2
To my Family,
always there for me
3
TABLE OF CONTENTS
INTRODUCTION_________________________________________________________________ 5
CHRONIC MYELOPROLIFERATIVE NEOPLASMS ______________________________________________ 5 THE CLA““IC PH-NEGATIVE MYELOPROLIFERATIVE NEOPLASMS _____________________________ 11
Polycythemia Vera ___________________________________________________________________________ 12 Essential Thrombocythemia ___________________________________________________________________ 16 Myelofibrosis _______________________________________________________________________________ 20
THE MOLECULAR PATHOGENESIS OF MPN _________________________________________________ 31
Mutations affecting the phenotypic driver genes __________________________________________________ 33 Subclonal mutations__________________________________________________________________________ 42
AIM OF THE STUDY _____________________________________________________________ 55 METHODS ____________________________________________________________________ 57
PATIENTS AND SAMPLES _______________________________________________________________ 57 SAMPLE PREPARATION ________________________________________________________________ 57 WHOLE GENOME AMPLIFICATION _______________________________________________________ 58
GENOTYPING _________________________________________________________________ 59
Polymerase chain reaction and direct sequencing ___________________________________________ 59 Quantitaive Real-Time PCR _____________________________________________________________ 63 Capillary electrophoresis _______________________________________________________________ 64 Next Generation Sequencing ____________________________________________________________ 65 Anti-calreticulin antibody preparation ____________________________________________________ 69 Production of recombinant mutated calreticulin ____________________________________________ 70 Western blot analysis __________________________________________________________________ 70 Immunostaining and silver impregnation __________________________________________________ 70 Calreticulin gene expression analysis _____________________________________________________ 71 Statistical analysis_____________________________________________________________________ 72
RESULTS _____________________________________________________________________ 73
PROGNOSTIC IMPACT OF MUTATIONS IN A LARGE SERIES OF PATIENTS WITH MYELOFIBROSIS _____________ 73
IMPACT OF DRIVER AND HMR MUTATIONS ON CLINICAL PHENOTYPE AND PROGNOSIS IN PATIENTS WITH POST-PV AND POST-ET MYELOFIBROSIS___________________________________________________________ 84
4
IMPACT OF CALRETICULIN MUTATIONS ON ET AND PMF _____________________________________ 90 DIFFERENTIAL PROGNOSTIC IMPACT OF TYPE 1/TYPE 1-LIKE VERSUS TYPE 2/TYPE2-LIKE CALR
MUTATIONS IN MYELOFIBROSIS __________________________________________________ 98
MUTATIONAL LANDSCAPE OF PATIENTS WITH MYELOFIBROSIS THAT DO NOT HARBOR MUTATIONS IN
JAK2, MPL AND CALRETICULIN DRIVER GENES _______________________________________ 102
CALRETICULIN MUTATION-SPECIFIC IMMUNOSTAINING IN MPNs _________________________ 106
DISCUSSION ___________________________________________________________ 112 CONCLUSION __________________________________________________________ 122
5
INTRODUCTION
CHRONIC MYELOPROLIFERATIVE NEOPLASMS
Myeloproliferative Neoplasms (MPNs) are a group of related hematological malignancies characterized by an overproduction of mature blood cells in one or more lineages and a tendency to transform to acute myeloid leukaemia (Cambpell, 2006). Single cell origin of hematopoiesis is considered to be a hallmark of all myeloid neoplasms and the genesis and evolution of diseases are strictly connected with molecular events such as translocated chromosomes, deleted or amplified regions and point mutations in single genes (Vannucchi, 2009; Rampal, 2014). In 1951, William Dameshek introduced the term elop olife ati e diso de s MPD to e o pass pol the ia e a PV , esse tial thrombocythemia (ET), primary myelofibrosis (PMF), chronic myelogenous leukemia (CML). Demeshek, was the first to suggest that these conditions might be related, recognising both the phenotypic overlap and tendency for phenotypic shift in these diseases (Dameshek, 1951). The association of the Philadelphia (Ph)-chromosome with CML i , disti guished the othe th ee diso de s as lassi Ph-negative MPD. The first systematic attempt to classify MPD and MPD-like clinicopathologic entities was undertaken by the World Health Organization (WHO) committee for the classification of hematologic malignancies. According to the 2001 WHO classification system, CML, PV, ET, and PMF were i luded u de the atego of h o i elop olife ati e diseases CMPD . The CMPD category also included othe o lassi MPD-like disorders such as chronic neutrophilic leukemia (CNL), chronic eosinophilic leukemia/hypereosinophilic syndrome (CEL/HES), and u lassified CMPD Vardiman, 2003). The identification of BCR-ABL as a CML-specific genetic event, in the context of CMPD, has facilitated accurate molecular diagnosis and effective targeted therapy. The lack of knowledge, until few years ago, on specific genetic defects in the other BCR-ABL-negative classic CMPDs necessitated that diagnosis rest on a combination of bone marrow histology and a few clinical and laboratory findings to distinguish clonal from reactive myeloproliferation and one CMPD from another.
The last 8 years have witnessed fundamental advances in understanding the molecular pathogenesis of classic BCR-ABL-negative CMPD, however the defining moment was the identification of an activating mutation in the JAK2 tyrosine kinase. As a result, WHO
6 diagnostic criteria have been revised, a d the te CMPD has ee ha ged to
eloproliferative eoplas s MPN Tefferi, 2008; Vardiman, 2009).
Apart from the BCR/ABL rearrangement in CML, originated by a reciprocal translocation between chromosomes 9 and 22, t(9;22)(q34; q11) (Hehlmann, 2009), or the chimeric FIP1L1-PDGFRA mRNA in some forms of eosinophilia, and kit mutations in cases with systemic mastocytosis, information concerning molecular abnormalities of MPN has been scanty until 2005, when four international groups described the same acquired point mutation in the JAK2 gene in most patients with PV and around half those with ET or MF (James, 2005; Levine, 2005) .Diverse approaches were used to identify this mutation, comprising dissection of signalling pathways in PV, high throughput sequencing of kinase genes and sequencing of candidate genes within a region of chromosome 9 known to undergo loss of heterozygosity in PV patients. JAK2, one of four JAK family cytoplasmic tyrosine kinases (comprising JAK1, JAK2, JAK3 and TYK2), is essential for signalling by the erythropoietin receptor (EpoR) and thrombopoietin receptor (MPL) , and is also involved in signalling through the granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony stimulating factor and interferon-γ receptors (Baxter, 2005; Kralovics, 2005; Levine, 2007; Delhommeau, 2006). Studies of EpoR indicate that ligand binding results in a conformation change in the receptor, with consequent phosphorylation of JAK2 and the receptor itself resulting in activation of downstream signalling pathways. JAK2 also plays a vital role in EpoR trafficking, with absence of JAK2 resulting in retention of EpoR within the endoplasmic reticulum (Dupont, 2007). The central role of JAK2 in haematopoiesis is highlighted by a JAK2 knock-out mouse, which dies at embryonic day 12.5 due to a complete absence of definitive erythropoiesis. The JAK2 V617F substitution, resulting from a single base change, affects the pseudokinase (JH2) domain of the protein. This domain is required for both JAK2 activation and inhibition of basal kinase activity. The JAK2 V617F mutation results in substitution of a phenylalanine at a highly conserved residue, and is thought to impair autoinhibition of JAK2, leading to constitutive activation of tyrosine kinase activity. Expression of JAK2 V617F leads to cytokine independent growth of various cytokine dependent cell lines, with constitutive activation of pathways implicated in the control of proliferation, differentiation and cell survival such as STAT5, PI3K/AKT and MAPK (Levine, 2005; James, 2005; Kralovics, 2005; Baxter, 2005).
7 In the following 2 years, additional mutations in JAK2 and MPL were reported (Table 1). These different mutant alleles all result in a gain of function due to the constitutive activation of tyrosine kinase-dependent cellular signaling pathways, particularly of the JAK-STAT pathway (Scott, 2007; Pikman, 2007).
