Progress in the study of cancer-associated venous thromboembolism

Abstract

Objective

Venous thromboembolism is a major cause of morbidity, mortality, and increased medical costs in tumor patients. In the current review, we summarize the progress made in the study of cancer-associated venous thromboembolism.

Methods

By searching cancer-associated venous thromboembolism-related literature on PubMed, the epidemiology, pathological mechanisms, risk factors, risk prediction models, and prevention and treatment of cancer-associated venous thromboembolism were reviewed.

Results

The pathophysiological mechanisms of cancer-associated venous thromboembolism are multifactorial. Various blood cell counts (such as platelets and white blood cells) and biomarkers (such as D-dimer and sP-selectin) were considered predictors of thrombosis in cancer patients and were incorporated into the venous thromboembolism risk stratification models. Thromboprophylaxis is currently recommended for all hospitalized cancer patients. In addition, outpatient thromboprophylaxis can be used for selected high-risk patients. Low-molecular-weight heparin was the preferred treatment for cancer-associated venous thromboembolism, but some issues arose in the long-term treatment. In this case, direct oral anticoagulants were a treatment option for tumor patients. The efficacy of direct oral anticoagulant in treating cancer patients is not inferior to low-molecular-weight heparin, but is associated with a higher risk of bleeding. Therefore, there were concerns regarding their safety.

Conclusion

Since thrombocytopenia, thrombosis recurrence, and bleeding are common in tumor patients, the selection of anticoagulants in this circumstance is a considerable challenge for clinicians.

Keywords

Cancer venous thromboembolism pulmonary embolism deep venous thrombosis

Venous thromboembolism (VTE), including pulmonary embolism and deep venous thrombosis (DVT), has an incidence of 4%–20% among different cancers¹ and is the second most common cause of death after tumor progression.²

This article reviews the progress made in the study of the epidemiology, pathological mechanisms, risk factors, risk prediction models, and prevention and treatment of cancer-associated venous thromboembolism (CAT), with the aim of drawing clinicians’ attention to CAT.

Epidemiology

With the development of anti-tumor technology, the incidence of VTE in tumor patients has exhibited an upward trend. A study analyzing 3,146,388 inpatient cancer patients in the United States between 1995 and 2012 indicated that the annual incidence of thrombosis has been increasing annually, from 3.5% in 1995 to >6.5% in 2012.³ Malignant tumors have also become a common cause of VTE in China. In an observational study conducted in Hong Kong, China, between 2004 and 2016, a total of 2214 new VTE patients were analyzed, and the leading cause of VTE was found to be active malignant tumors. The proportion of VTE caused by such tumors increased from 34.8% in 2005 to 60.9% in 2014.⁴ Cancer can shorten the survival time and is a poor prognostic factor for VTE patients. Studies have demonstrated that, compared to non-cancer patients, patients with CAT have a three- to eight-fold increased risk of death.⁵ Furthermore, thrombosis also imposes a substantial economic burden on cancer patients. The medical expenses of cancer patients with VTE increase by approximately two-fold compared to those without VTE.³

Pathological mechanism

The pathological mechanism of CAT is multifactorial and the related literature suggests the following underlying mechanisms: (1) Tumor cells release a variety of procoagulant substances, including tissue factor (TF), cancer procoagulant substances (CP), and TF-positive tumor-derived microplastics (MPs), which in turn form a hypercoagulable state. TF and factor VII combine to form a macromolecular complex, which in turn activates the coagulation pathway. CP directly activates factor X. Tumor shedding MP also has a procoagulant effect. The shedding of these procoagulant MPs may be genetically regulated, which further promotes the hypercoagulable state associated with malignant tumors.⁶ (2) Inhibition of the fibrinolytic system: tumor cells can express tissue-type plasminogen activator; urokinase-type plasminogen activator; plasminogen activator inhibitor (PAI) 1, 2; and other factors, thereby inhibiting the activity of the fibrinolytic system.⁷ (3) Cancer can induce related inflammation and release certain cytokines to trigger abnormalities in the body’s coagulation function. Tumor cells can produce inflammatory cytokines, such as interleukin 1β (IL-1β), tumor necrosis factor α (TNF-α), and lipopolysaccharides. These inflammatory factors promote the expression of PAI-1 and TF by endothelial cells, down-regulate the activity of thrombin-regulated protein, inhibit the activation of the protein C system, up-regulate the activity of TF, and release vascular endothelial growth factor, thereby enhancing vascular permeability and inducing TF expression.⁸ (4) Tumor cell-induced platelet aggregation^9,10: tumor cells activate platelets, induce platelet aggregation, and promote thrombus formation via direct cell–cell interactions and the secretion of platelet-activating factors.

