Mutation screening for JAK2V617F: When to order the test and how ...

Leukemia Research 30 (2006) 739–744Brief communicationMutation screening for JAK2 V617F : When to order the testand how to interpret the resultsAyalew Tefferi ∗ , Animesh PardananiDivision of Hematology, Mayo Clinic, Rochester, 200 First Street SW, Rochester, MN 55905, USAReceived 31 December 2005; received in revised form 5 January 2006; accepted 6 January 2006Available online 7 February 2006AbstractWith the application of adequately sensitive tests, it is now becoming evident that more than 90% of patients with conventionally-definedpolycythemia vera (PV) carry the somatic JAK2 V617F mutation in their granulocytes. However, the specific mutation is also found in other classicand atypical myeloproliferative disorders (MPD), albeit at a lesser frequency. In contrast, JAK2 V617F has not been reported in patients witheither reactive myeloproliferation or lymphoid disorders. Therefore, mutation screening for JAK2 V617F can be considered as a myeloid-specificclonality assay and it is diagnostically most useful in the evaluation of “polycythemia”.© 2006 Elsevier Ltd. All rights reserved.Keywords: JAK2 mutation; Polycythemia; Thrombocytosis; DiagnosisBetween March and June 2005, five studies describeda novel JAK2 somatic point mutation (a G-C to T-Atransversion, at nucleotide 1849 of exon 14, resulting inthe substitution of valine to phenylalanine at codon 617;JAK2 V617F ) in classic, BCR/ABL-negative myeloproliferativedisorders (MPD) [1–5]. Subsequently, the same mutationwas demonstrated in other myeloid disorders includingacute myeloid leukemia, the myelodysplastic syndrome,chronic myelomonocytic leukemia, juvenile myelomonocyticleukemia, chronic neutrophilic leukemia, hypereosinophilicsyndrome, systemic mastocytosis (SM), and otheratypical MPD (Table 1). In contrast, JAK2 V617F has notyet been reported in patients with secondary polycythemia,[1,6,7] acute [8–11] or chronic lymphoid disorders, [11–14]non-hematologic cancer, [10], or molecularly-defined MPD(e.g. chronic myeloid leukemia, [9] PDGFR-rearrangedeosinophilic disorders, [6] KIT D816V -positive SM) [6]. Furthermore,among several hematologic and solid tumorcell lines, JAK2 V617F was detected in only human erythroleukemia(HEL) cells [13,14].∗ Corresponding author: Tel.: +1 507 284 3159; fax: +1 507 266 4972.E-mail address: tefferi.ayalew@mayo.edu (A. Tefferi).The pathogenetic relevance of JAK2 V617F is currentlyunder intense investigation [15]. The specific mutation occurswithin the enzymatically inactive JH2 pseudo-kinase domainthat regulates the active JH1 kinase domain [16]. Such indirectJAK2 activation leads to constitutive JAK-STAT hyperactivationwith resultant induction of erythropoietin hypersensitivityand cell transformation in murine cell lines andin vivo erythrocytosis in mice [1]. These laboratory observationsare consistent with the clinical demonstration of significantlyhigher hemoglobin levels in JAK2 V617F -positiveversus JAK2 V617F -negative ET [17–19] and in polycythemiavera (PV) patients with homozygous versus heterozygousmutant allele [20]. Similarly, mutation burden has beenshown to be significantly higher in PV compared to otherMPD [21]. Most recently, activation of the enzyme as wellas downstream signaling involving JAK2 V617F was shown torequire the presence of an associated Type 1 cytokine receptor,which might explain the myeloid disorder-specificity aswell as phenotypic diversity that is associated with the specificmutation [22].A quick glance at Table 1 discloses a marked variationin mutational frequency, among different studies, in the contextof similarly defined clinicopathologic entities. Based on0145-2126/$ – see front matter © 2006 Elsevier Ltd. All rights reserved.doi:10.1016/j.leukres.2006.01.004