Genetic Abnormality Disease Frequency
BCR-ABL Chronic myelogenous leukemia ≅99%
JAK2V617F
Polycythemia vera >95%
Essential thrombocythemia ≅60%
Primary myelofibrosis ≅60%
MPN, unclassifiable ≅20%
Refractory anemia with sideroblasts
and thrombocytosis (RARS-T) ≅50%
JAK2 exon 12 Polycythemia vera ≅2%
MPLW515L/K
Primary myelofibrosis ≅8%
Essential thrombocythemia ≅8%†
Involving PDGFRA
Myeloid neoplasms with eosinophilia Unknown
Mast cell disease Unknown
Involving PDGFRB Myeloid neoplasms with eosinophilia Unknown Involving FGRF1 Myeloid neoplasms with eosinophilia Unknown Involving KIT (D816V as
the most frequent)
Mast cell disease
Unknown
Table 1. Recurrent molecular abnormalities associated with Myeloproliferative Neoplasms.
Transplantation of JAK2V617F mutated cells induced a PV-like phenotype in recipient mice, accompanied by leukocytosis of a different extent and eventually followed by changes suggestive of myelofibrotic transformation. More recently, by manipulating expression levels of the V617F allele, mice with an ET-like phenotype were also generated in the presence of low levels of mutated JAK2. Overall, these models indicated that the JAK2V617F mutation is sufficient to induce a MPN-like phenotype in mice and
8 suggested that the level of mutated allele may influence disease phenotype (Levine, 2007; Tiedt, 2008).
Mutational frequency of JAK2V617F is estimated to be more than 95% in PV, 60% in ET or PMF, 40% to 50% in refractory anemia with ringed sideroblasts and thrombocytosis (RARS-T), whereas it is very rare in AML or MDS. In most patients with PV or PMF, as opposed to a minority of those with ET, the mutation is harbored in a homozygous state, which is accomplished by mitotic recombination. In general, the highest V617F allele burden, that is the level of mutated allele relative to normal allele in a cell suspension such as granulocytes, is found in patients with PV followed by PMF and ET (Levine, 2005; Baxter, 2005); however, such variability in the allele burden does not represent a sufficient criterion for distinguishing among different clinical entities, nor does it satisfactorily help to e plai the appa e t pa ado of o e uta t allele-diffe e t li i al phe ot pes . In fact, how a single V617F mutation can be the basis of different clinical disorders, as in the classic MPN, is still unclear. Interestingly, single nucleotide polymorphisms (SNPs) in JAK2 have been associated preferentially with the diagnosis of PV, supporting the contribution of inherited host genetic characteristics to MPN phenotypic variability. Regardless, there is evidence to suggest that JAK2V617F may not be the initial clonogenic e e t i MPN a d that a p e-JAK2 utated ell a e ist. In support of this is also a finding that leukemic blasts in patients who evolve to AML from a pre-existing JAK2V617F-positive MPN are often negative for the JAK2V617F mutation. Conversely, JAK2V617F, or other JAK2 mutations, are likely a necessary component of the PV phenotype because they are detected in virtually all patients with the disease and are sufficient to reproduce the phenotype in mice. In summary, JAK2V617F mutation is integral to the classic MPN, but its exact hierarchical position in pathogenesis and its role in phenotypic variability remain to be clarified (Tefferi, 2010; Vainchenker , 2011).
In patients with a clinical picture suggestive of PV and who were found to be negative for the JAK2V617F mutation, several genetic abnormalities (ie, mutations, deletions, insertions) have been detected in a short region of JAK2 exon 12 (Scott, 2007). These mutations, which probably account for less than 2% of patients with PV, affect autonomous cell proliferation and differentiation in a fashion similar to that of the V617F allele. Another recurrent molecular abnormality of MPN is represented by somatic mutations at codon 515 of MPL, which, as is the case with JAK2V617F, involve early myeloid and lymphoid
9 progenitors. MPL (named after myeloproliferative leukemia virus oncogene homolog) is the receptor for the cytokine thrombopoietin (Tpo) and is highly expressed in early hematopoietic progenitors and in cells of the megakaryocytic lineage. MPLW515L induced both cytokine-independent growth and Tpo hypersensitivity in cell lines, resulting in constitutively activated JAK-STAT/ERK/Akt signaling pathways, and caused a PMF-like disease in mice (Pikman, 2006; Levine, 2007). The gene encoding for the receptor of platelet-derived growth factor A (PDGFRA) is involved in at least four different genetic abnormalities associated with eosinophilia. The most frequent and best characterized abnormality is due to a karyotypically occult microdeletion at chromosome 4q12, where PDGFRA is located, resulting in a chimeric FIP1L1-PDGFRA fusion gene. The latter encodes for an aberrantly activated tyrosine kinase as the consequence of disruption of the autoinhibitory activity encoded by PDGFRA exon 12, where the breakpoint is located; this constitutively active tyrosine kinase drives autonomous eosinophil progenitor proliferation, possesses transforming properties in vitro, and induces a myeloproliferative disorder with extensive eosinophil proliferation when expressed in transplanted mice. The fusion gene has been demonstrated at the level of hematopoietic stem cell compartment. In addition, the Beta type of PDGFR has been reported as being involved in rearrangements associated with imatinib-responsive eosinophilia. The PDGFRBis located at chromosome 5q31-32 and may fuse with different partners. One of the most common is the ETV6/TEL gene on chromosome 12p13, which encodes for a transcription factor with nonredundant roles in normal hematopoiesis. The fusion protein constitutively activates the cellular pathways normally associated with PDGFRB signaling and has transforming properties when expressed in cell lines (Skoda, 2007; Levine, 2008; Vannucchi, 2009). A D816V mutation located in the catalytic domain of the tyrosine kinase receptor c-Kit occurs in systemic mastocytosis. c-Kit is the receptor for stem cell factor, a key cytokine involved in the generation and differentiation of mast cells from primitive hematopoietic progenitors; it is encoded by kit, located at chromosome 4q12. The D816V and other homologous mutations induce growth factor independent growth and cell differentiation in mast cell lines through activation of STAT5/PI3K/AKT signaling pathways (Dai, 1994).
The main modification in the 2008 WHO classification has been the substitution of the att i ute eoplas fo disease . I fa t, ot ithsta di g the a al sis of the X chromosome inactivation pattern in informative females and other cytogenetic and/or
10 ole ula fi di gs that esta lished oth lassi a d o lassi eloproliferative disorders as being clonal stem cell disorders, and the finding that evolution to AML is part of their natural history, the neoplastic nature of these conditions has been mostly dismissed until recently. The 2008 WHO classification for myeloid neoplasms, which incorporates novel information derived from molecular discoveries in BCR-ABL negative lassi elop olife ati e states a d lo al eosi ophili diso de s, i ludes fi e ajo entities (Table 2) as follows: the Acute Myeloid Leukemia (AML) and the Myelodysplastic Syndromes (MDS) with their different subtypes, whose listing is outside the scope of this review; the Myeloproliferative Neoplasms (MPN); the category of overlapping Myelodysplastic/Myeloproliferative Neoplasms (MDS/MPN); and the Myeloid Neoplasms associated with eosinophilia and specific molecular abnormalities. AML is defined by the p ese e of eithe ≥ % last ells i the o e a o a d/o pe iphe al lood o e tai characteristic cytogenetic abnormalities. The MDSs are recognized and distinguished from MPN primarily on the basis of the presence of trilineage dyshematopoiesis in the absence of monocytosis in both bone marrow and peripheral blood.
