Available online 21 August 2006

Abstract

the normal r iagnosis ofPV. Th R testfor the sitive.The use

http://france.elsevier.com/direct/PATBIO/

Pathologie Biologie 55 (2007) 92104

LectNovemb

* CorrE-ma

0369-81doi:10.1ere is a significant overlap of serum EPO levels in PV versus control and controls versus SE. The specificity of a JAK2 V617F PCdiagnosis of MPD is high (near 100%), but only half of ET and MF (50%) and the majority of PV (up to 97%) are JAK2 V617F po

of

ure Iner 18esponil ad

14/$016/jEEC assays were performed. The diagnostic impact of low serum EPO levels (ELISA assay) in a large study of 186 patange (< 3.3 IU/l) had a sensitivity specificity and positive predictive value of 87%, 97% and 97.8%, respectively, for the dThe clinical criteria for the diagnosis of essential thrombocythemia (ET) according to the polycythemia vera study group (PVSG) do notdistinguish between ET and thrombocythemia associated with early stage PV and prefibrotic chronic idiopathic myelofibrosis (CIMF). Theclinical criteria of the PVSG for the diagnosis of polycythemia vera (PV) only detects advanced stage of PV with increased red cell mass. Thebone marrow criteria of the World Health Organization (WHO) are defined by pathologists to explicitly define the pathological criteria for thediagnostic differentiation of ET, PV, and prefibrotic and fibrotic CIMF. As the clinical PVSG and the pathological WHO criteria show significantshortcomings, an updated set of European Clinical and Pathological (ECP) criteria combined with currently available biological and molecularmarkers are proposed to much better distinct true ET from early PV mimicking ET, to distinguish ET from thrombocythemia associated withprefibrotic CIMF, and to define the various clinical and pathological stages of PV and CIMF that has important therapeutic and prognosticimplications. Comparing the finding of clustered giant abnormal megakaryocytes in a representative bone marrow as a diagnostic clue toMPD, the sensitivity for the diagnosis of MPD associated with splanchnic vein thrombosis was 63% for increased red cell mass, 52% for lowserum EPO level, 72% for EEC, and 74% for splenomegaly indicating the superiority of bone marrow histopathology to detect masked early andovert MPD in this setting. The majority of PV and about half of the ET patients have spontaneous EEC, low serum EPO levels and PRV-1 over-expression and are JAK2 V617F positive. The positive predictive value for the diagnosis of PV of spontaneous growth of endogenous erythroidcolonies (EEC) of peripheral blood (PB) and bone marrow (BM) cells is about 8085% when either PB or BM EEC assays, and up to 94% whenBM and PB ients belowbiological markers

ternational Symposiu, 2005, Hopital Avicding author.dress: postbus@good

- see front matter .patbio.2006.06.002Current diagnostic criteria for the chronic myeloproliferativedisorders (MPD) essential thrombocythemia (ET),

polycythemia vera (PV) and chronic idiopathic myelofibrosis (CIMF)

Critres diagnostiques actuels des syndromes myloprolifratifs (SMP),thrombocytmie essentielle (TE), polyglobulie de Vaquez (PV)

et mylofibrose idiopathique (MFI)

J.J. Michielsa,d,*, Z. Bernemaa, D. Van Bockstaeleb, H. De Raevec, W. Schroyensa

aDepartment of Hematology, University Hospital Antwerp Wilrijkstraat 10, 2650 Edegem/Antwerp, BelgiumbDepartment of Molecular Genetics, University Hospital Antwerp, Antwerp, Belgium

cDepartment of Pathology University Hospital Antwerp, Antwerp, BelgiumdGoodheart Institute, Hematology, Hemostasis and Thrombosis Research Center,

Rotterdam, MPD Center Europe, Erasmus Tower, Veenmos 13, 3069 AT Rotterdam, The Netherlandsincluding JAK2 V617 PCR test, serum EPO, PRV-1, EEC, leukocyte alkaline phosphatase score and peripheral

m on The V616F JAK2 mutation: A major step forward in the pathogenesis and management of myeloproliferative disorders,enne and Paris 13 University.

heartcenter.demon.nl (J.J. Michiels).

2006 Elsevier Masson SAS. All rights reserved.

h sePDs

(Te peagne prellelini, lag

iati%)

ia

and the megakaryocytes. Not only is PV a chronic disorder

giedisorder in which erythrocytosis, leukocytosis and thrombocy-tosis are all simultaneously present [1]. In PV, all stops toblood production in the bone marrow seem to have been pulledout. The marrow is crowded with great numbers of nucleatedred cells and granulocytes in all stages of maturation withmarked hyperplasia and clustering of enlarged mature mega-karyocytes. Some cases however show only a moderate eleva-tion of erythrocytes with an extreme degree of thrombocytosis,while in others the leukocyte counts may be at or close to leu-kemic levels, with only slight increase in red cells or platelets.As to the etiology of trilinear myeloproliferation in PV, Dame-shek proposed two highly speculative possibilities one, the pre-sence of excessive bone marrow stimulation by an unknownfactor or factors and two, a lack or a diminution in the normalinhibitory factor or factors [1]. This original view and hypoth-esis of Dameshek is recently confirmed by the discovery of theJAK2 V617F mutation by James, Ugo, Casadevall and Vainch-enker in Paris, France [2] demonstrating that the V617F muta-tion induces a loss of inhibitory activity of the JH2 pseudoki-nase part on the JH1 kinase part of JAK2 leading to enhancedactivity of the normal JH1 kinase activity of JAK2, whichmakes the mutated hematopoietic stem cells hypersensitive to

stages in the natural history of PV (Fig. 1) [3,4]:

stage 1). Pure erythrocythemia is featured by increasedhemoglobin, hematocrit and red cell mass with normal leu-kocytes, thrombocytes and spleen size, which is labeled asidiopathic erythrocytosis by Pearson and Whetherley-Mein[5] and Najean et al. [6]. This category of early PV maycomprise about 2030% of the cases at time of presentation;

stage 2). The polycythemic stage of PV is featured bythrombocythemia, erythrocythemia and no or slight myeloidmetaplasia, leukocytosis and/or splenomegaly (Fig. 1);

stage 3). Myeloid metaplasia in PV patients presents withno or different grades of reticulin and collagen fibrosis inthe bone marrow and progressive splenomegaly duringlong-term follow in about one third of the cases;

stage 4). The polycythemic stage with various degrees mye-lofibrosis and splenomegaly following PV may elapse 525 years before a period of normal red cell values so-called spontaneous remission of PV occurs. This stagemust be considered as the beginning of spent phase PVand may last a few to several years. At this point the spleenis frequently large and very firm to palpation, the liver isenlarged to a moderately degree in most patients, thrombo-without any evidence of invasiveness but it is a total marrowblood parameters combined with bone marrow histopathology has a higstages of ET, PV and CIMF in JAK2 V617F positive and negative M 2006 Elsevier Masson SAS. All rights reserved.

Rsum

Les critres cliniques de diagnostic de la thrombocytmie essentielleformes dbutantes avec thrombocytose de PV et de MFI. Ces critres nune claire augmentation du volume globulaire. Les critres OMS de dipar les pathologistes pour clairement diffrencier TE, PV, MFI au staddiagnostiques utilisant les marqueurs biologiques et molculaires actudbutantes de PV et de MFI mimant les TE, et de dfinir les stades cconsquences pronostiques et thrapeutiques. Sont notamment discutsdu dosage drythropotine, de lexistence de colonies rythrodes ou mla mutation V617F de JAK2. Lutilisation de ces marqueurs en assocdiagnostiquer avec une haute sensibilit et spcificit (proches de 100V617F JAK2 positifs ou ngatifs. 2006 Elsevier Masson SAS. All rights reserved.

Keywords: Myeloproliferative disorders; Essential thrombocythemia; Polycythemcolony assay; JAK2 V617F mutation; Bone marrow pathology

Mots cls : Thrombocytmie ; Polyglobulie ; JAK2 ; Biopsie mdulllaire

1. Introduction

In 1950, William Dameshek presented an original view onthe physiopathology and course of polycythemia vera (PV) [1].He described PV as a chronic disorder of the bone marrowcharacterized by excessive production of blood cells by themarrow elements i.e. the nucleated red cells the granulocytes

J.J. Michiels et al. / Patholohematopoietic growth factors TPO EPO, IGF1, SCF and GCSFresulting in trilinear myeloproliferation.nsitivity and specificity (almost 100%) to diagnose the early and overt.

E) dvelopps par le PVSG ne permettent pas de distinguer les TE desrmettent par ailleurs que le diagnostic de formes patentes de PV, avecostic des SMP incluent la biopsie mdullaire, avec des critres dfinisfibrotique ou fibrotique. Nous proposons une mise jour des critresment disponibles, dans le but de distinguer des TE pures les formesques et anatomopathologiques de PV et MFI, ce qui a dimportantesvaleur des donnes morphologiques sur la biopsie de moelle osseuse,acaryocytaires spontanes, de lhyperexpression du gne PRV-1, et deon avec les donnes morphologiques de la moelle osseuse permet deles stades patents mais aussi prcoces de TE, PV et MFI, quils soient

vera; Myeloid metaplasia; Myelofibrosis; Erythropoietin; Endogenous erythroid

2. The concept of PV as a trilinear MPD

Wasserman proposed in 1954 a hypothetical concept for thecourse of PV[3,4]. Accordingly PV is characterized by initialerythrocytosis (erythrocythemia), which is usually accompa-nied by leukocytosis of normal mature granulocytes (leuko-cythemia) with increased leukocyte alkaline phosphatasescore, by thrombocytosis (thrombocythemia), splenomegaly,and plethora. Wasserman distinguished at least five subsequent

Biologie 55 (2007) 92104 93cythemia is frequent and may be pronounced with bizarreand giant platelets, and white cells are usually increased

Based on extended clinical experience and review of theliterature, Glaser and Walker [7] concluded that the transitionof PV to myelofibrotic myeloid metaplasia occurs, but whetherpost-PV is the same or distinct from agnogenic myeloid meta-plasia (AMM) or chronic idiopathic myelofibrosis (CIMF) stillremains an open question and has never been solved in a pro-spective clinicopathological study, which nowadays becomespossible by the use of new molecular and biological markersincluding PRV-1 expression and JAK2 V617F mutation [2].

