This gene is a known cancer gene. ARID1A/BAF250A subunit of the SWI/SNF (BAF) chromatin remodeling complex has emerged as recurrently mutated in a broad array of tumor types and a potential tumor suppressor. There is evidence indicating that ARID1A-mutated cancers may be subjected to therapeutic intervention.

Activating mutations in JAK1 (SH2-, pseudokinase -and kinase- domains) have been reported in approximately 5-20% of cases of T-Cell Acute Lymphoblastic Leukemia, in "Ph-like ALL" and in less than 5% of Acute Myeloid Leukemia. Some, but not all, of these mutations have been shown to be inhibitable by ATP-competitive JAK inhibitors or Type I interferon.

NRAS encodes a membrane protein GTPase that is a central mediator of downstream growth factor receptor signaling, critical for cell proliferation and survival. Mutations in codons 12, 13, and 61 of NRAS have been reported in 7-22% of acute myeloid leukemia, 12% of chronic myelomonocytic leukemia, 20% of blastic plasmacytoid dendritic cell neoplasm, 16% of juvenile myelomonocytic leukemia, 4-10% of myelodysplastic syndromes, and 5% of primary myelofibrosis. In addition, NRAS mutations have been described in approximately 15% of cases of B-ALL and, interestingly, some cases of ALL may show more than one abnormality in the RAS pathway. NRAS mutations are associated with an unfavorable prognosis in chronic myelomonocytic leukemia and primary myelofibrosis. NRAS mutations are also associated with an unfavorable prognosis in myelodysplastic syndrome, particularly in patients predicted to have lower-risk myelodysplastic syndrome (NCCN Guidelines for Myelodysplastic Syndromes). NRAS mutations do not seem to have significant prognostic impact in AML.

DNMT3A is a DNA methyltransferase. Recurrent, somatic, heterozygous mutations in DNMT3A have been reported in approximately 18-25% of cases of acute myeloid leukemia (up to 34% of normal karyotype AML), 12-18% of cases of myelodysplastic syndrome, up to 15% of myeloproliferative neoplasms, less than 5% of cases of chronic myelomonocytic leukemia and 15% of cases of adult, eary T cell precursor acute lymphoblastic leukemia. DNMT3A is also one of the most frequently mutated genes in CHIP and CCUS. Mutations in DNMT3A may occur together with mutations in other genes including JAK2, FLT3, IDH1/IDH2, ASXL1, TET2 and NPM1. DNMT3A mutations have been associated with reduced enzymatic activity or altered histone binding, as well as reduced DNA methylation in various genomic regions and altered gene expression in some models. Codon R882 is a hotspot for mutations in DNMT3A. DNMT3A mutations may be associated with adverse prognosis in specific subtypes of AML according to some, but not all studies; the prognostic significance of DNMT3A in AML may depend on patient age, type of DNMT3A mutation (R882 or non-R882 mutation) and the co-existence (or absence) of specific mutations in other genes. DNMT3A mutations may also be associated with adverse prognosis in MDS according to some studies.

GATA2 is a member of the GATA transcription factors which play a role in hematopoiesis. GATA2 mutations in the zinc finger domains have been described in accelerated phase and blasts phase chronic myelogenous leukemia as well as 5-10% of acute myeloid leukemia and familial syndromes with a predisposition to acute myeloid leukemia and myelodysplastic syndromes. Co-existing mutations in ASXL1 have been reported in a subset of patients with mutations in GATA2 and are believed to represent an important step in myeloid transformation, particularly to chronic myelomonocytic leukemia in young female patients. Other reports suggest that in cases of AML, GATA2 mutations have a higher prevalence among cases with biallelic CEBPA mutations and were not observed in cases with monoallelic CEBPA mutations. In general, the GATA2 pathogenic mutations are loss-of-function mutations (nonsense, frameshift, splice site mutations or missense mutations(codons 349-398)) and are believed to result in impairment of granulocyte differentiation. In MDS, GATA2 mutations may be associated with a poor prognosis. If clinical findings and family history are concerning for an inherited disorder, then genetic counseling may be helpful, if clinically indicated.

Ten-Eleven Translocation-2 (TET2) encodes a dioxygenase that converts 5-methyl-cytosine (5-mC) to 5-hydroxymethyl-cytosine (5-hmC) and promotes DNA demethylation. TET2 is a tumor suppressor gene and loss-of-function via mutations, deletion and IDH1/2 (Isocitrate Dehydrogenase 1 and 2) gene mutations is a common event in myeloid and lymphoid malignancies. TET2 is also present in about 10% of otherwise healthy elderly individuals with clonal hematopoiesis of indeterminate potential (CHIP) and in some patients with unexplained cytopenia but who do not satisfy diagnostic criteria for MDS, so-called clonal cytopenia with undetermined significance (CCUS). Mutations in TET2 occur in 50-60% of chronic myelomonocytic leukemias. Comutation of TET2 and SRSF2 was highly predictive of a myeloid neoplasm characterized by myelodysplasia and monocytosis, including but not limited to, chronic myelomonocytic leukemia. TET2 mutations are also found in 20-40% of systemic mastocytosis, 36% of blastic plasamcytoid dendritic cell neoplasm, 12-32% of acute myeloid leukemia, 10-20% of primary myelofibrosis, 10-33% of myelodysplastic syndromes, 10% of myelodysplastic/myeloproliferative neoplasms with ring sideroblasts and thrombocytosis (MDS/MPN-RS-T), 22% of polycythemia vera, and 16% of essential thrombocythemia. TET2 mutations are absent in juvenile myelomonocytic leukemia and show a low prevalence (less than 5%) in pediatric AML. Among lymphoid neoplasms, TET2 mutations are reported in approximately 30% of angioimmunoblastic lymphomas and less than 15 % of other mature T cell lymphomas and mature B cell lymphomas. In general, the mutations in TET2 are typically loss of function variants (frameshift, missense, nonsense mutations) that may be monoallelic or biallelic and occur throughout the gene. TET2 mutations tend to be mutually exclusive of mutations in IDH1/IDH2. TET2 mutations are associated with unfavorable outcomes and shorter survival after hematopoietic stem cell transplantation in patients with myelodysplastic syndrome (PMID: 25092778). In acute myeloid leukemia with wild-type FLT3-ITD and normal karyotype or intermediate-risk cytogenetic abnormalities, TET2 mutations are associated with an unfavorable prognosis.

