Pediatric Polycythemia Vera 

Updated: Oct 22, 2020
Author: Josef T Prchal, MD; Chief Editor: Hassan M Yaish, MD 

Overview

Practice Essentials

Polycythemia vera (PV) is a disorder of the multipotent hematopoietic stem cell that manifests as excess production of normal erythrocytes and variable overproduction of leukocytes and platelets. It is grouped with the Philadelphia chromosome–negative myeloproliferative disorders and can usually be differentiated from them by the predominance of erythrocyte production.[1] (See the image below.) Once polycythemia vera is suspected, the first step in evaluating a patient is determining whether the patient has primary, secondary, or apparent polycythemia. Treatment of the disease depends on whether it is in the plethoric phase or the spent phase.

Bone marrow film at 400X magnification demonstrati Bone marrow film at 400X magnification demonstrating dominance of erythropoiesis. Courtesy of U. Woermann, MD, Division of Instructional Media, Institute for Medical Education, University of Bern, Switzerland.

Signs and symptoms of polycythemia vera

Some patients with polycythemia vera are asymptomatic, whereas others have various nonspecific symptoms. Thirty percent of patients report headache, weakness, dizziness, and sweating (in order of decreasing frequency). Many of these symptoms can be attributed to excess hematocrit.

Alternatively, the patient may present with a complication of polycythemia vera. Approximately 33% of patients present with thrombosis or hemorrhage; 75% of these have arterial thrombosis, and 25% have venous thrombosis.

Potential physical findings include plethora and ruddiness of the face, erythromelalgia of the distal extremities, bruising, and splenomegaly. Specific attention should be directed towards sternal tenderness, which may indicate transformation to acute myeloid leukemia.[2]

Workup in polycythemia vera

Laboratory studies

Once polycythemia vera is suspected, the first step in evaluating a patient is determining whether the patient has primary, secondary, or apparent polycythemia.

A complete blood count (CBC), arterial blood gas (ABG) measurement, venous blood gas (VBG), and erythropoietin level can be used to differentiate patients.

Ferritin levels may also help to differentiate between primary and secondary polycythemias. Typically in primary polycythemia, the ferritin level is low due to constant overproduction of erythrocytes. In contrast, the ferritin level is usually normal in secondary polycythemia.

Imaging studies

Computed tomography (CT) scanning or ultrasonography of the abdomen can be used to assess the size of the spleen, which is frequently enlarged in polycythemia vera.

Histologic findings

Bone marrow and aspirate in polycythemia vera tend to be hypercellular. Some evidence of myelofibrosis may also be present.

In the plethoric phase, the blood smear shows normal erythrocytes, variable neutrophilia with myelocytes, metamyelocytes, and varying degrees of immaturity, basophilia, and increased platelets.

In the spent phase, the blood smear shows abundant teardrop cells, leukocytosis (or leukopenia), and thrombocytosis (or thrombocytopenia).

Management of polycythemia vera

Treatment of polycythemia vera depends on whether the disease is in the plethoric phase or the spent phase.

In the plethoric phase, the goal of treatment is controlling thrombotic episodes by restraining monoclonal proliferation rather than restoring polyclonal growth and maturation of cells. Interferon alfa is an exception; a few case reports have reported restoration of polyclonality.

In the plethoric phase, polycythemia vera is treated first by performing phlebotomy until the hematocrit is under reasonable control. Most patients can tolerate removal of 450-500 mL of blood every 2-4 days. As more blood is removed and the patient becomes iron deficient, the hematocrit becomes easier to control, and the phlebotomy schedule should be adjusted accordingly.

Risk stratification is important in deciding whether or not chemotherapeutic cytoreductive therapy is indicated.

Because of the theoretically possible leukemogenic risk of hydroxyurea, anagrelide has been used to control increased platelet counts, with the aim of reducing thrombotic events. In the PT1 trial in the United Kingdom, patients with essential thrombocythemia were randomly assigned to receive hydroxyurea or anagrelide. The study demonstrated an increased risk of thrombosis with anagrelide. The implication for polycythemia vera is unclear, but a reduction in platelet count does not affect rate of thrombosis in essential thrombocytopenia; a similar result is expected in polycythemia vera.

Pathophysiology

Clonality and erythropoietin (Epo) independence are the two key aspects of polycythemia biology. A single clonal population of erythrocytes, granulocytes, platelets, and variable clonal B cells arises when a hematopoietic stem cell gains a proliferative advantage over other stem cells. The T lymphocytes and natural killer cells remain polyclonal in polycythemia vera; this is related to their longevity. Clonality can currently only be tested in females using X-chromosome polymorphisms that take advantage of X-chromosome inactivation.

Epo independence is the ability of polycythemia vera hematopoietic stem cells to grow erythroid colonies without Epo. Although the colonies do not require erythropoietin, they remain responsive to it, and the erythropoietin receptor (EpoR) is normal, without defects in its function or quantity. Experiments using antibodies to neutralize Epo or block the EpoR do not abolish Epo-independent erythroid colony formation.

