Preventing VTE in Cancer Outpatients: Are We There Yet?

Contemporary Oncology®Fall 2010
Volume 2
Issue 3

Cancer-associated thrombosis occurs commonly in patients with cancer, particularly during treatment with anticancer therapies.

Cancer-associated thrombosis occurs commonly in patients with cancer, particularly during treatment with anticancer therapies. Recent reports suggest a steep increase in the incidence of cancer-related thrombotic events starting in the late 1990s.1,2 Venous thromboembolism (VTE), which includes deep vein thrombosis (DVT) and pulmonary embolism (PE), has important consequences for patients with cancer: a requirement for long-term anticoagulation, a 12% annual risk of bleeding complications, a 21% annual risk of recurrent VTE,3 and the potential to affect chemotherapy delivery and quality of life. Most importantly, thrombotic events are the second leading cause of death in patients with cancer (after cancer itself) and are associated with decreased short-term and long-term survival.4-6 Thus, it is crucial to reduce the occurrence of VTE, especially considering that there are multiple available agents for thromboprophylaxis, including low-molecular-weight heparins (LMWHs) and warfarin.


The focus of thromboprophylaxis studies in the past has largely been in the hospital inpatient and postsurgical settings. Although these continue to be important settings, an increasing proportion of VTE events occur in the outpatient setting.7 Recent studies of thromboprophylaxis conducted primarily in the outpatient setting have implications for clinical practice.


PROTECHT (Prophylaxis of Thromboembolism During Chemotherapy Trial) was a multicenter, randomized, placebo-controlled, double-blind study in which prophylaxis was studied in patients with high-risk sites of cancer, including locally advanced or metastatic lung, gastrointestinal, pancreatic, breast, ovarian, and head/neck cancers, who were actively receiving chemotherapy.8 Patients receiving adjuvant or neoadjuvant chemotherapy, patients actively receiving antithrombotic treatment, and those with contraindications to anticoagulation were excluded.

Patients received either injections of nadroparin (an LMWH approved in Europe) or placebo in a 2:1 randomization ratio for the duration of chemotherapy or for a maximum of 4 months. The primary outcome in the trial was a combined endpoint of symptomatic VTE of the extremities, peripheral arterial thromboembolism, PE, cerebral or visceral venous thrombosis, myocardial infarction, ischemic stroke, and death secondary to a thromboembolic event. The study did not screen for asymptomatic events.

Overall, 2% of patients in the treatment group and 3.9% (n = 15) in the placebo group developed a thromboembolic event (one-sided; 95% CI, 0.303%; P = .02). DVTs comprised the majority of events in both arms, followed by PE, visceral venous thrombosis, and stroke/peripheral thrombosis (Table 1). No survival benefit was observed. Patients in the treatment arm were more likely than patients in the control arm to experience major bleeding (0.17% vs 0%, respectively; P = .18).8

This is the largest study of thromboprophylaxis conducted in outpatients with cancer to date, and it represents a significant advance in the prevention of VTE in this population. Although the study met its primary endpoint, the overall low event rate has not led to recommendations for the use of prophylaxis in outpatients. It is quite possible that the event rate is higher in specific subgroups of patients with cancer and that the risk-benefit ratio would have been more optimal for prophylaxis if higher-risk patients had been selected using either biomarkers or risk-assessment tools, which will be discussed later in this article.


It has been well established that patients with pancreatic cancer are at particularly high risk for VTE.9 The PROSPECTCONKO 004 (Prospective Randomized Trial of Simultaneous Pancreatic Cancer Treatment With Enoxaparin and Chemotherapy) study was a prospective, open, randomized multicenter phase III trial that evaluated thromboprophylaxis in patients with pancreatic cancer receiving gemcitabine (Gemzar)-based chemotherapy. The study was designed after an initial pilot study determined that adding enoxaparin to a chemotherapy regimen of gemcitabine, 5-fluorouracil, folinic acid, and cisplatin (GFFC) was safe and efficacious in advanced pancreatic cancer.10

Individuals with confirmed advanced pancreatic cancer and no prior chemotherapy or recent VTE events were eligible for enrollment. Patients were stratified according to Karnofsky performance status and renal function. Those with normal renal function and good performance status (>80%) received GFFC chemotherapy, whereas patients with elevated plasma levels of creatinine and worse performance status received gemcitabine alone. Patients were randomized to observation or prophylactic enoxaparin, receiving 1 mg/kg per day for 3 months, then 40 mg daily. The primary endpoint was reduction of symptomatic VTE. Time to progression (TTP) and overall survival (OS) were secondary outcomes.

VTE occurred in 5.0% (8 of 160) of patients in the enoxaparin arm compared with 14.5% (22 of 152) of patients in the observation arm (P <.01). The incidence of major bleeding was paradoxically higher in the observation arm, although this was not statistically significant (6.3% vs 9.9%, respectively; P = .18). No differences were observed between the groups in median OS or TTP.11 The results of this study show that thromboprophylaxis can safely and clinically produce a significant reduction in the rate of VTE in patients with pancreatic cancer.

