Version May 2024
INTRODUCTION — Patients with malignancy are at high risk of developing venous thromboembolism (VTE; deep venous thrombosis and pulmonary embolism). Thrombotic events are the second leading cause of death in cancer patients after death from cancer itself [1]. Thus, timely treatment with anticoagulant therapy is critical for managing this population.
Anticoagulation for acute VTE in patients with malignancy is discussed in this topic. Most of the recommendations for anticoagulation in this topic assume that treatment is in accordance with the patient's preference with regard to goals of care and life expectancy. The pathogenesis, epidemiology, and prevention of VTE in patients with malignancy are discussed separately. (See "Overview of the causes of venous thrombosis", section on 'Malignancy' and "Risk and prevention of venous thromboembolism in adults with cancer", section on 'Primary prevention'.)
PRINCIPLES OF THERAPY — The general principles of managing acute VTE in patients with cancer are the same as those for patients without cancer, with anticoagulant therapy as the cornerstone of treatment. However, treatment of VTE in patients with cancer has higher than usual rates of both VTE recurrence and bleeding events (table 1). In addition, therapy can be complicated by comorbidities, procedures, and medication interactions [2-9].
In general, our recommendations are consistent, for the most-part, with several evidence-based society guidelines [10-20].
AGENT CHOICE — Most patients with cancer-associated VTE are treated with anticoagulation unless there is active bleeding or a high risk of bleeding. While many are treated as inpatients, there is a growing population of patients that are suitable for outpatient therapy, which can affect agent selection. Selecting patients and a suitable anticoagulant for outpatient therapy are discussed separately. (See 'Outpatient therapy' below.)
Factors that influence agent choice — Many factors found in patients without malignancy also impact agent selection in patients with malignancy (table 2), and additional factors in patients with malignancy can affect the decision to initiate, continue, or stop anticoagulation (table 1).
Examples include but are not limited to the following:
●Cost and convenience
●Contraindications to the chosen agent (eg, allergy or adverse effects)
●The presence of renal or liver failure
●Local availability
●Desire to avoid injections
●Ability to take and absorb oral or subcutaneous (SC) medications
●Anticipated discontinuation or reversal of anticoagulation (eg, for procedures)
●Clot burden
●Special conditions (eg, pregnancy, heparin-induced thrombocytopenia (see 'Special populations' below))
●Cancer type and stage
●History of bleeding or bleeding diathesis
●Concurrent arterial thrombosis
●Once-daily versus twice-daily dosing
●Patient preferences and alignment with patient goals of care
●Life expectancy
Decision-making should be shared, taking into account the benefits and risks of anticoagulation with medications, life expectancy, and patient preference. For example, many patients may prefer to take an oral anticoagulant over daily injections or wish to avoid certain agents because of cost. On the other hand, the use of warfarin and direct oral anticoagulants (DOACs) in patients with cancer can be complicated by drug interactions and diminished absorption or poor oral intake, or troublesome laboratory draws (for warfarin only). As another example, some patients willingly take therapy in the context of reasonable survival, while patients in a hospice setting may prefer to avoid or stop therapy [21]. Further details regarding goals of care are provided separately. (See "Discussing goals of care".)
Estimating the bleeding risk in cancer-associated thrombosis is challenging. One study reported poor performance of several bleeding risk models [22]. (See "Selecting adult patients with lower extremity deep venous thrombosis and pulmonary embolism for indefinite anticoagulation", section on 'Bleeding risk models'.)
Selecting among the agents — Options for anticoagulation include SC low molecular weight (LMW) heparins (enoxaparin, dalteparin, or tinzaparin), oral factor Xa inhibitors (ie, apixaban, edoxaban, and rivaroxaban), the oral thrombin inhibitor dabigatran, intravenous (IV) or SC unfractionated heparin (UFH), the synthetic SC pentasaccharide fondaparinux, and the oral vitamin K antagonist warfarin.
For most hemodynamically stable patients with cancer-associated VTE who do not have severe renal insufficiency (eg, creatinine clearance <30 mL/minute) or a contraindication, we prefer an oral factor Xa inhibitor rather than LMW heparin or IV UFH followed by warfarin. Factor Xa inhibitors are frequently chosen to avoid the daily injections required for LMW heparin, although efficacy is equivalent between these agents and other factors may also weigh into this decision (table 1 and table 2). (See 'Factors that influence agent choice' above and 'Hemodynamically unstable patients' below and 'Cancer-specific issues' below.)
●Factor Xa inhibitors – Among the factor Xa inhibitors, apixaban is our preferred monotherapy option (ie, administered without an initial five days of parenteral heparin). This preference is based upon a single randomized trial that reported similar efficacy and safety compared with LMW heparin [23] and growing experience with this agent in this population; also, apixaban may carry a lower bleeding risk, particularly from the gastrointestinal (GI) tract, as compared with other oral factor Xa inhibitors (ie, rivaroxaban, edoxaban) [24]. (See 'Direct oral anticoagulant mono- or dual therapy' below.)
Rivaroxaban is another monotherapeutic factor Xa inhibitor alternative, but has only been prospectively compared with LMW heparin in a pilot study [25]. (See 'Direct oral anticoagulant mono- or dual therapy' below.)
As another alternative, edoxaban can be used, but was studied for the treatment of acute VTE following initial treatment with five days of LMW heparin [26]. (See 'Direct oral anticoagulant mono- or dual therapy' below.)
Impacting agent selection among the oral factor Xa inhibitors are several additional issues. For example, rivaroxaban and edoxaban are once-daily preparations, while apixaban is a twice-daily preparation. We tend to avoid edoxaban and rivaroxaban in those with upper GI cancers since more bleeding has been reported with these agents compared with apixaban or LMW heparin. Rivaroxaban must be taken with food to maximize its absorption.
The efficacy of factor Xa inhibitors is discussed separately. (See 'Direct oral anticoagulant mono- or dual therapy' below.)
●LMW heparin – Any of the LMW heparins, enoxaparin, dalteparin, or tinzaparin can be used. LMW heparin use as monotherapy is based on randomized trials typically comparing dalteparin or tinzaparin with warfarin as well as our extensive experience using chronic LMW heparin in cancer-associated thrombosis. (See 'Low molecular weight heparin' below.)
