The myeloproliferative neoplasms (MPNs) comprise a group of clonal hematopoietic stem cell disorders characterized by proliferation of one or more myeloid lineage. MPNs occur mainly in adults between 5^th and 7^th decade of life, and the annual incidence is estimated to be 6-10 per 100 000 population (1). The term “myeloproliferative disorders” (MPDs) was proposed in 1951 by Dameshek to encompass several disorders with similar clinical phenotype. They included polycythemia vera (PV), essential thrombocythemia (ET), primary myelofibrosis (PMF), chronic myelogenous leukemia (CML) and erythroleukemia (2). The last 5 years have brought a pivotal progress in our understanding of the pathogenesis of classical Ph-negative MPDs. It was associated with the identification of the specific molecular marker- the JAK2V617F point mutation, which seems to be involved in PV, ET and PMF (3). It is now clear that patients with this mutation are biologically distinct from those without the mutation and that the mutation is associated with different disease phenotype. Moreover, several novel mutations have been recently identified, but their pathogenic and prognostic relevance has not been established yet (4). The current World Health Organization (WHO) classification for hematological malignancies has incorporated the recently discovered mutations into the diagnostic approach. The term “MPDs” has been changed to “MPNs”. The WHO classification includes 8 clinical entities and it was presented in table 1 (5).

This current review will focus on classical Ph-negative MPNs and includes molecular basis of pathogenesis, risk stratification, diagnostic and therapeutic management.

JAK2 is located on chromosome 9p24 and it belongs to the Janus family of non-receptor protein tyrosine kinase. Janus kinase/signal transducer and activator of transcription (JAK-STAT) signaling remains crucial for survival and normal function of hematopoietic and other cells (6). The JAK2V617F mutation is of particular relevance to MPNs including mainly PV, ET and PMF (3). The JAK2V617F results from a somatic G to T mutation involving JAK2 exon 14. It leads to nucleotide shift at position 1849 and the substitution of valine to phenyloalanine at codon 617. Mutated JAK2 causes cytokine-independent growth of cells and their hypersensitivity to cytokines (7). The JAK2V617F point mutation is found in approximately 95% of patients with PV and 40-50% of patients with ET and PMF (3, 8). Two-step model may explain the development of cells which are homozygous for JAK2 mutation. Initially, the occurrence of the JAK2V617F gives rise to a heterozygous clone which replaces normal hematopoietic cells. The second step includes mitotic recombination within cells heterozygous for the JAK2V617F with subsequent uniparental disomy. As a consequence, the daughter cells are homozygous for JAK2V617F and they replace the heterozygous clone. It may happen that myeloid cell population includes a variable proportion of JAK2 mutant alleles; it means that cells with JAK2 mutation may coexist with cells lacking this abnormality. The first step is characterized by progressive increase of mutant alleles from 0 to 50%, whereas the second step from 50 to 100% (9). The experimental studies in mouse model demonstrated that both heterozygous and homozygous JAK2 mutations induce PV-like disease phenotype, but homozygous JAK2 mutation is associated with more aggressive disease with progression to post-PV myelofibrosis (10). The JAK2V617F homozygosity is frequently seen in PV and PMF. In contrast, high mutant allele burden is rarely seen in ET (11). Homozygosity in granulocytes can be detected in 30% of patients with PV (3), but it was demonstrated in approximately 90% of colonies of hematopoietic progenitor cells derived from patients with PV whereas homozygous progenitor colonies were not observed in ET (12). This difference in the occurrence of homozygosity may result from the substantial variation in the rate of mitotic recombination among persons and the role of genetic and environmental factors determining susceptibility to mitotic recombination is very likely (11). However, it should be clearly underlined that variability in the allele burden is not sufficient to distinguish different clinical entities. It does not help us to answer the question how one mutation gives rise to three different clinical phenotypes. Moreover, there is an increasing number of evidence that the JAK2V617F is not an initiating event in MPNs and that pre-JAK2 mutated clone may exist (13). Some studies suggest that deletions of 20q and other cytogenetic abnormalities are present in about 10% of patients with MPNs and that this deletion may occur before the JAK2 mutation. It was demonstrated that deletions 20q are exclusively V617F-positive (14). In conclusion, the pathogenic role of JAK2V617F mutation requires to be better defined, especially in the light of the presence of novel mutations which may occur in patients with MPNs (4).