. A ute eloid leuke ia AML a d elated p e u so eoplas s . M elod splasti s d o es MD“
. M elop olife ati e eoplas s MPN
. . Ch o i eloge ous leuke ia CML , BCR-ABL positi e . . Pol the ia e a PV
. . Esse tial th o o the ia ET . . P i a elofi osis PMF . . Ch o i eut ophili leuke ia CNL
. . Ch o i eosi ophili leuke ia, ot othe ise lassified CEL-NO“ . . H pe eosi ophili s d o e
. Mast ell disease . MPN u lassifi a le
. M elod splasti /M elop olife ati e eoplas s MD“/MPN
. . Ch o i elo o o ti leuke ia CMML . . Ju e ile elo o o ti leuke ia JMML
. At pi al h o i eloid leuke ia, BCR-ABL egati e . . M elod splasti / elop olife ati e eoplas , u lassifia le . . ‘ef a to a e ia ith i g side o lasts asso iated ith a ked th o o tosis
. M eloid a d l phoid eoplas s ith eosi ophilia a d a o alities of PDGFRA, PDGFRB, o FGFR
. . M eloid a d l phoid eoplas s asso iated ith PDGFRA ea a ge e t . . M eloid eoplas s ith PDGFRA ea a ge e t
. . M eloid a d l phoid eoplas s ith FGFR a o alities
11
The fou lassi MPNs ie, CML, PV, ET, a d PMF ha e ee disti guished f o the othe o lassi MPNs, which include chronic neutrophilic leukemia (CNL), chronic eosinophilic leukemia-not otherwise specified (CEL-NOS), systemic mastocytosis (SM), and unclassifiable forms of MPN. Classic MPNs have been further functionally classified based on the presence or absence of the t(9;22) chromosomal translocation in the Philadelphia (Ph) chromosome resulting in the BCR-ABL 1 fusion protein (hallmark feature of Chronic Myeloid Leukemia, CML). Thus, it s o ell esta lished that the th ee ai lassi al Ph-negative MPNs are PV, ET and PMF.
THE CLA““IC PH-NEGATIVE MYELOPROLIFERATIVE NEOPLASMS
Classic Ph-negative MPNs are among the most frequent hematologic neoplasms, usually affecting the adult elderly population; however, they can also be found in children, and in this instance, they raise specific diagnostic and management issues. Epidemiological studies on chronic myeloproliferative neoplasms conducted during the last five decades have reported variable annual incidence rates. In PV, annual incidence rate per 100.000 inhabitants varies in the range of 1.9-2.6.As regards ET and PMF the same variability is observed with annual incidence rates ranging from 0.6-2.5 and 0.3-1.5 per 100.000 inhabitants, respectively. The median age at diagnosis is 69-74 years in PV, 67-72 years in ET and 67-76 years in PMF. There is a male predominance in PV and PMF, whereas ET is most common in females. Familial clustering of these disorders is known, and even before the discovery of JAK2V617F mutation, this observation led to a suggestion of predisposition allele(s). This concept was further strenghtened when a clear molecular distinction of true familial MPN from other familial sindrome such as familial erythrocytosis and hereditary thrombocythemia has become possible using clonality markers, cellular studies and JAK2 mutation analysis. A clear example of germline genetic factors influencing MPN pathogenesis was the discovery of the GGCC (also known as 46/1) haplotype of the JAK2 gene (Jones, 2009; Olcaydu, 2009). Somatic mutations of JAK2 in MPN do not distribute equally between the two most common JAK2 gene haplotypes in Caucasian populations. The GGCC haplotype acquires over 80% of all V617F mutations as well as exon 12 mutations. The GGCC haplotype predisposes carries for JAK2 mutation positive MPN, and thus its major role is in the disease initiation. However, the hypothesis that GGCC haplotype
12 might account for familial clustering of MPN has recently been disproved in a study showing equal haplotype frequency in sporadic and familial MPN cases. The reason why the GGCC haplotype has negligible role in familial MPN is its weak ability to initiate the disease phenotype.
The three clinical entities share several common features, such as their origin in a multipotent hematopoietic stem cell, a relatively normal cellular maturation, a striking overlap in clinical presentation (apart from PMF, which has its own peculiar manifestations). Patie ts ith PV a d ET ha e a high isk of th o oti a d/o hae o hagi e e ts a d a p og ess to a a ele ated elofi osis phase post-polycythemic or post-thrombocythemic myelofibrosis), hile all th ee su t pes a e asso iated ith a lo g te isk of t a sfo atio to a ute eloid leuke ia AML ith a u ifo l poo p og osis. The isk of leuke i t a sfo atio is highest i p i a elofi osis PMF , he e it is esti ated to e app o i atel – % at ea s. I pol the ia e a PV the isk is . % at ea s a d . % at ea s Tefferi, 2013). T a sfo atio to AML is o side ed elati el u o o i esse tial th o o the ia
ET Vardiman, 2009).
P
OLYCYTHEMIAV
ERAPolycythemia Vera (PV) is characterized by abnormal expansion of erythroid lineage, possibly associated with leukocytosis and thrombocytosis. Diagnostic criteria for PV were initially established by the Polycythemia Vera Study Group (PVSG) and were based only on clinical parameters; in 2001 bone marrow panmyelosis was included in the minor criteria of WHO classification. Absence of BCR/ABL fusion gene is needed to exclude CML as well as causes of reactive erythrocytosis must be excluded. Karyotypic alterations do not impact on prognosis but are considered a main diagnostic criterion in that they point to the existence of clonal hematopoiesis and allow to exclude reactive erythrocytosis. Cytogenetic abnormalities occur in 30% of patients and are predominantly represented by chromosomes 8 and 9 trisomy and 20q deletion; alteration of chromosome 5, 6, 12, 13 and 1q duplication were reported. After the identification of JAK2 mutations, that occurs virtually in >95% of PV patients, some of the previous criteria became useless. Currently, according to WHO 2008 criteria for PV (Table 3) diagnosis requires both increase of red cell
13 mass and presence of JAK2 mutation, as the two major criteria, together with one of the minor criterion or the first major criteria and two minor criteria.
Major Criteria
Hemoglobin > 18.5 g/dL in men, 16.5 g/dL in women or other evidence of increased red cell volume
Presence of JAK2 V617F or other functionally similar mutation such as
JAK2 exon 12 mutation Minor Criteria
Bone marrow biopsy showing hypercellularity for age with trilineage growth (panmyelosis) with prominent erythroid, granulocytic, and megakaryocytic proliferation
Serum erythropoietin level below the reference range for normal Endogenous erythroid colony formation in vitro
Table 3 – WHO 2008 diagnostic criteria for PV.
The incidence of PV is estimated at 0.79 cases/100.000 per year and it represents almost a half of all MPN with a median 3 year survival of 88% (Rollison 2008). Median age at diagnosis is 60 years and no more than 5% of patients is under the age of 40(Berlin, 1975). A little prevalence in male gender was observed (Modan, 1995).
PV is characterized by global bone marrow hypercellularity, with a predominance of erythroid expansion which is largely independent from erythropoietin, the main erythrocytes growth factor. This is confirmed by the evidence of in vitro spontaneous formation of erythroid colonies in absence of EPO stimulation (Prchal, 1974). The identification of JAK2 mutation possibly explains the molecular basis of this phenomenon. Experimental data demonstrated the hypersensitivity of PV erythroid progenitors in response to growth factor such as EPO, GM-CSF, IL-3, IGF-1 and TPO, suggesting an underlying global defect of the intracellular signalling transduction system rather than a
14 dysfunction of specific receptors. Mutations of EPO receptor have been demonstrated in Hereditary Erythrocytosis patients that in 20% of cases are transmitted as a dominant autosomic defect that causes increased erythropoiesis but retains normal leucocyte and platelet counts with reduced EPO levels. Unlike PV progenitors, mononuclear cells from Hereditary Erythrocytosis patients do not form EEC and are sensitive to low levels of EPO (McMullin, 2008). Cases of Hereditary Erythrocytosis on autosomic recessive basis have been described, arising from loss-of-function mutations in VHL gene that determine increased levels of HIF1alpha and consequently enhanced production of EPO (Kralovics and Prchal, 2000). Neither mutations of EPO receptor nor VHL mutations were found in PV. In PV erythroid progenitors display overexpression of the anti-apoptotic protein bcl-xl which is also expressed in mature cells unlike in normal subjects (Fernandez-Luna 1999) Some hematopoietic growth factors, such as IGF-1, partially suppress apoptosis. Hypersensitivity to growth factors could thus derive from an intrinsic resistance to physiologic apoptosis mechanisms.