●Direct oral thrombin inhibitors (Dabigatran) – We generally do not use dabigatran in this population. While dabigatran has been compared with warfarin in patients without cancer, it has not been compared with chronic LMW heparin in patients with active cancer.
First-line options — Oral factor Xa inhibitors are the preferred first-line agents for anticoagulation therapy in patients with cancer-associated thrombosis. LMW heparin is a suitable alternative with equivalent efficacy and long-standing history of use. The data to support these agents are discussed in the sections below. (See 'Direct oral anticoagulant mono- or dual therapy' below and 'Low molecular weight heparin' below.)
Direct oral anticoagulant mono- or dual therapy — Among the DOACs, oral factor Xa inhibitors have proven to be efficacious in the treatment of cancer-associated VTE as compared with LMW heparin, which was the previous standard of care, and with warfarin. In addition, clinicians have gained experience and have become comfortable using factor Xa inhibitors (eg, apixaban, edoxaban, rivaroxaban) in this patient population.
●DOACs versus LMW heparin – LMW heparin has been compared with apixaban [23,27-29], edoxaban [26], and rivaroxaban [25,30,31]. Dabigatran has not been specifically studied in cancer-associated thrombosis. On balance, oral factor Xa inhibitors are associated with similar or lower VTE recurrence rates with similar or slightly increased major bleeding rates compared with LMW heparin. In addition, they may be comparatively more cost-effective [32].
•Meta-analyses – Several meta-analyses have examined the efficacy and safety of factor Xa inhibitors compared with LMW heparin [11,33-37]. As an example, one meta-analysis of four trials that included 2894 patients with active cancer reported that DOACs lowered the risk of recurrent VTE compared with dalteparin (odds ratio [OR] 0.59, 95% CI 0.41-0.86) without increasing the risk of major bleeding (OR 1.34, 95% CI 0.83-2.18) [37]. Apixaban and rivaroxaban appeared to have the lowest risk of VTE recurrence, and edoxaban was associated with an increased risk of major bleeding (OR 1.73, 95% CI 1.04-3.16). Another meta-analysis of the same four studies reported similar results (recurrent VTE in 5.2 versus 8.2 percent and major bleeding in 4.3 versus 3.3 percent for factor Xa inhibitors and LMW heparin, respectively) [35]. Mortality was no different.
•Apixaban – In a randomized trial (CARAVAGGIO) of 1170 patients with active cancer and VTE, two-thirds of whom had locally advanced or metastatic disease, apixaban administered as monotherapy (ie, no initial treatment with LMW heparin; 10 mg twice daily for seven days followed by 5 mg twice daily for six months) was compared with dalteparin (200 international units [IU]/kg SC for one month followed by 150 IU/kg once daily) [23]. Similar rates of VTE were seen in those receiving apixaban or LMW heparin (5.6 versus 7.9 percent, respectively; hazard ratio [HR] 0.63, 95% CI 0.37-1.07) without any significant difference in major bleeding (3.8 versus 4 percent; HR 0.82, 95% CI 0.40-1.69).
Among those with GI cancer, data are conflicting regarding the bleeding risk associated with apixaban compared with other agents. As examples:
-A post-hoc analysis of CARAVAGGIO reported no difference in the rates of bleeding among those with GI cancer (1.9 versus 1.7 percent) [28]. The number of patients with upper GI or pancreatic/hepatobiliary cancers in CARAVAGGIO, however, was limited (4 or 7.6 percent on apixaban and 5.4 or 7.4 percent on dalteparin, respectively).
-Another retrospective study suggested that apixaban had higher rates of GI bleeding among those with luminal GI cancers compared with non-GI cancer [29].
-The risk of major bleeding was lower in CARAVAGGIO than in the studies using rivaroxaban and edoxaban (see rivaroxaban and edoxaban bullets below).
•Edoxaban – An open-label noninferiority randomized trial (Hokusai VTE Cancer Investigators) of 1050 patients with cancer-associated VTE compared anticoagulation with LMW heparin (dalteparin 200 IU/kg for one month followed by 150 IU/kg for up to 12 months), with LMW heparin for five days followed by the oral factor Xa inhibitor edoxaban (60 mg once daily). Although there was no difference in the composite outcome of recurrent VTE and major bleeding, anticoagulation for up to 12 months (minimum six months) with edoxaban resulted in a nonsignificant reduction in the recurrence rate as compared with dalteparin (7.9 versus 11.3 percent, respectively), but at the expense of a higher rate of major bleeding (6.9 versus 4.0 percent, respectively) [26]. However, the rate of more serious bleeding events was equal between the groups, and most of the major bleeds occurred in patients with upper GI cancers. Death was no different, occurring in 40 percent in the edoxaban group and in 37 percent in the dalteparin group.
•Rivaroxaban – A pilot study (SELECT-D) randomized 406 patients to rivaroxaban (15 mg twice daily for three weeks followed by 20 mg once daily with food) given as monotherapy (ie, no initial treatment with LMW heparin) or to LMW heparin (dalteparin; 200 IU/kg daily during month 1, then 150 IU/kg daily for months 2 to 6) [25]. Rivaroxaban resulted in lower rates of recurrent VTE at six months (4 versus 11 percent). While the rate of major bleeding was similar (6 versus 4 percent), there was an excess number of clinically relevant nonmajor clinically relevant bleeds with rivaroxaban (13 versus 4 percent), especially in those with cancers of the upper GI tract.
•DOACs as a group – One randomized trial of 671 patients with VTE-related cancer (CANVAS) compared DOACs (mostly apixaban or rivaroxaban) with LMW heparin (mostly enoxaparin) [38]. Rates of recurrent VTE at six months were similar between the groups (6.1 [DOACs] versus 8.8 percent [LMW heparin]). Rates of major bleeding were also similar (5.2 [DOACs] versus 5.6 percent [LMW heparin]). However, there were more clinically significant nonmajor bleeding events in patients receiving DOACs compared with LMW heparin (5.8 versus 2.6 percent). Mortality was no different between treatment groups. While the trial was pragmatic, it was unblinded, randomization was relatively late (within 30 days of the diagnosis), and adherence rates were lower with LMW heparin, which could have biased the results in favor of DOACs. In addition, an assessment of the risks and benefits of specific DOACs as compared with LMW heparin is unclear.