3. Clinical, laboratory and pathological featuresof the MPDs

J.J. Michiels et al. / Pathologie Biologie 55 (2007) 9210494with granulocytic leukocytosis (leukocythemia) accompa-nied by a small percentage of immature forms;

stage 5). Post-PV myeloid metaplasia shows various degreesof leuko-erythroblastosis of the peripheral blood and mayprogress to extreme myelofibrosis with a dry tap on aspira-tion and massive splenomegaly (Fig. 1). At this end-stagehistopathology of bone marrow biopsy shows a similar pic-ture and can not been differentiated from agnogenic myeloidmetaplasia with no previous history of PV [1,2].

Fig. 1. The evolution and dynamics of the disease process in polycythemia vera(PV) according to the concept Wasserman, who defined PV as a trilinearmyeloproliferative disorder with various degrees of erythrocythemia, thrombo-cythemia, leukocythemia and prefibrotic myeloid metaplasia as initial stagesfollowed by spent phase PV and dry tap myelofibrosis after long-term follow-up in about one third of the cases [3,4].

Table 1The World Health Organization (WHO) [24] and the European Clinical and Patholthrombocythemia: ET

Clinical criteria Pathological critePersistent increase of platelet count WHO and ECP

Increase of dispehyperlobulated nreticulin

A typical ET pict

ECP: > 400 109/lWHO: > 600 109/lECP

Presence of large or giant platelets in peripheral blood smear

ECP

Absence of any underlying disorder for reactive thrombocytosis

ECP

No proliferation oWHO and ECP

No peripheral blood, bone marrow and cytogenetic evidence ofPV, CML, CIMF, MDS or reactive thrombocytosis

ECP

Absence of any cytogenetic abnormality

Molecular biolog

Clonality studies:

Acquired: JAK2

Congenital: polyTPO, cMPL geneIn the 1980s and 1990s the German pathologists Burkhardtet a1 [8], Georgii et al. [912] and Thiele et al. [1318] havedescribed typical histopathological features from bone marrowbiopsy material for the diagnosis and classification of each ofthe 3 different Ph-negative MPDs essential thrombocythemia(ET), PV and AMM. Georgii et al. [9,10] and Thiele et al.[13,14] drew attention to an authentic chronic megakaryocyticgranulocytic myeloproliferation (CMGM) as a separate patho-logical entity among the MPD, distinct from ET and distinctfrom chronic myeloid leukemia (CML) or myelodysplasticsyndrome (MDS). This condition has been described in 1996as Philadelphia-chromosome negative chronic megakaryocyticgranulocytic myeloproliferation (CMGM) by Georgii et al. [11,12] and as prefibrotic CIMF by Thiele et al. [18]. Since theearly 1980s Michiels and Juvonen [1921] combined clinicaland pathological features by including bone marrow histo-pathology as a pathognomonic, diagnostic clue to each of theMPDs. In 1997, the thrombocythemia vera study Group(TVSG) extended the clinical criteria of the polycythemiavera study Group (PVSG) for the diagnoses of ET and PV byincluding bone marrow histopathology as a diagnostic clue andspecific feature of ET, PV (Tables 1 and 5A) [21]. Thiele et al.[22,23] improved the TVSG criteria for ET, PV and added theCologne criteria for prefibrotic, early fibrotic and overt fibroticstages of CIMF (Table 2). These clinical and pathological cri-teria were diagnostic for MPD [2023] and inspired the WorldHealth Organization (WHO) [24] to clearly define the patholo-gical bone marrow criteria for the differentiation between true

ogical (ECP) [25] criteria for the diagnosis of acquired or congenital essential

ria

rsed or loosely clustered, predominantly enlarged mature megakaryocytes withuclei and mature cytoplasm, normal cellularity, no or borderline increase of

ure excludes PV, AMM, CML, MDS, and reactive thrombocytosis

r immaturity of granulo- or erythropoiesisy: ECP

polyclonal or monoclonal

V617F positive or negativeclonal and JAK2 V617F negative , caused by gain of function mutation ofs or of unknown etiology

al and Pathological (ECP) [25] Criteria for the diagnosis of prefibrotic and early

criteria, WHO, ECPakaryocytic and granulocytic myeloproliferation (CMGM) [912,20] and no ortion of erythroid precursors. Abnormal clustering and increase in atypical giant tomegakaryocytes containing bulbous (cloud-like) hypolobulated nuclei and

turation defects

sensus on grading of myelofibrosis (MF) [80]rotic CIMF: focal fine reticulin with no intersections (cross-over) and only rarein fibersfibroas,is

olog

pos

toge

gie Biologie 55 (2007) 92104 95ET, classical PV, and prefibrotic and fibrotic stages of CIMF.The WHO criteria are essentially drawn by internationalexperts in bone marrow morphology but not clinicians. Toovercome the shortcomings of the clinical PVSG and thepathological WHO [24] criteria, Michiels and Thiele [25,26]have defined a set of European Clinical and Pathological(ECP) criteria, which by including the available biologicaland molecular markers (Tables 1, 2, 5B, 7, 8) will much betterallow to diagnose and classify each of early and overt stages ofthe MPDs ET, PV and CIMF [25,26].

4. WHO pathological criteria for the diagnosis of ET, PVand CIMF

A typical bone marrow picture according to the WHO [24]for true ET affects mainly of megakaryocytic cell lineage,shows increased numbers of loosely clustered enlarged, maturemegakaryocytes with hyperploid staghorn-like nuclei togetherwith normal cellularity, normal erythropioesis, and normalgranulopoiesis, no increase of reticulin fibrosis, and there is

Table 2The World Health Organization (WHO) [24] and the updated European Clinicfibrotic agnogenic myloid metaplasia (AMM)

Clinical criteria, ECP PathologicalNo preceding or allied other subtype of myeloproliferativedisorders CML or MDS.Main presenting feature of prefibroticCIMF is thrombocythemia and slight splenomegaly, no dry tapon bone marrow aspiration and diagnosed as ET according tothe PVSG.

Chronic megrelative reducmedium sizeddefinitive ma

Clinical and laboratory features European conNormal hemoglobin or slight anemia, grade I: hemoglobin > 12g/dl

MF 0: Prefibcourse reticul

Slight or moderate splenomegaly on ultrasound scan or CT MF 1: Earlyperipheral are

Thrombocytosis, platelets in excess of 400, 600 or even1000 109/l

Note that MF

No leuko-erythroblastosis

No tear drop erythrocytes

Molecular bi

JAK2 V617F

Screen for cy

J.J. Michiels et al. / Patholono peripheral blood, bone marrow and cytogenetic evidenceof typical and atypical chronic myeloid leukemias (CML),PV, myelodysplastic syndrome (MDS) or reactive thrombocy-tosis (Table 1) [2026]. A typical bone marrow picture for trueET excludes any other MPD, as well as typical and atypicalCML, MDS and reactive thrombocytosis.

A typical bone marrow picture according to the WHO [24]for a prefibrotic form of CIMF is characterized by a prominentgranulocytic and megakaryocytic myeloproliferation (CMGM)[11,12,20] and no or borderline increase in reticulin (CIMF-0,Table 2). The prevalence of the (left-shifted) neutrophil andmegakaryocytic lineage is associated with reduction andmaturation arrest of erythroid precursors. Most conspicuous,however is the fact that the megakaryopoiesis is characterizednot only by a disturbance of bone marrow histotopography(loose to dense clustering and translocation to the endostealborders), but also by striking abnormalities of maturation.The immature megakaryocytes in prefibrotic CIMF consist ofvariations in size including giant forms and deviations of thenuclear-cytoplasmic ratio accompanied by bulbous and hyper-chromic cloud-like nuclei, which are never seen in ET and PV.Thiele et al. [18,2736] demonstrated that prefibrotic and earlyfibrotic CIMF usually presents with thrombocythemia (falseET), which has to be distinguished from true ET because clin-ical features, natural history and prognosis significantly differ(Table 2). The prefibrotic stage of CIMF is not only associatedwith pronounced thrombocythemia, but also show no leuko-erythroblastic blood picture, normal or increased LAF-scoreand no or minimal splenomegaly (Table 2) [18,2736], andtherefore diagnosed as ET according to the PVSG [37,38] cri-teria (Tables 14).

A typical picture in the bone marrow according to the WHO[24] pathognomonic and diagnostic for PV is featured byincrease of clustered enlarged mature megakaryocytes compar-able to ET, and a moderate to marked increased cellularity,erythropoiesis and granulopoiesis (i.e. panmyelosis) (Table 5A,5B) [2026]. Bone marrow features in polycythemic stage ofPV show a hypercellular bone marrow of prominent erythroid

tic CIMF: loose network of reticulin with intersections, especially inno collagenizationnot a disease but secondary to AMM

y:

itive or negative

netic abnormalitiesprecursor cells and neutrophil granulopoiesis in addition tomegakaryocytes proliferation with loose arrangements of clus-tered megakaryocytes. Megakaryocytes do not only show dif-ferent sizes, but fail to exhibit significant maturation defects.The megakaryocytes in PV usually have a rather pleiomorphicappearance with wide ranges of sizes including small and giantforms. A typical PV picture of the bone marrow is seen in

Table 3PVSG [38] criteria for the diagnosis of ET

Platelet count > 600 x 109/l: 1986 No known cause of Reactive Thrombocytosis Normal hemoglobin and red cell mass to exclude overt PV Stainable iron bone marrow to exclude PV No features of MDS in bone marrow smear and biopsy Absence of Ph1+ chromosome (brc/abl) to exclude CML Collagen fibrosis of bone marrow is allowedup to

onoc

one

sedB

B

gieTable 4ET according to PVSG [38] compared to WHO [24] and ECP [25] criteria

ET PVSG Hereditary ET True ETIncludes Incidence < 0.001 2030Serum EPO Normal Normal

Platelet / > 400 ECP> 600 WHO

Erythrocytes N NHematocrit N NBone marrow: ET picture ET pictureMegakaryocytes Normal large / giant and matureSplenomegaly JAK2 V617F Neg: EEC Neg: PVR-1 Not applicable Clonality polyclonal polyclonal m

Table 5AExtension of the PVSG [39] criteria of polycythemia vera (PV) by including b

The Rotterdam Criteria of Polycytemia Vera PropoA1 Raised red cell mass

males > 36mL/KgA2 Absence of any cause of secondary erythrocytosis by clinical and

laboratory investigations

J.J. Michiels et al. / Patholo96classical PV according to the PVSG [39], in erythrocythemiaor idiopathic erythrocytosis, in early or latent PV and inmasked PV (Fig. 2 and Tables 6 and 7) [25,26,4044]. In con-genital polycythemia (CP) and secondary erythrocytosis (SE),in which increased erythropoiesis is present, the number, size,morphology and distribution of megakaryocytes in bone mar-row smears and biopsies remain normal [2023,40,41].

