JAK2 V617F Mutation Status of 232 Patients Diagnosed With Chronic Myeloproliferative Neoplasms

Original Study

JAK2 V617F Mutation Status of 232 PatientsDiagnosed With Chronic

Myeloproliferative NeoplasmsKadriye Bahriye Payzin,1 Kaan Savasoglu,2 Inci Alacacioglu,3

Fusun Ozdemirkiran,1 Belgin Berber Mutlu,2 Sadi Bener,4

Aylin Orgen Calli,4 Betul Bolat Kucukzeybek,4 Saliha Haksun5

AbstractThe aim of this study was to investigate the presence of Janus kinase 2 (JAK2) V617F mutation in patientswith BCR-ABL negative chronic myeloproliferative neoplasms (CMPNs) in our center. JAK2 V617F mutationfrequencies in our PV and ET patients were similar to those reported previously. JAK2 V617F mutation fre-quency in our PMF patients was greater than in previous reports.Introduction/Background: The aim of this study was to investigate the presence of Janus kinase 2 (JAK2) V617Fmutation in patients with break point cluster region-abelson negative chronic myeloproliferative neoplasms(CMPNs) in our center. Patients and Methods: We compared patients with and without the mutation, and alsopatients with the homozygous and heterozygous mutation, in terms of different clinical and laboratory features.Results: The JAK2 V617F mutation was detected in 77 (95%), 88 (68%), and 17 (77%) of polycythemia vera (PV),essential thrombocythemia (ET), and primary myelofibrosis (PMF) patients, respectively. Among JAK2 V617F-positive patients, the homozygous genotype was found in 39 (50.6%) of the 77 PV, 23 (26.1%) of the 88 ET,and 11 (64.7%) of the 17 PMF patients. Bleeding was seen in 14 (6%) of all patients. Upper gastrointestinal bleedswere the most common, seen in 11 patients. Out of 232 CMPN patients, 44 (19%) had thrombosis. The mostcommon thrombotic event was transient ischemic attack (52%). Progression to myelofibrosis was seen in 1 (1.2%)PV and 3 (2.3%) ET patients, and progression to acute leukemia was seen in 2 (2.5%) PV and 3 (2.3%) ET patients.Three patients with PV (3.7%), 3 with ET (2.7%), and 5 with PMF (2.7%) died during follow-up. Conclusion: JAK2V617F mutation frequencies in our PV and ET patients were similar to those reported previously. JAK2 V617F mu-tation frequency in our PMF patients was greater than in previous reports. All of our PV patients with thrombosis andmost of our ET patients with thrombosis (76.1%) were JAK2 V617F mutation-positive. This mutation seems to becorrelated with thrombosis risk.

Clinical Lymphoma, Myeloma & Leukemia, Vol. -, No. -, --- ª 2014 Elsevier Inc. All rights reserved.Keywords: Essential thrombocythemia, JAK2 mutation, MPNs, Polycythemia vera, Primary myelofibrosis

1Division of Hematology, Department of Internal Medicine, Ataturk Training Hos-pital, Izmir, Turkey2Medical Genetics Laboratory, Ataturk Training Hospital, Izmir, Turkey3Department of Hematology, Dokuz Eylul University Faculty of Medicine, Izmir,Turkey4Department of Pathology, Ataturk Training Hospital, Izmir, Turkey5Department of Biochemistry, Ataturk Training Hospital, Izmir, Turkey

Submitted: Jan 28, 2014; Revised: Feb 19, 2014; Accepted: Feb 24, 2014

Address for correspondence: Kadriye Bahriye Payzin, MD, Division of Hematology,Department of Internal Medicine, Ataturk Training Hospital, Izmir, TurkeyE-mail contact: [email protected]

2152-2650/$ - see frontmatter ª 2014 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.clml.2014.02.013

IntroductionPolycythemia vera (PV), essential thrombocythemia (ET), and

primary myelofibrosis (PMF) are classified in the break point clusterregion-abelson (BCR-ABL1) negative myeloproliferative neoplasm(MPN) group according to World Health Organization (WHO)2008 classification.1 Their disease-causing mutations remain un-identified despite many mutations described beginning in 2005. PVpatients mostly harbor Janus kinase 2 (JAK2; 9p24) mutation (96%displaying mutation in exon 14, JAK2 V617F; 3% displayingmutation in exon 12; JAK2).2,3

Clinical Lymphoma, Myeloma & Leukemia Month 2014 - 1

mailto:[email protected]

http://dx.doi.org/10.1016/j.clml.2014.02.013

JAK2 V617F Mutation Status in CMPN

2 - Cli

Janus kinase 2 V617F also occurs in ET (55%) and PMF (65%),but JAK2 exon 12 mutations are rare in ET and PMF.4 Myelo-proliferative leukemia virus oncogene (MPL) (MPL515W>L/K)mutations are found in approximately 4% of ET, 8% of PMF, andrarely in PV.5 According to the 2008 WHO classification, the ex-istence of these mutations is the major criterion in diagnosis ofchronic MPNs.1 Nonetheless, evaluation of bone marrow histopa-thology is still required to define the specific morphological patterns,which are essential for the diagnosis of ET and PMF. Other mu-tations related to the tet oncogene family member 2 (TET2),additional sex comb-like 1 (ASXL1), and isocitrate dehydrogenase(IDH1/2) genes are occasionally observed in MPNs.6 Because themolecular pathogenesis of MPNs with nonmutated JAK2 is stillobscure and diagnosis remains a challenge, the mutation status ofthe gene encoding calreticulin (CALR) was studied recently byNangalia et al.7 CALR mutations were found in most patients withMPNs with nonmutated JAK2.7

Leukocytosis, splenomegaly, thrombohemorrhagic complica-tions, vasomotor disturbances, pruritus, and a small risk of diseaseprogression into acute myeloid leukemia or myelofibrosis are otherdisease features.4 In previous studies, the importance of mutationalstatus of JAK2 was evaluated.

Correlations have been found between high JAK2 V617F alleleburden and leukocytosis, spleen size, myelofibrosis severity,fibrotic transformation, and disease duration.8 Many researchershave reported thrombosis risk in JAK2 V617F-positive ET patientsto be greater than that of JAK2 V617F-negative ET patients.9-11

The effect of the JAK2 V617F mutation on clinical parametersof PMF such as prognosis and the need for transfusion is un-clear.12,13 JAK2 V617F mutated status has not been shown to havean effect on survival time, but patients with low JAK2 V617F alleleburden have been reported to have shorter survival times comparedwith those with high JAK2 V617F allele burden and wild typeJAK2.14

In this study, we aimed to investigate the relationship betweenJAK2 status/quantitation and clinic and laboratory findings in ourMPNs patients.