Risk factors

VTE in tumor patients is the result of multiple risk factors, thus a single risk factor cannot be used to predict the occurrence of thrombosis, which needs to be considered comprehensively. Throughout the disease course, the dynamic assessment of changes in risk factors and timely understanding of the risk of VTE will aid the early prevention of thrombosis and improve the prognosis of cancer patients.

Patient-related factors

Patient-related factors include long-term bed rest, advanced age (≥60 years), obesity, presence of complications, infection, and history of VTE.^2,11 When evaluating the risk of thrombosis in tumor patients, it is important to identify whether the patients have comorbidities. A study involving 3,146,388 cancer patients from >200 institutions revealed that the incidence of thrombosis in patients without comorbidities was 2.3%, and the incidence of thrombosis in patients with over three comorbidities increased to >11%.³

Tumor-related factors

Tumor-related factors include tumor type, grade, metastasis, and stage. The risk of thrombosis varies with different types of tumors. A meta-analysis including patients with different types of tumors between 1966 and 2011 statistically analyzed the occurrence of thrombosis in each tumor type. The results indicated that the incidence of thrombosis in hematological tumor, brain tumor, pancreatic cancer, and lung cancer was significantly higher than that of other cancers.¹² This suggests that different cancer types may have specific thrombogenic pathways in addition to the common thrombogenic mechanisms mentioned above. If such pathways can be found, they may provide benefits for targeted prevention and treatment of thrombus. Even in patients with the same tumor type, the risk of thrombosis differs with different tumor grades. In the Vienna Cancer and Thrombosis Study (CATS), the risk of thrombosis in patients with high-grade tumors was double that in patients with low-grade tumors (hazard ratio (HR) = 2.0, 95% confidence interval (CI) = 1.1–3.5, P = 0.015).¹³ In addition, activation of the coagulation system promotes tumor growth and angiogenesis, suggesting a link between thrombosis and poor prognosis. There is also evidence that metastatic cancer increases the risk of thrombosis compared to that associated with local cancer.¹⁴

Treatment-related factors

Treatment-related factors include chemotherapy, radiotherapy, surgery, central venous catheter, hormones, anti-angiogenic drugs, thalidomide, erythropoietin, and red blood cell or platelet transfusion.^2,11,15 Because patients can have multiple risk factors simultaneously during chemotherapy, the risk of thrombosis is very high. A case–control study found that chemotherapy increased the risk of thrombosis in cancer patients by 6.5-fold, exceeding the risk associated with the disease itself.¹⁶

Biomarkers

Among blood biomarkers, white blood cells participate in inflammatory responses, which may be related to thrombosis in tumor patients. A study by Pabinger and Posch found that for every 1 × 10⁹/L increase in the white blood cell count of tumor patients, the risk of thrombosis increased by 7% (risk ratio (RR) = 1.06, 95% CI = 1.02–1.10).¹⁷ Platelet aggregation is an important part of the process of thrombosis and hemostasis. Multiple studies have identified elevated platelet counts as an independent predictor of VTE. An international observational study including 665 cancer patients found that the risk of thrombosis increased by 3.5-fold in patients with platelet counts of ≥443 ×10⁹/L (HR = 3.50, 95% CI = 1.52–8.06, P = 0.0032).¹⁸ Moreover, in a study by Ay et al., the incidence of thrombosis in tumor patients with sP-selectin levels of >53.1 ng/mL was 11.9% after six months, while the incidence of thrombosis in tumor patients with sP-selectin levels of <53.1 ng/mL was only 3.7%. In a multivariate analysis, the association between sP-selectin levels and VTE was statistically significant (P = 0.003).¹⁵ Biomarkers that reflect intravascular thrombosis or hyperfibrinolysis, such as D-dimer and prothrombin fragment 1 + 2 (F1 + 2), are also major predictors of thrombosis in tumor patients.¹¹ However, due to the influence of the detection method and the ethnicity of the subjects, the predictive role of these biomarkers was not confirmed. It is currently not recommended to use these biomarkers alone for VTE risk assessment.

