Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

The Functional Role of PRC2 in Early T-cell Precursor Acute Lymphoblastic Leukemia (ETP-ALL) – Mechanisms and Opportunities

The Functional Role of PRC2 in Early T-cell Precursor Acute Lymphoblastic Leukemia (ETP-ALL) –... Review published: 18 May 2016 doi: 10.3389/fped.2016.00049 The Functional Role of PRC2 in early T-cell Precursor Acute Lymphoblastic Leukemia (eTP-ALL) – Mechanisms and Opportunities 1 2 1 Kathrin M. Bernt , Stephen P. Hunger and Tobias Neff * Department of Pediatrics, Center for Cancer and Blood Disorders, Children’s Hospital Colorado, University of Colorado Denver, Aurora, CO, USA, Department of Pediatrics, Center for Childhood Cancer Research, The Children’s Hospital of Philadelphia, Philadelphia, PA, USA Early T-Cell precursor acute lymphoblastic leukemia (ETP-ALL) is a relatively newly iden- tified subset of T-lineage ALL. There are conflicting results regarding prognosis, and the genetic basis of this condition is variable. Here, we summarize the current status of the field and discuss the role of mutations in the Polycomb Repressive Complex 2 frequently identified in ETP-ALL patients. Keywords: leukemia, lymphoid, epigenetics, polycomb repressive complex, Hox genes, JAK/STAT signaling pathway Edited by: Brenton Garrett Mar, e TP-ALL, A NOve L SUBTYPe OF ACUTe LYMPHOBLASTiC Dana Farber Cancer Institute, USA Le UKe MiA Reviewed by: Kathryn Grace Roberts, In 2009, Coustan-Smith and colleagues reported a subtype of T-lineage acute lymphoblastic leuke- St. Jude Children’s Research mia (T-ALL) with transcriptional and surface marker profile similarities to early T-cell precursors Hospital, USA Alejandro Gutierrez, (ETP-ALL) (1). Clinical characteristics of ETP-ALL include associations with lower WBC, older Boston Children’s Hospital, USA age, and a very high rate of poor MRD response or even induction failure at the end of induction *Correspondence: chemotherapy. Tobias Neff Initial reports suggested that ETP-ALL has an extraordinarily poor long-term prognosis, with tobias.neff@ucdenver.edu 2- to 10-year event free/overall survival (EFS/OS) rates in the 11–40% range, compared to 84–90% OS for all pediatric T-ALL in a similar time frame (1–3). However, patients in these cohorts were Specialty section: not treated on the most recent treatment protocols [e.g., 1992–2006 for the St. Jude cohort and This article was submitted to 2001–2006 for the AIEOP cohort (1)] and included only a small number of ETP-ALL patients: 17 Pediatric Hematology and in the St Jude, 13 patients in the AEIOP (1), 5 in the Tokyo Children’s Cancer Study Group L99-15 Hematological Malignancies, cohort (2), 12 in the Shanghai Children’s (3), and 7 (3 pediatric and 4 adult) treated at Columbia a section of the journal University (4). Poor outcomes were also reported for adult patients with immune-phenotypic ETP- Frontiers in Pediatrics ALL or T-ALL with a gene expression profile characteristic of ETP-ALL (5–7). Received: 13 April 2016 Accepted: 02 May 2016 Published: 18 May 2016 e TP-ALL iN Re LATiON TO OTHe R De FiNiTiONS OF Citation: iMMATURe  T-ALL Bernt KM, Hunger SP and Neff T (2016) The Functional Role of PRC2 Early T-cell precursor acute lymphoblastic leukemia is defined based on immunophenotyping in Early T-cell Precursor Acute (CD1a-negative, CD8-negative, weak CD5 expression with less than 75% positive blasts, and expres- Lymphoblastic Leukemia (ETP- sion of one or more of the following myeloid or stem cell markers on at least 25% of lymphoblasts: ALL) – Mechanisms and CD117, CD34, HLA-DR, CD13, CD33, CD11b, and/or CD65). It is noteworthy that other definitions Opportunities. of very immature T-ALL subtypes have been proposed. In an important study, using array-based Front. Pediatr. 4:49. doi: 10.3389/fped.2016.00049 genome-wide expression profiling, Ferrando et al. found that T-ALL with LYL1 expression clusters Frontiers in Pediatrics | www.frontiersin.org 1 May 2016 | Volume 4 | Article 49 Bernt et al. PRC2 in ETP-ALL TABLe 1 | Overlap between different methods to classify immature separate from other T-ALL subtypes and show immunopheno- T-lineage ALL [data from Ref. (13)]. typically immature features (8). The Meijerink-group identified an immature subgroup of T-ALL with high MEF2C expression, which based on expression profiling clustered separately from other T-ALL cases and was enriched for ETP-ALL cases (9). This same laboratory went on to further characterize the overlap between ETP-ALL defined by gene expression profile, ETP-ALL based on immunophenotyping, MEF2C-high T-ALL, and a third group of immature T-ALL originally described by Gutierrez et al. defined by absence of biallelic TCRgamma deletion (ABD) (10). Poor prognosis for T-ALL and T-lymphoblastic lymphoma with ABD was also found in two other studies (11, 12). Meijerink and colleagues systematically compared the different definitions of immature T-lineage ALL. There is excellent overlap between the immature cluster and ETP-ALL as defined by an ETP-ALL gene expression signature (13). There is less overlap between the immature cluster and ETP-ALL defined by immunophenotyping. There is excellent overlap between clustering and PAM based on a defined ETP-ALL Inclusion of the CD5 marker in flow-cytometry panel reduced signature. Stringent ETP-ALL criteria including CD5 low criterion exclude many samples the number of cases classified as phenotypic ETP-ALL. Omission classified as immature by gene expression profile. Relaxing the CD5 criterion (“near of the CD5 marker improved the number of expression-based ETP”) includes a number of samples that do not display an ETP-ALL expression profile. PAM, prediction analysis of microarrays. ETP-ALL cases that also were immunophenotypically defined as ETP-ALL. However, the less stringent flow criteria also led to the classification as immunophenotypic ETP-ALL of samples [methodology harmonized with the approach used by Coustan- without an ETP-ALL expression profile. There is also imperfect Smith et  al. (1)], near ETP (ETP-ALL flow profile but positive overlap between samples with ABD and immature as well as CD5 staining), and not ETP. The five-year EFS/OS for 130 (11.3%) transcriptionally defined ETP-ALL. ETP-ALL cases was 87.0/93.0%, compared to 84.4/91.6% for 195 In summary, there are four potential classification meth- (17.0%) near-ETP-ALL cases, and 86.9/92.0% for 819 (71.6%) ods for early/undifferentiated T-lineage ALL. (1) Immature not-ETP cases, showing that ETP-ALL is not associated with an expression profile cluster, (2) ETP-ALL expression profile, (3) inferior outcome with contemporary treatment regimens (15). ETP-ALL immunophenotype, and (4) ABD. A MEF2C-high er Th e are several possible reasons for these discrepant out - immature cluster seems to mostly overlap with transcription- comes. One obvious reason is that the flow-cytometric definition ally defined ETP-ALL, and a case has been made that this is of this subtype of ALL is not straightforward. In theory, differ - one biological entity (13). The ETP-ALL phenotype depends on ences of how this patient population is defined could account complex criteria, and single features such as the CD5 strongly for differences in outcome. For example, 69% of patients in the influence classification, limiting sensitivity and specificity of UKALL 2003 ETP-ALL cohort, which reported good outcomes the flow panel. ABD seems to only partially overlap with the for ETP-ALL, were classified as “probable ETP-ALL” (with CD5 other subgroups based on the data reported in Ref. (10, 13). The staining either positive or unavailable). Only 31% were classi- overlap between the different diagnostic samples in Ref. (13) is fied as “definitive ETP-ALL,” again speaking to the difficulty of shown in Table 1. accurately defining this patient population. Outcomes showed a non-statistically significant trend toward lower EFS (76.7 vs. iMPROve D OUTCOMe S iN e TP-ALL iN 84.6%) and OS (82.4 vs. 90.9%) for the group of ETP-ALL (“prob- able” + “definitive”), but the absolute number of definitive ETP- MORe Re Ce NT STUDie S ALL patients was small (11 patients) and outcomes for this group More recent studies from Europe and the US demonstrated a were not reported separately. However, it should be noted again substantially better survival for patients with immature T-lineage that the recent US data were classified by the reference laboratory ALL, compared to the reports referenced above. The Dutch that initially reported ETP-ALL. In the Dutch study, outcomes group reported non-inferior survival of immature ALL in their were non-inferior to typical T-ALL using gene expression-based study using all four definitions outlined above ( 13). Patrick et al. and immune-phenotypic classification criteria as well as ABD recently reported results for patients treated in England and criteria. Wales. In this series of patients, treated on protocol UKALL 2003 A second possible reason may be that more recent treatment between 2003 and 2011, results for patients with phenotypic ETP- regimens are in fact superior to older regimens. An important ALL were only slightly and non-significantly worse than results aspect of more recent protocols is the implementation of MRD- for cases with typical T-ALL phenotype (14). To address the based, risk-adapted treatment strategies. Analysis of clinical and outcome of different T-ALL subsets with contemporary therapy, outcomes characteristics of ETP-ALL patients treated on the the Children’s Oncology Group (COG) used a single central COG ALL0434 and UKALL 2003 showed that very few ETP-ALL