Table 1 Characteristics of 232 Patients With MPNs

MPN

M/F 109/123

Mean Age (Minimum-Maximum), Years 59.9 (18-89)

Splenomegaly, n (%) 125 (59.8)

Thrombosis, n (%) 44 (19)

Bleeding, n (%) 14 (6)

Diabetes Mellitus, n (%) 29 (12.5)

Hypertension, n (%) 42 (18.1)

Mean Hemoglobin Level ± SD, g/dL 14.6 � 3.0

Mean Hematocrit Level ± SD, % 13.6 � 10.3

Mean Leukocyte Count ± SD, 3 109/L 15.2 � 8.0

Mean Thrombocyte Count ± SD, 3 109/L 344.7 � 22.9

Mean Serum LDH ± SD, U/L 313.7 � 219

Abbreviations: ET ¼ essential thrombocythemia; F ¼ female; LDH ¼ lactic dehydrogenase; M ¼ mpolycythemia vera.

nical Lymphoma, Myeloma & Leukemia Month 2014

Patients and MethodsDemographic and clinical features of 233 patients diagnosed

with MPNs at our hematology department between November2000 and November 2012 were evaluated retrospectively. Themedian age was 63 years (range, 18 to 89 years) with a male tofemale ratio of 0.9 to 1 (110 men, 123 women). BCR-ABL mo-lecular analysis and fluorescence in situ hybridization analysis fort(9;22) were made in all MPN patients and found as not amplified.During retrospective evaluation, 1 male patient was excluded fromthe study because of diagnosis of chronic myelomonocytic leuke-mia. Eighty-one patients were rediagnosed with PV (34.9%), 129with ET (55.6%), and 22 with PMF (9.5%) according to poly-cythemia vera study group or WHO 2008 criteria. White bloodcell (WBC), hemoglobin (Hb), and hematocrit (htc) levels, andplatelet (plt) count, and serum lactic dehydrogenase (LDH) levelwere recorded. Spleen size was measured in centimeters below themidpoint of the left costal margin, and spleen size > 5 cm wasaccepted as gross splenomegaly. JAK2 V617F analysis was per-formed at the time or after the time of diagnosis depending on theavailability of this test. Thrombotic and hemorrhagic events beforeor after diagnosis were recorded. Clinical thrombotic eventsincluded ischemic stroke, cerebral transient ischemic attack (TIA),acute myocardial infarction (AMI), peripheral arterial thrombosis(PAT), pulmonary thromboembolism (PE), and venous throm-boembolism (VTE). Most patients received 1 or more of thefollowing: acetyl salicylic acid (ASA), some form of myelosup-pressive therapy as hydroxyurea (HU), anagrelide, interferon (IF),phosphorus-32 (P-32). Patients with PV were phlebotomized tomaintain an htc level � 45%. This study was approved by the localethics committee.

Quantification of JAK2 V617F mutation was detected usingan allele-specific real-time quantitative polymerase chain reaction(RQ-PCR) assay (JAK2 MutaQuant Kit, Ipsogen). RQ-PCR wasperformed using a Rotor-Gene (Corbett Research). Patientswith � 50% mutational load and those with < 50% mutationalload were considered ‘homozygotes’ and ‘heterozygotes,’ respec-tively, as has been previously designated.15,16

PV (81 Pts) ET (129 Pts) PMF (22 Pts)

30/51 65/64 14/8

62 (35-86) 58 (18-89) 62.8 (32-85)

46 (67.6) 57 (47.5) 22 (100)

23 (28.3) 21 (16.3) NR

4 (4.9) 10 (7.6) NR

12 (14.8) 16 (12.4) 1 (4.5)

15 (18.5) 25 (19.4) 2 (9)

17.6 � 1.8 13.26 � 2.0 11.3 � 1.9

53.6 � 5.9 39.8 � 8.3 33.6 � 5.5

15.2 � 5.6 13.7 � 5.9 23.9 � 16.6

646.9 � 268.4 971.0 � 319.9 344.6 � 73.5

297 � 115 271.9 � 110.5 648.1 � 556.2

ale; MPN ¼ myeloproliferative neoplasm; PMF ¼ primary myelofibrosis; pts ¼ patients; PV ¼

Kadriye Bahriye Payzin et al

Statistical AnalysisAll statistical analysis was performed using the SPSS 18.0 statis-

tical package. Analysis of variance test for continuous variables andc2 test for categorical variables as appropriate were used. Overallsurvival (OS) was defined as the time between diagnosis and death(from any causes) or end of follow-up (censored observations). OSwas estimated using the Kaplan-Meier product limit method. AllP values were 2-sided. A value of P < .05 was considered assignificant.

ResultsThe demographic, laboratory, and clinical data of all patients are

summarized in Table 1. There were significantly more female pa-tients than male patients in the PV group (P ¼ .01). In contrast,there were significantly more male patients than female patients(P ¼ .01) in the PMF group. Mean age was similar in all 3 groups.All patients with PMF had splenomegaly compared with others(P ¼ .01). Gross splenomegaly was observed in PV and ET patientswith similar frequency.

Out of 232 MPN patients, 182 (78.4%) were positive for theJAK2 V617F mutation. The distribution of allele burden ofmutation-positive patients is given in Table 2 and Figure 1. JAK2V617F mutations were found in 77 patients in the PV group(95%), 88 patients in the ET group (68%), and 17 patients in thePMF group (77%) (Table 2). The PV and ET groups showed nosignificant differences for JAK2 V617F mutation positivity amongthe 2 sexes (P¼ .8 for both). The JAK2 V617F mutation was foundsignificantly more frequently in male patients in the PMF group(P ¼ .049).

The heterozygous genotype was found significantly morefrequently in the ET group, whereas the homozygous genotype wasmore frequent in the PMF group (P ¼ .000 and 0.003, respec-tively). The percentage of PV patients with the homozygous ge-notype was significantly greater than the percentage of ET patientswith the same genotype (P ¼ .004) (Fig. 1). JAK2 V617F alleleburden status according to sex is shown in Figure 2.

