Thromboprophylaxis with low-molecular-weight heparin in medical patients with cancer


  • Gary H. Lyman MD, MPH

    Corresponding author
    1. Department of Medicine, Duke University School of Medicine and the Duke Comprehensive Cancer Center, Durham, North Carolina
    • Professor of Medicine and Director, Health Services, Effectiveness, and Outcomes Research, Division of Medical Oncology, Department of Medicine, Duke University and the Duke Comprehensive Cancer Center, Durham, NC 27710===

    Search for more papers by this author
    • Fax: (919) 681-7488


Venous thromboembolism (VTE) is a frequent complication of cancer and cancer treatment and is associated with multiple clinical consequences, including recurrent VTE, bleeding, and an increase in the risk of death. Although the risks associated with VTE have been well recognized in surgical cancer patients, there is also considerable and increasing risk in medical cancer patients. VTE risk factors in medical cancer patients include the type and stage of cancer, major comorbid illnesses, current hospitalization, active chemotherapy, hormone therapy, and antiangiogenic agents. Low-molecular-weight heparins (LMWHs) are recommended commonly for the prevention of VTE in hospitalized cancer patients and in higher risk ambulatory cancer patients because of their favorable risk-to-benefit profile. These agents have demonstrated effectiveness in both the primary and secondary prevention of VTE in medical cancer patients. Extended-duration anticoagulant therapy is often recommended to reduce the risk of VTE recurrence in patients with cancer. LMWHs are often used for long-term prophylaxis because of a reduced need for coagulation monitoring, few major bleeding episodes, and once-daily dosing. Despite clinical and practical benefits, a substantial proportion of medical cancer patients do not receive VTE prophylaxis. To improve the appropriate prevention and treatment of VTE in cancer patients, guidelines have been published recently by the American Society of Clinical Oncology and the National Comprehensive Cancer Network. Widespread dissemination and application of these guidelines are encouraged to improve the appropriate use of these agents and to improve clinical outcomes in medical cancer patients at risk for VTE and its complications. Cancer 2009. © 2009 American Cancer Society.

Venous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE), is a frequent complication of cancer and cancer treatment. VTE in patients with cancer is strongly associated with reduced survival,1-3 such that these patients are >3 times more likely to die within 6 months of VTE compared with patients who have VTE without cancer.1, 4 Furthermore, patients who are diagnosed with cancer at the same time or within 1 year of a VTE event are associated more often with advanced stage and poor prognosis compared with patients who are diagnosed with cancer without a preceding VTE event.2 Cancer patients with VTE are also more likely to develop recurrent VTE and major bleeding during anticoagulant treatment compared with VTE in patients without malignancy.5

The low-molecular-weight heparins (LMWHs) have demonstrated the ability to reduce the incidence of VTE and prevent recurrent VTE events in cancer patients.6-8 LMWH is prescribed frequently for the treatment and secondary prevention of VTE in patients with cancer because of the favorable benefit-to-risk profile and the minimal requirement for monitoring.9-13 Although they still require further investigation, several randomized clinical trials also have suggested that LMWHs may improve survival in certain populations of cancer patients.14-19

An association between active cancer, increased risk for VTE, and the need for thromboprophylaxis is commonly acknowledged in cancer patients who undergo major surgical procedures. However, the incidence and impact of VTE are also considerable in patients who receive nonsurgical cancer treatment.13, 20, 21 Hospitalized neutropenic patients with cancer and VTE have a greater mortality rate compared with cancer patients without VTE (odds ratio [OR], 2.01; 95% confidence interval [CI], 1.83-2.22; P < .0001), and the increased risk is similar in patients with either nonmetastatic or metastatic disease.22 In a recent study of ambulatory patients with cancer who were receiving chemotherapy, the cumulative risk of VTE was approximately 4%, and its occurrence was a significant, independent predictor for early mortality in multivariate analysis.4 Clinical risk models and new genetic and molecular biomarkers are under active investigation in an effort to improve the identification of cancer patients undergoing medical treatment who are at increased risk for VTE and may be candidates for thromboprophylaxis in the ambulatory setting. In this review, the incidence and risk factors associated with VTE in medical patients with cancer and the appropriate indications for LMWH thromboprophylaxis are discussed, and the current underuse of VTE prophylaxis in cancer patients who are at increased risk is highlighted.