740 A. Tefferi, A. Pardanani / Leukemia Research 30 (2006) 739–744Table 1Summary of published literature on JAK2 V617F mutational frequencies in myeloproliferative disorders. Homozygous mutational frequencies are indicated inparenthesisReference Polycythemia vera Essential thrombocythemia MMM Other myeloid disordersAssay type usedCell type usedLevine et al. [3]PCR sequencing 74% (25%) 32% (3%) 35% (9%) –Neutrophils n = 164 n = 115 n =46Kralovics et al. [4]PCR sequencing 65% (27%) 23% (3%) 57% (22%) CML 0%Neutrophils n = 128 n =93 n =23 (n =9)Baxter et al. [2]AS-PCR 97% (26%) 57% (0%) 50% (19%) –Neutrophils n =73 n =51 n =16All were AMMJones et al. [6]AS-PCR 81% (33%) 41% (7%) 43% (29%) CML 0%PB cells n =72 n =59 n =35 (n = 18)CMML/UMPD 20% (n = 152)HES/CEL 2% (n = 134)SM 0% (n = 28)CNL 33% (n =6)AML 0% (n = 17)James et al. [1]PCR sequencing 89% (30%) 43% 43% –Neutrophils n =45 n =21 n =7Tefferi et al. [7]PCR sequencing 95% (26%) 55% (0%) 30% (0%) –Neutrophils n =38 n =22 n =10All were AMMJelinek et al. [9]Pyrosequencing 86% 30% 95% CML 0%PB cells n =29 n =10 n =19 (n = 99)Post-MPD AML 55% (n = 22)AML M0-M5 0% (n = 20)AMLM60%(n =8)AML M7 18% (n = 11)MDS 1% (n = 68)CMML 13% (n = 52)Atypical MPD 19% (n = 16)Goerttler et al. [30]cDNA PCR sequencing 100% (27%) 33% (5%) 57% (21%) –Neutrophils n =22 n =42 n =14Passamonti et al. [21]Real-time AS-PCR 92% 53% 72% –Neutrophils n =25 n =19 n =46AMM 57%PPMM 100%James et al. [24]Real-time PCR 95% – – –Neutrophils n =80Campbell et al. [18]AS-PCR – 53% – –PB cells n = 776Wolanskyj et al. [17]PCR sequencing – 49% (0%) – –PB-MNC or BM cells n = 150Antonioli et al. [19]AS-PCR – 57% (6%) – –Neutrophils n = 130

A. Tefferi, A. Pardanani / Leukemia Research 30 (2006) 739–744 741Table 1 (Continued )Reference Polycythemia vera Essential thrombocythemia MMM Other myeloid disordersAssay type usedCell type usedZhao et al. [5]PCR sequencing 83% (NA) – – –PB-MNC n =24Tefferi et al. [31]PCR sequencing – – 51% (6%) –PB-MNC n = 157AMM-45% (3%)PTMM-40% (11%)PPMM-91% (18%)Campbell et al. [32]AS-PCR or sequencing – – 55% –PB or BM cells n = 152All were AMMTefferi et al. [20]PCR sequencing 92% (21%) – – –Neutrophils n =63Steensma et al. [28] – – – MDS 5% (n = 101)PCR sequencing CMML 3% (n = 119)PB and BM cells HES 0% (n = 11)SM 25% (n =8)CNL 17% (n =6)Levine et al. [11]PCR sequencing – – – De novo AML 0.5% (n = 219)PB and BM cells Post-MPD AML 100% (n =3)MDS4%(n = 48)CMML/atypical MPD 8% (n = 116)Johan et al. [33]AS-PCR – – 38% CMML 2%PB cells n =21 (n = 47)All were AMMOhyashiki et al. [27]PCR sequencing – – – MDS-MF 33% (n =6)BM or PB cellsLee et al. [10]PCR-based SSCP 65% (35%) – – AML 2% (n = 113)BM cells n =17Scott et al. [14] AML6%(n = 90)AS-PCR MDS 0.5% (n = 20)BM cells CMML 0% (n =5)CML 1%* (n = 91)Tono et al. [34]PCR sequencing – – – JMML 20% (n =5)BM cells* The single CML patient had actually BCR/ABL-negative ET prior to the development of CML. The mutational frequencies in different studies are notcomparable because of the use of assays with variable sensitivity. AS-PCR, allele-specific PCR; CMML, chronic myelomonocytic leukemia; JMML, juvenilemyelomonocytic leukemia; CML, chronic myeloid leukemia; MDS, myelodysplastic syndrome; HES, hypereosinophilic syndrome; CEL, chronic eosinophilicleukemia; SM, systemic mastocytosis; CNL, chronic neutrophilic leukemia; UMPD, unclassified myeloproliferative disorder; AML, acute myeloid leukemia;BM, bone marrow; PB, peripheral blood; MNC, mononuclear cells; MMM, myelofibrosis with myeloid metaplasia; AMM, agnogenic myeloid metaplasia;PTMM, post thrombocythemic myeloid metaplasia; PPMM, post