The dose of LMWH used in this setting was higher than the commonly accepted prophylaxis dose of 40 mg per day, although it was not as high as the therapeutic dose of 1 mg/kg twice daily. Given the unique hypercoagulability of patients with cancer, it is quite possible that higher doses of anticoagulants are needed to achieve significant reduction in VTE. One limitation of this study was the use of the GFFC combination, which is not considered standard of care; one of the regimen’s components—cisplatin&mdash;is known to increase the risk of VTE in patients with cancer.12 Full results in manuscript form are awaited.

The FRAGEM Study

In the FRAGEM trial, patients in the United Kingdom with unresectable pancreatic cancer were randomized to gemcitabine alone or combined with daily dalteparin (Fragmin) for 12 weeks.13 The primary outcome was reduction in VTE; secondary outcomes were fatal VTE events, sudden death, and early death burden. An interim analysis was presented in 2009, encompassing data for 59 evaluable patients in the dalteparin arm and 64 evaluable patients in the chemotherapy-only arm. Within the first 100 days after randomization, 25% of patients in the control arm and 3.5% of patients in the dalteparin group experienced a VTE (relative risk [RR], 0.14; 95% CI, 0.03 to 0.58; P = .002). Overall incidence of VTE was 31% in the chemotherapy-only group versus 12% in the dalteparin arm (RR, 0.38; 95% CI, 0.17 to 0.84; P = .019). Further, 9% of patients in the chemotherapy-alone arm experienced a fatal VTE or sudden death compared with none of the patients in the thromboprophylaxis arm (RR, 0.08; 95% CI, 0.005-1.45; P = .028).14 Final results and a full manuscript from this study are pending, but the reported data are consistent with the known high rate of VTE in patients with pancreatic cancer and with the results of the CONKO 004 study.

Trial of Prophylaxis in Patients With Myeloma

Patients with myeloma are at high risk for experiencing a VTE. This is particularly true of patients receiving thalidomideor lenalidomide (Revlimid)-based combination regimens. VTE rates of up to 34% have been observed in patients on these therapies.15,16

A recent prospective study addressed the efficacy of thromboprophylaxis in 991 newly diagnosed myeloma patients.17 Since there is still controversy regarding the most appropriate thromboprophylaxis regimen, the authors conducted a prospective, multicenter phase III trial evaluating the safety and efficacy of LMWH, low-dose aspirin (ASA), or low-intensity, fixed-dose warfarin as anticoagulant prophylaxis. The endpoints were incidences of VTE, acute cardiovascular events, sudden death, and major and minor bleeding. The study population comprised 991 patients with newly diagnosed myeloma who were randomized to one of 4 regimens (Table 2): Velcade, thalidomide, and dexamethasone (VTD); thalidomide and dexamethasone (TD); Velcade, melphalan, prednisone, and thalidomide (VMPT); or Velcade, melphalan, and prednisone (VMP). In a substudy, patients treated with VTD, TD, or VMPT were randomly assigned to receive 40 mg per day of LMWH (enoxaparin, n = 223), 100 mg per day of ASA (n = 227), or 1.25 mg per day of warfarin (n = 223) for the duration of induction therapy. Patients treated with VMP (n = 257) did not receive any prophylaxis and were used as controls; this was the only nonprophylaxis arm.

The incidence of VTE was 5% in the LMWH group, 6% in the ASA group, and 8% in the warfarin group (P value not significant). VTE occurred in only 2% of patients in the VMP nonprophylaxis control arm. The rate of cardiovascular events was 2% in the LMWH group, 1% in the ASA group, and 0.5% in the warfarin group. The incidence of major and minor bleeding was 2% in the LMWH group, 3% in the ASA group, and 1% in the warfarin group (P value not significant). The combined incidence of thrombosis, bleeding, and cardiovascular events was 9% in the LMWH group, 10% in the ASA group, and 9% in the warfarin group (P value not significant). Thus, patients receiving LMWH had a lower risk of VTE, although no statistical difference was observed. The authors concluded that LMWH, warfarin, and ASA are all likely to be effective thromboprophylactic regimens. Current American Society of Clinical Oncology (ASCO) guidelines on VTE prophylaxis in patients with cancer, however, recommend against the use of aspirin in this setting.18

Meta-Analysis of Prophylaxis Studies in Outpatients

Kuderer et al recently presented a meta-analysis investigating LMWH thromboprophylaxis in outpatients with cancer.19 This meta-analysis included all the recent studies discussed herein, with the exception of the myeloma prophylaxis study; older studies utilizing LMWH prophylaxis were also included. The authors identified a total of 7 randomized controlled trials (RCTs) of LMWH use in ambulatory patients with cancer (N = 2960), which consisted of 1685 patients who received LMWH and 1275 controls. The RCTs included 3 that enrolled patients with various solid tumors and 1 each for patients with breast cancer, lung cancer, pancreatic cancer, and glioblastoma.