Main clinical features of PV are direct consequences of the increased proliferation of the different hematopoietic lineages. Erythrocytosis causes augmented blood viscosity that leads to cerebral and peripheral microvascular involvement and related signs and symptoms. Water-related pruritus is typical and affects 50% of patients. Costitutional symptoms such as weight loss, night sweats, and fever, are present al diagnosis in <30% of cases. Hepatomegaly, splenomegaly, cyanosis and arterial hypertension are common. Main complications are thrombotic events that can occur also in uncommon sites such as principal abdominal vessels. PV is characterized by increased erythrocyte mass or hemoglobin values higher than 18.5 g/dL in men and 16.4 g/dL in women for at least 2 months. Increased plasma volume can lead to underestimation or failure to identify the increased erythrocyte mass (Spivak, 2002; Spivak, 2003, Spivak and Silver, 2008). Leukocytosis and thrombocytosis occur in 50% of patients. Erythrocyte morphology is usually normal and only in the advanced phase of the disease occasional erythroblasts or myeloid progenitor cells, such as myelocytes, can be found, suggesting transformation to post-PV myelofibrosis. Leucocyte alkaline phosphatase levels are increased in 70% of patients.
The most frequent cause of death in PV patients are cardiovascular events (41% of deaths). PV patients mortality increases with age being 1.6 fold higher than general
15 population in patients under 50 years and 3.6 fold higher in patients older than 50 (Passamonti, 2004; Cervantes, 2008). PV can evolve to secondary myelofibrosis and more rarely in acute myeloid leukemia. Evolution is responsible for 13% of deaths (Marchioli, 2005). Frequency of evolution to secondary myelofibrosis is estimated at 5% after 15 years from PV diagnosis(Marchioli, 2005). ECLAP study estimated that progression to AML occurs in 1.3% of patients after an average of 8.4 years from diagnosis.
‘e e tl it has ee i t odu ed the te masked PV for JAK-2 mutated patients who display PV – characteristic bone marrow morphology, despite lower Hb levels ranging between 16 and 18,5 g/dL for men and 15 and 16,5 g/dL for women. Barbui and colleagues subsequently determined a Hb level of 16,5 g/dL in men and 16 g/dL for women or a hematocrit level of 49% in men and 48% in women to be optimal cutoff levels for distinguishing JAK-2 mutated ET from masked PV patients(Barbui, 2014). Accordingly, WHO criteria might be revised to include bone marrow morphology and lower Hb thresholds levels as components of its major criteria. The EEC growth as minor criteria is no longer used and it has been proposed to delete it as minor criterion (Barbui, 2014). Conversely, another study indicates that bone marrow histology cannot be included as a major criterion of PV and should be reserved as minor criterion in selected patients because of high specificity and low sensitivity (Alvarez-Larran, 2014).
The goal of current therapy in PV is primarily to prevent thrombohemorrhagic complications, without increasing bleeding risk, and secondly to control the microcirculatory symptoms. The current risk stratification in PV divide patients in two risk categories, low risk (age < 60 years and absence of thrombosis history) and high risk (age >60 years and/or presence of thrombosis history). The correction of cardiovascular risk factors (smoke, hypertension, diabetes, hypercolesterolemia, hypertriglyceridemia) is always the basis of all therapeutic strategies. Among all risk categories two other therapy are recommended: low-dose (81-100 mg daily) aspirin and phlebotomy to maintain hematocrit of less than 45% (Landolfi, 2004). About it, a recent randomized study demonstrated that in patients with PV, those with a Hct target less than 45% had significantly lower rate of cardiovascular death and major thrombosis than did those with a Hct target of 45 to 50% (Marchioli, 2013). In the presence of extreme thrombocytosis (>1.000×109 /L) screening for acquired von Willebrand syndrome (AvWS) is recommended before administrating aspirin, for the increased risk of bleeding, but it is not used in clinical
16 practice. About the pruritus, the responses to antihistamines have been both unpredictable and variable. Other treatments that have been reported to be useful include paroxetine (Tefferi, 2002), JAK inhibitors(Pardanani, 2011), interferon-α(Muller, 1995) and narrow band ultraviolet B phototherapy(Baldo, 2002).
In high risk patients, treatment with cytoreductive drugs, such as hydroxyurea, is recommended. However, some patients have an inadequate response to the drug or have unacceptable side effects at the doses required to consistently control the hematocrit, platelet count, white cell count, splenomegaly or symptom burden. These can be shifted to interferon-α o usulfa , if olde tha . Ne t eat e t optio fo PV sele ted patie ts, who are resistant or unable to tolerate hydroxyurea, is Ruxolitinib, which is a JAK1 and 2 inhibitor, as demonstrated by Vannucchi and colleagues in RESPONSE phase 3 clinical trial (Vannucchi, 2015).
E
SSENTIALT
HROMBOCYTHEMIAEssential Thrombocythemia (ET) is characterized by abnormal megakaryocyte proliferation that leads to increased platelets count. Diagnostic criteria for ET according to WHO 2008 classification are reported in Table 4.
Major Criteria
Platelet ou t ≥ 9/L
Megakaryocyte proliferation with large and mature morphology.
Absent or poor granulocyte or erythroid proliferation.
Not meeting WHO criteria for CML, PV, PMF, MDS or other myeloid neoplasm
Demonstration of JAK2V617F or other clonal marker or no evidence of reactive thrombocytosis
Table 4. WHO 2008 diagnostic criteria for ET (diagnosis requires all major criteria).
ET pathogenetic mechanism is still poorly known. Thrombocytosis is caused by augmented platelet production in the bone marrow where megakaryocytes display larger volume and increased nuclear lobulations and ploidy, rather than from prolonged platelet (Buss, 1994; Thiele, 2003). Clonality analysis in female patients affected from ET, who were
17 heterozygous for G6PD isoenzymes, demonstrated the involvement of a pluripotent hematopoietic stem cell (Fialkow, 1981; Fialkow, 1990). Further studies, involving restriction-fragment length polymorphism analysis of X chromosome, evidenced that also granulocytes, platelets and occasionally B-lymphocytes are clonal (Chen, 2007). Nevertheless a subgroup of non-clonal lymphocytes has been demonstrated, indicating the presence of a residual normal population that is overwhelmed by the neoplastic clone during disease progression. These results suggest that neoplastic transformation can occur at different levels of the hematopoietic hierarchy.
TPO gene mutations have been described in cases of Hereditary Thrombocytosis. (Cazzola, 2000; Skoda, 2005: Liu, 2009; Skoda, 2009). Thrombopoietin is the main cytokine involved in megakaryocytes growth and differentiation and platelet production. Four different mutations have been reported, with autosomic dominant transmission and complete penetrance. The clinical phenotype of the disease is characterized by increased platelet count with normal hematocrit and high serum TPO levels and it manifests since the very early childhood. All the mutations induce hyperproduction of TPO by reducing or abolishing the inhibitory action of regulatory regions. It has been hypothesized that such mutations might affect also some ET patients that display high serum TPO levels.
Studies on the TPO receptor MPL showed its reduced expression in platelets derived from ET patients thus probably explaining TPO high levels (Li, 2000). Reduced expression of MPL is associated with altered glycosylation that leads to defective activation of the downstream kinase cascade and subsequent altered phosphorylation of JAK2 and STAT5 (Moliterno, 1998). JAK/STAT pathway activation may occur after activating mutation of cytokines receptor, such as MPL. Most studies investigated the transmembrane domain which is critical for dimerization and receptor activation and the juxta-membrane domain which is needed for JAK2 binding. These studies lead to the identification of W515L and W515K point mutations that cause constitutive activation of the downstream signalling pathway in the absence of TPO (Pardanani, 2006; Pikman, 2006). MPL mutations were first estimated to occur in 1% of ET patients (Pardanani, 2006) but more recent studies, including from our group, demonstrated that the percentage of mutated cases is up to 5-8% (Beer, 2008; Vannucchi, 2008). Very recently, in December 2013, two independent groups (Green/Campbell and Kralovics) discovered a new gene involved in the pathogenesis of MPN named Calreticulin (CALR). Mutations of CALR exon 9 were reported
18 in 20-35% of patients with ET and PMF who were wild type (wt) for JAK2 (Nangalia, 2013; Klampfl 2013).
The role of megakaryocytes (MK) in ET has been deeply studied. In vitro analysis on MK colonies growth demonstrated that in some patients it may be cytokine-independent (Grossi, 1987). Furthermore MK precursors are less responsive to the treatment with growth inhibitors such as TGF-β, that is one of the most powerful MK growth inhibitors (Zauli, 1993). TGF-β is produced by MKs themselves, thus its high levels could be a consequence of increased MK and platelets number. It has been hypothesized that in the bone marrow microenvironment TGF-β stimulates local production of TPO that in turn induces the expression of TGF-β receptor thus enhancing growth inhibition. Some studies suggest that spontaneous formation of MK colonies may be due to lack of platelet factor 4. It was demonstrated that this factor inhibits MK colonies growth in ET patients (Han, 1990). Both TGF-β and platelet factor 4 are stored in α granules that have known inhibition activity on MK colonies growth (Grossi, 1986).
