A Novel t(1;9)(p36;p24.1) JAK2 Translocation and Review of the Literature

The JAK (Janus kinase) proteins are a family of cytoplasmic tyrosine kinases involved in the JAK-STAT signaling pathway and are essential in maintaining normal hematopoiesis. The JAK2gene, located on chromosome 9p24, encodes for a receptor predominantly responsive to type I cytokine ligands, including erythropoietin, thrombopoietin, and granulocyte-macrophage colony-stimulating factor. Ligand binding to JAK2 leads to autophosphorylation and activation of STAT (signal transducers and activators of transcription) proteins, which mediate the expression of genes involved in hematopoiesis [1].

Constitutive activation of the JAK-STAT pathway through the acquired point activating mutation on exon 14 (JAK2V617F) has been implicated in the pathogenesis of myeloproliferative neoplasms (MPNs). This mutation is present in 98% of polycythemia vera (PV), 40–50% of essential thrombocythemia (ET), and 50–60% of primary myelofibrosis (PMF) [2, 3], and is present in 20% of patients with non-classical MPNs [4]. Additionally, a subset of patients with JAK2V617F-negative MPNs is found to have translocations involving JAK2, which result in gene fusion products that lead to JAK2 amplification or constitutive activation of the tyrosine kinase. These translocations are not limited to the MPNs, but have also been identified in de novo leukemias of both myeloid and lymphoid lineages (Table 2).

Here, we describe a case of JAK2V617F-negative PV with a novel JAK2translocation. We also provide a review of the known JAK2 translocations associated with PV and other MPNs.

A 52-year-old woman initially presented in 2008 with symptoms of headaches, dizziness, fatigue, shortness of breath, and paresthesias of the hands and feet with an HCT of 61% and normal WBC and PLT counts. Bone marrow biopsy showed a hypercellular marrow (95%) with increased megakaryocytes in clusters, without reticulin staining. The patient was found to meet the World Health Organization’s (WHO) diagnostic criteria for JAK2V617F mutation-negative PV and was initiated on treatment with aspirin and therapeutic phlebotomy, to maintain the HCT <42%.

A year later, the patient developed leukocytosis (WBC 30 × 10⁹/L) and thrombocytosis (PLT 600,000/L). She was started on hydroxyurea, with subsequent improvement in leukocytosis and thrombocytosis, but developed treatment-emergent anemia. Repeat bone marrow biopsy showed a hypercellular marrow (100%) with trilineage hematopoiesis, marked granulocytic hyperplasia, increased immature forms, markedly increased eosinophils (31%), and 7% blasts. Peripheral blood flow cytometry showed a small CD33+ and CD34+ myeloblast population (1.4%).

Several months later, the patient developed constitutional symptoms. The WBC was found to be elevated to 72 × 10⁹/L and peripheral blasts were 21%, consistent with transformation to MPN-blast phase (MPN-BP), a form of secondary acute myeloid leukemia (AML). Decitabine was initiated and after receiving 4 cycles the patient underwent hematopoietic stem cell transplantation (HSCT) in August 2010 with cells from a 10/10 matched unrelated donor, with successful engraftment. The post-transplant bone marrow biopsy, however, showed persistent hypercellularity (80–90%) and persistent myeloblast populations ranging from 1.1 to 8%, consistent with residual disease. Unfortunately, the patient’s course was complicated by graft-versus-host disease of the gastrointestinal system and central nervous system. Ultimately, the patient succumbed to sepsis on day +106 post-transplant.

Conventional cytogenetic preparations and FISH (fluorescent in situ hybridization) analyses were performed as described previously [5, 6]. To detect the exact region involved in the novel translocation, 12 FISH probes were used: 9p24.1 (RP11-3H3), 9p24.1 (RP11-28A9; BACPAC Resources, Oakland, CA, USA), 1p36.22 (RP11-483P2), 1p36.22 (RP11-1107P2), RP11-1134M20, RP11-483P2, RP11-1107P2 (Empire Genomics, Buffalo, NY, USA), 9q34 (ABL1), 22q11.2 (BCR), 1p36 (p58), 1p36 (CEP108/T7), 1q25, WCP19, telomere 1p, telomere 19p (Abbott Molecular, Abbott Park, IL, USA). FISH chimerism was detected using XX/XY probes (Abbott Molecular). FISH probes, including bacterial artificial chromosomes (BACs), were fluorescently labeled by the manufacturer excluding RP11-3H3 and RP11-28A9, which were fluorescently labeled using the Nick translation kit (Abbott Molecular) following the manufacturer’s procedure [5]. DNA isolated from bone marrow cells was subjected to whole-genome sequencing and analyzed by standard pipelines at the Sanger Institute, Hinxton, UK [7].