polycythemic myeloid metaplasia; PCR, polymerase chain reaction.our own experience that revealed, on careful re-review, anincorrect diagnosis for four reportedly JAK2 V617F -negativePV cases, [23] we trust that diagnostic imprecision is inpart to blame for the observed inconsistency in diseasespecificincidence figures reported in the literature. Such acontention is supported by a recent study that demonstrateddiscordant JAK2 V617F prevalence figures in WHO- versusPVSG-defined PV within the same study population [24].Differences in methodological assay sensitivity constituteanother potential cause of discrepancy in reported mutational

742 A. Tefferi, A. Pardanani / Leukemia Research 30 (2006) 739–744Table 2Test indications and result interpretation of mutation screening for JAK2 V617FJAK2 V617F test indications Positive result Negative result“Increased” hemoglobin/hematocrit Consistent with PV PV diagnosis unlikely but not impossibleSplanchnic or cerebral vein thrombosis Consistent with PV/MPD PV diagnosis unlikelyThrombocytosis Consistent with MPD Diagnostically unhelpfulUnexplained cytopenia or cytosis Consistent with MPD Diagnostically unhelpfulfrequencies. For example, use of a more sensitive assay in onestudy led to an increase in mutational frequency for PV from73% to 97% [2]. Assay sensitivity might also be affected byboth tissue type (neutrophils versus other cells) and nucleicacid source (genomic DNA versus RNA) of the test sample.Nevertheless, the practical impact of such factors is currentlyuncertain and systematic studies are needed for furtherclarification [24]. In the meantime, we strongly recommendavoidance of undue reliance on a single test report and encouragerepeat testing in the presence of either marginally positiveor clinically incompatible test results.Some generalizations and practical recommendations canbe made based on information from Table 1 and the abovediscussion (Table 2). First, either peripheral blood or bonemarrow mutation screening for JAK2 V617F appears to becurrently most useful for the evaluation of “polycythemia”.However, the particular test is intended to complement ratherthan substitute for clinical diagnosis that is based on history,physical examination, and pertinent laboratory tests includingserum erythropoietin level and bone marrow examination[25]. Accordingly, if clinical suspicion for PV is low, a negativeperipheral blood test result argues against the specificdiagnosis and should allow deferral of further investigationwith either bone marrow biopsy or red cell mass measurement.A bone marrow biopsy is, however, indicated in thepresence of a high clinical suspicion for PV, regardless ofmutational status [20,21,24]. Furthermore, in the context ofsuspected PV, information regarding JAK2 V617F mutationalstatus should help clarify the situation with either “equivocal”or “inconsistent” bone marrow histology.Peripheral blood mutation screening for JAK2 V617F is alsohelpful during the evaluation of an increased platelet countthat is clinically not consistent with reactive thrombocytosis.However, unlike the aforementioned situation with the workup of polycythemia, the negative predictive value of the testis too low to warrant deferral of bone marrow examination inthe absence of JAK2 V617F [17,18]. Similarly, although a positivetest result confirms clonality and makes the possibility ofFig. 1. A modern diagnostic algorithm for suspected polycythemia vera (PV) that incorporates mutation screening for JAK2 V617F . * PV-characteristic featuresinclude splenomegaly, leukocytosis, thrombocytosis, aquagenic pruritus, splanchnic or cerebral vein thrombosis, erythromelalgia, and microcytosis related toiron deficiency.

