Search

Introduction: Combined activation of specific oncogenes is a general feature of human cancer and suggests that co-occurrence of particular oncogenic factors provides a selective advantage during cellular transformation. However, the exact molecular mechanisms by which oncoproteins cooperate during malignant transformation often remains elusive. Here, we study the functional relationship between the cooperative oncogenes NOTCH1 and TLX1 in the context of T-cell acute lymphoblastic leukemia to better understand their cooperative mechanism of action during T cell transformation. Methods: In this study, we performed chromatin immunoprecipitation followed by deep sequencing (ChIP-seq) for the TLX1 homeobox oncoprotein in the T-ALL cell line ALL-SIL and analyzed the transcriptional response before and after TLX1 modulation using microarray based methods. We performed integration of TLX1 ChIPseq data with publically available transcription factor binding profiles in T-ALL and evaluated the immunophenotypic and transcriptional effects of ectopic TLX1 expression in thymus-derived CD34+ T-cell progenitors. Results: Integration of TLX1 ChIP-seq data with gene expression profiles after TLX1 knockdown in the TLX1 positive T-ALL cell line ALL-SIL, confirmed the previously established role for TLX1 as transcriptional repressor in T-ALL biology. In line with previous reports (Della-Gatta et al., Nature Medicine, 2012), de novo TLX1 motif discovery identified RUNX1 and ETS1 as important mediators of the global TLX1 transcriptional network. Next, we used TLX1 ChIP-seq data to define TLX1 super-enhancers including several loci critically involved in T-cell biology (e.g. T-cell receptor loci, RAG2, MYB). Furthermore, Gene set Enrichment Analysis (GSEA) showed that TLX1-defined super-enhancers were significantly affected by JQ1 treatment in ALL-SIL. Integration of our TLX1 ChIP-seq data with publically available ChIP-seq data for ICN1, RUNX1 and ETS1 in T-ALL cells (Wang et al., PNAS, 2013) showed a remarkable genome-wide overlap between the binding sites of these four transcription factors. Integration of these binding patterns with transcriptional read-out revealed an unprecedented transcriptional antagonism between TLX1 and NOTCH1, in which TLX1 suppresses the oncogenic NOTCH1 transcriptional program including IL7R, NOTCH3 and c-MYC. In line with this observation, ectopic TLX1 expression in CD34+ human thymic precursor T-cells broadly interfered with the normal T-cell differentiation program causing differentiation arrest, massive apoptosis and a significant downregulation of NOTCH1 target genes including reduced IL7R-alpha surface expression. Conclusion: In conclusion, our study reveals unexpected transcriptional antagonism between the cooperative oncogenes TLX1 and NOTCH1 in the biology of T-ALL. TLX1 mediated suppression of NOTCH signaling might be critically involved in the pre-leukemic phenotype (reduced thymus size and decreased cellularity) that has been observed in Lck-TLX1 transgenic mice (De Keersmaecker et al., Nature Medicine, 2010). Moreover, our results suggests that full malignant transformation of TLX1-driven leukemias might only be possible through acquisition of secondary NOTCH1 mutations that can overcome the initial TLX1 mediated suppression of NOTCH1 signaling, which is in line with the high frequency of gain-of-function NOTCH1 mutations in TLX1 positive human T-ALL. Disclosures No relevant conflicts of interest to declare.