The understanding of the molecular mechanisms underlying polycythemia vera has been greatly enhanced over last decade. Genome-wide scanning that compared clonal polycythemia vera and nonclonal cells from the same individuals revealed a loss of heterozygosity (LOH) in chromosome 9p. This is found in approximately 30% of patients with polycythemia vera. This is not a classical chromosomal deletion but rather a duplication of a portion of a chromosome and the loss of the corresponding parental region. This process is called uniparental disomy.[3]

The 9p region contains a gene that encodes for the JAK2 tyrosine kinase. The JAK family of kinases is critical in cytokine receptor signaling and transmits the activating signal in the Epo-EpoR signaling pathway. Inhibition of JAK2 has been shown to eliminate Epo independence of erythroid progenitors. Subsequently, these observations were followed by the identification of a loss-of-function somatic mutation in an auto-inhibitory JAK2 domain, which essentially produces a gain-of-function mutation that affects the kinase. This occurs when a point mutation in exon 14 leads to a valine-to-phenylalanine mutation at codon 617 of the JAK2 gene.[4, 5]

The JAK2V617F mutation leads to constitutive phosphorylation, thus constitutive activity and STAT recruitment, which provides the proliferative advantage seen in polycythemia vera. This process occurs in the absence of Epo and explains both the Epo independence and Epo hypersensitivity of polycythemia vera colonies. A mouse model of this mutation produced a clinical phenotype consistent with polycythemia vera. These data were rapidly confirmed by several groups; each reported that more than 90% of patients with polycythemia vera carry the JAK2V617F mutation.[4] However, compelling data strongly argue that this mutation is not a disease-initiating mutation.[6] Rather, an as-of-yet unidentified mutation or mutations predispose to the acquisition of polycythemia vera. Patients with the JAK2V617F mutation tend to have the clinical phenotype of essential thrombocythemia.The quantitative allele burden (ratio of mutant to wild type expression) of JAK2V617F also has a clinical impact. Data from Vannucchi et al reveal that higher quantitative levels of the JAK2V617F allele correlated with higher values for hematocrit.[7] WBC and lactate dehydrogenase levels were positively correlated with the level of the mutation. The highest JAK2V617F levels at diagnoses predicted patients more likely to have splenomegaly, develop presenting pruritus, or eventually require chemotherapy.

Also, the rate of presenting major thromboses was positively correlated with higher mutation values. In fact, a multivariate analysis that included age, leukocytosis, hematocrit, platelet count, and therapies indicated that JAK2V617F/JAK2 wild type ratio is an independent risk factor for major vascular events. This was also validated by Silver et al, suggesting greater JAK2V617F allele burden correlates with higher white cell count, splenomegaly, and thromboembolic disease.[8, 9] They also suggested a higher frequency of myelofibrosis.

Carobbio et al demonstrated that the JAK2V617F allelic burden in JAK2V617F-positive essential thrombocythemia and polycythemia vera is the only risk factor that correlated with increased vascular events 5 years after diagnosis.[10]

A study by Szuber et al indicated that patients aged 40 years or younger with polycythemia vera are more likely than older patients with the disease to demonstrate a normal karyotype.[11]

Epidemiology

Frequency

United States

The incidence of polycythemia vera is reported to be 4.9 cases per 100,000 population in Baltimore. A more recent review of polycythemia vera in Connecticut reported an incidence of 22 cases per 100,000 population.

International

The incidence of polycythemia vera is reported to be 6.7 cases per 1,000,000 population in Israel, and reviews have estimated 30 cases per 100,000 population in Sweden and Italy. In Norway, the prevalence of polycythemia vera is reported to be 9.2 cases per 1,000,000 inhabitants.[12]

Race

The disease appears more common in Jews of European extraction than in most non-Jewish populations. Some familial forms of polycythemia vera are noted, but the mode of inheritance is not clear.

Sex

Men are preferentially affected over women. The male-to-female ratio is 1.2-2.2:1.

Age

Onset is typically in the sixth decade, and the peak incidence is at age 60-80 years. In the previously mentioned study by Szuber et al, the incidence of polycythemia vera in patients aged 40 years or younger with myeloproliferative neoplasms was 12%.[11]

Prognosis

The course of polycythemia vera may or may not follow two phases. The plethoric phase usually occurs first and is characterized by hyperproliferation of cellular components. The principle manifestations during this phase are thrombosis and hemorrhage. Consequently, treatment is aimed at ameliorating symptoms. The plethoric phase can last for a few years to as many as 20. Following the plethoric phase, the spent phase is characterized by progressive anemia, fibrosis, and splenomegaly. The smear demonstrates anemia, thrombocytosis (or thrombocytopenia), and leukocytosis (or leukopenia/neutropenia). In contrast to the plethoric phase, patients in the spent phase are often transfusion dependent.

The aforementioned study by Szuber et al found the incidence of fibrotic progression in polycythemia vera to be higher in patients aged 40 years or younger (22%) than in older patients, a phenomenon the investigators attributed to the longer survival period in the younger group.[11]

Patients are at risk for leukemic transformation throughout the entire course of disease although the rate is higher during the spent phase. The incidence of leukemia was found by the Polycythemia Vera Study Group (PVSG) to be affected by the mode of treatment.[13]  Treatment with phlebotomy only, Phosphorus-32 (P32), and chlorambucil resulted in a leukemic incidence of 1.5%, 10%, and 13% respectively. The study by Szuber et al determined that over a median follow-up period of 11.3 years, 4% of patients aged 40 years or younger with polycythemia vera underwent documented leukemic transformation.[11]

Except for potential leukemic transformation, appropriately treated polycythemia vera is compatible with near normal life. Without treatment, 50% of patients die within 18 months of diagnosis, usually from a thrombotic event. Survival with treatment depends on modality. Median survival is 13.9 years for phlebotomy alone, 11.8 years for P32, and 8.9 years for chlorambucil.