reference laboratory to classify 1144 T-ALL cases enrolled on patients fell into the low-MRD group, similar to what has been the COG AALL0434 clinical trial between 2007 and 2014 as ETP observed by other investigators. Importantly, MRD response at Frontiers in Pediatrics | www.frontiersin.org 2 May 2016 | Volume 4 | Article 49 Bernt et al. PRC2 in ETP-ALL end consolidation on newer trials is not clearly different from KRAS, FLT3, JAK1, JAK3, and SH2B3), developmental genes MRD response observed in other types of T-ALL, suggesting that (including genes encoding the transcription factors GATA3, some of the poor therapeutic response observed in ETP-ALL may RUNX1, ETV6, and IKZF1), and epigenetic modifiers [includ - be related to the specific drugs used during induction chemother - ing inactivating alterations in components of the Polycomb apy. It is tempting to speculate that resistance to corticosteroids in Repressive Complex 2 (PRC2)] (16). The frequencies for altera - particular may be important in the poor response of ETP-ALL to tions individual components among ETP-ALL were as follows: induction chemotherapy. Yet, early identification of these patients SUZ12 11/64 cases, EED 8/64 cases, EZH2 10/64 cases, any PRC2 based on intermediate or high MRD, and increased intensity of alteration 26/64 cases. Of note, three out of the ETP-ALL patients therapy may have offset the higher-risk biology of ETP-ALL, had lesions in two PRC2 member. Fifteen out of a total of 29 PRC2 resulting in the good outcomes in these two trials. In addition, alterations in ETP-ALL were deletions, strongly suggesting they differences in choice of agents may influence results. As pointed are inactivating. Similarly, 6 out of 6 PRC2-alterations found in out by Zuurbier et al., COALL-97 includes high-dose Cytarabine 42 cases of typical T-ALL were deletions. While the EZH2 muta- (13). A subset of patients on COG ALL0434 was randomized to tions found were not all individually functionally characterized receive Nelarabine. The precise impact of including these drugs in this study, based on computer modeling they are thought to be on outcome is presently not known. The outcomes for the differ - inactivating (16). This is in contrast to the activating mutations of ent cohorts are summarized in Table 2. EZH2 described in B-lineage lymphomas (17, 18). The discovery (whole genome sequencing) cohort in this study consisted of a limited number of 12 patients, with targeted sequencing only in Ge Ne TiC He Te ROGe Nei TY AND the larger validation cohort. A larger cohort allowing for a more complete definition of the genomic landscape of ETP-ALL would OUTCOMe be very informative. In multi-variant analysis, mutations in PRC2 A third potential reason lies in the genetic heterogeneity of ETP- members were found to be an independent predictor of poor ALL. The seminal study by Zhang et al. ( 16) identified a very high outcome. It is possible that outcomes differences between ETP- rate of mutations in certain functional “clusters” of genes. These ALL patient cohorts could read out differences in the underlying include signaling-associated genes (particularly Il7R, NRAS, genomic landscape. The mutational landscape for the patients in TABLe 2 | Survival data from published studies in e TP-ALL and ABD-ALL. Reference Cohort Year Age e FS (%) OS (%) years follow Classification (years) up ETP-ALL (pediatric, initial reports) (1) US/SJCRH 2009 0.5–18.9 19 22 10 FC (1) Italy/AEIOP 45 22 2 FC (2) Japan/Tokyo Children’s Cancer Study 2012 1–18 40 ~70 4 FC Group L99-15 (3) China/Shanghai Children’s Medical Center 2012 11 13 3 FC (4) US/Columbia University 2013 7–49 14 86 5 FC ABD (10) US/COG, DFCI 2010 2–17 25 25 PCR (11) Taiwan/TPOG 2012 43 46 10 PCR (12) France/EuroLB02 2012 <20 0 0 3 PCR ETP-ALL (pediatric, recent reports) (13) Holland/DCOG 2014 1.5–17.8 70 5 multiple (15) US/COG 2014 87 93 5 FC (14) UK/UKALL2003 2014 1.0–24.9 77 82 5 FC Adult data d c (5) Germany/GMALL 2012 35 10 FC d c (6) Europe/E2993 ECOG 2013 34 5 GEP f g (7) US/MDACC 2016 13–79 5 FC (46) France/GRAALL 2016 25 31 5 FC The study by Callens et al. describes four cases of lymphoblastic lymphoma with ABD. ABD, absence of biallelic deletion of the gamma/delta T-cell receptor; FC, flow cytometry; GEP, Gene expression profiling. Custom score involving flow cytometry. 62% 1–9 years, 38% >10 years. Not available. Adult patients, exact age range not available. Median age 30.5. Median OS = 17 months. Median EFS = 20 months. Frontiers in Pediatrics | www.frontiersin.org 3 May 2016 | Volume 4 | Article 49 Bernt et al. PRC2 in ETP-ALL the study by Wood et al. (15) is currently being investigated (19). required for successful establishment of the model]. Cells were Preliminary analyses of patients in the UKALL2003 cohort sug- then differentiated toward a T-cell fate on stromal cells provid - gest ETP-ALL patients in this cohort showed some of the genetic ing a Notch-signal, and injected into lethally irradiated mice. We characteristics of ETP-ALL (less frequent NOTCH1/FBXW7 found that 100% of mice developed leukemia/lymphoma that mutations, no LMO1/2 or TLX1 mutations, and increased rate recapitulated some features of human ETP-ALL (e.g., lack of CD4 of KMT2A translocations). However, alterations of PRC2 com- and CD8 expression in most samples, low expression of CD5, ponents were not reported for this cohort. It is also noteworthy and co-expression of myeloid markers). Inactivation of Ezh2 that in adult ETP-ALL, in contrast to pediatric cases, recurrent or Eed in this model significantly shortened leukemia latency. DNMT3A alterations were found in 16% (20) and 7.5% (6). e fin Th dings from our model are summarized in Figure  1. We This is in keeping with low proportions of DNMT3A mutations noted that Ezh2 inactivation is associated with the expression of in pediatric acute leukemia (21, 22) and with high proportion a stem cell gene expression program. This includes HoxA9, and of DNMT3A alterations in age-related clonal hematopoiesis of genes correlated with HOXA9 in human leukemia. Interestingly, indeterminate potential (23). three recent reports suggest a prognostic role for HOXA9 in ETP-ALL: a subset of T-ALL patients enrolled in AALL00434 was analyzed, and the authors concluded that patients with the A MURiNe MODe L TO STUDY THe combination of MLL-rearrangement, high HOXA9 expression, ROLe  OF PRC2 iN e TP-ALL and ETP-ALL (by expression profiling) had a significantly worse outcome (44). Similarly, T-ALL patients with the CALM-AF10 The association between survival and PRC2 alterations in the study by Zhang et al. (16) was noteworthy. PRC2 is a multipro- fusion, which is associated with high HOXA expression (45), had a worse prognosis if blasts showed an immature phenotype (46). tein complex with important roles in cancer and development. PRC2 consists of the core components EZH2, EED, and SUZ12 In a French cohort, the combination of high HOXA expression and ETP-ALL immunophenotype showed significantly inferior (24). PRC2 di- and tri-methylates histone 3 on lysine 27, a genetic modification associated with low or no transcriptional outcomes. In this study, a number of the high HOXA expressors were explained by chromosomal translocations, but many were output. Importantly, high-level expression of EZH2 has been not. The PRC2 mutational status in this study was not investi - linked to poor outcomes in solid tumors, but recent reports have cast doubt on H3K27me3 as the culprit (25, 26). PRC2- gated (47). These data provide a good rationale for paying special attention to the combination of ETP-ALL and HOXA expression independent roles for EZH2 have been reported, illustrating the complexity of PRC2 biology (25, 27, 28). Wild-type EZH2 and in future studies. Our murine study also identified deregulation of growth naturally occurring hyperactive mutants have been found to be important in the development of Germinal center B-lineage and survival signaling as a direct consequence of loss or PRC2 components. Similarly to what was reported by De Raedt et  al. lymphomas, and EZH2 is a promising target for small molecule inhibition in this disease (17, 18). An epistatic relationship and others in NF1-associated sarcomas (36), we found that in a malignancy with activated RAS signaling, a RAS-downstream between PRC2 and the SWI/SNF chromatin remodeling com- plex (also known as BAF complex) has been reported (29, 30). signature is further enriched by compromise in PRC2 function. In addition, we found deregulated expression of cytokines and Given that approximately 20% of human cancer has alterations in SWI/SNF (31, 32), there is growing interest in EZH2 as a their receptors in response to Ezh2-inactivation. Il6ra is a known PRC2 silencing target, and its derepression in our model was therapeutic target (29, 30, 32). Despite these bona-fide oncogene credentials, recurrent accompanied by exaggerated STAT3Y705 phosphorylation in response to IL6. The clinically approved JAK2 inhibitor ruxoli - inactivating alterations of PRC2 genes have been described in ETP-ALL (16) and other hematologic and non-hematologic tinib inhibited growth in our model in vitro. Importantly, David Teachy’s group recently demonstrated pre-clinical efficacy of rux - malignancies (33–41), creating somewhat of a paradox: PRC2 can have oncogenic and tumor-suppressive roles in cancer depending olitinib in human ETP-ALL samples (48). Our data suggest that even samples without obvious activating mutations in the JAK/ on cellular context. How precisely EZH2/PRC2 exert their onco- genic and tumor suppressor roles mechanistically