Risk assessment model

Based on the above risk factors, many studies have established risk stratified models of CAT to identify the tumor patients who may be most suitable for thromboprophylaxis. The Khorana score¹⁹ is currently the most widely used VTE risk stratification tool in tumor patients to aid clinician decision-making regarding the necessity of prophylactic anticoagulation. The score contains the following five items: cancer site, white blood cell count of >11 × 10⁹/L, body mass index of ≥35 kg/m², platelet count of ≥350 × 10⁹/L, and hemoglobin level of <10 g/dL, and/or administration of erythropoietin. With the exception of the cancer site (two points for pancreatic and gastric cancer, one point for lung cancer, lymphoma, gynecological tumor, and urogenital cancer), one point is assigned for each item. According to the total score, patients are divided into low-risk (total score of 0 points), intermediate-risk (1–2 points), and high-risk (>2 points) groups. However, the predictive effect of the Khorana score primarily depends on the type of cancer and the blood cell count before chemotherapy. It does not consider the effect of chemotherapy on blood cell levels throughout the tumor treatment process, thus it can only be used when starting chemotherapy, which limits its predictive ability. Recently, a meta-analysis of 55 cohorts (34,555 outpatient cancer patients) revealed that the Khorana score was used to evaluate the risk of these patients. Low-risk patients accounted for 19%, intermediate-risk patients for 64%, and high-risk patients for 17%. At the six-month follow-up, it was found that the incidence of VTE was 5.0% in the low-risk category, 6.6% in the intermediate-risk category, and 11.0% in the high-risk category, with approximately 25% of VTE events distributed in the high-risk category. Although the Khorana score can distinguish patients in different risk categories, most VTE events occur outside of the high-risk population, illustrating its poor predictive power.²⁰ The fact that 64% of patients were in the intermediate-risk category is a serious deficiency, as it is easy to treat patients in low- or high-risk categories, but treating patients at intermediate risk is a challenge for doctors. Therefore, researchers have proposed stratifying the risk of tumor patients with a threshold value of 2 for thromboprophylaxis. In an observational study, the incidence of thrombosis was 11% in the intermediate-high risk category (Khorana score of ≥2 points) and 4% in the low-risk category, confirming that the critical value of 2 can effectively identify patients who easily develop VTE.²¹

In order to further improve the Khorana score, Ay et al. proposed the Vienna CATS score: adding D-dimer and sP-selectin on the basis of the Khorana score increased the positive predictive value of this score from 22.1% (Khorana score) to 42.9%, thereby significantly improving the predictive efficiency.¹¹ However, the determination of sP-selectin is not easy to obtain in routine clinical practice, which significantly reduces its extensive clinical application and increases the overall cost of VTE risk assessment. At present, many international scholars have also proposed novel VTE risk prediction models, such as the PROTECHT score,²² ONKOTEV score,²³ and Tic-ONCO score,²⁴ in order to improve the ability to predict VTE risk, correctly identify patients suitable for thrombosis prevention, and avoid unnecessary anticoagulant treatment. However, the performance of these models is limited by insufficient prediction ability and poor usability. Therefore, it is a vital and difficult point to find a simple, convenient, and effective risk model for different tumor populations.

Thromboprophylaxis in cancer patients

VTE seriously affects patients’ quality of life and survival prognosis, which is a severe challenge in the natural course of cancer and treatment process. However, VTE is preventable when appropriate anticoagulant treatment is administered in a timely manner. It is worth noting that thromboprophylaxis in tumor patients is a major medical challenge, as tumor patients also exhibit higher bleeding rates, which highlights the need for clinicians to carefully evaluate the risk/benefit ratio of preventative anticoagulation.