In the study by Szuber et al, the investigators found the median period of survival for polycythemia vera patients aged 40 years or younger to be 37 years, compared with 22 years for patients aged 41-60 years, and 10 years for patients over age 60 years.[11]

A European study, by Marchioli and colleagues, attempted to further define the prognosis of this disease;[14]  1,638 patients were prospectively followed in an attempt to describe the clinical history of polycythemia vera. The primary limitation of this study is a mean follow-up of 2.7 years. The overall mortality rate was 3.7 death per 100 persons per year. This was primarily caused by a moderate rate of cardiovascular death (1.7 deaths per 100 persons per year) and a high rate of death from noncardiovascular causes (1.8 deaths per 100 persons per year), primarily hematologic transformations. Cardiovascular mortality accounted for 45% of all deaths. Hematologic transformation (13% of all deaths) and solid tumors (19.5%) were also significant causes of mortality.

As previously seen in other studies, age older than 65 years and history of previous thrombosis were also significantly associated with mortality risk. Cumulative rate of cardiovascular events was 5.5 per 100 persons per year. Rates of combined malignancy, hematologic transformation, and non–polycythemia vera related malignancies were 3, 1.3, and 1.7 per 100 persons per year, respectively.

 

Presentation

History

Polycythemia vera (PV) frequently comes to the attention of clinicians because of an elevated hematocrit level found on routine laboratory testing. Some patients are asymptomatic, whereas others have various nonspecific symptoms that are recognized in the context of polycythemia vera. Thirty percent of patients report headache, weakness, dizziness, and sweating (in order of decreasing frequency). Many of these symptoms can be attributed to excess hematocrit.

Alternatively, the patient may present with a complication of polycythemia vera. Approximately 33% of patients present with thrombosis or hemorrhage; 75% of these have arterial thrombosis, and 25% have venous thrombosis. In the previously mentioned study by Szuber et al, the incidence of venous thrombosis was found to be higher in patients with polycythemia vera aged 40 years or younger than in older patients with the disease.[11] Cerebrovascular accidents, myocardial infarction, deep venous thrombosis, and pulmonary embolism (in order of decreasing frequency) can result from thrombosis due to polycythemia vera.

Budd-Chiari syndrome (hepatic vein thrombosis) is less common in polycythemia vera but is more specific to it. A patient who presents with Budd-Chiari syndrome should alert the physician to consider polycythemia vera because it is the most common underlying disease. Approximately 2-10% of patients with polycythemia vera have Budd-Chiari syndrome. Many presenting patients do not have an elevated hematocrit at the time of presentation but have other polycythemia vera laboratory abnormalities; if they survive, they eventually develop a myeloproliferative phenotype.

One study of 41 patients with idiopathic Budd-Chiari syndrome found that 58% of these patients had the JAK2V617F mutation.[15] Another group reported that patients with Budd-Chiari syndrome and the JAK2V617F mutation can have an elevated serum erythropoietin (Epo) level.[16] Classically, the presence of an elevated Epo level was believed to make the diagnosis of polycythemia vera extremely unlikely.

Less than 5% of patients have erythromelalgia (ie, erythema and warmth of the distal extremities, especially of the hands and feet, with a painful burning sensation that can result in digital ischemia if prolonged). A role for platelet aggregation has been proposed; in fact, the syndrome responds frequently (but not always) within hours to low-dose aspirin therapy.

Less commonly, polycythemia vera presents with cardiovascular symptoms due to myocardial infarction and congestive heart failure, pulmonary hypertension from chronic thromboembolic disease, neurological symptoms due to spinal cord compression by extramedullary hematopoiesis, and elevated uric acid with subsequent gout due to increased cell turnover.

Unrecognized hepatic or splenic vein thrombosis can result in portal hypertension and varices. Other GI symptoms include peptic ulcer disease, which occurs 4-5 times more frequently than in the general population. Hemorrhagic presentations are usually mild, with gum bleeding and easy bruising; however, serious GI hemorrhage can occur. Forty percent of patients experience life-altering pruritus. Typically, the pruritus is worse after a warm shower or bath; this is known as aquagenic pruritus. It has been attributed to increased numbers of mast cells and elevated histamine levels.

Physical

Potential physical findings include plethora and ruddiness of the face, erythromelalgia of the distal extremities, bruising, and splenomegaly. Specific attention should be directed towards sternal tenderness, which may indicate transformation to acute myeloid leukemia.[2]

Causes

See Pathophysiology.

 

DDx

Diagnostic Considerations

Polycythemia vera (PV) must be differentiated from other causes of polycythemia. The polycythemias can be subdivided by etiology into 3 groups: apparent or relative polycythemia, primary polycythemia, and secondary polycythemia.

A quick way to screen for polycythemia vera without excessive diagnostic testing is to determine if a hereditary pattern to the erythrocytosis is present. Because polycythemia vera is an acquired disorder, a familial pattern weighs against such a diagnosis; familial polycythemia vera has been reported, but in contrast with other familial polycythemias, the family clustering of polycythemia vera is associated with absence of phenotype at birth and an acquired polycythemic phenotype later in life. Rather, the phenotype at birth suggests the diagnoses such as high-affinity hemoglobin mutations, low 2,3 bisphosphoglycerate levels (BPG), primary familial and congenital polycythemia (PFCP), Chuvash polycythemia or rare mutations of HIF2a, or proline dehydrogenase type 2 genes. These disorders should be referred to a hematologist who is an expert in this area for specialized diagnostic testing and management.