5. Current clinical and pathological criteriafor the diagnoses of ET PV and CIMF

Since 1975 the minimum criterion of the Polycythemia VeraStudy Group (PVSG) for the diagnosis of ET was 1000 109/l(1 million) [37]. In 1985, we demonstrated that the majority ofsymptomatic ET patients diagnosed by increase and clusteringof enlarged megakaryocytes in bone marrow biopsies, had pla-telet counts between 400 and 1000 109/l (below 1 million).This prompted the PVSG to lower the platelet count for thediagnosis of ET to the arbitrary minimum of 600 (Table 3)[38]. We could consistently demonstrate that the arbitraryminimum of 600 109/l platelets according to the PVSG andWHO overlooks early stage MPD at platelets between 400 and600 with no or slight splenomegaly, with no or low serumEPO, and typical MPD features in the bone marrow consistent

A3 Histopathologie of bone marrow biopsy increase of: Ba. cellularity, panmyelosisb. enlarged megakaryocytes with hyperploid nuclei;c. reticulin fibers (optional)

BA1 + A2 + A3 is consistent with early stage PV (so-called "idiopathic erythrocytosA1 + A2+ A3 + any one from category B establishes overt PVA3 + B1 is consistent with essential thrombocythemiaA3 + B3 and/or B4 is consitent with a primary myeloproliferative disorderEarlyl PV Prefibrotic CIMFMimicking ET False ET2030 4060Decreased Normal

> 400 ECP > 400 ECP> 600 ECP > 600 WHON/ N/N/ N/PV picture MMM picture

abnormal ++ ++ ++

lonal Monoclonal Monoclonal

marrow histopathology as a diagnostic clue to early and overt stages PV [21]

by the thrombocithemia Vera Study group (TVSG)1 Thrombocytosis

Platelet count > 400 109/L2 Granulocytes > 10 109/L and/or raised neutrophil alkaline

phosphatase score of > 100 in the absence of fever or infection

Biologie 55 (2007) 92104with early ET (Table 1), early PV mimicking ET (Tables 4,5A, 5B and 6) or prefibrotic CIMF or unclassifiable MPD(Tables 1 and 3).

The PVSG criteria for ET [38] is a diagnosis of exclusion ofreactive thrombocytosis, PV, MDS and Ph1+ CML but includesby definition thrombocythemia associated with prefibrotic andearly fibrotic CIMF (Table 3) [1723]. The WHO [24] and theECP [25] criteria for the diagnoses of ET extend the PVSG[34] criteria by including histopathology from bone marrowbiopsies as a positive criterion of MPD and as powerful toolto differentiate between true ET, false ET and early PVmimicking ET (Table 4). Comparing the WHO and ECP withthe PVSG criteria for the diagnosis of ET show that the PVSGcriteria fail to distinguish ET from early PV mimicking ET[4044] and fail to distinguish ET from usually pronouncedthrombocythemia associated with prefibrotic CIMF (Tables 4)[2325,3336]. In consideration of disease-related complica-tions occurring at low platelet counts, the arbitrarily chosenlimit for platelet count (> 600 10/l) by the PVSG (Table 3)[38] and WHO (Table 1) [24] has been reduced to 400 109/lin the ECP [25] (Table 1) criteria for ET. The PVSG [38] cri-teria for ET, when compared to the WHO [24] and ECP [25]criteria, also include early PV mimicking ET (Tables 4 and 6)[4044].

3 Splenomegaly on palpation or isotope/ultrasound scan

4 Erythroid colony formation in absence of EPO: spontaneous EECis")

RCMean

dl in

gieTable 5BDiagnosis of PV according to the PVSG, WHO and ECP criteria

PVSG [39] WHO [24]Criteria Major criteria Major criteriaA1 Red cell mass: RCM Red cell mass:Overt

PV

Male > 36 ml/kg

Female > 32 ml/kg

> 25% above m

normal valueor Hb > 18.5 g/

J.J. Michiels et al. / PatholoThe PVSG have used increased red cell mass proposed byDameshek in 1950 [1] as the main inclusion criterion for diag-nosis of PV in the PVSG 01 study since 1969 [4], whichbecame a main diagnostic criterion for the diagnosis of PVsince 1975 (Tables 5A, 5B) [39]. Increased red cell mass isnot specific, because it includes patients with PV, congenitalpolycythemias (CP), and secondary erythrocytosis (SE),thereby introducing a main differential diagnostic problemand the need of extensive laboratory investigations [45,46].

women

Latent PV

A2 Normal arterial oxygen saturation > 92% Absence secondaryA3 Splenomegaly on palpation Splenomegaly on pA4 Clonal evidence othA5 Spontaneous EECCriteria Minor criteria Minor criteriaB1 Platelets > 400 Platelets > 400B2 Leukocytes > 12 Leukocytes > 12B3 Bone marrow biopsy has been disregarded by

the PVSG and included by the TVSG as adiagnostic clue for early and overt PV [21]

Bone marrow biops

Increased cellularityferation and clustermorphic) megakary

B4 Raised LAP score Low serum EPODiagnosis DiagnosisA1 + A2 + A3

A1 + A2 + two from B

A1 + A2 + any othe

A1 + A2 + two fromManifest PV:

Increased RCM

Manifest PV:

Increased RCM

Table 6Clinical staging of Polycythemia Vera: therapeutic implications

Polycythemia Vera EvolutionStaging of PV as one distinctdisease

ECP 1 Aspirin/phlebotomy ECP 2 Aspirin/ph

initial PV mimicking ET ErythrocythemicIncidence (%) 2025 2025Hemoglobin g/dl N/ Serum EPO

Hematocrit Male 0.43- < 0.51 Male > 0.51Female 0.42- < 0.48 Female > 0.48

Red cell mass N Thrombocytes (x 109/l) >400600 < 400Leukocytes (x 109/l) N NSpleen on echogram N15 cm NBone marrow: PV picture PV pictureJAK2 V617F + +EEC + +PRV-1 LAF score Myelofibrosis (MF) 0 0ECP [25]Clinical criteria

Red cell mass optional> 25% above mean normal value

men, Hb > 16.5 g/dl in or Hb > 18.5 g/dl in men

Biologie 55 (2007) 92104 97The PVSG [39] postulated three major and four minor criteria(Table 5B) to ensure that patients who entered the PVSG pro-spective trial indeed were suffering from PV and not from con-genital polycythemia or secondary erythrocytosis [39].Increased red cell mass corresponded with high hematocritvalues between 0.50 and 0.75 [39] (ECP stage 3 and 4 Table 6),which according to the PVSG 01 study [39] was associatedincreased platelet count in two third and with major thrombosisin one third of more than 400 PV patients at time of diagnosis

Hb > 16.5 g/dl in womenRed cell mass normaland Ht < 0.51 in men

Ht < 0.48 in femaleerythrocytosis Absence of secondary erythrocytosisalpation Splenomegaly on CT or ultrasound (> 12 cm)er than Ph1+ or BCR/ABL Clonal evidence other than Ph1+ or BCR/ABL

Spontaneous EECClinical criteriaPlatelets > 400 109/lLeukocytes >12 109/l

y with typical PV picture

with trilineage myeloproli-ing of small to giant (pleio-ocytes

Bone marrow biopsy with typical PV picture

Increased cellularity with trilineage myeloproli-feration and clustering of small to giant (pleio-morphic) megakaryocytesLow serum EPODiagnosis

r from A

B

B3 plus any other of the clinical criteria

Manifest PV:

Increased RCM

Early stage PV:

RCM Normal

Manifestationlebototm ECP 3 PVSG WHO A/P ECP4 PVSG WHO IFN/HU

PV Thrombo/erythro/leukocythemic PV4060

> 400 > 1,000N > 15N15 cm > 15 cmPV picture PV MF picture+/++ ++/LOH+ + 0/1 1/2

[4,39]. The thrombocythemia vera study group (TVSG,Table 5A) [21] extended the PVSG criteria by including bonemarrow histopathology, which prompted pathologists to definethe WHO [24] and clinicians to define ECP [25] criteria for thediagnosis of early or masked PV and overt and advancedstages of PV (Tables 4 and 6).

According to the TVSG [21] criteria in Table 5A a typicalPV picture of the bone marrow is seen in 4 different categoriesof MPD patients.

First, the combination of a typical PV picture, increased redcell mass, high hematocrit and one of the B criteria (Table 5A)is consistent with classical PV according the PVSG [39] andWHO [24] and as advanced stage 3 and 4 PV when applyingthe ECP [25] criteria (Table 6).