●DOACs (factor Xa inhibitors) versus warfarin – There are limited data comparing the efficacy and safety of DOACs and warfarin in the treatment of cancer-associated VTE. The data are largely derived from the initial randomized trials that compared oral factor Xa inhibitors with LMW heparin followed by warfarin. However, many of the included patients had a previous history of rather than "active" cancer. These trials are discussed separately. (See "Venous thromboembolism: Anticoagulation after initial management", section on 'Direct thrombin and factor Xa inhibitors'.)
Limited direct data comparing DOACS with warfarin include the following:
•In a meta-analysis of five studies enrolling 982 participants, DOACs reduced the rate of recurrent VTE when compared with warfarin (relative risk [RR] 0.66, 95% CI 0.33-1.31; 27 fewer to 12 more per 1000; low-certainty evidence) without any effect on survival or bleeding [39].
•In another meta-analysis of six studies that included 1132 patients with deep venous thrombosis (DVT) and cancer, DOACs resulted in similar rates of VTE recurrence compared with warfarin (4 versus 6 percent; OR 0.63, 95% CI 0.37-1.10) and major bleeding (OR 0.77, 95% CI 0.41-1.44) [40].
•Two separate pooled analyses of patients with DVT who were treated with rivaroxaban (EINSTEIN trials) [41] or dabigatran (RECOVER trials) [42] reported rates of recurrence and bleeding that were no different between patients with cancer who were treated with DOACs or warfarin. However, most of the "cancer" patients had completed therapy or had a past history of cancer rather than "active" cancer upon entry into the trials. Thus, there was a selection bias such that patients in these trials had a lower risk of recurrent thrombosis from the outset.
•A post-hoc analysis of the AMPLIFY trial reported outcomes in patients who had cancer and were treated with apixaban as compared with standard anticoagulation (ie, LMW heparin followed by warfarin) [43]. Among those with active malignancy receiving apixaban, similar rates of recurrent VTE (4 versus 6 percent; RR 0.56, 95% CI 0.13-2.17) and bleeding (2 versus 5 percent; RR 0.45, 95% CI 0.08-2.46) were reported.
●DOAC dosing – Doses of the factor Xa inhibitors are similar in patients with cancer-associated thrombosis to those in patients without cancer. (See "Venous thromboembolism: Anticoagulation after initial management", section on 'Direct thrombin and factor Xa inhibitors'.)
Low molecular weight heparin — LMW heparin is a monotherapeutic agent with long standing established efficacy in patients with cancer-associated VTE. LMW heparin has been compared with DOACs, unfractionated heparin (UFH), and warfarin. No study has directly compared different LMW heparins with each other. However, in clinical practice in the United States, dalteparin or enoxaparin are the most commonly used agents. (See "Heparin and LMW heparin: Dosing and adverse effects" and "Prevention of venous thromboembolism in adults undergoing hip fracture repair or hip or knee replacement".)
●LMW heparin versus DOACS – Trials that compared LMW heparin with a DOAC are discussed above. (See 'Direct oral anticoagulant mono- or dual therapy' above.)
●LMW heparin and IV UFH – Data consistently show similar efficacy of LMW heparin compared with UFH as an initial anticoagulant for the treatment of cancer-associated VTE. In a meta-analysis of 15 randomized controlled trials, LMW heparin as initial therapy in cancer-related VTE was associated with a possible reduction in mortality at three months as compared with UFH (RR 0.66, 95% CI 0.40-1.1; 57 fewer deaths per 1000, 95% CI 101 fewer to 17 more; moderate-certainty evidence) [44]. There was no difference in the risk of recurrent VTE or bleeding events.
●LMW heparin versus warfarin – Randomized trials and meta-analyses have demonstrated superiority of LMW heparin over warfarin [11,26,39,44-54].
•The largest of the meta-analyses (eight randomized trials of 2327 patients) reported reduced rates of recurrent VTE with LMW heparin compared with warfarin (risk ratio 0.58, 95% CI 0.43-0.77), a benefit that occurred without a survival advantage [39]. In addition, no difference in major bleeding events or reports of thrombocytopenia were reported between the groups. However, the quality of evidence was low for mortality, major bleeding, and minor bleeding, and moderate for recurrent VTE.
•Two sentinel randomized trials concur with the meta-analysis above:
-In one randomized trial (CLOT), dalteparin compared with LMW heparin followed by warfarin resulted in a reduction in the rate of recurrent VTE at six months (9 versus 17 percent) without an increase in the rate of major bleeding (6 versus 4 percent) or overall mortality (39 versus 41 percent) [46].
-In another trial (CATCH), tinzaparin compared with LMW heparin followed by warfarin also resulted in a lower rate of recurrent VTE that was not statistically significant (7 versus 11 percent) at six months, without an increase in mortality or rates of major bleeding [52].
●LMW heparin versus fondaparinux – Data suggest that LMW heparin is superior to fondaparinux, the details of which are discussed separately. (See 'Fondaparinux' below.)
●LMW heparin dosing – Dosing of LMW heparin is similar to that in patients without cancer, the details of which are provided separately. (See "Venous thromboembolism: Anticoagulation after initial management", section on 'Low molecular weight heparin'.)
While some early trials suggested that LMW heparin led to a survival advantage in cancer patients with and without VTE [55-60], this potential benefit of LMW heparin has not been confirmed, particularly in the extended-therapy setting [44].
Second-line options — Second line options include UFH (followed by warfarin) and fondaparinux, both of which may be used in select circumstances.
Unfractionated heparin followed by warfarin — UFH and warfarin, while used commonly in the past, have now been surpassed by LMW heparin and DOACs (specifically factor Xa inhibitors), respectively. However, they are still appropriate to use in select circumstances. Examples include the following:
●Patients with severe renal insufficiency – (see 'Renal failure' below and "Venous thromboembolism: Initiation of anticoagulation", section on 'Renal failure').
●Hemodynamically unstable patients (UFH) – (see 'Hemodynamically unstable patients' below and "Venous thromboembolism: Initiation of anticoagulation", section on 'Hemodynamic instability').
●Patients with marked obesity (eg, >150 kg) – (see 'Obesity' below and "Venous thromboembolism: Initiation of anticoagulation", section on 'Obesity or poor subcutaneous absorption').