Apparent or relative polycythemia is due to a decrease in plasma volume with a normal red cell mass. It is associated with hypertension, obesity, dehydration and stress, among other causes.[17]

Primary polycythemia is caused by intrinsic hyperproliferation of the hematopoietic stem cell independent of erythropoietin (Epo) stimulation or with exaggerated response to a low Epo level. Polycythemia vera, in which the hematopoietic stem cell proliferates independently of erythropoietin, is the most common primary polycythemia. The defining features of polycythemia vera are described in the Introduction. Another primary polycythemia is PFCP. The defect in PFCP is hyper-responsiveness to erythropoietin. One of its genetic causes has been defined: a hyperfunctional Epo receptor (a gain-of-function mutation) involving deletion of the negative regulatory subunit of the erythropoietin receptor (EpoR). Other mutations independent of the EpoR mutation are present but are as yet undefined.

Unlike polycythemia vera, which is a clonal acquired genetic mutation that can progress to leukemia, PFCP is a nonclonal germ line mutation that does not progress to acute leukemia. PFCP also differs from polycythemia vera in that only the erythroid lineage is affected.

Secondary polycythemia is due to elevated levels of Epo that induce erythrocyte proliferation; however, at the time of presentation, the increased RBC mass might have reached an equilibrium, and the Epo level is often within normal limits. The normal Epo level, however, would be inappropriately high for the elevated hematocrit. High Epo results from physiologically appropriate or inappropriate causes.

Physiologically appropriate secondary polycythemias result from hypoxia. Hypoxia is the common endpoint of the various causes of physiologically appropriate secondary polycythemias.

High-altitude polycythemia occurs because of lower ambient pO2 resulting in tissue hypoxia. Acute compensation occurs through hyperventilation, but chronic compensation involves elevation of hematocrit (although the degree of response varies between individuals). Not all populations accommodate to high altitude by polycythemia. The Tibetans have a lower hemoglobin level than expected but have high levels of exhaled nitric oxide, which may be the end product of a process improving oxygen delivery by inducing vasodilation and increasing blood flow to the tissues.

In cardiopulmonary disease, impaired respiration and circulation result in tissue hypoxia and subsequently increased erythropoietin.

Smoking results in the formation of carboxyhemoglobin that does not carry oxygen and results in higher oxygen affinity in other hemoglobin molecules. This results in tissue hypoxia, which induces Epo production. The rise in hematocrit is compounded by the reduction in plasma volume due to smoking.

Defects in bisphosphoglycerate mutase and phosphofructokinase result in decreased 2,3 BPG. BPG is necessary for hemoglobin to transition from a high oxygen affinity state to a low oxygen affinity state. Thus a decreased BPG level results in tissue hypoxia in erythrocyte enzyme defect polycythemia

Individuals with hemoglobinopathy with high affinity mutations (autosomal dominant inheritance) are unable to transition from high oxygen affinity to low oxygen affinity states due to impaired intramolecular rotation or BPG binding. Deoxygenation is impaired in some cases.

Methemoglobinemia is usually due to a cytochrome b5 reductase (methemoglobin reductase) deficiency but can also be caused by various mutations of globin genes, such as Hemoglobin M.

Cobalt is believed to inhibit oxidative metabolism controlling Epo production. It is not effective treatment for anemia. Cobalt has been used as a foam stabilizer in beer and has been shown to cause an acquired polycythemia when unintentionally ingested in high amounts.

Physiologically inappropriate polycythemia is often due to exogenous sources of erythropoietin.

Several malignancies have been shown to produce erythropoietin. These include hepatoma, renal cell carcinoma and cerebellar hemangiomas. Uterine myomas have been reported to produce erythropoietin. However, another mechanism by which these often large bulky tumors produce erythrocytosis is mechanical interference with the blood supply to the kidneys resulting in false sensing of hypoxia and Epo production.

Endocrine disorders such as pheochromocytomas, aldosterone producing adenomas, Barter syndrome, and dermoid cysts of the ovary can result in inappropriate Epo through mechanical interference with renal blood supply or hypertensive damage to renal parenchyma resulting in false sensing of hypoxia by the kidneys and subsequent Epo production, functional interaction between aldosterone, renin and erythropoietin, and inappropriate Epo secretion by the tumor. Androgens increase hematocrit by 2 mechanisms: stimulation of Epo production and an independent hyperproliferative effect on erythrocyte precursors.

Chuvash polycythemia is an endemic polycythemia found on the west bank of the Volga River in the Chuvash Autonomous Republic in western Russia. It is an autosomal recessive disorder characterized by a mutation in the VHL gene that prevents ubiquitin degradation of hypoxia inducible factor (HIF)-1, resulting in upregulation of downstream target genes, including Epo production. As such, Chuvash polycythemia can be grouped with the secondary inappropriate polycythemias. But because of a second defect resulting in hyper-responsiveness to erythropoietin, not linked to the EpoR, it also has some features of primary polycythemia. Clinically, patients with Chuvash polycythemia have normal ABG, normal calculated p50 of hemoglobin, normal to increased Epo levels, and no abnormal hemoglobins.

Rare mutations of HIF2a or proline dehydrogenase type 2 genes are associated with secondary congenital polycythemia; because of their rarity, the phenotype is not fully elucidated as yet.

Renal polycythemia is due to Epo produced by renal cysts, polycystic disease, or hydronephrosis.