J.J. Michiels et al. / Pathologie98Second, a typical PV picture of the bone and increased redcell mass, high hematocrit but with normal platelet count andspleen size is consistent with erythrocythemia according toWasserman [3] (Fig. 1), diagnosed as idiopathic erythrocytosisby the PVSG criteria [5,6], and diagnosed as erythrocythemicPV according to WHO [24] and stage 2 ECP [25] when com-bined with serum EPO levels and EEC (Tables 5B and 6). Incontrast to the PVSG [39] criteria, both the WHO [24] andECP [25] criteria clearly differentiates between PV, CP andSE without the need of red cell measurement in PV, and witha clear indication for red cell mass measurement in CP and SE[21,25,26].

Third, a typical PV picture of the bone marrow with ahematocrit in the upper limit of normal but with increased pla-telet count is consistent with ET according to PVSG [39] andWHO [24] but diagnosed as early stage 1 PV mimicking ETaccording to the ECP [25] criteria (clinical case 2, Fig. 2 andTables 4, 5A, 5B and 6) [25]. Early stage 1 PV usually pre-sents with microvascular disturbances at platelet count inexcess of 400 109l, increased LAF score, low serum EPO,spontaneous EEC and no or slight splenomegaly (Tables 4and 6) [26].

Fourth, the combination of a typical PV picture, normal redcell mass, normal hemoglobin and hematocrit, normal plateletbut slowly progressive splenomegaly, granulocytosis or even

Fig. 2. The evolution and dynamics of the disease process in polycythemia vera(PV) according to Thiele et al. [4044] indicating the sequential occurrence of

the early initial stage of PV mimicking ET, the overt polycythemic stage ofclassical PV, and progression to post-PV myeloid metaplasia or leukemia asterminal stages in about one third of the cases.slight anemia is not consistent with either ET, PV but withunclassifiable MPD or masked PV [21]. Such cases of unclas-sifiable MPD and masked PV are overlooked by clinicians andmay comprises about one quarter of patients with a typical PVpicture of the bone marrow. Such cases may present withthrombotic complications including splanchnic vein thrombosis[47,48], are described as masked PV [48,49], and frequentlyshow spontaneous EEC as the clue to the atypical presentationof MPD. Masked ET or PV may progress to so-called classi-cal CIMF without overt PV.

Using clusters of pleiomorphic giant megakaryocytes inbone marrow biopsy as a reference standard for the diagnosisMPD, a French study demonstrated that 46 out of 128 conse-cutive patients with splanchnic (hepatic, portal or mesenteric)vein thrombosis had features of MPD [48]. These 46 MPDpatients with splanchnic vein thrombosis had hepatic veinthrombosis (Budd Chiari syndrome) in 19 and portal or mesen-teric vein thrombosis in 27 [48]. As compared to a positivebone marrow finding of clustered giant megakaryocytes diag-nostic for MPD, the sensitivity for the diagnosis of MPD asso-ciated with splanchnic vein thrombosis was 63% for increasedred cell mass, 52% for low serum EPO level, 72% for EEC,and 74% for splenomegaly indicating the superiority of bonemarrow histopathology to detect masked early and overt stagesof MPD in this setting [48].

Thiele et al. [4044] confirmed the sequential occurrence ofearly PV mimicking ET, overt and advanced stages of PV(Fig. 2 and Table 6). The WHO [24] and ECP [25] criteriaclearly differentiate PV from CP, SP, and ET from reactivethrombocytosis, initial PV and prefibrotic CIMF. The PVSG[39] and the WHO [24] criteria used increased red cell as amandatory main criterion (Table 5B), which is a very crudeand therefore overlook by definition early PV mimicking trueET (ECP stage 1 in Table 6) and the erythrocythemic phase(ECP stage 2 in Table 6) of PV, formerly labeled as idiopathicerythrocytosis [5,6]. Accumulating data from the literatureshow that spontaneous endogenous erythroid colony formation(EEC) [5153], low serum EPO [5460] and PRV-1 expres-sion [6164] are the hall mark of PV. About 50% of ETpatients according to the PVSG are EEC positive [51,6569],PRV-1 positive [6164], and may be associated with lowserum EPO [50,5759] indicating early PV mimicking ET(Tables 6 and 7). The reports on EEC/PRV-1 positive ET diag-nosed according to the PVSG [39] may have with low serumEPO and therefore represent early ECP stage 1 of PV (Tables 6and 7). The cohorts of early ECP stage 1 and 2 PV and theovert ECP stage 3 PV patients are featured by spontaneousEEC, positive PRV-1, low serum EPO levels and a typicalPV bone marrow picture (Fig. 2 and Tables 6 and 7) [5466].Both the early (ECP stage 1 and 2) and overt (ECP stage 3) PVpatients are at high-risk for potential minor and major vascularcomplications, because they present with elevated plateletcount mimicking true ET [65,66]. EEC/PRV-1 positive ETpatients according to PVSG criteria have a high risk of devel-

Biologie 55 (2007) 92104oping microvascular and major thrombotic complication ascompared to EEC/PRV-1 negative ET [65,66]. PRV-1-negative

sis o

Pl P

f

P

A

N

P

P

P

P

P

D

C

P

Ps

gieET comprises a pathophysiologically distinct subgroup of trueET patients with no features of early PV, who are at lower riskfor the development of thrombotic complications and for emer-gence of PV [66]. Early or initial PV (or forme fruste of PV)according to updated ECP criteria in Table 7 is typically fea-tured by a PV picture in the bone marrow, positive results forEEC and PRV-1, and/or low serum Epo levels, and a highthrombotic risk, which is related to increase of hypersensitiveplatelet counts (thrombocythemia) and slight increase of hema-tocrit up to 0.50, and therefore candidates for low dose aspirinand phlebotomy (Table 6).

Spontaneous growth of endogenous erythroid colonies(EEC) and serum erythropoietine (EPO) levels are important,but have insufficient diagnostic specificity and sensitivity asisolated parameters to differentiate between PV and SE. Thereliability of peripheral blood (PB) and bone marrow (BM)EEC was investigated in a multicenter study including 140patients (80 PV according to the PVSG, 54 SE, 6 idiopathic

Table 7The updated European Clinical and Pathological (ECP) Criteria for the Diagno

Clinical (C) criteria suspected for PVC 1. Classical PV: Red Cell Mass optional Hemoglobin > 18.5/> 16.5 g/dmale/female Hematocrit (Ht) > 51/> 48% male/female

C 2. Early or latent stage PV

Hematocrit (Ht): 0.450.51 male and 0.430.48 female

C 3. Low plasma Epo level (ELISA)

C 4. Persistent increase of platelet count: grade I: 4001500, grade II: > 1500.

C 5. Splenomegaly on palpation or on ultrasound echogram (>12 cm length indiameter).

C 6. Granulocytes > 10 109/l or Leukocytes > 12 109/l and/or raised LAP-score or increased PRV-1 expression in the absence of fever or infection.

C 7. Platelet-mediated microvascular ischemic, thrombotic complications

C 8. Typical PV signs and symptoms of hypervolumemia

C 9. Itching, fatigue, upper abdominal complaints

C 10. Absence of any cause of secondary erythrocytosis.

J.J. Michiels et al. / Patholoerythrocytosis) and 10 healthy controls [52]. PB and BMEEC were positive for 81% and 84% of PV patients, and94% of PV were diagnosed when BM and PB EEC assayswere performed with a specificity of near 100%. The diagnos-tic impact of low serum EPO levels (ELISA assay) was eval-uated in a multicenter study including 186 patients (116 PVaccording to the PVSG, 66 SE and 4 idiopathic erythrocytosis)[60]. The majority of PV patients displayed a serum levelbelow the normal range of 3.3 IU/l with a sensitivity of 87%(101/116), a specificity of 97% and a positive predictive valueof 97.8%. Statistical analysis (ROC curves) could define twothresholds allowing a specific and correct diagnosis of PV byEPO values below 1.4 IU/l and SE by EPO values above 13.7IU/l. Only 66% (65/99) of PV and 20% (13/66) of SE werediagnosed with these cut-off points, indicating an overlap ofserum EPO in PV versus control and controls versus SE [60].Histopathology from bone marrow biopsies clearly distin-guishes PV from congenital polycythemias (CP) or secondaryerythrocytosis (SE) with a sensitivity and specificity of morethan 95% to almost 100% [4044].6. Myelofibrosis (MF) versus chronic idiopathicmyelofibrosis (CIMF)

Prefibrotic CIMF according to WHO [24] and ECP [25] is adual mixed proliferation of increased granulopoiesis and mega-karyopoiesis dominated by immature giant megakaryocyteswhich are conspicuously enlarged due to increase of nuclearas well as cellular size with bulky and irregular roundish-shaped nuclei, so-called cloud-like nuclei, which are neverseen in ET and PV (Table 2) [2236]. The PVSG criteria forthe diagnosis of ET include prefibrotic or early fibrotic CIMF,which is usually associated with pronounced thrombocythemia(false ET) (Table 4) [33,77]. Classical CIMF is defined as aclinicopathological entity not preceded by any other MPD,CML or MDS and characterized by various degrees of anemia,splenomegaly, a leuko-erythroblastic blood picture with teardrop-shaped erythrocytes and dry tap on bone marrow aspira-tion due to various degrees of bone marrow collagen fibrosis or

f Early, Overt and Advanced Stage Polycythemia Vera (PV)

athological (P)criteria diagnostic for PV1. Bone marrow pathology: increased cellularity with trilineage myeloproli-

eration (i.e. panmyelosis).

roliferation and clustering of small to giant (pleiomorphic) megakaryocytes.

bsence of stainable iron.

o pronounced inflammatory reaction (plasmacytosis, cellular debris).

2. Bone marrow biology: spontaneous erythroid colony (EEC) formation.

3. Molecular biology: heterozygous or homozygous JAK2 V617F mutation.

1 + P2 + P3 = true PV [1,2,91].

1 or P2 plus P3 plus C1 is classical PV

1 or P2 plus P3 plus any of C2 to C10 is early PV mimicking ET

iagnosis PV:

1 plus P1 and P2 or C1 and P2 plus any other criterion establish classical PV

1 and C2 plus any other C criterion establish masked, or early PV

1, 2, 3 plus C1, 3 plus none of the others is consistent with erythrocythemictage of PV

Biologie 55 (2007) 92104 99osteosclerosis (Table 8) [7076]. All studies and reports in theliterature on CIMF up to date have disregarded and excludedjust by definition the prefibrotic and early reticulin fibroticstage of CIMF (Table 2).