As it was aforementioned, JAK2 allele burden is associated with clinical phenotype and PV patients homozygous for V617F mutation have been found to have a more severe disease (15). Recently published report has demonstrated that patients with an allele burden equal or > 50% had higher white blood cell (WBC) count, greater spleen size and lower platelet count than those with less than 50% of mutant dosage. It was also proved that patients homozygous for JAK2 mutation had significantly higher risk of developing myelofibrosis (16).

JAK2 exon 12 mutations were identified in PV who were negative for JAK2V617F point mutation. Those patients presented with an isolated erythrocytosis, distinct bone marrow morphology and low serum erythropoietin level. It is noteworthy that erythroid colonies have grown from their blood samples in the absence of exogenous erythropoietin. In contrast to JAK2V617F-positive PV patients, JAK2 exon 12 mutated PV subset is often heterozygous for this mutation (17).

MPL (myeloproliferative leukemia virus oncogene) located on chromosome 1p34, is the thrombopoietin receptor. MPL is highly expressed in early hematopoietic progenitors and in cells of megakaryocytic lineage. MPL is the key growth and survival factor for megakaryocytes (18). The two most common MPL mutations are W515L and W515K. They were found in about 10% of patients with PMF and in less than 10% of ET-patients lacking the JAK2V617F mutation (19). MPL-mutated ET patients are older, have lower hemoglobin concentration, higher platelet count and higher risk of arterial thrombosis (20) whereas the presence of MPL mutation in patients with PMF is associated with female gender, older age, lower hemoglobin concentration and blood transfusion dependence (21).

TET2 is located on chromosome 4q24 and its function is probably associated with epigenetic regulation of transcription (22). TET2 mutations are present in about 10-15% of patients with MPNs, including both the JAK2V617F – positive and negative cases (23). It should be highlighted that TET2 mutations may coexist with other relevant mutation e.g. KITD816V for mast cell disease (24). Moreover, it was demonstrated that TET2 mutations may precede the acquisition of a JAK2 mutation. Up-to-date studies show that the presence of TET2 mutations in MPNs do not affect survival, leukemic transformation and the risk of thrombosis (23).

Several novel mutations have been recently found in patients with MPNs:

1) ASXL1 (Additional Sex Combs-Like 1,

2) CBL (Casitas B-lineage lymphoma proto-oncogene),

3) IDH1 and IDH2 (Isocitrate dehydrogenase 1 and 2),

4) IKZF1 (IKAROS family zinc finger 1),

5) LNK,

6) EZH2.

All these mutations are rarely seen in patients with chronic phase of MPNs whereas their frequency may exceed 20% in blast transformation (tab. 2) (4).

The results including the analysis of JAK2 haplotype may suggest that hereditary component confers susceptibility to MPNs and some of above-mentioned mutations may play a role at different stages of disease evolution. A strong association between the risk of developing a JAK2V617F-positive MPNs and a germline haplotype (46/1 or GGCC) has been recently demonstrated (fig. 1) (25).

