Primary Prophylaxis With a Direct Oral Anticoagulation Agent Reduces Venous Thromboembolism Rate in Ambulatory Patients With Cancer

Get Permission

LATE-BREAKING results from the large, randomized, placebo-controlled CASSINI trial showed that primary prophylaxis with the direct oral anticoagulation agent rivaroxaban reduced the incidence of venous thromboembolism (VTE) as well as VTE-related deaths in high-risk patients with cancer who were about to initiate a new systemic chemotherapy. These results are potentially practice-changing in terms of screening and prophylaxis against VTE with a direct oral anticoagulation agent.

Alok A. Khorana, MD, FASCO

Alok A. Khorana, MD, FASCO

“These results suggest that screening for VTE risk should be done for all patients with cancer undergoing ambulatory chemotherapy. The findings of this study1 and a second recently published study2 should inform future recommendations from the American Society of Hematology (ASH) and ASCO regarding thromboprophylaxis for this population,” noted lead author Alok A. Khorana, MD, FASCO, of the Cleveland Clinic Taussig Cancer Center, Cleveland.

“In this large study of over 800 high-risk patients, a direct oral anticoagulation agent reduced the risk of harmful clots. This is an important advance in management and primary prevention in a common illness that presents with cancer,” said Mark A. Crowther, MD, MSc, FRCPC, of McMaster University, Hamilton, Ontario, Canada, and Co-Chair of the ASH Educational Program. Dr. Crowther was moderator of a press conference at the 2018 ASH Annual Meeting & Exposition where this study and other late-breaking abstracts were discussed.

Mark A. Crowther, MD, MSc, FRCPC

Mark A. Crowther, MD, MSc, FRCPC


VTE is a common problem in patients with cancer, and it is potentially fatal. The condition can lead to complications that include hospitalization, morbidity, delays in initiating chemotherapy and/or radiation, and increased resource utilization.

“Most clots are now happening in the outpatient setting, because most patients get chemotherapy there. Two larger trials showed a reduction in VTE with prophylaxis with injectable low–molecular-weight heparins, but the event rates were low and the number needed to treat was high. An estimated 50 patients would have to take an injection every day to prevent 1 blood clot,” Dr. Khorana told listeners at a press conference.

Study Details

TEN YEARS AGO, Dr. Khorana and colleagues developed a risk-adapted approach to VTE to determine clinical benefit, and this score was subsequently validated by multiple investigators.


The Khorana score predicts thrombosis risk based on a variety of baseline clinical and laboratory values, including type of cancer, body mass index, and complete blood cell count.

In the CASSINI trial, a Khorana risk score of 2 or higher identified patients at high risk for VTE for enrollment in the trial. The study enrolled 841 high-risk patients with cancer about to start systemic chemotherapy and randomly assigned them 1:1 to oral rivaroxaban at 10 mg/d or placebo for 180 days of treatment. Rivaroxaban is approved for patients with atrial fibrillation, for prevention of recurrent VTE, and for prevention of a first VTE in selected high-risk patients, he explained. Patients had a variety of cancer types; they were stratified for pancreatic cancer or not, since pancreatic cancer is associated with a high risk of VTE.

Prior to randomization, all patients underwent ultrasound of the legs, and a clot was present in 4.5%, who were excluded from the study. “Thus, all patients included in the analysis were thrombosis-free at the start of the study,” he said. On study, ultrasound of the legs was performed every 8 weeks.

“These results suggest that screening for VTE risk should be done for all patients with cancer undergoing ambulatory chemotherapy.”
— Alok A. Khorana, MD, FASCO

Tweet this quote

For the primary efficacy analysis, which included all randomly assigned patients up to day 180, prophylactic rivaroxaban reduced the numerical incidence of VTE by about 3% (8.79% in the placebo group vs 5.95% in the rivaroxaban group), but this difference was not statistically significant. About 40% of patients overall did not take the drug for the full 6 months for a variety of reasons. “These were patients [in very poor health], maybe transitioning to a new cancer regimen or having side effects,” Dr. Khorana explained.