MF is not a disease because reticulin and collagen fibrosisare produced by polyclonal fibroblasts as the consequence ofcytokines released from the clonal granulocytic and megakar-yocytic proliferative cells in PV and CIMF. The well-knownBaumeister scoring system was developed on aspirated bonemarrow samples [78]. The Manoharan system scores thedegree of reticulin derived bone marrow biopsy [79]. A scoringsystem based on morphometric analysis (point intersectionwith an ocular grid) and quality of fibers (reticulin and col-lagen fibers) and the bone marrow fiber density (fine or coursereticulin and some or course bundles of collagen) have beenproposed by Georgii et al. [1113] and by Thiele et al. [18,23]. All these different scoring systems for MF use differentcriteria for grading of reticulin and collagen, are subjectiveand not comparable by lack of strict criteria. A panel of experi-enced European pathologists and a foreign expert reached a

consensus on how to grade bone marrow fibrosis (myelofibro-sis: MF) in bone marrow biopsies of patients with CIMF or PV[80]. Grading of MF was simplified by using four easily repro-ducible categories including differentiation between reticulinand collagen [80]. According to defined standardized semi-quantitative grading of reticulin and collagen fibrosis in thebone marrow, MF can reliably be graded at the pathologicalbone marrow level as 0 in prefibrotic, as 1 in early fibrotic,as 2 in classical fibrotic and as 3 in classical sclerotic CIMF(Tables 2 and 8) [80].

MF is not a feature of true ET and very few ET patients willdevelop myelofibrosis during long-term follow-up [11,12,8183]. MF is present in only a minority of PV patients at time ofdiagnosis, but all stages of myelofibrosis have been observedduring long-term follow-up [11,12,81,82]. As compared to anormal or near normal life expectancy in true ET and PV, pre-fibrotic and early fibrotic CIMF with maturation defects ofenlarged dense clustered megakaryocytes (pronounced dysme-gakaryopoiesis) are featured by slowly progressive myelofibro-sis (MF) progressive splenomegaly, thrombocytopenia and/or

ities [89]. The Lille scoring system is derived from patientswith classical CIMF or agnogenic myeloid metaplasia(AMM) and based on two adverse prognostic factors namelyhemoglobin < 10 g/dl and leukocytes count < 4 or> 30 109/l88. The Lille score is widely used and able to sepa-rate patients with advanced (classical) CIMF into three groupswith low (score 0), intermediate (score 1) and high risk (score 2) of progressive disease and loss of life expectancy. TheCologne score is derived from patients with prefibrotic, earlyfibrotic and classical CIMF and uses age (70 years), hemoglo-bin (10 g/dl), platelets (< 300 109/l) and leukocytes(> 20 109/l) myeloblasts (> 2%) or erythroblasts (> 2%) asfactors with prognostic impact. The Cologne scoring systemreveals a strong discriminating power and is applicable for pre-fibrotic and early CIMF (CIMF 0-1) as well as advanced stagesof CIMF (CIMF 2-3). Applying the Lille score to a cohort of458 patients with prefibrotic or early fibrotic CIMF (CIMF 0-1), 398 belonged to the low risk group with a very good prog-nosis (60% survival after 15 years) and only 57 and 4 belongedto the intermediate and high risk group with a much less favor-

clas

P, D

Ba

Acd

E

Mt

Ma

H

Ol

E

J.J. Michiels et al. / Pathologie Biologie 55 (2007) 92104100anemia and a significantly shortened life expectancy as com-pared to a normal or near normal life expectancy in ET and PV[8487]. It is reasonable to assume that the prognosis of CIMFmay depend on the grade of MF in the bone marrow at time ofdiagnosis or evaluation. Early CIMF with MF grade 0 (prefi-brotic) and grade 1 (early fibrotic) show a more favorable prog-nosis than advanced stage CIMF with grade 3 MF. However,the survival curves of CIMF patients with grade 0, 1 and 2 arenot significantly different. Age, anemia, leukocyte and plateletcounts, but not the degree of MF (except MF grade 3 and ahypocellular bone marrow) appeared to be the most reliableand important parameters for prognosis and survival [8789].The Sheffield scoring system is based on age, hemoglobin(10g/dl) and the presence or absence of cytogenetic abnormal-

Table 8The WHO [24] and updated ECP [25,26] Criteria for the clinical diagnosis of

ClinicalNo preceding or allied other subtype of myeloproliferative disorders ET, PVCML, CMML atypical CML or MDS.

Intermediate clinical stage

Anemia grade II:

hemoglobin > 10g/dl

Definitive leuko-erythroblastic blood picture and/or tear drop erythrocytes

Various degrees of thrombocythemia or normal platelet counts

Various degrees of splenomegaly

No adverse signsa

Advanced clinical stage

Anemia grade III:

hemoglobin < 10 g/l

Various degrees of thrombocytopenia

plus one or more adverse signsa

Molecular biology

JAK2 V617 positive: post-PV MM?

JAK2 V617F negative: classical CIMF?

Screen for cytogenetic abnormalities

a Adverse signs: age > 70 years, hemoglobin < 10 g/dl, myeloblasts PB > 2%, ery

0 109/l, severe constitutional symptoms, massive splenomegaly, cytogenetic abnoable prognosis (less than 50% survival after 5 years) [87].There may be an observer disagreement with regard to the

classification and natural history of prefibrotic and early CIMFand its differentiation from true ET [87,90]. Those cases withprefibrotic CIMF and slight dysmegakaryocytopoiesis are fea-tured by slowly progressive myelofibrosis and splenomegalyand have a life expectancy close to normal similar as in PV[87]. Discussions between clinicians and pathologists revealthat diagnostic differentiation between true ET and thrombo-cythemia as the presenting feature of prefibrotic MM withslight maturation defect of enlarged clustered megakaryocytes(slight dysmegakryopoiesis) and no or slight increased cellular-ity is subjective and due to a rather high inter-observer dis-agreement between pathologists [87,90].

sical fibrotic or sclerotic chronic idiopathic myelofibrosis (CIMF)

athologicalry tap on bone marrow aspiration is consistent with MF grade 2 or 3.

one marrow pathology: megakaryocytic and granulocytic myeloproliferationnd relative reduction of erythroid precursors.

bnormal clustering and increase in atypical giant to medium sized megakaryo-ytes containing clumsy (cloud-like) lobulated nuclei and definitive maturationefects.

uropean consensus on grading of myelofibrosis (MF) [80]

F 2, manifest CIMF: Diffuse increase in reticulin with extensive intersec-ions and only focal bundles of collagen. Hypercellular bone marrow

F 3, overt CIMF: Diffuse and dense increased reticulin with extensive inter-ctions with course bundles of collagen and significant osteosclerosis.

ypercellular bone marrow

steomyelosclerosis: Slerosis, endophytic bone formation and decreased cellu-arity.

ndstage hypocellular bone marrowthro-normoblasts PB > 2%, leukocytosis > 2 0 109/l, thrombocytopenia < 30rmalities.

7. The molecular etiology of the MPDs ET PV and MF

As to the etiology of trilinear myeloproliferation in PV,Dameshek proposed in 1950 two highly speculative possibili-ties: the presence of excessive bone marrow stimulation by anunknown factor or factors, and a lack or a diminution in thenormal inhibitory factor or factors [1]. The JAK2 V617F muta-tion reflects a gain of function mutation, which is in line withthe concept of Dameshek [1] that all stops to blood produc-tion in the bone marrow seem to have been pulled out, whichnow appears to due to one factor, the JAK2 V617F mutationdiscovered Vainchenker et al. [91] and rapidly confirmed byseveral investigators [9296]. The JAK2 V617F mutationcauses hypersensitivity of hematopoietic progenitor cells togrowth factors, which readily can explain the trilinear myelo-proliferation and the interrelationships between ET, PV and

whether the JAK2 V617F mutation, homozygous in particularor additional genetic events and/or the use of potential leuke-mogenic drugs contribute to the leukemic transformation ofMPDs.

The current concept is that heterozygous JAK2 V617Fmutation with increased kinase activity is enough for megakar-yocyte proliferation and increased hypersensitive plateletscomplicated by platelet-mediated microvascular events withno or slightly increased erythropoiesis in ET and in early PVmimicking ET (Table 9) [2]. Homozygous JAK2 mutation withpronounced kinase activity is associated with trilinear mega-karyocyte, erythroid and granulocytic myeloproliferation, mye-loid metaplasia and secondary myelofibrosis (MF) with themost frequent clinical picture of classical PV complicated bymajor thrombosis on top of the platelet-mediated microvascularthrombotic syndrome of thrombocythemia (Table 9). The

cell

pa

e 0

Microvascular Macrovascular

J.J. Michiels et al. / Pathologie Biologie 55 (2007) 92104 101Thrombosis ThrombosisCIMF (Table 9) [2,9799]. Applying the PVSG criteria, 4250% of the ET patients, 4267% of CIMF patients, 7292%of PV patients have the mutated the JAK2 allele as detectedby DNA sequencing [9296]. A much higher frequency ofJAK2 of 97% in PV, 4957% in ET and 57% in CIMF wasdescribed in three studies that used allele-specific polymerasechain reaction (PCR) analysis [94,96,100,101]. The presenceof the JAK2 V617F mutation was strongly correlated withPRV-1 over expression and the ability to form spontaneousEEC in all three subtypes of MPD (Tables 6 and 7) [102,103]. The JAK2 V617F mutation was absent in more than600 healthy controls, in patients with Ph1+ CML, in patientswith reactive thrombocytosis [9296], and in patients with denovo AML, CLL, B-ALL and T-ALL [96,104]. The mutationhas been found rarely in MDS (3/116 = 2.6%) [92,96], hyper-eosinophilic syndrome (2/145 = 1.4%) [92,96,105,106], butsomewhat more frequent in chronic myelomonocytic leukemia,and atypical CML 31/408 = 7.6%) [92,96,105,106]. The JAK2V617F mutation is rather frequent in unclassified MPD(13/53 = 25%) [96] and frequent in AML with preceding MPD(12/22 = 55%) [106]. The latter finding arises a key question

Table 9Molecular etiology of platelet-mediated microvascular thrombosis, increased redVainchenker & Michiels 2005

JAK2 V617F gain of function mutation in trilinear hematopoietic cells of MPDStep 1 V617F+ Step 12 V617F++

LOH Spontanuous SpontanuousCFU-MK / EEC EEC, CFU-MKET PV Increase of enlargedhypersensitive platelets

Increase of hematocrit to abovhigher platelets

Already at platelet >400

Clinical Step 1 Clinical Step 2Aspirin sensitive Aspirin/Phlebotomy

See figures 1 and 2 regarding the natural history of true PV according to Wassermsequential occurrence of heterozygous and homozygousV617F mutation can readily explain the progression of ETand early PV to overt PV and may explain at least in part theaccompanied granulocytic proliferation, leukocyte activation(increased LAP score and PRV-1 expression), leukocytosis(leukocythemia), and secondary myelofibrosis (Table 9) [1,2].

