Patients diagnosed with myeloproliferative neoplasms (MPNs) — including polycythemia vera (PV), essential thrombocythemia (ET), and myelofibrosis (MF) — are at a higher risk of both venous and arterial thromboembolism. Management of thrombosis is central to the care of these patients; however, there is a dearth of recommendations specific to MPNs on the duration and selection of anticoagulation for the management of these events.1
In this edition of Drawing First Blood, ASH Clinical News invited Chi-Joan How, MD, clinical chief of hematology at Brigham and Women’s Faulkner Hospital and assistant professor in medicine at Harvard Medical School in Boston, and Brady L. Stein, MD, professor of medicine at Northwestern University’s Feinberg School of Medicine in Chicago, to debate and discuss how long a patient with an MPN should be treated with anticoagulation and other factors to consider when determining the appropriate management of thrombosis. Dr. How was asked to argue for indefinite anticoagulation therapy; Dr. Stein was assigned to argue for limited-duration anticoagulation therapy.
Chi-Joan How, MD Brady L. Stein, MD
What are the benefits and risks of indefinite anticoagulant therapy?
Chi-Joan How, MD: Anytime someone is on anticoagulation, we are trying to reduce the risk of thrombosis and balance that against the risk for bleeding. MPNs create a procoagulant state, so patients are at a higher risk of thrombosis. Patients with an MPN who had a prior thrombosis are at an especially high risk for recurrence, up to 10% per year in some patients.
The anticoagulant should be effective for however long someone is on it, but once they stop, there is a risk of thrombotic recurrence. Because patients with an MPN have a higher ongoing risk of subsequent blood clots, staying on the blood thinner can help prevent these events, such as pulmonary embolism (PE), deep vein thrombosis (DVT), or blood clots in the splanchnic venous system.
The risk of indefinite anticoagulant therapy is bleeding. We know that patients with an MPN also have more bleeding issues. We know, for instance, that patients with ET who have a very high platelet count might be predisposed to bleeding rather than thrombosis. A blood thinner would add to this bleeding risk.
Finally, a lot of patients might not want to be on indefinite treatment. Even though it may seem like a small matter, one benefit of a finite duration of anticoagulation is that it is one fewer medication a patient has to take.
What are the benefits and risks of time-limited anticoagulant therapy?
Brady L. Stein, MD: The benefits of a time-limited approach to anticoagulation include a reduced exposure to bleeding risks associated with anticoagulation.
The risk of this approach, though, is that anticoagulation only works while the patient is taking it. Once it is stopped, there is risk of a recurrent thrombotic event, which can be severe or life threatening in some cases.
What anticoagulant regimen should be used for patients with an MPN?
Dr. How: Historically speaking, vitamin K antagonists had been used as the standard anticoagulant regimen for MPNs. They have been shown to be safe and effective. However, now we have more experience with direct oral anticoagulants (DOACs), and these are a lot easier for patients to take.
There has been a lot of research on DOACs in patients with cancer generally, and they are now a preferred anticoagulant in the general oncology population. These studies don’t include patients with an MPN though, or even many patients with leukemia. So, despite the comfort we have with DOACs in MPNs, it’s still important to acknowledge the relative lack of data.
Dr. Stein: It’s hard to be dogmatic about MPN-associated thrombosis, whether it is duration or choice of anticoagulant, because we suffer from a lack of high-quality studies. It’s not for a lack of effort, but we have a relatively rare disease with a relatively low annual risk for thrombosis, and designing studies that lead to high-quality data is difficult, even to answer some of the most important questions in MPNs. We must rely on studies that are generally retrospective.
That being said, hematologists are comfortable using DOACs for patients with thrombosis at-large, including in patients with cancer-associated thrombosis. Patients with an MPN are not the same as patients with pancreatic or lung cancer, but we are comfortable using DOACs in patients with an MPN because they are generally safer, more predictable, and more reliable than warfarin.
The only situation where I would say we are less comfortable is for a patient with abdominal venous thrombosis. If a patient has Budd-Chiari and serious liver disease associated with this, then DOAC use may be contraindicated.
