ALBERT

Description: Key Points SH2B3 is a recessive tumor suppressor gene with germline and somatic mutations in ALL.

Description: Abstract 294 T-cell acute lymphoblastic leukemia (T-ALL) is a highly aggressive hematologic tumor associated with poor prognosis. Over the last decade, microarray gene expression studies have shown that T-ALL encompasses distinct molecular groups defined by characteristic gene expression signatures and molecular studies have uncovered major mechanisms of T-cell transformation and numerous oncogenes and tumor suppressors involved in the pathogenesis of T-ALL. This molecular and genetic heterogeneity suggests that distinct groups of T-ALL may respond differently to chemotherapy. However, today there are no well established prognostic markers for patient stratification in this disease and all T-ALL patients are treated under the same therapeutic scheme. To overcome this obstacle here we performed a comprehensive analysis of the prognostic significance of molecular groups defined gene microarray expression profiling, copy number alterations identified by array-Comparative Genomic Hybridization, immunophenotypic markers and mutation analysis of all major adult T-ALL oncogenes (NOTCH1, IL7R, FLT3, NRAS) and tumor suppressor genes (FBXW7, PTEN, DNM2, PHF6, BCL11B, WT1, EZH2, ETV6, IDH1, IDH2, DNMT3A, GATA3 and RUNX1) in a clinical series of 53 primary leukemia samples uniformly treated according to the Eastern Cooperative Oncology Group (ECOG) E2993 protocol. Unsupervised analysis of gene expression oligonucleotide microarrays in this series revealed the presence of 2 stable gene expression clusters corresponding to early immature (n = 28) and cortical/mature (n = 25) adult T-ALLs respectively. Early immature T-ALLs show gene expression signatures related to hematopoietic stem cells and myeloid progenitors and are associated with poor prognosis and reduced overall survival compared with cortical/mature adult T-ALLs (P = 0.011). Immunophenotypic analysis showed that CD13 expression was strongly associated with poor survival in our patient series (P=0.002), whereas other early immature or myeloid antigens including CD34 and CD33 showed no clinical impact. Array comparative genomic hybridization (aCGH) analysis in this series identified absence of biallelic deletion of TCRγ, a cytogenetic feature associated with early induction failure and inferior overall survival rates in pediatric T-ALL (Gutierrez et al., JCO, 2010), in 27 out of 53 (51%) adult T-ALL samples analyzed. Notably, 22 of 27 T-ALLs with absence of biallelic deletion of TCRγ were found in the early immature adult T-ALL group, and as in pediatric T-ALL, this molecular marker was associated with inferior overall survival (P = 0.022). In addition, heterozygous deletions of the short arm of chromosome 17 encompassing the TP53 tumor suppressor gene predicted for worse clinical outcome (P = 0.0005); while, homozygous deletions of the CDKN2A/CDKN2B locus on the short arm of chromosome 9 was associated with favorable prognosis (P = 0.012). Most notably, the favorable prognostic effect of homozygous CDKN2A/CDKN2B deletions in our series was restricted to the good prognostic subtype of leukemia samples with a mature/cortical gene expression profile. Finally, comprehensive mutation analysis of adult T-ALL oncogenes and tumor suppressors demonstrated a favorable prognosis for patients with activation of NOTCH1 signaling as result of mutations in NOTCH1 and/or FBXW7 (P = 0.021) and for those harboring heterozygous inactivating mutations or deletions in the BCL11B tumor suppressor gene (P = 0.041). In contrast, somatic mutations in genes targeting the epigenetic regulators DNMT3A (P = 0.003) and IDH1/2 (P = 0.011) were associated with adverse prognosis. Multivariate analyses highlighted the association of NOTCH1 and/or FBXW7 mutations with favorable outcome and that of TP53 deletions and DNMT3A mutations with poor prognosis in adult T-ALL. Importantly, homozygous CDKN2A/CDKN2B deletions and CD13 expression may serve as prognostic markers to further stratify low-risk cortical/mature adult T-ALL leukemias, whereas DNMT3A mutations might be useful for risk stratification within high-risk early immature adult T-ALLs. Overall, our comprehensive analysis of molecular prognostic markers identify for the first time a subset of adult T-ALLs with very poor response to intensified chemotherapy, highlighting the need to introduce alternative therapies aiming to improve the therapeutic outcome in this group. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 406 T-cell acute lymphoblastic leukemia (T-ALL) is a heterogeneous disease in which genetic lesions coordinately affect cell proliferation, differentiation and survival of thymocytes. Although pediatric T-ALL has been extensively studied on the genomic level during the last decade, large scale genomic studies in adult T-ALL are sparse. Importantly, significant differences in treatment outcome are present between pediatric and adult T-ALL. Here we performed integrative analysis of gene expression, copy number alterations and mutation analysis on 58 adult T-ALL samples treated in the Eastern Cooperative Oncology Group (ECOG) E2993 protocol. Unsupervised and consensus clustering of microarray gene expression data in this series revealed the presence of 2 stable gene expression clusters resembling immature/ETP (n=30) and typical (n=28) T-ALL gene signatures. Indeed, gene set enrichment analysis (GSEA) confirmed that our ETP/immature adult T-ALL signature closely resembled the previously published ETP profiles. In this adult patient cohort, the immature/ETP patient samples were associated with poor outcome (5 year OS 36% for ETP and 58% for non ETP TALL, P=0.088). Additional supervised analyses revealed differential expression of ERAP2, a gene involved in the processing of MHC peptides, between early relapse and long time survivor adult T-ALL patients. Consistently low ERAP2 expression was associated poor prognosis in adult T-ALL (OS 26%, 36% and 78% for 1st 2nd and 3rd terciles respectively, P=0.016). In addition, absence of TCR gamma deletion (p=0.013) as well as the absence of biallelic CDKN2A/CDKN2B deletion (p=0.009) were shown to be poor prognostic markers in adult T-ALL. To gain additional insight in the biology of adult ETP T-ALL, we performed GSEA analysis using the immature/ETP gene expression profiles. This analysis revealed that the adult ETP T-ALL signature is more closely related to acute myeloid leukemia (AML) profiles as compared to ALL signatures. Given this, we hypothesized that myeloid specific genetic alterations might be uniquely present in ETP adult T-ALL. Mutation analysis of AML oncogenes and tumor suppressor genes revealed the presence of myeloid mutations targeting IDH1, IDH2, DNMT3A, FLT3 and N-RAS in 13/30 (43%) of adult ETP T-ALL cases, whereas only 1/28 (3.5%) typical adult T-ALL showed a mutation in N-RAS. Although these ETP adult T-ALL samples seem to have myeloid properties, they still retain T-lymphoid characteristics including activational NOTCH1 mutations in 12/30 (40%) of ETP adult T-ALLs. Of note, the myeloid specific genetic alterations were preferentially present in NOTCH1 wild type ETP adult T-ALL samples. Next, we used a systems biology approach to identify master regulators that might be able to drive the ETP signature identified in adult T-ALL. For this, we interrogated a T-ALL interactome that was constructed from gene expression profiles of 228 human T-ALL samples. This master regulator analysis identified MN1 as the key transcription factor driving the ETP signature. The MN1 gene was initially identified as fusion partner of TEL (ETV6) in patients with AML. Notably, MN1-TEL expression in multipotent progenitors induces both AML and T-ALL and mutation analysis of ETV6 in adult T-ALL revealed the presence of 10 ETV6 mutants (9 frameshift and 1 splice site mutation) in adult T-ALL. Analysis of bone marrow remission genomic DNA confirmed the somatic origin of ETV6 mutations (3 frameshift and 1 splice site) in each of the 4 patient samples with available material. Most notably, ETV6 mutations were exclusively present in the ETP subtype of adult T-ALL. In addition, 8/10 (80%) of ETV6 mutant ETP T-ALL cases were characterized mutations in NOTCH1, suggesting a specific interaction between the oncogenic NOTCH1 programs and the mutational loss of ETV6 in early T-cell transformation. In conclusion, our integrative genomic analyses in adult T-ALL unraveled a high prevalence of immature/early T-cell progenitor (ETP) T-ALLs, associates the pathogenesis of ETP-TALL with myeloid mutations and identifies ETV6 truncating alleles as the first ETP specific gene mutation that strongly interact with NOTCH1 in T-cell transformation. Disclosures: No relevant conflicts of interest to declare.