Depending on the set of laboratory tests and diagnostic cri-teria used, the population of the MPD patients defined as ET,PV and CIMF are heterogeneous at the molecular and biologi-cal level (Tables 1, 2, 7, 8). As the JAK2 V617F mutation isthe cause of a distinct trilinear MPD in its manifold clinicalmanifestations during long-term follow-up [2], the specificityof a positive JAK2 V617F PCR test for the diagnosis ofMPD is high (near 100%), but only half of ET and CIMF (sen-sitivity 50%) and the majority of PV (sensitivity 8597%) areJAK2 V617F positive. Kiladjian et al. [107] searched for theJAK2 V617F mutation in a cohort of 44 female ET patientswith previously documented clonality data (X-chromosomeinactivation pattern: XCIP assay). The mutation was found infour of the nine patients with polyclonal hematopoiesis (prob-ably due to better sensitivity of mutation detection than XCIPassays), in 13 of the 31 patients with monoclonal hematopoi-

mass and secondary myelofibrosis in JAK2 V617F positive MPDs ET, PV, MF

tients is detectable in platelets, erythroblast and granulocytesStep 12 V617F++

LOH

Myeloid Metaplasia: MMLeukocyte activationPVR-1 = LAF

.45-0.50: PV Leukocythemia/cytokinesFatigue, splenomegaly, unclassified MPD,MMM

Clinical Step 3 Secondary MF: 30%

Thrombosis ThrombosisConstitutional symptoms

IFN/Hydroxyurea

an and Thiele.

[7]

[8]

[9]

[10

[11

[12

[13

[14

[15

[16

[17

[18

[19

[20

[21

[22

[23

[24

[25

gie Bioesis and in two of the four patients with undetermined clonalitystatus due to non-random XCIP. These results show that themajority of patients with ET according to the PVSG withoutV617F JAK2 mutation have nevertheless a clonal hematopoie-tic stem cell disease. Combining results of JAK2 sequencingand XCIP assays allowed demonstration of a clonal prolifera-tion in 90% of ET cases. Buck et al. [108] observed a lowerincidence of 25% JAK2 V617F positivity in patients with trueET according to the WHO and ECP criteria (personal commu-nication). In a large prospective study of 806 ET patients diag-nosed according to the PVSG criteria, about half of the ETpatients (53%) were JAK2 V617F positive. ET patients posi-tive for the JAK2 V617F mutation have higher hemoglobinlevel and a significantly higher rate of transformation into PVduring follow-up as compared to ET patients negative for theJAK2 V617F mutation [108]. Comparing the laboratory fea-tures of JAK2 V617 positive ET and JAK2 V617 negativeET patients in the PT-1 study clearly showed that JAK2V617 positive ET is featured by higher values for hemoglobin,hematocrit, neutrophil counts, lower values for serum EPOlevels, serum ferritine and MCV, and hypercellularity of thebone marrow in biopsy material indicating that JAK2 V617positive ET patients diagnosed according to the PVSG criteriarepresent a forme fruste of PV [108] consistent with early PVmimicking ET (ECP stage 1 in Table 6). In contrast, the JAK2negative ET patients had significantly higher platelet countsand showed a clinical picture of true ET with normal serumEPO levels and increased of clustered megakaryocytes in anormocellular bone marrow consistent with a diagnosis oftrue ET (Table 1). These clinical data indicate that JAK2V617F mutation defines one disease with several sequentialsteps of ET, PV and MF during long-term follow-up (Table 9).Patients with either JAK2 V617F positive or negative prefibro-tic and fibrotic CIMF may represent two distinct entities with arelated etiology for one and the same disorder. Bone marrowhistopathology when used in combination with specific mar-kers like JAK2 V617F, serum EPO, PRV-1, EEC, peripheralblood parameters has a high sensitivity and specificity (almost100%) to detect each of the early stages of MPD and to differ-entiate between ET, PV and CIMF in both JAK2 V617F posi-tive and negative MPDs.

References

[1] Dameshek W. Physiopathology and course of polycythemia vera asrelated to therapy. J Am Ass Med 1950;142:7907.

[2] James C, Ugo V, Casadevall N, Constantinescu SN, Vainchenker WA.JAK2 mutation in myeloproliferative disorders: pathogenesis and thera-peutic and scientific prospects. Trends Mol Med 2005 (On line).

[3] Wasserman LR. Polycthemia vera, its course and treatment: relation tomyeloid metaplasia and leukemia. Bull N Y Acad Med 1954;30:34375.

[4] Wasserman LR, Berk PD. Berlin NI. Polycythemia vera and the myelo-proliferative disorders. Philadelphia: WB Saunders; 1995 (ISBN 0-7216-4213-6).

[5] Pearson TC, Wetherley-Mein G. The course and complications of idio-

J.J. Michiels et al. / Patholo102pathic erythrocytosis. Clin Lab Haematol 1979;1:18996.[6] Najean Y, Triebel F, Dresch C. Pure erythrocytosis: reappraisal of a

study of 51 patients. Am J Hematol 1981;10:12936.[26Glaser RM, Walker RI. Transitions among the myeloproliferative disor-ders. Ann Intern Med 1969;71:285307.Burkhardt R., Bartl R., Jaeger K., Frisch B., Kettner G., Mahl G., SundM. Chronic myeloproliferative disorders. Path Res Pract 1084;179:131-186.Georgii A, Vykoupil KF, Thiele J. Chronic megakaryocytic granulocy-tic myelosis: CMGM. Virch Arch A Path Anat Histol 1980;389:25368.

] Georgii A, Vykoupil KF, Buhr T, Choritz H, Doehler U, Kaloutsi V,Werner M. Chronic myeloproliferative disorders in bone marrow biop-sies. Path Res Pract 1990;186:327.

] Georgii A, Buhr T, Buesche G, Kreft A, Choritz H. Classification andstaging of Ph-negative myeloproliferative disorders by histopathologyfrom bone marrow biopsies. Leuk Lymphoma 1996;22(Suppl 1):1529.

] Georgii A, Buesche G, Kreft A. The histopathology of chronic myelo-proliferative diseases. Bailires Clin Haematol 1998;11:72149.

] Thiele J, Ballard AC, Georgii A, Vykoupil KF. Chronic megakaryocyticgranulocytic myelosis: An electron microscopic study. Megakaryocytesand thrombocytes. Virch Arch 1977;373:191211 (Path Anat Histol).

] Thiele J, Holgado S, Choritz H, Georgii A. Density distribution and sizeof megakaryocytes in inflammatory reactions of the bone marrow (mye-litis) and chronic myeloproliferative disorders. Scand J Haematol 1983;31:32941.

] Thiele J, Zankovich R, Schneider G, Kremer B, Fischer R, Diehl V.Primary (essential) thrombocythemia versus polycythemia vera rubra.A histomorphometric analysis of bone marrow features in trephine biop-sies. Analyt Quat Cytol Histol 1988;10:37582.

] Thiele J, Zankovich R, Steinberg T, Kremer B, Fischer R, Diehl V. Pri-mary (essential) thrombocythemia versus hyperplastic stages of agno-genic myeloid metaplasia with thrombocytosis: a critical evaluation ofclinical and histomorphological data. Acta Haematol 1989;81:192202.

] Thiele J, Wagner S, Degel C, et al. Megakaryocyte precursors (pro- andmegakaryoblasts) in bone marrow tissue from patients with reactivethrombocytosis, polycythemia vera and primary (essential) thrombo-cythemia. Virch Arch 1990;58:295302 (Cell Pathol).

] Thiele J, Kvasnicka HM, Werden C, Zankovich R, Diehl V, Fischer R.Idiopathic primary osteo-myelofibrosis: A clinico-pathological study on208 patients wit special emphasis on evolution of disease features, dif-ferentiation from essential thrombocythemia and variables of prognosticimpact. Leuk Lymphoma 1996;22:30317.

] Michiels JJ. The myeloproliferative disorders. An historical appraisaland personal experiences. Leuk Lymphoma 1996;22(Suppl 1):114.

] Michiels JJ. Diagnostic criteria of the myeloproliferative disorders(MPD) essential thrombocythemia, polycythemia vera and chronicmegakaryocytic granulocytic metaplasia. Neth J Med 1997;51:5764.

] Michiels JJ, Juvonen E. Proposal for revised diagnostic criteria ofessential thrombocythemia and polycythemia vera by the Thrombo-cythemia Vera Study Group. Semin Thromb Hemostas 1997;23:33947.

] Thiele J, Kvasnicka HM, Fischer R. Histochemistry and morphometryon bone marrow biopsies in chronic myeloproliferative disorders: aidsto diagnosis and classification. Ann Hematol 1999;78:495506.