Would you select a different anticoagulant for limited versus indefinite duration?
Dr. Stein: Not necessarily. I’m comfortable using the DOACs short or long term. The only thing I might not use long term is enoxaparin because of the burden of long-term injections.
Dr. How: I would agree. In general, the duration won’t affect my decision-making, but other clinical factors would affect my anticoagulant choice. Most patients with an MPN do well with DOACs, but a few have a complex thrombotic history and would do better on something else.
You need a lot of flexibility when approaching anticoagulation with a patient. Every patient will be different and have different indications, whether it is a patient with liver dysfunction and Budd-Chiari, or one who has failed a DOAC. There are no hard and fast rules. Flexibility is important.
Which patients can stop anticoagulants after several months, and which should remain on anticoagulants indefinitely?
Dr. Stein: When considering which patient might be appropriate for limited-duration therapy, I think of a couple scenarios. One is if there is a major transient provoking risk factor — such as elective surgery (i.e., joint replacement) complicated by a DVT or PE. Those are patients for whom I might consider use of time-limited duration.
Another scenario is a patient with a lower-risk thrombotic event. Not all thrombotic events are equal. A distal DVT below the knee is different than a PE or abdominal venous thrombosis. For patients with upper extremity or distal DVT, I’m usually comfortable with time-limited duration.
Finally, most patients present with thrombosis either just before or at the time of their MPN diagnosis. If the patient presents with a thrombotic event and the MPN is relatively uncontrolled, I will try to control the MPN, start a treatment plan, and then consider a time-limited duration of anticoagulation.
Dr. How: Another thing to consider is a patient’s bleeding risk. If you have a patient who you think is at high risk for bleeding, that might push you toward a limited course.
Should patients with an MPN be on a lifelong therapy if they’re not likely to have another clot?
Dr. How: Unfortunately, you can never really know if a patient will have another clot, but there are some general guidelines. The main thing to consider is the provoking event. If you have a clear provoking event like surgery, there is generally a lower risk for recurrence, but even that may not be a sure thing.
It is easier to identify the patient in whom we know there is going to be a high risk of recurrence. For example, a patient could have an unprovoked clot despite optimal management of their MPN with normal blood counts. Although you can never be 100% sure, I’d be worried about a recurrence if patients go off anticoagulation.
Dr. Stein: The field really suffers in terms of vascular risk assessment. What we use for patients with ET and PV is age and thrombosis history. However, they lack specificity. Those variables can be applied to any patient you see in the hematology clinic. We know that it is a lot more nuanced than that. In ET, we do also look at mutations — like JAK2 V617F— in addition to age and thrombosis history, but in general, we are stuck with two to three variables. In other aspects of the disease, we are much more nuanced in how we predict prognosis. For MF, when we are trying to predict longevity or transformation, we use so many variables, but in thrombosis risk assessment, we are extraordinarily generic.
Is there a platelet count goal, especially in ET, and does that make a difference?
Dr. How: Platelets can be tough because unlike hematocrit for PV, where we have good data recommending a standardized level to shoot for, platelet goals are less clear. A lot of times, my platelet goal is dependent on a patient’s symptoms and side effects from cytoreduction. We try to get as close to normal as we can without causing other cytopenias or side effects.
Conversely, we know that as the platelets get high, there is an increased bleeding risk, but there is no magic number that will protect against bleeding.
In general, if someone has had a thrombotic event, that automatically puts them at higher risk for thrombosis in terms of risk stratification scores, so that becomes an indication for cytoreduction and to lower their blood counts. The definition of a clinicohematologic response is a platelet count of less than 400, which is pretty aggressive. I aim for platelet counts to be as close to normal without causing other problems and may push harder if there is a greater risk of thrombosis.
Dr. Stein: There is no such thing as a perfect platelet count. There are no high-quality clinical studies that suggest that having a perfect platelet count protects a patient from recurrent thrombosis. It’s frustrating because often the platelet count is what gets patients acquainted with hematology, but it’s a poor surrogate for thrombotic risk. The goal must be individualized for each patient.
