Making a Differential Diagnosis of Polycythemia Vera


Shared insight on making a differential diagnosis of polycythemia vera and properly risk-stratifying patients who present with this disease.


Pankit Vachhani, MD: Polycythemia vera is diagnosed at a median age of roughly about 60 years. In most patients, it is actually diagnosed in an incidental fashion. Someone goes to their primary care physician or some other physician for a regular blood check and is found to have elevated hemoglobin or hematocrit. However, there is a fraction of patients who do present with disease-related symptoms. Now, these could be headaches, or visual disturbances, or pruritis. Or the more complicated ones, including, for example, thrombotic episodes or bleeding episodes. For example, in a large series of about 1,500 patients who had WHO [World Health Organization]–defined polycythemia vera, almost a third of patients presented with palpable spleen, had pruritis, also had vasomotor symptoms like erythromelalgia. And in terms of thrombotic episodes, about 16% had an arterial thrombotic episode at the time or prior to their diagnosis. For venous thrombosis that incidence was 7%, and for major hemorrhagic episodes, it was 4%. But in addition to those symptoms, one should also know that elevated white blood cell count and platelet counts can also be present in polycythemia vera. Something on physical exam that we noticed is a facial platter, which is quite classic, but definitely not entirely specific to polycythemia vera. And erythromelalgia is something that we very well know at this point to be associated with polycythemia vera and essential thrombocytosis. Generally, it correlates with a platelet count of more than 400,000 at diagnosis.

Now, in terms of the diagnostic approach, or testing for polycythemia vera, the clinicians generally tend to follow and try to meet the parameters, the criteria that are part of the WHO major and minor criteria. What that means is that we want to have a sustained hemoglobin of 16.5 for men or more than 16 for women. The numbers in terms of hematocrit are more than 49% for men or 48% for women. That’s one, in other words, sustained erythrocytosis. But also, erythropoietin level should be below the reference range. Turns out to be the minor criteria. These are usually the first set of labs. In addition, one should check for the JAK2 V617F mutation, or in the rare case that it is not present, the JAK2 exon 12 mutations. These could be done from either peripheral blood or from a bone marrow sample, and typically most next-generation sequencing panels have these. And last but not the least, as part of the diagnostic workup is the very important point about bone marrow biopsy. In a classic polycythemia vera case, they should show hypercellularity with panmyelosis. And also, pleomorphic, meaning different sized mature megakaryocytes. The JAK2 mutations, the classic one is V617F mutation found in about a 95% of polycythemia vera patients, while the exon 12 mutation is found in a very small fraction, maybe about 1% to 2%, or maybe around, depending on the series, slightly more percent of patients. But again, the vast majority have the JAK2 V617F mutation.

Now, in terms of some of the differential diagnosis, one should note that if we are talking about WHO definition of polycythemia vera, the differential diagnosis is really vanishingly small. In fact, it’s really confined to patients who have the other JAK mutation positive myeloproliferative neoplasms, or in the rare event, for example, MDS [myelodysplastic syndrome], MPN [myeloproliferative neoplasm] overlap syndromes, where too, one may find JAK mutations. In particular, let me mention that essential thrombocytosis, and in some cases, myelofibrosis, needs to be verified on bone marrow biopsy and effectively made sure that those are not the underlying conditions. Finally, one must also remember that these other MPNs can also have unrelated secondary causes for having polycythemia vera in any individual patient. Outside of these more common scenarios, very rarely we find familial or hereditary cases of erythrocytosis. For example, EPOR [erythropoietin] receptor mutations can present, those patients can present, with a picture very akin to polycythemia vera. And for every polycythemia vera patient there are probably about 10 secondary polycythemia or secondary erythrocytosis patients. The 3 most common scenarios in which one finds secondary polycythemia are obstructive sleep apnea, which contributes to the chronic hypoxia. Also, tobacco smoking or smoking in general can cause a secondary polycythemia. And last but not the least, testosterone use. And yes, there are many other causes as well, but these are the 3 most common causes of secondary polycythemia.

Ruben Mesa, MD: When defining risk for polycythemia vera, I’m mindful of a couple of different concepts. First, risk as it relates to mortality related to the disease, and second, which is the more common situation, risk of having a vascular event, thrombosis or bleeding. Now, our assessment of risk, I think, has evolved over time. First, the assessment of risk of having thrombosis or bleeding has been primarily, in the past, age over 60 or individuals that have had a prior thrombotic event. Now, we recognize that this is not nearly granular enough. Over time, I think there have been several additional parts to the model. First, other contributing factors, an elevated white blood cell count through multiple different studies has been associated with a higher risk of thrombosis. Second, other contributing factors, cardiovascular risk, other poor areas of fitness, strong family history, and hypercholesterolemia. All of these other factors can also really mingle with those other P-vera [polycythemia vera]–related risks in terms of risks of vascular events. Over time, there have been additional prognostic scores developed by colleagues such as Ayalew Tefferi [MD at the Mayo Clinic] that have included age, the degree of leukocytosis, etc, in terms of more of a mortality type assessment.

Then finally, increasingly our new molecular mutations that we obtain on molecular profiling panels are helping to refine some longer-term risk of progression to myelofibrosis or to acute leukemia. I would say at the current time management decisions as it relates to treatment primarily around the risk of thrombotic or vascular events, as opposed to those markers that may predict mortality in 20 years versus 35 years, and lower risk being younger individuals without prior vascular events, without cardiovascular risk, and higher being those individuals that have those factors. I would share as well as I work to make my treatment decisions, I think about both risk and disease burden, meaning risk is how likely are you to have a vascular event, but disease burden can include difficult symptoms that can relate to the disease or sometimes difficult toxicities from the therapeutic approach. Let’s say you do not tolerate phlebotomies. I’m mindful as I come up with treatment decisions, both what is the risk and what is the disease burden. An individual may be low risk, but if they have significant disease burden symptoms and don’t tolerate phlebotomies, there still may well be an indication for beginning them on medical therapy.

Transcript edited for clarity.

Related Videos
Amer Zeidan, MBBS
Marina Kremyanskaya, MD, PhD
Douglas A. Tremblay, MD
Anthony M. Hunter, MD
Courtney D. DiNardo, MD, MSCE
Andrew Kuykendall, MD
Amy W. Zhou, MD
Amy W. Zhou, MD
Ricardo D. Parrondo, MD
Amy W. Zhou, MD