●Patients in whom discontinuation or reversal of anticoagulation is anticipated (UFH) – (see 'Anticipated discontinuation or reversal of anticoagulation' below and "Venous thromboembolism: Initiation of anticoagulation", section on 'Anticipated need for discontinuation or reversal').
●Patients with an extensive clot burden (eg, massive pulmonary embolism [PE], phlegmasia cerulea dolens) (UFH) – (see 'Extensive clot burden' below and "Venous thromboembolism: Initiation of anticoagulation", section on 'Extensive clot burden').
●Patients in whom tight regulation of anticoagulation using readily available tests (ie, partial thromboplastin time or anti-factor Xa levels for heparin and prothrombin time/international normalized ratio [for warfarin]) is required (eg, those with a bleeding diathesis in whom an atypical range of anticoagulation is targeted).
Data comparing UFH and warfarin with LMW heparin are discussed above. (See 'Low molecular weight heparin' above.)
Dosing of UFH and warfarin is similar to that in patients without cancer, the details of which are provided separately. (See "Venous thromboembolism: Initiation of anticoagulation", section on 'Unfractionated heparin' and "Venous thromboembolism: Initiation of anticoagulation", section on 'Transitioning to maintenance therapy' and "Venous thromboembolism: Anticoagulation after initial management", section on 'Warfarin'.)
Fondaparinux — Fondaparinux, a synthetic pentasaccharide, is infrequently used in patients with cancer-associated VTE but may be considered in those with heparin-induced thrombocytopenia who need parenteral anticoagulation. (See "Management of heparin-induced thrombocytopenia", section on 'Fondaparinux'.)
In patients with cancer, fondaparinux has only been compared with LMW heparin as an initial anticoagulant prior to the administration of warfarin and has not been compared with LMW heparin in the extended-treatment setting. In a post-hoc analysis of two randomized trials of acute cancer-associated VTE, patients treated initially with fondaparinux (followed by vitamin K antagonists) had higher rates of recurrence at three months with no difference in bleeding rates, when compared with enoxaparin (followed by vitamin K antagonists; 12.7 versus 5.5 percent) [61].
Dosing of fondaparinux is similar to that in patients without cancer, the details of which are provided separately. (See "Venous thromboembolism: Anticoagulation after initial management", section on 'Fondaparinux' and "Fondaparinux: Dosing and adverse effects".)
SPECIAL POPULATIONS — Principles of anticoagulant selection in special populations of patients with cancer-associated VTE are similar to that in patients without cancer. Agent choice in these populations is mostly discussed in the linked sections below.
Contraindications: Inferior vena cava filter — When therapeutic anticoagulation is absolutely contraindicated (eg, active bleeding), we evaluate patients for placement of an inferior vena cava (IVC) filter. Importantly, an IVC filter should not routinely be placed for the initial management of VTE in any population, including patients with intracranial malignancy. Importantly, the decision to place a filter should weigh the benefit of preventing death from PE against the risks of recurrent DVT, since the risk of recurrent DVT is higher in those in whom a filter is placed in the absence of anticoagulation. (See 'Management of recurrence' below and 'Bleeding risk' below.)
Although retrospective cohort studies have found a lower fatality rate in some patients with cancer-associated VTE who had IVC filters compared with those with cancer-associated VTE who did not have an IVC filter placed [62,63], there are no randomized trials that have studied the safety and efficacy of IVC filters in the treatment and prevention of VTE in patients with cancer.
Studies performed in the general population have reported an increased risk of DVT (see "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)", section on 'Inferior vena cava filter'). Observational studies of patients with cancer-associated VTE report a similar increased DVT risk with IVC filters. As examples:
●In a retrospective analysis of patients, patients with cancer-associated DVT in whom an IVC filter was placed had a higher rate of recurrence compared with those who did not have a filter placed (32 versus 17 percent) [64].
●In a retrospective series of 274 patients undergoing surgery for ovarian cancer, one-half of whom had a newly diagnosed VTE, patients who had a filter placed for VTE had an increased rate of recurrent DVT (25 versus 7.2 percent) and decreased overall survival (22 versus 47 months) compared with patients who had VTE in whom no filter was placed [65].
Renal failure — For patients with cancer-associated VTE who have severe renal insufficiency (eg, creatinine clearance <30 mL/minute) and in whom renally dosed anticoagulants are contraindicated, IV UFH followed by warfarin is typically used. (See "Venous thromboembolism: Initiation of anticoagulation", section on 'Renal failure' and "Venous thromboembolism: Anticoagulation after initial management", section on 'Warfarin' and "Anticoagulation for continuous kidney replacement therapy" and "Heparin and LMW heparin: Dosing and adverse effects" and "Warfarin and other VKAs: Dosing and adverse effects".)
For those unable to ingest or absorb warfarin, renally dosed LMW heparin with monitoring of anti-factor Xa levels or SC UFH are alternatives.
Hemodynamically unstable patients — IV UFH is our preferred anticoagulant in those who are hemodynamically unstable since thrombolysis, an interventional procedure, or surgery may need to be considered. (See "Treatment, prognosis, and follow-up of acute pulmonary embolism in adults", section on 'Hemodynamically unstable' and "Approach to thrombolytic (fibrinolytic) therapy in acute pulmonary embolism: Patient selection and administration".)
Anticipated discontinuation or reversal of anticoagulation — For patients in whom a need to discontinue or reverse anticoagulation is anticipated in the near future, we prefer UFH due to its shorter half-life as compared with other anticoagulant therapies. (See "Risks and prevention of bleeding with oral anticoagulants".)
Extensive clot burden — IV UFH is our preferred anticoagulant for those patients with extensive DVT (eg, phlegmasia cerulea dolens), or those with massive or submassive PE. This choice is based upon an anticipated need for a procedural or surgical intervention as well as the ease with which it can be monitored in the laboratory. In addition, DOACs and LMW heparin have not been rigorously evaluated in this population. (See "Treatment, prognosis, and follow-up of acute pulmonary embolism in adults", section on 'Hemodynamically unstable' and "Approach to thrombolytic (fibrinolytic) therapy in acute pulmonary embolism: Patient selection and administration" and "Catheter-directed thrombolytic therapy in deep venous thrombosis of the lower extremity: Patient selection and administration".)