Incidence of Venous Thromboembolism in Medical Patients With Cancer

Clinically detectable VTE is diagnosed in approximately 15% of all cancer patients, and the number is likely to be even higher when subclinical thromboembolism is taken into account.23 Several studies have investigated the clinical incidence of VTE in medical cancer patients.22, 24, 25 In a large, retrospective study of patients who were hospitalized between 1979 and 1999 (n = 40,787,000), the rate of VTE was 2% in patients with cancer compared with 1% in similar patients without malignancy.24 In an analysis of neutropenic cancer patients who were hospitalized between 1995 and 2002, 5.4% of patients developed VTE during their first hospitalization, and the rates increased significantly over the study period.21 Khorana and colleagues also noted the increasing frequency of VTE in a recent study of 1,015,598 hospitalized cancer patients who were included in the discharge database of the University Health System Consortium.25 The proportion of patients with VTE increased from 3.6% per hospitalization in 1995 to 4.6% in 2002-2003, an increase of 28% (P < .0001) (Fig. 1). A near doubling of the rate of PE in that study was reported from 0.8% in 1995 to 1.5% in 2002/2003.

Figure 1.

This chart illustrates the rate of venous thromboembolism (VTE) over time in patients with cancer (adapted from Khorana AA, Francis CW, Culakova E, Kuderer NM, Lyman GH. Frequency, risk factors, and trends for venous thromboembolism among hospitalized cancer patients. Cancer. 2007;110:2339-2346.25) Error bars represent 95% confidence intervals. DVT indicates deep vein thrombosis; PE, pulmonary embolism.

Risk Factors for Venous Thromboembolism in Medical Cancer Patients

Identifying clinical characteristics that place cancer patients at increased risk of initial and recurrent VTE and its complications is important if their outcomes are to be improved. The main risk factors for VTE in cancer patients can be categorized as either related to individual patient characteristics, to the presence of cancer itself, or to the cancer treatment received (Table 1).13

Table 1. Risk Factors for Venous Thromboembolism in Patients With Cancer*
  • *

    Reprinted with permission. © 2008 American Society of Clinical Oncology. All rights reserved. Lyman GH, Khorana AA, Falanga A, et al. American Society of Clinical Oncology guideline: recommendations for venous thromboembolism prophylaxis and treatment in patients with cancer. J Clin Oncol. 2007;25:5490-5505.13

  • Bevacizumab clearly is associated with an increased risk of arterial thrombotic events; an association with venous thrombosis is not fully established.

Patient-related factors
 Advanced age
 Ethnicity (higher in African Americans; lower in Asian-Pacific Islanders)
 Comorbid conditions (obesity, infection, renal disease, pulmonary disease, arterial thromboembolism)
 Prior history of venous thromboembolism
 Elevated prechemotherapy platelet count
 Heritable prothrombotic mutations
Cancer-related factors
 Primary site of cancer (gastrointestinal, brain, lung, gynecologic, renal, hematologic)
 Initial 3-6 mo after diagnosis
 Current metastatic disease
Treatment-related factors
 Recent major surgery
 Current hospitalization
 Active chemotherapy
 Active hormone therapy
 Current or recent angiogenic therapy (thalidomide, lenalidomide, bevacizumab)
 Current erythropoiesis-stimulating agents
 Presence of central venous catheters

In the study of hospitalized cancer patients by Khorana et al., advanced age (≥65 years), being a woman, and black ethnicity were patient-related risk factors associated with VTE in multivariate analysis (P < .0001).25 The presence of comorbid conditions—in particular, neutropenic complications, infection, obesity, anemia, pulmonary disease, and renal disease—also contributed to the risk of VTE.25 Other patient-related risk factors include a prior history of VTE or inherited prothrombotic mutations, such as factor V Leiden and prothrombin 20210A mutations.26

The primary site of cancer is an important risk factor for VTE, and high rates of VTE are reported in hospitalized patients who have cancer of the pancreas (8.1%), kidney (5.6%), ovary (5.6%), lung (5.1%), stomach (4.9%), and brain (4.7%).25 A high incidence of VTE has also been reported in patients with hematologic malignancies, such as multiple myeloma (5%), non-Hodgkin lymphoma (4.8%), and Hodgkin disease (4.6%).25 Patients with cancer have a particularly increased risk of VTE in the first few months after diagnosis and when distant metastases are present.26, 27

Active medical treatments, such as chemotherapy and hormone therapy, have been shown to increase the risk of VTE, including the risk of recurrent VTE. A population survey demonstrated a 6-fold increase in the risk of VTE in cancer patients who were receiving chemotherapy compared with a 4-fold increase in cancer patients who were not receiving chemotherapy.20 Moreover, the risk of recurrent VTE increased by >4-fold in cancer patients who were receiving chemotherapy and by 2-fold in patients who were not receiving chemotherapy.28 It has been estimated that cancer patients who are receiving chemotherapy represent 12% of the total cases of VTE in the community, whereas cancer patients who are not receiving chemotherapy constitute some 6%.29