Thromboprophylaxis in hospitalized cancer patients

Since it is well known that hospitalized tumor patients are at high risk of venous thromboembolism, the 2019 version of the National Comprehensive Cancer Network (NCCN) guideline²⁵ suggests the following: in the absence of contraindications, VTE prevention for hospitalized cancer patients can be considered, and the prevention should continue throughout the hospitalization period. However, there is a large gap between the consensus of the guidelines and reality. The rate of thromboprophylaxis in hospitalized cancer patients, in reality, is low because clinicians are often limited by the contraindications associated with thromboprophylaxis, including active bleeding, thrombocytopenia, and a high risk of bleeding. In a recent study evaluating thromboprophylaxis in hospitalized cancer patients, approximately one-third of cancer patients were unable to undergo thromboprophylaxis due to relative contraindications to anticoagulation. The most common contraindication was thrombocytopenia, which accounts for 65.2% of all contraindications, followed by active bleeding, accounting for 17.4%.²⁶

Thromboprophylaxis in outpatient cancer patients

There is still a certain VTE risk for cancer patients outside the hospital. Due to doubts regarding the risk/benefit ratio of preventative anticoagulation, there is no consensus on thromboprophylaxis in outpatient cancer patients. We have conducted multiple randomized controlled trials to explore the role of thromboprophylaxis in such patients. A study²⁷ included 1150 outpatient cancer patients with solid tumors and evaluated the efficacy of low-molecular-weight heparin (LMWH) in preventing arteriovenous thromboembolic events. Patients included had locally advanced or metastatic stomach, pancreas, breast, lung, ovarian, and head and neck cancers, with patients administered placebos as a control group and those receiving up to four months of preventative treatment as the experimental group. The results indicated that, compared to the placebo, LMWH significantly reduced the incidence of thrombosis (2% vs. 3.9%, P = 0.02). Although prevention is effective, indiscriminate VTE prevention for all outpatient cancer patients is clearly unreasonable, and further research is needed, specifically in the high-risk tumor populations. The AVERT study²⁸ included 574 outpatient tumor patients with Khorana scores of ≥2 points, who were randomly assigned to the apixaban prevention group or the placebo group for 180 days. The results revealed that the apixaban prophylaxis group reduced the risk of thrombosis by approximately 50%, but at the expense of a higher risk of major bleeding. The above two trials indicate that prophylactic anticoagulation is significantly related to a reduction in the relative risk of thrombosis in outpatient tumor patients and also emphasize the need for risk assessment before initiating thromboprophylaxis. Therefore, the current guidelines^25,29 suggest that outpatient tumor patients should not routinely undergo anticoagulant prophylaxis, while high-risk outpatient tumor patients can.

In addition, the combination of thalidomide, lenalidomide, and dexamethasone increases the risk of thrombosis. When patients with multiple myelomas are treated with lenalidomide, pomalidomide, or thalidomide, if the risk of thrombosis is high, LMWH or warfarin can be used for prevention; if the risk of thrombosis is low, aspirin should be used for prevention.^25,30 Notably, the only malignant tumor in which aspirin is applied as a thromboprophylaxis is multiple myeloma.

Thromboprophylaxis in cancer patients undergoing surgery

Malignant tumors and surgery are independent predictors of thrombosis, thus when tumor patients undergo surgery, the risk of thrombosis complications is higher. One study has demonstrated that among tumor patients undergoing surgery, the risk of postoperative DVT is more than two-fold that of non-tumor patients, and the risk of fatal PTE is at least three-fold that of non-tumor patients.³¹ Simultaneously, there are other risk factors for such patients, and thus anticoagulant prophylaxis is reasonable in tumor patients undergoing surgery. In cases where anticoagulation contraindications are excluded, preventive anticoagulation is needed before cancer patients undergo major surgery. LMWH or unfractionated heparin is administered 12 to 2 h before surgery, and postoperative prophylaxis continues for 7 to 10 days.²⁹ Research has indicated that when co-existing risk factors such as gastric, pancreatic, or metastatic cancer; sepsis; or congestive heart failure are present, cancer patients undergoing surgery are at an increased risk of thrombosis within 30 days after surgery, approximately one-third of which occur after discharge, suggesting that further research is needed to determine the most reasonable duration of postoperative VTE prophylaxis in tumor patients.³² A recent study revealed that in tumor patients who underwent major abdominal surgery and prolonged anticoagulation for four weeks after surgery, the VTE risk was reduced by 91%.³³ This finding supports the current guideline suggestion; when cancer patients undergo major operations such as abdominal and pelvic surgery, anticoagulation is recommended to be extended for four weeks.²⁹

