Conundrums in Managing Cancer-Associated Thrombosis

Get Permission

Although there may be a clear role for prophylactic anticoagulation in hospitalized patients and those undergoing surgery, indication for prophylactic anticoagulation for patients with cancer seen in the ambulatory setting has yet to emerge. Until clinical trial data on prophylactic anticoagulation in the ambulatory setting emerge, the main focus of clinicians is how to treat the patient who develops a cancer-associated venous thromboembolism, according to Christine L. Kempton, MD, MSc. Dr. Kempton is Associate Professor in the Department of Hematology and Medical Oncology and Director of the Hemophilia of Georgia Center for Bleeding and Clotting Disorders of Emory University.

Christine L. Kempton, MD, MSc

Christine L. Kempton, MD, MSc

Persons with cancer are four to seven times more likely than the general population to develop thrombosis. Venous thromboembolism, in fact, can develop in up to 10% of individuals with cancer, especially among those with risk factors that include advanced age, African American race, and comorbidities.1,2 Certain cancer types also increase the risk—gastric, pancreatic, brain, lung, and ovarian tumors—as do metastatic disease and high tumor burden. Cisplatin-based chemotherapy, antiangiogenic agents, and hormonal therapy are the treatments most associated with risk for venous thromboembolism.

Even knowing the risk factors, “It’s still a conundrum as to who would benefit from anticoagulation to prevent venous thromboembolism,” Dr. Kempton said at the 2019 Debates and Didactics Conference in Sea Island, Georgia, sponsored by Emory.3

Direct Oral Anticoagulants Gaining Favor

For many years, low–molecular-weight heparin has been the standard of care for preventing venous thromboembolism after an index thrombosis.

The 2018 study by the Hokusai Venous Thromboembolism Cancer Investigators4 found a significant benefit to the direct oral anticoagulant edoxaban vs dalteparin in preventing recurrent cancer-associated venous thromboembolism (11.3% vs 7.8%; hazard ratio [HR] = 0.71), but major bleeding events were increased in the edoxaban arm. A subset analysis showed that this increase was caused by upper gastrointestinal tract in patients with gastrointestinal cancers (HR = 4.0).5

In a similar trial, SELECT-D, the recurrence rate of cancer-associated venous thromboembolism was significantly less with rivaroxaban (4%) than dalteparin (11%; HR = 0.43). However, as in the Hokusai trial, major bleeding was increased (HR = 1.83).6

“The overall take home point is that direct oral anticoagulants can be used in the cancer setting, although their ease of use should be weighed against the potential for more bleeding, particularly with gastrointestinal malignancies,” Dr. Kempton said.

Drug-Drug Interactions

Dr. Kempton cautioned that drug-drug interactions are a consideration when using direct oral anticoagulants, especially inducers and inhibitors of P-glycoprotein and CYP3A4, and recommended working with a pharmacist to ensure drug-drug interactions are identified.

Renal dysfunction is a contraindication to the use of direct oral anticoagulants, and clinicians should become familiar with the dose adjustments needed for certain levels of creatinine clearance. Patients with a body mass index greater than 40 kg/m2 may also not be good candidates for direct oral anticoagulants due to lower drug levels. Caution is needed in this population, with careful weighing of risks and benefits.

Choice of Anticoagulant: ISTH Recommendations

The International Society on Thrombosis and Haemostasis (ISTH) has made the following recommendations for anticoagulation in patients with cancer:

  • Specific direct oral anticoagulants are recommended for patients with a low risk of bleeding and for patients with no drug-drug interactions with current systemic therapy.
  • Low–molecular-weight heparin is recommended for patients with a high risk of bleeding, including for those with luminal gastrointestinal cancers and an intact primary; risk of bleeding from the genitourinary tract, bladder, or nephrostomy tubes; active gastrointestinal mucosal abnormalities such as duodenal ulcers, gastritis, esophagitis, or colitis.

The ISTH also emphasizes that patients should understand the risk/benefit scenario with direct oral anticoagulants, and treatment should incorporate patient preferences and values. “All this needs to be put together with shared decision-making. Some patients see injections as the worst thing in the world and want an oral anticoagulant. Others just want the best therapy and don’t care if it’s an injection or an oral agent,” Dr. Kempton noted.

She cautioned that the previously mentioned considerations—use of concomitant medications, bleeding risk, comorbidities—are likely to change at some point in a patient’s trajectory. “So, your choice of anticoagulant may also change, and you can adapt as needed,” suggested Dr. Kempton.

As for cost-effectiveness, Dr. Kempton added that, based on studies to date, “overall, it’s not more expensive to use a direct oral anticoagulant than enoxaparin, and the quality of life is fairly similar (aside from the patient who has real issues with injections).”

How Long Should Anticoagulation Be Administered?