If anticoagulation therapy isn’t the way to go, what’s the alternative?
Dr. Stein: Let’s say we feel anticoagulation is contraindicated or a patient wants to stop anticoagulation after a minimal duration. In those patients, we certainly encourage cytoreduction. We don’t know that there is any one cytoreductive that is best at reducing the risk of thrombosis. For patients with PV, we also make sure to control the hematocrit because this can reduce the risk for thrombotic events. In general, a patient at high risk is likely going to be on cytoreduction to hopefully reduce their risk of a recurrent event.
Dr. How: The most common reason a patient can’t be on anticoagulation is bleeding risk. Other strategies for these patients include looking at whether we can lower the dose of the anticoagulant. Can we give prophylactic dosing? We have a lot of good data for that in patients without cancer but not in patients with an MPN.
Most patients with an MPN should be on aspirin. If they are on an anticoagulant, I discontinue aspirin and leave them on an anticoagulant alone. Can they tolerate an antiplatelet agent if they can’t tolerate an anticoagulant? These are other things you could think about if someone can’t be on full-dose anticoagulation.
What about the use of antiplatelets?
Dr. Stein: Aspirin is often recommended in patients with ET, but we extrapolate the safety and efficacy from patients with PV. There are not strong data to suggest that we use aspirin for all patients with ET. We often cite a meta-analysis that showed uncertainty about the risk and benefits of aspirin in ET.2 This idea of universally recommended aspirin for patients with ET is something that is done but not always indicated. Sometimes we look by mutation. Oftentimes patients with CALR mutations have the highest platelet count, which tends to worry the patient and the clinician the most, but patients with CALR mutations can have higher bleeding with aspirin compared with thrombosis protection.
Dr. How: Aspirin is not always needed and can increase bleeding risk in certain patients with an MPN. Conversely, there has been some research suggesting that a higher dose of twice-daily aspirin has benefits.3 A recent study showed greater platelet inhibition with twice-daily aspirin and some decrease in symptoms, which might be helpful in some patients with ET.4
How do you address thrombosis in MF?
Dr. Stein: When we have been discussing thrombosis, we have been implicitly discussing ET and PV for the most part, but thrombosis is also seen in MF, though it is not the greatest complication for patients with MF. These patients are more likely to have inflammatory symptoms, problems from splenomegaly, problems with cytopenias, and progression. However, they can have thrombosis, but the rates are a bit lower than ET and PV. MF is an area where we have even less guidance, especially in patients with early or prefibrotic MF — here, we have less knowledge of which of those patients benefit from antiplatelet or cytoreductive therapy.
Dr. How: When you have a patient with ET progress to MF, it is still the same patient. Often, though, thrombotic risk becomes less of a priority because there are so many other competing risks with patients with MF, and you may be starting them on new drugs or considering transplant. These patients can also be thrombocytopenic, which further complicates anticoagulant use. I agree that there is definitely more that needs to be done to reduce thrombosis and bleeding risk in MF.
References
- National Comprehensive Cancer Network. Myeloproliferative neoplasms. Version 2.2024. Accessed September 26, 2024. https://www.nccn.org/professionals/physician_gls/pdf/mpn.pdf.
- Chu DK, Hillis CM, Leong DP, et al. Benefits and risks of antithrombotic therapy in essential thrombocythemia: a systematic review. Ann Intern Med. 2017;167(3):170-180.
- Rocca B, Tosetto A, Betti S, et al. A randomized double-blind trial of 3 aspirin regimens to optimize antiplatelet therapy in essential thrombocythemia. Blood. 2020;136(2):171–182.
- Tosetto A, Rocca B, Petrucci G, et al. Association of platelet thromboxane inhibition by low-dose aspirin with platelet count and cytoreductive therapy in essential thrombocythemia. Clin Pharmacol Ther. 2022;111(4):939–949.
Disclaimer: The following positions were assigned to the participants and do not necessarily reflect ASH opinions, the participant’s opinion, or what they do in daily practice.
Agree? Disagree? We want to hear from you! Send your thoughts and opinions on this controversial issue to [email protected].