CAT treatment

Traditional anticoagulant therapy

The anticoagulant treatment rate of cancer-related PE patients is significantly lower than that of patients with PE of different etiologies (62% vs. 81.7%, P < 0.001). This is due to clinician concerns regarding the tendency of high bleeding in patients with advanced cancer or the possible interaction between anticoagulants and anticancer drugs, leading to doubts regarding cancer-related thrombosis treatment. However, one study demonstrated that the recurrence rate of thrombus in tumor patients is high, while anticoagulation-related bleeding complications do not increase correspondingly.³⁴ These observations support current guidelines for anticoagulation therapy, even in cancer patients. Lee et al. published a meta-analysis on the choice of anticoagulant drugs for CAT, which revealed that LMWH aids the prevention of thrombosis recurrence in tumor patients to a greater extent than warfarin (HR = 0.48; P = 0.002), but there is no significant difference between the two in terms of major bleeding.³⁵ The NCCN and ASCO guidelines^25,29 suggest that once a cancer patient is diagnosed with VTE, anticoagulant therapy should be initiated, with LMWH as the first choice for treatment.

Although several guidelines have proposed LMWH as the preferred treatment for CAT, one study reported that cancer patients treated with LMWH for six months and bridged to warfarin treatment did not exhibit an increased incidence of thrombosis recurrence and bleeding. This treatment strategy may be beneficial to cancer patients. One study divided cancer patients who had been treated with LWMH for six months into two groups. One group would continue to receive LWMH anticoagulation therapy, and the other would be bridged to warfarin. The results revealed no significant difference in the risk of recurrent VTE (HR = 0.67, 95% CI = 0.44–1.02) and major bleeding (HR = 1.05, 95% CI = 0.79–1.55) between the two groups.³⁶ This suggests that warfarin may be a more acceptable option in tumor patients who cannot tolerate long-term LMWH subcutaneous injections or those in poor economic conditions.

The optimal duration of therapeutic anticoagulation has not been defined. The NCCN guideline emphasizes ≥3 months of anticoagulation,²⁵ while the ASCO guideline suggests ≥6months of anticoagulation.²⁹ We should dynamically assess whether the cancer has metastasized or progressed or is being treated with systemic chemotherapy based on the actual situation of each patient. When these factors are present, the risk of thrombus recurrence is high, and therefore anticoagulation is required indefinitely.

Direct oral anticoagulant therapy

Direct oral anticoagulants (DOACs) are becoming more widely used in clinical practice. They offer many advantages over traditional anticoagulants, including the need for less frequent monitoring, the convenience of oral administration, fixed doses, reduced risk of drug-food interactions, and more predictable pharmacodynamics. Previous studies have directly compared the efficacy of DOACs and LMWH in the treatment of tumor patients. The Select-D trial³⁷ applied rivaroxaban and LMWH to treat CAT and found that rivaroxaban was associated with a lower cumulative recurrence rate of thrombosis (HR = 0.43, 95% CI = 0.19–0.99) than LMWH. The incidence of major bleeding in the two groups was 6% and 4%, respectively. It was also observed that rivaroxaban had a higher incidence of major bleeding in digestive tract tumors. Another clinical study³⁸ compared the efficacy of edoxaban and LMWH in CAT, including 1046 tumor patients for ≥6 months of treatment. In this study, the recurrence rates of VTE with edoxaban and LMWH were comparable (7.9% vs. 11.3%, respectively), but edoxaban had a higher incidence of major bleeding than LMWH (9% vs. 4%, respectively). The ADAM VTE Trial found that DOACs treatment for cancer patients reduced the incidence of major bleeding and recurrent VTE compared with LMWH.³⁹ The above three trials found that DOACs were not inferior to LMWH in treating CAT. Therefore, DOACs can provide a novel option for the long-term treatment of CAT patients, particularly in patients with better prognosis. However, the comparative study of DOACs and LMWH included a small sample size, and DOACs may be related to a higher risk of major bleeding, particularly in patients with digestive tract tumors.⁴⁰ Therefore, the evidence for its application in cancer patients is insufficient, and further research is needed.