In the PV group, there was no significant difference among maleand female patients in terms of the homozygous genotype (P ¼ .2).In the ET group the homozygous genotype was found significantlymore in JAK2 V617F-positive male patients, than in JAK2 V617F-positive female patients (P ¼ .04). In the PMF group, no significantdifference was found among the 2 sexes (P ¼ .06) (Fig. 2).

Mean plt count was statistically greater but Hb level was lower inPV patients with the JAK2 V617F mutation. When compared withET patients, mean spleen size, Hb and LDH level, and WBC count

Table 2 Distribition of Allele Burden of 182 (78.4%) MPN Patients W

JAK2 V617FAllele Burden 78%-100% 50%-78%

HomozygousGenotypea 31%-5

PV: 77 (95%) 12 (15.6) 27 (35) 39 (50.6) 21(27

ET: 88 (68%) 14 (16) 9 (10) 23 (26) 25 (28

PMF: 17 (77%) 5 (29.4) 6 (35.2) 11 (64.6) 6 (35

Data are presented as n (%).Abbreviations: ET ¼ essential thrombocythemia; MPN ¼ myeloproliferative neoplasm; PMF ¼ primaJAK2 V617F allele burden � 50%.bJAK2 V617F allele burden < 50%.

were found to be significantly greater in patients with the JAK2V617F mutation (P ¼ .006, .001, .01, and .001, respectively).When a similar comparison was made among PMF patients, meanage was older for the JAK2 V617F-positive group and mean spleensize was greater for the JAK2 V617F-negative group (P ¼ .04 forboth). There were no significant differences among the 2 groups interms of LDH and Hb levels, or WBC and plt counts (Table 3).

Myeloproliferative neoplasm patients homozygous and hetero-zygous for the JAK2 V617F mutation were also compared for theirHb and serum LDH levels, WBC and plt counts, and spleen size(Table 4). Mean spleen size, LDH level, and WBC count weresignificantly greater in the JAK2 V617F homozygous PV patients(P ¼ .006, .03 and .005, respectively). Among the ET patients,mean WBC count and spleen size were significantly greater in pa-tients homozygous for the JAK2 V617F mutation (P ¼ .001 and.001, respectively). In contrast, mean Hb level of the homozygousET patients was significantly lower compared with the heterozygousET patients (P ¼ .01). PMF patients with homozygous and het-erozygous JAK2 V617F mutations did not show significant differ-ences in age, spleen size, Hb and serum LDH levels, and WBC andplt counts (Table 4).

Among all MPN patients, 44 (19%) had thrombosis. Throm-bosis was not observed in any PMF patients. In the PV group, only2 of the 23 patients (28.3%) with thrombosis developed it after thediagnosis (1 had PE and VTE, the other had TIA). The remaining21 patients (28.4%) either had thrombosis for years, or werediagnosed with MPNs when searching for the cause of theirthromboses. TIA was the most common vascular event in thePV group (15 patients). Portal vein thrombosis was observed in3 patients, AMI in 2 patients, deep vein thrombosis (DVT) in 2patients (1 with a history of pulmonary thromboembolism), andhepatic vein thrombosis (Budd-Chiari syndrome) in 1 patient.Eleven PV patients with thrombosis were homozygous for the JAK2V617F mutation; the remaining patients were heterozygous. Ahomozygous JAK2 V617F mutation was found in 6 patients withTIA, all AMI and DVT patients, and 1 patient with portal veinthrombosis. There were no PV patients with thrombosis who testednegative for the JAK2 V617F mutation. When PV patients withand without thrombosis were compared, no significant differenceswere found in terms of age, spleen size, htc, Hb, and serum LDHlevels, and WBC and plt counts.

Twenty-one (16.3%) of the ET patients had thrombosis. Nine-teen patients had a history of thrombosis before their ET diagnoses,and 2 patients were diagnosed with ET along with clinical throm-bosis. Three of these patients were homozygous, 13 were

ho Were JAK2eV617F-Positive

0% 12.5%-31% 5%-12.5% 2%-5%HeterozygousGenotypeb

.4) 17 (22) 0 0 38 (49.4)

.4) 24 (27.2) 14 (15.9) 2 (2.5) 65 (74)

.4) 0 0 0 6 (27)

ary myelofibrosis; PV ¼ polycythemia vera.


Figure 1 The Percent Interval of JAK2 V617F Mutation Allelesare Shown on the Y-Axis; 1 Represents 2%-5%, 2Represents 5%-12.5%, 3 Represents 12.5%-31%, 4Represents 31%-50%, 5 Represents 50%-78%, and 6Represents 78%-100%. Patients With ‡ 50%Mutational Load and Those With < 50% MutationalLoad Were Homozygotes and Heterozygotes,Respectively. The Heterozygous Genotype was FoundSignificantly More Frequently in the ET Group, andthe Homozygous Genotype was More Frequent in thePMF Group (P [ .000 and .003, Respectively). ThePercentage of PV Patients With the HomozygousGenotype was Significantly Greater Than thePercentage of ET Patients With the Same Genotype(P [ .004)

Abbreviations: ET ¼ essential thrombocythemia; PMF ¼ primary myelofibrosis; PV ¼ poly-cythemia vera.


4 - Cli

heterozygous, and 5 were negative for the JAK2 V617F mutation. Avariety of thrombotic events were observed in ET patients. Startingfrom the most common was TIA, observed in 8 patients, PAT(2 patients with aortofemoral bypass) and AMI, both observed in5 patients, Budd-Chiari syndrome observed in 3 patients, and PEand portal vein thrombosis, both observed in 1 patient.

No significant differences were found among ET patients with orwithout thrombosis with respect to age, spleen size, htc, Hb, andserum LDH levels, and WBC and plt counts. There was no sig-nificant difference between PV and ET patients with regard tothrombosis frequency (P ¼ .07).

Bleeding was seen in 14 (6%) MPN patients. Upper gastroin-testinal bleeding was the most common, seen in 11 patients. Rectalor vaginal hemorrhage and epistaxis were also observed, each seen in1 patient. Only 3 of the 10 ET patients with bleeding were negativefor the JAK2 V617F mutation. Of the remaining 7, 3 were ho-mozygous and 4 were heterozygous for the JAK2 V617F mutation.Although 95% of our PV patients were positive for the JAK2V617F mutation, 1 of the 4 PV patients with bleeding were JAK2V617F-negative. Two of the remaining patients were homozygous,and the remaining 1 was heterozygous for the JAK2 V617Fmutation.