In a series of 153 children with T-cell malignancies enrolled in 2 consecutive European Organization for Research and Treatment of Cancer (EORTC) trials, we assessed the HOX11L2 expression and/or the presence of a t(5;14)(q35;q32). Additionally, in 138 of these patients, HOX11 expression and SIL-TAL rearrangement were also assessed. These alterations were mutually exclusive, and their frequency was 23% (n = 35), 7% (n = 10), and 12% (n = 17), respectively. HOX11L2/t(5;14) positivity was more frequent in acute lymphoblastic leukemia (ALL) with cortical T immunophenotype and in children aged between 6 and 9 years. In contrast with previously reported data, patients positive and negative for HOX11L2/t(5;14) were comparable with regard to clinical outcome as well as to the response to a 7-day prephase treatment or to residual disease at completion of induction therapy. The 3-year event-free survival (EFS) rate (+/- SE percentage) for patients positive and negative for HOX11L2/t(5;14) was 75.5% (+/- 8.1%) and 68.3% (+/- 5.0%), respectively; the hazard ratio was 0.84 (95% confidence interval, 0.40-1.80). Patients with HOX11-high expression and those with SIL-TAL fusion had low levels of residual disease at the end of induction and a favorable prognosis: the 3-year EFS rate was 83.3% (+/- 8.5%) and 75.3% (+/- 12.6%), respectively. The results obtained in HOX11L2/t(5;14) patients in this study do not confirm the unfavorable prognosis reported in previous studies.

MLL amplification was recently recognized as a recurrent aberration in acute myeloid leukemia (AML) and myelodys-plastic syndrome (MDS), associated with adverse prognosis and karyotype complexity. Here we present detailed results of fluorescence in situ hybridization (FISH) and expression analyses of MLL and 5 selected 11q candidate oncogenes (CBL, DDX6, ETS1, FLI1, and PLZF) in 31 patient samples and one cell line with 11q23 gain. FISH analyses revealed that the 11q23 amplicon invariably encompassed MLL, DDX6, ETS1, and FLI1, whereas expression analyses identified MLL and DDX6 as the most differentially expressed genes among samples with and without 11q23 copy gain or amplification. In MLL-amplified samples, a significant transcriptional up-regulation of MEIS1, PROML1, ADAM10, NKG2D, and ITPA was noted. Further analyses, designed to elucidate a possible role of the 11q overexpressed genes (MLL, DDX6, FLI1, and ETS1) in unselected MDS and AML samples, revealed a significant upregulation of MLL in MDS. Our findings confirm the MLL gene as a prominent target of 11q23 amplification and provide further evidence for an etiologic role for MLL gain of function in myeloid malignancies. In addition, our results indicate that the transcriptional program associated with MLL rearrangements and MLL overexpression displays significant similarities.

Abstract 565FN2 Hyperdiploidy >50 chromosomes (HD>50) has long been recognized as favorable group in childhood B-ALL but there is still debate on the factors contributing to heterogeneity of prognosis observed within this entity. Better outcome has been reported for patients (pts) presenting DNA index (DI) >1.16 (Blood 1985;10:213), ≥56 chromosomes (Leukemia 1996;10:213), triple trisomies (TT) +4,+10,+17, double trisomies +4,+10 (Leukemia 2005;19:734) and trisomy 18 (Blood 2003;102:2756) but there is no consensus between the reports and these factors are differently applied in current protocols. We studied these factors in the pts with HD>50 enrolled in the ALL 58951 trial, BFM related. HD>50 were detected by cytogenetics (karyotype/FISH) and/or flow cytometry (DI). In order to analyze the outcome of HD>50 itself, pts with recurrent unfavorable translocations t(9;22), 11q23/MLL+, with