Conclusion

The prevention and treatment of CAT are challenging and unique. The development of appropriate risk assessment tools to identify the cancer patients best suited for thrombosis prevention is needed. However, the existing risk stratification tools for cancer patients have poor predictive power and applicability. Building a more complete and effective model is thus a necessity. At present, LMWH monotherapy is considered the standard for treating CAT, but the long-term subcutaneous injection of LMWH represents great inconvenience to patients and hinders its clinical use. Therefore, the emergence of DOACs provides a promising alternative treatment for CAT patients. However, current data on the safety and efficacy of DOACs in the treatment of tumor patients are limited, and DOACs are associated with a higher risk of bleeding. Further large-scale clinical trials are needed to support the clinical application of DOACs and the update of guidelines.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by Precision Medical Special National Key Research and Development Plan (2016YFC0905600).

ORCID iD

Yuting Yao

References

Wang

, et al. Clinical characteristics and prognosis of cancer patients with venous thromboembolism. J Cancer Res Ther 2019; 15: 344–349.

Faiz

Khan

Beckman

, et al. Characteristics and risk factors of cancer associated venous thromboembolism. Thromb Res 2015; 136: 535–541.

Lyman

Culakova

Poniewierski

, et al. Morbidity, mortality and costs associated with venous thromboembolism in hospitalized patients with cancer. Thromb Res 2018; 164: S112–S118.

Huang

Chan

She

, et al. Secular trends and etiologies of venous thromboembolism in Chinese from 2004 to 2016. Thromb Res 2018; 166: 80–85.

Uppuluri

Burke

Haaf

, et al. Assessment of venous thromboembolism treatment in patients with cancer on low molecular weight heparin, warfarin, and the direct oral anticoagulants. J Oncol Pharm Pract 2019; 25: 261–268.

Schmaier

Ambesh

Campia

Venous thromboembolism and cancer. Curr Cardiol Rep 2018; 20: 89.

Kwaan

Keer

HN.

Fibrinolysis and cancer. Semin Thromb Hemost 1990; 16: 230–235.

Colucci

Balconi

Lorenzet

, et al. Cultured human endothelial cells generate tissue factor in response to endotoxin. J Clin Invest 1983; 71: 1893–1896.

Bastida

Ordinas

Platelet contribution to the formation of metastatic foci: the role of cancer cell-induced platelet activation. Haemostasis 1988; 18: 29–36.

10.

Bakewell

Nestor

Prasad

, et al. Platelet and osteoclast beta3 integrins are critical for bone metastasis. Proc Natl Acad Sci USA 2003; 100: 14205–14210.

11.

Dunkler

Marosi

, et al. Prediction of venous thromboembolism in cancer patients. Blood 2010; 116: 5377–5382.

12.

Horsted

West

Grainge

MJ.

Risk of venous thromboembolism in patients with cancer: a systematic review and meta-analysis. PLoS Med 2012; 9: e1001275.

13.

Ahlbrecht

Dickmann

, et al. Tumor grade is associated with venous thromboembolism in patients with cancer: results from the Vienna cancer and thrombosis study. J Clin Oncol 2012; 30: 3870–3875.

14.

Dickmann

Ahlbrecht

, et al. Regional lymph node metastases are a strong risk factor for venous thromboembolism: results from the Vienna Cancer and Thrombosis Study. Haematologica 2013; 98: 1309–1314.

15.

Simanek

Vormittag

, et al. High plasma levels of soluble P-selectin are predictive of venous thromboembolism in cancer patients: results from the Vienna Cancer and Thrombosis Study (CATS). Blood 2008; 112: 2703–2708.

16.

Heit

Silverstein

Mohr

, et al. Risk factors for deep vein thrombosis and pulmonary embolism: a population-based case-control study. Arch Intern Med 2000; 160: 809–815.

17.

Pabinger

Posch

Flamethrowers: blood cells and cancer thrombosis risk. Hematol Am Soc Hematol Educ Prog 2014; 2014: 410–417.

18.

Simanek

Vormittag

, et al. High platelet count associated with venous thromboembolism in cancer patients: results from the Vienna Cancer and Thrombosis Study (CATS). J Thromb Haemost 2010; 8: 114–120.

19.

Khorana

Kuderer

Culakova

, et al. Development and validation of a predictive model for chemotherapy-associated thrombosis. Blood 2008; 111: 4902–4907.

20.

Mulder

Candeloro

Kamphuisen

, et al. The Khorana score for prediction of venous thromboembolism in cancer patients: a systematic review and meta-analysis. Haematologica 2019; 104: 1277–1287.

21.

Lustig

Rodriguez

Wells

PS.

