Int J Hematol (2015) 101:148–153 DOI 10.1007/s12185-014-1721-9
ORIGINAL ARTICLE
Consequences of the JAK2V617F allele burden for the prediction of transformation into myelofibrosis from polycythemia vera and essential thrombocythemia Shuichi Shirane · Marito Araki · Soji Morishita · Yoko Edahiro · Yoshitaka Sunami · Yumi Hironaka · Masaaki Noguchi · Michiaki Koike · Eriko Sato · Akimichi Ohsaka · Norio Komatsu
Received: 14 October 2014 / Revised: 27 November 2014 / Accepted: 28 November 2014 / Published online: 19 December 2014 © The Japanese Society of Hematology 2014
Abstract Patients diagnosed with polycythemia vera (PV) or essential thrombocythemia (ET) sometimes suffer transformation of the disease into myelofibrosis (MF), which is associated with a poorer prognosis. This study investigated the prognostic value of the allele burden of JAK2V617F, a somatic driver mutation in these diseases, by comparing the allele burden between formalin-fixed paraffin-embedded bone marrow collected at initial diagnosis and peripheral blood from follow-up visits. Although the annual changes in the JAK2V617F allele burden were comparable between MF-transformed (n = 11) and untransformed (n = 23) patients, the burden was significantly increased in MF-transformed patients exhibiting a longer disease duration than untransformed patients. Furthermore, MF transformation was only observed in patients Electronic supplementary material The online version of this article (doi:10.1007/s12185-014-1721-9) contains supplementary material, which is available to authorized users. S. Shirane · Y. Edahiro · Y. Sunami · Y. Hironaka · N. Komatsu (*) Department of Hematology, Juntendo University Graduate School of Medicine, 2‑1‑1 Hongo, Bunkyo‑ku, Tokyo 113‑8421, Japan e-mail: [email protected] M. Araki · S. Morishita · A. Ohsaka Department of Transfusion Medicine and Stem Cell Regulation, Juntendo University Graduate School of Medicine, Tokyo, Japan M. Noguchi Juntendo University Urayasu Hospital, Chiba, Japan M. Koike Juntendo University Shizuoka Hospital, Shizuoka, Japan E. Sato Juntendo University Nerima Hospital, Tokyo, Japan
13
whose JAK2V617F allele burden exceeded the mean values for each disease (PV, 71.7 %; ET, 35.5 %) at initial diagnosis or during follow-up. Finally, we showed that hydroxycarbamide treatment exerted neither a preventive effect on MF transformation nor a suppressive effect on the increased JAK2V617F allele burden. In conclusion, a high JAK2V617F allele burden at initial diagnosis or during follow-up is predictive of MF transformation in PV and ET. Therefore, routine measurement of the JAK2V617F allele burden using an accurate assay system is recommended to predict MF transformation. Keywords Polycythemia vera (PV) · Essential thrombocythemia (ET) · Secondary myelofibrosis · Myeloproliferative neoplasms (MPN) · JAK2V617F · Allele burden
Introduction A subset of patients diagnosed with polycythemia vera (PV) or essential thrombocythemia (ET), which is a subgroup of myeloproliferative neoplasms (MPN), paradoxically develops secondary myelofibrosis (MF) [1]. Secondary MF is strongly associated with a poorer prognosis, and patients suffer a low quality of life due to the associated anemia and splenomegaly. Therefore, the prognosis for MF transformation is a critical clinical issue [2, 3]. The driver mutation JAK2V617F was identified in a majority of MPN patients [4–7], and the potential of the JAK2V617F allele burden as a prognostic marker for MF transformation was studied. Homozygous status or a high mutant allele burden of JAK2V617F at the initial diagnosis or study entry was associated with MF transformation in both PV [8–12] and
149
Consequences of the JAK2V617F allele burden
ET [12]. However, the prognostic value of a chronological change in the JAK2V617F allele burden during follow-up was found to vary between studies. An association between an increased JAK2V617F allele burden during follow-up and MF transformation has been described [13], although another study reported the opposite result [14]. One drawback of these studies is that bone marrow (BM) biopsy was not performed in all cases, and the possibility that MF transformation occurred more frequently in patients who were defined as untransformed without BM observations therefore remains. Thus, whether chronological changes in the allele burden have prognostic value for MF transformation in PV and ET patients is controversial. We previously developed an alternative-binding probe competitive polymerase chain reaction (ABC-PCR) method that determines the JAK2V617F allele burden more accurately compared with other quantitative measurement methods, such as allele-specific PCR [15]. We used this method to assess the predictive value of the JAK2V617F allele burden for MF transformation in a Japanese MPN cohort in whom MF transformation was examined through BM biopsy.
Materials and methods Sample collection and preparation We analyzed a cohort of 34 Japanese MPN patients who were followed up at the Department of Hematology at the Juntendo University School of Medicine or at partner institutions in Japan. The patients were diagnosed with a clinical record at the initial diagnosis, and retrospective identification of the JAK2V617F mutation was performed in formalin-fixed paraffin-embedded (FFPE)-BM following the WHO 2008 criteria. The diagnosis of MF transformation was defined using the criteria of the International Working Group on Myelofibrosis Research and Treatment [16]. This study was conducted in accordance with the Declaration of Helsinki, and the ethics committee of the Juntendo University School of Medicine approved this study (IRB #2012208 and #2013020). Patients provided written informed consent, and genomic DNA was prepared from peripheral blood (PB) collected during followup and from FFPE-BM obtained at the initial diagnosis, using the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) and the QIAamp DNA FFPE Tissue Kit (Qiagen), respectively. The DNA concentration was measured using a NanoDrop LITE spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA), and the samples were stored at −80 °C until use. Measurement of the JAK2V617F allele burden The JAK2V617F allele burden was first determined via ABC-PCR [15]. In brief, 10–100 ng of genomic DNA
was subjected to 50 cycles of PCR (Titanium Taq PCR kit, Takara), with primers and a fluorescence-conjugated ABprobe, in a CFX-96 real-time PCR system (Bio-Rad, Hercules, USA). The allele burden was determined from the detected fluorescence intensity based on a standard curve plotted for a control with a known JAK2V617F allele burden. ABC-PCR accurately determines JAK2V617F allele burdens above 10 %, but allele-specific PCR (AS-PCR) is more accurate for the measurement of allele burdens below 10 %. Therefore, AS-PCR was performed when the allele burden determined using ABC-PCR was below 10 %; for these assays, genomic DNA (1 µl) was mixed with Universal PCR Master Mix (Applied Biosystems), primers and a TaqMan probe, and a total of 50 PCR cycles were performed, as previously reported [17]. Statistics Statistical values for the comparison of allele burdens between MF-transformed and untransformed patients were examined using Student’s t test. Statistical values for chronological changes in the allele burdens for hydroxycarbamide (hydroxyurea; HU)-treated and untreated patients were determined using a mixed-model ANCOVA with JMP PRO11 (SAS Institute, Cary, USA). The MF transformation frequencies between subgroups defined in the text were evaluated using Fisher’s exact test, and p values below 0.05 were considered significant.