A. Tefferi, A. Pardanani / Leukemia Research 30 (2006) 739–744 743chronic myeloid leukemia unlikely, it cannot further specifythe underlying MPD [18]. Therefore, a bone marrow examinationshould not be skipped over during the work up ofapparently clonal thrombocytosis, regardless of JAK2 V617Fmutational status. Approximately 50% of patients with ETare expected to carry the JAK2 V617F mutation [17,18] anda positive mutation screening test confirms and/or clarifiesthe otherwise largely subjective histological diagnosis of ET.Furthermore, although heterozygous JAK2 V617F is seen inboth ET and PV, homozygosity for the mutant allele favorsthe latter diagnosis [18].In general, our enthusiasm is lukewarm regarding the utilityof mutation screening for JAK2 V617F in MPD other thanPV or ET [6,26,27]. In this regard, it is apparent from theinformation from Table 1 that the particular test cannot reliablydistinguish classic MPD from atypical MPD. However,mutational frequency in both MDS and de novo AML islow enough to consider limited utility in differentiating thesedisorders from MPD as a whole [9,11,28]. Also, in routineclinical practice, it is not infrequent to encounter casesof unexplained cytopenia, cytosis, or deep vein thrombosisinvolving splanchnic or cerebral vessels. In such instances,mutation screening for JAK2 V617F is indicated and a positivetest is consistent with the presence of occult MPD includinginapparent PV whereas a negative test result does not ruleout the possibility. Finally, our overall discussion assumesthe utilization of adequately sensitive (i.e. sensitivity of 5%or less) mutation screening assays as well as a Bayesianapproach to result interpretation. The latter becomes especiallyimportant when using super-sensitive assays that mightdetect small copies of JAK2 V617F in healthy individuals, a situationthat is analogous to BCR/ABL [29]. Finally, the overallobjectivity and accessibility of both JAK2 mutation screeningand erythropoietin determination from the peripheralblood allows for a practical modern approach in evaluatingpolycythemia that does not require red cell mass measurementand limits the number of bone marrow examinations(Fig. 1).AcknowledgementsWe acknowledge Dr. Rebecca F. McClure’s help indata sharing and scientific discussion. There are no potentialauthor conflicts of interest that relate to the currentmanuscript.Contributions. Ayalew Tefferi came up with the concept,designed the study, and wrote the paper. Dr. Animesh Pardananihelped with the design of the study and approved thefinal draft of the paper.References[1] James C, Ugo V, Le Couedic JP, Staerk J, Delhommeau F, LacoutC, et al. A unique clonal JAK2 mutation leading to constitutivesignalling causes polycythaemia vera. Nature 2005;434:1144–8.[2] Baxter EJ, Scott LM, Campbell PJ, East C, Fourouclas N, SwantonS, et al. Acquired mutation of the tyrosine kinase JAK2 in humanmyeloproliferative disorders. Lancet 2005;365:1054–61.[3] Levine RL, Wadleigh M, Cools J, Ebert BL, Wernig G, HuntlyBJ, et al. Activating mutation in the tyrosine kinase JAK2 in polycythemiavera, essential thrombocythemia, and myeloid metaplasiawith myelofibrosis. Cancer Cell 2005;7:387–97.