IKZF1(Ikaros) is a transcriptional regulator of B cell development and is believed to have tumor suppressor-like properties. Deletions (whole gene and/or partial gene deletions) of IKZF1 have been reported in approximately 15-28% of BCR-ABL1-Negative_B-cell ALL, 70-90% of BCR-ABL1-Positive B-cell ALL. IKZF1 mutations are also found in approximately 40% of "Ph-like" ALL. Loss of functions mutations (missense, nonsense, frameshift mutations) have also been reported in IKZF1 in ALL and appear to be much less common (less than 5% of cases) than deletions. Deletions and mutations in IKZF1 have been associated with adverse prognosis and greater risk of relapse.

EZH2 encodes the histone methyltransferase subunit of the polycomb repressive complex 2 (PRC2) that leads to H3K27me3 and promotes transcriptional repression. EZH2 loss of function mutations (nonsense, frameshift mutations, occasionally occurring as homozygous mutations) may occur throughout the gene and have been reported in less than 10% of patients with acute myeloid leukemia, myelodysplasia, atypical chronic myelogenous leukemia, primary myelofibrosis and up to 12% of patients with chronic myelomonocytic leukemia. EZH2 loss of function mutations may be more frequent (15%) among cases of T cell acute lymphoblastic leukemia. EZH2 mutations have been independently associated with adverse prognosis in MDS and MDS/MPN. Therapeutic targeting of EZH2 is currently under study for some types of lymphoma and solid tumors.

RAD21 belongs to the cohesin complex family of genes that encode protein subunits of the cohesion complex, which regulates chromosomal segregation. is a member of the cohesin complex that regulates chromosome segregation during meiosis and mitosis. Loss of function mutations of RAD21 have been described throughout the gene in approximately 1% of cases of myelodysplasia, 1-5% of acute myeloid leukemia (AML), 1% of chronic myeloid leukemia and tend to be mutually exclusive of other mutations in the other components of the cohesin complex (ie, STAG1, SMC3, STAG2, SMC1A). In AML, mutations in the cohesin complex genes tend to be associated with mutations in NPM1. Cohesin complex mutations do not have clear prognostic impact in AML. Cohesin complex mutations are associated with an unfavorable prognosis in myelodysplastic syndrome, and are more frequently found in patients with high IPSS scores and secondary acute myeloid leukemia.

ABL1 kinase domain mutations in Philadelphia chromosome positive acute lymphoblastic leukemia and chronic myelogenous leukemia are associated with resistance to some types of tryosine kinase inhibitor therapy. The various mutations span several hundred amino acids (M237 thru E507) and vary in their response to later generation tyrosine kinase inhibitors.

SMC3 is a member of the cohesin complex and has been found to be mutated in approximately 1% of acute myeloid leukemia and 1% myelodysplastic syndromes. The mutations of SMC3 described tend to be missense mutations that occur throughout the gene. Mutations of the various members of the cohesin complex appear to occur in a mutually exclusive manner. Cases of AML with mutations of the cohesin complex may be associated with mutations of NPM1. Currently there does not appear to be any clear prognosistic impact of cohesin complex gene mutations in AML. Cohesin complex mutations are associated with an unfavorable prognosis in myelodysplastic syndrome, and are more frequently found in patients with high IPSS scores and secondary acute myeloid leukemia.

WT1 encodes for a transcription factor containing an N-terminal transactivation domain and a C-terminal zinc-finger domain necessary for the development of the urogenital system. The precise roles of WT1 in normal and malignant hematopoiesis remain uncertain. New emerging supports a novel role of WT1 in the regulation of epigenetic programs through its interaction with TET proteins in the 5=hydroxymethylation of cytosines. WT1 mutations are found in 6% of acute myeloid leukemia overall, and about 8-13% in cytogenetically normal AML. Higher frequencies are present in biallelic CEBPA mutated acute myeloid leukemia (14%), followed by t(15;17)/PML-RARA (11.0%), and FLT3-ITD (8.5%,). WT1 mutations are associated with younger age in AML. WT1 mutations are typically putative loss of function mutations and most frequently occur in exon 7 or exon 9. About 75% of these mutations are frameshift, and the remaining are missense, nonsense, splice site or inframe indel mutations. In some cases two or more mutations in WT1 may occur. In addition, WT1 mutations may coexist with mutations in NPM1, FLT3, among others. WT1 is overexpressed in the majority of AML, giving rise to the concept that it may act as both a tumor suppressor and oncogene, depending on the context. Several studies showed that WT1 mutations are associated with a worse prognosis in cytogenetically normal acute myeloid leukemia, although one study including patients from three German-Austrian AML study protocols demonstrated no association with overall survival or relapse-free survival. Given its over-expression in AML, clinical trials employing peptide vaccination strategy against WT1 has been ongoing in AML patients.

SF1 is a member of the spliceosome machinery and missense mutations have been described in approximately 1% of myeloid neoplasms including myelodysplasia, acute myeloid leukemia and myeloproliferative neoplasms.

CBL (casitas-B-lineage lymphoma) gene mutations have been identified in approximately 15% of chronic myelomonocytic leukemia, 15% of juvenile myelomonocytic leukemia, 15% of secondary AML(from MDS or MDS/MPN overlap syndrome) and rare or absent in polycythemia vera, essential thrombocythemia, primary myelofibrosis, chronic eosinophilic leukemia and MDS. Also, CBL mutations are found in only 1% of de novo acute leukemias and tend to be associated with core binding factor acute myeloid leukemia (AML) among AML cases. CBL is a Ras pathway gene and has been associated with hereditary myeloid disorders. CBL ubiquitinylates and degrades activated receptor and non-receptor tyrosine kinases via the E3-ligase activity of its RING domain. CBL also acts as an adaptor for downstream cell signal transduction, via its tyrosine kinase binding domain. Most variants of the CBL protein are missense substitutions in the zinc binding RING domain (amino acids 366-420) (exons 8-9) that abrogate CBL ubiquitin ligase activity but retain other functions. Pathogenic mutations are believed to be oncogenic by a variety of potential mechanisms including increased Ras pathway activation, aberrant phosphoSTAT5 and/or increased KIT expression in different cellular contexts. Occasionally, two CBL mutations may be present or CBL mutations may be associated with uniparental disomy. In addition, CBL mutations may occur together with mutations in other genes ( RUNX1, ASXL1, TET2 or EZH2 ). According to some studies, mutations of CBL may be associated with reduced overall survival in MDS.

