Cancer and thrombosis: Improvements in strategies for prediction, diagnosis, and treatment

Venous thromboembolism, which comprises deep vein thrombosis and pulmonary embolism, is a common complication in cancer patients and is associated with substantial morbidity and mortality. This thesis aims to provide innovative, personalized strategies to prevent, adequately diagnose, and effectively and safely treat cancer-associated venous thromboembolism. In Part I, we demonstrate that current risk scores for cancer-related venous thromboembolism perform suboptimal. We therefore introduce a new, simple clinical prediction model based on tumor type and D-dimer level, which could aid clinicians in selecting high-risk patients for thromboprophylaxis. In Part II, we show that pulmonary embolism can be safely ruled out without additional imaging in a larger proportion of patients when applying an age-adjusted D-dimer threshold instead of the conventional fixed threshold. In Part II we also show that one in every twenty patients with unprovoked venous thromboembolism is diagnosed with cancer within a year, which calls for vigilant occult cancer screening in these patients. Finally, in Part III, the optimal treatment for venous thromboembolism is evaluated. Based on a meta-analysis including 27,000 patients with acute deep vein thrombosis or pulmonary embolism, direct oral anticoagulants appear to be as effective as vitamin K antagonists in preventing recurrence, but they are associated with significantly less major bleeding. This finding is consistent in patients with active cancer, although low-molecular-weight heparin remain the recommended treatment. The ongoing Hokusai VTE-cancer study, of which the rationale and design are discussed in Part III, will evaluate whether direct oral anticoagulants are also safe and effective for treating cancer-associated venous thromboembolism.