is incompletely STAT signaling pathway may respond to JAK/STAT small mol- ecule inhibitors. This was indeed the case for two patient samples understood. A better understanding of the molecular details would likely aid in the development of targeted therapies. without JAK/STAT pathway mutation, but with an inactivating PRC2 component mutation, in the xeongraft study by Maude et al To clarify the functional role of PRC2 in T-ALL, we recently generated a murine model of ETP-ALL that allows for the (48). The incorporation of JAK-inhibitors in ETP-ALL should be evaluated in more detail in pre-clinical trials and possibly also in comparison of leukemias with and without genetic compromise in PRC2 function (42). Since RAS-activation occurs frequently carefully designed clinical studies, with a thoughtful selection of patients most likely to benefit. in ETP-ALL, we overexpressed oncogenic mutant NRAS in immature hematopoietic cells. We used mice with homozygously floxed alleles for the PRC2 components Eed and Ezh2, allowing ADDiTiONAL Ge Ne TiC Le SiONS iN homozygous inactivation of either PRC2-member. The mice also e TP-ALL contained homozygously deleted Cdkn2a alleles, to facilitate leukemogenesis [genetic Cdkn2a inactivation is very com- It is noteworthy that certain genetic lesions are inversely corre- lated with PRC2 alterations. es Th e include mutations in GATA3, mon in T-ALL, is not typical for human ETP-ALL (43), but is Frontiers in Pediatrics | www.frontiersin.org 4 May 2016 | Volume 4 | Article 49 Bernt et al. PRC2 in ETP-ALL FiGURe 1 | Cellular consequences of compromised PRC2 function in e TP-ALL. Inactivation of PRC2 components in ETP-ALL results in cellular loss of global cellular H3K27me3, a chromatin mark associated with silent genes. A subset of these genes show increased transcription. Noteworthy groups of target genes with increased expression in Ras-transformed cells with compromised PRC2 function include transcription factors and epigenetic regulators associated with early hematopoiesis (e.g., HoxA, Gata2, and Bmi1) and growth factors and their receptors (e.g., Il6ra). which also carry a poor prognosis (16). It is tempting to speculate A BROADe R ROLe FOR POLYCOMB that GATA3 and PRC2 collaborate in T-cell development, and Ge Ne S iN TUMOR BiOLOGY? that their inactivation in ETP-ALL has some degree of functional redundancy. Furthermore, it is interesting to note that PRC2 The combination of PRC2 inactivation and oncogenic RAS alterations and FLT3 alterations are mutually exclusive in the pathway activation now has been observed in several contexts: study by Zhang et  al. (16). More experimental work and larger NF1-associated malignant peripheral nerve sheath tumors studies that interrogate these suggestive molecular findings and (MPNST) (36–38), juvenile myelomonocytic leukemia correlate molecular alterations with survival data seem warranted. (JMML) (39, 40), megakaryocytic AML of Down syndrome Based on the data available today, alterations in treatment are (41), and ETP-ALL (16). In addition, mutations in genes other not warranted for children with an ETP-ALL immunophenotype than EED, SUZ12, and EZH2 with a documented effect on detected at the time of initial diagnosis (and data in adults may canonical PRC2 function are inactivated in cancer (including be too sparse to draw definitive conclusions). However, there is a hematologic neoplasms). This includes, e.g., Jarid2 (49– 51) higher rate of poor early response, and overt induction failure in and Asxl1 (52). Finally, genes involved in Polycomb Repressive ETP-ALL, and most relapses that occur in ETP-ALL occur early Complex 1 have shown tumor suppressor function in model in therapy. Thus, there is a subset of ETP-ALL patient (and/or systems (53, 54). In summary, these findings suggest com - those with an immature immunophenotype) who relapse quickly monalities between the roles of different Polycomb genes in with extremely refractory disease, or never go into remission in cancer that may potentially be targetable. As an example, the the first place. This raises the following pressing questions for chromatin-binding protein BRD4 has been suggested as such the field: a therapeutic target (36). The careful study of patient material and genetic models holds great promise in this regard. We 1. Is flow cytometry plus early treatment response the best strat - are optimistic that a deeper understanding of the molecular egy to identify high-risk patients, or is there a less equivocal mechanisms underlying this collaboration will result in marker (ETP-transcriptional signature? PRC2 mutation? successful future clinical trials and ultimately in improved Kinase signature?) that can identify the group of patients with outcomes for patients. ETP-ALL and a poor outcome? 2. Assuming that ultra-high-risk patients can be identified at AUTHOR CONTRiBUTiONS diagnosis or very early in therapy can advances in understand- ing the biology of the disease pave the way toward targeted All authors listed, have made substantial, direct, and intellectual and more effective therapies? contribution to the work and approved it for publication. Frontiers in Pediatrics | www.frontiersin.org 5 May 2016 | Volume 4 | Article 49 Bernt et al. PRC2 in ETP-ALL lymphoid transformation. Cancer Cell (2013) 23:677–92. doi:10.1016/j. Re Fe Re NCe S ccr.2013.04.011 18. Caganova M, Carrisi C, Varano G, Mainoldi F, Zanardi F, Germain PL, 1. Coustan-Smith E, Mullighan CG, Onciu M, Behm FG, Raimondi SC, Pei et  al. Germinal center dysregulation by histone methyltransferase EZH2 D, et  al. Early T-cell precursor leukaemia: a subtype of very high-risk acute promotes lymphomagenesis. J Clin Invest (2013) 123:5009–22. doi:10.1172/ lymphoblastic leukaemia. Lancet Oncol (2009) 10:147–56. doi:10.1016/ JCI70626 S1470-2045(08)70314-0 19. Liu Y, Easton J, Shao Y, Wilkinson M, Edmonson M, Ma X, et al. The genomic 2. Inukai T, Kiyokawa N, Campana D, Coustan-Smith E, Kikuchi A, Kobayashi landscape of childhood T-lineage acute lymphoblastic leukemia (abstract). M, et  al. Clinical significance of early T-cell precursor acute lymphoblastic Blood (2015) 126:691. doi:10.1172/JCI70626 leukaemia: results of the Tokyo Children’s Cancer Study Group Study L99-15. 20. Neumann M, Heesch S, Schlee C, Schwartz S, Gokbuget N, Hoelzer D, Br J Haematol (2012) 156:358–65. doi:10.1111/j.1365-2141.2011.08955.x et  al. Whole-exome sequencing in adult ETP-ALL reveals a high rate of 3. Ma M, Wang X, Tang J, Xue H, Chen J, Pan C, et  al. Early T-cell precursor DNMT3A mutations. Blood (2013) 121:4749–52. doi:10.1182/blood-2012-11- leukemia: a subtype of high risk childhood acute lymphoblastic leukemia. Front Med (2012) 6:416–20. doi:10.1007/s11684-012-0224-4 21. Li W, Gao C, Cui L, Liu S, Zhao X, Zhang R, et al. DNMT3A mutations and 4. Allen A, Sireci A, Colovai A, Pinkney K, Sulis M, Bhagat G, et al. Early T-cell prognostic significance in childhood acute lymphoblastic leukemia. Leuk precursor leukemia/lymphoma in adults and children. Leuk Res (2013) Lymphoma (2015) 56:1066–71. doi:10.3109/10428194.2014.947607 37:1027–34. doi:10.1016/j.leukres.2013.06.010 22. Ho PA, Kutny MA, Alonzo TA, Gerbing RB, Joaquin J, Raimondi SC, et  al. 5. Neumann M, Heesch S, Gokbuget N, Schwartz S, Schlee C, Benlasfer O, et al. Leukemic mutations in the methylation-associated genes DNMT3A and Clinical and molecular characterization of early T-cell precursor leukemia: a IDH2 are rare events in pediatric AML: a report from the Children’s Oncology high-risk subgroup in adult T-ALL with a high frequency of FLT3 mutations. Group. Pediatr Blood Cancer (2011) 57:204–9. doi:10.1002/pbc.23179 Blood Cancer J (2012) 2:e55. doi:10.1038/bcj.2011.49 23. Jaiswal S, Fontanillas P, Flannick J, Manning A, Grauman PV, Mar BG, et al. 6. Van Vlierberghe P, Ambesi-Impiombato A, De Keersmaecker K, Hadler M, Age-related clonal hematopoiesis associated with adverse outcomes. N Engl Paietta E, Tallman MS, et al. Prognostic relevance of integrated genetic pro- J Med (2014) 371:2488–98. doi:10.1056/NEJMoa1408617 filing in adult T-cell acute lymphoblastic leukemia. Blood (2013) 122:74–82. 24. Di Croce L, Helin K. Transcriptional regulation by Polycomb group proteins. doi:10.1182/blood-2013-03-491092 Nat Struct Mol Biol (2013) 20:1147–55. doi:10.1038/nsmb.2669 7. Jain N, Lamb AE, O’Brien S, Ravandi F, Konopleva M, Jabbour E, et al. Early 25. Xu K, Wu ZJ, Groner AC, He HH, Cai C, Lis RT, et al. EZH2 oncogenic activity T-cell precursor acute lymphoblastic leukemia/lymphoma (ETP-ALL/LBL) in castration-resistant prostate cancer cells is Polycomb-independent. Science in adolescents and adults: a high-risk subtype. Blood (2016) 127:1863–9. (2012) 338:1465–9. doi:10.1126/science.1227604 doi:10.1182/blood-2015-08-661702 26. Wassef M, Rodilla V, Teissandier A, Zeitouni B, Gruel N, Sadacca B, et  al. 8. Ferrando AA, Neuberg DS, Staunton J, Loh ML, Huard C, Raimondi SC, Impaired PRC2 activity promotes transcriptional instability and favors breast et  al. Gene expression signatures define novel oncogenic pathways in T cell tumorigenesis. Genes Dev (2015) 29:2547–62. doi:10.1101/gad.269522.115 acute lymphoblastic leukemia. Cancer Cell (2002) 1:75–87. doi:10.1016/ 27. Shi B, Liang J, Yang X, Wang Y, Zhao Y, Wu H, et al. Integration of estrogen and S1535-6108(02)00018-1 Wnt signaling circuits by the polycomb group protein EZH2 in breast cancer 9. Homminga I, Pieters R, Langerak AW, de Rooi JJ, Stubbs A, Verstegen M, cells. Mol Cell Biol (2007) 27:5105–19. doi:10.1128/MCB.00162-07 et al. Integrated transcript and genome analyses reveal NKX2-1 and MEF2C 28. Lee ST, Li