Incidental and small subsegmental pulmonary embolism — Asymptomatic PE and small subsegmental PE (SSPE) are sometimes diagnosed on interval computed tomography (CT) scans done for cancer restaging or other reasons. Despite limited data, we generally consider incidental PE and SSPE in patients with cancer as an indication for therapeutic anticoagulation. This preference is based upon the presumption that in this population in the absence of therapeutic anticoagulation, there is a high incidence of developing symptomatic PE in the future due to clot extension or recurrence.
Symptoms may be more common than originally thought among those diagnosed incidentally by CT. One retrospective review of 59 patients with unsuspected PE discovered during routine staging for cancer reported that 75 percent of patients had symptoms consistent with PE [66].
In patients with incidental or subsegmental PE in association with cancer, limited data suggest a high recurrence rate despite anticoagulation, ranging from 4.3 to 7.9 percent. Although adequate comparison with patients with incidental or subsegmental PE in the absence of cancer have not been published, these rates are higher than that in general population (eg, ≤1 percent on anticoagulation at three months) [63-67].
●In one prospective observational study of 695 patients with active cancer and a diagnosis of incidental PE, the recurrence rate was 6 percent at one year despite anticoagulation, and there was no difference between those with subsegmental or more proximal emboli [64]. The bleeding rate was also high at 6 percent.
●In a post-hoc analysis of a trial that compared edoxaban with dalteparin in patients with cancer, the recurrence rate in patients with incidental PE was 7.9 percent and the bleeding rate was 6.6 percent [65].
●In another post-hoc analysis of a second randomized trial that compared apixaban with enoxaparin in the same population, recurrence rate was 4.3 percent in those who had PE as an incidental finding on imaging and 5.3 percent had major bleeding [63].
Further data on the incidence and management of incidental and subsegmental PE in patients without cancer are provided separately. (See "Treatment, prognosis, and follow-up of acute pulmonary embolism in adults", section on 'Patients with subsegmental PE'.)
Distal deep venous thrombosis — In patients with active cancer, we initiate treatment with anticoagulation, unless there is a contraindication. (See "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)", section on 'Evidence of or risk factors for proximal extension'.)
Data describing outcomes in patients with cancer-associated distal DVT are limited. One registry study of 886 patients with cancer-associated distal DVT reported that the adjusted risks of death, VTE recurrence, or major bleeding were similar in patients with cancer-associated proximal DVT [68]. However, more patients with cancer-associated proximal DVT developed fatal PE (risk difference 0.40 percent, 95% CI 0.23-0.58).
Pregnancy — For pregnant women with acute VTE, adjusted-dose SC LMW heparin is the preferred agent for initial anticoagulation because it has a more favorable safety profile, especially when compared with warfarin and DOACs. (See "Venous thromboembolism in pregnancy and postpartum: Treatment" and "Use of anticoagulants during pregnancy and postpartum".)
Obesity — There is no preferred agent in patients with class 2 or 3 obesity.
Therapeutic anticoagulation can be assured with IV UFH and warfarin. IV UFH is an alternative to SC LMW heparin when SC absorption is potentially poor (eg, massive edema, anasarca).
Doses for LMW heparin in patients with class 2 or 3 obesity are found in the table (table 3), although therapeutic efficacy of such dosing is unclear. (See "Intensive care unit management of patients with obesity", section on 'Anticoagulants' and "Venous thromboembolism: Initiation of anticoagulation", section on 'Obesity or poor subcutaneous absorption'.)
Dosing for DOACs in patients with class 2 or 3 obesity is discussed separately. (See "Venous thromboembolism: Anticoagulation after initial management", section on 'Dosing'.)
Thrombosis in other locations — Management of thrombosis in other locations is discussed separately:
●Patients with catheter-related upper extremity venous thrombosis – (see "Catheter-related upper extremity venous thrombosis in adults", section on 'Treatment').
●Patients with splanchnic or visceral vein thrombus – (see "Epidemiology and pathogenesis of portal vein thrombosis in adults" and "Acute portal vein thrombosis in adults: Clinical manifestations, diagnosis, and management" and "Chronic portal vein thrombosis in adults: Clinical manifestations, diagnosis, and management").
CANCER-SPECIFIC ISSUES
Specific cancers — Certain cancers may require special consideration including:
●Patients with intracranial malignancy who may be at increased risk of significant complications from bleeding – (see "Treatment and prevention of venous thromboembolism in patients with brain tumors").
●Patients with pancreatic and upper gastrointestinal cancer who are at increased risk of recurrence and bleeding, respectively – (see "Supportive care for locally advanced or metastatic exocrine pancreatic cancer", section on 'Venous thromboembolism').
Rates of recurrence and bleeding may vary depending upon the cancer site. In one post-hoc analysis of a randomized trial that compared dalteparin with apixaban [23], rates of VTE recurrence were 11 percent in patients with gynecological, 9 percent with gastrointestinal, 7 percent with genitourinary, and 6 percent with lung cancer [24]; rates were lower rates in the other sites of cancer. Rates of major bleeding were highest in those with genitourinary cancer (7 percent) and gastrointestinal cancer (5 percent). Although rates or recurrence were lower with apixaban at each site, none was statistically significant and rates of bleeding were similar.
Thrombocytopenia — Patients with cancer frequently have thrombocytopenia placing them at increased risk of bleeding on anticoagulants.
Thrombocytopenia is not a contraindication to anticoagulation for individuals with platelet counts >50,000/microL. However, anticoagulation is typically contraindicated in those with platelet counts <20,000/microL.
For many individuals with cancer and acute VTE who have platelet counts between 20,000 and 50,000/microL, the decision to anticoagulate should be individualized and based upon the risk of serious complications from VTE and the risk of bleeding associated with anticoagulation, especially when anticoagulant dose modifications or platelet transfusions are used.
This subject and anticoagulation in patients with heparin-induced thrombocytopenia are discussed in more detail separately. (See "Anticoagulation in individuals with thrombocytopenia", section on 'Cancer-associated VTE' and "Management of heparin-induced thrombocytopenia", section on 'Anticoagulation'.)
Arterial thromboembolism — Arterial thromboembolism is less common than venous thromboembolism in patients with cancer. Retrospective cohort analyses in patients with cancer suggest that the incidence of arterial thromboembolism ranges from 2 to 6 percent [69,70].
