PerspectiveIn the Journals Plus

Direct oral anticoagulants do not increase incidence of intracranial hemorrhage

Show Citation

February 25, 2019

Direct oral anticoagulants did not appear associated with a higher incidence of intracranial hemorrhage compared with low-molecular-weight heparin in patients with brain metastases or primary brain tumors, according to results of a retrospective study published in Journal of Thrombosis and Haemostasis.

“There is a growing body of data supporting the use of [direct oral anticoagulants] in patients with cancer, but prior to the publication of our study there was little evidence supporting their safety in terms of intracranial hemorrhage risk in patients with brain tumors,” Brian J. Carney, MD, clinical fellow in hematology and oncology at Beth Israel Deaconess Medical Center and Harvard Medical School, told HemOnc Today. “Our work suggests that [direct oral anticoagulants] have an acceptable safety profile in patients with brain tumors and may even be preferable to [low-molecular-weight heparin] in patients with gliomas.”

Direct oral anticoagulants are effective in treating cancer-associated thrombosis but pose a higher risk for hemorrhage compared with low-molecular-weight heparin in certain cancers, according to study background.

Carney and colleagues sought to compare incidence of intracranial hemorrhage with the two treatment options among 172 patients with primary (n = 67) or metastatic (n = 105) brain tumors.

In the primary brain tumor group, 47 patients received enoxaparin and 20 patients received direct oral anticoagulants, including apixaban (Eliquis; Bristol-Myers Squibb, Pfizer; n = 15) and rivaroxaban (Xarelto, Janssen; n = 5).

In the metastatic brain tumor group, 84 patients received enoxaparin and 21 patients received direct oral anticoagulants including rivaroxaban (n = 11), apixaban (n = 5) and dabigatran (Pradaxa, Boehringer Ingelheim; n = 5).

“More patients received enoxaparin because prior to late 2017 because [direct oral anticoagulants] did not have a substantial evidence base supporting their use in cancer-associated thrombosis,” Carney told HemOnc Today. “The patients described in our study were treated between 2011 and 2018, hence the larger numbers receiving [low-molecular-weight heparin].”

Subgroups appeared well-balanced regarding age, gender and, in the brain metastases subgroups, tumor diagnosis, including tumors with a high incidence of intracranial hemorrhage. However, in both groups, a greater proportion of those treated with direct oral anticoagulants had additional risk factors for hemorrhage, including concomitant aspirin use and diagnosis of hypertension.

Among patients with primary brain tumors, none who received direct oral anticoagulants experienced intracranial hemorrhage within 12 months of treatment, compared with 36.8% (95% CI, 22.3-51.3) of patients who received enoxaparin (P = .007). Eight patients (18.2%) who received enoxaparin experienced a major intracranial hemorrhage.


The only intracranial hemorrhage among patients with primary brain tumors who took direct oral anticoagulants occurred 21 months after initial treatment and was minor (0.3 mL). Median size of intracranial hemorrhage in patients who received enoxaparin was 2 mL (range, trace to 62 mL).

Among patients with primary brain tumors, direct oral anticoagulants did not increase the incidence of any intracranial hemorrhage events compared with enoxaparin (27.8%; 95% CI, 5.5-56.7 vs. 52.9%; 95% CI, 37.4-66.2) nor the incidence of major intracranial hemorrhage (11.1%; 95% CI, 0.540.6 vs. 17.8%; 95% CI, 10.2-27.2).

The retrospective nature of the study served as a limitation because there could have been selection bias.

“A randomized clinical trial would be needed to definitively establish the relative benefit of [direct oral anticoagulants] over [low-molecular-weight heparin] in patients with glioma,” Carney and colleagues wrote. “However, considering [direct oral anticoagulants] are efficacious in patients [with cancer] and favored by patients because of their ease of administration, and in view of our safety data, we anticipate that it would be very difficult to recruit patients to such a trial.” – by John DeRosier

For more information:

Brian J Carney, MD, can be reached at

Disclosures: Carney reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.

itj+ Infographic

itj+ Perspective

Photo of Ashley Sumrall

For years, we have managed outpatients in the neuro-oncology setting who have venous thromboembolism and pulmonary embolism with low-molecular-weight heparin. My own perception was that these products are best for patients with cancer, as well as safest.

In my cohort of patients with primary brain tumors — where VTE is felt to be present in at least one-third — I have been rigid in my recommendations. Our patients tend to hate the injections but tolerate the therapy.

When direct oral anticoagulants (DOACs) entered the scene, I was leery of transitioning patients, given an increased risk for bleeding. Patients would see commercials and ask, then a difficult conversation would follow. In SELECT-D, only 1% of included patients had brain cancer. Over half of the patients had metastatic disease, but brain metastases were not specified. In the paper published in The New England Journal of Medicine, we had more information, but still not enough to feel comfortable with a change in therapy.

I have felt the best about transitioning to DOACs after reading the study by Carney and colleagues. There is no significant increase in bleeding and, among patients with primary glioma, the incidence of intracranial hemorrhage of any grade was zero at 12 months.

For those patients with brain metastases — remembering that some subtypes of brain metastases are very prone to bleeding — DOACs did not increase incidence of any-grade intracranial hemorrhage relative to low-molecular-weight heparin. Interestingly, patients with melanoma and renal cell carcinoma did not have significant increases in rates of intracranial hemorrhage. To the authors’ credit, they painstakingly analyzed reports to ensure that episodes of intracranial hemorrhage were appropriately characterized.

The caveat to making a complete change in our recommendations for anticoagulation for patients with brain tumors and VTE is that the number of patients included in this paper, unfortunately, was small. I would also note that patients with bleeds within 4 weeks after a neurosurgical procedure were excluded — and this is highest-risk time to develop a clot. DOACs in this situation have not been appropriately studied.


Raskob GE, et al. N Engl J Med. 2018;doi:10.1056/NEJMoa1711948.

Young AM, et al. J Clin Oncol. 2018;doi:10.1200/JCO.2018.78.8034.

Ashley L. Sumrall, MD

HemOnc Today Editorial Board Member
Levine Cancer Institute at Atrium Health

Disclosure: Sumrall reports no relevant financial disclosures.