Abstract A2-09: Oncogenic dysregulations in neuroblastoma are associated with distal large chromosomal aberrations

Abstract

Abstract Neuroblastoma (NB) is a prenatal malignancy diagnosed in infants, arising from neural crest cells, with heterogeneous etiology and prognosis. High risk tumors harbor large chromosomal alterations that substantially impact the expression of approximately one quarter of the genome; this challenges the distinction between driver and passenger copy number mutations, hindering the discovery of new therapeutic targets. Our lab has established a paradigm known as the bottleneck hypothesis in which multiple disease driver genetic elements integrate their aberrant signal through regulatory bottleneck, typically formed by a few transcriptional regulators, responsible of maintaining aggressive phenotypes. A new algorithm developed in our lab DIGGIT (Driver-Gene Inference by Genetical-Genomic Information Theory) has been able to successfully identify the impact of deletions in KLHL9 on the transcriptional activity of C/EBPβ and C/EBPδ, established master regulators of mesenchymal subtype in glioblastoma(1). To gain understanding of the association between genetics and the molecular phenotype which drives NB disease we analyze genome wide expression and copy number profiles from primary tumors from two independent cohorts with clinical information available; TARGET (n=250) and SIOPEN (n2=278); first, we identify genomic regions that prevalently suffer gain/loss aberrations which genetic dosage is associated with patient survival using a Cox hazards model. The main covariates are chromosome 1p (P=2.3e-9), 3p (P = 1.1e-2), 6q (P=), 11q (P=9.7e-7), 17p(P=5.3e-4) and 17q(P=4.5e-3). Also, we observed a multiplicative affect of chromosome 17 imbalance between p and q arms (P=7.8e-6). All measured with independence of MYCN amplification (the main hallmark of NB aggressive tumors). We used the above-mentioned dosage of these regions as trait loci to perform trans-aQTL analysis using the algorithm DIGGIT. We also included p and q arms combined cox linear model of chromosomes 11 and 17 to study the imbalance effect. The LIM domain only 1 protein (LMO1) is a validated NB oncogene which expression is increased in 9% of patients due to duplication events of its chromosome 11p15 locus. These events only partially explain LMO1 de-regulation. Our findings show that deletions on the distal arm 11q are implicated in LMO1 increased activity (P=4.3e-8), this effect is significantly stronger than its own loci duplication acting in cis (P=7.07e-5). The linear combination of p and q arms strikes with a p-value P=7.8e-10 supporting the additive effect of chromosome 11 imbalance. In addition to this finding, which confirms the oncogenic role of LMO1 in this disease, our integrated analysis also identified a plethora of additional findings providing plausible hypotheses for genetic alterations that contribute to dysregulation of driver genes in NB. The approach presented here is especially well suited to study tumors characterized by genomic instabilities leading to large chromosomal rearrangements, despite a paucity of recurrent point mutations. 1. James C. Chen, et al. Identification of Causal Genetic Drivers of Human Disease through Systems-Level Analysis of Regulatory Networks. Cell, Volume 159, Issue 2, p402–414, 9 October 2014. Citation Format: Gonzalo Lopez, Mariano Alvarez, James Chen, Presha Rajbhandari, Kristina A. Cole, Edward F. Attiyeh, Sharon Diskin, Pieter Mestdagh, Jo Vandesompele, John M. Maris, Andrea Califano. Oncogenic dysregulations in neuroblastoma are associated with distal large chromosomal aberrations. [abstract]. In: Proceedings of the AACR Special Conference on Translation of the Cancer Genome; Feb 7-9, 2015; San Francisco, CA. Philadelphia (PA): AACR; Cancer Res 2015;75(22 Suppl 1):Abstract nr A2-09.