In patients with malignancy, the most common sources for arterial thromboembolism are nonbacterial thrombotic endocarditis and paradoxical embolism arising from deep venous thrombosis [71]. (See "Nonbacterial thrombotic endocarditis".)
Embolism to the digits, brain, or solid organs (in the absence of a source) can also be a paraneoplastic manifestation of solid tumors and myeloproliferative disorders [72,73]. (See "Clinical manifestations, pathogenesis, and diagnosis of essential thrombocythemia", section on 'Thrombosis and hemorrhage' and "Clinical manifestations and diagnosis of polycythemia vera", section on 'Thrombosis and hemorrhage'.)
We typically anticoagulate patients with cancer-associated arterial thrombosis provided that the patient's bleeding risk is not high and treatment is consistent with the patient's overall management goals.
OUTPATIENT THERAPY — We consider the following as appropriate in patients with cancer-associated VTE:
●Deep venous thrombosis (DVT) – Many patients with cancer-associated DVT of the lower extremity are treated as outpatients. We select patients for outpatient treatment using similar principles to those in the noncancer population (table 4), the details of which are provided separately. (See "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)", section on 'Outpatient versus inpatient therapy'.)
●Pulmonary embolism (PE) – By contrast, we generally treat patients with cancer-associated PE as an inpatient or with observation for a short period of time before discharge (eg, 12 to 24 hours). However, we often treat clinically stable patients with asymptomatic or small subsegmental PE as outpatients. (See "Treatment, prognosis, and follow-up of acute pulmonary embolism in adults", section on 'Outpatient anticoagulation'.)
Principles of agent selection are similar to those in patients who do not have malignancy. The most commonly used agent is an oral factor Xa inhibitor. (See "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)", section on 'Outpatient versus inpatient therapy'.)
DURATION OF ANTICOAGULATION
General approach — For a first episode of acute VTE in patients with active cancer, anticoagulant therapy should be administered for at least three to six months, provided the bleeding risk is low and no clinically relevant complications from anticoagulation have occurred (bleeding, heparin-induced thrombocytopenia). Considering when to extend beyond this finite period involves clinical judgement and reassessment of the recurrence and bleeding risk. Our general approach is the following:
●In most patients with active cancer and proximal deep venous thrombosis (DVT) or pulmonary embolism (PE), we administer indefinite anticoagulant therapy, unless the bleeding risk is high or VTE was precipitated by a major provoking event such as surgery. Provided the patient has no other justification for indefinite anticoagulation, treatment is typically continued until the patient is in remission or has a major bleeding complication. Whether the intensity of anticoagulant therapy might be lowered after some initial (eg, 6 to 12 months) treatment is the subject of at least one ongoing randomized controlled trial (NCT03080883) [74]. Active cancer is defined by the International Society on Thrombosis and Hemostasis as cancer diagnosed within the previous six months; recurrent, regionally advanced, or metastatic cancer; cancer for which treatment has been administered within six months; or hematologic cancer that is not in complete remission [15].
●The optimal duration of anticoagulant therapy for a patient with cancer-related isolated distal DVT that is not associated with surgery is not known; however, we typically approach such patients similar to cancer patients with proximal DVT.
●Exceptions include those at high bleeding risk, those with a clear provoking event and no other persistent risk factors (eg, surgery, subtherapeutic anticoagulation), patients in remission from their cancer, and patients with a strong preference not to be on extended anticoagulation.
●Follow-up with yearly reevaluation of the benefit/risk ratio for anticoagulation is important for this population to avoid unnecessary morbidity from bleeding or a recurrence. (See 'Follow-up' below.)
This recommendation is based upon the rationale that patients with active cancer have a high risk of recurrence. (See 'Recurrence risk' below.)
When selecting patients for extended anticoagulation, we follow the same general principles as in noncancer patients (table 5). Further details are provided separately. (See "Selecting adult patients with lower extremity deep venous thrombosis and pulmonary embolism for indefinite anticoagulation".)
Recurrence and bleeding risk assessment for patient selection — In patients with cancer-associated VTE, we individualize recurrence and bleeding risk assessment and take into consideration the following:
●General factors that apply to all patients
●Cancer-specific factors
In this section, we will discuss factors that are common in or specific to patients with cancer. Factors used to assess all patients with VTE for indefinite anticoagulation are provided separately (table 5). (See "Selecting adult patients with lower extremity deep venous thrombosis and pulmonary embolism for indefinite anticoagulation".)
Recurrence risk — VTE recurrence rates are high in patients with active cancer despite adequate anticoagulation [51,75-80]. However, the risk varies considerably. Factors that may influence this risk include the following:
●Active versus in-remission cancer – Patients with active cancer have a higher risk for recurrence on therapeutic anticoagulation compared with those in remission. Although the lower risk among those in remission may justify discontinuing anticoagulation, the risk of recurrence in this patient population is unknown. The definition of active cancer is provided above. (See 'Duration of anticoagulation' above.)
●Risk factors for subtherapeutic anticoagulation – For patients on warfarin, subtherapeutic anticoagulation is a common cause of recurrence. In patients with cancer, it is particularly common due to decreased absorption, cessation of therapy for procedures or bleeding, and varying dose requirements (eg, drug interactions), all of which contribute to longer periods of time outside of the target therapeutic range.
●Others – Other cancer-specific factors that may increase the recurrence risk in patients with cancer include the following [60,81]:
•Concomitant anticancer medications (eg, hormonal medications, bevacizumab).
•Common risk factors in patients with cancer (eg, immobility, pathologic fracture, recent surgery or invasive procedures, presence of a central venous access line, oral intolerance of medications due to nausea/vomiting).
•Younger age (<65 years).
•PE at presentation.
•Newly diagnosed cancer (<3 months).
•Specific cancer site (eg, pancreatic cancer has one of the highest rates of thrombosis). (See "Supportive care for locally advanced or metastatic exocrine pancreatic cancer", section on 'Venous thromboembolism'.)
Models have been proposed to help identify patients with cancer and acute VTE who are at greatest risk of recurrence. However, they are not widely used, and further studies are needed to examine whether they can be used to select patients who may benefit from extended anticoagulation [76]. Data that support using models such as the Khorana score to calculate the risk of VTE in patients with cancer who have not yet had VTE are described separately. (See "Risk and prevention of venous thromboembolism in adults with cancer", section on 'VTE risk assessment/Khorana score'.)