Z, Wu Z, Aau M, Guan P, Karuturi RK, et  al. Context-specific as potential oncogenes in T cell acute lymphoblastic leukemia. Cancer Cell regulation of NF-kappaB target gene expression by EZH2 in breast cancers. (2011) 19:484–97. doi:10.1016/j.ccr.2011.02.008 Mol Cell (2011) 43:798–810. doi:10.1016/j.molcel.2011.08.011 10. Gutierrez A, Dahlberg SE, Neuberg DS, Zhang J, Grebliunaite R, Sanda T, 29. Kim KH, Kim W, Howard TP, Vazquez F, Tsherniak A, Wu JN, et  al. SWI/ et  al. Absence of biallelic TCRgamma deletion predicts early treatment SNF-mutant cancers depend on catalytic and non-catalytic activity of EZH2. failure in pediatric T-cell acute lymphoblastic leukemia. J Clin Oncol (2010) Nat Med (2015) 21:1491–6. doi:10.1038/nm.3968 28:3816–23. doi:10.1200/JCO.2010.28.3390 30. Wilson BG, Wang X, Shen X, McKenna ES, Lemieux ME, Cho YJ, et  al. 11. Yang YL, Hsiao CC, Chen HY, Lin KH, Jou ST, Chen JS, et al. Absence of bial- Epigenetic antagonism between polycomb and SWI/SNF complexes during lelic TCRgamma deletion predicts induction failure and poorer outcomes in oncogenic transformation. Cancer Cell (2010) 18:316–28. doi:10.1016/j. childhood T-cell acute lymphoblastic leukemia. Pediatr Blood Cancer (2012) ccr.2010.09.006 58:846–51. doi:10.1002/pbc.24021 31. Kadoch C, Hargreaves DC, Hodges C, Elias L, Ho L, Ranish J, et al. Proteomic 12. Callens C, Baleydier F, Lengline E, Ben Abdelali R, Petit A, Villarese P, et al. and bioinformatic analysis of mammalian SWI/SNF complexes identifies Clinical impact of NOTCH1 and/or FBXW7 mutations, FLASH deletion, and extensive roles in human malignancy. Nat Genet (2013) 45:592–601. TCR status in pediatric T-cell lymphoblastic lymphoma. J Clin Oncol (2012) doi:10.1038/ng.2628 30:1966–73. doi:10.1200/JCO.2011.39.7661 32. Knutson SK, Warholic NM, Wigle TJ, Klaus CR, Allain CJ, Raimondi A, et al. 13. Zuurbier L, Gutierrez A, Mullighan CG, Cante-Barrett K, Gevaert AO, de Rooi J, et al. Immature MEF2C-dysregulated T-cell leukemia patients have an early Durable tumor regression in genetically altered malignant rhabdoid tumors by inhibition of methyltransferase EZH2. Proc Natl Acad Sci U S A (2013) T-cell precursor acute lymphoblastic leukemia gene signature and typically have non-rearranged T-cell receptors. Haematologica (2014) 99:94–102. 110:7922–7. doi:10.1073/pnas.1303800110 33. Nikoloski G, Langemeijer SM, Kuiper RP, Knops R, Massop M, Tonnissen doi:10.3324/haematol.2013.090233 14. Patrick K, Wade R, Goulden N, Mitchell C, Moorman AV, Rowntree C, et al. ER, et  al. Somatic mutations of the histone methyltransferase gene EZH2 in myelodysplastic syndromes. Nat Genet (2010) 42:665–7. doi:10.1038/ Outcome for children and young people with Early T-cell precursor acute lymphoblastic leukaemia treated on a contemporary protocol, UKALL 2003. ng.620 34. Ernst T, Chase AJ, Score J, Hidalgo-Curtis CE, Bryant C, Jones AV, et  al. Br J Haematol (2014) 166:421–4. doi:10.1111/bjh.12882 15. Wood BL, Winter SS, Dunsmore KP, Devidas M, Chen S, Asselin B, et  al. Inactivating mutations of the histone methyltransferase gene EZH2 in myeloid disorders. Nat Genet (2010) 42:722–6. doi:10.1038/ng.621 T-Lymphoblastic Leukemia (T-ALL) Shows Excellent Outcome, Lack of Significance of the Early Thymic Precursor (ETP) Immunophenotype, and 35. Guglielmelli P, Biamonte F, Score J, Hidalgo-Curtis C, Cervantes F, Maffioli M, et al. EZH2 mutational status predicts poor survival in myelob fi rosis. Blood Validation of the Prognostic Value of End-Induction Minimal Residual Disease (MRD) in Children’s Oncology Group (COG) Study AALL0434 (2011) 118:5227–34. doi:10.1182/blood-2011-06-363424 (abstract). Blood (2014) 124:1. 36. De Raedt T, Beert E, Pasmant E, Luscan A, Brems H, Ortonne N, et al. PRC2 16. Zhang J, Ding L, Holmfeldt L, Wu G, Heatley SL, Payne-Turner D, et al. The loss amplifies Ras-driven transcription and confers sensitivity to BRD4-based genetic basis of early T-cell precursor acute lymphoblastic leukaemia. Nature therapies. Nature (2014) 514:247–51. doi:10.1038/nature13561 (2012) 481:157–63. doi:10.1038/nature10725 37. Zhang M, Wang Y, Jones S, Sausen M, McMahon K, Sharma R, et al. Somatic 17. Beguelin W, Popovic R, Teater M, Jiang Y, Bunting KL, Rosen M, et al. EZH2 is mutations of SUZ12 in malignant peripheral nerve sheath tumors. Nat Genet required for germinal center formation and somatic EZH2 mutations promote (2014) 46:1170–2. doi:10.1038/ng.3116 Frontiers in Pediatrics | www.frontiersin.org 6 May 2016 | Volume 4 | Article 49 Bernt et al. PRC2 in ETP-ALL 38. Lee W, Teckie S, Wiesner T, Ran L, Prieto Granada CN, Lin M, et al. PRC2 is 48. Maude SL, Dolai S, Delgado-Martin C, Vincent T, Robbins A, Selvanathan A, recurrently inactivated through EED or SUZ12 loss in malignant peripheral et  al. Efficacy of JAK/STAT pathway inhibition in murine xenograft models nerve sheath tumors. Nat Genet (2014) 46:1227–32. doi:10.1038/ng.3095 of early T-cell precursor (ETP) acute lymphoblastic leukemia. Blood (2015) 39. Caye A, Strullu M, Guidez F, Cassinat B, Gazal S, Fenneteau O, et al. Juvenile 125:1759–67. doi:10.1182/blood-2014-06-580480 myelomonocytic leukemia displays mutations in components of the RAS 49. Shen X, Kim W, Fujiwara Y, Simon MD, Liu Y, Mysliwiec MR, et al. Jumonji pathway and the PRC2 network. Nat Genet (2015) 47:1334–40. doi:10.1038/ modulates polycomb activity and self-renewal versus differentiation of stem ng.3420 cells. Cell (2009) 139:1303–14. doi:10.1016/j.cell.2009.12.003 40. Stieglitz E, Taylor-Weiner AN, Chang TY, Gelston LC, Wang YD, Mazor T, 50. Puda A, Milosevic JD, Berg T, Klampfl T, Harutyunyan AS, Gisslinger B, et al. et al. The genomic landscape of juvenile myelomonocytic leukemia. Nat Genet Frequent deletions of JARID2 in leukemic transformation of chronic myeloid (2015) 47:1326–33. doi:10.1038/ng.3400 malignancies. Am J Hematol (2012) 87:245–50. doi:10.1002/ajh.22257 41. Yoshida K, Toki T, Okuno Y, Kanezaki R, Shiraishi Y, Sato-Otsubo A, et  al. 51. Simon C, Chagraoui J, Krosl J, Gendron P, Wilhelm B, Lemieux S, et al. A key e l Th andscape of somatic mutations in Down syndrome-related myeloid role for EZH2 and associated genes in mouse and human adult T-cell acute disorders. Nat Genet (2013) 45:1293–9. doi:10.1038/ng.2759 leukemia. Genes Dev (2012) 26:651–6. doi:10.1101/gad.186411.111 42. Danis E, Yamauchi T, Echanique K, Zhang X, Haladyna JN, Riedel SS, et al. 52. Abdel-Wahab O, Adli M, LaFave LM, Gao J, Hricik T, Shih AH, et  al. Ezh2 controls an early hematopoietic program and growth and survival sig- ASXL1 mutations promote myeloid transformation through loss of PRC2- naling in early T cell precursor acute lymphoblastic leukemia. Cell Rep (2016) mediated gene repression. Cancer Cell (2012) 22:180–93. doi:10.1016/j. 14:1953–65. doi:10.1016/j.celrep.2016.01.064 ccr.2012.06.032 43. Haydu JE, Ferrando AA. Early T-cell precursor acute lymphoblastic 53. Ross K, Sedello AK, Todd GP, Paszkowski-Rogacz M, Bird AW, Ding L, et al. leukaemia. Curr Opin Hematol (2013) 20:369–73. doi:10.1097/MOH. Polycomb group ring finger 1 cooperates with Runx1 in regulating differen - 0b013e3283623c61 tiation and self-renewal of hematopoietic cells. Blood (2012) 119:4152–61. 44. Matlawska-Wasowska K, Kang H, Devidas M, Wen J, Harvey RC, Nickl CK, doi:10.1182/blood-2011-09-382390 et al. MLL rearrangements impact outcome in HOXA-deregulated T-lineage 54. Oguro H, Yuan J, Tanaka S, Miyagi S, Mochizuki-Kashio M, Ichikawa H, et al. acute lymphoblastic leukemia: a children’s oncology group study. Leukemia Lethal myelob fi rosis induced by Bmi1-deficient hematopoietic cells unveils a (2016). doi:10.1038/leu.2016.60 tumor suppressor function of the polycomb group genes. J Exp Med (2012) 45. Dik WA, Brahim W, Braun C, Asnafi V, Dastugue N, Bernard OA, et  al. 209:445–54. doi:10.1084/jem.20111709 CALM-AF10+ T-ALL expression profiles are characterized by overexpression of HOXA and BMI1 oncogenes. Leukemia (2005) 19:1948–57. doi:10.1038/ Conflict of Interest Statement: The authors declare that the research was con - sj.leu.2403891 ducted in the absence of any commercial or financial relationships that could be 46. Ben Abdelali R, Asnafi V, Petit A, Micol JB, Callens C, Villarese P, et al. The construed as a potential conflict of interest. prognosis of CALM-AF10-positive adult T-cell acute lymphoblastic leukemias depends on the stage of maturation arrest. Haematologica (2013) 98:1711–7. doi:10.3324/haematol.2013.086082 Copyright © 2016 Bernt, Hunger and Neff. i Th s is an open-access article distributed 47. Bond J, Marchand T, Touzart A, Cieslak A, Trinquand A, Sutton L, et al. An under the terms of the Creative Commons Attribution License (CC BY). The use, early thymic precursor phenotype predicts outcome exclusively in HOXA- distribution or reproduction in other forums is permitted, provided the original overexpressing adult T-cell acute lymphoblastic leukemia: a group for author(s) or licensor are credited and that the original publication in this journal research in adult acute lymphoblastic leukemia study. Haematologica (2016). is cited, in accordance with accepted academic practice. No use, distribution or doi:10.3324/haematol.2015.141218 reproduction is permitted which does not comply with these terms. Frontiers in Pediatrics | www.frontiersin.org 7 May 2016 | Volume 4 | Article 49 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Frontiers in Pediatrics Pubmed Central

The Functional Role of PRC2 in Early T-cell Precursor Acute Lymphoblastic Leukemia (ETP-ALL) – Mechanisms and Opportunities

Frontiers in Pediatrics , Volume 4 – May 18, 2016

Loading next page...