It has been demonstrated that erythroid progenitors of ET patients can originate BFU-E colonies when cultured in absence of BFU-EPO as PV patients progenitors do (BFU-Eridani, 1987). ET clinical phenotype is characterized by thromboembolic events affecting arteries and veins of large and medium calibre as well as peripheral microvessels (Vannucchi, 2007; Vannucchi, 2010). Microvascular involvement leads to acrocyanosis, paresthesias and erythromelalgia while involvement of the cerebral district causes neurological symptoms such as headache, tinnitus, dizziness, frequent TIA and only occasionally epileptic and seizure attacks (Cortelazzo, 1990). Haemorrhagic events mainly affect gastrointestinal tract but can also involve skin, conjunctivae, urinary tract and rhino-pharyngeal mucosa. Systemic symptoms such as weight loss, sweating, fever and itching affect 20-30% of patients. About a half of them displays splenomegaly and 20% has hepatomegaly.
Laboratory data suggesting ET diagnosis are platelet count higher than 600.000/µL for 6 months and usually moderate leukocytosis. Blood smear analysis may reveal altered platelet morphology and presence of megathrombocytes. Functional abnormalities have been described in ET platelets such as defective metabolism of arachidonic acid, decreased expression of prostaglandin D2 receptor and adrenalin receptor (Kaywin, 1978). Platelets derived from ET patients with previous history of thrombotic events display increased
19 production of B2 thromboxane and improved affinity for fibrinogen (Landolfi, 1988). The i eased le els of β-thromboglobulin and serum thromboxane suggest an enhanced platelets activation mechanism (Cortelazzo, 1981). Bone marrow biopsy reveals hypercellularity associated with marked expansion of megakaryocytes that are pleomorphic and clustered. MKs are mature and do not show dysplasia as in pre-fibrotic myelofibrosis that must be considered in the differential diagnosis with ET. In the last years many studies have been performed separating true ET from early PMF by means of bone marrow morphology; this separation has a prognostic impact on patients. In fact true ET is characterized by lower white blood cell counts, lower Hb and LDH levels and, importantly, a better prognosis, which is close to normal (Barbui, 2011). About early-PMF and true-ET in a recent study Barbui and colleagues have shown substantial differences after 15 years of follow up with regard to survival rates (59% vs 80%) and leukemic transformation rates (11.7% vs 2.1%) respectively (Barbui, 2012). About morphological features megakaryocytes in ET are large and mature appearing, whereas those in prefibrotic PMF display abnormal maturation with hyperchromatic and irregularly folded nuclei.
According to WHO description, in early prefibrotic myelofibrosis there is hypercellularity by a prominent neutrophil granulocytic and megakaryocytic proliferation often associated with a concomitant reduction of nucleated red cell precursors in the absence or minor reticulin MF, consistent with MF-0 and MF-1.
In ET disease course is chronic and survival is similar to the healthy population adjusted for age and sex. Progression to acute leukemia is rare and usually associated with the onset of cytogenetic abnormalities. Progression to secondary myelofibrosis is also possible. About the risk stratification, categories are the same of PV patients. Current ET treatment is based on antiplatelet drugs for patients with vascular involvement (Beer, 2009). Hydroxyurea is the most used myelosuppressive treatment since it quickly and steadily reduces platelet count thus preventing thrombotic events (Cortelazzo, 1996). Other chemotherapeutic agents such as busulfan, melphalan, chlorambucil and thiotepa have been used in ET although with lower efficacy and higher risk of blastic evolution (Berk, 1981). Anagrelide treatment inhibits MK maturation and platelet release (Fruchtman, 2005) but it has been demonstrated that hydroxyurea more efficiently reduces platelet
20 count and risk of thrombotic events or progression to myelofibrosis compared to anagrelide (Harrison, 2005).
M
YELOFIBROSISMyelofibrosis has an incidence of 0.5-1.5 new cases over 100.000/year. Median age at diagnosis is 60 and both sex are equally affected (Cervantes, 2008). Myelofibrosis can be primary (PMF) or secondary occurring as the natural evolution of pre-existent PV or ET (PPV-MF and PET-MF respectively) in about 15% of cases, usually after the first decade or so (Mesa, 2010). Clinical features of the disease are splenomegaly, leukoerythroblastosis, erythrocyte anisopoichilocytosis, variable degree of bone marrow fibrosis and extramedullary erythropoiesis mainly occurring in spleen and liver (Barosi, 2003; Tefferi, 2003; Tefferi, 2006). 20% of patients are asymptomatic at diagnosis. Laboratory data can reveal anemia, thrombocytosis or thrombocytopenia and variable leukocytosis or leukopenia. Most cases are diagnosed in more advanced phase when hematological parameters are clearly altered with marked anemia, leukocytosis and usually thrombocytopenia. Medullar fibrosis worsens and hepatosplenomegaly due to extramedullary hematopoiesis increases. Blasts may be found in peripheral blood implying evolution to acute myeloid leukemia. Myelofibrosis is characterized by abnormally increased number of mature cells but also by differentiation abnormalities mainly involving megakaryocyte lineage. MK clone is thought to release fibrogenic factors that promote fibers deposition in the bone marrow. Thus myelofibrosis may originate as a consequence of the augmented MK population whose defective maturation may lead to local release of g o th fa to sto ed i the α g a ules pa ti ula l TGF-beta) (Chagraoui, 2002; Vannucchi, 2005).
Myelofibrosis originates when a multipotent hematopoietic stem cell is damaged. Monoclonality of erythrocytes, granulocytes, MK, monocytes and B-lymphocytes was demonstrated studying G6PDH expression, inactivation pattern of X-chromosome, polymorphisms and RAS family mutations (Jacobson, 1978; Buschle, 1988; Reilly, 1994). T-lymphocytes have been shown to be monoclonal in a subgroup of patients suggesting that abnormal proliferation originates from a partially committed progenitor (Tsukamoto, 1994). On the contrary, bone marrow fibroblasts are policlonal therefore not being part of
21 the malignant clone (Wang,1992; Castro-Malaspina,1982; Greenberg,1987). Bone marrow of myelofibrosis patients shows increased amount of stromal cells, extracellular matrix proteins, angiogenesis and osteosclerosis besides the alrered expression of many cytokines (Reilly,1997; Mesa,1999; Martyre,1997).
Bone marrow fibrosis is caused by aumented production of extracellular matrix proteins that organize in fibers. This stack of fibers leads to loss of hematopoietic function of the bone marrow. Extramedullary hematopoiesis expands predominantly in the spleen and li e ut it does t fu tio ally replace medullar hematopoiesis. Splenectomy is indeed used as a therapeutic option to improve blood cells count by removing a collection district. Fibrosis is a secondary event caused by growth factor released by the malignant clone. This hypothesis is supported by the evidence of fibrosis regression after bone marrow transplantation or prolonged treatment with chemotherapy or interferon (McCarthy, 1985; Manoharan, 1984). Cytokines mostly involved in fibrosis development are platelet derived growth factor (PDGF), fibroblasts growth factor (b-FGF), platelet factor 4, transforming growth factor beta (TGF-β , β-thromboglobulin, calmodulin, interleukin-1 and vessel endothelium growth factor (VEGF). These cytokines are produced by MKs and monocytes. The role of megakaryocytes in the development of bone marrow fibrosis is supported by several observation: 1- in bone marrow biopsies of MF patients MK hyperplasia, dysplasia and necrosis are reported; 2- bone marrow fibrosis occurs in megakaryocytic acute leukemia; 3- fibrosis is also reported in cases of gray platelet syndrome which is an he edita diso de affe ti g platelet α g a ules (Martyre, 1994; Reilly, 1994); 4- animal models of myelofibrosis always show increase MK proliferation. MKs and platelets display e ha ed p odu tio a d a o alous elease of α g a ules o tai i g PDGF, platelet factor 4, FGF, TGF-beta and calmodulin. Cytoplasmic fragmentation of MKs determines elease of these toki es i the edulla i oe i o e t ithout affe ti g α g a ules concentration in circulating platelets (Villeval, 1997; Taskin, 1998; Vannucchi, 2005). More recently other factors have been proved to play a key role in this complex pattern of disease. In a recent paper has been demonstrated an association between PMF and a non-canonical MAPK activation of TGF- β sig ali g i di ati g a p o a le autoi u e mechanism for bone marrow fibrosis (Ciaffoni, 2015).