The discovery of JAK2 mutations (either V617F or JAK2 exon 12 mutations) in more than 95% of patients with PV was a turning point in the classification and diagnosis of all classical MPNs. There are several important issues. This genetic marker has been found only in myeloid neoplasms, but not in non-myeloid malignancies (8). Therefore, JAK2 mutation remains a sensitive marker of PV, but it is not specific for any one MPNs. The presence of JAK2 mutation may confirm the diagnosis of myeloid malignancy, but its absence does not exclude the MPNs. It should be underlined that this mutation is not seen in healthy volunteers (26). In PV and PMF patients the absence of JAK2 mutation has limited diagnostic utility since about half the patients lacking this marker (5). A small subset of patients with ET and PMF possess MPL mutations, but their incidence remains very low and therefore the diagnostic utility is highly limited (20). It is noteworthy, that there are lacking standardized methods of JAK2 assessment and therefore we should realize the possibility of false positive and negative results. It may occur when very sensitive diagnostic method was implemented or in case of low tumor burden e.g. mononuclear cells from peripheral blood were used for assay (27). In conclusion, the JAK2 and MPL mutations are not necessary to establish a diagnosis of MPNs, but they could be helpful when present. If not, the diagnosis of MPNs is still likely. A diagnostic diagram using JAK2 genotyping was proposed in patients suspected to have MPNs (28) (fig. 2).

Due to above-mentioned ambiguities concerning JAK2 mutations, a new WHO classification system requires bone marrow histology which may help in distinguishing between 3 classical MPNs. Importantly, the 2008 WHO criteria include several additional, clinical and laboratory findings which are specific for particular MPNs.

Because of similarity with secondary causes, diagnosis of different MPNs remains challenging. The introduction of the new molecular markers seem to facilitate the final diagnosis, but the presence of mutations should be integrated with other characteristic findings.

Definition: PV is a chronic myeloproliferative neoplasm characterized by increased red cell production. The presence of JAK2 mutations in virtually all patients with PV results in proliferation of not only erythroid, but also granulocytic and megakaryocytic lineages. We can distinguish three phases of disease: 1) pre-polycythemic phase with mild erythrocytosis 2) overt polycythemia with an increase of red cell mass and 3) myelofibrotic “spent” phase. PV may infrequently progress to acute myeloid leukemia or myelodysplastic syndrome (5).

Epidemiology: The annual incidence of PV varies between 0.7 to 2.6 per 100 000 population with a median age at diagnosis of 60 years and male predominance (5).

Clinical manifestations: The main clinical features encompass venous and arterial thrombosis. Most thrombotic events occur in the two years before diagnosis (29). Arterial thrombosis includes myocardial infraction, ischemic stroke and peripheral arterial occlusion. Venous thrombosis is represented by lower extremity deep venous thrombosis, pulmonary embolism and splanchnic vein thrombosis (28). The other symptoms include headache, visual disturbances, pruritus and erythromelalgia. There are hepatomegaly, splenomegaly and plethora on physical examination (30).

Diagnostic criteria: see table 3 (5).

Risk-stratification: Older age (> 60 years) and a previous history of thrombosis remain risk factors for thrombosis. In high-risk patients both factors are present whereas in low-risk neither of these is demonstrated. Intermediate risk group includes patients with cardiovascular risk factors, such as hypertension, smoking, hyperlipidemia and diabetes (28, 29). Moreover, recently published reports have showed that WBC count > 15.0 x 10⁹/L and high JAK2V617F allele burden are independent risk factor for thrombosis (15).

Treatment: The treatment should adhere the standard risk stratification. In low and intermediate-risk patients, phlebotomy and low dose of aspirin are recommended. High-risk patients require phlebotomy, low-dose aspirin and myelosuppressive agents (hydroxyurea or interferon-alpha). Either hydroxyurea or interferon-alpha is first-line cytoreductive therapy at any age (31).

Therapeutic management in pregnancy: Female-PV patients should be stratified according to risk factors which include 1) previous major thrombotic or bleeding complications and 2) previous pregnancy complications (pre-eclampsia, stillbirth, > 3 first-trimester or > 1 second and third-trimester losses). High-risk pregnancy is defined by the presence of at least one of the above-mentioned risk factors. This subgroup should be treated with phlebotomy (target hematocrit below 45%), aspirin, low molecular weight heparin and interferon-alpha. In low-risk group (no risk factors present) phlebotomy, aspirin and low molecular weight heparin are recommended (31).

