Risk prediction for cancer-associated thrombosis in ambulatory patients with cancer: past, present and future.

Venous thromboembolism (VTE) is a preventable disease, thus, in a number of clinical situations primary thromboprophylaxis has been proposed. Although we now know that cancer is one of the most important risk factors for VTE, primary prophylactic anticoagulation is only widely established for high-risk hospitalized patients and peri- and postoperatively after major cancer surgery. Long-term primary thromboprophylaxis in ambulatory cancer patients has been demonstrated to be effective. However, drawbacks are the additional burden of drug use, the lack of a reduced mortality benefit and costs. Only with reliable risk prediction the recommendation of primary thromboprophylaxis will convince oncologists and patients of its usefulness. This review deals with clinical and laboratory parameters and their combination in risk assessment models to define patients at high and low risk of VTE, in whom targeted thromboprophylaxis could best be applied. At present 90% of patients in the so-called intermediate- to high-risk group according to the Khorana score still do not develop VTE during the first 6 months, whereas there is a high absolute number of patients in the so-called low-risk groups that develop VTE. Improvements in risk assessment have been made by new risk prediction models. However, additional refinements to further improve risk prediction and their applicability in clinical practice are still needed.