Diabetes mellitus (DM) was identified in 14.8%, 12.4%, and4.5% of the PV, ET, and PMF patients, respectively. DM wasobserved significantly more frequently in PV than in PMF patients


(P ¼ .02). No significant differences were found among differentdisease groups in terms of DM frequency. Essential hypertensionwas found in 18.5% of the PV patients, 19.4% in the ET patients,and 9% in the PV patients, with no significant differences amongthe 3 groups.

Considering the time from diagnosis to the determination of theJAK2 V617F mutation status, no significant differences were foundamong the number of patients tested within 6 months or after6 months of diagnosis in all MPN, PV, ET, and PMF patients(Fig. 3).

The cytoreductive treatment (HU, IF, anagrelide) and anti-aggregant agent were given according to risk classification5 andtreatments of all patients are summarized in Table 5. There was 1 PVpatient who received P-32 radioisotope treatment and is not listed inTable 5. Phlebotomy was practiced on PV patients when necessary.Patients receiving warfarin or clopidogrel either had venous/arterialthrombus or coronary angioplasty. Increased reticular fibrosis wasobserved in 1 patient who developed myelofibrosis (MF) secondaryto PV 52months after diagnosis and in 3 patients who developedMFsecondary to ET 98, 48, and 66 months after diagnosis according tobone marrow biopsy.

Eleven (4.7%) of the total 232 patients; 3 with PV (3.7%), 3with ET (2.7%), and 5 with PMF (2.7%) died for reasons sum-marized in Table 6. Acute leukemia was observed in 2 PV (2.5%)and 3 ET (1.6%) patients. One PV patient developed acute myeloidleukemia (AML) in the 52nd month of follow-up and was lost tofollow-up during chemotherapy. Another PV patient developedmyelodysplastic syndrome (MDS) (refractory anemia with excessblasts-2) 127 months after diagnosis and azacytidine therapy wasstarted. The patient progressed into AML and was lost to follow-upwith sepsis. Three patients diagnosed with ET progressed into AML13, 23, and 84 months after diagnosis and died despitechemotherapy.

The median follow-up of the 232 MPN patients was 28 months(range, 1 to 219 months). At the end of the follow-up period, 11patients (4.7%) had died. Estimated median OS could not bedetermined in the MPN, PV, or ET patient groups. Estimatedmean OS was 147.8 months in the MPN group (95% confidenceinterval [CI], 164.9-210.7), 191.3 months in the PV group (95%CI, 155.5-227.1), and 143 months in the ET group (95% CI,132.9-153.2). In the PMF group, estimated median and mean OSwas 62 months (estimation is limited because of the longest survivaltime being censored) and 79.6 months (95% CI, 58.4-100.7),respectively (Fig. 4).

DiscussionJanus kinase 2 V617F is the most frequent mutation in MPN

patients. In ET and PMF patients, the JAK2 V617F mutation hasbeen found in approximately 50%-60%, and the JAK2 V617Fmutation was detected in approximately 95% of PV patients.17 Inour study, JAK2 V617F mutation frequencies in our PV and ETpatients were similar to those previously reported. Because our PMFgroup was small for an accurate evaluation, we reasoned that thefrequency of the JAK2 V617F mutation in our PMF group wouldmatch those of larger studies if we had more PMF patients. Inaddition, we detected and quantified the JAK2 V617F mutation inmore than 25% of the PMF patients who were at the 36th month

Figure 2 In the PV Group, There was no Significant Difference Among Male and Female Patients in Terms of the HomozygousGenotype (P [ .2). Twelve (42.8%) of the 28 Male and 27 (55.1%) of the 49 Female PV Patients Positive for JAK2 WereHomozygous. In the ET Group the Homozygous Genotype was Found in 15 (33%) of 45 JAK2-Positive Male Patients, Whichwas Significantly Greater Than the 8 (18.6%) of 43 JAK2-Positive Female Patients With the Same Genotype (P [ .04). In thePMF Group, 4 of 5 (80%) Female and 7 of 12 (58.3%) Male Patients had the Homozygous Genotype. Although no SignificantDifference was Found Among the 2 Sexes (P [ .06), There Were More Female Than Male Patients With the HomozygousGenotype

Abbreviations: ET ¼ essential thrombocythemia; PMF ¼ primary myelofibrosis; PV ¼ polycythemia vera.


or more after their diagnosis. We were able to do that in only half ofthe PMF patients at the time of their diagnosis. The GIMEMAgroup has reported that heterozygous primary cases have progressedto a homozygous V617F status in the time from diagnosis to yearsafter it.18 This might be one of the reasons we found a high JAK2V617F allele burden in PMF.

Polycythemia vera is defined as a disease seen predominantlyin men,19 although there are PV groups reported to be pre-dominantly female.20,21 Our PV group was also predominantlyfemale (female to male ratio, 1.7-1). Female predominance in ETand male predominance in PMF has been reported in the liter-ature.22 In our patient groups, we had almost an equal number ofmale and female ET patients, whereas the PMF patients werepredominantly male.

In clinical studies, patients with � 50% JAK2 mutant alleleburden are considered homozygous for the mutation, whereas pa-tients with < 50% are considered heterozygous. Although evidenceof greater JAK2 V617F allele burden in PV than ET have beenreported in a few studies, data on allele burden in PMF remain to bevalidated.16,22-24 In our study, homozygous JAK2 V617F mutationwas also more frequently found in PV compared with ET patients(50.6% and 26.1%, respectively, P ¼ .004).

Patients with the JAK2 V617F mutation were shown to have atendency for higher Hb levels, higher WBC counts, and lower pltcounts.9,10 In a study conducted in multiple centers, JAK2 V617Fallele burden in PV patients was shown to be correlated withleukocytosis, greater htc values, larger spleen size, greater cell count,and fibrosis development.3,25 In a PV group of 105 patients, higherJAK2 V617F allele burden has been shown to be correlated withgreater splenomegaly and higher WBC count, but not to age, sex, orhtc levels.8 We found WBC count, LDH level, and spleen size to begreater in homozygous PV patients with high allele burden than inheterozygous PV patients with lower allele burden. In ET, severalstudies have shown that V617F-positive patients presented witholder age at diagnosis, higher Hb level and WBC count, but lowerplt count.3,9 In PMF, we found the same differences, except for thelower plt count. In our ET group, we found that patients with theJAK2 V617F mutation were younger, with larger splenomegaly, andhigher Hb and WBC counts, compared with patients without themutation.