 
/lp/pubmed-central/the-functional-role-of-prc2-in-early-t-cell-precursor-acute-7ZN9c5oQw6

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Pubmed Central
Copyright
Copyright © 2016 Bernt, Hunger and Neff.
ISSN
2296-2360
eISSN
2296-2360
DOI
10.3389/fped.2016.00049
Publisher site
See Article on Publisher Site

Abstract

Review published: 18 May 2016 doi: 10.3389/fped.2016.00049 The Functional Role of PRC2 in early T-cell Precursor Acute Lymphoblastic Leukemia (eTP-ALL) – Mechanisms and Opportunities 1 2 1 Kathrin M. Bernt , Stephen P. Hunger and Tobias Neff * Department of Pediatrics, Center for Cancer and Blood Disorders, Children’s Hospital Colorado, University of Colorado Denver, Aurora, CO, USA, Department of Pediatrics, Center for Childhood Cancer Research, The Children’s Hospital of Philadelphia, Philadelphia, PA, USA Early T-Cell precursor acute lymphoblastic leukemia (ETP-ALL) is a relatively newly iden- tified subset of T-lineage ALL. There are conflicting results regarding prognosis, and the genetic basis of this condition is variable. Here, we summarize the current status of the field and discuss the role of mutations in the Polycomb Repressive Complex 2 frequently identified in ETP-ALL patients. Keywords: leukemia, lymphoid, epigenetics, polycomb repressive complex, Hox genes, JAK/STAT signaling pathway Edited by: Brenton Garrett Mar, e TP-ALL, A NOve L SUBTYPe OF ACUTe LYMPHOBLASTiC Dana Farber Cancer Institute, USA Le UKe MiA Reviewed by: Kathryn Grace Roberts, In 2009, Coustan-Smith and colleagues reported a subtype of T-lineage acute lymphoblastic leuke- St. Jude Children’s Research mia (T-ALL) with transcriptional and surface marker profile similarities to early T-cell precursors Hospital, USA Alejandro Gutierrez, (ETP-ALL) (1). Clinical characteristics of ETP-ALL include associations with lower WBC, older Boston Children’s Hospital, USA age, and a very high rate of poor MRD response or even induction failure at the end of induction *Correspondence: chemotherapy. Tobias Neff Initial reports suggested that ETP-ALL has an extraordinarily poor long-term prognosis, with tobias.neff@ucdenver.edu 2- to 10-year event free/overall survival (EFS/OS) rates in the 11–40% range, compared to 84–90% OS for all pediatric T-ALL in a similar time frame (1–3). However, patients in these cohorts were Specialty section: not treated on the most recent treatment protocols [e.g., 1992–2006 for the St. Jude cohort and This article was submitted to 2001–2006 for the AIEOP cohort (1)] and included only a small number of ETP-ALL patients: 17 Pediatric Hematology and in the St Jude, 13 patients in the AEIOP (1), 5 in the Tokyo Children’s Cancer Study Group L99-15 Hematological Malignancies, cohort (2), 12 in the Shanghai Children’s (3), and 7 (3 pediatric and 4 adult) treated at Columbia a section of the journal University (4). Poor outcomes were also reported for adult patients with immune-phenotypic ETP- Frontiers in Pediatrics ALL or T-ALL with a gene expression profile characteristic of ETP-ALL (5–7). Received: 13 April 2016 Accepted: 02 May 2016 Published: 18 May 2016 e TP-ALL iN Re LATiON TO OTHe R De FiNiTiONS OF Citation: iMMATURe  T-ALL Bernt KM, Hunger SP and Neff T (2016) The Functional Role of PRC2 Early T-cell precursor acute lymphoblastic leukemia is defined based on immunophenotyping in Early T-cell Precursor Acute (CD1a-negative, CD8-negative, weak CD5 expression with less than 75% positive blasts, and expres- Lymphoblastic Leukemia (ETP- sion of one or more of the following myeloid or stem cell markers on at least 25% of lymphoblasts: ALL) – Mechanisms and CD117, CD34, HLA-DR, CD13, CD33, CD11b, and/or CD65). It is noteworthy that other definitions Opportunities. of very immature T-ALL subtypes have been proposed. In an important study, using array-based Front. Pediatr. 4:49. doi: 10.3389/fped.2016.00049 genome-wide expression profiling, Ferrando et al. found that T-ALL with LYL1 expression clusters Frontiers in Pediatrics | www.frontiersin.org 1 May 2016 | Volume 4 | Article 49 Bernt et al. PRC2 in ETP-ALL TABLe 1 | Overlap between different methods to classify immature separate from other T-ALL subtypes and show immunopheno- T-lineage ALL [data from Ref. (13)]. typically immature features (8). The Meijerink-group identified an immature subgroup of T-ALL with high MEF2C expression, which based on expression profiling clustered separately from other T-ALL cases and was enriched for ETP-ALL cases (9). This same laboratory went on to further characterize the overlap between ETP-ALL defined by gene expression profile, ETP-ALL based on immunophenotyping, MEF2C-high T-ALL, and a third group of immature T-ALL originally described by Gutierrez et al. defined by absence of biallelic TCRgamma deletion (ABD) (10). Poor prognosis for T-ALL and T-lymphoblastic lymphoma with ABD was also found in two other studies (11, 12). Meijerink and colleagues systematically compared the different definitions of immature T-lineage ALL. There is excellent overlap between the immature cluster and ETP-ALL as defined by an ETP-ALL gene expression signature (13). There is less overlap between the immature cluster and ETP-ALL defined by immunophenotyping. There is excellent overlap between clustering and PAM based on a defined ETP-ALL Inclusion of the CD5 marker in flow-cytometry panel reduced signature. Stringent ETP-ALL criteria including CD5 low criterion exclude many samples the number of cases classified as phenotypic ETP-ALL. Omission classified as immature by gene expression profile. Relaxing the CD5 criterion (“near of the CD5 marker improved the number of expression-based ETP”) includes a number of samples that do not display an ETP-ALL expression profile. PAM, prediction analysis of microarrays. ETP-ALL cases that also were immunophenotypically defined as ETP-ALL. However, the less stringent flow criteria also led to the classification as immunophenotypic ETP-ALL of samples [methodology harmonized with the approach used by Coustan- without an ETP-ALL expression profile. There is also imperfect Smith et  al. (1)], near ETP (ETP-ALL flow profile but positive overlap between samples with ABD and immature as well as CD5 staining), and not ETP. The five-year EFS/OS for 130 (11.3%) transcriptionally defined ETP-ALL. ETP-ALL cases was 87.0/93.0%, compared to 84.4/91.6% for 195 In summary, there are four potential classification meth- (17.0%) near-ETP-ALL cases, and 86.9/92.0% for 819 (71.6%) ods for early/undifferentiated T-lineage ALL. (1) Immature not-ETP cases, showing that ETP-ALL is not associated with an expression profile cluster, (2) ETP-ALL expression profile, (3) inferior outcome with contemporary treatment regimens (15). ETP-ALL immunophenotype, and (4) ABD. A MEF2C-high er Th e are several possible reasons for these discrepant out - immature cluster seems to mostly overlap with transcription- comes. One obvious reason is that the flow-cytometric definition ally defined ETP-ALL, and a case has been made that this is of this subtype of ALL is not straightforward. In theory, differ - one biological entity (13). The ETP-ALL phenotype depends on ences of how this patient population is defined could account complex criteria, and single features such as the CD5 strongly for differences in outcome. For example, 69% of patients in the influence classification, limiting sensitivity and specificity of UKALL 2003 ETP-ALL cohort, which reported good outcomes the flow panel. ABD seems to only partially overlap with the for ETP-ALL, were classified as “probable ETP-ALL” (with CD5 other subgroups based on the data reported in Ref. (10, 13). The staining either positive or unavailable). Only 31% were classi- overlap between the different diagnostic samples in Ref. (13) is fied as “definitive ETP-ALL,” again speaking to the difficulty of shown in Table 1. accurately defining this patient population. Outcomes showed a non-statistically significant trend toward lower EFS (76.7 vs. iMPROve D OUTCOMe S iN e TP-ALL iN 84.6%) and OS (82.4 vs. 90.9%) for the group of ETP-ALL (“prob- able” + “definitive”), but the absolute number of definitive ETP- MORe Re Ce NT STUDie S ALL patients was small (11 patients) and outcomes for this group More recent studies from Europe and the US demonstrated a were not reported separately. However, it should be noted again substantially better survival for patients with immature T-lineage that the recent US data were classified by the reference laboratory ALL, compared to the reports referenced above. The Dutch that initially reported ETP-ALL. In the Dutch study, outcomes group reported non-inferior survival of immature ALL in their were non-inferior to typical T-ALL using gene expression-based study using all four definitions outlined above ( 13). Patrick et al. and immune-phenotypic classification criteria as well as ABD recently reported results for patients treated in England and criteria. Wales. In this series of patients, treated on protocol UKALL 2003 A second possible reason may be that more recent treatment between 2003 and 2011, results for patients with phenotypic ETP- regimens are in fact superior to older regimens. An important ALL were only slightly and non-significantly worse than results aspect of more recent protocols is the implementation of MRD- for cases with typical T-ALL phenotype (14). To address the based, risk-adapted treatment strategies. Analysis of clinical and outcome of different T-ALL subsets with contemporary therapy, outcomes characteristics of ETP-ALL patients treated on the the Children’s Oncology Group (COG) used a single central COG ALL0434 and UKALL 2003 showed that very few ETP-ALL reference laboratory to classify 1144 T-ALL cases enrolled on patients fell into the low-MRD group, similar to what has been the COG AALL0434 clinical trial between 2007 and 2014 as ETP observed by other investigators. Importantly, MRD response at Frontiers in Pediatrics | www.frontiersin.org 2 May 2016 | Volume 4 | Article 49 Bernt et al. PRC2 in ETP-ALL end consolidation on newer trials is not clearly different from KRAS, FLT3, JAK1, JAK3, and SH2B3), developmental genes MRD response observed in other types of T-ALL, suggesting that (including genes encoding the transcription factors GATA3, some of the poor therapeutic response observed in ETP-ALL may RUNX1, ETV6, and IKZF1), and epigenetic modifiers [includ - be related to the specific drugs used during induction chemother - ing inactivating alterations in components of the Polycomb apy. It is tempting to speculate that resistance to corticosteroids in Repressive Complex 2 (PRC2)] (16). The frequencies for altera - particular may be important in the poor response of ETP-ALL to tions individual components among ETP-ALL were as follows: induction chemotherapy. Yet, early identification of these patients SUZ12 11/64 cases, EED 8/64 cases, EZH2 10/64 cases, any PRC2 based on intermediate or high MRD, and increased intensity of alteration 26/64 cases. Of note, three out of the ETP-ALL patients therapy may have offset the higher-risk biology of ETP-ALL, had lesions in two PRC2 member. Fifteen out of a total of 29 PRC2 resulting in the good outcomes in these two trials. In addition, alterations in ETP-ALL were deletions, strongly suggesting they differences in choice of agents may influence results. As pointed are inactivating. Similarly, 6 out of 6 PRC2-alterations found in out by Zuurbier et al., COALL-97 includes high-dose Cytarabine 42 cases of typical T-ALL were deletions. While the EZH2 muta- (13). A subset of patients on COG ALL0434 was randomized to tions found were not all individually functionally characterized receive Nelarabine. The precise impact of including these drugs in this study, based on computer modeling they are thought to be on outcome is presently not known. The outcomes for the differ - inactivating (16). This is in contrast to the activating mutations of ent cohorts are summarized in Table 2. EZH2 described in B-lineage lymphomas (17, 18). The discovery (whole genome sequencing) cohort in this study consisted of a limited number of 12 patients, with targeted sequencing only in Ge Ne TiC He Te ROGe Nei TY AND the larger validation cohort. A larger cohort allowing for a more complete definition of the genomic landscape of ETP-ALL would OUTCOMe be very informative. In multi-variant analysis, mutations in PRC2 A third potential reason lies in the genetic heterogeneity of ETP- members were found to be an independent predictor of poor ALL. The seminal study by Zhang et al. ( 16) identified a very high outcome. It is possible that outcomes differences between ETP- rate of mutations in certain functional “clusters” of genes. These ALL patient cohorts could read out differences in the underlying include signaling-associated genes (particularly Il7R, NRAS, genomic landscape. The mutational landscape for the patients in TABLe 2 | Survival data from published studies in e TP-ALL and ABD-ALL. Reference Cohort Year Age e FS (%) OS (%) years follow Classification (years) up ETP-ALL (pediatric, initial reports) (1) US/SJCRH 2009 0.5–18.9 19 22 10 FC (1) Italy/AEIOP 45 22 2 FC (2) Japan/Tokyo Children’s Cancer Study 2012 1–18 40 ~70 4 FC Group L99-15 (3) China/Shanghai Children’s Medical Center 2012 11 13 3 FC (4) US/Columbia University 2013 7–49 14 86 5 FC ABD (10) US/COG, DFCI 2010 2–17 25 25 PCR (11) Taiwan/TPOG 2012 43 46 10 PCR (12) France/EuroLB02 2012 <20 0 0 3 PCR ETP-ALL (pediatric, recent reports) (13) Holland/DCOG 2014 1.5–17.8 70 5 multiple (15) US/COG 2014 87 93 5 FC (14) UK/UKALL2003 2014 1.0–24.9 77 82 5 FC Adult data d c (5) Germany/GMALL 2012 35 10 FC d c (6) Europe/E2993 ECOG 2013 34 5 GEP f g (7) US/MDACC 2016 13–79 5 FC (46) France/GRAALL 2016 25 31 5 FC The study by Callens et al. describes four cases of lymphoblastic lymphoma with ABD. ABD, absence of biallelic deletion of the gamma/delta T-cell receptor; FC, flow cytometry; GEP, Gene expression profiling. Custom score involving flow cytometry. 62% 1–9 years, 38% >10 years. Not available. Adult patients, exact age range not available. Median age 30.5. Median OS = 17 months. Median EFS = 20 months. Frontiers in Pediatrics | www.frontiersin.org 3 May 2016 | Volume 4 | Article 49 Bernt et al. PRC2 in ETP-ALL the study by Wood et al. (15) is currently being investigated (19). required for successful establishment of the model]. Cells were Preliminary analyses of patients in the UKALL2003 cohort sug- then differentiated toward a T-cell fate on stromal cells provid - gest ETP-ALL patients in this cohort showed some of the genetic ing a Notch-signal, and injected into lethally irradiated mice. We characteristics of ETP-ALL (less frequent NOTCH1/FBXW7 found that 100% of mice developed leukemia/lymphoma that mutations, no LMO1/2 or TLX1 mutations, and increased rate recapitulated some features of human ETP-ALL (e.g., lack of CD4 of KMT2A translocations). However, alterations of PRC2 com- and CD8 expression in most samples, low expression of CD5, ponents were not reported for this cohort. It is also noteworthy and co-expression of myeloid markers). Inactivation of Ezh2 that in adult ETP-ALL, in contrast to pediatric cases, recurrent or Eed in this model significantly shortened leukemia latency. DNMT3A alterations were found in 16% (20) and 7.5% (6). e fin Th dings from our model are summarized in Figure  1. We This is in keeping with low proportions of DNMT3A mutations noted that Ezh2 inactivation is associated with the expression of in pediatric acute leukemia (21, 22) and with high proportion a stem cell gene expression program. This includes HoxA9, and of DNMT3A alterations in age-related clonal hematopoiesis of genes correlated with HOXA9 in human leukemia. Interestingly, indeterminate potential (23). three recent reports suggest a prognostic role for HOXA9 in ETP-ALL: a subset of T-ALL patients enrolled in AALL00434 was analyzed, and the authors concluded that patients with the A MURiNe MODe L TO STUDY THe combination of MLL-rearrangement, high HOXA9 expression, ROLe  OF PRC2 iN e TP-ALL and ETP-ALL (by expression profiling) had a significantly worse outcome (44). Similarly, T-ALL patients with the CALM-AF10 The association between survival and PRC2 alterations in the study by Zhang et al. (16) was noteworthy. PRC2 is a multipro- fusion, which is associated with high HOXA expression (45), had a worse prognosis if blasts showed an immature phenotype (46). tein complex with important roles in cancer and development. PRC2 consists of the core components EZH2, EED, and SUZ12 In a French cohort, the combination of high HOXA expression and ETP-ALL immunophenotype showed significantly inferior (24). PRC2 di- and tri-methylates histone 3 on lysine 27, a genetic modification associated with low or no transcriptional outcomes. In this study, a number of the high HOXA expressors were explained by chromosomal translocations, but many were output. Importantly, high-level expression of EZH2 has been not. The PRC2 mutational status in this study was not investi - linked to poor outcomes in solid tumors, but recent reports have cast doubt on H3K27me3 as the culprit (25, 26). PRC2- gated (47). These data provide a good rationale for paying special attention to the combination of ETP-ALL and HOXA expression independent roles for EZH2 have been reported, illustrating the complexity of PRC2 biology (25, 27, 28). Wild-type EZH2 and in future studies. Our murine study also identified deregulation of growth naturally occurring hyperactive mutants have been found to be important in the development of Germinal center B-lineage and survival signaling as a direct consequence of loss or PRC2 components. Similarly to what was reported by De Raedt et  al. lymphomas, and EZH2 is a promising target for small molecule inhibition in this disease (17, 18). An epistatic relationship and others in NF1-associated sarcomas (36), we found that in a malignancy with activated RAS signaling, a RAS-downstream between PRC2 and the SWI/SNF chromatin remodeling com- plex (also known as BAF complex) has been reported (29, 30). signature is further enriched by compromise in PRC2 function. In addition, we found deregulated expression of cytokines and Given that approximately 20% of human cancer has alterations in SWI/SNF (31, 32), there is growing interest in EZH2 as a their receptors in response to Ezh2-inactivation. Il6ra is a known PRC2 silencing target, and its derepression in our model was therapeutic target (29, 30, 32). Despite these bona-fide oncogene credentials, recurrent accompanied by exaggerated STAT3Y705 phosphorylation in response to IL6. The clinically approved JAK2 inhibitor ruxoli - inactivating alterations of PRC2 genes have been described in ETP-ALL (16) and other hematologic and non-hematologic tinib inhibited growth in our model in vitro. Importantly, David Teachy’s group recently demonstrated pre-clinical efficacy of rux - malignancies (33–41), creating somewhat of a paradox: PRC2 can have oncogenic and tumor-suppressive roles in cancer depending olitinib in human ETP-ALL samples (48). Our data suggest that even samples without obvious activating mutations in the JAK/ on cellular context. How precisely EZH2/PRC2 exert their onco- genic and tumor suppressor roles mechanistically is incompletely STAT signaling pathway may respond to JAK/STAT small mol- ecule inhibitors. This was indeed the case for two patient samples understood. A better understanding of the molecular details would likely aid in the development of targeted therapies. without JAK/STAT pathway mutation, but with an inactivating PRC2 component mutation, in the xeongraft study by Maude et al To clarify the functional role of PRC2 in T-ALL, we recently generated a murine model of ETP-ALL that allows for the (48). The incorporation of JAK-inhibitors in ETP-ALL should be evaluated in more detail in pre-clinical trials and possibly also in comparison of leukemias with and without genetic compromise in PRC2 function (42). Since RAS-activation occurs frequently carefully designed clinical studies, with a thoughtful selection of patients most likely to benefit. in ETP-ALL, we overexpressed oncogenic mutant NRAS in immature hematopoietic cells. We used mice with homozygously floxed alleles for the PRC2 components Eed and Ezh2, allowing ADDiTiONAL Ge Ne TiC Le SiONS iN homozygous inactivation of either PRC2-member. The mice also e TP-ALL contained homozygously deleted Cdkn2a alleles, to facilitate leukemogenesis [genetic Cdkn2a inactivation is very com- It is noteworthy that certain genetic lesions are inversely corre- lated with PRC2 alterations. es Th e include mutations in GATA3, mon in T-ALL, is not typical for human ETP-ALL (43), but is Frontiers in Pediatrics | www.frontiersin.org 4 May 2016 | Volume 4 | Article 49 Bernt et al. PRC2 in ETP-ALL FiGURe 1 | Cellular consequences of compromised PRC2 function in e TP-ALL. Inactivation of PRC2 components in ETP-ALL results in cellular loss of global cellular H3K27me3, a chromatin mark associated with silent genes. A subset of these genes show increased transcription. Noteworthy groups of target genes with increased expression in Ras-transformed cells with compromised PRC2 function include transcription factors and epigenetic regulators associated with early hematopoiesis (e.g., HoxA, Gata2, and Bmi1) and growth factors and their receptors (e.g., Il6ra). which also carry a poor prognosis (16). It is tempting to speculate A BROADe R ROLe FOR POLYCOMB that GATA3 and PRC2 collaborate in T-cell development, and Ge Ne S iN TUMOR BiOLOGY? that their inactivation in ETP-ALL has some degree of functional redundancy. Furthermore, it is interesting to note that PRC2 The combination of PRC2 inactivation and oncogenic RAS alterations and FLT3 alterations are mutually exclusive in the pathway activation now has been observed in several contexts: study by Zhang et  al. (16). More experimental work and larger NF1-associated malignant peripheral nerve sheath tumors studies that interrogate these suggestive molecular findings and (MPNST) (36–38), juvenile myelomonocytic leukemia correlate molecular alterations with survival data seem warranted. (JMML) (39, 40), megakaryocytic AML of Down syndrome Based on the data available today, alterations in treatment are (41), and ETP-ALL (16). In addition, mutations in genes other not warranted for children with an ETP-ALL immunophenotype than EED, SUZ12, and EZH2 with a documented effect on detected at the time of initial diagnosis (and data in adults may canonical PRC2 function are inactivated in cancer (including be too sparse to draw definitive conclusions). However, there is a hematologic neoplasms). This includes, e.g., Jarid2 (49– 51) higher rate of poor early response, and overt induction failure in and Asxl1 (52). Finally, genes involved in Polycomb Repressive ETP-ALL, and most relapses that occur in ETP-ALL occur early Complex 1 have shown tumor suppressor function in model in therapy. Thus, there is a subset of ETP-ALL patient (and/or systems (53, 54). In summary, these findings suggest com - those with an immature immunophenotype) who relapse quickly monalities between the roles of different Polycomb genes in with extremely refractory disease, or never go into remission in cancer that may potentially be targetable. As an example, the the first place. This raises the following pressing questions for chromatin-binding protein BRD4 has been suggested as such the field: a therapeutic target (36). The careful study of patient material and genetic models holds great promise in this regard. We 1. Is flow cytometry plus early treatment response the best strat - are optimistic that a deeper understanding of the molecular egy to identify high-risk patients, or is there a less equivocal mechanisms underlying this collaboration will result in marker (ETP-transcriptional signature? PRC2 mutation? successful future clinical trials and ultimately in improved Kinase signature?) that can identify the group of patients with outcomes for patients. ETP-ALL and a poor outcome? 2. Assuming that ultra-high-risk patients can be identified at AUTHOR CONTRiBUTiONS diagnosis or very early in therapy can advances in understand- ing the biology of the disease pave the way toward targeted All authors listed, have made substantial, direct, and intellectual and more effective therapies? contribution to the work and approved it for publication. Frontiers in Pediatrics | www.frontiersin.org 5 May 2016 | Volume 4 | Article 49 Bernt et al. PRC2 in ETP-ALL lymphoid transformation. Cancer Cell (2013) 23:677–92. doi:10.1016/j. Re Fe Re NCe S ccr.2013.04.011 18. Caganova M, Carrisi C, Varano G, Mainoldi F, Zanardi F, Germain PL, 1. Coustan-Smith E, Mullighan CG, Onciu M, Behm FG, Raimondi SC, Pei et  al. Germinal center dysregulation by histone methyltransferase EZH2 D, et  al. Early T-cell precursor leukaemia: a subtype of very high-risk acute promotes lymphomagenesis. J Clin Invest (2013) 123:5009–22. doi:10.1172/ lymphoblastic leukaemia. Lancet Oncol (2009) 10:147–56. doi:10.1016/ JCI70626 S1470-2045(08)70314-0 19. Liu Y, Easton J, Shao Y, Wilkinson M, Edmonson M, Ma X, et al. The genomic 2. Inukai T, Kiyokawa N, Campana D, Coustan-Smith E, Kikuchi A, Kobayashi landscape of childhood T-lineage acute lymphoblastic leukemia (abstract). M, et  al. Clinical significance of early T-cell precursor acute lymphoblastic Blood (2015) 126:691. doi:10.1172/JCI70626 leukaemia: results of the Tokyo Children’s Cancer Study Group Study L99-15. 20. Neumann M, Heesch S, Schlee C, Schwartz S, Gokbuget N, Hoelzer D, Br J Haematol (2012) 156:358–65. doi:10.1111/j.1365-2141.2011.08955.x et  al. Whole-exome sequencing in adult ETP-ALL reveals a high rate of 3. Ma M, Wang X, Tang J, Xue H, Chen J, Pan C, et  al. Early T-cell precursor DNMT3A mutations. Blood (2013) 121:4749–52. doi:10.1182/blood-2012-11- leukemia: a subtype of high risk childhood acute lymphoblastic leukemia. Front Med (2012) 6:416–20. doi:10.1007/s11684-012-0224-4 21. Li W, Gao C, Cui L, Liu S, Zhao X, Zhang R, et al. DNMT3A mutations and 4. Allen A, Sireci A, Colovai A, Pinkney K, Sulis M, Bhagat G, et al. Early T-cell prognostic significance in childhood acute lymphoblastic leukemia. Leuk precursor leukemia/lymphoma in adults and children. Leuk Res (2013) Lymphoma (2015) 56:1066–71. doi:10.3109/10428194.2014.947607 37:1027–34. doi:10.1016/j.leukres.2013.06.010 22. Ho PA, Kutny MA, Alonzo TA, Gerbing RB, Joaquin J, Raimondi SC, et  al. 5. Neumann M, Heesch S, Gokbuget N, Schwartz S, Schlee C, Benlasfer O, et al. Leukemic mutations in the methylation-associated genes DNMT3A and Clinical and molecular characterization of early T-cell precursor leukemia: a IDH2 are rare events in pediatric AML: a report from the Children’s Oncology high-risk subgroup in adult T-ALL with a high frequency of FLT3 mutations. Group. Pediatr Blood Cancer (2011) 57:204–9. doi:10.1002/pbc.23179 Blood Cancer J (2012) 2:e55. doi:10.1038/bcj.2011.49 23. Jaiswal S, Fontanillas P, Flannick J, Manning A, Grauman PV, Mar BG, et al. 6. Van Vlierberghe P, Ambesi-Impiombato A, De Keersmaecker K, Hadler M, Age-related clonal hematopoiesis associated with adverse outcomes. N Engl Paietta E, Tallman MS, et al. Prognostic relevance of integrated genetic pro- J Med (2014) 371:2488–98. doi:10.1056/NEJMoa1408617 filing in adult T-cell acute lymphoblastic leukemia. Blood (2013) 122:74–82. 