Anticoagulation in ambulatory cancer patients with no indication for prophylactic or therapeutic anticoagulation

Abstract

SummaryVenous thromboembolism (VTE) occurs frequently in cancer patients and is a major cause of mortality. Although anticoagulation (AC) may have anti-tumour effects, it is recommended in cancer patients only for indications unrelated to these effects. A Markov model was constructed to estimate when prophylactic AC, given during four months of chemotherapy following a new cancer diagnosis, would be economically reasonable in ambulatory cancer patients with no VTE history. Risks and costs of major and minor bleeding, VTE, and death were obtained from the literature, including a 2011 Cochrane review of AC in cancer inpatients with no conventional indication. In the model, AC was low-molecular-weight-heparin (LMWH), and effects were modeled over a 24-month period. Model outputs were quality-adjusted life years (QALY) and the incremental cost-effectiveness ratio (ICER). In the base case analysis, when prophylactic AC had a two-year relative mortality risk of 0.92, there was a gain of 0.0354 QALY, for an ICER of $90,893/QALY gained, compared with no AC. In sensitivity analyses, AC prophylaxis cost less than $100,000/QALY, if two-year mortality exceeded 75%, or if AC costs were less than $1,076 per month, or if LMWH relative mortality risk was less than 0.927. Results were not sensitive to variation in VTE relative risk on AC, nor to major or minor bleeding risk on AC. A probabilistic sensitivity analysis also favoured prophylactic AC. In conclusion, prophylactic LMWH given to decrease cancer-related mortality, with no conventional indication, appears economically reasonable if its suggested mortality benefit is confirmed in future trials.