Erythrocytosis can occur after renal transplant and is thought to be due to increased activity of the angiotensin II-angiotensin receptor 1 pathway. Angiotensin II may also modulate release of Epo and insulinlike growth factor (IGF)-1. Venous canalization studies have shown the source to be the "nonfunctional" native kidneys. Removal of the native kidneys can normalize the hematocrit; however ACE inhibitors can also control this typically transient erythrocytosis, thus avoiding surgery.

Neonatal polycythemia is an appropriate secondary polycythemia due to increased oxygen affinity of fetal hemoglobin and subsequent tissue hypoxia. This response can become excessive and inappropriate in the setting of maternal diabetes or placenta to child transfusion.

Differential Diagnoses

 

Workup

Laboratory Studies

Once polycythemia vera (PV) is suspected, the first step in evaluating a patient is determining whether the patient has primary, secondary, or apparent polycythemia.

A CBC, ABG measurement, VBG, and erythropoietin level can be used to differentiate patients. A CBC typically reveals increased leukocytes, platelets, and erythrocytes in primary polycythemia, whereas, in secondary and apparent polycythemia, only the erythrocytes are elevated.[18] Primary familial congenital polycythemia (PFCP) is an exception; it only has elevated erythrocytes but not leukocytes or platelets. However, it can be distinguished from secondary polycythemia by its erythropoietin level. An erythropoietin (Epo) level is almost always low or low-normal in primary polycythemia; in secondary polycythemia, it is elevated or high-normal when hematocrit is high.

Budd-Chiari syndrome in patients with the JAK2V617F mutation and elevated Epo levels has changed this absolute criteria in the diagnosis of polycythemia vera. An ABG reveals secondary appropriate polycythemia if it reveals hypoxia. Finally,the VBG allows calculation of the P50 value; if this is low, it suggests a high oxygen affinity hemoglobin or 2,3-bisphosphoglycerate (BPG) deficiency.

Ferritin levels may also help to differentiate between primary and secondary polycythemias. Typically in primary polycythemia, the ferritin level is low due to constant overproduction of erythrocytes. In contrast, the ferritin level is usually normal in secondary polycythemia.

Red cell mass has been used to distinguish apparent polycythemia from secondary and primary polycythemia. However, the test is expensive and requires expertise. Also, the131 I-albumin used to measure plasma volume is not available in the United States and is difficult to handle because of its radioactivity. Consequently, the authors do not routinely use this test at the University of Utah School of Medicine because its diagnostic value is limited when the hematocrit level is clearly abnormal. The use of this test is frequently limited in clinical practice due to availability; however, some clinicians feel very strongly about it, especially in patients with borderline hemoglobin levels. It can occasionally identify a patient with an elevated red cell mass whose hematocrit is normal because of an increased plasma volume and can also identify patients whose hematocrit is only elevated due to a reduced plasma volume.

Secondary polycythemia must be differentiated into appropriate and inappropriate causes. As mentioned above, an elevated Epo level with a hypoxic ABG suggests secondary appropriate polycythemia, whereas an elevated Epo level without hypoxia suggests secondary inappropriate polycythemia. Determining the cause of appropriate polycythemia can proceed using the history, although specialized testing such as a p50 curve can be used to identify high-affinity hemoglobins due to structural hemoglobin defects or enzyme deficiencies. Hemoglobin electrophoresis is insufficient to identify hemoglobin structural defects because some hemoglobin mutants are missed. Secondary inappropriate polycythemia causes can be sorted out using judicious imaging and specialized endocrine testing.

Of the primary polycythemias, PFCP must be differentiated from polycythemia vera. These 2 diagnoses differ in clonality and in vitro responsiveness of peripheral blood erythroid progenitors to erythropoietin. Clonality testing relies on polymorphisms based on X chromosome inactivation and therefore can only be done in females. Polycythemia vera is clonal; PFCP is not. Endogenous erythroid colony responsiveness to Epo also differentiates polycythemia vera from PFCP. Polycythemia vera is characterized by growth independence from erythropoietin. In contrast, PFCP is not growth independent from erythropoietin, although it is hyperresponsive. Endogenous erythroid colony testing is not routinely available and can only be done in specialized laboratories. The authors frequently use it in our laboratory in difficult cases.

Imaging Studies

Computed tomography (CT) scanning or ultrasonography of the abdomen can be used to assess the size of the spleen, which is frequently enlarged in polycythemia vera.

Renal pathology, cerebellar hemangioblastomas, and pheochromocytomas that can cause secondary polycythemia may also be detected.

Histologic Findings

Bone marrow and aspirate in polycythemia vera tend to be hypercellular.

Some evidence of myelofibrosis may also be present.

In the plethoric phase, the blood smear shows normal erythrocytes, variable neutrophilia with myelocytes, metamyelocytes, and varying degrees of immaturity, basophilia, and increased platelets.

In the spent phase, the blood smear shows abundant teardrop cells, leukocytosis (or leukopenia), and thrombocytosis (or thrombocytopenia).

In a 2016 revision to its classification of myeloid neoplasms and acute leukemia, the World Health Organization (WHO) recognized bone marrow morphology’s usefulness “as a reproducible criterion for the diagnosis of” polycythemia vera.[19, 20]

Staging

Because many practitioners do not have access to specialized clonality testing or erythroid colony assays, many of the criteria for diagnosis of polycythemia vera do not require them, although they are taken into account. No consensus has been reached on diagnostic criteria.

The World Health Organization (WHO) criteria for polycythemia vera diagnosis requires 2 components: reasonable elimination of apparent and secondary polycythemia and confirmation of polycythemia vera. However, the discovery of the JAK2V617F mutation have made these criteria insufficient. A proposed set of revised criteria have recently been published.