t(1;19), t(12;21) or Down Syndrome were excluded, as well as near-triploidies/duplication of hypodiploidies 30–39 chromosomes. Pts were stratified into 4 risk groups (VLR/AR1/AR2/VHR) according to DI, Modal Number of Chromosomes (MNC), WBC, CNS/gonadal involvement, presence of VHR features (unfavorable translocations, poor response to prephase, residual disease (MRD) at the end of induction >10-2). VLR was defined as: DI>1.16 or MNC>50, WBC Overall Results: Out of 1651 B-lineage pts registered in the 58951 study over a 10-year period (1998-2008), a total of 541 pts had HD>50. Median age was 3 years and median WBC was 5.6×109/L. After prephase, 3% (N=17) were poor responders; initial risk group distribution into VLR/AR1/AR2/VHR was 45%/47%/5%/3%. After induction, 540 (99.8%) reached complete remission, 455 of whom had an MRD evaluation: MRD10-3 (N=26;6%) or MRD≥10-2 (N=13;2%). At median follow-up of 6 years, 48 pts (9%) relapsed, 22 pts (4%) died and 6 (1%) died without relapse. The 6-yr EFS was 89% (SE=1.5%). MNC was assessed in 446 pts (82%) with successful karyotypes, MNC ranged from 51 to 66 chromosomes and peaked at 55–56; 87 pts had 51 to 53 chromosomes (HD51-53), 258 pts had 54 to 57 (HD54-57) and 101 pts had ≥58 chromosomes (HD≥58). In these 3 groups, VLR regimen was given to 16%, 50% and 64% respectively. Structural abnormalities were detected in 46% of pts and associated with all MNCs. DI, assessed in 460 pts was Prognostic Factors: The only significant prognostic factors for EFS were MNC, DI, TT, DT and MRD. EFS was clearly improved (p10-2 and 2 pts: MRD≥10-3) and only 1 pt (MNC=58) relapsed. No specific profile of chromosome gains was identified in HD ≥58 since all chromosomes contributed to tri/tetrasomies, except chromosome 1. Likewise, the higher the DI the better the outcome (p=0.01): 6-yr EFS was 83% (DI TT were detected in 168 out of 468 pts, and DT in 242 out of 466 pts. There was no overlap with HD ≥58 since TT and DT were distributed from 52 to 66 chromosomes. One third of TT and of DT had MNC≥58. TT and DT were also of prognostic importance for outcome: the 6-yr EFS rate was 96% (TT) vs 86% (non-TT) (p=0.005) and 94% (DT) vs 84% (non-DT) (p=0.003). A hierarchical variable based on presence of HD≥58, TT or DT showed that HD ≥58 (N=101; 6-yr EFS: 99%) group had a better outcome than TT without HD≥58 (N=115; 6-yr EFS: 93%) and DT without HD≥58 and without +17 (N=55; 6-yr EFS: 84%) groups (p=0.04). We can infer from our results that the good outcome observed for all of the TT and DT was partially due to their association with HD '58. Consequently, the best indicator for excellent outcome was a high MNC (≥58 chromosomes). From our 58951 trial, we can assume that among children with B-ALL and HD>50, those with ≥58 chromosomes, stand every chance of being cured. Our results stress the necessity of karyotype for identifying them since this is the only way to assess MNC. They can also be detected (less accurately) by DI (DI ≥1.24). Therefore, both MNC and DI should be used for stratifying pts in the very low risk groups. Disclosures: No relevant conflicts of interest to declare.

Abstract 471 T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive malignancy associated with the activation of oncogenic transcription factors. Thus, 5% to 10% of pediatric and up to 30% of adult T-ALL patients show aberrant expression of the TLX1 transcription factor oncogene. Aberrant expression of TLX1 in Lck-TLX1 transgenics induces transformation of T-cell progenitors and the development of clonal T-cell lymphoblastic tumors with over 80% penetrance and a latency of 29 weeks. The prolonged latency before T-ALL development in TLX1 transgenic mice and the clonal