Furthermore, the rate of VTE in hospitalized patients receiving chemotherapy has increased over time from 3.9% in 1995 to 5.7% in 2002/2003, a significant rise of 46% (P < .0001).25 Although the reasons for this increase largely remain unknown, they may relate in part to the increased use and sensitivity of imaging procedures used in staging cancer patients and to the use of more intensive combination systemic therapies for the treatment of cancer.30, 31 The observed rise in VTE events may also relate in part to the introduction of new antiangiogenic cancer treatments, such as bevacizumab and thalidomide or lenalidomide-based regimens, which have demonstrated that ability to increase the risk of VTE complications.32, 33 In a meta-analysis of the use of thalidomide in patients with multiple myeloma, thalidomide and dexamethasone reportedly increased the risk of VTE by 2.6 times and 2.8 times, respectively, whereas the combination of agents along with chemotherapy and corticosteroids increased the risk of VTE 8-fold.31 An increased risk of VTE also has been demonstrated in cancer patients who received hormone therapy. VTE rates as high as 8% have been reported in a meta-analysis of patients with breast cancer who received treatment with tamoxifen.34 Supportive treatment with the erythropoiesis-stimulating agents epoetin and darbepoetin has been associated with an increase in VTE (relative risk, 1.67; 95% CI, 1.35-2.06).35 However, a recent study of 504,208 cancer patients who were hospitalized between 1995 and 2003 demonstrated an increased risk of VTE with a diagnosis of anemia (OR, 1.24; 95% CI, 1.19-1.29; P < .001) and with the use of erythrocyte transfusions (OR, 1.60; 95% CI, 1.53-1.67; P < .001) after adjustment for other risk factors for VTE.36

A simple risk model for predicting VTE based on baseline clinical and laboratory variables has been developed using data from a multicenter, prospective, observational study of ambulatory cancer outpatients who were receiving chemotherapy.27 Risk factors for VTE were studied in a randomly selected development cohort of 2701 ambulatory patients with cancer. A risk score for VTE was derived in this population and then was validated in a separate group of 1365 patients from the same study. Five predictive variables were identified as significant independent risk factors for VTE in a multivariate model, as shown in Table 2.27 Three risk categories were defined based on the score from the risk model (low [score 0], intermediate [score 1 or 2], and high [score ≥3]). The observed rates of VTE according to risk category were similar in the development cohort and the validation cohort, with approximately 7% of patients deemed at high risk of VTE (Fig. 2).27 This risk model may be used by clinicians to assess risk for VTE in clinical practice and also in the selection of cancer outpatients for trials of thromboprophylaxis. Additional investigation is underway to validate this risk model prospectively and to demonstrate the influence of prophylactic anticoagulation on rates of VTE in high-risk ambulatory cancer patients. The potential impact of adding various biomarkers, such as tissue factor, on the predictive accuracy of the model also is being studied. The VTE risk score also recently demonstrated the ability to discriminate patients at low risk, intermediate risk, and high risk for early all-cause mortality in ambulatory cancer patients who were receiving chemotherapy.37

Figure 2.

Observed rates of venous thromboembolism (VTE) are illustrated according to risk category for derivation and validation cohorts in the development of a predictive model for chemotherapy-associated VTE. Reprinted with permission from the American Society of Hematology as originally published in Khorana AA. Development and validation of a predictive model for chemotherapy-associated thrombosis. Blood. 2008;111:4902-4907.27 © 2008 American Society of Hematology.

Table 2. Predictive Model for Chemotherapy-Associated Venous Thromboembolism*
VariableRisk Score
  • *

    Reprinted with slight adaptation by permission from the American Society of Hematology as originally published in Khorana AA. Development and validation of a predictive model for chemotherapy-associated thrombosis. Blood. 2008;111:4902-4907.27 © The American Society of Hematology.