An on-treatment analysis found an important reduction in VTE with rivaroxaban (P = .007): 6.41% for placebo vs 2.42% with rivaroxaban. For the secondary endpoint of VTE and all-cause mortality, a substantial benefit was found for rivaroxaban up to day 180. The rate of VTE plus all-cause mortality was 29% with placebo and 23.1% with rivaroxaban, an absolute reduction of 6% (P = .03).

“This shows that high-risk patients with cancer can benefit from a primary prevention approach,” he stated 

Major and Nonmajor Bleeding

“SAFETY IS A KEY ISSUE,” Dr. Khorana added. Among 809 patients who took at least 1 dose of study drug and were evaluable for safety, the rate of major bleeding was 0.99% in the placebo arm and 1.98% in the rivaroxaban arm, about twice as high. The rate of clinically relevant nonmajor bleeding was 1.98% vs 2.72%, respectively, about 50% higher with rivaroxaban.

“We have never before had a study showing that an intervention that is relatively inexpensive compared to the costs of other cancer drugs and the costs of treating complications from VTE can prevent VTE.”
— Mark A. Crowther, MD, MSc, FRCPC

Tweet this quote

Looking at risk vs benefit, the number needed to treat to prevent 1 VTE was 35 for the whole study population; in the on-treatment population, the number needed to treat was 26, and the number needed to treat decreased to 17 for a composite endpoint of VTE secondary efficacy thromboembolic events (including arterial) on treatment. “This compares favorably with the [number needed to treat] of 50 in the older trials of VTE prophylaxis,” Dr. Khorana noted.

The number needed to harm—how many patients need to be treated in order for 1 person to have a particular adverse effect— for major bleeding was 101. The number needed to harm for nonmajor bleeding was 135.

Current practice is to perform risk assessment in patients starting chemotherapy on a case-by-case basis. Patients with a Khorana risk score of 3 are typically considered to be at high risk, Dr. Khorana said. The CASSINI study used a score of 2 or higher to identify high-risk patients.

Additional Comments

“IN AN IDEAL WORLD, VTE in this setting would be almost completely preventable,” said Dr. Crowther. “This is a very important study. We have never before had a study showing that an intervention that is relatively inexpensive compared to the costs of other cancer drugs and the costs of treating complications from VTE can prevent VTE. Direct oral anticoagulation agents are convenient to use and can prevent deaths and morbidity.”

He concluded, “This study, along with a paper by Carrier et al published recently, supports the fact that high-risk patients benefit from prophylaxis.”

DISCLOSURE: Dr. Khorana is a consultant for Pfizer, Sanofi, Janssen, and Bayer; has received nonfinancial support for travel from Parexel, Pfizer, Sanofi, Halozyme, Seattle Genetics, AngioDynamics, Leo Pharma, Medscape/Web MD, and Bayer; personal fees from Parexel, Pfizer, Sanofi, Janssen, Halozyme, Seattle Genetics, AngioDynamics, Leo Pharma, Medscape/WebMD, Pharmacyclics, PharmaCyte, TriSalus, and Bayer; and research funding from Janssen. Dr. Crowther has equity ownership in Alnylam; is an advisor or consultant or on the board of directors for Daiichi Sankyo, Bristol-Myers Squibb, Octapharma, Bayer, and Shionogi; is on the speakers bureau for Alexion; and has received research funding from Leo Pharma and Bayer as well as honoraria from Pfizer.


1. Khorana AK, Soff GA, Kakkar AK, et al: Rivaroxaban thromboprophylaxis in high-risk ambulatory cancer patients receiving systemic therapy. Results of a randomized clinical trial (CASSINI). 2018 ASH Annual Meeting. Abstract LBA-1. Presented December 4, 2018.

2. Carrier M, Abou-Nassar K, Mallick R, et al: Apixaban to prevent venous thromboembolism in patients with cancer. N Engl J Med. December 4, 2018 (early release online).