Bleeding risk — Treatment of VTE with anticoagulation is associated with higher rates of bleeding than in patients without cancer [2-9]. Rates of major bleeding range from 6.5 to 18 percent [5,6,51]. In a trial of 334 patients treated with dalteparin, rates of bleeding were highest in the first month and declined over the subsequent 11 months (4 versus 1 percent) [51]. Another study reported that the case fatality rates from bleeding were 0.8 per 100 patient-years of follow-up [82].
Cited risk factors for bleeding in patients with cancer overlap with those in patients without cancer (table 6) and include the following [81,83,84]:
●Older age (>65 years)
●Supra therapeutic anticoagulation
●Higher-intensity anticoagulation regimens (see 'Management of recurrence' below)
●Thrombocytopenia
●Prior gastrointestinal bleeding
●Immobilization
●Presence of metastases
●Recent bleeding
●Creatinine clearance <30 mL/minute
●Intracranial neoplasms
Risk factors for bleeding in the general population and assessment for bleeding risk when considering indefinite anticoagulation are discussed separately. (See "Venous thromboembolism: Anticoagulation after initial management", section on 'Risk of bleeding' and "Selecting adult patients with lower extremity deep venous thrombosis and pulmonary embolism for indefinite anticoagulation", section on 'Assessing the risk of bleeding'.)
Agent selection — Agent selection for extended anticoagulation is similar to that for initial therapy, which is generally an oral factor Xa inhibitor (ie, a direct oral anticoagulant) or low molecular weight heparin. Agent selection for indefinite anticoagulation in patients without cancer is discussed separately. (See "Selecting adult patients with lower extremity deep venous thrombosis and pulmonary embolism for indefinite anticoagulation", section on 'Low-intensity regimens following a finite course'.)
There is a paucity of data evaluating extended anticoagulation beyond six months in patients with active cancer. In a post-hoc analysis of one major trial [26] that compared dalteparin with edoxaban in patients with active cancer, VTE recurrence rates were no different between the groups (<0.7 percent for edoxaban versus 1.1 percent for dalteparin) [85]. Similar rates of major bleeding were also reported (1.7 percent for edoxaban versus 1.1 percent for dalteparin).
We do not administer low-intensity regimens in patients with active cancer since the baseline risk of recurrence is too high and lower-intensity regimens have not been studied in this population. Data to support low-intensity regimens in select patients without cancer are provided separately. (See "Selecting adult patients with lower extremity deep venous thrombosis and pulmonary embolism for indefinite anticoagulation", section on 'Low-intensity regimens following a finite course'.)
SWITCHING ANTICOAGULANTS — Interruptions should be limited, if feasible, especially during the first three months of anticoagulation. However, anticoagulants may need to be changed for medical reasons or patient preference. Reasons for changing anticoagulants and how best to transition are discussed separately. (See "Venous thromboembolism: Anticoagulation after initial management", section on 'Switching anticoagulants during therapy'.)
FOLLOW-UP — In patients with cancer-associated VTE, follow-up is important since the course is often punctuated by issues that affect continuation or discontinuation of anticoagulation [2-9]. This includes recurrence, bleeding, procedures, thrombocytopenia, new medications, conditions that affect oral intake or drug absorption, cancer progression, or achievement of remission.
All patients on anticoagulation should be monitored clinically for recurrences and bleeding, and for the development of adverse effects or conditions that affect the metabolism of the administered anticoagulant. During evaluation, we carefully weigh the benefit of preventing VTE recurrence (see 'Recurrence risk' above) against the bleeding risk (see 'Bleeding risk' above), while taking into consideration the type of cancer, burden of disease, treatment received, patient preference, and life expectancy.
MANAGEMENT OF RECURRENCE — VTE recurrence is more common in patients with cancer compared with patients who do not have cancer (see 'Recurrence risk' above). For patients who develop a recurrence despite therapeutic anticoagulation, involvement of a hematologist and/or oncologist is prudent.
The causes and principles of management are similar to patients without cancer. In brief, the etiology should be identified and treated accordingly (eg, subtherapeutic anticoagulation, nonadherence, impaired oral absorption, marked obesity, ongoing thrombotic stimuli). (See "Treatment, prognosis, and follow-up of acute pulmonary embolism in adults", section on 'Management of recurrence on therapy'.)
The following is a strategy that can be applied to patients with and without cancer who have recurrence despite therapeutic anticoagulation and in whom a reversible etiology cannot be identified. This strategy involves one or both of the following:
●Switching or escalating anticoagulant therapy – A common management strategy involves using an alternate agent or increasing the dose or frequency of administration of the anticoagulant. Such management is empiric, as randomized trials addressing this problem are lacking, and consultation with a thrombosis specialist is advised. As examples:
•For patients with a recurrence on the recommended dose of a direct oral anticoagulant or warfarin with an international normalized ratio (INR) in the therapeutic range (ie, 2 to 3), we generally switch to low molecular weight (LMW) heparin at therapeutic doses (see 'Low molecular weight heparin' above). Rarely, for patients on warfarin in whom switching to LMW heparin or an oral factor Xa inhibitor is not feasible, LMW heparin can be used followed by warfarin with a higher INR target range (eg, 2.5 to 3.5).
•For patients with recurrence on a single daily dose of LMW heparin, we switch to every-12-hour administration at a therapeutic dose or increase to 120 percent of the therapeutic dose, knowing that the bleeding risk is increased with the latter. Another option would be to switch to either apixaban or rivaroxaban at therapeutic doses, but there is little published experience with this approach. (See 'Direct oral anticoagulant mono- or dual therapy' above.)
There are limited data to support this strategy and larger, prospective studies with longer follow-up are needed. One retrospective study evaluated the efficacy of escalating the dose of LMW heparin in 70 patients with active cancer who developed a documented episode of recurrent VTE while receiving anticoagulation (LMW heparin or warfarin) [86]. Twenty-one percent of patients were on therapeutic doses of LMW heparin and were increased to 120 percent of the therapeutic dose for at least four weeks. The remaining patients, who were on subtherapeutic doses of LMW heparin (46 percent) or were warfarin failures (33 percent), were switched over to therapeutic LMW heparin. At three months, 8.6 percent of patients had a recurrent episode of VTE within two months. Among those treated with an increased dose of LMW heparin, no further recurrences were reported during the three-month follow-up period. Bleeding complications were reported in 4 percent of patients, with only one major bleed (in a patient with intracranial malignancy). Deaths in this study were not related to either recurrent VTE or bleeding.