Site of cancer
Very high risk (stomach, pancreas)2
High risk (lung, lymphoma, gynecologic, bladder, testicular)1
Prechemotherapy platelet count ≥350×109/L1
Hemoglobin <100 g/L or use of erythrocyte growth factors1
Prechemotherapy leukocyte count >11×109/L1
Body mass index ≥35 kg/m21

Prophylaxis Rates in Medical Patients With Cancer

Evidence from population-based surveys indicates that the rates of VTE prophylaxis are low or, when provided, that VTE prophylaxis is often prescribed inappropriately for cancer patients in general and for medical patients with cancer in particular (Table 3).38-42 In an analysis of 1096 patients with active cancer from a prospective US DVT registry, 28.2% of medical oncology patients received prophylaxis before they developed DVT, a rate that was significantly less than that reported in patients without cancer (34.6%; P < .0001).38 A 4-year survey of prophylaxis rates in >2 million US medical patient discharges indicated that hospitalized cancer patients had the lowest rates of prophylaxis (range, 18%-25%) compared with other medical conditions, including acute myocardial infarction (range, 71%-74%), heart failure (range, 29%-38%), severe lung disease (range, 24%-32%), and ischemic stroke (range, 27%-32%).39 In addition, a Canadian audit of hospital VTE prophylaxis reported that, among 1894 patients, those with cancer had a significantly reduced likelihood of receiving prophylaxis compared with acutely ill medical patients without cancer (OR, 0.40; 95% CI, 0.24-0.68; P = .0007).42 Likewise, the Fundamental Research in Oncology and Thrombosis (FRONTLINE) survey assessed prophylaxis by clinicians involved in cancer care, and marked differences were observed in the use of prophylaxis for surgical and medical cancer patients.41 More than 50% of surgeons reported that they initiated prophylaxis routinely, whereas most medical oncologists stated that they used prophylaxis in <5% of patients. The use of appropriate prophylaxis was assessed in a recent survey of 196,104 medical patient discharges from 227 US hospitals, including 30,708 patients with cancer.40 Although approximately half of patients with cancer (56.4%) received some prophylaxis, only 27.6% of patients received prophylaxis in accordance with the Sixth American College of Chest Physicians (ACCP) guidelines. Possible reasons for the poor prophylaxis rates reported in cancer patients include concerns over the risks of bleeding and thrombocytopenia, belief that the cancer-associated VTE risk is low, the perception that treatments are not very effective, and cost considerations.41, 42 Consequently, physicians may be reluctant to prescribe anticoagulation despite evidence that appropriate prophylaxis can confer greater benefits than risks. To address such concerns, clinical practice guidelines specific to VTE prophylaxis in cancer patients have been developed based on emerging evidence from several randomized, controlled clinical trials that were designed to evaluate the efficacy and safety as well as the optimal approach to prophylactic anticoagulation in medical patients with cancer.

Table 3. Rates Prophylaxis Use in Patients With Cancer at Risk of Venous Thromboembolism
ReferenceTotal At-Risk Medical Patients Studied Who Received Any Prophylaxis, %At-Risk Medical Cancer Patients Who Received Any Prophylaxis, %
  • NA indicates not applicable.

  • *

    Total medical patients without cancer.

Seddighzadeh 20073834.6*28.2
Burleigh 20063926-3318-25
Amin 20074061.856.4
Kakkar 200341NA<5

Impact of Venous Thromboembolism Prophylaxis

Efficacy and safety of prophylaxis with low-molecular-weight heparin

Studies have demonstrated the efficacy and safety of LMWH versus unfractionated heparin (UFH) or placebo in surgical cancer patients, including Enoxaparin and Cancer (ENOXACAN) study I and ENOXACAN study II.6, 43 There are fewer data, however, on the efficacy of LMWH as primary prophylaxis in medical cancer patients. Nevertheless, 3 large randomized, controlled trials of hospitalized acutely ill medical patients have demonstrated that enoxaparin (the Prophylaxis in Medical Patients With Enoxaparin [or MEDENOX] trial), dalteparin (the Prospective Evaluation of Dalteparin Efficacy for Prevention of VTE in Immobilized Patients Trial [PREVENT]), and fondaparinux (the Arixtra for Thromboembolism Prevention in a Medical Indications Study [ARTEMIS]) are effective in the prevention of screen-detected VTE.7, 8, 44

Several studies have assessed the efficacy and safety of LMWH treatment in the secondary prevention of VTE (Table 4).45-49 In the Randomized Comparison of LMWH versus Oral Anticoagulant Therapy for the Prevention of Recurrent Venous Thromboembolism in Patients With Cancer (CLOT) study, patients with cancer who had acute, symptomatic, proximal DVT, PE, or both were randomized to receive either the LMWH dalteparin for 5 to 7 days plus a coumarin derivative for 6 months or dalteparin alone for 6 months (Table 4).46 The hazard ratio (HR) for recurrent VTE in the dalteparin group compared with the oral anticoagulant group was 0.48 (95% CI, 0.30-0.77; P = .002), and no significant differences were reported in the rate of major bleeding. Likewise, when the long-term effects of usual care versus tinzaparin were investigated, cancer patients who were receiving tinzaparin experienced a lower risk of recurrent VTE than patients who were receiving usual care (relative risk, 0.44; absolute difference, −9%; 95% CI, −21.7 to −0.7; P = .044), and similar numbers of patients reportedly experienced major or fatal bleeding complications.47 However, the US Food and Drug Administration cautioned against the use of tinzaparin to treat VTE in elderly patients with renal insufficiency. Celgene (Summit, NJ) has issued a letter describing a controlled clinical study suggesting that tinzaparin may increase the risk for death compared with UFH in elderly patients with renal insufficiency. The American Society of Clinical Oncology (ASCO) guidelines recommend alternatives to tinzaparin when treating such patients for DVT with or without PE.50