“One of the more difficult questions is how long to have a patient stay on an anticoagulant, and how do we predict recurrence?” Dr. Kempton continued. “We’ve often said that as long as the patient has cancer, he or she should remain on anticoagulation, but there may be some ways we can tailor therapy.”

One consideration is the presence of residual venous thrombosis, ie, thrombus occupying more than 40% of the lumen. Overall, the annual risk of recurrent venous thromboembolism is about 15%; this rises to about 22% for patients with residual venous thrombosis but drops to 3% for those without this degree of residual clot. Patients who test negative for D-dimer and high-sensitivity C-reactive protein are also less likely to recur.7

“Taking into account your patients—the presence of residual venous thrombosis as well as their D-dimer levels before and after they stop anticoagulation—can help you predict recurrence risk,” Dr. Kempton said.

Overall, the recommendation is to continue anticoagulation as long as risk factors for recurrence are present, including active malignancy and ongoing treatment—at least for 6 months. One could consider discontinuing anticoagulation in patients with stable, low-risk cancer, those without residual venous thrombosis or signs of chronic thromboembolic pulmonary hypertension, and patients with normal D-dimer levels and C-reactive protein 21 days off anticoagulation. “With shared decision-making, these patients could be considered candidates for stopping anticoagulation, particularly if anticoagulation is burdensome to them in some way,” commented Dr. Kempton.

Incidental Pulmonary Embolism

“We’ve all gotten a call from a radiologist saying that he or she has found an incidental pulmonary embolism in a patient we sent for screening. What do we do with this?” asked Dr. Kempton.

Up to 50% of pulmonary emboli in patients with cancer are incidental findings. In the Hokusai trial, the rate of recurrent venous thromboembolism (and major bleeding) in patients with incidental thromboembolism was similar to those with symptomatic events.4 These patients are treated the same as any other patient with thrombosis, particularly if the thrombus is single segmental, lobar, central, or multiple subsegmental.

The recommendations are not clear, however, for treating the patient with a single subsegmental incidental pulmonary embolus. Such patients should be evaluated for evidence of a proximal deep vein thrombosis, segmental or more proximal pulmonary emboli, or multiple subsegmental pulmonary emboli. If the patient has a single subsegmental pulmonary embolism without lower extremity deep vein thrombosis, the ISTH recommends managing on a case-by-case basis, either by anticoagulation or monitoring.

Direct Oral Anticoagulants and Catheter-Associated Thrombosis

Catheter-associated thrombosis occurs in approximately one-third of patients with cancer and is often asymptomatic. If the catheter is functioning and the patient can tolerate anticoagulation, the catheter can remain in place. Anticoagulation is given for 3 to 6 months regardless of whether or not the catheter is removed, and longer if it is removed.

In a study of rivaroxaban, the recurrence rate was 1.4%, but the rate of major and clinically relevant nonmajor bleeding was 12.8%.8 Historical rates of bleeding in this population, had low–molecular-weight heparin been used, would be closer to 3% to 4%, Dr. Kempton pointed out.

“This was just one small study, but I think if you are going to be using anticoagulation in a patient with catheter-related thrombosis, know that it’s easy to prescribe direct oral anticoagulants, but keep in mind the risk for bleeding,” Dr. Kempton advised. 

DISCLOSURE: Dr. Kempton reported no conflicts of interest.


1. Heit JA: Epidemiology of venous thromboembolism. Nat Rev Cardiol 12:464-474, 2015.

2. Timp JF, BraekkanSK, Versteeg HH, et al: Epidemiology of cancer-associated venous thrombosis. Blood 122:1712-1723, 2013.

3. Kempton CL: Updates in the management of cancer-associated thrombosis. 2019 Debates and Didactics Conference. Lecture. Presented July 27, 2019.

4. Raskob GE, van Es N, Verhamme P, et al: Edoxaban for the treatment of cancer-associated venous thromboembolism. N Engl J Med 378:615-624, 2018.

5. Kraaijpoel N, Di Nisio M, Mulder FI, et al: Clinical impact of bleeding in cancer-associated venous thromboembolism: Results from the Hokusai VTE Cancer Study. Thromb Haemost 118:1439-1449, 2018.

6. Young AM, Marshall A, Thirlwall J, et al: Comparison of an oral factor Xa inhibitor with low molecular weight heparin in patients with cancer with venous thromboembolism: Results of a randomized trial (SELECT-D). J Clin Oncol 36:2017-2023, 2018.

7. Jara-Palomares L, Solier-Lopez A, Elias-Hernandez T, et al: D-dimer and high-sensitivity C-reactive protein levels to predict venous thromboembolism recurrence after discontinuation of anticoagulation for cancer-associated thrombosis. Br J Cancer 119:915-921, 2018.

8. Davies GA, Lazo-Langner A, Gandara E, et al: A prospective study of rivaroxaban for central venous catheter associated upper extremity deep vein thrombosis in cancer patients (Catheter 2). Thromb Res 162:88-92, 2018.