CUX1 Mutations Research Articles

RUNX1 associated Familial Platelet Disorder with Myeloid Malignancy (FPD-MM) in Children: A Novel New Phenotype with Juvenile and Chronic Myelomonocytic Leukemia (JMML/CMML) Characteristics

Introduction:RUNX1 (aka AML1; 21q22.12) is indispensable in the establishment of definitive hematopoiesis in humans. Activating RUNX1 mutations are associated with both Acute Myeloid and Lymphoblastic Leukemias (AML, ALL). On the other hand, hypofunctioning RUNX1 mutations cause dominantly inherited Familial Platelet Disorder (FPD). RUNX1 FPD has a high risk for progression to pancytopenia, myeloproliferative disorders (MPD) or AML, hence the new WHO category FPD with myeloid malignancy (FPD-MM). Those with MM carry mutations in other genes seen in AML, MDS. It is a relatively rare disorder with ~75 affected kindreds reported worldwide (Sood, et al. Blood 2017). Detailed reviews of pediatric cases are few.Case Histories: We encountered two children with RUNX1 associated thrombocytopenia; the mutations are novel. The first family is that of 14 yr old AAF, presenting with fainting- blood counts are shown in Table 1; fetal hemoglobin (HbF) was elevated; bone marrow was hypercellular with 6% type 1 blasts, extreme paucity of megakaryocytes, erythroid hyperplasia and large numbers of sea blue histiocytes. The high HbF suggested JMML while the monocyte CD16;14 profile (95.6% CD14+ cells) was similar to that seen in the adult type Chronic Myelomonocytic Leukemia (CMML). Her mother has pancytopenia without excess blasts in the marrow. The second case presented with neonatal thrombocytopenia; father has history of excessive bruising.Results: Blood counts and values for HbF are listed in Table 1.Molecular testing:Case1: A Myeloid gene panel showed RUNX1 - NM_001754.4:c.501delT, p.Ser167Argfs*9; PHF6 - NM_032458.2:c.902dupA, p.Tyr301*; CUX1- NM_001202543.1:c.2378delC, p.Pro793Argfs*26. No mutations were noted in PTPN11, CBL or RAS genes, the latter confirmed by JMML panel done at University of California, San Francisco. UCSF panel identified a mutation in SH2B3, a gene linked to erythrocytosis not caused by JAK2 mutations. Her mother has the same RUNX1 mutation, thus identifying a germline mutation of RUNX1 in her and her child but not the PHF6, CUX1 or the SH2B3 mutations seen in her daughter. A half sibling is unaffected and is a potential transplant donor for the mother.Case2: No coding sequence mutations were detected in genes associated with familial thrombocytopenia including ETV6, GATA1 and RUNX1. Array Comparative Genomic Hybridization studies (Prevention Genetics) identified a heterozygous deletion of the entire exon 5 of RUNX1.To understand the complex findings in family 1 additional studies were done- DRAQ5, CD71, Fetal Hb staining showed that NRBC in Case 1 contained predominantly high HBF cells. LIN28B was markedly elevated in the proband but not the mother (HbF- normal); LIN28B expression was normal in Case 2.Treatment/Outcome: In Case 1, low dose decitabine therapy resulted in the control of MPD features with good Hb recovery and normalization of the monocyte CD16;14 profiles. There was no platelet response to decitabine nor to a course of valproic acid. The child died of fulminant acute graft vs host disease affecting the liver following a 4/6 cord mismatch transplantation. Mother continues to show moderately severe pancytopenia requiring frequent transfusion support. The second child is symptom free with mild thrombocytopenia.Discussion: The hybrid JMML/CMML features in the index child are likely caused by the concurrent CUX1/PHF6/SH2B3 mutations. We are unable to establish if these are true de novo mutations as the father was not available for study; she had no full siblings. Neither high HbF nor high LIN28B are known feature of FPD by itself nor CMML or Polycythemia Vera (p Vera). Recently, the high HbF in JMML has been linked to high expression of LIN28B. SH2B3 mutation may have contributed to the high erythroid proliferation observed in our case. Induced CUX1 haploinsufficiency in mice causes MPD akin to CMML and megakaryocytic (Meg) proliferation (An N, et al. Blood 2018). The virtual absence of Megs in our case indicates that the CUX1 mutation was unable to overcome the Meg ploidization defect caused by the RUNX1 mutation. PHF6 mutations occur in T-ALL and AML but have not been linked to high HbF.Conclusions: Normal HbF and normal LIN28B expression in the mother of Case1 and in Case2 indicate that increased LIN28B is linked to the high HbF in Case 1 and that high LIN28B itself is a consequence of the malignant transformation caused by the concurrent CUX1/PHF6/SH2B3 mutations. DisclosuresChitlur:Baxter, Bayer, Biogen Idec, and Pfizer: Honoraria; Novo Nordisk Inc: Consultancy. Ravindranath:AGIOS: Other: Site Investigator for Pyruvate Kinase Deficiency.