nature of these tumors suggested the presence of cooperating mutations involved in the pathogenesis of these leukemias. Array Comparative Genomic Hybridization (aCGH) analysis identified several recurring chromosomal alterations in these tumors including three heterozygous deletions on chromosome 12, with a common deleted region containing only the Bcl11b gene. Moreover, DNA sequence analysis of Bcl11b showed the presence of Bcl11b mutations in 4/15 (27%) mouse TLX1 induced T-ALLs, which together with the 3 Bcl11b deletions identified in our aCGH analysis brings the prevalence of Bcl11b alterations in mouse TLX1 induced tumors to 7/15 (47%). Notably, mutation analysis of Bcl11b in a panel of 21 non TLX1-transgenic mouse T-ALL tumors failed to detect any Bcl11b mutations (P Bcl11b encodes a zinc finger transcription factor with a critical role in the differentiation and survival of T-cell progenitors in the thymus. Given the similarities between our mouse model of TLX1-induced leukemia and human T-ALL, we hypothesized that BCL11B alterations could also be implicated in the pathogenesis of human T-ALL. Most notably, aCGH analysis of human T-ALL samples showed the presence of focal heterozygous deletions encompassing the BCL11B locus in 2/69 (3%) T-ALL cases. In addition, mutation analysis of BCL11B in human T-ALL demonstrated the presence of somatic truncating frameshift mutations and somatic missense point mutations in 9/71 (13%) cases analyzed. Interestingly, all 6 BCL11B missense mutations involved zinc finger domains. One sample showed compound heterozygous BCL11B mutations while the remaining cases harbored heterozygous BCL11B lesions. Finally, expression analysis of TLX1 or TLX3 in 8 T-ALLs harboring mutations or deletions in BCL11B with RNA available showed aberrant expression of these transcription factor oncogenes in 5/8 (63%) [TLX1 (4/8); TLX3 (1/8)] of these samples. The identification of a high prevalence of Bcl11b mutations and deletions in TLX1-induced tumors suggested a specific genetic interaction between these two genes in T-cell transformation. We contemplated two alternative scenarios. First, TLX1 could induce partial inactivation of BCL11B in T-cell transformation, an effect that would be reinforced by secondary genetic alterations in the BCL11B locus during tumor progression. Alternatively, BCL11B mutations and TLX1 overexpression could operate in parallel pathways that contribute to distinct but complementary aspects of tumor development. To test if BCL11B could be a direct transcriptional target of TLX1 in T-ALL, we analyzed the binding of TLX1 to the BCL11B promoter via chromatin immunoprecipitation analysis in ALL-SIL, a T-ALL cell line expressing high levels of TLX1 as result of the t(10;14)(q24;q11.2) translocation. This analysis revealed that indeed TLX1 binds to the BCL11B promoter. Moreover, siRNA knockdown of TLX1 in this cell line resulted in transcriptional upregulation of BCL11B, indicating that BCL11B is a direct transcriptional target downregulated by TLX1 in T-ALL. The model that emerges from these results is that the TLX1 oncogene directly downregulates the BCL11B tumor suppressor gene, and that the oncogenic effects of this transcriptional regulatory axis can be fixed and reinforced by secondary deletions and mutations in the BCL11B locus acquired during tumor progression. Altogether, these results highlight the power of integrated genomic analysis of human and mouse leukemias, identify a TLX1-BCL11B transcriptional regulatory axis in the pathogenesis of T-ALL and define BCL11B as a tumor suppressor gene recurrently deleted and mutated in 16% of human T-ALL patients. Disclosures: No relevant conflicts of interest to declare.

