Pages Navigation Menu

Meet 'em, greet 'em, treat 'em and street 'em

SGEM#235: Edoxaban for Cancer Associated VTE – Would the NEJM Lie to You?

SGEM#235: Edoxaban for Cancer Associated VTE – Would the NEJM Lie to You?

Podcast LinkSGEM235

Date: October 30th , 2018

Reference: Raskob GE et al. Edoxaban for the Treatment of Cancer-Associated Venous Thromboembolism. NEJM 2018

Guest Skeptic: Dr. Anand Swaminathan is an assistant professor of Emergency Medicine at the St. Joseph’s Regional Medical Center in Patterson, NJ. He is a deputy editor for EM: RAP and, associate editor for REBEL EM.

Case: A 43-year old woman with a history of breast cancer currently undergoing chemotherapy presents with mild chest pain. She is hemodynamically stable except for a heart rate of 105 and her pain is increased when she takes a deep breath. The chest x-ray is unremarkable, and you order a CT pulmonary angiogram (CTPA) which demonstrates a right segmental pulmonary embolism. You write a prescription for low molecular weight heparin (LMWH) and advise the patient that she will be taking shots for a couple of months. She tells you that a friend of hers had a clot in her leg and was given an oral blood thinner. She wants to know if you can prescribe that pill, so she doesn’t have to take a shot.

Background: Venous thromboembolism (VTE) occurs frequently in patient with cancer. Treatment in this group entails a number of challenges including a higher rate of thrombosis recurrence and a higher risk of bleeding. Standard therapy at this time for both symptomatic and asymptomatic VTE is with LMWH based on results from the CLOT trial (Lee 2003).

In non-cancer patients, new oral anticoagulants (NOACs) like rivaroxaban have been shown to be effective in treatment without increasing bleeding events. The NOACs also add ease of use for the patient.

We covered using rivaroxaban on SGEM#126 with VTE guru Dr. Jeff Kline. This study suggested it was safe and effective to dry start (no LMWH needed) in certain patients with DVTs and PEs.

Though these agents are frequently used in the treatment of cancer-associated VTE, there is a dearth of evidence supporting this practice, in fact, none of the major agents – dabigitran, rivaroxaban, apixaban or edoxaban have undergone a well-done, randomized controlled trial.


Clinical Question: Is edoxaban non-inferior to LMWH in the treatment of cancer-associated VTE?


Reference: Raskob GE et al. Edoxaban for the Treatment of Cancer-Associated Venous Thromboembolism. NEJM 2018

  • Population: Adult patients with active cancer or cancer diagnosed within the previous two years with acute symptomatic or asymptomatic deep-vein thrombosis (DVT) or pulmonary embolism (PE).
  • Intervention: LMWH for five days followed by oral edoxaban 60 mg daily for at least six months.
  • Comparison: Subcutaneous (SQ) dalteparin 200 IU/kg daily (maximum dose 18,000IU) for one month followed by 150 IU/kg daily for at least five months.
  • Outcome:
    • Primary: Composite of recurrent VTE (DVT or segmental or more proximal PE) or major bleeding (overt bleeding associated with 2g/dL drop in hemoglobin or a transfusion of two or more units of blood during twelve-month follow up.
    • Secondary: Clinically relevant non-major bleeding (CRNB), event-free survival, VTE-related death, all-cause mortality, recurrent DVT, recurrent PE. The complete list can also be found in the Supplementary Appendix.

Authors’ Conclusions: “Oral edoxaban was noninferior to subcutaneous dalteparin with respect to the composite outcome of recurrent venous thromboembolism or major bleeding. The rate of recurrent venous thromboembolism was lower but the rate of major bleeding was higher with edoxaban than with dalteparin.”

Quality Checklist for Randomized Clinical Trials:

  1. The study population included or focused on those in the emergency department. Unsure.
  2. The patients were adequately randomized. Yes
  3. The randomization process was concealed. No
  4. The patients were analyzed in the groups to which they were randomized. Yes
  5. The study patients were recruited consecutively (i.e. no selection bias). Unsure
  6. The patients in both groups were similar with respect to prognostic factors. Yes
  7. All participants (patients, clinicians, outcome assessors) were unaware of group allocation. No
  8. All groups were treated equally except for the intervention. No
  9. Follow-up was complete (i.e. at least 80% for both groups). Yes
  10. All patient-important outcomes were considered. Yes
  11. The treatment effect was large enough and precise enough to be clinically significant. Unsure

Key Results: This trial included 1,050 patients with the average age in the early 60’s and close to a 50/50 male/female split. More than 50% had metastatic disease with almost 1/3 with recurrent disease.


Edoxaban was non-inferior to dalteparin for the primary outcome of recurrent VTE or major bleeding.


  • Primary Outcome:Recurrent VTE or major bleeding
    • 12.8% vs 13.5 % HR 0.97 (95% CI 0.70 to 1.36 P=0.006 for non-inferiority)
  • Secondary Outcomes: 

1. Consecutive Patients– There is no mention in the manuscript if there were consecutive patients enrolled in the trial. Selection bias could have been introduced making the results harder to interpret. However, the Supplemental Appendix says: “Adult subjects presenting with VTE associated with cancer (other than basal-cell or squamous-cell carcinoma of the skin) for whom long-term treatment with LMWH is intended are eligible to participate in the study.”It does not say “all” so we are unsure.

2. Emergency Patients – It appears that these were patients recruited from outpatient clinics. Could the protocol be applied to emergency department patients being diagnosed with VTEs and would the outcomes be the same?

3. Lack of Blinding –The patients were not blinded while the outcome assessors for major bleeding were unaware of group assignments. It is unsure if patients knowing what group they were assigned would have impacted the results. Why not just have placebo pills and SQ injections? This could have minimized this bias.