Pan-Myeloid Leukemia Analysis: Machine Learning-Based Approach to Predict Phenotype and Clinical Outcomes Using Mutation Data

Pan-Myeloid Leukemia Analysis: Machine Learning-Based Approach to Predict Phenotype and Clinical Outcomes Using Mutation Data

Mechanisms of Azacitidine Chemotherapy Resistance in AML and MDS and New Therapy Options

BACKGROUND:TET2 (Ten-eleven-translocation-2) mutations found in 10-30% of MDS and AML patients are actionable for Azacitidine (AZA) treatment (PMID: 21828143). Although response rate of AZA in older AML patients is ~47%, rate of relapse is high due to evolution of disease clones and development of new mutations during treatment cycle. Currently, there is no standard of care for AZA-resistant MDS and AML, and treating these individuals is challenging. Thus, there is an urgent need to identify mechanisms for resistance to AZA for TET2-mutant cancers.AIM: The objective of the study was to predict response to AZA in OCI-M2 disease model and its AZA-resistant (AZA-R) sub-clones using computational biology modeling (CBM) to determine mechanisms of resistance to AZA and identify new therapy options for AZA chemoresistance.METHODS: Parental OCI-M2 cells, representing an AML transformed from MDS, were treated with 8 mM AZA added to media every 2 days (IC50=2µM) for a period of 6 weeks to generate OCI-M2 AZA resistant subclones. AZA-R subclones together with parental AZA-sensitive cells were profiled using cytogenetics and NGS (TruSight Myeloid Sequencing, Illumina). Genomic data were analyzed via CBM system (Cellworks Group), which generates disease-specific protein network maps and enables digital drug simulations. Impact of digital drug treatments were quantified by measuring predicted effects on AML cell growth score, which is a composite of cell proliferation, viability and apoptosis.RESULTS: CBM correctly predicted the response of parental OCI-M2 AML cells to be sensitive to AZA and all its AZA-resistant subclones to be resistant to AZA treatment. Genomic analysis using CBM showed that OCI-M2 AML model harboured TET2mutation (Q1084P), KAT6A amplification and DNMT3B amplification, which are putative determinants of AZA sensitivity as the AZA-resistant subclones lacked cells harbouring these mutations (TET2, KAT6Aamp and DNMT3Bamp). Instead, AZA treatment selected clones with frameshift or nonsense mutations in the following genes: ASXL1, EZH2, CUX1, ATRX, BCOR, PTPN11, FLT3 along with preserved TET2 in all the clones. Mutations in ASXL1 and EZH2 were previously shown to reduce the importance of methylation in driving disease, thus potentially leading to subclone proliferation regardless of AZA. Also 4 of 6 AZA-resistant subclones had DNMT3A frameshift mutations which may enhance AZA resistance.CBM identified new drug combinations based on overlapping as well as unique mutations in AZA-resistant subclones. These combinations include Ruxolitinib + Venetoclax, Nelfinavir + Ruxolitinib or Venetoclax, Sorafenib or Dasatinib + Venetoclax. Some of these drugs were shown to effectively substitute AZA on the parental cell lines (Venetoclax IC=0.9µM, Ruxolitinib IC50=1.7µM, Sorafenib IC50=5.3µM). Whether AZA enhanced sensitivity to these drugs in AZA-R clones is currently under investigation.Conclusions: CBM-based genomic analysis identified secondary mutations in ASXL1, EZH2, CUX1, ATRX, BCOR, PTPN11 and FLT3 that associated with AZA chemoresistance. Digital drug screening identified new drug combinations including Venetoclax for validation testing in AZA-resistant MDS and AML. DisclosuresAbbasi:Cell Works Group Inc.: Employment. Singh:Cellworks Research India Private Limited: Employment. Kumar:Cellworks Research India Private Limited: Employment. Suseela:Cellworks Research India Private Limited: Employment. Patil:Cellworks Research India Private Limited: Employment. Dattatraya:Cellworks Research India Private Limited: Employment. Tiwari:Cellworks Research India Private Limited: Employment. Vali:Cell Works Group Inc.: Employment. Cogle:Celgene: Other: Steering Committee Member of Connect MDS/AML Registry.

Cux1 promotes cell proliferation and polycystic kidney disease progression in an ADPKD mouse model.

Autosomal dominant polycystic kidney disease (ADPKD) is one of the most common monogenic hereditary disorders in humans characterized by fluid-filled cysts, primarily in the kidneys. Cux1, a cell cycle regulatory gene highly expressed during kidney development, is elevated in the cyst-lining cells of Pkd1 mutant mice, and in human ADPKD cells. However, forced expression of Cux1 is insufficient to induce cystic disease in transgenic mice or to induce rapid cyst formation after cilia disruption in the kidneys of adult mice. Here we report a double mutant mouse model that has a conditional deletion of the Pkd1 gene in the renal collecting ducts together with a targeted mutation in the Cux1 gene (Pkd1CD;Cux1tm2Ejn). While kidneys isolated from newborn Pkd1CD mice exhibit cortical and medullary cysts, kidneys isolated from newborn Pkd1CD;Cux1tm2Ejn-/- mice did not show any cysts. Because Cux1tm2Ejn-/- are perinatal lethal, we evaluated Pkd1CD mice that were heterozygote for the Cux1 mutation. Similar to the newborn Pkd1CD;Cux1tm2Ejn-/- mice, newborn Pkd1CD;Cux1tm2Ejn+/- mice did not show any cysts. Comparison of Pkd1CD and Pkd1CD;Cux1tm2Ejn+/- mice at later stages of development showed a reduction in the severity of PKD in the Pkd1CD;Cux1tm2Ejn+/- mice. Moreover, we observed an increase in expression of the cyclin kinase inhibitor p27, a target of Cux1 repression, in the rescued collecting ducts. Taken together, our results suggest that Cux1 expression in PKD is not directly involved in cystogenesis but promotes cell proliferation required for expansion of existing cysts, primarily by repression of p27.