Table 4. Randomized Controlled Trials for Treatment of Cancer-Related Venous Thromboembolism*
Study/No. of PatientsTreatment RegimensProportion of Patients Experiencing Event During Treatment Period, %Statistical Inference
Recurrent VTEMajor BleedingDeath
  • VTE indicates venous thromboembolism; IU, international units; OD, once daily; INR, international normalized ratio; NS, not significant; IV, intravenous; UFH, unfractionated heparin; APTT, activated partial thromboplastin time; BID, twice-daily; ND, not determined.

  • *

    From Falanga A, Lee AY, Streiff MB, Lyman GH. Cancer and thrombosis 3: anticoagulation in the treatment of venous thromboembolism in patients with cancer [abstract]. Am Soc Clin Oncol Ed Book. 2008;2008:258-268.45 Adapted with permission. © 2008 American Society of Clinical Oncology. All rights reserved.

  • The major bleeding rate does not include fatal bleeding.

  • The rate of recurrent VTE or major hemorrhage.

Lee 200346    P = .002 for recurrent VTE; NS for major bleeding and for death
 336Dalteparin (200 IU/kg OD) for 5-7 d and warfarin (target INR, 2-3) for 6 mo16441 
 336Dalteparin (200 IU/kg OD) for 1 mo and dalteparin (˜150 IU/kg OD) for 5 mo8639 
Hull 200647    NS
 100IV UFH (APTT adjusted) for 6 d and warfarin (target INR, 2-3) for 3 mo10719 
 100Tinzaparin (175 IU/kg OD) for 3 mo6720 
Meyer 200248    NS
 71Enoxaparin (1.5 mg/kg OD) for ≥4 d and warfarin (target INR, 2-3) for 3 mo21.11622.7 
 67Enoxaparin (1.5 mg/kg OD) for 3 mo10.5711.3 
Deitcher 200649    ND for recurrent VTE or death; NS for major bleeding
 30Enoxaparin (1 mg/kg BID) for ≥5 d and warfarin (target INR, 2-3) for 178 d10.02.932.4 
 29Enoxaparin (1 mg/kg BID) for 5 d and enoxaparin (1 mg/kg OD) for 175 d6.96.522.6 
 32Enoxaparin (1 mg/kg BID) for 5 d and enoxaparin (1.5 mg/kg OD) for 175 d6.311.141.7 

Another study of patients with cancer and VTE compared enoxaparin for 3 months with enoxaparin bridged to warfarin therapy (Table 4).48 Overall, 21% of patients who received warfarin experienced the composite outcome of major hemorrhage or recurrent VTE (95% CI, 12.3-32.4) compared with 10.5% of patents who received enoxaparin alone (95% CI, 4.3-20.3; P = .09). When the time to major hemorrhage or recurrent VTE event was analyzed, enoxaparin was more effective than warfarin (P = .04; log-rank test). Major bleeding occurred in 16% of patients in the warfarin group compared with 7% in the enoxaparin alone arm (P = .09). It should be noted, however, that the definition of major bleeding varied between the trials cited.46-49 Unlike the other studies that were considered, Hull and colleagues did not classify bleeding resulting in death as major bleeding but as a separate outcome.47 In addition, Deitcher and colleagues considered major bleeding the need for surgery or decompression of a closed space and ecchymosis or hematoma >10 cm in greatest dimension.49

The potential role of LMWH in VTE prophylaxis for ambulatory cancer patients remains an area of active investigation. Of the randomized, controlled trials of LMWH in ambulatory cancer patients reported to date, none have demonstrated a significant reduction in VTE, and only 2 have been published.15, 51-53 It is worth noting that the recently presented Prophylaxis for Thromboembolism in Critical Care Trial (PROTECHT) did indicate a significant reduction in all thrombotic events combined.53 The relatively low risk of VTE in unselected ambulatory patients and the small sample size of most studies preclude definitive conclusions and further suggest that better methods for risk stratification in this patient population are needed.