[4] Kralovics R, Passamonti F, Buser AS, Teo SS, Tiedt R, PasswegJR, et al. A gain-of-function mutation of JAK2 in myeloproliferativedisorders. N Engl J Med 2005;352:1779–90.[5] Zhao R, Xing S, Li Z, Fu X, Li Q, Krantz SB, et al. Identificationof an acquired JAK2 mutation in polycythemia vera. J Biol Chem2005;280:22788–92.[6] Jones AV, Kreil S, Zoi K, Waghorn K, Curtis C, Zhang L, et al.Widespread occurrence of the JAK2 V617F mutation in chronicmyeloproliferative disorders. Blood 2005;106:2162–8.[7] Tefferi A, Sirhan S, Lasho TL, Schwager SM, Li CY, Dingli D,et al. Concomitant neutrophil JAK2 mutation screening and PRV-1 expression analysis in myeloproliferative disorders and secondarypolycythaemia. Br J Haematol 2005;131:166–71.[8] Sulong S, Case M, Minto L, Wilkins B, Hall A, Irving J. The V617Fmutation in Jak2 is not found in childhood acute lymphoblasticleukaemia. Br J Haematol 2005;130:964–5.[9] Jelinek J, Oki Y, Gharibyan V, Bueso-Ramos C, Prchal JT, VerstovsekS, et al. JAK2 mutation 1849G > T is rare in acute leukemiasbut can be found in CMML, Philadelphia chromosome-negativeCML, and megakaryocytic leukemia. Blood 2005;106:3370–3.[10] Lee JW, Kim YG, Soung YH, Han KJ, Kim SY, Rhim HS, et al.The JAK2 V617F mutation in de novo acute myelogenous leukemias.Oncogene 2005.[11] Levine RL, Loriaux M, Huntly BJ, Loh ML, Beran M, StoffregenE, et al. The JAK2V617F activating mutation occurs in chronicmyelomonocytic leukemia and acute myeloid leukemia, but notin acute lymphoblastic leukemia or chronic lymphocytic leukemia.Blood 2005;106:3377–9.[12] Lee JW, Soung YH, Kim SY, Nam SW, Park WS, Lee JY, et al.JAK2 V617F mutation is uncommon in non-Hodgkin lymphomas.Leuk Lymphoma 2006;47:313–4.[13] Melzner I, Weniger MA, Menz CK, Moller P. Absence of the JAK2V617F activating mutation in classical Hodgkin lymphoma and primarymediastinal B-cell lymphoma. Leukemia 2005.[14] Scott LM, Campbell PJ, Baxter EJ, Todd T, Stephens P, EdkinsS, et al. The V617F JAK2 mutation is uncommon in cancers andin myeloid malignancies other than the classic myeloproliferativedisorders. Blood 2005;106:2920–1.[15] Tefferi A, Gilliland DG. JAK2 in myeloproliferative disorders is notjust another kinase. Cell Cycle 2005;4:1053–6.[16] Saharinen P, Vihinen M, Silvennoinen O. Autoinhibition of Jak2tyrosine kinase is dependent on specific regions in its pseudokinasedomain. Mol Biol Cell 2003;14:1448–59.[17] Wolanskyj AP, Lasho TL, Schwager SM, McClure RF, Wadleigh M,Lee SJ, et al. JAK2 mutation in essential thrombocythaemia: clinicalassociations and long-term prognostic relevance. Br J Haematol2005;131:208–13.[18] Campbell PJ, Scott LM, Buck G, Wheatley K, East CL, Marsden JT,et al. Definition of subtypes of essential thrombocythaemia and relationto polycythaemia vera based on JAK2 V617F mutation status:a prospective study. Lancet 2005;366:1945–53.[19] Antonioli E, Guglielmelli P, Pancrazzi A, Bogani C, Verrucci M,Ponziani V, et al. Clinical implications of the JAK2 V617F mutationin essential thrombocythemia. Leukemia 2005;19:1847–9.[20] Tefferi A, Lasho TL, Schwager SM, Strand JS, Elliott M,Mesa R, et al. The clinical phenotype of wild-type, heterozygous,and homozygous JAK2(V617F) in polycythemia vera. Cancer2005.[21] Passamonti F, Rumi E, Pietra D, Della Porta MG, Boveri E, PascuttoC, et al. Relation between JAK2 (V617F) mutation status,