In another study it has been highlighted the over-expression of lysyl-oxidase (LOX) gene family members in the fibrogenesis (Tadmor, 2013). LOX oxidizes the PDGF receptor on
22 smooth muscle cells, fibroblasts and megakaryocytes and enhances cell proliferation signals (Lucero, 2008). Production of higher levels of LOX subsequently stabilize the extracellular matrix, and contribute to enhanced fibrosis; in addition PDGF and TGF-β also increase LOX expression (Hong, 1999).
I the follo i g pi tu e the e is a su a of Bad seeds i ad soil o ept (Le Bousse-Kerdilès, 2012), about PMF and connection between HSCs and stromal cells in medullar micro-environment.
Figure 1. The "Bad Seeds in Bad Soil" model in PMF.
Circulating hematopoietic progenitor cells can be observed in MPNs, mainly in myelofibrosis. These cells can be identified by the expression of CD34 surface marker that is not present on differentiated cells. CD34+ cells can leave bone marrow at different maturation stages as suggested by their variable expression (23 to 99%) of CD38 transmembrane molecule which is poor or absent in hematopoietic cells with high self-renewal and differentiation potential. Circulating CD34 positive cells count is meaningful for MF diagnosis since it is 360 fold higher compared to that of healthy population and 18 to 30 fold higher in respect with PV and ET count. Threshold value of 15x106/L discriminates untreated myelofibrosis from other MPN with positive predictive value of 98,4% and negative predictive value of 85%. CD34+ cells count increase during disease progression and correlates with circulating myeloid blasts count and with splenomegaly . CD34+ count
23 decreases after cytoreductive therapy suggesting that it correlates with the extent of tumor mass. CD34+ value is associated with leukemic transformation as patients with more than 300x106/L have 50% probability to develop leukemia in the next 11 months (Barosi, 2001). Myelofibrosis patients also display higher levels of pluripotent progenitors, such as megakaryocytes colony forming unit (CFU-MK), granulocytic/monocytic-CFU (CFU-GM),
erythroid burst forming unit (BFU-E) and
granulocytic/erythroid/monocytic/megakaryocytic-CFU (CFU-GEMM) when compared to healthy donors (Partanen, 1982; Wang,1983; Douer,1983; Hibbin,1984).
Molecular mechanisms responsible for clonal proliferation in myelofibrosis are still poorly understood. Cytogenetical analysis revealed that the most common defects, representing 65% of all cytogenetic abnormalities in Myelofibrosis, are deletions of chromosome 13q which correlate with early leukemic transformation (Mesa, 2009), deletion of chr 20q and partial trisomy of 1q . Chromosome 8 trisomy and chr 12p deletion seem to represent unfavourable prognostic factors (Tefferi, 2001). Many tumor-suppressor genes might be involved, such as RB-1 (retinoblastoma 1) located on chr 13q14 whose loss of heterozygosity has been demonstrated in 25% of PMF patients (Juneau, 1998). Occurrence of chromosomic alterations has been shown to own ominous prognosis in many studies (Demory, 1988; Dupriez, 1996), and to reduce therapy response rate (Besa, 1982). Cytogenetic abnormalities can occur during disease progression and possibly contribute to leukemic transformation. Such role has been hypotized for p53 and RAS family genes lesions (Gaidano, 1994; Wang, 1998).
The proliferative advantage is demonstrated by cytokine-independent in vitro growth of mutated cells and hypersensitivity to cytokines (Taskin, 1998). For stem cell factor receptor (c-KIT) over-expression and a point mutation were both reported. The mutation occurs in the domain that stabilizes the binding to SCF and could explain the increased tyrosine-kinase activity downstream of c-KIT. It has been hypothesized that myelofibrosis is caused by alteration of the signalling pathway involving the fibroblasts growth factor (b-FGF). This cytokine is produced by several cell types including hematopoietic and stromal cells and has mitogenic activity on bone marrow stromal cells. FGF is also a strong angiogenic factor and promotes stem cell growth acting through a tyrosine-kinase associated receptor. In CD34+ cells of myelofibrosis patients both b-FGF and its receptor are overexpressed in contrast with the reduced expression of TGF-beta (LeBousse-Kerdilès,
24 1996). Recent studies suggested the involvement of JAK2/STAT signalling pathway since the discovery of the point mutation JAK2 V617F that causes constitutive activation of the downstream pathway. Several experiments demonstrated that V617F leads to in vitro spontaneous growth of erythroid colonies (BFU-E) in PV patients. This mutation can indeed represent an important molecular marker for myelofibrosis although its higher frequency in PV suggests that other mechanisms may play an important role in the pathogenesis of myelofibrosis (Baxter, 2005; Ugo, 2005; Kralovics, 2005; Levine, 2005). Further studies identified mutations in the TPO receptor MPL. These mutations are found in 5% of myelofibrosis patients that do not display JAK2 V617F mutation and lead to constitutive activation of TPO signalling (Pardanani, 2006; Pikman, 2006).
Several studies have indicated a role for mutant JAK2 in driving genetic instability, with increases in homologous recombination, point mutations and small deletions observed in cell lines expressing mutant compared to wildtype JAK2. In fact, other factors have been demonstrated to be implicated in the development of myelofibrosis including reduced GATA-1 expression, increased thrombopoietin and transforming growth factor β signalling and altered expression of molecules involved in stem cell trafficking such as CXCR4. The mutational landscape of PMF is complex, about 50-60% of patients carry JAK2 mutations (Levine, 2005; Kralovics, 2005), MPL mutations are present in 5-8% of patients and more recently mutations in the gene encoding the endoplasmic reticulum protein calreticulin have been reported in about 20% of PMF patients (Nangalia, 2013; Klampf, 2013) accounting for 70–80 % of those lacking the JAK2V617F allele. The above mutations are defi ed d i e as the o fe the ells the a ility of proliferate independently from any stimuli; further their in vivo expression induces MPN (Levine, 2005; Li, 2011). Despite these evidences a large number of MPN are still molecularly not characterized. Recently high-throughput genome analysis have identified several genes encoding for proteins involved in the epigenetic regulation of transcription: TET oncogene family member 2 (TET2), Additional Sex Combs-Like 1 (ASXL1), Isocitrate dehydrogenase (IDH1/IDH2), Enhancer of zeste homolog 2 (EZH2) (Tefferi, 2010), DNA Methyl transferase 3A (DNMT3A) (Stegelmann, 2011), other components of PCR2 (SUZ12, EED and others)(Score, 2011) which have been reported to be mutated in a small subset of patients with PMF (Figure 2) and can occur with or without the JAK mutation, suggesting a common strike contribution in MPN (Vainchenker, 2011).
25 Figure 2. Frequencies of mutations in commonly affected genes in patients during chronic myeloproliferative phase (a–c) and acute leukemic phase (d) of the disease (Milosevic and Robert Kralovics, 2012).
However, the specific role of these new gene alterations as possible new diagnostic molecular markers and/or as new prognostic factors is still inconclusive. In fact, considering that most of these mutations are shared by other myeloid malignances, in particular myelodysplastic syndromes and chronic myelomonocytic leukemia, other myeloid neoplasia, and acute leukemia, they have no specific diagnostic value, while, on the other hand, they contribute remarkably to the prognosis of patients with PMF.