Results and discussion Comparison of the JAK2V617F allele burden between PB and BM samples We performed a retrospective analysis of chronological changes in the JAK2V617F allele burden in MPN patients with or without MF transformation over a long period of time to examine the potential of the JAK2V617F allele burden to predict MF transformation. Genomic DNA from PB collected at initial diagnosis was not available for most of the patients, and we therefore intended to measure the JAK2V617F allele burden in genomic DNA obtained from FFPE-BM collected at the initial diagnosis. During followup, we routinely examined the JAK2V617F allele burden in genomic DNA from PB, but not BM, which allowed us to determine chronological changes in the JAK2V617F allele burden obtained from two different tissues. Therefore, we first examined whether the JAK2V617F allele burden in BM and PB was comparable when the biopsies were taken at the same time. We purified a set of genomic DNAs from FFPE-BM and PB that were collected within a 3-month period from the
13
150
Fig. 1 Correlation of the JAK2V617F allele burden between PB and FFPE-BM. Each dot represents a patient’s JAK2V617F allele burden in genomic DNA purified from PB (x-axis) and FFPE-BM (y-axis) within 3 months. The line represents a fitted regression line (y = 0.8866x + 6.2413, R2 = 0.966) and reveals a strong correlation
same patient (PV, n = 8; ET, n = 20) and determined the allele burden in both types of tissue samples (see “Materials and methods”). Figure 1 shows that the mean JAK2V617F allele burdens in PB and FFPE-BM were strongly correlated (n = 28, R2 = 0.97), which was recently reported by other groups as well [18–20]. Therefore, the baseline mutant allele burden was determined from FFPE-BM in subsequent analyses, which was then compared with the allele burden detected in PB during the duration of the disease. Assessment of the prognostic value of the JAK2V617F allele burden at initial diagnosis for MF transformation We purified genomic DNA from FFPE-BM (initial diagnosis) and PB (follow-up) and determined the allele burden (Supplemental Fig. 1 for individual data) for 14 PV and 20 ET patients, who were defined based on the WHO 2008 MPN criteria at the initial diagnosis (the clinical features at the initial diagnosis are summarized in Supplemental Table 1) and showed a disease duration of more than 1 year (mean duration 69.2 months), to investigate the correlation between the JAK2V617F allele burden and the frequency of MF transformation. Based on BM biopsy examination of all 34 MPN patients, we diagnosed 6 PV and 5 ET patients who developed secondary MF. The higher frequency of MF transformation observed in this cohort compared with other MPN cohorts [8, 10–14, 21–24] was likely because the BM biopsies were performed in patients with MF transformation suspected from clinical symptoms or who exhibited a longer disease duration. Several reports have indicated that PV patients with a higher JAK2V617F allele burden at initial diagnosis show
13
S. Shirane et al.
a higher risk for MF transformation (see “Introduction”). Therefore, we sub-classified the PV patients (n = 14) in our cohort based on the criteria of JAK2 mutant allele burdens of 50 % [8, 10, 13] and 80 % [9] used in previous studies and assessed the frequency of MF transformation. By either criterion, we found that there was no significant difference in terms of MF transformation between PV patients exhibiting high and low mutant allele burdens at the initial diagnosis (p = 1.000 or 0.277, respectively). The inconsistency between our data and previous data is presumably due to differences in the determination of the JAK2V61F allele burden, as evident in the wide range of mean allele burdens (64.8–98.2 %) associated with PV in different studies [6, 14]. To avoid the use of a fixed value from other cohorts in the classification of our patients, we sub-classified PV and ET patients according to the mean values of the JAK2V617F allele burden (PV, 71.7 %; ET, 35.5 %) determined using identical measurement techniques in a large Japanese MPN cohort [25]. We observed that patients exhibiting a higher JAK2 mutant allele burden than the mean value for each disease exhibited a significantly higher risk of MF transformation in ET (p = 0.033), but not in PV (p = 0.165). These data imply that a high JAK2 mutant allele burden at the initial diagnosis is a prognostic marker for MF transformation in ET, but not PV. An increased JAK2V617F allele burden during a long observation period is associated with MF‑transformed patients To further investigate the prognostic value of the JAK2V617F allele burden for MF transformation, we examined chronological changes in the allele burden during the disease duration. By comparing the JAK2V617F allele burden between the initial and last diagnoses, we found that the allele burden was significantly increased upon MF transformation compared with the initial diagnosis (Fig. 2a) (p = 0.004). Furthermore, the increase in the JAK2V617F allele burden in the MF-transformed patients (6 PV and 5 ET) was 19.5 ± 17.3 %, which was significantly higher than in patients without MF transformation (8 PV, 15 ET), in whom the increase was 3.9 ± 16.1 % (Fig. 2b) (p = 0.014). However, the disease duration in patients showing MF transformation (8.1 ± 4.3 years) was significantly longer than in patients without transformation (4.4 ± 2.4 years), which suggests that the JAK2V617F allele burden may increase over time, independent of MF transformation. As suspected, the annual change in the allele burden showed no significant difference between MF-transformed and untransformed patients (Fig. 2c) (p = 0.209). These observations suggest that the JAK2V617F allele burden increases over a long period
Consequences of the JAK2V617F allele burden
151
Fig. 2 Chronological changes in the JAK2V617F allele burden during MF transformation. a The JAK2V617F allele burden at initial and last diagnosis was plotted for MF-transformed patients (n = 11). b Chronological changes in the JAK2V617F allele burden were compared between MF-transformed (n = 11) and untransformed (n = 23) patients. c The calculated annual changes in the JAK2V617F allele
burden were compared between MF-transformed and untransformed patients. Boxes represent the interquartile range containing 50 % of the subjects. The open symbol indicates the mean value; the horizontal line inside the box marks the median; and the bars show the upper and lower range of values
and is associated with an increased number of incidents of MF transformation, as previously reported [21, 26]. In addition, because we did not routinely perform bone marrow biopsies to examine MF transformation, it is not clear whether the JAK2V617F allele burden increases before MF transformation. To address this question and to define the prognostic value of the JAK2V617F allele burden for MF transformation, a prospective study involving routine bone marrow biopsy and measurement of the allele burden is required.