24. Di Croce L, Helin K. Transcriptional regulation by Polycomb group proteins. doi:10.1182/blood-2013-03-491092 Nat Struct Mol Biol (2013) 20:1147–55. doi:10.1038/nsmb.2669 7. Jain N, Lamb AE, O’Brien S, Ravandi F, Konopleva M, Jabbour E, et al. Early 25. Xu K, Wu ZJ, Groner AC, He HH, Cai C, Lis RT, et al. EZH2 oncogenic activity T-cell precursor acute lymphoblastic leukemia/lymphoma (ETP-ALL/LBL) in castration-resistant prostate cancer cells is Polycomb-independent. Science in adolescents and adults: a high-risk subtype. Blood (2016) 127:1863–9. (2012) 338:1465–9. doi:10.1126/science.1227604 doi:10.1182/blood-2015-08-661702 26. Wassef M, Rodilla V, Teissandier A, Zeitouni B, Gruel N, Sadacca B, et  al. 8. Ferrando AA, Neuberg DS, Staunton J, Loh ML, Huard C, Raimondi SC, Impaired PRC2 activity promotes transcriptional instability and favors breast et  al. Gene expression signatures define novel oncogenic pathways in T cell tumorigenesis. Genes Dev (2015) 29:2547–62. doi:10.1101/gad.269522.115 acute lymphoblastic leukemia. Cancer Cell (2002) 1:75–87. doi:10.1016/ 27. Shi B, Liang J, Yang X, Wang Y, Zhao Y, Wu H, et al. Integration of estrogen and S1535-6108(02)00018-1 Wnt signaling circuits by the polycomb group protein EZH2 in breast cancer 9. Homminga I, Pieters R, Langerak AW, de Rooi JJ, Stubbs A, Verstegen M, cells. Mol Cell Biol (2007) 27:5105–19. doi:10.1128/MCB.00162-07 et al. Integrated transcript and genome analyses reveal NKX2-1 and MEF2C 28. Lee ST, Li Z, Wu Z, Aau M, Guan P, Karuturi RK, et  al. Context-specific as potential oncogenes in T cell acute lymphoblastic leukemia. Cancer Cell regulation of NF-kappaB target gene expression by EZH2 in breast cancers. (2011) 19:484–97. doi:10.1016/j.ccr.2011.02.008 Mol Cell (2011) 43:798–810. doi:10.1016/j.molcel.2011.08.011 10. Gutierrez A, Dahlberg SE, Neuberg DS, Zhang J, Grebliunaite R, Sanda T, 29. Kim KH, Kim W, Howard TP, Vazquez F, Tsherniak A, Wu JN, et  al. SWI/ et  al. Absence of biallelic TCRgamma deletion predicts early treatment SNF-mutant cancers depend on catalytic and non-catalytic activity of EZH2. failure in pediatric T-cell acute lymphoblastic leukemia. J Clin Oncol (2010) Nat Med (2015) 21:1491–6. doi:10.1038/nm.3968 28:3816–23. doi:10.1200/JCO.2010.28.3390 30. Wilson BG, Wang X, Shen X, McKenna ES, Lemieux ME, Cho YJ, et  al. 11. Yang YL, Hsiao CC, Chen HY, Lin KH, Jou ST, Chen JS, et al. Absence of bial- Epigenetic antagonism between polycomb and SWI/SNF complexes during lelic TCRgamma deletion predicts induction failure and poorer outcomes in oncogenic transformation. Cancer Cell (2010) 18:316–28. doi:10.1016/j. childhood T-cell acute lymphoblastic leukemia. Pediatr Blood Cancer (2012) ccr.2010.09.006 58:846–51. doi:10.1002/pbc.24021 31. Kadoch C, Hargreaves DC, Hodges C, Elias L, Ho L, Ranish J, et al. Proteomic 12. Callens C, Baleydier F, Lengline E, Ben Abdelali R, Petit A, Villarese P, et al. and bioinformatic analysis of mammalian SWI/SNF complexes identifies Clinical impact of NOTCH1 and/or FBXW7 mutations, FLASH deletion, and extensive roles in human malignancy. Nat Genet (2013) 45:592–601. TCR status in pediatric T-cell lymphoblastic lymphoma. J Clin Oncol (2012) doi:10.1038/ng.2628 30:1966–73. doi:10.1200/JCO.2011.39.7661 32. Knutson SK, Warholic NM, Wigle TJ, Klaus CR, Allain CJ, Raimondi A, et al. 13. Zuurbier L, Gutierrez A, Mullighan CG, Cante-Barrett K, Gevaert AO, de Rooi J, et al. Immature MEF2C-dysregulated T-cell leukemia patients have an early Durable tumor regression in genetically altered malignant rhabdoid tumors by inhibition of methyltransferase EZH2. Proc Natl Acad Sci U S A (2013) T-cell precursor acute lymphoblastic leukemia gene signature and typically have non-rearranged T-cell receptors. Haematologica (2014) 99:94–102. 110:7922–7. doi:10.1073/pnas.1303800110 33. Nikoloski G, Langemeijer SM, Kuiper RP, Knops R, Massop M, Tonnissen doi:10.3324/haematol.2013.090233 14. Patrick K, Wade R, Goulden N, Mitchell C, Moorman AV, Rowntree C, et al. ER, et  al. Somatic mutations of the histone methyltransferase gene EZH2 in myelodysplastic syndromes. Nat Genet (2010) 42:665–7. doi:10.1038/ Outcome for children and young people with Early T-cell precursor acute lymphoblastic leukaemia treated on a contemporary protocol, UKALL 2003. ng.620 34. Ernst T, Chase AJ, Score J, Hidalgo-Curtis CE, Bryant C, Jones AV, et  al. Br J Haematol (2014) 166:421–4. doi:10.1111/bjh.12882 15. Wood BL, Winter SS, Dunsmore KP, Devidas M, Chen S, Asselin B, et  al. Inactivating mutations of the histone methyltransferase gene EZH2 in myeloid disorders. Nat Genet (2010) 42:722–6. doi:10.1038/ng.621 T-Lymphoblastic Leukemia (T-ALL) Shows Excellent Outcome, Lack of Significance of the Early Thymic Precursor (ETP) Immunophenotype, and 35. Guglielmelli P, Biamonte F, Score J, Hidalgo-Curtis C, Cervantes F, Maffioli M, et al. EZH2 mutational status predicts poor survival in myelob fi rosis. Blood Validation of the Prognostic Value of End-Induction Minimal Residual Disease (MRD) in Children’s Oncology Group (COG) Study AALL0434 (2011) 118:5227–34. doi:10.1182/blood-2011-06-363424 (abstract). Blood (2014) 124:1. 36. De Raedt T, Beert E, Pasmant E, Luscan A, Brems H, Ortonne N, et al. PRC2 16. Zhang J, Ding L, Holmfeldt L, Wu G, Heatley SL, Payne-Turner D, et al. The loss amplifies Ras-driven transcription and confers sensitivity to BRD4-based genetic basis of early T-cell precursor acute lymphoblastic leukaemia. Nature therapies. Nature (2014) 514:247–51. doi:10.1038/nature13561 (2012) 481:157–63. doi:10.1038/nature10725 37. Zhang M, Wang Y, Jones S, Sausen M, McMahon K, Sharma R, et al. Somatic 17. Beguelin W, Popovic R, Teater M, Jiang Y, Bunting KL, Rosen M, et al. EZH2 is mutations of SUZ12 in malignant peripheral nerve sheath tumors. Nat Genet required for germinal center formation and somatic EZH2 mutations promote (2014) 46:1170–2. doi:10.1038/ng.3116 Frontiers in Pediatrics | www.frontiersin.org 6 May 2016 | Volume 4 | Article 49 Bernt et al. PRC2 in ETP-ALL 38. Lee W, Teckie S, Wiesner T, Ran L, Prieto Granada CN, Lin M, et al. PRC2 is 48. Maude SL, Dolai S, Delgado-Martin C, Vincent T, Robbins A, Selvanathan A, recurrently inactivated through EED or SUZ12 loss in malignant peripheral et  al. Efficacy of JAK/STAT pathway inhibition in murine xenograft models nerve sheath tumors. Nat Genet (2014) 46:1227–32. doi:10.1038/ng.3095 of early T-cell precursor (ETP) acute lymphoblastic leukemia. Blood (2015) 39. Caye A, Strullu M, Guidez F, Cassinat B, Gazal S, Fenneteau O, et al. Juvenile 125:1759–67. doi:10.1182/blood-2014-06-580480 myelomonocytic leukemia displays mutations in components of the RAS 49. Shen X, Kim W, Fujiwara Y, Simon MD, Liu Y, Mysliwiec MR, et al. Jumonji pathway and the PRC2 network. Nat Genet (2015) 47:1334–40. doi:10.1038/ modulates polycomb activity and self-renewal versus differentiation of stem ng.3420 cells. Cell (2009) 139:1303–14. doi:10.1016/j.cell.2009.12.003 40. Stieglitz E, Taylor-Weiner AN, Chang TY, Gelston LC, Wang YD, Mazor T, 50. Puda A, Milosevic JD, Berg T, Klampfl T, Harutyunyan AS, Gisslinger B, et al. et al. The genomic landscape of juvenile myelomonocytic leukemia. Nat Genet Frequent deletions of JARID2 in leukemic transformation of chronic myeloid (2015) 47:1326–33. doi:10.1038/ng.3400 malignancies. Am J Hematol (2012) 87:245–50. doi:10.1002/ajh.22257 41. Yoshida K, Toki T, Okuno Y, Kanezaki R, Shiraishi Y, Sato-Otsubo A, et  al. 51. Simon C, Chagraoui J, Krosl J, Gendron P, Wilhelm B, Lemieux S, et al. A key e l Th andscape of somatic mutations in Down syndrome-related myeloid role for EZH2 and associated genes in mouse and human adult T-cell acute disorders. Nat Genet (2013) 45:1293–9. doi:10.1038/ng.2759 leukemia. Genes Dev (2012) 26:651–6. doi:10.1101/gad.186411.111 42. Danis E, Yamauchi T, Echanique K, Zhang X, Haladyna JN, Riedel SS, et al. 52. Abdel-Wahab O, Adli M, LaFave LM, Gao J, Hricik T, Shih AH, et  al. Ezh2 controls an early hematopoietic program and growth and survival sig- ASXL1 mutations promote myeloid transformation through loss of PRC2- naling in early T cell precursor acute lymphoblastic leukemia. Cell Rep (2016) mediated gene repression. Cancer Cell (2012) 22:180–93. doi:10.1016/j. 14:1953–65. doi:10.1016/j.celrep.2016.01.064 ccr.2012.06.032 43. Haydu JE, Ferrando AA. Early T-cell precursor acute lymphoblastic 53. Ross K, Sedello AK, Todd GP, Paszkowski-Rogacz M, Bird AW, Ding L, et al. leukaemia. Curr Opin Hematol (2013) 20:369–73. doi:10.1097/MOH. Polycomb group ring finger 1 cooperates with Runx1 in regulating differen - 0b013e3283623c61 tiation and self-renewal of hematopoietic cells. Blood (2012) 119:4152–61. 44. Matlawska-Wasowska K, Kang H, Devidas M, Wen J, Harvey RC, Nickl CK, doi:10.1182/blood-2011-09-382390 et al. MLL rearrangements impact outcome in HOXA-deregulated T-lineage 54. Oguro H, Yuan J, Tanaka S, Miyagi S, Mochizuki-Kashio M, Ichikawa H, et al. acute lymphoblastic leukemia: a children’s oncology group study. Leukemia Lethal myelob fi rosis induced by Bmi1-deficient hematopoietic cells unveils a (2016). doi:10.1038/leu.2016.60 tumor suppressor function of the polycomb group genes. J Exp Med (2012) 45. Dik WA, Brahim W, Braun C, Asnafi V, Dastugue N, Bernard OA, et  al. 209:445–54. doi:10.1084/jem.20111709 CALM-AF10+ T-ALL expression profiles are characterized by overexpression of HOXA and BMI1 oncogenes. Leukemia (2005) 19:1948–57. doi:10.1038/ Conflict of Interest Statement: The authors declare that the research was con - sj.leu.2403891 ducted in the absence of any commercial or financial relationships that could be 46. Ben Abdelali R, Asnafi V, Petit A, Micol JB, Callens C, Villarese P, et al. The construed as a potential conflict of interest. prognosis of CALM-AF10-positive adult T-cell acute lymphoblastic leukemias depends on the stage of maturation arrest. Haematologica (2013) 98:1711–7. doi:10.3324/haematol.2013.086082 Copyright © 2016 Bernt, Hunger and Neff. i Th s is an open-access article distributed 47. Bond J, Marchand T, Touzart A, Cieslak A, Trinquand A, Sutton L, et al. An under the terms of the Creative Commons Attribution License (CC BY). The use, early thymic precursor phenotype predicts outcome exclusively in HOXA- distribution or reproduction in other forums is permitted, provided the original overexpressing adult T-cell acute lymphoblastic leukemia: a group for author(s) or licensor are credited and that the original publication in this journal research in adult acute lymphoblastic leukemia study. Haematologica (2016). is cited, in accordance with accepted academic practice. No use, distribution or doi:10.3324/haematol.2015.141218 reproduction is permitted which does not comply with these terms. Frontiers in Pediatrics | www.frontiersin.org 7 May 2016 | Volume 4 | Article 49

Journal

Frontiers in PediatricsPubmed Central

Published: May 18, 2016

There are no references for this article.