About animal models, transgenic and knock-in mice expressing mutant JAK2 have provided compelling evidence that mutated JAK2 (typically JAK2-V617F) is a driver in this major subset of myeloproliferative neoplasms, however these mice are poor models for PMF. PMF characteristics such as megakaryocyte proliferation and fibrosis have been
26 recapitulated in mice expressing thrombopoietin (Villeval,1997) , the NF-E2 transcription factor (Kaufmann, 2012), vascular endothelial growth factor (Maes, 2010)or reduced levels of GATA1 (Vannucchi, 2002) suggesting that abnormal erythroid/megakaryocyte development and/or abnormal release of cytokines may be a key factor in the disease. The Table 5 below reports the current 2008 WHO criteria for PMF diagnosis and some updates at the light of the most recent papers.
WHO diag osti ite ia fo P i a M elofi osis
Majo ite ia Megaka o te p olife atio a d at pia a o pa ied eithe
eti uli a d/o ollage fi osis or I the a se e of eti uli fi osis, the egaka o te ha ges ust e a o pa ied i eased a o ellula it , g a ulo ti p olife atio a d ofte de eased e th opoiesis i.e. p e-fi oti PMF .
Not eeti g WHO ite ia fo CML, PV, MD“, o othe eloid eoplas
De o st atio of JAK V F or othe lo al a ke or o e ide e of ea ti e BM fi osis
Mi o ite ia Leukoe th o lastosis I eased se u LDH le el A e ia
Palpa le sple o egal
PMF diag osis e ui es eeti g all ajo ite ia a d t o i o ite ia.
Table . WHO diag osti ite ia fo PMF.
The diagnosis of post-PV and post-ET MF should adhere to criteria published in 2008 by the International Working Group for MPN Research and treatment (IWG-MRT) (Barosi, 2008) (Table 6).
27 C ite ia fo post - PV MF
‘e ui ed ite ia
Do u e tatio of p e ious diag osis of PV as defi ed WHO ite ia. Bo e Ma o fi osis g ade - o g ade - .
Additio al ite ia t o a e e ui ed
A e ia o sustai ed loss of e ui e e t fo phle oto i the a se e of to edu ti e the ap
A leukoe th o lasti pe iphe al lood pi tu e
I easi g sple o egal defi ed as eithe a i ease i palpa le sple o egal of ≥ dista e of the tip of splee f o the left ostal
a gi o the appea a e of a e l palpa le sple o egal .
De elop e t of ≥ of th ee o stitutio al s pto s: > % eight loss i o ths, ight s eats, u e plai ed fe e > . °C .
C ite ia fo post - ET MF
‘e ui ed ite ia
Do u e tatio of p e ious diag osis of ET as defi ed the WHO ite ia Bo e Ma o fi osis g ade - o g ade - .
Additio al ite ia t o a e e ui ed
A e ia a d a ≥ g/dL de ease f o aseli e H le el A leukoe th o lasti pe iphe al lood pi tu e.
I easi g sple o egal defi ed as eithe a i ease i palpa le sple o egal of ≥ dista e of the tip of splee f o the left ostal
a gi o the appea a e of a e l palpa le sple o egal . I eased LDH le el
De elop e t of ≥ of th ee o stitutio al s pto s: > % eight loss i o ths, ight s eats, u e plai ed fe e > . °C .
Table . Diag osti ite ia fo post-PV a d post-ET elofi osis.
MF should be distinguished from other closely related myeloid neoplasms including chronic myeloid leukemia (CML), PV, ET, MDS, chronic myelomonocytic leukemia (CMML) and also hairy cell leukemia. Other rare causes of splenomegaly and marrow fibrosis should be considered.
About the risk stratification, robust prognostic models in PMF started in 2009 with the development of the International Prognostic Scoring System (IPSS) (Cervantes, 2009). It is applicable to patients being evaluated at the time of diagnosis and uses five independent predictors of inferior survival: age > 65 years, Hb < 10 g/dL, leukocyte count > 25*109/L, circulating blasts ≥ 1% and the presence of constitutional symptoms. The presence of 0, 1, 2 and ≥3 adverse factors defines low, intermediate 1, intermediate 2 and high risk disease. The IWG-MRT subsequently developed a dynamic prognostic model (DIPSS) that uses the same prognostic variables used in IPSS but can be applied at any time during the disease course. DIPSS assigns two, instead of one adverse points for Hb < 10 g/dL and risk
28 categorization is accordingly modified: low (0 adverse points), intermediate 1 (1 or 2 points), intermediate 2 (3 or 4 points) and high (5 or 6 points) (Passamonti, 2010).
In the following table 7 the scores and the IPSS corresponding median survival.
Pa a ete IP““ poi t DIP““ poi t
-Age > ea s
-p ese e of o stitutio al s pto s -He oglo i < g/dL
-WBC ou t > * /L -Blood lasts ≥ %
‘isk g oup IP““ s o e Media su i al
ea s -Lo -I te ediate -I te ediate -High ≥ . . . Table . IP““ a d DIP““ plus s o es.
Figure . IP““ st atifi atio i PMF.
Other risk factors for survival in MF were subsequently identified and included unfavourable karyotype (include +8, -7/7q-, i(17q), inv(3), -5/5q-, 12p-, or 11q23 rearrangement), red cell transfusion need and platelet count < 100*109/L. So DIPSS was modified into DIPSS-plus by incorporating these three additional independent risk factors. The four DIPSS-plus score risk categories based on aforementioned eight risk factors are low (no risk), intermediate-1 (one risk factor), intermediate 2 (2-3 risk factors) and high
29 (four or more risk factors) with respective median survivals of 15.4, 6.5, 2.9 and 1.3 years (Gangat, 2011). Since the publication of DIPSS-plus score, several studies that suggest additional prognostic information have been published. For example Tefferi and colleagues demonstrated that a > 80% 2-year mortality in PMF patients was predicted by monosomal karyotype, inv(3)/i(17q) abnormalities, or any two of circulating blasts >9%, leukocytes ≥ 40*109/L or other unfavourable karyotype (Tefferi, 2011)(Figure 4).
Figure . ‘isk-st atified su i al data fo Ma o Cli i patie ts ith p i a elofi osis.
I fe io su i al i PMF has ee also asso iated ith ulliz gosit fo JAK / haplot pe Teffe i, , lo JAK V F allele u de Gugliel elli, , o the p ese e of IDH Teffe i, , EZH Gugliel elli, , “R“F Lasho, utatio s. Co e sel the p ese e o a se e of MPL Pa da a i, , JAK V F Gugliel elli, a d TET
Teffe i, utatio s did ot appea to affe t su i al.
A out the the ap , the e is o u ati e t eat e t fo PMF a d PPV-PET/MF othe tha allo – ste ells t a spla tatio “CT . Co se ue tl it is esse tiall palliati e a d guided the p edo i a t s pto s su h as a e ia a d sple o egal . The p ese e a d se e it of a e ia i PMF sig if a lo all ad a ed a d iologi all o e agg essi e disease. The eed of esol i g a e ia is fu da e tal to i p o e the ualit of life i these patie ts a d
30 a oid i o o e load elated to ed ell t a sfusio s. A H < g/dL usuall t igge s o side atio of t eat e t, ut the e a e i di idual a iatio s depe di g upo age a d o o idities.
A out sple o egal , h d o u ea as the fi st-li e the ap ith a o e all espo se of % Ma ti ez-T illos, ut afte o e ea of t eat e t app o i atel % of patie ts e ui e a alte ati e the ap su h as JAK i hi ito s. “ple e to is i di ated i patie ts ith la ge a d pai ful sple o egal he e JAK-i hi ito s a e ot a aila le o p o e i effe ti e Teffe i, . “ple i i adiatio a also e applied i patie ts ho do ot tole ate JAK i hi ito s a d a e poo a didates fo su ge Boua dallah, .
The JAK-i hi ito s a t ai l i hi iti g d s egulated JAK-“TAT sig alli g, p ese t i all MF patie ts. The a e ot sele ti e of the utated JAK so the a e i di ated i oth JAK utated a d u utated MF patie ts. ‘u oliti i is the fi st d ug app o ed fo MF t eat e t Ve sto sek, . T o phase III studies o pa ed u oliti i ith pla e o
COMFO‘T I Ve sto sek, o est a aila le the ap COMFO‘T II Ha iso , a d oth attai ed the p i a e d poi t of ≥ % edu tio i splee olu e i agi g te h i ues at a d eeks of t eat e t, espe ti el . It has ee sho also a su i al ad a tage fo patie ts t eated ith u oliti i Passa o ti, .