Effects of low-molecular-weight heparin on survival

In addition to preventing VTE complications, several randomized, controlled trials and meta-analyses of those trials have addressed the question of whether treatment with an LMWH in cancer patients without recognized VTE may improve their survival.18, 19, 54 One-year mortality rates from studies that were included in a meta-analysis performed in support of the ASCO Guidelines Panel are presented in Figure 3.19 For cancer patients without a concurrent diagnosis of VTE, the relative risk for mortality compared with controls was 0.877 (95% CI, 0.789-0.975; P = .015) for LMWH and 0.942 (95% CI, 0.854-1.040; P = .239) for warfarin. The estimated absolute risk difference compared with controls was 8% for LMWH and 3% for warfarin. Although it was studied in different patient populations, the estimated absolute increase in major bleeding episodes was 1% (95% CI, 0.3%-2.3%) in studies of LMWH and 11.5% (95% CI, 8.5%-14.5%) in studies of warfarin.

Figure 3.

Results for 1-year overall mortality are illustrated by type of anticoagulation treatment from a meta-analysis of patients with cancer without venous thromboembolism (adapted from Kuderer NM, Khorana AA, Lyman GH, Francis CW. A meta-analysis and systematic review of the efficacy and safety of anticoagulants as cancer treatment. Cancer. 2007;110:1149-116019). Rate 1 indicates anticoagulation group; Rate 2, control group; RR, relative risk; CL, confidence limits; SCLC, small cell lung cancer; Mixed, mixture of solid tumors; LMWH, low-molecular-weight heparin; UFH, unfractionated heparin; NSCLC, nonsmall cell lung cancer; CRC, colorectal cancer; HN, head and neck cancer; LMWH, low-molecular-weight heparin.

In the Fragmin Advanced Malignancy Outcome Study (FAMOUS), a significant difference in survival was not observed with dalteparin 5000 IU once daily versus placebo when the entire study population was analyzed.15 However, a post-hoc analysis of a subgroup of patients who had a better prognosis and who were alive 17 months after randomization suggested that the 2-year and 3-year survival rates were improved for the patients who received dalteparin (78% vs 55%, respectively) versus placebo (60% vs 36%, respectively; P = .03). Likewise, a subgroup analysis was conducted of the CLOT trial to assess the effect of dalteparin on survival in patients with and without metastases.17 In patients without metastases, the probability of death at 12 months was 20% in the dalteparin group compared with 36% in the oral anticoagulant group (HR, 0.50; 95% CI, 0.27-0.95; P = .03), whereas no significant difference was observed in patients with metastatic disease (HR, 1.1; 95% CI, 0.87-1.4; P = .46). Such post-hoc subgroup analyses must be interpreted with caution and should be considered hypothesis-generating. It is noteworthy that a recent meta-analysis of 5 randomized, controlled trials of LMWH in patients with small cell lung cancer who had limited disease suggested a survival benefit with LMWH.55

Studies of the impact of LMWHs on survival in patients with metastatic disease have produced inconsistent results. An 18-week study compared chemotherapy alone (n = 42 patients) with chemotherapy plus the LMWH dalteparin at a once-daily dose of 5000 IU (n = 42 patients) for the treatment of small cell lung cancer.14 The median overall survival in that trial was 8 months with chemotherapy alone and 13 months with the addition of dalteparin (P = .01), and similar improvements in survival were observed among patients with limited stage and extensive stage disease. In a 6-week study of an LMWH in patients with advanced cancer, the median survival was 6.6 months in the placebo group compared with 8 months in the LMWH group (HR for mortality, 0.75; 95% CI, 0.59-0.96; P = .021).16 However, Sideras et al. observed no difference in survival with LMWH therapy versus placebo in patients with cancer who had advanced disease.51 Further studies are required to better define the clinical setting, including stage of disease, in which LMWHs may be considered. Additional research also is warranted investigating the effects of LMWHs across different cancer sites. Zacharski et al. concluded that warfarin treatment was particularly beneficial in patients who had small cell lung cancer compared with patients who had nonsmall cell lung, colorectal, head and neck, and prostate cancers.56 However, the comparative effects of the LMWHs on different tumor types remain to be elucidated.