JAK2V617F allele burden is directly associated with MF transformation, the prognostic potential of JAK2V617F should be examined in a prospective study (see previous section). However, our findings suggest that an increased allele burden in PB may be an indicator for MF transformation in MPN patients exhibiting a lower allele burden at the initial diagnosis. Such a diagnosis should lighten the patient’s burden for invasive bone marrow biopsy and may identify MF transformation before the appearance of clinical symptoms such as anemia.
A higher JAK2V617F allele burden at initial diagnosis or during follow‑up is a risk factor for MF transformation
HU treatment does not affect the JAK2V617F allele burden or MF transformation
Because we found a correlation between an increased JAK2V617F allele burden and MF transformation (Fig. 2a), we classified all of the patients into two groups according to the chronological change or the baseline value of the JAK2V617F allele burden. In group A (n = 26), either the allele burden was higher at initial diagnosis than the mean values for each disease (Supplemental Figure 1A), or the allele burden was lower than the mean values but later increased to a value higher than the mean values during follow-up (Supplemental Figure 1B). In group B (n = 8), the allele burden was lower than the mean values for each disease throughout the study (Supplemental Figure 1C). Under this classification, MF transformation was more frequently observed in group A (11/26) than group B (0/8) (Fig. 3) (p = 0.034). This analysis implies that a JAK2V617F allele burden higher than the mean value for each disease at initial diagnosis or a dynamic increase in the allele burden during the follow-up period could be a predictive factor for MF transformation, similar to the results of a study performed in Caucasian populations [13]. Again, because it is not clear whether the increase in the
Some patients in our cohort were treated with HU, which is a cytoreductive agent with the potential to lower the tumor burden [27–30]. Therefore, we examined whether HU treatment modulates the JAK2V617F allele burden and the frequency of MF transformation. The change in the JAK2V617F allele burden in the HU-treated group (n = 16) was 10.0 ± 20.0 %, while that in the untreated group was 7.8 ± 16.1 % (n = 18) (Fig. 4). There was no significant difference in the change in the allele burden between these two groups (p = 0.740), which indicates that HU treatment does not affect the JAK2V617F allele burden, as previously reported [9, 31–33]. MF transformation was observed in 7 of 16 and 4 of 18 HU-treated and untreated patients, respectively. There was no significant difference in the frequencies of MF transformation detected, despite the small cohort size (p = 0.166), which confirms that HU had no preventative effect against MF transformation, as proposed previously [8]. In summary, we report the following results: (1) an increased JAK2V617F allele burden is associated with MF transformation (Fig. 2); (2) the annual change in the
13
152
Fig. 3 A schematic representation of patient classification based on the JAK2V617F allele burden during the disease duration and its association with MF transformation. MPN patients were classified based on JAK2V617F allele burden at initial diagnosis and its chronological changes, as follows. Group A patients exhibiting an allele burden higher than the mean value for each disease in a Japanese cohort
S. Shirane et al.
(PV, 71.7 %; ET, 35.5 %) at initial diagnosis or patients exhibiting an allele burden lower than the mean value at initial diagnosis in whom the burden increased to a value higher than the mean value during follow-up; and group B patients exhibiting allele burdens lower than the mean value throughout the disease duration
allele burden is not significantly different between MFtransformed and untransformed patients; (3) a higher allele burden than the mean burden for each disease at initial diagnosis or during follow-up is strongly associated with MF transformation; and (4) HU treatment has no preventive effect on MF transformation and no effect on changes in the JAK2V617F allele burden. Our findings suggest that an increase in the JAK2V617F allele burden promotes MF transformation in some, if not all, patients who develop secondary MF. We showed that the JAK2V617F allele burden in PB is strongly correlated with the burden in BM (Fig. 1). Although the prognostic value of the JAK2V617F allele burden needs to be assessed in a prospective study, routine measurement of the JAK2V617F allele burden in PB using an accurate assay system is recommended to predict secondary MF in patients with a low JAK2V617F allele burden.
Fig. 4 Comparison of changes in the JAK2V617F allele burden in HU-treated and untreated patients. The changes in the JAK2V617F allele burden in each patient between the initial (FFPE-BM) and last (PB) diagnoses were determined. The HU-treated (upper panel) and untreated (lower panel) patients were not significantly different (p = 0.740). Dotted and solid lines represent MF-transformed and untransformed patients, respectively
13
Acknowledgments We thank Kensuke Usuki (NTT Kanto Medical Center), Masafumi Ito (Japanese Red Cross Nagoya Daiichi Hospital), Makoto Kashimura (Matsudo Municipal Hospital), Shiro Tsuchiya (Soka Municipal Hospital) and Takao Hirano (Juntendo Nerima Hospital) for providing patient specimens and clinical information, Joe Matsuoka (Clinical Research Support Center, Graduate School of Medicine) for performing statistical analysis, and Satoshi Tsuneda (Department of Life Science and Medical Bioscience, Waseda University) for fruitful discussions. We also thank Kyoko Kubo, Kazuko Kawamura, Junko Enomoto, and Megumi Hasegawa for providing
Consequences of the JAK2V617F allele burden secretarial assistance and other members of the Department of Hematology for providing support in this study. We also acknowledge the Laboratory of Morphology and Image Analysis, Biomedical Research Center, Juntendo University Graduate School of Medicine, for technical assistance. This work was funded in part by JSPS (http://www. jsps.go.jp/english/e-grants/)KAKENHI grant #25860416 (SM) and Ministry of Education, Culture, Sports, Science and Technology, the Project for Development of Innovative Research on Cancer Therapeutics (NK, MA, SM, YE, YS). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. Conflict of interest SM and NK hold a patent related to the method for the measurement of the JAK2V617F allele burden that was used in the present study.