] Thiele J, Kvasnicka HM, Diehl V, Fischer R, Michiels JJ. Clinicopatho-logical diagnosis and differential criteria of thrombocythemias in var-ious myeloproliferative disorders by histopathology, histochemistryand immunostaining from bone marrow biopsies. Leuk Lymphoma1999;33:20718.

] WHO classification of the chronic myeloproliferative diseases (CMPD)polycythemia vera, chronic idiopathic myelofibrosis, essential thrombo-cythemia and CMPD unclassifiable. WHO Classification of Tumours.Tumours of Haemtopoiesis and Lymphoid Tissues. Lyon: IARC;2001. p. 3142.

] Michiels JJ, Thiele J. Clinical and pathological criteria for the diagnosisof essential thrombocythemia, polycythemia vera and idiopathic myelo-fibrosis (agnogenic myeloid metaplasia). Int J Hematol 2002;76:133

logie 55 (2007) 9210445.] Michiels JJ. Bone marrow histopathology and biological markers as

specific clues to the differential diagnosis of essential thrombocythemia,

[27

[28

[29

[30

[31

[32

[33

[34

[35

[36

[37

[38

[39

[40

[41

[42

[43

[44

[45

[46

[47

[48

[49

[50

[51

[52

[53

[54

[55

[56

[57

[58

[59

[60

[61

[62

[63

[64

[65

[66

[67

gie Biopolycythemia vera and prefibrotic or fibrotic myeloid metaplasia.Hematol J 2004;5:93102.

] Thiele J, Kvasnicka HM, Boeltken B, Zankovich R, Diehl V, Fischer R.Initial (prefibrotic) stages of idiopathic (primary) myelofibrosis (IMF): aclinicopathological study. Leukemia 1999;13:174117428.

] Thiele J, Kvasnika HM, Zankovich R, Diehl V. Relevance of bone mar-row features in the differential diagnosis between essential thrombo-cythemia and early stage idiopathic myelofibrosis. Haematlogica 2000;85:112634.

] Thiele J, Kvasnicka HM, Zankovich R, Diehl V. Clinical and morpho-logical criteria for the diagnosis of prefibrotic idiopathic (primary) mye-lofibrosis. Ann Hematol 2001;80:1605.

] Thiele J, Kvasnicka HM, Zankovich R, Diehl V. Early stage idiopathic(primary) myelofibrosis: current issues of diagnostic features. LeukLymphoma 2002;43:103641.

] Thiele J, Kvasnicka HM. Diagnostic differentiation of essential throm-bocythemia from thrombocythemia associated with chronic idiopathicmyelofibrosis by discriminate analysis of bone marrow features. HistolHistopathol 2003;18:93102.

] Thiele J, Kvasnicka HM, Schmitt-Graeff A, Diehl V. Dynamics offibrosis in chronic idiopathic (primary) myelofibrosis during therapy: afollow-up on 308 patients. Leuk Lymphoma 2003;44:94953.

] Thiele J, Kvasnicka HM. Chronic myeloproliferative disorders withthrombocythemia: a comparative study of two classifications systems(PVSG-WHO) on 839 patients. Ann Hematol 2003;82:14852.

] Thiele J, Kvasnicka HM. Prefibrotic chronic myelofibrosis a diagnos-tic enigma? Acta Haematol 2004;111:1559.

] Thiele J, Kvasnicka HM, Diehl V. Standadization of bone marrow fea-turesdoes it work in hematopathology for histological discriminationof different disease patterns? Histopathol 2005;20:63344.

] Thiele J, Kvasnicka HM, Orazi A. Bone marrow histopathology inmyeloproliferative disorderscurrent diagnostic approacj. SeminHematol 2005;42:18495.

] Laszlo J. Myeloproliferative disorders (MPD): myelofibrosis, myelo-sclerosis, extramedullary hematopoiesis, undifferentiated MPD andhemorrhagic thrombocythemia. Semin Hematol 1975;12:40932.

] Murphy S, Iland H, Rosenthal D, Laszlo J. Essential thrombocythemia:an interim report from the Polycythemia Vera Study Group. SeminHematol 1986;23:17782.

] Berlin NI. Diagnosis and classification of the polycythemias. SemHematol 1975;12:33951.

] Thiele J, Kvasnicka HM, Zankovich R, Diehl V. The value of bonemarrow histopathology for the differentiation between early stage poly-cythemia vera and secondary (reactive) polycythemias. Haematologica2001;86:36874.

] Thiele J, Kvasnicka HM, Muehlhausen K, Walter S, Zankovich R,Diehl V. Polycythemia rubra vera versus secondary polycythemias. Aclinicopathological evaluation of distinctive features in 199 patients.Pathol Res Pract 2001;197:7784.

] Thiele J, Kvasnicka HM, Diehl V. Bone marrow features of diagnosticimpact in erythrocytosis. Ann Haematol 2005;84:3627.

] Thiele J, Kvasnicka HM, Diehl V. Initial (latent) polycythemia verawith thrombocytosis mimicking essential thrombocythemia. Acta Hae-matol 2005;113:2139.

] Thiele J, Kvasnicka HM. Diagnostic impact of bone marrow histo-pathology in polycythemia vera (PV). Histol Histpathol 2005;20:31728.

] Tefferi A. Polycythemia vera: a comprehensive review and clinicalrecommendations. Mayo Clin Proc 2003;78:17494.

] McMullin MF, Bareford D, Campell P, Green AR, et al. Guidelines forthe diagnosis, investigation and management of polycythaemia/erythro-cytosis. Br J Haematol 2005;130:17495.

] De Stefano V, Teofili L, Leone G, Michiels JJ. Spontaneous erythroid

J.J. Michiels et al. / Patholocolony formation as the clue to an underlying myeloproliferative disor-der in Budd-Chiari syndrome or portal vein thrombosis. Semin ThrombHemostas 1997;23:4118.] Chait Y, Condat B, Cazals-Hatem D, et al. Relevance of the criteriacommonly used to diagnose myeloproliferative disorders in patientswith splanchnic vein thrombosis. Br J Haematol 2005;129:55360.

] Liu E, Jelinek J, Pastore YD, Guan Y, Prchal JF, Prchal JT. Discrimina-tion of polycythemias and thrombocythemias by novel, simple accurateclonality assays and comparison with PRV-1 expression and BFU:Eresponse to erythropoietin. Blood 2003;101:3294301.

] Shih L-Y, Lee C-T. Identification of masked polycythemia vera frompatients with idiopathic marked thrombocytosis by endogenous ery-throid colony assay. Blood 1994;83:7448.

] Westwood NB, Pearson TC. Diagnostic applications of haematopoieticprogenitor culture techniques in polycythaemias and thrombocythae-mias. Leuk Lymphoma 1996;22:95103.

] Dobo I, Donnard M, Giridon F, Mossuz P, et al. Standardization andcomparison of endogenous erythroid colony assays performed withbone marrow or blood progenitors for the diagnosis of polycythemiavera. Hematol J 2004;5:1617.

] Florensa L, Besses C, Zamora L, Bellosillo B, Espinet B, Serrano S,et al. Endogenous erythroid and megakaryocytic circulating progenitors,HUMARA clonality assay, and PRV-1 expression are useful tools fordiagnosis of polycythemia vera and essential thrombocythemia. Blood2004;103:24278.

] Messinezy M, Westwood NB, El-Hemaida I, Marsden JT, Sher-wood RS, Pearson TC. Serum erythropoietin values in erythrocytosisand in primary thrombocythaemia. Br J Haematol 2002;117:4753.

] Cotes PM, Dore CJ, Tin JA, Lewis SM, Messinezy M, Pearson TC,et al. Determination of serum immunoreactive erythropoietin in theinvestigation of erythrocytosis. N Engl J Med 1986;315:2837.

] Birgegard G, Wide L. Serum erythropoietin in the diagnosis of poly-cythemia and after phlebotomy treatment. Br J Haematol 1992;81:6036.

] Messinezy M, Westwood NB, Woodstock SP, Strong RM, Pearson TC.Low serum erythropoietin: a strong diagnostic criterion of primary poly-cythaemia even at normal haemoglobin levels. Clin Lab Haematol1995;17:21720.

] Carneskog J, Kutti J, Wadenvik H, Lundberg PA, Lindstedt G. Plasmaerythropoietin by high-detectability immunoradiometric assay inuntreated patients with polycythemia vera and essential thrombocythe-mia. Eur J Haematol 1998;60:27882.

] Griesshammer M, Kubanek B, Beneke H, Heimpel H, Bangerter M,Bergmann L, Shrezenmeier H. Serum erythropoietin and thrombopoie-tin levels in patients with essential thrombocythemia. Leuk Lymphoma2000;36:5338.

] Mossuz P, Giridon F, Latger-Cannoard V, Dobo I, et al. In: Diagnosticvalue of serum erythropoietin level in patients with absolute erythrocy-tosis. Haematlogica; 2004. p. 11948.

] Temerinac S, Klippel S, Strunck E, Rder S, Lbbert M, Lange M,et al. Cloning of PRV-1, a novel member of the uPAR receptor super-family, which is over expressed in polycythemia rubra vera. Blood2000;95:256976.

] Pahl HL. Towards a molecular understanding of polycythemia rubravera. Eur J Biochem 2000;267:3395401.

] Klippel S, Strunck E, Termerinac S, et al. Quantification of PRV-1expression, a molecular marker for the diagnosis of polycythemia vera.Blood 2001;98:470a.

] Pahl HL. Polycythaemia vera: will new markers help us answer oldquestions? Acta Haematol 2002;108:12031.

] Johanson P, Andreason B, Safai-Kutti S, Wennstrom L, Palmqvist L,Rickson A, et al. The presence of a significant association between ele-vated PRV-1 mRNA expression and low plasma erythropoietin concen-tration in essential thrombocythemia. Eur J Haematol 2003;70:35862.

] Griesshammer M, Klippel S, Strunk E, Temeric S, Mohr U, Heimpel H,et al. PRV-1 mRNA expression discriminates two types of essentialthrombocythemia. Ann Hematol 2004;83:36470.