Antitumor effects

Heparin and heparin-like compounds appear to possess anticancer properties. Heparin may influence malignant cell growth through different interrelated mechanisms, including inhibiting heparin-binding growth factors that drive malignant cell growth; inhibiting tumor cell heparinases that mediate tumor cell invasion and metastasis; and blocking cell-surface, selectin-mediated tumor cell metastasis and blood coagulation.57 Several experimental studies have suggested that the LMWHs may inhibit angiogenesis, block thrombin-induced platelet aggregation, inhibit platelet interaction with vascular endothelium, and stimulate platelet production.58 In contrast to UFH, it has been demonstrated that the LMWHs can hinder the binding of growth factors to their high-affinity receptors.57 For example, small molecular heparin fractions have demonstrated the ability to inhibit vascular endothelial growth factor-mediated and basic fibroblast growth factor-mediated angiogenesis in vivo.57 Further evidence from clinical trials in patients with cancer is needed to confirm these findings and to further clarify the potential impact of LMWH on the natural history of the disease. Currently, anticoagulants, including the LMWHs, are not indicated for use as anticancer treatment.13

Practical aspects of low-molecular-weight prophylaxis

Although this review is focused on LMWH, vitamin K antagonists and UFH continue to be used in some settings. Extended-duration anticoagulant treatment often is recommended to reduce the risk of recurrence of VTE in patients with cancer. Although it is less costly, long-term treatment with vitamin K antagonists not only may be less effective but also may be impractical for many patients.59 Unpredictable anticoagulant responses can result from warfarin treatment because of multiple food and drug interactions, and this is particularly likely in cancer patients who are receiving multiple additional medications, including chemotherapeutic agents and antiemetics.31 Responses to vitamin K antagonists are affected by liver dysfunction, borderline vitamin K deficiency, and gastrointestinal disorders (eg, vomiting, diarrhea), which are commonly observed among patients who are receiving treatment for cancer. Vitamin K antagonists are also difficult to manage for patients who need anticoagulant interruptions because of chemotherapy-induced thrombocytopenia or invasive procedures, such as spinal taps, paracentesis, and various surgical procedures.60

Unlike vitamin K antagonists, LMWHs have a longer half-life, greater bioavailability, and a more predictable anticoagulant effect, and dose monitoring and adjustment normally are required only in patients with severe renal impairment or obesity.61 A reduced need for regular coagulation monitoring for the majority of patients who receive LMWH therapy makes these agents suitable for outpatient treatment62 and extended-duration VTE prophylaxis.42 Outpatient treatment and freedom from coagulation monitoring offer improved convenience for cancer patients. LMWH therapy appears to be cost-effective for long-term, secondary prophylaxis in cancer patients, because higher drug costs appear to be offset in part by the potential for reduced hospital stays, reduced need for coagulation monitoring, and fewer bleeding complications.63

Unfortunately, there are very limited data on the use of UFH for primary prophylaxis in either hospitalized or ambulatory cancer patients. Prophylaxis studies of UFH in acutely ill hospitalized medical patients have yielded variable results, but none have been conducted specifically in a cancer population.64, 65 Although UFH still is used in some settings, the ASCO VTE guidelines recommend LMWHs for both the initial treatment and the extended treatment of VTE in cancer patients with established VTE. To the author's knowledge, there have been no trials of UFH for primary prophylaxis in ambulatory cancer patients.

Clinical Practice Guidelines

Despite compelling evidence, a significant proportion of medical cancer patients who are at increased risk for VTE do not receive appropriate VTE prophylaxis. In recognition of the importance of VTE prevention in patients with cancer, specific VTE management guidelines recently have been developed by ASCO13 and the National Comprehensive Cancer Network,12 in addition to the updated recommendations from the ACCP9 and the International Union of Angiology.11 A summary of the ASCO guideline recommendations is presented in Table 5,13 and algorithms for the prevention and treatment of VTE for patients with cancer are shown in Figure 4.13, 66 ASCO recommendations include: 1) all hospitalized cancer patients should be considered for VTE prophylaxis with anticoagulants in the absence of bleeding or other contraindications; 2) routine prophylaxis of ambulatory cancer patients with anticoagulation is not recommended, with the exception of patients with myeloma who are receiving thalidomide or lenalidomide with chemotherapy or dexamethasone; 3) patients undergoing major surgery for malignant disease should be considered for pharmacologic thromboprophylaxis; 4) LMWH represents the preferred agent for both the initial treatment and the extended treatment of cancer patients with established VTE; and 5) determining the impact of anticoagulants on the survival of patients with cancer requires additional study and currently cannot be recommended (Table 5).13, 67 Widespread, active dissemination of these guidelines is needed to improve appropriate prescribing rates for prophylactic anticoagulation in patients with cancer who are at risk of VTE. Improved thromboprophylaxis in medical cancer patients should significantly reduce patient morbidity and consumption of healthcare resources while improving the delivery of cancer therapy and cancer-related outcomes, including, most importantly, the survival of patients with cancer.68

Figure 4.