Clinical and Biological Implications of CUX1 Mutations in Myeloid Neoplasms

Recurrent somatic mutations of CUX1 are described in myeloid neoplasms. CUX1 is located at chromosome 7q22.1; -7/del(7q) involving CUX1 locus are common abnormalities in myelodysplastic syndromes (MDS). Mutations and loss of heterozygosity involving CUX1 have been also described in breast, lung and uterine cancers. Preliminary functional studies, lack of a mutational hotspot and coincidental deletions suggest loss of function/hypomorphic consequences of these molecular defects. CUX1 (p200), contains 4 evolutionarily conserved DNA-binding domains, including 3 CUT repeats and a CUT homeodomain. Functionally, CUX1 regulates many genes involved in DNA replication and chromosome segregation. Cell-based assays have established a role for CUX1 in the control of cell-cycle progression, cell motility, and invasion .The objective of this study is to assess the molecular context and clinical significance of CUX1 mutations and deletions in myeloid neoplasms.We analyzed a subset of 1478 patients [24% lower-risk MDS, 17% higher-risk MDS, 22% primary (p)AML, 14% secondary AML, 14% MDS/myeloproliferative neoplasms (MPN) and 9% MPN] for the presence of CUX1 mutations and deletions. No CUX1 mutations were found in core binding factor AML. We correlated the presence of these lesions with clinical parameters, cytogenetic abnormalities, and molecular features including clonal architecture and associated somatic mutations. Copy number variation and their boundaries were analyzed by Single Nucleotide Polymorphism (SNP) arrays and mutations by multiamplicon deep sequencing utilizing a panel targeting 60 most commonly mutated genes in myeloid neoplasms.In total cohort 4 % of patients had CUX1 mutations and 6% had locus deletions (affecting ch 7q commonly deleted region: 7q22.1) including 90% of del (7q) cases. Expression of CUX1 is significantly lower in AML with -7/del(7q) compared to AML with normal cytogenetics (p<.00001) and also in MDS with -7/del(7q) compared tohealthy controls (p=.004). Additionally, decreased expression of CUX1 was found in 15% of MDS and 8% of AML patients without -7/del(7q) or related mutations. Cases with lower expression had worse OS compared to patients with higher expression (p=.002). In terms of configuration, most mutations were heterozygous, 5% of mutations were hemizygous and 4% were homozygous (due to UPD). Among 75 somatic CUX1mutations; 72% were missense, 20% where frame shift and 8% where non sense.CUX1 mutations were associated with either lower-risk MDS (p=.0001) and pAML (p=.04) while deletions involving the CUX1 locus were significantly related to higher-risk MDS (p=.05). Heterozygous CUX1 mutations were more commonly associated with normal cytogenetics (p=.01). Patients with -7/del(7q) frequently represented del(5q) (p=.04) and thrombocytopenia (p=.001). The OS of patients with CUX1 mutations was shorter (p=.04) as was that of patients with CUX1/deletions (p=.02) when compared to wild type.We subsequently studied the molecular background of CUX1 alterations. CUX1 mutations (vs. wild type) were associated with TET2 (31% vs. 14%, p=.006), ASXL1 (29% vs. 9%, p=.0005), BCOR (28% vs. 8%, p=.0004), and cohesion mutations (26%, vs. 5%, p=.0005), while NPM1 mutations showed the reverse relationship (1% vs. 7%, p=.03). RAS and CUX1 mutations were mutually exclusive (0% vs. 6%, p=.03).When we analyzed clonal hierarchy in the context of CUX1 mutations; dominant CUX1 mutations (24%; mean VAF=49%); were accomplished by ASXL1 (21%) and SRSF2 (14%) mutations which were the most common secondary events in this context. Phenotypically, dominant CUX1 mutations were associated with MDS/MPN (42%) and MDS (33%). 14% of CUX1 mutant cases did not harbor any other alterations and were not associated with a discernable phenotype. Secondary CUX1 lesions (62%; mean VAF=22%) were found in the context of dominant TET2 mutations (16%). The pathomorphologic context of secondary CUX1 mutation did not differ from that of primary lesions. AML seemed to be underrepresented (p=.006) and MPN overrepresented (p=.019) among dominant CUX1 mutant cases.In conclusion, CUX1 lesions including locus deletions with haploinsuffciency, mutations and a fraction of cases with decreased CUX1 expression can be encountered in MDS and related neoplasms, chiefly AML. CUX1 dysfunction is associated with poor survival likely due to its distinct molecular background. DisclosuresMakishima:The Yasuda Medical Foundation: Research Funding. Sekeres:Millenium/Takeda: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees.

Somatic Mutations Are Frequently Detected in Chronic Myeloid Leukemia in Chronic Phase and Do Not Affect Response to Tyrosine-Kinase Inhibitors