●Inferior vena cava filter – For patients with VTE recurrence in whom anticoagulation therapy is maximized, we consider adding an inferior vena cava (IVC) filter to therapeutic anticoagulation. For patients with recurrent cancer-associated VTE in whom anticoagulation has been associated with bleeding complications, we also consider an IVC filter. Data supporting the placement of an IVC filter are limited and discussed above. (See 'Contraindications: Inferior vena cava filter' above and "Placement of vena cava filters and their complications".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Superficial vein thrombosis, deep vein thrombosis, and pulmonary embolism" and "Society guideline links: Anticoagulation".)
SUMMARY AND RECOMMENDATIONS
●General principles – Anticoagulation is the cornerstone of therapy for cancer-associated venous thromboembolism (VTE). However, anticoagulation in this population is complicated by higher than usual rates of recurrent VTE and bleeding and is commonly impacted by comorbidities, procedures, and medication interactions. (See 'Principles of therapy' above and 'Agent choice' above.)
●Agent selection – Choosing among the agents is dependent upon several factors that are described in the tables (table 1 and table 2). (See 'Selecting among the agents' above.)
Our approach is the following:
•Most patients – For most hemodynamically stable patients with cancer-associated VTE who do not have severe renal insufficiency (eg, creatinine clearance <30 mL/minute) or a contraindication, we suggest an oral factor Xa inhibitor (ie, a direct oral anticoagulant) rather than low molecular weight (LMW) heparin (Grade 2B) or conventional anticoagulation (ie, intravenous unfractionated heparin followed by warfarin) (Grade 2C).
This choice is based upon data that show similar efficacy and safety for oral factor Xa inhibitors and LMW heparin in cancer-associated VTE. While several factors influence agent selection, factor Xa inhibitors are frequently chosen to avoid the need for daily injections that are required for LMW heparin. (See 'First-line options' above.)
-Factor Xa inhibitors – Among the factor Xa inhibitors, we suggest apixaban, as an oral monotherapy option (ie, administered without an initial five days of parenteral LMW heparin) (Grade 2C). This preference is based upon a randomized trial that reported similar efficacy and safety compared with LMW heparin. However, apixaban has the disadvantage of twice-daily dosing.
As an alternative, the factor Xa inhibitors rivaroxaban (a once-daily monotherapy agent) or edoxaban (a once-daily dual-therapy agent; ie, administered with an initial five days of parenteral heparin) can be used. However, we typically avoid these two agents in patients with upper gastrointestinal cancers since more bleeding has been reported with these agents than with apixaban or LMW heparin. (See 'Direct oral anticoagulant mono- or dual therapy' above.)
-LMW heparin – LMW heparin may be an alternative in patients who cannot take a factor Xa inhibitor (eg, not covered on insurance). Several LMW heparin formulations are available. While they have not been compared with one another, dalteparin or enoxaparin are used most commonly in the United States. (See 'Low molecular weight heparin' above.)
•Special populations – For patients with hemodynamic instability or severe renal insufficiency, considerations regarding agent selection are similar to those in patients without cancer. Other cancer-specific issues requiring special consideration include intracranial malignancy and thrombocytopenia. Agent selection in specific populations is discussed separately. (See 'Special populations' above and "Venous thromboembolism: Initiation of anticoagulation", section on 'Special populations' and 'Cancer-specific issues' above.)
●Outpatient therapy – We typically treat patients with cancer and uncomplicated deep venous thrombosis (DVT; eg, DVT without concomitant pulmonary embolism [PE]) as outpatients using similar principles to those in the noncancer population (table 4). Most patients with cancer who have PE are admitted or undergo a short observation period. (See "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)", section on 'Outpatient versus inpatient therapy' and "Treatment, prognosis, and follow-up of acute pulmonary embolism in adults", section on 'Outpatient anticoagulation' and 'Outpatient therapy' above.)
●Duration of anticoagulation – For most patients with active cancer-associated VTE, we suggest extending therapy beyond the conventional three-to-six-month period (Grade 2C), unless the bleeding risk is high or VTE was precipitated by a major provoking event such as surgery. Provided the patient has no other justification for indefinite anticoagulation, treatment is typically continued until the patient is in remission or has a major bleeding complication. (See 'Duration of anticoagulation' above.)
•This recommendation is based upon the rationale that most of these patients have a high risk of recurrence such that the benefit of VTE prevention outweighs the risk of bleeding with anticoagulant therapy. However, the decision is challenging since this population is also at a higher risk of bleeding on anticoagulant therapy and patients may have personal factors that weigh into the decision (eg, life expectancy) (table 1). Thus, we individualize the approach to risk assessment using similar principles in patients without cancer. (See 'Recurrence risk' above and 'Bleeding risk' above and "Selecting adult patients with lower extremity deep venous thrombosis and pulmonary embolism for indefinite anticoagulation", section on 'Clinical assessment'.)
•Agent selection is similar to that for initial selection (ie, oral factor Xa inhibitor or LMW heparin). We do not administer low-intensity regimens in patients with active cancer. (See 'Agent selection' above and 'Switching anticoagulants' above and "Venous thromboembolism: Anticoagulation after initial management", section on 'Switching anticoagulants during therapy'.)
●Follow-up – All patients with cancer on anticoagulant therapy should be monitored clinically since their course is often punctuated with issues that affect continuation or discontinuation of anticoagulation including VTE recurrence, bleeding, procedures, thrombocytopenia, new medications, conditions that affect oral intake or drug absorption, local and metastatic progression, or achievement of remission (table 1). (See 'Follow-up' above.)
●Managing recurrences – Management options for those who recur despite therapeutic anticoagulation include switching to LMW heparin for patients taking an oral anticoagulant or escalation of the dose of LMW heparin for those already on LMW heparin. Involvement of a hematologist is prudent. (See 'Management of recurrence' above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