Although thrombosis seems to be more common in our PVpatients than in our ET patients, this was not statistically significant(28% and 16%, respectively; P ¼ .07). None of our PMF patientshad clinical thrombosis before or after diagnosis. We reported 44


Table 3 Comparison of Mean Serum LDH and Hemoglobin Level, Mean age, and Mean Leukocyte and Thrombocyte Count in PatientsWith (Positive) and Without (Negative) the JAK2 V617F Mutation

Patient Population LDH, U/L Age, YearsSpleenSize, cm Hemoglobin, g/dL

LeukocyteCount, 3 109/L

ThrombocyteCount, 3 109/L

PV

JAK2 V617Fþ 301 � 115 62 � 11.8 3.1 � 3.6 17.5 � 1.7 15.3 � 5.5 660 � 264

JAK2 V617F� 194 � 42 62.2 � 18.4 0.8 � 1.4 19.6 � 2.2 11 � 5.1 312 � 128

P .2 .9 .2 .02a .1 .02a

ET

JAK2 V617Fþ 288 � 111 52.2 � 16.6 1.9 � 2.3 13.6 � 2 15 � 6.3 961 � 342

JAK2 V617F� 236 � 101 60.8 � 15.9 0.7 � 1.2 12.4 � 1.6 10.9 � 3.2 991 � 268

P .2 .006a .006a .001a .001a .6

PMF

JAK2 V617Fþ 561 � 352 65.6 � 11 6.6 � 4.1 11.5 � 2 17.3 � 4.2 640 � 369

JAK2 V617F� 874 � 924 53.2 � 13.5 12.3 � 7.4 10.7 � 0.9 8.5 � 3.8 647 � 276

P .2 .04a .04a .4 .1 .9

Abbreviations: ET ¼ essential thrombocythemia; LDH ¼ lactic dehydrogenase; PMF ¼ primary myelofibrosis; PV ¼ polycythemia vera.aP < .05 is statistically significant.


6 - Cli

(19%) cases of thrombosis among our 232 MPN patients. TIA wasthe most common thrombotic event in our PV and ET patients. Itwas observed in a little more than half (52%) of the 44 patients withthrombosis complications.

Peripheral arterial thrombosis was not observed in PV, whereas itwas observed in 5 (23.8%) of our 21 ET patients. Among our 23PV patients with thrombosis, 21 had a history of thrombosis yearsbefore their PV diagnoses. Despite phlebotomy, ASA, and cytoreduc-tive treatment, 1 patient developed PE and another developed TIA.

Thrombosis was observed at the time of diagnosis in 3 of our 21ET patients. In these 3 patients, JAK2 V617F allele burden wasquantified as 78% to 100% in the 69-year-old male patient withPAT, and 12.5% to 31% in the 41-year-old male patient with PATand the 18-year-old female patient with Budd-Chiari syndrome. A56-year-old ET patient developed a TIA while she was receiving

Table 4 Comparison of Mean Serum LDH and Hemoglobin Level, MeWith the JAK2 V617F Homozygous Genotype and Patients

JAK2 V617FMutation Genotype LDH, U/L

Age,Years

SpleSize,

PV

Homozygous 329 � 143 64 � 10.7 4.2 �Heterozygous 271 � 64 60 � 10.6 1.8 �P .03a .1 .00

ET

Homozygous 313 � 122 63 � 17.1 3.5 �Heterozygous 280 � 107 60 � 15.4 1.4 �P .2 .9 .00

PMF

Homozygous 542 � 351 64 � 12 7.1 �Heterozygous 583 � 385 68 � 70.5 5.6 �P .8 .5 .5

Abbreviations: ET ¼ essential thrombocythemia; LDH ¼ lactic dehydrogenase; PMF ¼ primary myeaP < .05 is statistically significant.


ASA and HU treatment and had slightly higher than normal pltcount. This patient also had additional conditions such as essentialhypertension and DM, which could lead to vascular complications.In ET treatment, age older than 60 years is considered as high-risk,but in younger patients we also observed other conditions that couldlead to vascular events. We therefore suggest more frequent moni-toring of these patients for antiaggregant and cytoreductive thera-pies, and for the management of their other diseases.

Some researchers have found thrombosis risk to be greater in JAK2V617F-positive ET patients compared with JAK2 V617F-negativeET patients.9-11,16,26 Our results agree with these reports. All ofour PV patients with thrombosis and most of our ET patients withthrombosis (76.1%) were JAK2 V617F mutation-positive.

Recent studies in PV and ET have indicated that the frequency ofthrombosis progressively increased with the amount of allele burden

an Age, and Mean Leukocyte and Thrombocyte Count in PatientsWith the JAK2 V617F Heterozygous Genotype

encm Hemoglobin, g/dL

LeukocyteCount, 3 109/L

ThrombocyteCount, 3 109/L

4.4 17.5 � 2 17.1 � 4.8 644 � 281

1.9 17.4 � 1.4 13.5 � 5.7 675 � 249

6a .8 .005a .6

2.5 12.8 � 2.4 20.8 � 7 1026 � 422

1.9 13.9 � 1.8 12.8 � 4.5 938 � 308

1a .01a .001a .2

4 11.1 � 1.7 31.7 � 18 590 � 398

4.5 12.1 � 2.6 18.3 � 10 732 � 344

.3 .1 .4

lofibrosis; PV ¼ polycythemia vera.

Figure 3 Patients Identified as JAK2 Mutation Homozygous (1)and Heterozygous (2) in the First 6 Months AfterDiagnosis, and Patients Identified as Homozygous (3)and Heterozygous (4) After 6 Months or Later. TheFrequency of JAK2 Mutation Positivity and theFrequency of Various Allele Burden Levels in ourCMPNs Series did not Significantly Change in theBeginning or Later

Abbreviations: CMPNs ¼ chronic myeloproliferative neoplasms; et ¼ essential thrombocy-themia; mf ¼ myelofibrosis; pv ¼ polycythemia vera.


in ET and PV.3,27-29 It has been reported that the highest rate ofvascular complications was seen in groups with allele burden> 50%, found in most of PV patients and in a small proportion ofET patients.3

We had similar results in our patients. Homozygous genotype forthe JAK2 V617F mutation was found in almost half (48.7%) of ourPV patients with thrombosis. We identified the heterozygous JAK2V617F mutation in 13 of 21 (61.9%), which is more than half ofour ET patients with thrombosis. The homozygous genotype wasfound in 3 of our ET patients with thrombosis, whereas 5 of the ETpatients with thrombosis were negative for the JAK2 V617Fmutation.