References 1. Mesa RA, et al. MPN-associated myelofibrosis (MPN-MF). Leuk Res. 2011;35:12–3. 2. Tam CS, et al. Dynamic model for predicting death within 12 months in patients with primary or post-polycythemia vera/essential thrombocythemia myelofibrosis. J Clin Oncol. 2009;27:5587–93. 3. Passamonti F, et al. A dynamic prognostic model to predict survival in post-polycythemia vera myelofibrosis. Blood. 2008;111:3383–7. 4. Levine RL, et al. Activating mutation in the tyrosine kinase JAK2 in polycythemia vera, essential thrombocythemia, and myeloid metaplasia with myelofibrosis. Cancer Cell. 2005;7:387–97. 5. Baxter EJ, et al. Acquired mutation of the tyrosine kinase JAK2 in human myeloproliferative disorders. Lancet. 2005;365:1054–61. 6. Kralovics R, et al. A gain-of-function mutation of JAK2 in myeloproliferative disorders. N Engl J Med. 2005;352:1779–90. 7. James C, et al. A unique clonal JAK2 mutation leading to constitutive signalling causes polycythaemia vera. Nature. 2005;434:1144–8. 8. Alvarez-Larran A, et al. Postpolycythaemic myelofibrosis: frequency and risk factors for this complication in 116 patients. Br J Haematol. 2009;146:504–9. 9. Silver RT, et al. JAK2(V617F) allele burden in polycythemia vera correlates with grade of myelofibrosis, but is not substantially affected by therapy. Leuk Res. 2011;35:177–82. 10. Passamonti F, et al. A prospective study of 338 patients with polycythemia vera: the impact of JAK2 (V617F) allele burden and leukocytosis on fibrotic or leukemic disease transformation and vascular complications. Leukemia. 2010;24:1574–9. 11. Tefferi A, et al. The clinical phenotype of wild-type, heterozygous, and homozygous JAK2V617F in polycythemia vera. Cancer. 2006;106:631–5. 12. Vannucchi AM, et al. Clinical profile of homozygous JAK2 617 V > F mutation in patients with polycythemia vera or essential thrombocythemia. Blood. 2007;110:840–6. 13. Alvarez-Larran A, et al. JAK2V617F monitoring in polycythemia vera and essential thrombocythemia: clinical usefulness for predicting myelofibrotic transformation and thrombotic events. Am J Hematol. 2014;89:517–23. 14. Hussein K, et al. JAK2(V617F) allele burden discriminates essential thrombocythemia from a subset of prefibrotic-stage primary myelofibrosis. Exp Hematol. 2009;37(1186–1193):e7. 15. Morishita S, et al. Alternately binding probe competitive PCR as a simple, cost-effective, and accurate quantification method for
153 JAK2V617F allele burden in myeloproliferative neoplasms. Leuk Res. 2011;3512:1632–6. 16. Barosi G, et al. Proposed criteria for the diagnosis of post-polycythemia vera and post-essential thrombocythemia myelofibrosis: a consensus statement from the International Working Group for Myelofibrosis Research and Treatment. Leukemia. 2008;22:437–8. 17. Rapado I, et al. Validity test study of JAK2 V617F and allele burden quantification in the diagnosis of myeloproliferative diseases. Ann Hematol. 2008;87:741–9. 18. Takahashi K, et al. JAK2 p. V617F detection and allele burden measurement in peripheral blood and bone marrow aspirates in patients with myeloproliferative neoplasms. Blood. 2013;122:3784–6. 19. Larsen TS, et al. Quantitative assessment of the JAK2 V617F allele burden: equivalent levels in peripheral blood and bone marrow. Leukemia. 2008;22:194–5. 20. Lange T, et al. JAK2 p. V617F allele burden in myeloproliferative neoplasms 1 month after allogeneic stem cell transplantation significantly predicts outcome and risk of relapse. Haematologica. 2013;98:722–8. 21. Passamonti F, et al. Prognostic factors for thrombosis, myelofibrosis, and leukemia in essential thrombocythemia: a study of 605 patients. Haematologica. 2008;93:1645–51. 22. Alvarez-Larran A, et al. Essential thrombocythemia in young individuals: frequency and risk factors for vascular events and evolution to myelofibrosis in 126 patients. Leukemia. 2007;21:1218–23. 23. Barbui T, et al. Survival and disease progression in essen tial thrombocythemia are significantly influenced by accurate morphologic diagnosis: an international study. J Clin Oncol. 2011;29:3179–84. 24. Koren-Michowitz M, et al. JAK2V617F allele burden is associated with transformation to myelofibrosis. Leuk Lymphoma. 2012;53:2210–3. 25. Edahiro Y, et al. JAK2V617F mutation status and allele burden in classical Ph-negative myeloproliferative neoplasms in Japan. Int J Hematol. 2014;99:625–34. 26. Marchioli R, et al. Vascular and neoplastic risk in a large cohort of patients with polycythemia vera. J Clin Oncol. 2005;23:2224–32. 27. Ricksten A, et al. Rapid decline of JAK2V617F levels during hydroxyurea treatment in patients with polycythemia vera and essential thrombocythemia. Haematologica. 2008;93:1260–1. 28. Theocharides A, et al. The allele burden of JAK2 mutations remains stable over several years in patients with myeloproliferative disorders. Haematologica. 2008;93:1890–3. 29. Girodon F, et al. Frequent reduction or absence of detection of the JAK2-mutated clone in JAK2V617F-positive patients within the first years of hydroxyurea therapy. Haematologica. 2008;93:1723–7. 30. Besses C, et al. Modulation of JAK2 V617F allele burden dynamics by hydroxycarbamide in polycythaemia vera and essential thrombocythaemia patients. Br J Haematol. 2011;152:413–9. 31. Larsen TS, et al. Limited efficacy of hydroxyurea in lowering of the JAK2 V617F allele burden. Hematology. 2009;14:11–5. 32. Antonioli E, et al. Hydroxyurea does not appreciably reduce JAK2 V617F allele burden in patients with polycythemia vera or essential thrombocythemia. Haematologica. 2010;95:1435–8. 33. Zalcberg IR, et al. Hydroxyurea dose impacts hematologic parameters in polycythemia vera and essential thrombocythemia but does not appreciably affect JAK2-V617F allele burden. Haematologica. 2011;96:e18–20.