ETV6 is a transcriptional repressor and is frequently involved in translocations with a variety of different partner genes in a range of hematologic malignancies. Mutations of ETV6 have been described in <5% of myelodysplastic syndromes and appear to be more frequent (ie, 10-24% of cases) in early T cell precursor type (immature) acute lymphoblastic leukemias. In addition, ETV6 mutations have been reported in association with hereditary myeloid disorders. These mutations occur throughout the gene and typically correspond to loss of function mutations (nonsense and frameshift mutations). ETV6 mutations may occasionally occur in a homozygous/hemizygous manner and tend to occur with mutations in NOTCH1 in lymphoblastic leukemia. In MDS, ETV6 mutations have been independently associated with an adverse prognosis. If clinical findings and family history are concerning for the presence of an inherited disorder, then genetic counseling may be helpful, if clinically indicated.

SH2B3 encodes the lymphocyte adaptor protein (LNK) which attenuates JAK2 activation and negatively regulates the signaling of the thrombopoietin receptor (MPL) and the erythropoietin receptor. LNK has also shown to bind and regulate mutant signaling molecules found in myeloproliferative neoplasms (MPNs) like MPL-W515L and JAK2-V617F. Several acquired SH2B3 frameshift and missense mutations in the pleckstrin homology domain and NH2-terminal region have been reported in myeloproliferative neoplasms. Somatic mutations of SH2B3 are infrequent in MPN and reported in ~5 to 7% of MPN patients. These mutations can be found in all MPN subtypes and co-exist with other driver genes. In one study, SH2B3 mutations are associated with JAK2, CALR and MPL mutations in 50%, 18% and 13.6% of positive cases. All identified mutations were found throughout the gene, without evidence of a hot spot. SH2B3 mutations may be enriched in blast phase disease with a frequency up to 13% of such cases. Also, approximately 1% of T cell and B cell acute lymphoblastic leukemias have been shown to carry homozygous frameshift or nonsense mutations in SH2B3. Partial deletions of SH2B3 have also been reported in ALL. In addition, the T allele of a nonsynonymous single-nucleotide polymorphism (SNP), rs3184504 (p;W262R, c.784T>C), in exon 2 of the SH2B3 gene has been reported to be more prevalent among JAK2V617F-positive patients than control patients or JAK2 wild type patients. Loss of function mutations in SH2B3 have been shown to lead to increased JAK2/STAT3 signaling and cell proliferation.

PTPN11 encodes SHP2, a member of the non-receptor protein tyrosine phosphatase (PTP) family that regulates growth factor and cytokine signaling and plays a key role in the proliferation and survival of hematopoietic cells. PTPN11 mutation is directly associated with the pathogenesis of Noonan syndrome and childhood leukemias. Despite its direct function in protein dephosphorylation, SHP2 plays an overall positive role in transducing signals. Germline and somatic mutations that result in increased activity of PTPN11 have been described in Noonan's syndrome (approximately 50%), juvenile myelomonocytic leukemia (35-42%), pediatric and adult myelodysplasic syndromes (4-10%), B cell acute lymphoblastic leukemia (5-10%), as well as pediatric and adult acute myeloid leukemia (5-10%). These gain of function mutations most often occur as heterozygous missense mutations located in exon 3 (SH2 domain) or exon 13 (phosphatase domain) . Within cases of juvenile myelomonocytic leukemia, mutations of PTPN11 tend to be exclusive of mutations in RAS, CBL and NF-1. PTPN11 mutations in adult AML are associated with a normal karyotype and concurrent NPM1 mutation, but no alteration of the FLT3. In one study, myelodysplastic syndromes with PTPN11 mutations were shown to have a worse overall survival. Small molecule inhibitors of PTPN11 are currently being developed.

SETBP1 encodes a protein which is believed to inhibit PP2A phosphatase activity through SET stabilization. In addition, SETBP1 binds to gDNA in AT-rich promoter regions, causing activation of a network of development genes through recruitment of a HCF1/KMT2A/PHF8 epigenetic complex. Heterozygous, somatic, missense mutations are predominantly hot-spot mutations within the SKI homologous region in exon 4, which result in the functional loss of the degron motif responsible for the short half-life of the protein. Therefore, these mutations result in an increased half-life and accumulation of the mutated SETBP1, and thus increased inhibition of the oncosupressor PP2A through the SETBP1-SET-PP2A axis. In addition, mutations in SETBP1 potentially deregulate gene transcription mediated by SETBP1. SETBP1 mutations have been described in approximately 25% of atypical chronic myelogenous leukemia, 30% of juvenile myelomonocytic leukemia, 17% of secondary acute myeloid leukemia, 13% of myeloproliferative/myelodysplastic syndrome with ring sideroblasts and thrombocytosis, and 5-15% of chronic myelomonocytic leukemia. SETPB1 mutations appear to be rare (< 5%) or absent among cases of primary acute myeloid leukemia, acute lymphoblastic leukemia, myelodysplastic syndromes, myeloproliferative neoplasms and chronic lymphocytic leukemia. SETBP1 mutations may be seen together with mutations in other genes such as ASXL1. Mutated SETBP1 provides supportive evidence for the diagnosis of atypical chronic myeloid leukemia, BCR-ABL1-negative in the 2016 revision of the WHO classification.SETBP1 mutations are associated with disease progression in myelodysplastic syndrome (NCCN Guidelines for Myelodysplastic Syndromes), and unfavorable prognosis in chronic myelomonocytic leukemia, juvenile myelomonocytic leukemia, atypical chronic myeloid leukemia, and MDS/MPN with ring sideroblasts associated and thrombocytosis.

Mutations of the transcription factor CEBPA (CCAAT/enhancer binding protein alpha) have been reported in approximately 15% of patients with acute myeloid leukemia (AML) with a normal karyotype. CEBPA plays a role in the differentiation of granulocytes. Two types of mutations have been reported: N-terminal changes which result in a truncated dominant negative isoform lacking one of the N terminal domain transactivation domains and C-terminal mutations which are in-frame insertions or deletions affecting the leucine zipper and preventing dimerization and DNA binding. Patients may carry both N- and C-terminal mutations affecting different alleles. Isolated, biallelic ("double") mutations (not single mutation) of CEBPA appear to be associated with a favorable-risk group of normal karyotype AML. CEBPA mutations have also been reported in association with hereditary myeloid disorders; 5-10% of CEBPA double mutant AML cases may harbor germline mutations. Recommend correlation with clinical findings including family history and genetic counseling, if there is clinical suspicion of an inherited disorder.