In their large study of 806 patients, Campbell et al have foundthat the JAK2 V617F mutation has a relatively small effect on

Table 5 Therapy and Follow-Up Data of 232 MPN Patients

Characteristic HUASA/Warfarin/Clopidogrel Anagrelide

PV (81)a 74 (91%) 69 (85%); 5/1 3 (3.7%)

ET (129)a 91 (71%) 111 (86%); 5/4 23 (18%)

PMF (22) 19 (86%) 13 (59%); 0/0 4 (18.2%)

Total: 232 184 (79%) 193 (83%); 10/5 30 (13%)

Abbreviations: ASA ¼ acetyl salicylic acid; ET ¼ essential thrombocythemia; HU ¼ hydroxyurea; Mpolycythemia vera.aMF secondary to PV, 1 patient (1.2%), and MF secondary to ET, 3 patients (2.3%).

vascular complications that are observed in patients before their ETdiagnoses.9 This group also observed that ET patients with theJAK2 V617F mutation do not have an increased risk of arterial orvenous thrombosis after entry into the trial. Similarly, in a retro-spective study of 867 ET patients, the GIMEMA group found thatthe JAK2 V617F mutation at diagnosis was not a predictor foroccurrence of thrombosis during follow-up.30

We found that almost one-quarter (23.8%) of our ET patientswith thrombosis were negative for the JAK2 V617F mutation.Additionally, we quantified the allele burden as < 31% (12.5%-31%in 6 patients, 2%-5% in 1 patient, and < 2% in 1 patient) in morethan half (57%) of our JAK2 V617F-positive patients with throm-bosis. These low values made us question the effect of the degree ofJAK2 V617F allele burden on vascular events in ET patients.

Although it is accepted that the occurrence of thrombosis in PVor ET is not directly linked to increased htc or plt counts, leuko-cytosis has recently been associated with increased risk of majorvascular events and eventually could represent a major target ofcytoreductive therapy.31 It is reported that a leukocyte count >

8.4 � 109/L in ET is also strictly associated with an increased risk ofthrombosis as recently shown by Barbui et al.32,33 Several studieshave confirmed that standard risk factors (age > 60 years and/orprevious thrombosis) and leukocytosis (> 11.3 � 109/L in theGIMEMA study) were significantly associated with an increased riskof events in follow-up.9,34-36

Although we observed an increased frequency of thrombosis inpatients with > 11 � 109/L WBC count consistent with the pre-vious reports, we could not show the correlation of age > 60 yearswith increased frequency of thrombosis in PV and ET.

Few data have been published on the influence of JAK2 V617Fmutation on bleeding risk in MPNs. In PV patients, no correlationhas been found between homozygous and heterozygous genotypeand the occurrence of bleeding complications.15 In a study of 106ET patients, minor bleeds from various mucocutaneous sites wereobserved in 7 patients, 6 of which were JAK2 V617F-negative.Analysis from this study indicated that JAK2 V617F does nothave a protective role on hemorrhagic risk.31 In our study, gastro-intestinal system (GIS) bleeding was observed in all PV patientswith bleeding. GIS hemorrhage or epistaxis was observed in ETpatients with bleeding. We have thus concluded, different fromprevious reports, that JAK2 V617F negativity is not correlated withbleeding complications in MPNs. We think that in most patients,taking antiaggregants or anticoagulants are more important inbleeding complications than is the JAK2 V617F mutation. In the

InterferonAcute

Leukemia Died

Median Time FromDiagnosis, Months

(Range)

4 (4.9%) 2 (2.5%) 3 (3.7%) 38 (1-219)

6 (4.6%) 3 (1.6%) 3 (2.3%) 24.5 (1-152)

5 (22.7%) e 5 (2.7%) 36 (1-112)

15 (6.5%) e 11 (4.7%) 28 (1-219)

F ¼ myelofibrosis; MPN ¼ myeloproliferative neoplasm; PMF ¼ primary myelofibrosis; PV ¼


Table 6 Data of 11 CMPN Patients Who Died During the Follow-Up Period

PatientNumber Diagnosis

Age,Years Sex

JAK2 V617FGenotype Therapy Events Before Death

Follow-Up,Months

1 PV 66 M Homozygous HU, IF, ASA AML 53

2 PV 59 F Homozygous HU, ASA, Phlebotomy MDS (RAEB-2) 132

3 PV 70 F Heterozygous Phlebotomy Portal hypertension, renalinsuffiency, pneumonia

1

4 ET 71 F Heterozygous HU, ASA, anagrelide AML 93

5 ET 57 M Heterozygous HU, ASA AML 54

6 ET 73 M Homozygous HU, ASA AML 35

7 PMF 67 M Heterozygous HU, splenoctomy,lenalidomide

Sepsis 56

8 PMF 61 M Homozygous HU, splenectomy Heart failure 72

9 PMF 62 M Homozygous HU, splenectomy Heart failure 62

10 PMF 78 M Heterozygous HU Coronary artery disease, renalfailure, pneumonia

1

11 PMF 60 F Heterozygous HU Renal failure, diabetes mellitus,essential hypertension

61

Abbreviations: AML ¼ acute myeloid leukemia; ASA ¼ acetyl salicylic acid; CMPN ¼ chronic myeloproliferative neoplasm; ET ¼ essential thrombocythemia; F ¼ female; HU ¼ hydroxyurea;IF ¼ interferon; M ¼ male; MDS ¼ myelodysplastic syndrome; PMF ¼ primary myelofibrosis; PV ¼ polycythemia vera; RAEB ¼ refractory anemia with excess blasts.


8 - Cli

PV and ET group, frequencies of clinical bleeding in patients ho-mozygous and heterozygous for the JAK2 mutation were similar,consistent with results of previous studies.

Hematologic transformations to MF, MDS, or AML are recog-nized long-term complications of PV and ET.37 One (1.2%) of our81 PV patients progressed into MF secondary to PV, and 3 (2.3%)of our 129 ET patients progressed into MF secondary to ET. Ourresults were consistent with a study of 605 ET patients in which17 (2.8%) had progressed to MF. In the same study, acute leukemiawas seen in 14 (2.3%) patients, whereas in our study, slightly fewer

Figure 4 Kaplan-Meier Overall Survival Curves of PV, ET, andPMF Patients. PMF Patients Were Found to HaveShorter Survival Times Than PV and ET Patients

Abbreviations: Cum ¼ cumulative; ET ¼ essential thrombocythemia; ex ¼ exitus;PMF ¼ primary myelofibrosis; pts ¼ patients; PV ¼ polycythemia vera.


patients, 3 (1.6%) in the ET group and 1 (1.2%) in the PV group,progressed into acute leukemia.