Prognosis: Mortality rate is age-dependent and it is estimated to be 1.6 fold and 3.3 fold higher than in the reference population in patients younger and older than 50 years respectively. Major causes of death are fatal thrombotic events and progression to AML (32).

Definition: ET is a chronic myeloproliferative neoplasm that involves the megakaryocyte lineage. ET is characterized by significant thrombocytosis in peripheral blood and increased numbers of mature megakaryocytes in the bone marrow (5).

Epidemiology: The annual incidence of ET is estimated to be 0.6-2.5 per 100 000 persons per year. Male and female are equally affected. A median age at diagnosis is 60 years, but it may occur in patients at the age of 30 years and infrequently in children (33).

Clinical manifestations: Thrombosis and hemorrhage represent main clinical symptoms. Some studies demonstrated that the cumulative rate of thrombosis during the disease course ranged from 1.9% to 3% per year (34). Involvement of the microcirculatory system is typical for ET and it may manifest as erythromelalgia, ischemic attacks, visual defects, headache and paresthesia (5). Bleeding resulting from acquired von Willebrand disease, occurs most commonly from mucosa of gastrointestinal and upper respiratory tracts (35).

Diagnostic criteria: see table 4 (5).

Risk stratification: Older age (> 60 years) and history of thrombosis are standard risk factors for ET (the same as it was mentioned for PV) (28, 29). Some reports suggested that patients with high WBC count had higher risk for thrombosis. The best leukocyte cutoff values for predicting the thrombotic events was found to be 9.4 x 10⁹/L (36). The British Study Group (MRC-PT1) presented risk stratification system for ET patients. High-risk group includes age> 60 years, prior thrombotic or bleeding events, vascular risk factors and platelet count > 1000 x 10⁹/L, for intermediate group belong ET patients at the age between 40-60 years with no vascular risk factors and no history of thrombosis/bleeding and with a platelet count less than 1500 x 10⁹/L. The low-risk patients fulfill the criteria for intermediate group but are younger than 40 years (37).

Treatment: Low and intermediate risk patients with asymptomatic disease do not require therapy. Low-dose aspirin is an option. High-risk group needs the introduction of myelosuppressive therapy. It was proved that hydroxyurea is superior than anagrelide in preventing arterial thrombosis, although venous thrombosis was reduced in the anagrelide arm. It should be kept in mind that aspirin is contraindicated in extreme thrombocytosis due to acquired von Willebrand disease (38). Interferon-alpha is recommended for younger patients and in pregnancy, but this agent is often poor tolerated. Anagrelide may effectively control platelet count especially in patients resistant or intolerant to hydroxyurea (34).

Prognosis: This is an indolent disease, but survival of ET patients is reduced by about 2-fold compared with general population. Major causes of death include thrombotic events and transformation to AML. Bone marrow fibrosis may develop as a disease termination, but it is rarely seen (5, 19).

Definition: Primary myelofibrosis is a myeloproliferative neoplasm with an increased proliferation of megakaryocytes and granulocytes in the bone marrow associated with marrow fibrosis and extramedullary hematopoiesis. Initially, bone marrow is hypercellular with minimal deposition of fibrous connective tissue (prefibrotic phase). Subsequently, after several years of disease course, marrow picture evolves to a fibrotic phase with collagen fibrosis or even osteosclerosis (5).

Epidemiology: The incidence is estimated to occur at 0.5-1.5 per 100 000 persons per year. A median age at diagnosis is 6^th-7^th decade, but it may occur in young adults. Both genders are equally affected (39).

Clinical manifestations: About 30% of patients are asymptomatic at the time of diagnosis. The remaining ones complain of fatigue, dyspnea, weight loss, night sweats, fever, thrombotic and bleeding events. Splenomegaly and hepatomegaly are detected in 90% and 50% respectively (5, 39).

Diagnostic criteria: see table 5 (5).