4. Combined Endpoint – They made a composite outcome of efficacy (VTE recurrence) and safety (major bleed). Why not just have one primary outcome? They could have asked the patient what they thought the most important thing is from the list of all the secondary outcomes. Power the study to answer that question.

5. Patient Oriented Primary Outcome– Let’s drill down into the idea of a patient-oriented primary outcome. There was a lower recurrent VTE rate with edoxaban (7.9% vs. 11.3) but a higher major bleed rate (6.9% vs. 4.0%). This did not translate into a statistical difference in all-cause mortality or event free survival. So ultimately what is more important to patients?

6. Changed Primary Endpoint and Time Frame– If you go to ClinicalTrials.gov you can see that they originally had a co-primary outcome. The primary efficacy outcome was incidence of recurrent VTE and the primary safety outcome was clinically relevant bleeding while on treatment. This was changed to adjudicated recurrent VTE or major bleeding event.

In addition, the original recurrent VTE time frame was six months. This was extended to twelve months. The original primary outcomes at six months can be found in the Supplementary Appendix. It showed non-inferiority of edoxaban compared to dalteparin for recurrent VTE but an increased HR for clinically relevant bleed (major or clinically relevant nonmajor bleeding). There was no difference in all-cause mortality or event-free survival.

Searching through the changes on ClinicalTrials.gov site the it seems like the time frame change to 12 months for the outcome was only introduced September 28th, 2018. This is four years after the study began and on the day the results were 1st reported.

It’s unusual to see a combination of the efficacy and safety as the primary outcome and given that we know the authors changed this, it almost seems like they were trying to hide the increased risk of bleeding.

Personally, I’m more concerned about major bleeding in my patients than recurrent VTE which is common and expected in patients with cancer.

7. Cost – There was no mention of the cost between oral edoxaban vs. SQ dalteparin? This was a multinational trial done in twelve different counties. Funding for medication is done differently from country-to-country. This issue may come into play when thinking about applying these results. A quick check of GoodRx.com showed edoxaban 60mg costs $4,200 for a one-year supply while Dalteparin costs SQ $36,800 (assume 80kg x 200IU/kg OD x 1 month + 150IU/kg OD x 11 months).

8. Non-Inferiority – This was a non-inferiority trial design. They wanted to demonstrate that oral edoxaban was not worse than dalteparin in the parameters they measured. What about patient satisfaction? They did not ask the patient if they were happy with their care and if they would have liked to have been randomized into the other group. Was avoiding needles important to most of the patients? In the “real world” application would it mean more nursing visits for those getting SQ injections and less for those on oral medications? Would patients value the RN visit more than they disliked having a needle? These are all things to consider when contemplating how to apply these results. All that being said, compliance with the regimen suggests that patients preferred pills to shots.

9. Conflicts of Interest – The authors reported multiple conflicts of interest. The lead author and senior author both disclosed getting consultant fees and honoraria from Daiichi Sankyo during the conduct of the study. This does not make the data wrong but should make us more skeptical.

10. Sponsorship – This trial was sponsored by Daiichi Sankyo the maker of edoxaban. This too does not make the data wrong. This pharmaceutical company, in collaboration with the coordinating committee, was responsible for the trial design, protocol and oversight. They were responsible for collection of the data and maintenance of the data. They also performed all the statistical analysis in collaboration with the writing committee. Daichi Sankyo would have a clear bias for demonstrating non-inferiority.

 Comment on Authors’ Conclusion Compared to SGEM Conclusion: We would reversed the conclusions to emphasize the first to do no harm principle (primum non nocere). We also would have used the original co-primary outcome at six-months rather than the changed composite outcome at twelve months. Our conclusion would have been “oral edoxaban had a higher rate of clinically relevant bleeding compared to dalteparin, was noninferior with respect to recurrent VTE and no statistically significant different was observed in all-cause mortality and event-free survival.”


SGEM Bottom Line: Oral edoxaban may be a reasonable option to discuss with patients who have a cancer associated VTE, but the decision should probably be left up to the patient and their oncologist.


Case Resolution: The patient is given LMWH as usual and referred back to her oncologist to discuss the possibility of going on an oral agent.

Clinical Application: The evidence for using edoxaban, or any other DOAC, in cancer-associated VTE is inconclusive. Though some of these agents are being preferentially used for treatment, we eagerly await future studies on the topic to determine if there is a most appropriate management. Consideration of DOACs on an individual patient basis is reasonable particularly if compliance and cost will be an issue but, should be done in concert with the patient’s oncologist.

What Do I Tell My Patient? I understand why you might want a pill instead of a shot. A recent medical study shows that a pill was not inferior than a shot for preventing future blood clots. However, it did cause more bleeding. All the patients in the study got a shot once a day for at least the first five days before starting the pill. This option is something you can talk more about with your oncologist next week. I will send her a note saying you want to discuss taking a pill instead of a shot.

Keener Kontest: Last weeks’ winner was Aaron Morgan and PGY-1 Emergency medicine resident. He knew the tusk of the narwhal was sometimes sold as unicorn horn.

Listen to the SGEM podcast on iTunes to hear this the new keener question. If you know the answer send an email to TheSGEM@gmail.com with “keener” in the subject line. The first correct answer will receive a cool skeptical prize.

Other FOAMed:

  • REBEL EM – Edoxaban in Cancer-Associated VTE
  • PulmCrit – DVT-PE in cancer: Oral anticoagulant edoxaban non-inferior to enoxaparin

Remember to be skeptical of anything you learn, even if you heard it on the Skeptics’ Guide to Emergency Medicine.