ASXL1 regulates epigenetic functions including histone and chromatin modifications. ASXL1 mutations have been reported in 40-50% of chronic myelomonocytic leukemia(CMML), 20% of myelodsyplastic syndromes, 20-35% of primary myelofibrosis, 15% of systemic mastocytosis, 30% of patients with secondary acute myeloid leukemia and 5-10% of primary acute myeloid leukemia. ASXL1 mutations have also been described in CHIP and CCUS. In CMML, missense mutations of ASXL1 appear to be less common (less than 10% of cases). Nonsense and frameshift mutations (but apparently not missense mutations) of ASXL1 have been reported to carry an adverse prognostic impact in cases of chronic myelomonocytic leukemia. In addition, ASXL1 mutations have been associated with adverse outcome in myelodysplasia, primary myelofibrosis and systemic mastocytosis. Among cases of AML, ASXL1 mutations appear to be associated with adverse prognosis in some subtypes of AML according to some, but not all, studies. ASXL1 mutations may coexist with mutations of splicing factor components, TET2 and RUNX1; for example, co-existence of U2AF1 and ASXL1 mutations have been described in CMML and primary myelofibrosis; While in AML, ASXL1 mutations have been reported to be exclusive of NPM1 mutations according to some studies.

RUNX1(AML1, CBFA2) encodes the alpha subunit of core binding factor and is a transcription factor important in normal hematopoietic development. RUNX1 mutations have been reported in approximately 10% of myelodysplastic cases, 5-15% of acute myeloid leukemia, 8-37% of chronic myelomonocytic leukemia, 10% of T cell acute lymphoblastic leukemia, 3% of systemic mastocytosis, 2% of essential thrombocythemia and 2% of polycythemia vera. The mutations include frameshift, missense, nonsense, and splice site mutations. Typically, the Runt domain and the region just downstream of the Runt domain are affected and the mutations tend to be monoallelic. AML with RUNX1 mutation which does not fulfill the diagnostic criteria for other specific AML subtypes in the categories of AML with recurrent genetic abnormalities, therapy-related myeloid neoplasms, or AML with myelodysplasia-related changes is now classified the provisional entity of AML with mutated RUNX1. RUNX1 mutations may be associated with Trisomy 8 or MLL-PTD in AML according to some studies. They tend not to occur in AML cases with favorable cytogenetic findings and appear to be exclusive of NPM1 or CEBPA mutations in AML. Myeloid neoplasms, predominantly MDS/AML, developing in patients, usually at a young age, with a familial platelet disorder and germline monoallelic RUNX1 mutations are categorized as myeloid neoplasms with germline RUNX1 mutation. Of note, RUNX1 may also be involved in large intragenic deletions and translocations (e.g., t(8;21)(RUNX1-ETO), t(3;21)(RUNX1-EVI1), t(12;21)(TEL-RUNX1) which are not detected by this assay. Mutated RUNX1 is a poor-risk prognostic marker in AML unless it co-occurs with favorable-risk AML subtypes (NCCN Guidelines for AML). RUNX1 nonsense or frameshift mutations are associated with an unfavorable prognosis in myelodysplastic syndrome, independent of IPSS, IPSS-R, age, and other gene mutations (NCCN Guidelines for Myelodysplastic Syndromes). RUNX1 mutations are independently associated with unfavorable outcomes and shorter survival after hematopoietic stem cell transplantation in patients with myelodysplastic syndrome and myelodysplastic syndrome/acute myeloid leukemia. RUNX1 mutations are also associated with an unfavorable prognosis chronic myelomonocytic leukemia and systemic mastocytosis.

EP300 is a histone acetyltransferase and transcriptional coactivator. Somatic mutations of EP300 have been described in cases of T cell acute lymphoblastic leukemia and other lymphoid malignancies including diffuse large B cell lymphoma and follicular lymphoma. Occasional splice site mutations have also been described which are not detected by this assay. EP300 mutations have also been described in the Rubinstein-Taybi developmental syndrome.

ZRSR2 endoes a component of the RNA splicing machinery which associates with the U2 auxillary factor and is involved in the recognition of the 3'-splice site during the stages of spliceosome assembly. Mutations in ZRSR2 have been reported in approximately 3-11% of myelodysplasia, 4-8% of chronic myelomonocytic leukemia, 8% of blastic plasmacytoid dendritic cell neoplasm and less than 5% of acute myeloid leukemia and myeloproliferative neoplasms. Unlike other spliceosomal genes, ZRSR2 mutations do not occur in select "hot spots". Nearly all reported mutations are nonsense or frameshift mutations, compatible with loss-of-function mutations and suggesting a tumor suppressor role of ZRSR2. ZRSR2 mutations tend to be exclusive of mutations in most other components of the RNA splicing machinery. Aberrant splicing of U12-type introns has been shown to be a hallmark feature of MDS with ZRSR2 mutations. ZRSR2 mutations are associated with an unfavorable prognosis in myelodysplastic syndrome (NCCN Guidelines for Myelodysplastic Syndromes). ZRSR2 mutations are also reported to be highly specific for secondary acute myeloid leukemia, and may also be helpful in identifying a subset of therapy-related AML and elderly de novo AML with worse clinical outcomes.

BCOR is a ubiquitously expressed nuclear protein that is a transcriptional corepressor important in several cellular processes. Somatic, nonsense and frameshift mutations throughout BCOR have been reported in approximately 7% of chronic myelomonocytic leukemia, 4% of patients with myelodysplastic syndrome(MDS), 4% of primary acute myeloid leukemia and appear to be associated with RUNX1 and DNMT3A mutations . Also, BCOR mutations may be enriched among cases of AML lacking NPM1, CEBPA, FLT3-ITD, IDH1 and MLL-PTD alterations. BCOR mutations tend to be subclonal in MDS, clonal in primary AML and are believed to have significance as loss of function mutations in a tumor suppressor gene that affect the functional allele in male and female patients. The presence of BCOR mutation in patients with MDS and AML has been associated with poorer overall survival according to some studies.

The GATA1 transcription factor is important in the development of erythroid and megakaryocytic lineages. Amino-terminal, small insertion/deletion(frameshift), nonsense and missense mutations of GATA1 have been described in almost all patients with transient abnormal myelopoiesis(TAM) and acute megakaryoblastic leukemia associated with Down syndrome (Trisomy 21)(DS-AMKL). Studies suggest that the cases of TAM which progress to DS-AMKL are associated with the acquisition of additional driver mutations in other genes including the cohesin complex genes as well as CTCF and EZH2. The amino-terminal GATA1 mutations lead to a lack of the N-terminal amino acids and translation from an alternate start codon (methionine at position 84 in exon 3). GATA1 mutations appear to be rare in acute megakaryoblastic leukemia not associated with Down syndrome. GATA1 mutations have also been reported in the context of hereditary myeloid disorders. If clinical findings and family history are concerning for an inherited disorder, then genetic counseling may be helpful, if clinically indicated.