Diagnosis requires the presence of both major criteria and one minor criterion or the presence of the first major criterion together with 2 minor criteria.[21, 22]

Major criteria

See the list below:

  • Hemoglobin level of more than 18.5 g/dL in men, more than 16.5 g/dL in women, or other evidence of increased red cell volume (hemoglobin or hematocrit levels >99th percentile of method-specific reference range for age, sex, altitude of residence; hemoglobin level >17 g/dL in men, >15 g/dL in women [if associated with a documented and sustained increase of at least 2 g/dL from the individual’s baseline value that cannot be attributed to correction of iron deficiency], or elevated red cell mass >25% above mean normal value).

  • Presence of JAK2V617F or other functionally similar mutation such as JAK2 exon 12 mutation

Minor criteria

See the list below:

  • Bone marrow biopsy showing hypercellularity for age, with trilineage growth (panmyelosis) with prominent erythroid, granulocytic, and megakaryocytic proliferation (not validated in prospective studies)

  • Serum erythropoietin level below the reference range for normal

  • Endogenous erythroid colony formation in vitro

Other groups have proposed and are preparing other potential diagnostic criteria. Criticism of the new WHO 2008 revised criteria revolves around the substantial interobserver variability in diagnosing polycythemia vera by bone marrow histology and the difficulty of community practitioners to test for endogenous erythroid colonies.

 

Treatment

Medical Care

Treatment of polycythemia vera (PV) depends on whether the disease is in the plethoric phase or the spent phase.

In the plethoric phase, the goal of treatment is controlling thrombotic episodes by restraining monoclonal proliferation rather than restoring polyclonal growth and maturation of cells. Interferon alfa is an exception; a few case reports have reported restoration of polyclonality.

In the plethoric phase, polycythemia vera is treated first by performing phlebotomy until the hematocrit is under reasonable control. Most patients can tolerate removal of 450-500 mL of blood every 2-4 days. As more blood is removed and the patient becomes iron deficient, the hematocrit becomes easier to control, and the phlebotomy schedule should be adjusted accordingly. Although phlebotomy is effective for controlling erythrocytosis, it does not affect the variable leukocytosis, thrombocytosis, or thromboembolic events found in polycythemia vera.

For many years, the mainstay of therapy of polycythemia vera has been phlebotomy with a goal hematocrit level of less than 45% in men and less than 42% in women. This recommendation is based on retrospective data that are now almost 30 years old and, the authors believe, potentially inaccurate. Recently, DiNisio and colleagues published data in patients with polycythemia vera that suggested that differences in hematocrit in the range of 40-55% were not associated with the risk of thrombosis nor with mortality.[23]

Landolfi et al performed an extensive retrospective review of 1638 PV patients studied as part of the European collaboration study on low-dose aspirin in polycythemia (ECLAP).[24] In this trial, no correlation was observed between hematocrit and risk of thrombosis.

Limitations of this study include its retrospective nature and relatively short follow-up (2.8 y median). Therefore, the authors of this Medscape Reference article believe that the true hematocrit goal in polycythemia vera is not clear (if it is present at all). This issue remains to be sorted out in prospective fashion. Although phlebotomy is still recommended by many experts, it is clearly a controversial issue.

In most patients, low-dose aspirin is started to reduce the risk of thromboembolic events, and phlebotomy is continued as necessary to control the hematocrit. This recommendation is based on results of the ECLAP study, in which patients with polycythemia vera and no clear indications for aspirin were randomly assigned to receive aspirin 100 mg/d or no aspirin.[24]

The study showed a minor but statistically significant decreased risk of cardiovascular death, nonfatal myocardial infarction, nonfatal stroke, pulmonary embolism, and major venous thrombosis with aspirin therapy. Also, low-dose aspirin did not significantly increase rates of hemorrhage.

The retrospective review of the ECLAP data confirmed older age and previous thrombosis as risk factors for thrombotic events.

Although no correlation was demonstrated between thrombosis risk and hematocrit, what was shown was that a WBC count of more than 15,000/μL, when compared with a WBC < 10,000/μL, was an independent risk factor for major thrombosis, primary manifested as an increased risk of myocardial infarction.

Risk stratification is important in deciding whether or not chemotherapeutic cytoreductive therapy is indicated.

Most agree that high risk for thrombosis is present when the patient is older than 70 years and/or has a previous history of thrombosis. Note that the age is different than in risk stratification of essential thrombocythemia.

A platelet count of more than 1.5 million/μL is a risk factor for bleeding and is frequently considered a high risk indication favoring chemotherapeutic cytoreduction. Uncontrolled traditional cardiac risk factors, especially smoking, are considered by most to place a patient in a high-risk category. However, when these factors are well controlled, therapy for high risk may not be justified. Whether an additional agent should be given depends on the patient's thrombotic risk. The decision is a tradeoff between a reduction in thrombotic events and an increased incidence of malignancy. The initial Polycythemia Vera Study Group (PVSG) investigators compared phlebotomy, phlebotomy with32 P, and phlebotomy with chlorambucil. Median survival was 13 years, 11 years, and 9 years, respectively. The incidence of thrombosis was 23% in the phlebotomy-only group versus 16% in the32 P-and-phlebotomy group.

