**4. Risk factors for cancer-associated venous thromboembolism**

Incidence rates of VTE increase with age in the general population. Likewise, cancer-associated VTE occurs more often in the elderly population than younger population. Khorana et al. reported that age 65 or older is an independent risk factor for developing cancer-associated VTE. Cancer-associated VTE is more common in female sex and black race [3]. Obesity; complications such as respiratory disease, kidney disease, and infection; and poor performance status are also risk factors for cancer-associated VTE [10–12].

The risk of VTE varies by cancer site, and meta-analysis of several studies proves that the incidence of VTE is highest in the pancreatic cancer patients, followed by hematological malignancy and brain tumor patients [13]. VTE more often occurs in patients with advanced cancer than in patients with early cancer [14]. The incidence of VTE also varies by cancer histology, is higher in lung adenocarcinoma patients than in lung squamous cell carcinoma patients [15], and is higher in high-grade lymphoma patients than in low-grade lymphoma patients [16, 17].

Cancer treatment also affects VTE incidence. VTE is a common complication of surgery, regardless of whether it is cancer surgery or not, and adequate prophylaxis is recommended in guidelines including Japanese guideline [18]. However, among cancer patients who received adequate VTE prophylaxis after surgery, 2.1% of them developed massive VTE and 0.8% of them died [19]. Many types of anticancer drugs, such as cisplatin, l-asparaginase, and bevacizumab, also increase risk of thrombosis in cancer patients. Especially, the incidence of VTE is very high in multiple myeloma patients receiving immunomodulatory drugs (e.g., thalidomide, lenalidomide, and pomalidomide), and these patients need primary prevention of VTE by using antithrombotic drugs. Cancer patients often need indwelling central venous catheter (CVC) for delivery of intravenous drugs, parenteral nutrition, and collecting blood samples. Indwelling CVC increases risk of developing VTE, and it is estimated that the risk of symptomatic catheter thrombosis is 0.3–28% [20].

Considering these factors, several risk models to predict the occurrence of cancer-associated VTE have been published. Khorana score is the most widely used risk model among them [21]. Five predictive variables are identified in this score: site of cancer (2 points for very high-risk site, 1 point for high-risk site), platelet

count of 350 × 109 /L or more, hemoglobin less than 100 g/L (10 g/dL) and/or the use of erythropoiesis-stimulating agents, leukocyte count more than 11 × 109 /L, and body mass index of 35 kg/m2 or more (1 point each). Rates of VTE have been reported to be 0.3% in low-risk (score = 0), 2% in intermediate-risk (score = 1–2), and 6.7% in high-risk (score ≥ 3) category over a median of 2.5 months.