13
ORIGINAL ARTICLE
Consequences of the JAK2V617F allele burden for the prediction of transformation into myelofibrosis from polycythemia vera and essential thrombocythemia Shuichi Shirane · Marito Araki · Soji Morishita · Yoko Edahiro · Yoshitaka Sunami · Yumi Hironaka · Masaaki Noguchi · Michiaki Koike · Eriko Sato · Akimichi Ohsaka · Norio Komatsu
Received: 14 October 2014 / Revised: 27 November 2014 / Accepted: 28 November 2014 / Published online: 19 December 2014 © The Japanese Society of Hematology 2014
Abstract Patients diagnosed with polycythemia vera (PV) or essential thrombocythemia (ET) sometimes suffer transformation of the disease into myelofibrosis (MF), which is associated with a poorer prognosis. This study investigated the prognostic value of the allele burden of JAK2V617F, a somatic driver mutation in these diseases, by comparing the allele burden between formalin-fixed paraffin-embedded bone marrow collected at initial diagnosis and peripheral blood from follow-up visits. Although the annual changes in the JAK2V617F allele burden were comparable between MF-transformed (n = 11) and untransformed (n = 23) patients, the burden was significantly increased in MF-transformed patients exhibiting a longer disease duration than untransformed patients. Furthermore, MF transformation was only observed in patients Electronic supplementary material The online version of this article (doi:10.1007/s12185-014-1721-9) contains supplementary material, which is available to authorized users. S. Shirane · Y. Edahiro · Y. Sunami · Y. Hironaka · N. Komatsu (*) Department of Hematology, Juntendo University Graduate School of Medicine, 2‑1‑1 Hongo, Bunkyo‑ku, Tokyo 113‑8421, Japan e-mail: [email protected] M. Araki · S. Morishita · A. Ohsaka Department of Transfusion Medicine and Stem Cell Regulation, Juntendo University Graduate School of Medicine, Tokyo, Japan M. Noguchi Juntendo University Urayasu Hospital, Chiba, Japan M. Koike Juntendo University Shizuoka Hospital, Shizuoka, Japan E. Sato Juntendo University Nerima Hospital, Tokyo, Japan
13
whose JAK2V617F allele burden exceeded the mean values for each disease (PV, 71.7 %; ET, 35.5 %) at initial diagnosis or during follow-up. Finally, we showed that hydroxycarbamide treatment exerted neither a preventive effect on MF transformation nor a suppressive effect on the increased JAK2V617F allele burden. In conclusion, a high JAK2V617F allele burden at initial diagnosis or during follow-up is predictive of MF transformation in PV and ET. Therefore, routine measurement of the JAK2V617F allele burden using an accurate assay system is recommended to predict MF transformation. Keywords Polycythemia vera (PV) · Essential thrombocythemia (ET) · Secondary myelofibrosis · Myeloproliferative neoplasms (MPN) · JAK2V617F · Allele burden
Introduction A subset of patients diagnosed with polycythemia vera (PV) or essential thrombocythemia (ET), which is a subgroup of myeloproliferative neoplasms (MPN), paradoxically develops secondary myelofibrosis (MF) [1]. Secondary MF is strongly associated with a poorer prognosis, and patients suffer a low quality of life due to the associated anemia and splenomegaly. Therefore, the prognosis for MF transformation is a critical clinical issue [2, 3]. The driver mutation JAK2V617F was identified in a majority of MPN patients [4–7], and the potential of the JAK2V617F allele burden as a prognostic marker for MF transformation was studied. Homozygous status or a high mutant allele burden of JAK2V617F at the initial diagnosis or study entry was associated with MF transformation in both PV [8–12] and
149
Consequences of the JAK2V617F allele burden
ET [12]. However, the prognostic value of a chronological change in the JAK2V617F allele burden during follow-up was found to vary between studies. An association between an increased JAK2V617F allele burden during follow-up and MF transformation has been described [13], although another study reported the opposite result [14]. One drawback of these studies is that bone marrow (BM) biopsy was not performed in all cases, and the possibility that MF transformation occurred more frequently in patients who were defined as untransformed without BM observations therefore remains. Thus, whether chronological changes in the allele burden have prognostic value for MF transformation in PV and ET patients is controversial. We previously developed an alternative-binding probe competitive polymerase chain reaction (ABC-PCR) method that determines the JAK2V617F allele burden more accurately compared with other quantitative measurement methods, such as allele-specific PCR [15]. We used this method to assess the predictive value of the JAK2V617F allele burden for MF transformation in a Japanese MPN cohort in whom MF transformation was examined through BM biopsy.
Materials and methods Sample collection and preparation We analyzed a cohort of 34 Japanese MPN patients who were followed up at the Department of Hematology at the Juntendo University School of Medicine or at partner institutions in Japan. The patients were diagnosed with a clinical record at the initial diagnosis, and retrospective identification of the JAK2V617F mutation was performed in formalin-fixed paraffin-embedded (FFPE)-BM following the WHO 2008 criteria. The diagnosis of MF transformation was defined using the criteria of the International Working Group on Myelofibrosis Research and Treatment [16]. This study was conducted in accordance with the Declaration of Helsinki, and the ethics committee of the Juntendo University School of Medicine approved this study (IRB #2012208 and #2013020). Patients provided written informed consent, and genomic DNA was prepared from peripheral blood (PB) collected during followup and from FFPE-BM obtained at the initial diagnosis, using the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) and the QIAamp DNA FFPE Tissue Kit (Qiagen), respectively. The DNA concentration was measured using a NanoDrop LITE spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA), and the samples were stored at −80 °C until use. Measurement of the JAK2V617F allele burden The JAK2V617F allele burden was first determined via ABC-PCR [15]. In brief, 10–100 ng of genomic DNA
was subjected to 50 cycles of PCR (Titanium Taq PCR kit, Takara), with primers and a fluorescence-conjugated ABprobe, in a CFX-96 real-time PCR system (Bio-Rad, Hercules, USA). The allele burden was determined from the detected fluorescence intensity based on a standard curve plotted for a control with a known JAK2V617F allele burden. ABC-PCR accurately determines JAK2V617F allele burdens above 10 %, but allele-specific PCR (AS-PCR) is more accurate for the measurement of allele burdens below 10 %. Therefore, AS-PCR was performed when the allele burden determined using ABC-PCR was below 10 %; for these assays, genomic DNA (1 µl) was mixed with Universal PCR Master Mix (Applied Biosystems), primers and a TaqMan probe, and a total of 50 PCR cycles were performed, as previously reported [17]. Statistics Statistical values for the comparison of allele burdens between MF-transformed and untransformed patients were examined using Student’s t test. Statistical values for chronological changes in the allele burdens for hydroxycarbamide (hydroxyurea; HU)-treated and untreated patients were determined using a mixed-model ANCOVA with JMP PRO11 (SAS Institute, Cary, USA). The MF transformation frequencies between subgroups defined in the text were evaluated using Fisher’s exact test, and p values below 0.05 were considered significant.