The rate of acute myeloid leukemia was 1.5%, 10%, and 13% for phlebotomy, phlebotomy with32 P, and phlebotomy with chlorambucil, respectively. Because of the increased rate of acute myeloid leukemia in polycythemia vera treated with chlorambucil, this drug is no longer used for myelosuppression. Rates of GI and skin cancers also increased 4%, 9%, and 12% when patients were treated with phlebotomy, phlebotomy with32 P, and phlebotomy with chlorambucil, respectively. Clearly, myelosuppression reduces the incidence of thrombotic events but increases the risk of malignancy.

Because of these results, a phase II efficacy trial was performed by using hydroxyurea instead of chlorambucil or32 P to see if a less leukemogenic agent could control thrombosis. In that trial, 51 patients with polycythemia vera were given hydroxyurea 30 mg/kg/day for 1 week then 15 mg/kg/day with the goal to maintain a platelet count of < 600,000/cm3 and a hematocrit of < 50% with minimal phlebotomy. The incidence of thrombosis in the first 2 years of treatment (when most thromboses occur) was 9%, significantly lower than the historical control of 23% for phlebotomy alone in the PVSG trial. At a median follow up of 8.6 years, the incidence for acute myeloid leukemia was 6% for hydroxyurea compared with 1.5% for phlebotomy only. At the time of analysis, this difference was not statistically significant but the later addition of 2 cases of myelodysplasia in the hydroxyurea arm increased the incidence to a significant 8%.

Because of the theoretically possible leukemogenic risk of hydroxyurea, anagrelide has been used to control increased platelet counts, with the aim of reducing thrombotic events. In the PT1 trial in the United Kingdom, patients with essential thrombocythemia were randomly assigned to receive hydroxyurea or anagrelide. The study demonstrated an increased risk of thrombosis with anagrelide. The implication for polycythemia vera is unclear, but a reduction in platelet count does not affect rate of thrombosis in essential thrombocytopenia; a similar result is expected in polycythemia vera.

Interferon has been used for myeloproliferative diseases with efficacy in the past, but toxicity/tolerance has always limited its use in patients. However, in a recent phase II study by Kiladjian et al, pegylated interferon alfa-2a (Pegasys) was administered to 40 patients with polycythemia vera (median follow-up, 31.4 mo).[25] A completed hematologic response was achieved in 94.6%, with 7 patients achieving complete molecular response of the JAK2V617F that was durable. Most patients tolerated interferon well, and no vascular events were recorded. The acceptable tolerability, efficacy and extremely low leukemogenic risk may make interferon alfa first line therapy in the future.

Based on the above data, in the authors' clinical practice, chemotherapeutic cytoreductive therapy is used in all patients who are high risk. Generally, the drug of choice is hydroxyurea. The authors attempt to titrate the drug to achieve normalization of the WBC count. This is based on the data stated above. The authors are honest with patients that this treatment is based on retrospective data that still need to be prospectively proven. The authors also monitor the hematocrit level, and although phlebotomy is performed for symptoms or very high values, the authors do not feel that fully achieving a goal of 45% in men and 42% in women is required.

Patients who are low risk generally do not require chemotherapeutic cytoreductive therapy. However, the concern of increased risk of thrombosis (primarily myocardial infarction) due to leukocytosis brings into question whether or not low-risk patients with a WBC count of more than 15,000/uL should receive cytoreductive therapy. This question remains to be addressed in a prospective fashion; currently, cytoreductive therapy in this situation cannot be firmly recommended.

As stated above, the authors believe that current recommendations to phlebotomize to a goal hematocrit of 45% in men and 42% in women may be inaccurate. Clearly, patients with symptomatic hyperviscosity should receive phlebotomy sufficient to relieve their symptoms. The authors also consider phlebotomy in patients with a very high hematocrit (>55%), but do not feel bound by current guidelines, based on the above data.

All patients receive low-dose aspirin, usually 81 mg, unless a contraindication is noted.

As more clinical data is collected on pegylated interferon alfa-2a (Pegasys), some recommend its use as a potential first-line agent for cytoreduction; however, a planned randomized prospective study of Pegasys versus Hydrea should resolve the issue of optimal first-line polycythemia vera therapy.

A literature review by Bewersdorf et al indicated that pegylated and nonpegylated interferon alfa can be safely and effectively used as long-term therapy for polycythemia vera and essential thrombocythemia. Polycythemia vera and essential thrombocythemia had overall response rates to interferon alfa (as calculated via a composite of complete response, partial response, complete hematologic response, and partial hematologic response) of 76.7% and 80.6%, respectively. Meta-regression analysis demonstrated no significant difference in these rates between the pegylated and nonpegylated agents. For polycythemia vera, annualized rates of 0.5% and 6.5% were found for thromboembolic complications and treatment discontinuation resulting from adverse events, respectively, with the figures for essential thrombocythemia being 1.2% and 8.8%.[26]

Special treatment situations

P32 is a reasonable option in the patient who is unreliable or who has a limited lifespan because of the convenience of one injection resulting in long term control. The principle drawback is the increased risk of malignancy.

In pregnant women, interferon can be used to treat polycythemia vera. The mechanism is unclear, the side effects are moderate and often severe, and the drug is expensive. However, it is not teratogenic, it can reasonably control symptoms, and there are rare case reports of restoration of polyclonality. Interferon is also a reasonable consideration in a young patient because of possible concerns of leukemogenicity of hydroxyurea. Recent data suggest that pegylated interferon alfa-2a (Pegasys) is likely to be equally (and possibly more) effective. It is dosed once weekly which likely improves compliance. Some suggest that it may also be better tolerated than standard interferon.

Erythromelalgia responds to low-dose aspirin or reduction of the platelet count to normal with low-dose myelosuppressive agents.