The development of chronic myeloid leukemia (CML) is driven by a unique genetic aberration, i.e. the BCR-ABL rearrangement. Somatic mutations typical of myeloid neoplasms have been described in patients with CML, mainly associated with blastic phase (BP) or with additional chromosomal aberrations (ACA). Recently, somatic mutations have been identified in a significant proportion of healthy subjects without a hematological phenotype, resulting in a significantly higher risk of developing a hematologic cancer. However, the clinical relevance of these mutations in patients with chronic phase (CP)-CML needs to be clarified. In this study, we performed a comprehensive mutation analysis of genes implicated in myeloid malignancies in a large cohort of CP-CML patients with the aim to estimate their prevalence and the impact on response to TKI and clinical outcomes.Seventy-four patients (pts) with CP-CML were enrolled, after providing written informed consent (M/F: 44/30; median age 54 years, range 18-82; low or intermediate Sokal risk score: 85%; no ACA at diagnosis or during follow-up).Samples were collected in 53 pts (group1, median time from TKI start: 66 months, range 3-177). In addition, in order to better clarify the clonal architecture, 21 consecutive pts were analyzed at diagnosis and sequentially during follow-up (group 2, median follow-up 16 months, range 3-47, median number (no.) of samples per pt 3, range 2-5). Targeted sequencing of 54 genes was performed using an Illumina HiSeq platform (Illumina, San Diego, CA) in bone marrow or peripheral blood granulocytes, whereas CD3+T lymphocytes were used as control tissue.At least one somatic mutations was identified in 14/53 (26.4%) patients in the group1; out of these, 5 pts showed 2 mutations, and 1 pt showed 3 concomitant mutations. The most frequently mutated genes were DNMT3A (11%), TET2 (6%), ASXL1 (6%), as reported in age-related clonal hematopoiesis. In addition, somatic mutations in CUX1 and KDM6A were observed (4% each). No mutations in splicing factors were identified. Median variant allele frequency (VAF) was 0.07 (range 0.02-0.49). Most of mutated pts (11/14, 78.6%) were in MR3 at the time of sampling (median IS 0.014%, range 0-0.09%), thus suggesting that additional clones were independent from the BCR-ABL-positive one.In pts from group 2 somatic mutations were identified in 12/21 (57.1%) cases: 2 pts harboured mutations at diagnosis that remained stable despite of clearance of BCR-ABL (JAK2 and MPL, in pts with associated Ph-negative myeloproliferative neoplasm). The other 10 pts showed transient small clones (median VAF 0.06, range 0.02-0.1), not detectable at diagnosis and fluctuating during follow-up. Only one of these pts had a STAG2 mutation at diagnosis, which disappeared during follow-up along with clearance of BCR-ABL.The mutations most frequently identified involve genes demonstrated to be recurrently mutated in healthy subjects and associated with increasing age. In our cohort, no significant association was found between age and either presence (median age of wild-type pts 60 years vs 61 in mutated pts, P=0.35) and number (rho=0.13, P=0.26) of somatic mutations. However, the incidence of mutations was significantly higher in CP-CML pts in comparison with the age-matched general population (Jaiswal et al, NEJM, 2014), especially in young patients (P<0.001).After a median follow-up in the whole cohort of 5.3 years (range 0.4-38.4), no progression to accelerated phase/BP was observed. In multivariate analysis, mutational status did not affect the probability of achieving MR3 or better under TKI therapy, neither when evaluating the presence of somatic mutations (HR=4.1, 95%CI: 0.7-23.3, P=0.111) nor when we considered the no. of mutations per pt (HR=1.9, 95%CI: 0.8-4.4, P=0.148).In conclusion, this study showed that somatic mutations in genes frequently mutated in myeloid neoplasms and associated with age-related clonal hematopoiesis are detectable in CP-CML pts with a significantly higher incidence compared with the general population. These mutations are likely driving small transient clones, independent from the Ph-positive one, and become detectable when BCR-ABL is suppressed by TKI therapy. The presence of these mutations do not affect the probability of achieving an optimal molecular response to TKIs, although this observation needs to be confirmed in larger studies. DisclosuresOrlandi:Ariad: Honoraria; BMS: Honoraria; Novartis: Honoraria.

The Homeodomain Protein Cux1 is Required for Cyst Development in an ADPKD Mouse Model

Cux1 is a homeobox gene involved in cell cycle regulation and kidney development. Cux1 is upregulated in the kidneys of both mice and humans with autosomal dominant polycystic kidney disease (ADPKD). However, Cux1 transgenic mice do not develop cystic kidneys, indicating that upregulation of Cux1 alone is insufficient to develop PKD. Mice carrying a deletion of the C‐terminus of Cux1 (Cux1tm2Ejn‐/‐) express a truncated protein that does not localize to the nucleus, resulting in functional loss of Cux1 transcriptional activity. To test the hypothesis that Cux1 is required to develop PKD, we generated collecting duct specific Pkd1 mutant mice (Pkd1CD) that were also homozygous for the Cux1 mutation. When kidneys isolated from newborn Pkd1CD/Cux1tm2Ejn ‐/‐ mice were examined, we found a marked reduction in cystic dilations compared to the kidneys isolated from newborn Pkd1CD mice. We confirmed the Cre activity by crossing the Pkd1CD and Pkd1CD/ Cux1tm2Ejn ‐/‐ mice to the membrane‐targeted Tomato/Green Fluorescent Protein (mT/mG) double fluorescent reporter mouse strain (Gt(ROSA)26Sortm4(ACTB‐tdTomato,‐EGFP)Luo/J), which carries a loxP‐flanked Tomato cassette. Immunofluorescence of kidney sections indicated that collecting duct derived cysts were positive for the excision event in the Pkd1CD mice. In contrast, although collecting ducts were positive for the excision event in Pkd1CD/ Cux1tm2Ejn ‐/‐ mice, none of the positive tubules were cystic. Taken together, our results suggest that Cux1 gene dosage influences the progression of cystic disease in an ADPKD mouse model.

Somatic Mutation-Detecting Algorithm Enables Analysis of MDS Patients during Azacitidine Therapy