Results and discussion Comparison of the JAK2V617F allele burden between PB and BM samples We performed a retrospective analysis of chronological changes in the JAK2V617F allele burden in MPN patients with or without MF transformation over a long period of time to examine the potential of the JAK2V617F allele burden to predict MF transformation. Genomic DNA from PB collected at initial diagnosis was not available for most of the patients, and we therefore intended to measure the JAK2V617F allele burden in genomic DNA obtained from FFPE-BM collected at the initial diagnosis. During followup, we routinely examined the JAK2V617F allele burden in genomic DNA from PB, but not BM, which allowed us to determine chronological changes in the JAK2V617F allele burden obtained from two different tissues. Therefore, we first examined whether the JAK2V617F allele burden in BM and PB was comparable when the biopsies were taken at the same time. We purified a set of genomic DNAs from FFPE-BM and PB that were collected within a 3-month period from the
13
150
Fig. 1 Correlation of the JAK2V617F allele burden between PB and FFPE-BM. Each dot represents a patient’s JAK2V617F allele burden in genomic DNA purified from PB (x-axis) and FFPE-BM (y-axis) within 3 months. The line represents a fitted regression line (y = 0.8866x + 6.2413, R2 = 0.966) and reveals a strong correlation
same patient (PV, n = 8; ET, n = 20) and determined the allele burden in both types of tissue samples (see “Materials and methods”). Figure 1 shows that the mean JAK2V617F allele burdens in PB and FFPE-BM were strongly correlated (n = 28, R2 = 0.97), which was recently reported by other groups as well [18–20]. Therefore, the baseline mutant allele burden was determined from FFPE-BM in subsequent analyses, which was then compared with the allele burden detected in PB during the duration of the disease. Assessment of the prognostic value of the JAK2V617F allele burden at initial diagnosis for MF transformation We purified genomic DNA from FFPE-BM (initial diagnosis) and PB (follow-up) and determined the allele burden (Supplemental Fig. 1 for individual data) for 14 PV and 20 ET patients, who were defined based on the WHO 2008 MPN criteria at the initial diagnosis (the clinical features at the initial diagnosis are summarized in Supplemental Table 1) and showed a disease duration of more than 1 year (mean duration 69.2 months), to investigate the correlation between the JAK2V617F allele burden and the frequency of MF transformation. Based on BM biopsy examination of all 34 MPN patients, we diagnosed 6 PV and 5 ET patients who developed secondary MF. The higher frequency of MF transformation observed in this cohort compared with other MPN cohorts [8, 10–14, 21–24] was likely because the BM biopsies were performed in patients with MF transformation suspected from clinical symptoms or who exhibited a longer disease duration. Several reports have indicated that PV patients with a higher JAK2V617F allele burden at initial diagnosis show
13
S. Shirane et al.
a higher risk for MF transformation (see “Introduction”). Therefore, we sub-classified the PV patients (n = 14) in our cohort based on the criteria of JAK2 mutant allele burdens of 50 % [8, 10, 13] and 80 % [9] used in previous studies and assessed the frequency of MF transformation. By either criterion, we found that there was no significant difference in terms of MF transformation between PV patients exhibiting high and low mutant allele burdens at the initial diagnosis (p = 1.000 or 0.277, respectively). The inconsistency between our data and previous data is presumably due to differences in the determination of the JAK2V61F allele burden, as evident in the wide range of mean allele burdens (64.8–98.2 %) associated with PV in different studies [6, 14]. To avoid the use of a fixed value from other cohorts in the classification of our patients, we sub-classified PV and ET patients according to the mean values of the JAK2V617F allele burden (PV, 71.7 %; ET, 35.5 %) determined using identical measurement techniques in a large Japanese MPN cohort [25]. We observed that patients exhibiting a higher JAK2 mutant allele burden than the mean value for each disease exhibited a significantly higher risk of MF transformation in ET (p = 0.033), but not in PV (p = 0.165). These data imply that a high JAK2 mutant allele burden at the initial diagnosis is a prognostic marker for MF transformation in ET, but not PV. An increased JAK2V617F allele burden during a long observation period is associated with MF‑transformed patients To further investigate the prognostic value of the JAK2V617F allele burden for MF transformation, we examined chronological changes in the allele burden during the disease duration. By comparing the JAK2V617F allele burden between the initial and last diagnoses, we found that the allele burden was significantly increased upon MF transformation compared with the initial diagnosis (Fig. 2a) (p = 0.004). Furthermore, the increase in the JAK2V617F allele burden in the MF-transformed patients (6 PV and 5 ET) was 19.5 ± 17.3 %, which was significantly higher than in patients without MF transformation (8 PV, 15 ET), in whom the increase was 3.9 ± 16.1 % (Fig. 2b) (p = 0.014). However, the disease duration in patients showing MF transformation (8.1 ± 4.3 years) was significantly longer than in patients without transformation (4.4 ± 2.4 years), which suggests that the JAK2V617F allele burden may increase over time, independent of MF transformation. As suspected, the annual change in the allele burden showed no significant difference between MF-transformed and untransformed patients (Fig. 2c) (p = 0.209). These observations suggest that the JAK2V617F allele burden increases over a long period
Consequences of the JAK2V617F allele burden
151
Fig. 2 Chronological changes in the JAK2V617F allele burden during MF transformation. a The JAK2V617F allele burden at initial and last diagnosis was plotted for MF-transformed patients (n = 11). b Chronological changes in the JAK2V617F allele burden were compared between MF-transformed (n = 11) and untransformed (n = 23) patients. c The calculated annual changes in the JAK2V617F allele
burden were compared between MF-transformed and untransformed patients. Boxes represent the interquartile range containing 50 % of the subjects. The open symbol indicates the mean value; the horizontal line inside the box marks the median; and the bars show the upper and lower range of values
and is associated with an increased number of incidents of MF transformation, as previously reported [21, 26]. In addition, because we did not routinely perform bone marrow biopsies to examine MF transformation, it is not clear whether the JAK2V617F allele burden increases before MF transformation. To address this question and to define the prognostic value of the JAK2V617F allele burden for MF transformation, a prospective study involving routine bone marrow biopsy and measurement of the allele burden is required.