SMC1A belongs to the cohesin complex family of genes that encode protein subunits of the cohesion complex, which regulates chromosomal segregation. SMC1A has been reported to show somatic, missense mutations throughout the gene in less than 5% of cases of acute myeloid leukemia and less than 5% of chronic myeloid leukemia. Mutations of SMC1A are mostly mutually exclusive of mutations in other components of the cohesin complex. Mutations of SMC1A may be enriched in male patients since the gene is located on the X chromosome. Cohesin complex mutations are associated with an unfavorable prognosis in myelodysplastic syndrome, and are more frequently found in patients with high IPSS scores and secondary acute myeloid leukemia.

STAG2 belongs to the cohesin complex family of genes that encode protein subunits of the cohesion complex, which regulates chromosomal segregation. STAG2 has been reported to show somatic, nonsense, frameshift and occasional missense mutations throughout the gene in 6% of cases of myelodysplasia, 10% of chronic myelomonocytic leukemia, 6% of cases of acute myeloid leukemia, and less than 5% of chronic myeloid leukemia and myeloproliferative neoplasms. Mutations of STAG2 are mostly mutually exclusive of mutations in other components of the cohesin complex. STAG2 mutation is associated with a poor prognosis in myelodysplastic syndrome (NCCN Guidelines for Myelodysplastic syndromes). STAG2 mutations are also reported to be highly specific for secondary acute myeloid leukemia, and may also be helpful in identifying a subset of elderly patients with de novo acute myeloid leukemia or therapy-related AML with worse clinical outcomes.

PHF6 encodes a member of the plant homeodomain (PHD)-like finger (PHF) family with two PHD-type zinc finger domains, indicating a potential role in transcriptional regulation. It is localized to the nucleolus and may play a role in recognizing methylation status of histone lysines. Somatic, nonsense, frameshift and occasional missense mutations throughout PHF6 have been reported in up to 38% of cases of T cell acute lymphoblastic leukemia. In T-ALL, PHF6 mutations often co-exist with NOTCH1 mutations. PHF6 mutations have also been reported in approximately 3% of cases of acute myeloid leukemia, less than 5% of chronic myeloid leukemia in blast phase, and 3% of myelodysplastic syndrome. In acute myeloid leukemia, PHF6 mutations have been associated with male preponderance and reduced overall survival in patients with normal karyotype or intermediate-rish cytogenteics abnormalities. Mutated PHF6 is more frequent in MDS cases with excess blasts, but there appears to be no association with survival (NCCN Guidelines for Myelodysplastic Syndromes). PHF6 mutation status does not appear to affect outcome in T-ALL according to some studies.

MPL encodes the thrombopoietin receptor, an important growth and survival factor for megakaryocytes. Somatic activating mutation in MPL (W515L, W515K) has been reported in approximately 1%-10% of cases of JAK2 V617F-negative myelofibrosis, essential thrombocythemia, a subset of cases of acute megakaryoblastic leukemia and has been associated with sensitivity to JAK inhibitors. The W515 mutations are typically not observed in polycythemia vera or other myeloid disorders (chronic myelomonocytic leukemia, myelodysplastic syndrome). A S505N activating mutation has also been described in familial essential thrombocythemia. MPL mutation is included as one of the major diagnostic criteria for primary myelofibrosis and essential thrombocythemia in the 2016 revision of the WHO classification.

SF3B1 encodes a core component of the U2 small nuclear ribonucleoprotein, involved in the recognition of the branchpoint sequence during RNA splicing. SF3B1 is one of several genes involved in RNA splicing that has been identified as recurrently mutated in MDS and other malignancies. SF3B1 is the most commonly mutated gene found in MDS (20-33% of MDS overall). SF3B1 mutations are highly associated with subtypes of MDS characterized by ring sideroblasts (MDS with ring sideroblasts and MDS with multilineage dysplasia and ring sideroblasts), present in ~80% of these patients. In addition, many cases (60-80%) of myelodysplastic/myeloproliferative neoplasm with ring sideroblast and thrombocytosis (MDS/MPN-RS-T) harbor SF3B1 mutations. SF3B1 mutations are also found in 12% of blastic plasmacytoid dendritic cell neoplasm, 4-7% of primary myelofibrosis, 5% of CMML, less than 5% of de novo AML and less than 5% of essential thrombocythemia and polycythemia vera. SF3B1 mutations tend to occur in exons 13-16 and appear to be enriched at codons Lys700, Lys666, His662, Arg625 and Glu622. Missense mutations have been reported in approximately 5-10% of cases of chronic lymphocytic leukemia (CLL) and are reported to be associated with del11q , unmutated IGHV and may predict an adverse prognosis in CLL. Mutations in splicing factor components are usually mutually exclusive. Among cases of CLL, SF3B1 mutations tend to be exclusive of NOTCH1 mutations according to one study. The presence of SF3B1 mutation has been included in the diagnostic criteria for MDS/MPN-RS-T and MDS-RS diagnosis in the 2016 revision of the WHO classification. SF3B1 mutations are independently associated with a more favorable prognosis in MDS (NCCN Guidelines for Myelodysplastic Syndromes) and are highly predictive for the presence of ring sideroblasts. SF3B1 mutations are also reported to be highly specific for secondary acute myeloid leukemia, and may also be helpful in identifying a subset of therapy-related AML or elderly patients with de novo acute myeloid leukemia with worse clinical outcomes. SF3B1 mutations are associated with an unfavorable prognosis in essential thrombocythemia. SF3B1 has a critical role in MDS by affecting the expression and splicing of genes involved in specific cellular processes/pathways, many of which are relevant to the known MDS-RS pathophysiology, suggesting a causal link.