Introduction: Somatic mutation detection in myelodysplastic syndrome (MDS) is very important in deciphering clonal pathogenesis of every patient and if determined correctly will become useful tool in followup studies such as testing individual susceptibility to epigenetic therapy with azacitidine (AZA). While some patients respond to AZA by restoring hematologic parameters, others progress to AML. Recent identification of quite heterogeneous sets of mutated genes (Bejar R et al. 2013) suggested that: patients with specific mutation pattern/s may respond to epigenetic therapy differently.Aim: We herein set to determine mutation profiles of MDS cohort indicated to and treated by AZA and utilized TrueSight DNA amplicon NGS sequencing approach containing 54 genes all previously associated with MDS or AML.Patients: We analyzed immunomagnetically CD3-depleted bone marrows of two MDS patients - AZA responders. First patient (male, 68y), was diagnosed with RAEB2, IPSS int-2, transfusion dependent (4 TU/Mo), intermediate cytogenetics (tri21). Following 4 cycles of AZA (75 mg/m2 s.c., 5+2) the patient responded by partial remission, and AZA was discontinued after 17 cycles. Twelve months after discontinuation he progressed and AZA was readministered for additional 3 cycles and the patient achieved again partial remission. Analyzed are samples after 11 (P394) and 20 (P1380) cycles of Vidaza. Second patient (female, 64y), was diagnosed with RAEB2, IPSS high, transfusion dependent (2 TU/Mo), favorable cytogenetics (46XX). Following 4 cycles of AZA (75 mg/m2 s.c., 5+2) she responded by hematology improvement and later by partial remission. Analyzed is a sample after 4 (P1510) cycles of Vidaza. As negative controls we used two normal donor bone marrows from 41y male and 32y female. As a positive control we also used: 1 MDS/AML cell line MOLM-13 with previously identified mutations of CBL and FLT3 (DSMZ; ACC 554).Methods and approach: Samples were sequenced on Illumina MiSeq sequencer. The mapping was performed using Burrows-Wheeler Aligner algorithm. Illumina Somatic variant caller was used to identify mutations. Then we applied following filters on the data: sequencing coverage should be higher than 1000 per mutation (~80% data left), mutation should be heterozygous (~95% data left), mutation frequency should be higher than 10% (~10% data left), Illumina Somatic variant caller should flag the mutation as "PASS" (~50% data left), mutation should not be synonymous (~75% data left) and mutation should be exonic (~40% data left). These filters were also applied to find mutations in the two control samples. Those mutations which were identified also in the control samples were removed from the analysis of patient samples (~50% data left).Results: the MDS/AML cell line MOLM-13 contained mutations (SNVs or InDels) in ABL1 (SNV/frequency=46.5%), ASXL1 (SNV/49.8%), CEBPA (In/47.9%), HRAS (SNV/54.5%), TET2 (SNV/49.7%), and as expected also in the genes encoding CBL (delta-exon8/52.4%) and FLT3 (ITD/50.6%). Patient’ sample P1510 contained mutations in CBL (SNV/67.9%), CUX1 (SNV/51.8%), IKZF1 (2 different SNVs/41.9 and 50.7%), KDM6A (SNV/51.6%), SF3B1 (SNV/38.9%), and SMC3 (SNV/33.1%). Patient samples P394 and P1380 contained mutations in the ASXL1 (SNV/ 35.5% and 32.6% respectively), CUX1 (2x SNVs, first SNV/46.1->64.5%, second 48.4->57.9%), and IKZF1 (SNV, 50.4->44.5%) in similar frequencies in the sample before and after 2.5 years (including 9 cycles of AZA) suggesting limited genetic heterogeneity in this AZA-responding patient. Consequently, to gain more insight into how AZA modulates mutation pattern in MDS, we now analyze a set of fourty nine additional patients before and following at least 4 cycles on AZA treatment.Conclusions: Our data support use of immunomagnetic CD3-depletion of bone marrow and addition of normal control samples in the sequencing of MDS patient samples and support this approach for testing genetic heterogeneity during MDS disease course upon AZA treatment. DisclosuresStopka:GAČR P305/12/1033 and UNCE 204021: Research Funding; Celgene: Research Funding; PersMed ltd.: Equity Ownership. Vargova:GAČR P305/12/1033 and P305/11/1745: Research Funding; UNCE 204021: Research Funding; PRVOUK P24/LF1/3: Research Funding. Kulvait:PersMed ltd.: Equity Ownership. Jonasova:PRVOUK P24/LF1/1: Research Funding; Celgene: Research Funding.

Genomic Profile of Patients with Triple Negative (JAK2, CALR and MPL) Essential Thrombocythemia and Primary Myelofibrosis