JAK2V617F allele burden is directly associated with MF transformation, the prognostic potential of JAK2V617F should be examined in a prospective study (see previous section). However, our findings suggest that an increased allele burden in PB may be an indicator for MF transformation in MPN patients exhibiting a lower allele burden at the initial diagnosis. Such a diagnosis should lighten the patient’s burden for invasive bone marrow biopsy and may identify MF transformation before the appearance of clinical symptoms such as anemia.
A higher JAK2V617F allele burden at initial diagnosis or during follow‑up is a risk factor for MF transformation
HU treatment does not affect the JAK2V617F allele burden or MF transformation
Because we found a correlation between an increased JAK2V617F allele burden and MF transformation (Fig. 2a), we classified all of the patients into two groups according to the chronological change or the baseline value of the JAK2V617F allele burden. In group A (n = 26), either the allele burden was higher at initial diagnosis than the mean values for each disease (Supplemental Figure 1A), or the allele burden was lower than the mean values but later increased to a value higher than the mean values during follow-up (Supplemental Figure 1B). In group B (n = 8), the allele burden was lower than the mean values for each disease throughout the study (Supplemental Figure 1C). Under this classification, MF transformation was more frequently observed in group A (11/26) than group B (0/8) (Fig. 3) (p = 0.034). This analysis implies that a JAK2V617F allele burden higher than the mean value for each disease at initial diagnosis or a dynamic increase in the allele burden during the follow-up period could be a predictive factor for MF transformation, similar to the results of a study performed in Caucasian populations [13]. Again, because it is not clear whether the increase in the
Some patients in our cohort were treated with HU, which is a cytoreductive agent with the potential to lower the tumor burden [27–30]. Therefore, we examined whether HU treatment modulates the JAK2V617F allele burden and the frequency of MF transformation. The change in the JAK2V617F allele burden in the HU-treated group (n = 16) was 10.0 ± 20.0 %, while that in the untreated group was 7.8 ± 16.1 % (n = 18) (Fig. 4). There was no significant difference in the change in the allele burden between these two groups (p = 0.740), which indicates that HU treatment does not affect the JAK2V617F allele burden, as previously reported [9, 31–33]. MF transformation was observed in 7 of 16 and 4 of 18 HU-treated and untreated patients, respectively. There was no significant difference in the frequencies of MF transformation detected, despite the small cohort size (p = 0.166), which confirms that HU had no preventative effect against MF transformation, as proposed previously [8]. In summary, we report the following results: (1) an increased JAK2V617F allele burden is associated with MF transformation (Fig. 2); (2) the annual change in the
13
152
Fig. 3 A schematic representation of patient classification based on the JAK2V617F allele burden during the disease duration and its association with MF transformation. MPN patients were classified based on JAK2V617F allele burden at initial diagnosis and its chronological changes, as follows. Group A patients exhibiting an allele burden higher than the mean value for each disease in a Japanese cohort
S. Shirane et al.
(PV, 71.7 %; ET, 35.5 %) at initial diagnosis or patients exhibiting an allele burden lower than the mean value at initial diagnosis in whom the burden increased to a value higher than the mean value during follow-up; and group B patients exhibiting allele burdens lower than the mean value throughout the disease duration
allele burden is not significantly different between MFtransformed and untransformed patients; (3) a higher allele burden than the mean burden for each disease at initial diagnosis or during follow-up is strongly associated with MF transformation; and (4) HU treatment has no preventive effect on MF transformation and no effect on changes in the JAK2V617F allele burden. Our findings suggest that an increase in the JAK2V617F allele burden promotes MF transformation in some, if not all, patients who develop secondary MF. We showed that the JAK2V617F allele burden in PB is strongly correlated with the burden in BM (Fig. 1). Although the prognostic value of the JAK2V617F allele burden needs to be assessed in a prospective study, routine measurement of the JAK2V617F allele burden in PB using an accurate assay system is recommended to predict secondary MF in patients with a low JAK2V617F allele burden.
Fig. 4 Comparison of changes in the JAK2V617F allele burden in HU-treated and untreated patients. The changes in the JAK2V617F allele burden in each patient between the initial (FFPE-BM) and last (PB) diagnoses were determined. The HU-treated (upper panel) and untreated (lower panel) patients were not significantly different (p = 0.740). Dotted and solid lines represent MF-transformed and untransformed patients, respectively
13
Acknowledgments We thank Kensuke Usuki (NTT Kanto Medical Center), Masafumi Ito (Japanese Red Cross Nagoya Daiichi Hospital), Makoto Kashimura (Matsudo Municipal Hospital), Shiro Tsuchiya (Soka Municipal Hospital) and Takao Hirano (Juntendo Nerima Hospital) for providing patient specimens and clinical information, Joe Matsuoka (Clinical Research Support Center, Graduate School of Medicine) for performing statistical analysis, and Satoshi Tsuneda (Department of Life Science and Medical Bioscience, Waseda University) for fruitful discussions. We also thank Kyoko Kubo, Kazuko Kawamura, Junko Enomoto, and Megumi Hasegawa for providing
Consequences of the JAK2V617F allele burden secretarial assistance and other members of the Department of Hematology for providing support in this study. We also acknowledge the Laboratory of Morphology and Image Analysis, Biomedical Research Center, Juntendo University Graduate School of Medicine, for technical assistance. This work was funded in part by JSPS (http://www. jsps.go.jp/english/e-grants/)KAKENHI grant #25860416 (SM) and Ministry of Education, Culture, Sports, Science and Technology, the Project for Development of Innovative Research on Cancer Therapeutics (NK, MA, SM, YE, YS). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. Conflict of interest SM and NK hold a patent related to the method for the measurement of the JAK2V617F allele burden that was used in the present study.