Confirmation of a PEX14-JAK2genomic DNA fusion was performed using a forward primer in PEX14 exon 8 (5′CCACCAACTGGATCCTGGAGT) and reverse primer in JAK2exon 19 (5′AACCCCAGGGCACCTATCCT), on 25 ng of DNA using the Expand Long-Template LT-PCR system 2 (Roche, Burgess Hill, Sussex, UK) at an annealing temperature of 64°C and an elongation time of 4 min. Sequencing across the breakpoint was performed using primer PEX14 exon 9 (5′GTTCCCTCCATCCCCATCAG), using an Applied Biosystems 3130 analyzer (Foster City, CA, USA).

The presence of PEX14-JAK2 fusion mRNA was confirmed on random hexamer reverse-transcribed cDNA, using forward primer PEX14 exon 8 and reverse primer in JAK2 exon 21 (5′ TTTTAGATTACGCCGACCAGCA) using the Expand High Fidelity PCR System, an annealing temperature of 64°C, and an elongation time of 1 min. The product was sequenced in both directions using the same primers.

A summary of the conventional cytogenetic analysis is shown in Table 1. Initial bone marrow cytogenetic analysis, a year after the diagnosis of PV, showed 50% of evaluated metaphase cells to have the t(1;9)(p36;p24.1) karyotype. Subsequent metaphase FISH analyses using BAC FISH probes (RP11-3H3 and RP11-28A9) revealed that the 3′ portion of JAK2 was translocated to 1p36, while the 5′ portion remained on 9p24.1, indicating a JAK2 structural rearrangement (Fig. 1). To investigate the exact breakpoint on chromosome 1, we used a FISH BAC probe, and as shown in Figure 1, the BAC FISH probe RP11-4832P2 normally localized on chromosome 1p36 was detected on 9p24.1 (aqua), whereas BAC RP11-1107P2 remained on 1p36. Therefore, the breakpoint on chromosome 1, involved in the JAK2translocation, was determined to be within band p36.22 on chromosome 1 [7].

Five months from the initial analysis, 100% of the cells had t(1;9) and 10% had developed a subclone consisting of balanced t(7;17)(q22.1;25.3) and trisomy 1q in the form of unbalanced der(15)t(1;15)(q12;q26). Following HSCT, host cells were never eradicated (Table 1). A year after the initial cytogenetic analysis, the original abnormal host clone and a subclone were present in 100% of the cells with additional chromosomal abnormalities consistent with complex subclonal evolution.

To characterize the t(1;9) in detail, we performed whole-genome sequencing analysis of patient bone marrow DNA to identify the translocation breakpoints. Focusing on the analysis of JAK2, we identified two split reads that mapped to PEX14 exon 9 and JAK2 intron 18. PEX14maps to 1p36.22, and was thus a strong candidate to be fused to JAK2. To confirm a PEX14-JAK2 fusion, we amplified patient and control DNA using primers located in PEX14exon 9 and JAK2 exon 19. A product was obtained from the t(1;9) case only (not shown), which, upon sequencing, confirmed a break within PEX14 exon 9 and JAK2intron 18 (Fig. 2). Amplification from cDNA also yielded a specific product from the t(1;9) case but not the controls (Fig. 3), sequencing of which showed a fusion between a truncated PEX14exon 9 and JAK2exon 19 (Fig. 2). Comparison of the cDNA and genomic sequences indicates that two nucleotides (TA) from JAK2 intron 18 were retained in the mature mRNA that result in maintenance of the correct reading frame (Fig. 2). The TA dinucleotide is immediately followed by GT, which must have acted as a splice donor site.

Subsequent to the 2005 discovery of JAK2V617F, knowledge of the genetic underpinnings of MPNs and the role mutations play in diagnosis, prognosis and therapy increased exponentially [8‒11]. In stark contrast to the wealth of knowledge about the JAK2V617F-positive MPNs, there is a relative dearth of information on how to approach the small subset of patients with MPNs who lack this point mutation but contain JAK2 translocations.

Translocations at 9p24.1 and resultant gene fusion products have been identified in a wide spectrum of hematologic malignancies of myeloid and lymphoid origin, including acute lymphoblastic leukemia (ALL), chronic myeloid leukemia (CML), chronic eosinophilic lymphoma (CEL), and the BCR-ABL1-negative MPNs (Table 2). Additionally, several translocations involving 9p24.1 with subsequent gene fusion products involving JAK2 have been seen in solid malignancies, such as breast cancer [12] and small cell lung cancer [13].