Introduction: Essential Thrombocythemia (ET) and Primary Myelofibrosis (PMF) are myeloproliferative neoplasms (MPN) with similar driver mutations. The three hallmark molecular alterations in these diseases are JAK2, MPL and CALR mutations. Patients with myelodysplastic/myeloproliferative (MPN/ MDS) neoplasms such as refractory anemia with ring sideroblasts and thrombocytosis (RARS-T) also present with the same hallmark genetic changes. Nevertheless, roughly 10% of these patients do not present mutations in neither of these genes. Recent data suggest that triple negative patients have a more aggressive clinical course. While the molecular alterations present in patients with MPN have been extensively studies, the genomic profile of triple negative TE/PMF/RARS-T has not been described. To better characterize these patients we performed whole exome / genome sequencing of paired granulocytes and skin from 15 triple negative MPN patientsMethods: A total of 15 patients with triple negative MPN or MPN/ MDS [PMF (N=6)/TE (N=8)/RARS-T (N=1)] were analyzed. DNA was extracted from CD66b+ magnetic bead selected granulocytes (EasySep, Stem Cell Technologies) and matched skin biopsies with QiaAmp DNA Mini kit (Qiagen).Whole-exome targeted capture was carried out on 3 μg of genomic DNA, using the SureSelect Human Exome Kit 51Mb version 4 (Agilent Technologies, Inc., Santa Clara, CA, USA). The exome library was sequenced with 100 bp paired-end reads on an Illumina HiSeq2000. Somatic variants calls were generated by combining the output of Somatic Sniper (Washington University), Mutect (Broad Institute) and Pindel (Washington University). Tumor coverage was 150x and germline coverage was 60x. The combined output of these 3 softwares was further filtered by in-house criteria in order to reduce false-positive calls (minimum coverage at both tumor/germline ≥8 reads; fraction of reads supporting alternate allele ≥5% in tumor and ≤10% in germline; ratio of allele fraction tumor:germline >2). All JAK2 and CALR mutations were validated through Sanger sequencing. Validations of other somatic mutations are under way at this point.Results: First we asked whether other hematopoietic related genes could be responsible for the pathogenesis of the triple negative cases. With that goal we searched for high confidence mutations in genes that are mutated in at least 1% of patients with hematopoietic tumors on COSMIC (catalog of somatic mutations in cancer) database and also genes known to be recurrently mutated in myeloid malignancies. Only 6 out of 15 patients presented mutations in other myeloid related genes. The diagnosis of these patients were PMF=4, TE=2. The hematopoietic related genes mutated in these patients were: ASXL1 (n=4), CUX1 (n=3), NRAS (n=2) and ATM, CBL, CSFR3, CREBBP, DNMT3A, ETV6, EZH2, JARID2, MLL2, PHF6, SRSF2, STAG2, TET2, GNAS, U2AF1 (n=1).Noteworthy, we have found one known oncogenic mutation in CSFR3, an alteration supposed to be specific for chronic neutrophilic leukemia and atypical CML, in a patient with ET, and an oncogenic mutation in GNAS in a patient with PMF. In addition, the patient with RSRA-T had a putative oncogenic mutation in PTPN11Remarkably, the average number of hematopoietic related mutations in these patients was 5, significantly higher than the total number of mutations found in another cohort of patients with either JAK2 (average = 1.7) or CALR mutations (average = 1.9). Although our numbers are small, we may speculate that the high incidence of ASXL1 mutations (28%) associated with a high number of prognostically detrimental mutations can partially explain the worse outcomes associated with triple negative MPN. Noteworthy is also the high prevalence of CUX1 mutations in this subset of patients (21%) when compared to other myeloid malignancies in general.Regarding the other 9 patients for whom no hematopoietic mutations could be identified, 6 patients had ET, 2 patients PMF and one patient had RARS-T.Conclusion: We have shown that: i-patients with triple negative MPN are molecularly heterogeneous, with one group presenting a high number of hematopoietic related mutations, ii-the most common mutations present in these patients are ASXL1, CUX1 and NRAS, iii-The majority of these patients do not present mutations in hematopoietic related genes, what suggests that non-described molecular mechanisms are operating in these patients. DisclosuresNo relevant conflicts of interest to declare.

Somatic Mutations and Loss-Of-Heterozygosity Impair The DNA Repair Functions Of CUX1 in Myelodysplastic Syndromes (MDS)

Chromosome 7 lesions are common and are associated almost uniformly with a deleterious outcome in MDS and related myeloid neoplasms. We analyzed a large cohort of these patients (pts) (n=1595) and identified those with loss of heterozygosity (LOH) of chromosome 7, including del(7q), monosomy7 (-7) and UPD(7q) in 171 cases. Using single nucleotide polymorphism (SNP)-array karyotyping, 3 commonly deleted regions (CDR) have been isolated, including 7q22, 7q34, and 7q35-36. CDR 7q22 was involved in 119 cases and spanned among others PLOD3, RBL5 and CUX1 genes, which could play a role in del(7q) pathogenesis. To further investigate the molecular pathogenesis of -7/del(7q), we performed whole exome new generation sequencing (NGS) in 428 pts with MDS and related conditions; this cohort included 72 -7/del(7q) or UPD7. When we studied the mutational status of genes in 7q22, we noted 5 cases of CUX1mutations, including 2 with heterozygous and 3 with homozygous mutations (associated with UPD7q).These mutations were validated by Sanger sequencing and targeted deep sequencing of DNA from both tumor and normal cells. All CUX1 mutations revealed critical structural and functional determinants such as nonsense mutations resulting in premature stop codons (n=3), nonsense mutation located in the DNA-binding Cut homeodomain (p.R1296K, n=1) and splice site mutation (n=1). CUX1 mutation occurred in MDS (n=2) and MDS/MPN (n=3), and concomitant TET2 mutations were seen in 4 cases. No hemizygous CUX1mutations were found in cases with del(7q).CUX1 is haploinsufficient in cases with -7/del(7q). In total 166 cases (10.9%) had -7/del(7q) involving CUX1, in addition to pts in whom CUX1 was affected by likely hypomorphic/inactivating mutations. Pts with decreased expression of CUX1 had poor survival compared to pts without CUX1 (p<.01, HR=1.99), suggesting that deficient function (deletion or mutation) of CUX1affects disease progression.In addition to somatic mutations, LOH of CUX1 has been reported in several cancers. A role of CUX1 as a haploinsufficient tumor suppressor cannot be explained by the known functions of CUX1 in stimulating cell proliferation, motility and resistance to apoptosis. We recently identified a novel molecular function of CUX1 in DNA repair that may explain how haplo-deficient expression of CUX1 contributes to leukemic transformation. We used single cell gel electrophoresis (comet assay) to show that Cux1-/- mouse embryo fibroblasts (MEFs) are deficient, while Cux1+/- MEFs are haploinsufficient, in the repair of oxidative DNA damage. Using an inverse-PCR assay, following etoposide exposure, the frequency of chromosomal translocations involving the mixed lineage leukemia (Mll) gene is significantly and progressively increased in Cux1+/- and Cux1-/- MEFs compared to Cux1+/+ MEFs. We then performed comet assays using primary leukemic cells that harbor a frameshift mutation predicted to inactivate CUX1 by producing a C-terminally truncated protein devoid of a nuclear localization signal. Repair of oxidative DNA damage was delayed in leukemic cells compared to bone marrow cells from a healthy donor. Using an in vitrobase excision repair assay, we show that repair of 8-oxoguanine is reduced in a cell line displaying LOH of CUX1. Similarly, repair of 8-oxoguanine is reduced following siRNA-mediated CUX1 knockdown, but is rescued by the addition of a purified CUX1 protein. Together these results demonstrate that CUX1 plays a direct role in DNA repair and that inactivation of one CUX1 allele reduces the DNA repair capability of cells.In conclusion, novel somatic mutations of CUX1 as a candidate gene are associated with poor prognosis in MDS pts with -7/del(7q) and UPD7. Somatic events constitute loss of function of CUX1, resulting in insufficiency of DNA repair mechanisms, which is associated with leukemogenesis and could be considered as a new therapeutic target. CUX1 mutations may affect base excision repair, and dysfunction of CUX1 could theoretically predispose to chromosomal translocations and complex karyotype, seen in conjunction with del(7q) cases. Disclosures:Makishima:AA & MDS international foundation: Research Funding; Scott Hamilton CARES grant: Research Funding. Maciejewski:NIH: Research Funding; Aplastic anemia&MDS International Foundation: Research Funding.