References 1. Mesa RA, et al. MPN-associated myelofibrosis (MPN-MF). Leuk Res. 2011;35:12–3. 2. Tam CS, et al. Dynamic model for predicting death within 12 months in patients with primary or post-polycythemia vera/essential thrombocythemia myelofibrosis. J Clin Oncol. 2009;27:5587–93. 3. Passamonti F, et al. A dynamic prognostic model to predict survival in post-polycythemia vera myelofibrosis. Blood. 2008;111:3383–7. 4. Levine RL, et al. Activating mutation in the tyrosine kinase JAK2 in polycythemia vera, essential thrombocythemia, and myeloid metaplasia with myelofibrosis. Cancer Cell. 2005;7:387–97. 5. Baxter EJ, et al. Acquired mutation of the tyrosine kinase JAK2 in human myeloproliferative disorders. Lancet. 2005;365:1054–61. 6. Kralovics R, et al. A gain-of-function mutation of JAK2 in myeloproliferative disorders. N Engl J Med. 2005;352:1779–90. 7. James C, et al. A unique clonal JAK2 mutation leading to constitutive signalling causes polycythaemia vera. Nature. 2005;434:1144–8. 8. Alvarez-Larran A, et al. Postpolycythaemic myelofibrosis: frequency and risk factors for this complication in 116 patients. Br J Haematol. 2009;146:504–9. 9. Silver RT, et al. JAK2(V617F) allele burden in polycythemia vera correlates with grade of myelofibrosis, but is not substantially affected by therapy. Leuk Res. 2011;35:177–82. 10. Passamonti F, et al. A prospective study of 338 patients with polycythemia vera: the impact of JAK2 (V617F) allele burden and leukocytosis on fibrotic or leukemic disease transformation and vascular complications. Leukemia. 2010;24:1574–9. 11. Tefferi A, et al. The clinical phenotype of wild-type, heterozygous, and homozygous JAK2V617F in polycythemia vera. Cancer. 2006;106:631–5. 12. Vannucchi AM, et al. Clinical profile of homozygous JAK2 617 V > F mutation in patients with polycythemia vera or essential thrombocythemia. Blood. 2007;110:840–6. 13. Alvarez-Larran A, et al. JAK2V617F monitoring in polycythemia vera and essential thrombocythemia: clinical usefulness for predicting myelofibrotic transformation and thrombotic events. Am J Hematol. 2014;89:517–23. 14. Hussein K, et al. JAK2(V617F) allele burden discriminates essential thrombocythemia from a subset of prefibrotic-stage primary myelofibrosis. Exp Hematol. 2009;37(1186–1193):e7. 15. Morishita S, et al. Alternately binding probe competitive PCR as a simple, cost-effective, and accurate quantification method for
153 JAK2V617F allele burden in myeloproliferative neoplasms. Leuk Res. 2011;3512:1632–6. 16. Barosi G, et al. Proposed criteria for the diagnosis of post-polycythemia vera and post-essential thrombocythemia myelofibrosis: a consensus statement from the International Working Group for Myelofibrosis Research and Treatment. Leukemia. 2008;22:437–8. 17. Rapado I, et al. Validity test study of JAK2 V617F and allele burden quantification in the diagnosis of myeloproliferative diseases. Ann Hematol. 2008;87:741–9. 18. Takahashi K, et al. JAK2 p. V617F detection and allele burden measurement in peripheral blood and bone marrow aspirates in patients with myeloproliferative neoplasms. Blood. 2013;122:3784–6. 19. Larsen TS, et al. Quantitative assessment of the JAK2 V617F allele burden: equivalent levels in peripheral blood and bone marrow. Leukemia. 2008;22:194–5. 20. Lange T, et al. JAK2 p. V617F allele burden in myeloproliferative neoplasms 1 month after allogeneic stem cell transplantation significantly predicts outcome and risk of relapse. Haematologica. 2013;98:722–8. 21. Passamonti F, et al. Prognostic factors for thrombosis, myelofibrosis, and leukemia in essential thrombocythemia: a study of 605 patients. Haematologica. 2008;93:1645–51. 22. Alvarez-Larran A, et al. Essential thrombocythemia in young individuals: frequency and risk factors for vascular events and evolution to myelofibrosis in 126 patients. Leukemia. 2007;21:1218–23. 23. Barbui T, et al. Survival and disease progression in essen tial thrombocythemia are significantly influenced by accurate morphologic diagnosis: an international study. J Clin Oncol. 2011;29:3179–84. 24. Koren-Michowitz M, et al. JAK2V617F allele burden is associated with transformation to myelofibrosis. Leuk Lymphoma. 2012;53:2210–3. 25. Edahiro Y, et al. JAK2V617F mutation status and allele burden in classical Ph-negative myeloproliferative neoplasms in Japan. Int J Hematol. 2014;99:625–34. 26. Marchioli R, et al. Vascular and neoplastic risk in a large cohort of patients with polycythemia vera. J Clin Oncol. 2005;23:2224–32. 27. Ricksten A, et al. Rapid decline of JAK2V617F levels during hydroxyurea treatment in patients with polycythemia vera and essential thrombocythemia. Haematologica. 2008;93:1260–1. 28. Theocharides A, et al. The allele burden of JAK2 mutations remains stable over several years in patients with myeloproliferative disorders. Haematologica. 2008;93:1890–3. 29. Girodon F, et al. Frequent reduction or absence of detection of the JAK2-mutated clone in JAK2V617F-positive patients within the first years of hydroxyurea therapy. Haematologica. 2008;93:1723–7. 30. Besses C, et al. Modulation of JAK2 V617F allele burden dynamics by hydroxycarbamide in polycythaemia vera and essential thrombocythaemia patients. Br J Haematol. 2011;152:413–9. 31. Larsen TS, et al. Limited efficacy of hydroxyurea in lowering of the JAK2 V617F allele burden. Hematology. 2009;14:11–5. 32. Antonioli E, et al. Hydroxyurea does not appreciably reduce JAK2 V617F allele burden in patients with polycythemia vera or essential thrombocythemia. Haematologica. 2010;95:1435–8. 33. Zalcberg IR, et al. Hydroxyurea dose impacts hematologic parameters in polycythemia vera and essential thrombocythemia but does not appreciably affect JAK2-V617F allele burden. Haematologica. 2011;96:e18–20.
13