] Juvonen E, Ikkala E, Oksanen K, Ruutu T. Megakaryocyte and ery-

logie 55 (2007) 92104 103throid colony formation in essential thrombocythemia and reactivethrombocytosis: diagnostic value and correlation to complications. Br JHaematol 1993;83:1927.

[69

[70

[71

[72

[73

[74

[75

[76

[77

[78

[79

[80

[81

[82

[83

[84

[85

[86

[87

[88

[90

[91

[92

[93

[94

[95

[96

[97

[98

[99

[10

[10

[10

[10

[10

[10

[10

[10

[10genous megakaryocyte and erythroid colony formation from blood inessential thrombocythemia. Leukemia 1995;9:2713.

] Jantunen R, Juvonen E, Ikkala E, Oksanen K, Anttila P, Jormila P, et al.Essential thrombocythemia at diagnosis: causes of diagnostic evaluationand presence of positive diagnostic findings. Ann Hematol 1998;77:1016.

] Reilly JT. Pathogenesis of idiopathic myelofibrosis: present status andfuture directions. Br J Haematol 1994;88:18.

] Tefferi A, Silverstein MN, Nol P. Agnogenic myeloid metaplasia.Semin Hematol 1995;22:32733.

] Barosi G, Ambrosetti A, Finelli, et al. The Italian consensus conferenceon diagnostic criteria for myelofibrosis with myeloid metaplasia. Br JHaematol 1999;104:7307.

] Barosi G. Myelofibrosis with myeloid metaplasia: diagnostic definitionand prognostic classification for clinical studies and treatment guide-lines. J Clin Oncol 1999;17:295470.

] Tefferi A. Myelofibrosis with myeloid metaplasia. N Engl J Med 2000;342:125565.

] Dingli D, Mesa RA, Tefferi A. Myelofibrosis with myeloid metaplasia:new developments in pathogenesis and treatment. Intern Med 2004;43:5407.

] Thiele J, Kvasnicka HM, Diehl V. Bone marrow CD34+ progenitorcells in Philadelphia chromosome-negative chronic myeloproliferativedisorders: a clinicopathological study on 575 patients. Leuk Lymphoma2005;46:70915.

] Florena AM, Tripodi C, Iannitto E, Porcasi R, Ingrao S, Franco V.Value of bone marrow biopsy in the diagnosis of essential thrombo-cythemia. Haematologica 2004;89:9119.

] Bauermeister DE. Quantification of bone marrow reticulin. Am J ClinPathol 1971;56:2431.

] Manoharan A, Smart RC, Pitney WR. Prognostic factors in myelofibro-sis. Pathology 1982;14:44561.

] Thiele J, Kvasnicka HM, Facchetti F, Franco V, Van Der Walt J, OraziA. European consensus for grading of bone marrow fibrosis and assess-ment of cellularity in myeloproliferative disorders. Haematologica 2005;90:112832.

] Buhr T, Georgi A, Choritz H. Myelofibrosis in chronic myeloprolifera-tive disorders. Incidence among subtypes to the Hannover Classifica-tion. Patol Res Pract 1993;189:12132.

] Thiele J, Kvasnicka HM, Schmitt-Graeff A, Zankovich R, Diehl V.Follow-up examinations including sequential bone marrow biopsies inessential thrombocythemia (ET): a retrospective clinicopathologicalstudy of 120 patients. Am J Hematol 2002;70:28391.

] Kreft A, Buesche G, Ghalibafian M, Buhr T, Fischer T, Kirkpatrick CJ.The incidence of myelofibrosis in essential thrombocythemia, poly-cythemia vera and chronic idiopathicmyelofibrosis: a retrospective eva-luation of sequential bone marrow biopsies. Acta Haematol 2005;113:13743.

] Thiele J, Kvanicka HM, Schmitt-Graeff A, Diehl V. Dynamics of fibro-sis in chronic idiopathic (primary) myelofibrosis during therapy: afollow-up study on 309 patients. Leuk Lymphoma 2003;44:94953.

] Buhr T, Buesche G, Choritz H, Langer F, Kreipe H. Evolution of mye-lofibrosis in chronic idiopathic myelofibrosis as evidenced in sequentialbone marrow biopsy specimens. Am J Clin Pathol 2003;119:1528.

] Thiele J, Kvasnicka HM. Hematopathologic findings in chronic idio-pathic myelofibrosis. Semin Oncol 2005;32:38094.

] Michiels JJ, Kvasnicka HM, Thiele J. Myeloproliferative Disorders:current perspectives on diagnostic criteria, histopathology and treat-ment. Munich Germany: Verlag ME Uwe Grunwald; 2005 (ISBN 3-9898075-6-8).

] Dupriez B, Morel P, Demory JL, Lai JL, Simon M, Plantier I, et al.Prognostic factors in agnogenic myeloid metaplasia: a report on 195cases with a bew scoring syste. Blod 1996;88:10138.management. Blood Rev 1997;11:23342.] Michiels JJ. Clinical, pathological and molecular features of myelopro-

liferative disorders: MPD 2005 and beyond. Hematology (Am SocHematol Educ Program) 2005;10(Suppl 1):21523.

] James C, Ugo V, Le Couedic PF, Staerk J, Delhommeau F, Lacout C,et al. A unique clonal JAK2 mutation leading to constitutive signallingcauses polycythemia vera. Nature 2005;434:11448.

] Levine RL, Wadleigh M, Cools J, Ebert BL, Wernig G, Huntly BJ,et al. Activating mutation in the tyrosine kinase JAK2 in polycythemiavera, essential thrombocythemia and myeloid metaplasia with myelofi-brosis. Cancer Cells 2005;7:38797.

] Kralovics R, Passamonti F, Buser AS, Teo SS, Tiedt R, Passweg JR,et al. A gain-of-function mutation of JAK2 in myeloproliferative disor-ders. N Engl J Med 2005;352:177990.

] Baxter EJ, Scott LM, Campbell PJ, East C, Fourouclas N, Swanton S,et al. Acquired mutation of the tyrosine kinase in human myeloproli-ferative disorders. Lancet 2005;365:105461.

] Zhao R, Xing S, Li Z, Fu X, Li Q, Krantz AS, et al. Identification of anacquired JAK2 mutation in polycythemia vera. J Biol Chem 2005;29.

] Kaushanky K. On the origin of the chronic myeloproliferative disorders:it makes all sense. Blood 2005;105:418790.

] Goldman JM. A unifying mutation in chronic myeloproliferative disor-ders. N Engl J Med 2005;352:17445.

] Cazzola M, Skoda R. Gain of function, loss of control. A molecularbasis for chronic myeloproliferative disorders. Haematologica HematolJ 2005;90:8714.

] Jones AV, Kreil S, Xoi K, Waghorn K, Curtis C, Zhang L, et al. Wide-spread occurrence of the JAK2 V617F mutation in chronic myeloproli-ferative disorders. Blood 2005;106:21628.

0] Wolanskyj AP, Lasho TL, Schwager SM, McClure RF, Wadleigh M,Lee SJ, et al. JAK2 V617 mutation in essential thrombocythaemia: clin-ical associations and long-term relevance. Br J Haematol 2005;131:20813.

1] Antonioli E, Guglielmelli P, Pancrazzi A, Bogani C, Verrucci M, Pon-ziani V, et al. Clinical implications of the JAK2 V617F mutation inessential thrombocythemia. Leukemia 2005;19:18479.

2] Goerttler PS, Steimle C, Maerz E, Johanson PL, Andreasson B, Gries-shammer M, et al. The JAK2 V617F mutation, PRV-1 over expressionand EEC formation define a similar cohort of MPD patients. Blood2005;106:28624.

3] Kralovics R, Teo SS, Buser AS, et al. Altered gene expression in mye-loproliferative disorders correlates with activation of signalling by theV617F mutation of JAK2. Blood 2005;106:33746.

4] Levine RL, Loriaux M, Huntly BJP, Loh ML, et al. The JAK2 V617Factivating mutation occurs in chronic myelomonocytic leukemia andacute myeloid leukemia, but not in acute lymphoblastic leukemia orchronic lymphocytic leukemia. Blood 2005;106:33779.

5] Steensma DP, Dewald M, Lasho TL, Powell HL, McClure RF. LevineRL et al. The JAK2 V617F activating tyrosine kinase mutation is aninfrequent event in both atypical myeloproliferative disorders and themyelodysplastic syndrome. Blood 2005;106:12079.

6] Jelinek J, Oki Y, Gharibyan V, Bueso-Ramos C, Prchal JT, VerstovsekS, et al. JAK2 mutation 1849G >T is rare in acute leukemias but can befound in CMML, Philadelphia-chromosome negative CML and mega-karyocytic leukemia. Blood 2005 [Epub ahead of print].

7] Kiladjian JJ, Elkassar N, Hetet G, Balitrand N et al. Analysis of JAK2mutation in essential thrombocythemia (ET) patients with monoclonaland polyclonal X-chromosome inactivation patterns (XCIPs). Blood(Suppl Abstract 807-II).

8] Campbell P, Scott LM, Buck G. Definition of subtypes of essentialthrombocythaemia and relation to polycythaemia vera based on JAK2V617F mutation status: a prospective study. Lancet 2005;366(9501):194553.[68] Florensa L, Bessis C, Woessner S, Sole F, Acin P, Pedro C, et al. Endo- [89] Reilly JT. Idiopathic myelofibrosis: pathogenesis, natural history and

J.J. Michiels et al. / Pathologie Biologie 55 (2007) 92104104

Current diagnostic criteria for the chronic myeloproliferative disorders (MPD) essential thrombocythemia (ET), polycythe..IntroductionThe concept of PV as a trilinear MPDClinical, laboratory and pathological features of the MPDsWHO pathological criteria for the diagnosis of ET, PV and CIMFCurrent clinical and pathological criteria for the diagnoses of ET PV and CIMFMyelofibrosis (MF) versus chronic idiopathic myelofibrosis (CIMF)The molecular etiology of the MPDs ET PV and MFReferences