Risk stratification: There are three key prognostic scoring systems. The most used is “Lille score” which includes anemia (Hgb < 10g/dL) and abnormal leukocyte count (< 4 or > 30 x 10⁹/L). Based on these values we may distinguish three prognostic categories: low (no risk factors), intermediate (1 risk factor) and high (2 risk factors) with a median survival of 93, 26 and 13 months respectively (40). The new International Prognostic Scoring System (IPSS) uses 5 risk factors present at diagnosis: 1) age > 65 years, 2) hemoglobin< 10 g/L, 3) leukocyte count > 25 x 10⁹/L, 4) circulating blasts ≥ 1% and 5) the presence of constitutional symptoms. The presence of 0,1,2 and ≥ 3 factors defines low, intermediate-1, intermediate-2 and high risk disease with median survivals of 11.3, 7.9, 4 and 2.3 years, respectively (41). Recently, a Dynamic IPSS-plus scoring system (modified IPSS) was introduced for use at any time of disease course. This system includes three additional adverse risk factors: 1) red cell transfusion dependence, 2) platelet count ≤ 100 x 10⁹/L and 3) unfavorable cytogenetics. Using DIPSS-plus classification we may define low-risk category (no risk factors), intermediate-1 (1 risk factor), intermediate-2 (2 or 3 risk factors) and high (≥ 4 risk factors). Median survivals were 15.4, 6.5, 2.9 and 1.3 years, respectively (42). It was also demonstrated that presence or absence of molecular markers such JAK2, IDH and TET2 did not influence either survival and leukemic transformation whereas unfavorable karyotype or platelet count ≤ 100 x 10⁹/L did (43).

Treatment: The only therapeutic approach that has resulted favorably in survival is the application of allogeneic stem cell transplantation (allo-SCT), but one should kept in mind the high risk of death or chronic graft versus host disease (GVHD). Therefore, in every particular case one must consider risk/benefit ratio (43). Currently, it seems that this procedure should be reserved for high-risk patients and both myeloablative and reduced-intensity conditioning regimens have been used with similar efficacy. In patients who relapse after allo-SCT, a graft-versus-myelofibrosis effect could be demonstrated (44). Tefferi et al. (43) proposed to adjust the treatment to risk-groups. In asymptomatic patients with low or intermediate-1 DIPSS-plus score “watch and wait” strategy is recommended; whereas if these patients present symptomatic disease (it is unusual in this group), one of the conventional agents is preferred e.g. in anemic transfusion-independent patients with no splenomegaly and low serum erythropoietin level, the use of erythropoietin may result in a response rate of more than 50% (45). In the remaining cases one can choose between several agents including steroids, androgens, thalidomide and lenalidomide (43). PMF-patients with intermediate-2 or high risk category can be managed by allo-SCT (if fit enough), conventional and experimental agents, splenectomy and radiotherapy. All treatment modalities except of allo-SCT remain palliative. The use of lenalidomide is justified only in patients with del(5q) (46). It should be mentioned that several new agents are in clinical trials now including pomalidomide, hypomethylating agents and JAK inhibitors (43). The most promising is JAK inhibitor-CYT 387 with an 40% response rate in anemia, splenomegaly and constitutional symptoms (47). Myelofibrosis treatment algorithm was presented on figure 3 (43).

Prognosis: Life expectancy is 30% lower than in an age-matched and sex-matched population, with a median survival of 5 years (32). Major causes of death include thrombotic and bleeding events, infections, pulmonary hypertension and transformation to AML (48).

The breakthrough in the understanding of molecular basis of classical Philadelphia-negative MPNs was the identification of the JAK2V617F point mutation. It has changed dramatically our current view on disease pathogenesis and has direct implications in diagnostic algorithm as well as in therapeutic management. The 2008 WHO classification system for MPNs has integrated clinical, histological and genetic information and therefore facilitate the diagnostic strategy. A pivotal progress in molecular investigations we have witnessed, will certainly speed up further studies and subsequent revisions of WHO classifications are likely.