B fa the o l u ati e t eat e t fo MF is ste ell t a spla tatio allo“CT . It a o du e to esolutio of BM fi osis, ole ula e issio a d esto atio of o al he atopoiesis Balle , . A o di g to ELN e o e datio s, it is justified to offe allo“CT to eligi le patie ts ith MF hose edia su i al is e pe ted to e < ea s. This i ludes patie ts ith i te ediate- a d high isk a o di g to IP““ p og osti s o e. P e-t a spla t JAK-i hi ito t eat e t a edu e splee size a d i p o e o stitutio al s pto s, ut is u e tl ei g tested i li i al studies a d should e ega ded as e pe i e tal.
31
THE MOLECULAR PATHOGENESIS OF MPN
Myeloproliferative neoplasms are characterized by a high molecular complexity. Although JAK2 mutations have been shown to be the phenotypic drivers in MPN, there is evidence of clonality and mutational events preceding the acquisition of JAK2V617F (K alo i s, ; “ hau , ; Ortmann, 2015). An increasing number of mutations in genes distinct from JAK2 have been identified in patients with MPN.
Clonality is established by the finding of an acquired mutation or cytogenetic abnormality, although the diagnosis of each specific subtype depends on additional clinical, laboratory and morphological information. Some genetic abnormalities are specific for MPNs, some are also seen in other myeloid malignancies. The same mutations can also occur in all subtypes of MPNs, which are clearly three diseases with considerably different phenotypes but with some similarities; additional genetic and epigenetic factors are involved in the characterization of clinical pattern.
Mutations in MPNs can be divided in phe ot pi d i e a d su lo al utatio s. JAK2, MPL and CALR mutations are considered phenotypic driver since the expression of the mutated gene in cell lines caused cytokine independent or hypersensitive growth, as known to occur in primary cells from MPN patients, and in animal models phenotypes closely resembling a myeloproliferative disease were observed in transgenic or conditional animals (Li, 2011). The subclonal mutations usually occur in hematopoietic cell subclones of variable size, often but not invariably together with one of the phenotypic driver mutations, and may either antedate or follow the acquisition of phenotypic driver mutations (Vainchenker, 2011). The order by which these mutations appear in the mutational phylogenesis of MPN contribute to the disease phenotype, while the prognostic relevance if any is far from being appreciated (Ortmann, 2015; Nangalia, 2015). Since these mutations are commonly represented also in myelodysplastic syndromes, other myeloid neoplasia and acute leukemias, they have no specific diagnostic value, while, on the other hand, they contribute remarkably to the prognosis of patients with PMF.
There are four different classes of somatic acquired mutations in the MPNs, with an increased number of involved genes. Some of these mutations are gain of function (GOF), others are loss of function (LOS). These classes involves signaling pathway, epigenetic mechanisms, mRNA splicing, transcription factors and leukemic transformation.
32 In the following Table I reported the most relevant involved gene from the recent literature and below a picture of the major related molecular pathways in MPNs.
33
Figure . A ui ed a o alities i MPN.
M
UTATIONS AFFECTING THE PHENOTYPIC DRIVER GENESJAK2 is a member of the Janus kinase family composed by four tyrosin-kinases (JAK1, 2,
and 3 and TYK2) that attach to cytokine receptor cytosolic domains. JAK kinases possess two highly homologous domains at the carboxyl terminus: an active kinase domain (JAK ho olog , JH a d a atal ti all i a ti e pseudoki ase do ai JH hi h egati e regulates the JH1 kinase activity. At the N-terminus, the JH5-JH7 domains contain a FERM (Band-4.1, ezrin, radixin, and moesin)–like motif, which plays a role in the binding to the cytosolic domain of cognate cytokine receptors. JAK2 plays a central role in the signaling f o eloid toki e e epto s. I fa t, it i ds to the th ee ho odi e i eloid receptors (erythropoietin receptor [EPO-R], myeloproliferative leukemia [MPL; TPO-R], G-CSF receptor [G-G-CSF-R]), to the prolactin and growth hormone receptors, to heterodimeric receptors (GM-CSF-R, IL-3-R, and IL-5-R, which share the common β chain of IL-3-R and the gp130 family of receptors), and to IFN-γ R2. JAK2 is the only JAK capable of mediating the signaling of EPO-R and MPL. JAK2 also functions as a chaperone for trafficking of these 2 receptors to the cell surface and their stability. More recently, JAK2 was also shown to promote G-CSF-R cell-surface localization. Therefore, JAK2 a d the eloid e epto s34 form functional units and have been shown to be required for the promotion of JAK2V617F signaling. (Lu, 2008) (Figure 6).
Figure 6. Cytokine receptors and JAK2.
The JAK2V617F point mutation at base pair 1849 (GT), cause substitution of the normal valine to phenylalanine in codon 617 (V617F) of exon 14 of the gene. This valine is located at one of the predicted interfaces between JH1 and JH2 domains, and the change to a phenylalanine appears to relieve the inhibition of the JH2 domain on the JH1 kinase domain. JAK2V F utatio as defi ed as gai - of- fu tio e ause the expression of mutated allele in cytokine-dependent cell lines conferred cytokine independence and cytokine hypersensitivity through the constitutive activation of STAT5, Akt and ERK-dependent pathway (Ugo, 2005; Levine, 2005; Baxter, 2005; Kralovics, 2005). The JAK2 gene is located on chromosome 9p24. Following the observation that acquired uniparental disomy (UPD) of chromosome 9p is present in 30% of patients with PV, the JAK2V617F mutation was discovered as the prominent genetic aberration in patients with BCR-ABL– negative MPNs (∼ 95% of patients with PV, 50%-70% with ET, and 40%-50% with PMF), as well as in some cases of atypical MPN (30%-50% splanchnic vein thrombosis and
35 sideroblastic anemia associated with a thrombocytosis) (Milosevic, 2013; Kralovics, 2002). The high prevalence of the V617F mutation in three clinically distinct MPNs begs the question of what additional factors contribute to phenotypic diversity between PV, ET, and PMF. First, mutant allele burden of V617F may modulate phenotype: the mutation can be found on one or both alleles (homozygosity) due to a mitotic recombination process that occurs in most patients with PV or PMF and a minority only of ET. Recently, in-vivo studies have corroborated these findings: murine retroviral transplant models resulting in high levels of V617F expression produced a PV-like phenotype with marked erythrocytosis. Conversely, transgenic models with more physiologic levels of V617F expression resulted in phenotypes resembling ET and PMF. For ET patients, positivity for V617F tends to confer a PV-like phenotype, with higher hemoglobin and lower platelet counts than in V617F-negative ET patients. Differences in intracellular signaling arising from V617F may also explain the development of PV compared with ET: preferential activation of STAT1 constrains erythroid differentiation and promotes megakaryocytic development, leading to an ET phenotype (Xing, 2008; Vannucchi, 2008)). In contrast, reduced STAT1 phosphorylation promotes erythroid development as observed in PV. Host genetic a kg ou d a also i flue e disease presentation. In retroviral transplant models, disparate phenotypes were observed depending on the mouse strain. In C57Bl/6 mice, transplantation with JAK2 V617F transduced cells resulted in a PV-like disease predominantly characterized by erythrocytosis (Wernig, 2006). However, in Balb/C mice, similar experiments yielded mice with erythrocytosis, but also leukocytosis and the su se ue t de elop e t of elofi osis (Bumm, 2006). However, most recently several lines of evidence support that the most compelling basis for MPN diversity comes from the additional molecular abnormalities that either precede or follow the acquisition of V617F. The aggregate data suggest that there is no strict temporal order of mutation occurrence that defi es the de elop e t o atu al histo of spe ifi MPNs a d V F a a ise o a pre-existing abnormal clonal substrate: 1) in some patients, the V617F burden is relatively small compared with the proportion of cells with a coexistent clonal karyotypic abnormality; and 2) in AML arising from a V617F-positive MPN, the mutant V617F allele can frequently no longer be detected (Theocharides, 2007). This suggests that the MPN and AML share a clonal origin that likely preceded the acquisition of V617F. Furthermore, in V617F-positive PV and ET patients, both JAK2 wild-type and V617F-positive EECs have been