Patients receiving LMWH experienced 47 VTE events compared with 74 VTE events for control subjects, for crude rates of 2.79% and 5.80%, respectively. The RR for VTE across all trials was estimated at 0.54 (95% CI, 0.38-0.78; P = .001), while the absolute risk decrease was 2.55% (95% CI, 1.06%-4.05%; P <.001). A total of 30 patients receiving LMWH experienced major bleeding events compared with 15 control subjects, for crude rates of 1.78% and 1.18%, respectively. The RR for major bleeding across all trials was estimated at 1.74 (95% CI, 0.95-3.18; P = .071), while the absolute risk increase was 0.75% (95% CI, 0.17%-1.33%; P = .011). The RR for major bleeding across the 5 primary prophylaxis trials was 2.27 (95% CI, 1.12-4.59; P = .022) with the absolute risk increase estimated at 1.27% (95% CI, 0.27%-2.27%; P = .013).

The authors concluded that LMWH thromboprophylaxis in ambulatory patients with cancer is effective and results in a significant 46% reduction in the RR of VTE. The overall risk of VTE is low in this setting, however, and the absolute RR with prophylactic anticoagulation is only 2.6%; the increase in major bleeding events invites concern. This meta-analysis points to the need for risk stratification when selecting patients for thromboprophylaxis.


The results of recent studies strongly suggest a benefit from using thromboprophylaxis in select patient populations. The results from PROTECHT, which had a low rate of VTE events, are more applicable to a general population of patients with cancer. The studies in patients with pancreatic cancer and myeloma demonstrate high event rates, but these data apply to small subsets of patients with cancer. For ambulatory patients with myeloma, current ASCO guidelines recommend prophylaxis only for those receiving thalidomide- or lenalidomide-based regimens and not for other ambulatory populations.18 A consensus statement from major guidelines panels emphasizes a similar selection of outpatients for prophylaxis.20

To reduce the public health burden of cancer-associated VTE, it is crucial to develop ways to stratify cancer patients according to their risk for VTE and identify appropriate prophylactic regimens. A risk assessment model that applies to a broad population of patients with cancer was created (Table 3)21 after studying risk factors for VTE in a development cohort of 2701 ambulatory patients from a prospective registry. A risk score for VTE was devised using regression coefficients estimated in the multivariate model. The score was then validated in an independent cohort of 1365 patients from the same study.

The stage-adjusted multivariate model identified 5 predictive variables as significant risk factors for chemotherapy-associated VTE and categorized risk as low (score of 0), intermediate (score of 1 or 2), or high (score ≥3). Observed rates of VTE in the development and validation cohorts were 0.8% and 0.3%, respectively, in the low-risk category; 1.8% and 2%, respectively, in the intermediate-risk category; and 7.1% and 6.7%, respectively, in the high-risk category.

A recent report from Vienna CATS (Cancer and Thrombosis Study), a trial that enrolled 834 patients with cancer, has now externally validated this model.22 The study determined that the 6-month cumulative probability of developing VTE was 1.5% for patients with a score of 0, compared with 3.8% for patients with a score of 1, 9.4% for patients with a score of 2, and 17.7% for patients with a score ≥3.


Three large RCTs are underway that will add to the available data on thromboprophylaxis for patients with cancer (Table 4). The largest is SAVE-ONCO (Evaluation of AVE5026 in the Prevention of Venous Thromboembolism in Cancer Patients Undergoing Chemotherapy), which is investigating the efficacy of the novel agent semuloparin (AVE5026) at preventing VTE. Patients undergoing chemotherapy for locally advanced or metastatic solid tumors of the lung, gastrointestinal tract, bladder, or ovary are eligible for the study.

The National Institutes of Health is sponsoring PHACS (a Prospective Randomized Multicenter Study of Dalteparin Prophylaxis in High-Risk Ambulatory Cancer Patients) at the University of Rochester in New York and Duke University in Durham, North Carolina. The study uses the risk model discussed above to identify outpatients with cancer who have a high risk of experiencing a VTE during chemotherapy. Patients with a risk score ≥3 are being randomized to 12 weeks of dalteparin or observation.

The third trial, referred to as MicroTEC (Microparticles and Thromboprophylaxis With Enoxaparin in Cancer), is investigating enoxaparin in patients with pancreatic, lung, and colorectal cancer who have elevated plasma levels of tissue factor—bearing microparticles. The results of these novel approaches to risk stratification and prophylaxis will add to the growing information in this area and allow clinicians to individualize decisions regarding thromboprophylaxis for ambulatory patients with cancer.

Related Videos