Implementation and validation of a risk stratification method at The Ottawa Hospital to guide thromboprophylaxis in ambulatory cancer patients at intermediate-high risk for venous thrombosis. Thromb Res 2015; 136: 1099–1102.

22.

Verso

Agnelli

Barni

, et al. A modified Khorana risk assessment score for venous thromboembolism in cancer patients receiving chemotherapy: the Protecht score. Intern Emerg Med 2012; 7: 291–292.

23.

Cella

Di Minno

Carlomagno

, et al. Preventing venous thromboembolism in ambulatory cancer patients: the ONKOTEV study. The Oncol 2017; 22: 601–608.

24.

Munoz Martin

Ortega

Font

, et al. Multivariable clinical-genetic risk model for predicting venous thromboembolic events in patients with cancer. Br J Cancer 2018; 118: 1056–1061.

25.

NCCN Guidelines Version 1. Cancer-associated venous thromboembolic disease [DB/OL]. Plymouth: National Comprehensive Cancer Network, 2019.

26.

Zwicker

Rojan

Campigotto

, et al. Pattern of frequent but nontargeted pharmacologic thromboprophylaxis for hospitalized patients with cancer at academic medical centers: a prospective, cross-sectional, multicenter study. J Clin Oncol 2014; 32: 1792–1796.

27.

Agnelli

Gussoni

Bianchini

, et al. Nadroparin for the prevention of thromboembolic events in ambulatory patients with metastatic or locally advanced solid cancer receiving chemotherapy: a randomised, placebo-controlled, double-blind study. Lancet Oncol 2009; 10: 943–949.

28.

Carrier

Abou-Nassar

Mallick

, et al. Apixaban to prevent venous thromboembolism in patients with cancer. N Engl J Med 2019; 380: 711–719.

29.

Key

Khorana

Kuderer

, et al. Venous thromboembolism prophylaxis and treatment in patients with cancer: ASCO clinical practice guideline update. J Clin Oncol 2020; 38: 496–520.

30.

Larocca

Cavallo

Bringhen

, et al. Aspirin or enoxaparin thromboprophylaxis for patients with newly diagnosed multiple myeloma treated with lenalidomide. Blood 2012; 119: 933–939; quiz 1093.

31.

Khorana

Francis

CW.

Risk prediction of cancer-associated thrombosis: appraising the first decade and developing the future. Thromb Res 2018; 164: S70–S76.

32.

Ruff

Weber

Khader

, et al. Venous thromboembolism in patients with cancer undergoing surgical exploration. J Thromb Thrombolysis 2019; 47: 316–323.

33.

Vedovati

Becattini

Rondelli

, et al. A randomized study on 1-week versus 4-week prophylaxis for venous thromboembolism after laparoscopic surgery for colorectal cancer. Ann Surg 2014; 259: 665–669.

34.

Kim

Chang

Lee

, et al. Differential clinical manifestations and clinical outcome of cancer-related pulmonary embolism. Korean J Intern Med 2020; 35: 360–368.

35.

Lee

Levine

Baker

, et al. Low-molecular-weight heparin versus a coumarin for the prevention of recurrent venous thromboembolism in patients with cancer. N Engl J Med 2003; 349: 146–153.

36.

Chai-Adisaksopha

Iorio

Crowther

, et al. Vitamin K antagonists after 6 months of low-molecular-weight heparin in cancer patients with venous thromboembolism. Am J Med 2018; 131: 430–437.

37.

Young

Marshall

Thirlwall

, et al. Comparison of an oral factor Xa inhibitor with low molecular weight heparin in patients with cancer with venous thromboembolism: results of a randomized trial (SELECT-D). J Clin Oncol 2018; 36: 2017–2023.

38.

Raskob

van Es

Verhamme

, et al. Edoxaban for the treatment of cancer-associated venous thromboembolism. N Engl J Med 2018; 378: 615–624.

39.

McBane

Wysokinski

Le‐Rademacher

, et al. Apixaban and dalteparin in active malignancy‐associated venous thromboembolism: The ADAM VTE trial. J Thromb Haemost 2019; 18: 411–421.

40.

Agnelli

Becattini

Meyer

, et al. Apixaban for the treatment of venous thromboembolism associated with cancer. N Engl J Med 2020; 382: 1599–1607.