Chromosomal rearrangements involving the EVI1 gene are a recurrent finding in malignant myeloid disorders. These translocations or inversions contribute to ectopic expression of, or to the formation of fusion genes involving the EVI1 gene. EVI1 transcriptional activation has been reported in up to 10% of acute myeloid leukemia (AML) patients, and is a prognostic marker predictive of a poor outcome. Recently, microRNA deregulation was identified as a major contributor to cancer initiation and progression. Furthermore, miRNA genes were shown to be directly regulated by activated proto-oncogenes. In this study, we investigated which miRNAs are implicated in the transcriptional pathways governed by the EVI1 oncogene as well as possible microRNAs regulating the expression of the EVI1 gene itself. A total of 384 miRNAs were profiled through automated qRT-PCR using high-throughput quantitative stem-loop RT-PCR. Our patient series consisted of 18 EVI1 rearranged and overexpressing samples confirmed by FISH, karyotyping and qRT-PCR, 11 normal bone marrow samples and 2 CD34+ cord blood fractions. Through integrated statistical analysis we were able to identify 36 significantly upregulated and 44 significantly downregulated miRNAs (p

Numerous studies have shown that the presence, number and type of chromosomal aberrations represent an independent predictor of prognosis in B-cell chronic lymphocytic leukemia (CLL). Consequently, cytogenetic analysis is routinely performed in this disease. However, CLL lymphocytes have a poor mitotic index, generating only 40–50% of abnormal karyotypes. The rate of detection can be increased to 80% by interphase FISH analysis. Since some aberrations, i.e. those not involving the classical regions (13q, 12, 11q, 17p, and 6q), can escape FISH detection, there has been great interest in improved culturing methods with an immunostimulatory CpG oligonucleotide (CpG). We performed a multicentric cytogenetic study to assess the impact of 2 different culturing procedures on the detection of clonal abnormalities in 159 consecutive unselected cases with CLL referred to our centers for routine analysis from October 2007 to July 2008. Dual 72 hours cultures of bone marrow or peripheral blood were set up with the addition of either a conventional B cell mitogen (TPA) or CpG and interleukin-2 (IL2). Cytogenetic analysis was performed on both cultures. FISH analysis using a CLL probe panel analyzing 1–6 regions (13q, centromere 12, 11q, 17p, 6q and 14q32) was also applied on uncultured material, on CpG and/or on TPA culture in 146 cases. Quality of banding and proliferation capacity were assessed in 38 cases. The quality was good in 17 (CpG) and 8 (TPA), intermediate in 12 (CpG) and 18 (TPA), and poor in 9 (CpG) and 12 (TPA) cases. The mitotic index was slightly higher in CpG cultures: 20 mitoses/slide in 13 (CpG) and 9 (TPA) cases, respectively. Clonal abnormalities were identified in 89 cases (56%). In 56 cases, the aberrant clone was detected in both cultures. Of these, the percentage of aberrant metaphases was similar in both cultures in 13, higher in CpG culture in 32 and higher in TPA culture in 11 cases. In 29 and 4 additional cases, a clonal abnormality was detected only in CpG or TPA culture, respectively. Thus, the percentage of abnormal karyotypes with CpG and TPA was 53 and 38%, respectively (p=0,005). “Typical” aberrations, including del(13q), +12, del(11q), del(17p) and del(6q), were detected in 16, 26, 15, 9, and 7 (CpG) cases, and in 10, 20, 16, 8, and 3 (TPA) cases, respectively. Translocations (balanced and unbalanced) were observed in 46 (CpG) and 27 (TPA) cases, whereas aberrations involving 14q32 were seen in 5 (CpG) and 4 (TPA) cases, respectively. Interphase FISH detected del(13q), +12, del(11q), del(17p), del(6q) and del(14q) in 53, 1, 5, 0, 0 and 1 cases, respectively, in which cytogenetic analysis was either normal or abnormal but did not show the specific change. Conversely, FISH did not detect del(13q), +12, del(11q), del(17p), del(6q) and del(14q) in 1, 1, 4, 1, 2 and 0 cases in which the specific change was visible at karyotypic level. FISH was performed on both CpG and TPA cultures in 9 selected cases harbouring an aberration for which a specific commercial probe was available (+12, n=3; del(13q), n=4; del(11q), n=1; and del(17p), n=1) and showing different percentages of aberrant cells in both cultures. In 3 of these cases, fresh uncultured material was also analyzed. The highest percentage of abnormal nuclei was observed in TPA culture in all cases. In the 3 uncultured samples the percentage of aberrant nuclei was similar to the CpG culture. In conclusion, our results confirm an increased detection rate of abnormalities in CLL by using CpG/IL2 stimulation. Interphase FISH can further increase the detection rate of recurrent abnormalities. However, neither cytogenetics nor FISH detected all aberrations, demonstrating the complementary nature of these techniques and the necessity of performing both.