Leukemic clones emerging from V617F-positive chronic MPNcan be either V617F-positive or V617F-negative.38 V617F-positiveAML more frequently occurs after transition to the MF phase,whereas V617F-negative AML seems to arise directly from chronic-phase JAK2-positive MPN.39 In our ET patient group, 1 progressedinto AML after MF development. The remaining 2 patients directlydeveloped AML. Three ET and 1 PV patient who had acute leu-kemia also had the JAK2 V617F mutation.

The question of a possible effect of the JAK2 V617F mutation onOS of patients with MPN has not been clearly answered by pre-viously published studies, which have provided conflicting results.27

So far, we only have the median survival in our PMF patients(62 months; 95% CI, 57.9-66). In the PV and ET groups, alongwith all MPN patients, we have not yet reached median survivalstatistically because of less patient loss in these goups.

A subset of patients with PV, ET, and PMF harbor mutations innight lighteinducible and clock-regulated genes and the CasitasB-cell lymphoma gene, TET2, ASXL1, and IDH,40 but pathoge-netic contributions of these and other mutations are unclear at thistime.41 Nowadays, calreticulin mutations have been described.These mutations have been observed in approximately 70% ofpatients with ET or PMF who do not carry mutations in either theJAK2 or MPL gene, and have not been observed in patients withPV. Because all of these mutations are still being researched, wecould not study these in our patients.

ConclusionThrombosis was found in 19% of all of our MPN patients. TIA

was the most common thrombotic event, seen in 52% of the pa-tients with thrombosis. When our center began assessing patientsfor the JAK2 V617F mutation, we quantified the JAK2 V617Fmutation in patients diagnosed at that time and previously. Amongthose 2 groups with a similar number of patients, there were no


significant differences in JAK2 V617F positivity or quantificationvalues. However, our results were based on different patientsassessed at different times for the JAK2 V617F mutation. Our re-sults are limited by this, and it is apparent that JAK2 V617F as-sessments need to be done at certain time points after diagnosis todetermine the effects of follow-up times and cytoreductive treat-ments on JAK2 V617F positivity and allele burden quantification.

With longer follow-up and larger patient populations, this studyaimed to investigate our patients’ progression into myelofibrosis oracute leukemia, the effect of clinical and laboratory parameters onpatient survival, and finally to compare our results with currentliterature findings.

Clinical Practice Points
� In this study, we aimed to investigate the presence of JAK2V617F in CMPNs patients in our center. JAK2 V617F mutationfrequencies of our PV and ET patients were similar to thosereported previously.
� JAK2 V617F mutation frequency of our PMF patients washigher than in previous reports. We attribute this difference toour low number of PMF patients, and think that it will matchother reports if we have a larger data set. Transient ischemicattack was the most common thrombotic event in our PV andET patients. In ET treatment age over 60 is considered a standardrisk, but in younger patients we also observed other conditionssuch as essential hypertension, diabetes mellitus, hereditarythrombophilia that could lead to vascular events.

� We therefore suggest more frequent monitoring of these patientsfor antiaggregant and cytoreductive therapies, and for the man-agement of their other diseases.

DisclosureThe authors have stated that they have no conflicts of interest.

References1. Tefferi A, Vardiman JW. Classification and diagnosis of myeloproliferative neo-

plasms: the 2008 World Health Organization criteria and point-of-care diagnosticalgorithms. Leukemia 2008; 22:14-22.

2. Pardanani A, Lasho TL, Finke C, et al. Prevalence and clinicopathologic correlatesof JAK2 exon 12 mutations in JAK2 V617F-negative polycythemia vera. Leukemia2007; 21:1960-3.

3. Vannucchi AM, Antonioli E, Guglielmelli P, et al. Clinical correlates of JAK2V617F presence or allele burden in myeloproliferative neoplasms: a critical reap-praisal. Leukemia 2008; 22:1299-307.

4. Tefferi A. Polycythemia vera and essential thrombocythemia: 2012 update ondiagnosis, risk stratification, and management. Am J Hematol 2012; 87:285-93.

5. Pardanani A, Lasho TL, Finke CM, et al. Infrequent occurrence of MPL exon 10mutations in polycythemia vera and post-polycythemia vera myelofibrosis. Am JHematol 2011; 86:701-2.

6. Martínez-Avilés L, Besses C, Álvarez-Larrán A, Torres E, Serrano S, Bellosillo B.TET2, ASXL1, IDH1, IDH2, and c-CBL genes in JAK2- and MPL-negativemyeloproliferative neoplasms. Ann Hematol 2012; 91:533-41.

7. Nangalia J, Massie CE, Baxter EJ, et al. Somatic CALR mutations in myelopro-liferative neoplasms with nonmutated JAK2. N Engl J Med 2013; 369:2391-405.

8. Silver RT, Vandris K, Wang YL, et al. JAK2(V617F) allele burden in polycythemiavera correlates with grade of myelofibrosis, but is not substantially affected bytherapy. Leuk Res 2011; 35:177-82.

9. Campbell PJ, Scott LM, Buck G, et al. Definition of subtypes of essentialthrombocythaemia and relation to polycythaemia vera based on JAK2 V617Fmutation status: a prospective study. Lancet 2005; 366:1945-53.

10. Finazzi G, Rambaldi A, Guerini V, et al. Risk of thrombosis in patients withessential thrombocythemia and polycythemia vera according to JAK2 V617Fmutation status. Haematologica 2007; 92:135-6.

11. Wolansky AP, Lasho TL, Schwager SM, et al. JAK2 V617F mutation in essentialthrombocythaemia: clinical associations and long-term prognostic relevance. Br JHaematol 2005; 131:208-13.

12. Campbell PJ, Griesshammer M, Dohner K, et al. V617F mutation in JAK2 isassociated with poorer survival in idiopathic myelofibrosis. Blood 2006; 107:2098-100.

13. Tefferi A, Lasho TL, Schwager SM, et al. The JAK2(V617F) tyrosine kinasemutation in myelofibrosis with myeloid metaplasia: lineage specificity and clinicalcorrelates. Br J Haematol 2005; 131:320-8.