IDH1 is an enzyme localized to the cytoplasm and peroxisomes and involved in citrate metabolism. Mutations at Arg132 of IDH1 are typically heterozygous mutations and considered gain of function mutations that lead to increased levels of 2-hydroxyglutarate which are believed to alter epigenetic regulation (ie, DNA methylation) in AML. Mutations of IDH1 appear to be mutually exclusive of mutations in TET2, another gene involved in regulation of DNA methylation, and also exclusive of mutations in IDH2. Mutations of IDH1 have been shown to lead to increased DNA methylation in AML. Recurrent missense mutation of Arg 132 in IDH1 has been reported in approximately 5-15% of cases of acute myeloid leukemia and is often associated with a normal karyotype, wild type CEBPA, wild type FLT3 and the presence of NPM1 mutations. In addition, this mutation has been reported in approximately 10-20% of cases with leukemic transformation of myeloproliferative neoplasms and has been reported in less than 5% of chronic phase primary myelofibrosis, less than 5% of myelodysplastic syndrome and rare cases of polycythemia vera, essential thrombocytosis and chronic myelomonocytic leukemia. The prognostic impact of IDH1 mutations in AML appears uncertain, however, in the settings of primary myelofibrosis and polycythemia vera, the presence of IDH1 mutation is independently associated with inferior survival. Mutant IDH1 represents a therapeutic target in some clinical settings.

KIT(cKIT) mutations are present in approximately 8-25% of cases of acute myeloid leukemia and have a higher prevalence in the favorable cytogenetic risk group including core binding factor (CBF) AMLs (ie, (t(8;21)(q22;q22)(RUNX1-RUNX1T1), inv(16)(p13q22)(CBFB-MYH11)) or normal karyotype AML. Mutations of KIT in AML are most common in KIT exon 17 (within the activation loop of the tyrosine kinase domain) but may also occur in KIT exons 8 (extracellular portion of the receptor implicated in dimerization), 9-11 (juxtamembrane/transmembrane domains). The presence of KIT mutations has been reported to be associated with a poorer survival and/or higher risk of relapse than expected for patients with the t(8;21)(q22;q22)(RUNX1-RUNX1T1), and to a lesser extent, in inv(16) AML. KIT mutations are also important in systemic mastocytosis and various mast cell disorders; over 90% of cases of systemic mastocytosis carry mutations in exon 17 of KIT (most commonly D816V or rarely D816H, D816Y or other variants). In patients with mastocytosis, the KIT mutations may be detectable in non-mast cell hematopoietic cells. The KIT D816V mutation has been shown to be resistant to imatinib; other KIT mutations may show variable responses to imatinib. The KIT D816V mutant has been reported to be sensitive to other tyrosine kinase inhibitors. In the context of core binding factor AMLs, the KIT mutation status can help direct therapeutic management.

Mutations of NPM1 have been reported in approximately 25-35% of cases of acute myeloid leukemia (AML). The mutations of NPM1 are frameshift mutations in the C-terminus of the protein that alter the C-terminal amino acid sequence and are associated with aberrant cytoplasmic localization of the protein. NPM1 mutations in AML are typically associated with a normal karyotype and may co-exist with FLT3 mutations. The presence of NPM1 mutations has been associated with improved complete remission rates, but not necessarily overall survival, in multivariate analysis including assessment of the variety of more recently discovered mutations that may be present in AML. In addition, cytogenetically normal AML with mutated NPM1, without FLT3 ITD or mutated DNMT3A, has been considered to be a favorable genetic risk group according to some studies, although other studies suggest that coexistant mutations in IDH1 or IDH2 may be required for the favorable risk effect of NPM1.

JAK2 is a non-receptor tyrosine kinase that mediates signaling via the JAK-STAT pathway. The somatic, activating mutation V617F in the pseudokinase domain of JAK2 has been reported in over 90% of patients with polycythemia vera, 40-70% of essential thrombocythemia, 40-60% of primary myelofibrosis and 50% of MDS/MPN with Ring Sideroblasts and Thrombocytosis. JAK2 mutations have also been reported in CHIP. The small percentage of cases of polycythemia vera that do not carry the JAK2 V617F mutation have somatic, activating mutations in JAK2 exon 12 which typically affect the region encompassing codons 536-547 and are in-frame deletions/insertions, duplications of 8-12 amino acids or missense mutations. Mutations in JAK2 are typically mutually exclusive with mutations in the thrombopoietin receptor (MPL) and calreticulin (CALR), but JAK2 mutations may co-exist with mutations in other genes (ie, IDH1, SF3B1, TET2, ASXL1, etc). Ruxolitinib is a JAK2 inhibitor that has been approved for use in patients with intermediate- and high-risk primary myelofibrosis and is under study in other JAK2+ MPNs. Other JAK2 inhibitors (eg, Pacritinib) are also in various stages of study.

NOTCH1 encodes a member of the NOTCH family of proteins, which is a group of transmembrane receptors involved in the Notch signaling pathway. Notch signaling regulates cell fate decisions during development, and plays a crucial role in T cell development. Activating mutations of NOTCH1 including missense mutations and inframe inserstions/deletions in the heterodimerization(HD) domains either alone or together with missense, nonsense or frameshift (in/del) mutations in the C terminal PEST domain have been described in approximately 50% of cases of T-ALL. The HD domain or PEST domain mutations may occur together in cis (on the same allele) in ALL. NOTCH1 mutations are very rare in AML. However, NOTCH1 mutations are present in about 27% patients with T-myeloid mixed-phenotypeacute leukemia. The potential utility of therapeutic targeting of activating NOTCH1 mutations in these diseases remains to be elucidated.

KRAS is a well known proto-oncogene that belongs to the small GTPase family and functions as a central mediator of downstream growth factor receptor signaling, with a critical role for cell proliferation and survival. Pathogenic mutations in KRAS typically occur in codons 12-13 of exon 2 and codon 61 of exon 3; however, other, non-canonical, pathogenic mutations in KRAS have also been reported in acute myeoid leukemia. KRAS mutations have been described in approximately 3-15% of acute myeloid leukemia, 8-20% of chronic myelomonocytic leukemia, 14% of juvenile myelomonocytic leukemia, 8% of blastic plasmacytoid dendritic cell neoplasm 4% of patients with myelodysplastic syndrome, 2% of primary myelofibrosis, 12% of B cell acute lymphoblastic leukemia (often associated with MLL rearrangement) and 1-2% of T cell acute lymphoblastic leukemia. Investigation into the targetability of this pathway in leukemia has been attempted in some disease models.