Algorithms for the prevention and treatment of venous thromboembolism (VTE) according to American Society of Clinical Oncology (ASCO) are illustrated for patients with cancer: (Top) Primary prophylaxis, (Bottom Left) secondary prophylaxis, and (Bottom Right) contraindications applicable to primary and secondary prophylaxis are illustrated. Tx indicates treatment; LMWH, low-molecular-weight heparin; VTE, venous thromboembolism; IVC, inferior vena cava; GI, gastrointestinal. From American Society of Clinical Oncology 2007 Clinical practice guideline recommendations for venous thromboembolism prophylaxis and treatment in patients with cancer. J Oncol Pract. 2007;3:326-329.66 Reprinted with permission. © 2008 American Society of Clinical Oncology. All rights reserved.

Table 5. Summary of American Society of Clinical Oncology 2007 Guidelines on Venous Thromboembolism Prophylaxis and Treatment in Patients With Cancer*
Patient GroupRecommendedNot Recommended
  • VTE indicates venous thromboembolism; LMWH, low-molecular-weight heparin; UFH, unfractionated heparin; INR, international normalized ratio.

  • *

    Adapted with permission from Lyman GH, Khorana AA, Falanga A, et al. American Society of Clinical Oncology guideline: recommendations for venous thromboembolism prophylaxis and treatment in patients with cancer. J Clin Oncol. 2007;25:5490-5505.13 © 2007 American Society of Clinical Oncology. All rights reserved.

Hospitalized patients with cancerVTE prophylaxis with anticoagulants (LMWH, UFH, or fondaparinux)If presence of bleeding or other contraindications to anticoagulation
Ambulatory patients with cancer who are receiving chemotherapyLMWH or adjusted-dose warfarin for patients with multiple myeloma receiving thalidomide or lenalidomide plus chemotherapy or dexamethasoneOtherwise, no routine VTE prophylaxis
Patients with cancer who are undergoing surgeryProphylaxis with low-dose UFH, LMWH, or fondaparinux for at least 7-10 d; combined prophylaxis with mechanical methods for patients at very high riskIf presence of bleeding or other contraindications to anticoagulation
Combined prophylaxis with mechanical methods for patients at very high riskConsider mechanical methods alone for those with contraindications to pharmacologic methods
Patients who have cancer with established VTELMWH for the initial 5-10 d 
LMWH for ≥6 mo or vitamin K antagonists (target INR, 2-3) when LMWH unavailable 
Consider continued anticoagulation beyond 6 months in those with active cancer 
Patients who have cancer without VTE, to improve survivalProphylaxis not recommended 

Further Studies

Further study of the potential roles for these agents in patients with cancer is clearly needed. Very few data are available on the prevention of VTE in ambulatory patients with cancer. Studies also are needed to better define the optimum dose and duration of LMWH therapy in specific clinical settings in cancer patients.49, 69 Additional studies are needed to better define the benefits and risks associated with prolonged anticoagulation, especially in high-risk patients, such as the elderly or those with central nervous system malignancies.13 Finally, although intriguing data have emerged from several trials, further research is needed to explore the potential impact of the LMWHs on the survival of patients with cancer.


Although the value of primary prophylaxis with LMWHs in patients with cancer who undergo surgical treatment is well established, medical cancer patients also represent a population at significant risk for VTE and its complications. The risk of VTE in this population is increasing in frequency, in part because of treatment with more aggressive systemic cancer therapies, including new, targeted antiangiogenic agents. The LMWHs have demonstrated clinical efficacy in medical patients with cancer, including as primary VTE prophylaxis in high-risk patients, for secondary prevention of recurrent episodes of VTE, and potentially for improvement in overall survival. The LMWHs appear to be suitable for long-term, secondary prophylaxis as a result of the reduced need for coagulation monitoring, low rates of bleeding complications, and once-daily dosing. Clinical practice guidelines from ASCO and other professional organizations have provided recommendations for the appropriate and safe use of VTE prophylaxis for the high-risk cancer patient.


I also acknowledge the countless discussions and many useful suggestions provided by Dr. Nicole Kuderer in the preparation of this article

Conflict of Interest Disclosures

Dr. Lyman is supported by a grant from the National Heart, Lung and Blood Institute (1R01HL095109-01). The author received no compensation for the preparation of this article. The author received editorial support in the preparation of this manuscript, funded by Sanofi-Aventis, Bridgewater, NJ. However, the author had complete independence in defining content and in all editorial decisions with respect to this article.