JAK2translocations are likely more common in hematologic malignancies than previously recognized. Patnaik et al. [3] screened over 24,000 patient cytogenetic reports and found 5 patients harboring translocations at 9p24 with gene fusion products involving JAK2. All 5 of the translocations described in this subset of patients had not previously been reported in the literature, but the JAK2 fusion partners could not be identified. Four of the 5 patients carried diagnoses of JAK2V617F-positive MPNs (PMF and PV).

Though the above study found JAK2 translocations only in patients with JAK2V617F-positive MPNs, structural rearrangements, including JAK2translocations, have been found frequently in chromosomal analysis of samples from patients with JAK2V617F-negative MPNs [14]. The discovery of several of these translocations has shed light on theories of the pathogenesis and unique patterns of the disease. A prime example of this is the identification of t(8;9)(p22;p24), resulting in a PCM1-JAK2fusion gene product in a variety of hematologic diseases, including many in the spectrum of MPNs, such as atypical CML (aCML), CEL, myelodysplastic syndrome/MPN-unclassified (MDS/MPN-U), and myelofibrosis. Identifying the common translocation in cases of these disparate disorders has led to the recognition that patients with t(8;9)(p22;p24) are more likely to be male, have prominent erythroid dysplasia, and have high rates of peripheral blood and bone marrow eosinophilia [15]. PCM1-JAK2 was identified in both myeloid and lymphoid neoplasms, leading to the conclusion that malignancies with acquired t(8;9)(p22;p24) result from disorders of the pluripotent hematopoietic stem cell [16]. Increased recognition of the role of PCM1-JAK2 in hematologic malignancies has led to a relatively new categorization of “myeloid/lymphoid neoplasms with eosinophilia and rearrangement of PDGFRA, PDGFRB, or FGFR1, or with PCM1-JAK2” in the 2016 revision to the WHO classification of myeloid neoplasms and acute leukemia [17].

Similarly, both myeloid and lymphoid neoplasms with ETV6-JAK2 fusions have been identified. Animal modeling using this translocation has led to a proposed pathogenic mechanism involving the rearranged genes, which appears to cause constitutive activation of several STATs [18]. A number of cases have also been noted in which JAK2 fuses with the BCR gene on 9p24, resulting in aCML and unclassified MPNs [19, 20].

Here, we presented a case of a noveltranslocation t(1;9)(p36;p24.1) involving JAK2and peroxisomal biogenesis factor 14 (PEX14) found in a patient with JAK2V617F-negative PV, which transformed to an aggressive form of MPN-BP. PEX14 is a membrane protein involved in protein docking on peroxisomes, with a role in peroxisome formation and degradation. It also has a unique function as a transcriptional corepressor and a polypeptide transport modulator. Upregulation of PEX14 has been demonstrated in tissue from lung, rectal, ovarian, and esophageal carcinomas; however, its precise role in these malignancies remains unclear and information about its role in hematologic malignancies is absent [21]. Interestingly, like many other cases of JAK2V617F-negative, JAK2 translocation-positive MPNs, this patient’s disease was notable for prominent bone marrow eosinophilia of undetermined significance. JAK2 was identified as a fusion gene partner with a gene on chromosome 1 in only 1 other case of a hematologic malignancy – a TPM3-JAK2 fusion was found in a case of T-cell ALL [22].

Although commonalities between hematologic malignancies with JAK2 rearrangements have been identified, the prognostic significance and therapeutic implications of the translocations are not fully elucidated. As the disease of our patient progressed to MPN-BP, cytogenetic analysis showed a gain of 1q. We recently reported this to be associated with progression of MPNs to myelofibrosis and AML [23]. In vitro studies utilizing cell lines containing JAK2 rearrangements suggest a promising role for JAK inhibitors in halting malignant cell proliferation [22, 24]. Early case reports of ruxolitinib treatment in patients with MPNs containing JAK2 trans­locations also show positive outcomes in cytogenetic response and hematologic remission, although the durability of remission is variable and differences in outcomes are not well studied [25‒27]. As data accumulates about newly identified JAK2 translocations, further connections can be made between the specific translocations, disease course, and prognosis. Further information is needed to understand the translocations’ oncogenicity and the potential for novel targeted therapies aimed at targeting the specificJAK2 partners for this minority of MPN patients. In the future, MPNs may be stratified further into distinct entities that take into account specific JAK2 translocations.

The authors have no ethical disclosures to declare.

The authors have no conflicts of interest to declare.

H.L. and B.M. wrote the report. J.T. and D.G. performed cytogenetic and FISH studies. A.V.J. and N.C.P.C. performed molecular studies and analyses. J.M. was involved in clinical care and studies, and V.N. conceived and organized the work, and helped in preparing the report.