Spliceosomal Gene LUC7L2 Mutation Causes Missplicing and Alteration Of Gene Expression In Myeloid Neoplasms

While deletion of the long arm of chromosome 7 (del(7q)) along with monosomy 7 (-7) are common in myeloid neoplasms and especially MDS, their associated pathogenetic consequences and the genes responsible for the clinico-morphologic phenotypes remain unknown. To characterize the molecular defects resultant from del(7q), we applied a combined analysis of SNP-array karyotyping, whole exome NGS, targeted deep NGS and deep whole RNA NGS to facilitate identification of somatic mutations, loss of heterozygosity (LOH) and haploinsufficiency. First, using a cohort of 1595 patients, we precisely defined 3 commonly deleted regions in 7q22, 7q34, and 7q35-36. To identify genes involved in the pathogenesis of del(7q), we applied whole exome NGS to 428 patients with MDS and related conditions, including 72 cases of -7/del(7q) or UPD7. We found that both recurrent deletions with haploinsufficient expression and somatic nonsense/frameshift mutations were present in the genes CUX1 (7q22), LUC7L2 (7q34) and EZH2 (7q36), located on 7q. For this project we focused our study on the LUC7L2 gene encoding a spliceosomal protein that interacts with U1 snRNP to recognize 5’ splice sites. This function is in contrast to other spliceosomal genes mutated in MDS, such as SF3B1, U2AF1, SRSF2 and SF3A1that interact with the U2 snRNP complex and the 3’ splice site. Initially, we found 8 cases of LUC7L2 mutation including 2 cases of hemizygous and 1 case of homozygous mutations; all LUC7L2 mutations result in premature stop codons. The concomitant mutational spectrum in LUC7L2 mutant cases differed from those with EZH2 or CUX1 mutations. For example, TP53 and TET2 are mostly mutually exclusive with LUC7L2 mutations. While LUC7L2 mutant cases were low grade-MDS (RCUD or RCMD) or CMML, the survival impact of LUC7L2 mutation was similar to that seen in -7/del(7q) (LUC7L2; HR=2.36). Most of the mutations are heterozygous (diploid chr7) or homozygous (UPD7q). However, in addition to a few hemizygous mutations, wild type expression of LUC7L2 is haploinsufficient in del(7q). In total we found 117 cases (9%) with del(7q) and resultant decreased expression of LUC7L2, showing comparably poor survival as in the mutant cases (HR=1.99). Next, we hypothesized that dysfunction of LUC7L2 causes splicing defects (missplicing) in specific genes involved in leukemogenesis. Using deep RNA NGS we compared splicing patterns of 201,837 exons between the cases with deficient function of LUC7L2 (mutation/low-expression; n=11) versus wild-type (n=11), and identified concordant alternative splicing patterns in 44 genes (increased exon skipping in 27 genes; increased exon retention in 17 genes). For example, we found abnormal splicing of genes involved in functionally important pathways including the RAS pathway (NF1) and the TGF-β pathway (SMAD5). In cases of mutation/low-expression of LUC7L2, as a result of increased skipping of NF1 exon 31(NM_001042492), the type I isoform of NF1 predominates as compared to wild type (types I/II ratio; 0.76 vs. 0.45, p&lt;.01, respectively). Missplicing also occurred in SMAD5 exon 2 (5’UTR site) resulting in a significantly lower SMAD5 expression. In agreement with RNA sequencing results, lentivirus-mediated shRNA knockdown of LUC7L2 in K562 cells resulted in the concordant effects on NF1 type 1 isoform and SMAD5 exon 2 skipping. Similar splicing patterns were observed in MDS cases with -7/del(7q), involving the LUC7L2gene. In conclusion, novel somatic mutations of LUC7L2 suggest that it could be a candidate gene associated with the poor prognosis of -7/del(7q) and UPD7. Loss of function or low expression of LUC7L2 results in distinctly altered splicing patterns involving genes associated with proliferation or leukemogenesis. Disclosures: Makishima: AA &amp; MDS international foundation: Research Funding; Scott Hamilton CARES grant: Research Funding.