FLT3 is a receptor tyrosine kinase important in hematopoietic cell proliferation and differentiation. In-frame, FLT3 internal tandem duplications (ITD), which show a wide range of number of nucleotides duplicated and/or inserted (eg, 18-204 bp), affect exons 14 and 15 in the FLT3 juxtamembrane/tyrosine kinase domain and have been reported in 20-30% of patients with acute myeloid leukemia. FLT3 ITDs usually occur in cases of AML with a normal karyotype and but may also occur in cases with chromosome abnormalities including t(15;17) or other cytogenetic groups. Individuals with FLT3 mutations are more likely to have AML than MDS and FLT3 mutations in MDS are associated with a poor prognosis. Functional studies have shown that FLT3 ITDs are ligand-independent, gain-of-function mutations. In addition, activating mutations at codon D835 in exon 20 (A-loop of the tyrosine kinase domain, TKD) of FLT3 have been reported in approximately 7-10% of AML. The FLT3 D835 mutations are also ligand-independent, gain-of-function mutations. FLT3 ITD and D835 mutations tend to occur in a mutually exclusive manner; however, the FLT3 D835 mutation or other mutations in the A-loop(eg, D839, Y842) may occur as an acquired resistance mutation in the setting of patients with FLT3 ITD mutations being treated with targeted therapy. In addition, variants at codon N676 and in exon 17 F691, G697 have been associated with resistance to FLT3 targeted therapy. The presence of FLT3 ITD mutation in young patients with AML and normal cytogenetics is thought to be associated with a poor prognosis. On the other hand, the prognostic significance of the FLT3 D835 and TKD mutations appears less clear. More recently, therapeutic targeting of FLT3 in combination with other chemotherapy is available in certain settings for acute myeloid leukemia with FLT3 ITD or TKD mutations. Lastly, in T cell acute lymphoblastic leukemia, up to 20% of cases have been reported to show a FLT3 mutation (TKD or ITD) and are often associated with an ETP(early T cell precursor) phenotype. In addition, FLT3 mutations have also been reported in up to 15% of B-ALL cases ("Ph-like phenotype") and may also be associated with hyperdiploidy and MLL rearrangement. Targetting FLT3 in acute lymphoblastic leukemia therapy may also be possible in some settings.

IDH2 is a mitochondrial enzyme involved in citrate metabolism. Mutations at Arg140 and Arg172 of IDH2 are typically heterozygous mutations and considered gain of function mutations that lead to increased levels of 2-hydroxyglutarate which is believed to alter epigenetic regulation (ie, DNA methylation) in AML. Mutations of IDH2 appear to be mutually exclusive of mutations in TET2, another gene involved in regulation of DNA methylation, and also exclusive of mutations in IDH1. Mutations of IDH2 have been shown to lead to increased DNA methylation in AML. IDH2 mutations have been reported in 10-20% of AML and are often associated with a normal karyotype. IDH2 mutations have been reported in less than 5% of cases of MDS and less than 10% of myeloproliferative neoplasms. The prognostic impact of IDH2 mutations in AML appears uncertain due to conflicting reports. In the setting of essential thrombocytosis, primary myelofibrosis and MDS, the presence of IDH2 mutations is associated with reduced survival. Therapeutic targeting with an FDA approved mutant IDH2 inhibitor (enasidenib (AG-221)) has been reported for patients with relapsed or refractory IDH2-mutated AML.

TP53 is a well known tumor suppressor gene that is mutated in wide variety of cancers. Loss of function mutations (missense, nonsense and frameshift mutations) of TP53 have been described in 10-20% of CLL cases and TP53 gene defects tend to be enriched among cases with unmutated IGH variable regions; in some series, TP53 mutations have been reported in approximately 15%-18% of IGHV unmutated CLL cases . TP53 mutations appears to be less common in other types of CLL (eg, less than 5% of IGHV3-21-expressing CLL carried a TP53 defect according to one study). Mutations of TP53 in CLL have been found together with del17p and mutations in other genes such as NOTCH1 and SF3B1. Mutations and deletions of TP53 appear to represent adverse prognostic markers in chronic lymphocytic leukemia.

SRSF2 is a member of the serine/arginine-rich family of pre-mRNA splicing factors, which constitute part of the spliceosome. It interacts with other spliceosomal components bound to both the 5- and 3-splice sites during spliceosome assembly. SRSF2 mutations typically occur as missense mutations at Pro95. SRSF2 mutations have been reported in approximately 40% of cases of chronic myelomonocytic leukemia, but they may not have prognostic significance in that entity. Comutation of TET2 and SRSF2 was highly predictive of a myeloid neoplasm characterized by myelodysplasia and monocytosis, including but not limited to, chronic myelomonocytic leukemia. In addition, SRSF2 mutations have been reported in approximately 15-20% of cases of myelodysplastic syndrome. SRSF2 mutations have also been described in 5-20% of patients with acute myeloid leukemia and appear to be enriched among AML patients with reduced blast counts. SRSF2 has been found to be mutated in approximately 10% of cases of primary myelofibrosis where mutations may occur together with mutations in JAK2, MPL, TET2, CBL, ASXL1, EZH2, IDH1/2. SRSF2 mutations are also present in 8% of blastic plasmacytoid dendritic cell neoplasm and 3% of polythemia vera. SRSF2 mutations tend to be (although are not entirely) exclusive of mutations in other splicing factor components. SRSF2 mutations are associated with a poor prognosis in myelodysplastic syndrome (NCCN Guidelines for Myelodysplastic Syndromes), primary myelofibrosis, polycythemia vera, and KIT D816V-mutated advanced systemic mastocytosis. SRSF2 mutations are also reported to be highly specific for secondary acute myeloid leukemia, and may also be helpful in identifying a subset of elderly patients with de novo acute myeloid leukemia and therapy-related AML with worse clinical outcomes.

U2AF1 encodes for the small subunit of the U2 auxiliary factor, which is a non-small nuclear ribonucleoprotein (non-snRNP) required for the binding of U2 snRNP to the pre-mRNA branch site and plays critical role in RNA splicing. U2AF1 is one of several spliceosome complex genes frequently mutated in a variety of hematologic malignancies. Two hotspot mutations (S34 in exon 2 and Q157 in exon 6) occur within the two zinc-finger domains of the U2AF1 protein. These mutations have been reported in approximately 4- 9% of chronic myelomonocytic leukemia, 8-11% of cases of myelodysplastic syndrome (typically without ring sideroblasts), 16% of primary myelofibrosis, 12% of blastic plasmacytoid dendritic cell neoplasm, 4% of acute myeloid leukemia and 1% of essential thrombocythemia. U2AF1 mutations are associated with an unfavorable prognosis in myelodysplastic syndrome (NCCN Guidelines for Myelodysplastic Syndromes) and essential thrombocythemia, and decreased response to lenalidomide in myeloid neoplasms with and without del(5q). U2AF1 mutations are also reported to be highly specific for secondary acute myeloid leukemia, and may also be helpful in identifying a subset of therapy-related AML or elderly de novo AML with worse clinical outcomes. U2AF1 mutations have been associated with altered splicing patterns in vitro and in vivo, and may play a significant role in the pathogenesis of myeloid malignancies due to selective mis-splicing of tumor-associated genes.

