3% (95% CI 15.6–39.6%) at 1 year; 31.1% (95% CI 16.5–43.8%) at 2 years; 31.9% (95% CI 16.8–45.0%) at 3 years; and 35.0% (95% CI 16.8–47.4%) at 5 years after discontinuation of anticoagulant therapy. Interpretation: This meta-an