744 A. Tefferi, A. Pardanani / Leukemia Research 30 (2006) 739–744granulocyte activation and constitutive mobilization of CD34-positivecells into peripheral blood in myeloproliferative disorders. Blood2005.[22] Lu X, Levine R, Tong W, Wernig G, Pikman Y, Zarnegar S, et al.Expression of a homodimeric type I cytokine receptor is requiredfor JAK2V617F-mediated transformation. Proc Natl Acad Sci USA2005;102:18962–7.[23] McClure R, Mai M, Lasho T. Validation of two clinically usefulassays for evaluation of JAK2 V617F mutation in chronic myeloproliferativedisorders. Leukemia 2005.[24] James C, Delhommeau F, Marzac C, Teyssandier I, Couedic JP,Giraudier S, et al. Detection of JAK2 V617F as a first intentiondiagnostic test for erythrocytosis. Leukemia 2005.[25] Tefferi A, Spivak JL. Polycythemia vera: scientific advances andcurrent practice. Semin Hematol 2005;42:206–20.[26] Vardiman JW, Brunning RD, Harris NL. WHO histological classificationof chronic myeloproliferative diseases. In: Jaffe ES, HarrisNL, Stein H, Vardiman JW, editors. World Health Organization classificationof tumors: tumours of the haematopoietic and lymphoidtissues. Lyon, France: International Agency for Research on Cancer(IARC) Press; 2001. p. 17–44.[27] Ohyashiki K, Aota Y, Akahane D, Gotoh A, Miyazawa K, KimuraY, et al. The JAK2 V617F tyrosine kinase mutation in myelodysplasticsyndromes (MDS) developing myelofibrosis indicates themyeloproliferative nature in a subset of MDS patients. Leukemia2005;19:2359–60.[28] Steensma DP, Dewald GW, Lasho TL, Powell HL, McClure RF,Levine RL, et al. The JAK2 V617F activating tyrosine kinase mutationis an infrequent event in both “atypical” myeloproliferativedisorders and myelodysplastic syndromes. Blood 2005;106:1207–9.[29] Bose S, Deininger M, Gora-Tybor J, Goldman JM, Melo JV.The presence of typical and atypical BCR-ABL fusion genes inleukocytes of normal individuals: biologic significance and implicationsfor the assessment of minimal residual disease. Blood1998;92:3362–7.[30] Goerttler PS, Steimle C, Marz E, Johansson PL, Andreasson B,Griesshammer M, et al. The Jak2V617F mutation, PRV-1 overexpression,and EEC formation define a similar cohort of MPD patients.Blood 2005;106:2862–4.[31] Tefferi A, Lasho TL, Schwager SM, Steensma DP, Mesa RA, Li CY,et al. The JAK2(V617F) tyrosine kinase mutation in myelofibrosiswith myeloid metaplasia: lineage specificity and clinical correlates.Br J Haematol 2005;131:320–8.[32] Campbell PJ, Griesshammer M, Dohner K, Dohner H, Kusec R,Hasselbalch HC, et al. The V617F mutation in JAK2 is associatedwith poorer survival in idiopathic myelofibrosis. Blood 2005.[33] Johan MF, Goodeve AC, Bowen DT, Frew ME, Reilly JT.JAK2 V617F Mutation is uncommon in chronic myelomonocyticleukaemia. Br J Haematol 2005;130:968.[34] Tono C, Xu G, Toki T, Takahashi Y, Sasaki S, Terui K, et al.JAK2 Val617Phe activating tyrosine kinase mutation in juvenilemyelomonocytic leukemia. Leukemia 2005;19:1843–4.