14. Guglielmelli P, Barosi G, Specchia G, et al. Identification of patients with poorersurvival in primary myelofibrosis based on the burden of JAK2 V617F mutatedallele. Blood 2009; 114:1477-83.

15. Tefferi A, Lasho TL, Schwager SM, et al. The clinical phenotype of wild-type,heterozygous, and homozygous JAK2V617F in polycythemia vera. Cancer 2006;106:631-5.

16. Vannucchi AM, Antonioli E, Guglielmelli P, et al. Clinical profile of homozygousJAK2 617V>F mutation in patients with polycythemia vera or essential throm-bocythemia. Blood 2007; 110:840-6.

17. Cross NC. Genetic and epigenetic complexity in myeloproliferative neoplasms.Hematology Am Soc Hematol Educ Program 2011; 2011:208-14.

18. Barosi G, Bergamaschi G, Marchetti M, et al. JAK2 V617F mutational statuspredicts progression to large splenomegaly and leukemic transformation in primarymyelofibrosis. Blood 2007; 110:4030-6.

19. McNally RJ, Rowland D, Roman E, Cartwright RA. Age and sex distributions ofhematological malignancies in the U.K. Hematol Oncol 1997; 15:173-89.

20. Stein BL, Williams DM, Wang NY, et al. Sex differences in the JAK2 V617F alleleburden in chronic myeloproliferative disorders. Haematologica 2010; 95:1090-7.

21. Johansson P, Kutti J, Andreasson B, et al. Trends in the incidence of chronicPhiladelphia chromosome negative (Ph-) myeloproliferative disorders in the city ofGoteborg, Sweden, during 1983-99. J Intern Med 2004; 256:161-5.

22. Passamonti F, Rumi E, Pietra D, et al. Relation between JAK2 (V617F) mu-tation status, granulocyte activation and constitutive mobilization of CD34-positive cells into peripheral blood in myeloproliferative disorders. Blood 2006;107:3676-82.

23. Tefferi A, Strand JJ, Lasho TL, et al. Bone marrow JAK2 V617F allele burden andclinical correlates in polycythemia vera. Leukemia 2007; 21:2074-5.

24. Kittur J, Knudson RA, Lasho TL, et al. Clinical correlates of JAK2V617F alleleburden in essential thrombocythemia. Cancer 2007; 109:2279-84.

25. Vannucchi AM, Antonioli E, Guglielmelli P, et al. Prospective identification ofhigh-risk polycythemia vera patients based on JAK2(V617F) allele burden. Leu-kemia 2007; 21:1952-9.

26. Cheung B, Radia D, Pantelidis P, Yadegarfar G, Harrison C. The presence of theJAK2 V617F mutation is associated with a higher haemoglobin and increased riskof thrombosis in essential thrombocythaemia. Br J Haematol 2006; 132:244-5.

27. Levine RL, Wadleigh M, Cools J, et al. Activating mutation in the tyrosine kinaseJAK2 in polycythemia vera, essential thrombocythemia, and myeloid metaplasiawith myelofibrosis. Cancer Cell 2005; 7:387-97.

28. Vainchenker W, Constantinescu SN. A unique activating mutation in JAK2(V617F) is at the origin of polycythemia vera and allows a new classification ofmyeloproliferative diseases. Hematology Am Soc Hematol Educ Program 2005:195-200.

29. Prchal JF, Axelrad AA. Letter: bone-marrow responses in polycythemia vera.N Engl J Med 1974; 290:1382.

30. Carobbio A, Finazzi G, Antonioli E, et al. JAK2 V617F allele burden andthrombosis: a direct comparison in essential thrombocythemia and polycythemiavera. Exp Hematol 2009; 37:1016-21.

31. Barbui T, Carobbio A, Rambaldi A, Finazzi G. Perspectives on thrombosis inessential thrombocythemia and polycythemia vera: is leukocytosis a causative fac-tor? Blood 2009; 114:759-63.

32. Patriarca A, Pompetti F, Malizia R, et al. Is the absence of JAK2 mutation a riskfactor for bleeding in essential thrombocythemia? An analysis of 106 patients.Blood Transfus 2010; 8:21-7.

33. Barbui T, Barosi G, Grossi A, et al. Practice guidelines for the therapy of essentialthrombocythemia. A statement from the Italian Society of Hematology, the ItalianSociety of Experimental Hematology and the Italian Group for Bone MarrowTransplantation. Haematologica 2004; 89:215-32.

34. Wilkins BS, Erber WN, Bareford D, et al. Bone marrow pathology in essentialthrombocythemia: interobserver reliability and utility for identifying disease sub-types. Blood 2008; 111:60-70.

35. Levine RL, Loriaux M, Huntly BJ, et al. The JAK2 V617F activating mutationoccurs in chronic myelomonocytic leukemia and acute myeloid leukemia, but notin acute lymphoblastic leukemia or chronic lymphocytic leukemia. Blood 2005;106:3377-9.

36. Moliterno AR, Williams DM, Rogers O, Isaacs MA, Spivak JL. Phenotypicvariability within the JAK2 V617F-positive MPD: roles of progenitor cell andneutrophil allele burdens. Exp Hematol 2008; 36:1480-6.

37. Campbell PJ, Green AR. The myeloproliferative disorders. N Engl J Med 2006;355:2452-66.

38. Theocharides A, Boissinot M, Girodon F, et al. Leukemic blasts in transformedJAK2 V617F-positive myeloproliferative disorders are frequently negative for theJAK2 V617F mutation. Blood 2007; 110:375-9.

39. Beer PA, Delhommeau F, LeCouédic JP, et al. Two routes to leukemic trans-formation after a JAK2 mutation-positive myeloproliferative neoplasm. Blood2010; 115:2891-900.

40. Tefferi A, Vainchenker W. Myeloproliferative neoplasms: molecular pathophysi-ology, essential clinical understanding, and treatment strategies. J Clin Oncol 2011;29:573-82.

41. Vainchenker W, Delhommeau F, Constantinescu SN, Bernard OA. New muta-tions and pathogenesis of myeloproliferative neoplasms. Blood 2011; 118:1723-35.


http://refhub.elsevier.com/S2152-2650(14)00055-X/sref1





















































































































Documents

JAK2 V617F Mutation Status of 232 Patients Diagnosed With Chronic Myeloproliferative Neoplasms