Pruritus can be disabling and life altering in polycythemia vera. High water temperatures and vigorous skin rubbing are factors in inciting itching. Taking cooler baths and patting the skin dry can provide some symptomatic relief. Also, starch baths (half a box of Linet starch in a tub of water) can be effective. Pharmacologic treatment options include antihistamines (eg, cyproheptadine at 4 mg orally 3 times daily), histamine 2 (H2) receptor blockers (eg, cimetidine at 300 mg orally 4 times daily), photochemistry, danazol, and interferon alfa. Serotonin reuptake inhibitors (eg, paroxetine at 20 mg orally daily or fluoxetine at 10 mg orally daily) can also be used. In severe refractory cases, myelosuppression may be required.

Patients with symptomatic hyperuricemia (gout, urate kidney stones) receive allopurinol. The authors also obtain uric acid levels and treat asymptomatic hyperuricemia if the level is significantly elevated.

Surgical Care

Polycythemia vera is not treated surgically, except in the spent phase when splenectomy may be performed to relieve symptoms related to mass effect and pancytopenia.

Patients undergoing surgery have a very high risk of postoperative thrombosis.

Consultations

All patients suspected of having polycythemia vera should be referred to a hematologist.

Diet

Beer with cobalt foam stabilizers should be avoided. Otherwise, a normal, healthy diet is recommended.

Activity

Patients with polycythemia vera can have a normal active life.

 

Medication

Medication Summary

Chemotherapeutic cytoreductive therapy is used in all patients who are high risk but not in those who are low risk. All patients receive low-dose aspirin.

Biologic response modifiers

Class Summary

Biologic response modifiers elicit antiproliferative, antiviral, and immunomodulating effects. They inhibit cellular growth and alter cellular differentiation.

Interferon alfa-2a, recombinant (Pegasys)

Protein produced in response to viral infection and other inflammatory stimuli. The exact mechanism is unknown, but it is believed to exert an antiproliferative effect. Produced by recombinant DNA techniques in E coli. In PV and ET, it has shown to have long term efficacy, and reduces JAK2 allelic burden. Sporadic case reports have noted cytogenetic remissions, suggesting a possible biologic effect. 

Antiplatelet agents

Class Summary

These agents prevent formation of thrombi-associated polycythemia.

Aspirin (Anacin, Ascriptin, Bayer)

Irreversibly acetylates platelet cyclooxygenase, resulting in a decrease in thromboxane A2, the prostaglandin responsible for platelet shape change, granule release, and aggregation. Prescribed for most patients with polycythemia vera.

Anagrelide (Agrylin)

Inhibits post mitotic megakaryocyte maturation. Unlike hydroxyurea, selectively inhibits platelet proliferation. In polycythemia vera used only to control platelet counts, but shows slight decrease in mean hemoglobin and hematocrit while white cell counts maintained. Inhibits cyclic nucleotide phosphodiesterase and the release of arachidonic acid from phospholipase, possibly by inhibiting phospholipase A2. Can be used in addition to hydroxyurea for particularly difficult to control thrombocytosis in polycythemia vera.

Antineoplastic Agent

Class Summary

These agents are used off-label for polycythemia, but pediatric doses are extrapolated from pediatric treatment regimens, including leukemia and myelodysplastic syndrome.

Hydroxyurea (Droxia, Hydrea)

Nonalkylating, myelosuppressive, S-phase agent. Inhibits ribonucleotide reductase, the enzyme that converts ribonucleotides into deoxynucleotides, thereby depleting deoxynucleotide and inhibiting DNA synthesis. Cellular proliferation is ultimately inhibited, and leukocytes, erythrocytes, and platelets are decreased. The mechanism of action is probably different than the one exerted by hydroxyurea in the treatment of sickle cell disease. A misconception is that hydroxyurea is leukemogenic. No studies have conclusively demonstrated that hydroxyurea is more leukemogenic than baseline in myeloproliferative disease.

 

Follow-up

Further Outpatient Care

In the plethoric phase, patients initially require close follow-up for monitoring of blood counts. The dose of hydroxyurea must be closely monitored until a steady state is achieved and phlebotomy may occasionally be required in symptomatic patients.

In the spent phase, therapy needs to switch from removal of cells to transfusion of cells to relieve anemia. Leukemic transformation needs to be managed expectantly but has a poor prognosis.

Further Inpatient Care

Most patients with polycythemia vera (PV) can be managed as outpatients.

Occasionally a symptomatic patient who has an extremely high hematocrit may need hospitalization for emergent phlebotomy. These patients should receive aggressive volume replacement with saline. Their CBC counts should be closely monitored.

Complications

As outlined in detail above, potential complications of this disorder are primarily thromboembolic and hematologic.

Increased cardiovascular morbidity and mortality can be significant, although appropriate treatment is felt to significantly reduce these risks. Unfortunately, whether current therapies can reduce the risk of hematologic transformation is unclear. Therapies that reduce the mutant clone population, such as interferon and possibly JAK2 inhibitors in the future, will hopefully decrease this risk.

Prognosis

For many patients, a normal or near normal life span can be anticipated. However, polycythemia vera carries significant potential morbidity and mortality, even when correctly treated.

As outlined by Marchioli et al, cardiovascular events are more common in this population and pose an ever present risk for these patients.[14]

Transformation to myelofibrosis decreases anticipated survival and, although uncommon, transformation to acute leukemia portends a very poor prognosis.

Please refer to Morbidity/Mortality for further details.