High Throughput Targeted Gene Sequencing in 738 Myelodysplastic Syndromes Patients Reveals Novel Oncogenic Genes, Rare Driver Mutations and Complex Molecular Signatures with Potential Impact for Patient Diagnosis and Prognosis in the Clinic

AbstractAbstract LBA-5Myelodysplastic syndromes (MDS) are a diverse group of chronic hematological malignancies characterised by dysplasia, ineffective hematopoiesis and a variable risk of progression to acute myeloid leukemia (AML). Recent screens of MDS genomes have identified a diverse range of mutations in genes implicated in RNA splicing, DNA modification, chromatin regulation and cell signalling. Whilst these findings improve our understanding in the molecular aetiology of this disease a detailed understanding of the specific genetic contributions, genotype-phenotype relationships and prognostic value of this heterogeneous genomic landscape is warranted.To evaluate the genomic architecture of MDS in more detail, targeted sequencing analysis was performed for 111 genes implicated in myeloid malignancy across 738 MDS patients. Sequencing data were assessed for the presence of oncogenic mutations and detection of regional copy number changes. High-confidence oncogenic mutations (n = 1155) were identified in 41 genes. Of these 41 genes, only 4 (SF3B1, TET2, SRSF2, ASXL1) were mutated in >10% of patients with 15 genes mutated between 10% and 1% of patients and 22 in < 1%. Overall, oncogenic lesions are identified in 78% of MDS patients with 43% of the patients displaying two or three oncogenic events and approximately 10% of patients harbouring four or more oncogenic mutations. Amongst the less characterised genes, we confirm inactivating mutations in CUX1 and IRF1 in 2% and 0.5% of the patients respectively and provide strong evidence that these two genes are bona-fide rare tumour suppressors in MDS. We observe complex patterns of pairwise association between genes, including co-mutation and mutual-exclusivity. Of 325 possible pairwise associations, 47 driver gene mutations pairs were significant (FDR < 10%). Mutations in components of the U2-spliceosome machinery are characteristic of 51% of patients with MDS and whilst mutually exclusive with one another, spliceosome mutations show a strong tendency to co-occur with genes implicated in DNA methylation.Of 24 genes mutated in >5 patients, 8 genes were associated with significantly worse leukemia-free survival if mutated and one gene (SF3B1) with a better leukemia-free survival replicating findings from previous studies. For the new MDS genes described, loss of function truncating mutations in CUX1 are associated with a poor prognosis (p=0.02). Additionally, clinical outcome analysis identifies functional non-canonical mutations within well-characterised driver genes. We find that there is an inverse correlation of leukemia-free survival with the total number of oncogenic lesions observed (p<0.0001). Patients with fewer oncogenic events display lower rates of leukemic transformation and better overall survival, whereas patients with higher number of mutations have a significantly worse outcome (p<0.0001). We evaluate the prognostic potential of targeted genomic sequencing data by means of the area under the receiver operating characteristic (AUC) and compare the predictive value to that of conventional markers utilised in the clinic, including the IPSS classification, peripheral blood counts, bone marrow morphology and cytogenetic analysis. Strikingly we find that the prognostic information contained within the gene sequencing data and peripheral blood counts at diagnosis appears to be equivalent to that derived from existing outcome indicators, suggestive of a potential redundancy between the molecular markers and the morphological variables used to classify patients. Most importantly the number of oncogenic point mutations is an independent biomarker and may therefore be used to refine the prognosis based on existing IPSS categories.The present study provides an unprecedented characterisation of the genomic lesions implicated in MDS pathogenesis, refines the classification of oncogenic driver mutations and reveals a diverse molecular landscape of mostly rare mutations with valuable insights on the operating biological networks. The absence of known oncogenic mutations in 22% of MDS indicates that additional yet uncharacterised genomic events are likely to be operating in MDS pathogenesis. Most importantly our data highlight the potential of well-defined gene mutations as reproducible diagnostic and predictive biomarkers to support MDS patient management and clinical decision-making in the future. Disclosures:No relevant conflicts of interest to declare.

Acceleration of polycystic kidney disease progression in cpk mice carrying a deletion in the homeodomain protein Cux1

Polycystic kidney diseases (PKD) are inherited as autosomal dominant (ADPKD) or autosomal recessive (ARPKD) traits and are characterized by progressive enlargement of renal cysts. Aberrant cell proliferation is a key feature in the progression of PKD. Cux1 is a homeobox gene that is related to Drosophila cut and is the murine homolog of human CDP (CCAAT Displacement Protein). Cux1 represses the cyclin kinase inhibitors p21 and p27, and transgenic mice ectopically expressing Cux1 develop renal hyperplasia. However, Cux1 transgenic mice do not develop PKD. Here, we show that a 246 amino acid deletion in Cux1 accelerates PKD progression in cpk mice. Cystic kidneys isolated from 10-day-old cpk/Cux1 double mutant mice were significantly larger than kidneys from 10-day-old cpk mice. Moreover, renal function was significantly reduced in the Cux1 mutant cpk mice, compared with cpk mice. The mutant Cux1 protein was ectopically expressed in cyst-lining cells, where expression corresponded to increased cell proliferation and apoptosis, and a decrease in expression of the cyclin kinase inhibitors p27 and p21. While the mutant Cux1 protein altered PKD progression, kidneys from mice carrying the mutant Cux1 protein alone were phenotypically normal, suggesting the Cux1 mutation modifies PKD progression in cpk mice. During cell cycle progression, Cux1 is proteolytically processed by a nuclear isoform of the cysteine protease cathepsin-L. Analysis of the deleted sequences reveals that a cathepsin-L processing site in Cux1 is deleted. Moreover, nuclear cathepsin-L is significantly reduced in both human ADPKD cells and in Pkd1 null kidneys, corresponding to increased levels of Cux1 protein in the cystic cells and kidneys. These results suggest a mechanism in which reduced Cux1 processing by cathepsin-L results in the accumulation of Cux1, downregulation of p21/p27, and increased cell proliferation in PKD.