Calreticulin(CALR) is an endoplasmic reticulum chaperone protein. Somatic insertions and deletions in exon 9 of calreticulin that cause a +1bp frameshift and a novel carboxy-terminal peptide in mutant calreticulin have been reported in 70% of JAK2/MPL-negative essential thrombocythemia (ET)and 56-88% of JAK2/MPL-negative primary myelofibrosis(PMF). In addition, CALR mutations have been reported in approximately 10% of patients with myelodysplasia, including JAK2/MPL-negative refractory anemia with ring sideroblasts (RARS-T) where it may co-occur with mutations in SF3B1. In ET, PMF, and RARS-T, calreticulin mutations appear to be mutually exclusive of mutations in JAK2 or MPL. The CALR mutations lead to loss of the endoplasmic reticulum retention motif (KDEL) sequence in the carboxy-terminal portion of mutant CALR. Calreticulin mutations appear to be absent in polycythemia vera, acute myeloid leukemia, chronic myeloid leukemia, systemic mastocytosis, lymphoid malignancies and are rare in atypical chronic myeloid leukemia and chronic myelomonocytic leukemia. The most common 52bp deletion mutation (Type 1) in CALR has been shown to lead to cytokine-independent growth and activation of STAT5. Type 2 (5 bp insertion) mutations have also been described. This represents a potentially targettable pathway alteration. Patients with some types of mutant CALR may show improved survival and lower risk of thrombosis compared to patients with mutant JAK2, according to some, but not all studies.

B-RAF is a member of the RAF-family of kinases which plays an important role in the RAS-RAF-MEK-ERK mitotic signaling pathway. Mutations of B-RAF have been described in up to 40-70% of Langerhans cell histiocytosis and approximately 50% of Erdheim-Chester disease. The hotspot for mutations in BRAF is at codon Val600 and these are activating mutations. The most common activating mutation is p.Val600Glu(V600E). Various B-Raf inhibitors(Vemurafenib, Dabrafenib) have been FDA approved for therapy for some tumor types in certain settings, and clinical trials for advanced BRAF V600 mutation-positive tumors using targeted therapy (often in combination with other therapy) may be available (clinical trials.gov).

TP53 encodes p53, a tumor suppressor protein that consists of transactivation domain, proline-rich domain, DNA-binding domain, oligomerization domain, and regulatory domain. p53 responds to diverse cellular stresses to maintain genomic stability and to induce cell cycle arrest, apoptosis, DNA repair and metabolic changes. TP53 mutations represent an important mechanism of resistance to DNA-damaging chemotherapeutic agents. Somatic TP53 mutations are found in a variety of cancers with various frequencies depending on cancer type; overall, TP53 is mutated in over one-half of human cancers. Missense mutations were the most frequent (~70-80%), followed by frameshift and nonsense mutations. Most TP53 mutations are clustered in the DNA-binding domain encompassing exons 5 and 8. These mutations either directly disrupt the DNA-binding domain of TP53 or cause conformational changes of the TP53 protein, thus leading to severely impaired TP53 function. Overall in myeloid malignancies, TP53 mutations are found in 5% to 15% of de novo MDS and AML but 20% of myelodysplastic syndrome with isolated del(5q) and ~50% of MDS/AML with complex karyotype. TP53 mutations are also more frequent in therapy-associated myeloid neoplasm (21-38%) compared to de novo MDS and AML. TP53 mutations are also found in 8% of blastic plasmacytoid dendritic cell neoplasm, and less than 5% in myeloproliferative neoplasms (ET, PV and PMF) and chronic myelomonocytic leukemia. TP53 mutations are independently associated with a poor prognosis in myelodysplastic syndrome (NCCN Guidelines for Myelodysplastic Syndromes) and is a poor risk factor in AML (NCCN Guildelines for AML). TP53 mutations are also associated with resistance to lenalidomide or relapse during lenalidomide treatment. TP53 mutations are independently associated with unfavorable outcomes and shorter survival after hematopoietic stem cell transplantation in patients with myelodysplastic syndrome and myelodysplastic syndrome/acute myeloid leukemia, but an increased response to decitabine in patients with myelodysplastic syndrome or acute myeloid leukemia.

SUZ12 is one of the core components of the polycomb repressive complex 2 (PRC2), which is a highly conserved histone H3 lysine 27 methyltransferase that regulates the expression of developmental genes. SUZ12 mutations are present infrequently (<2%) in myeloproliferative neoplasms (MPN) and myelodysplastic syndrome/myeloproliferative neoplasm (MDS/MPN). The mutations are missense and tend to be located at the highly conserved VEFS domain, is required for the interaction between SUZ12 and EZH2. These mutations reduced PRC2 histone methyltransferase activity in vitro. Inactivating mutations of the catalytic component of PRC2, EZH2, can also be seen in myeloid neoplasms. Mice with loss of function mutations in PRC2 components display enhanced activity of their hematopoietic stem cell/progenitor population and loss of SUZ12 function in particular enhances hematopoietic stem cell activity.

NFE2 codes for a subunit of the NF-E2 (nuclear factor, erythroid 2) complex essential for regulating erythroid and megakaryocytic maturation and differentiation. This submit regulates a number of erythroid and megakaryocytic promoters. Rare insertion and deletion mutations leading to premature translation termination in NFE2 were found in 2% of myeloproliferative neoplasms (MPN). NFE2 mutatoins appear to be enriched in isolated myeloid sarcomas (MS). Of the six investigated cases of MS without previous or concurrent AML in the bone marrow, 4 (67%) harbored mutations in NFE2, 3 of which were missense and 1 of which was frameshift. In addition, NFE2 is overexpressed in the majority of patients with MPNs. In murine models, over-expression of NFE2 caused an MPN phenotype with spontaneous leukemic transformation, supporting a direct role of NFE2 in the pathogenesis of MPN. Over-expression of NFE2 in MPN may be attributed to histone H3Y41 phosphorylation by V617F-mutated JAK2 and transcriptional activation of NFE2 by JMJD1C in a positive feedback loop.