Whole Exome Sequencing Identifies Novel MPL and JAK2 M utations in Triple Negative Myeloproliferative Neoplasms

Essential thrombocythemia (ET) and primary myelofibrosis (PMF) are chronic myeloproliferative neoplasms (MPN) characterized by clonal hematopoiesis and hyperproliferation of terminally differentiated myeloid cells. Most of the cases are sporadic and driven by somatic mutations, although familial clustering is observed. The most common mutation affecting 50-60% of the cases is JAK2-V617F, while 25-30% of the patients carry somatic mutations in exon 9 of CALR. MPL exon 10 mutations affect ~5% of the cases. JAK2, CALR and MPL mutations are mutually exclusive and account for >90% of ET and PMF cases. In 12% of ET and 5% of PMF cases the disease drivers remain unknown. These patients are termed as triple negative. The mutational analysis for diagnostic purposes is limited to exons 14 of JAK2, exon 10 of MPL and exon 9 of CALR. The aim of this study was to identify disease causing mutation in triple negative cases of ET and PMF.

To identify the somatic mutations that are potential disease drivers in triple negative MPN we performed whole exome sequencing (WES) on paired samples from the tumor and control tissue of 4 patients with ET and 4 patients with PMF. We identified somatic mutations in 3/8 analyzed cases. In two PMF cases we identified somatic mutations in genes relevant for MPN- TET2, ASXL1, CBL, SRSF2 and a mutation in MPL-S204P. We did not identify a novel recurrent mutation. In the 5 cases without somatic mutations, we looked for germline mutations in genes relevant for MPN. We identified germline mutations MPL-V285E and JAK2-G571S in one PMF case and one case of ET, respectively. SNP microarray analysis for presence of chromosomal aberrations revealed a uniparental disomy of chromosome 6p in the case with MPL -V285E mutation, suggesting clonal hematopoiesis. To determine the frequency of MPL and JAK2 mutations outside exons 10 and 14 in triple negative MPN, we performed Sanger sequencing of all coding exons of MPL in 62 patients and of JAK2 in 49 patients. We detected variants outside exon 10 of MPL in 6/62 cases (9.7%). MPL-T119I, MPL-S204F, MPL-E230G and MPL-Y591D were somatic mutations, while MPL-R321W was germline. We identified an additional patient with MPL-S204P mutation, however the control tissue was not available. JAK2 variants were found in 4/49 cases (8.1%). JAK2-G335D and JAK2-V625F were germline mutations, while for the patients with JAK2-F556V and JAK2-G571S the control tissue was unavailable. In total, we identified non-canonical MPL mutations in 8/70 (11.4%) and JAK2 mutations in 5/57 (8.8%) triple negative cases of ET and PMF. All mutations were heterozygous. The mutations in MPL and JAK2 were mutually exclusive in our patient cohort.

The expression of identified MPL mutants did not induce cytokine independent growth of Ba/F3 cells, but the MPL-Y591D expressing cells showed marked hypersensitivity to TPO compared to the wild type. Using a luciferase reporter assay in JAK2-deficient gamma 2A cells, where we transiently expressed the wild type or mutant MPL cDNAs, JAK2, STAT5, STAT5 reporter Spi-Luc, and pRL-TK for transfection control, we could demonstrate that all identified MPL mutations lead to constitutive activation of JAK/STAT signaling. As the detection of activity required longer times (48h) than for the MPL-W515K (24h), we concluded that the identified mutations have a milder effect of the function of MPL. By Western immunodetection we could demonstrate that expression of JAK2-F556V and JAK2-V625F in Ba/F3-MPL cells, lead to the increased phosphorylation of STAT5 in the absence of cytokines. We also observed increased sensitivity to TPO in the Ba/F3 MPL cell lines expressing JAK2-F556V and JAK2-V625F. JAK2-V625F and JAK2-F556V are mild gain-of-function mutations, while JAK2-G335D and JAK2-G571S do not seem to alter the function on the JAK2 protein.

The results of our study suggest that sequencing of all coding exons of MPL and JAK2 is recommended for the diagnostic work-up of the ET and PMF patients who do not carry other more common mutations. The lack of evidence for clonal disease in 50% of the triple negative cases and presence of germline mutations suggests that a proportion of cases are likely to be hereditary MPN-like disorders. Application of whole genome sequencing or RNA sequencing for fusion oncogene detection will likely fill in the gap of the remaining triple negative MPN cases with clonal hematopoiesis in which we did not identify a recurrent driving mutation using WES.