T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive malignancy of thymocytes that accounts for about 15 percent of ALL cases. Leukemic transformation of immature thymocytes is caused by a multistep pathogenesis involving numerous genetic abnormalities providing uncontrolled cell growth. Accumulating evidence suggests the presence of at least 5 different molecular-cytogenetic subgroups in T-ALL, ie. TAL/LMO , TLX1 , TLX3 , HOXA and MYB . Recently, non coding microRNAs were discovered as important regulators of gene and/or protein expression and subsequently shown to be directly implicated in cancer. Nevertheless, it is currently unclear in which way deregulated miRNA expression may contribute to the pathogenesis of T-cell acute leukemia. In this study, we investigated whether different genetic subgroups in T-ALL are characterized by distinct miRNA expression patterns. Therefore, we profiled a total of 360 miRNAs through automated qRT-PCR using high-throughput quantitative stem-loop RT-PCR in a genetically well characterized T-ALL patient cohort (n=52), including 11 HOXA (3 MLL rearranged, 5 inv(7)(p15q35) and 3 CALM - AF10 ), 16 TAL/LMO (7 LMO2 rearranged, 8 TAL1 rearranged, 1 LMO2/TAL1 rearranged), 11 TLX3 and 5 TLX1 rearranged patient samples. Since T-ALL blasts originate from maturating T lymphocytes, we also profiled different subsets of sorted T-cell populations (CD34 + , CD4 + /CD8 + /CD3 − , CD4 + /CD8 + /CD3 + , CD4 + SP and CD8 + SP). These miRNA profiles of normal T-cells served as a negative control for the identification of deregulated miRNA expression that may be truly leukemia associated. SAM analysis (t-test and wilcoxon, FDR=0) identified significant and differentially expressed miRNAs between the HOXA, TLX3 and TAL/LMO subgroups. No significant and differentially expressed miRNAs were obtained for the TLX1 subgroup, probably due to the limited number of patient samples. The HOXA subgroup showed specific up-regulation of miR-196a and miR-196b , which are encoded at the HOXB and HOXA cluster, respectively, but no significantly down-regulated miRNAs could be identified in this subgroup. The TLX3 subgroup was uniquely characterized by the up-regulation of miR-99a , miR-125b , let-7c , miR-508 and miR-509 , and down-regulation of miR-127 and miR-182 . Finally, specific up-regulation of miR-424 , miR-148a , miR-422 , miR-362 , miR-148a , miR-502 , miR-10a , miR-200c , miR-31 , miR-660 and miR-15b , was identified in the TAL/LMO rearranged subgroup, which was also characterized by the specific down-regulation of miR-99b , miR-155 , miR-125a , miR-153 , miR-135a , miR-34a and miR-193b . Next, we evaluated the expression pattern of all significant and differentially expressed miRNAs in the different subsets of sorted T-cell populations. The expression patterns of these miRNAs could be classified into consistently active, completely absent or temporally regulated during T-cell development. For the miRNAs showing a temporal regulation during T-cell maturation, their differential expression in T-ALL subtypes may reflect differences in the maturation arrest of the T-cell of origin, rather than pointing to an oncogenic event. Nevertheless, their constitutive (in)activation in primary T-ALL patients could still be of oncogenic relevance, similar to transcription factors like TAL1 or LMO2 which also show a temporal regulation during T-cell maturation. In contrast, some other miRNAs showed no expression in any of the T-cell populations, providing stronger evidence that their activation in specific T-ALL subtypes may contribute to T-ALL pathogenesis. In conclusion, this study shows that molecular-cytogenetic subgroups in T-ALL are characterized by a specific miRNA expression signature. In addition, correlation of our findings to the expression of these miRNAs in normal T-cell subsets may guide us to the miRNAs with true oncogenic potential. This report paves the way for further investigation directed at the role of these miRNAs in the pathogenesis of T-ALL, which may provide us with further insight in the oncogenic pathways that are (in)activated in different T-ALL subgroups. Ultimately, these deregulated miRNAs may offer new targets for therapeutic intervention.