The Neuroblastoma Genome and Epigenome - Patient Stratification and Identification of Candidate Genes

Abstract

Neuroblastoma (NB) is a tumor of the sympathetic nervous system, and the most common extracranial tumor of childhood. The prognosis for high-stage NBs is still poor, with survival rates of about 35%. Side-effects of treatment in these young children can also be severe. It is therefore important to develop better tools for improved patient stratification as well as to identify new targets for therapy. Aims: Using genetic and epigenetic approaches, this thesis aimed to analyze candidate genes with potential involvment in the initation/progression of NB and to identify genes that can be used for improved patient stratification. Results: The six candidate genes located in chromosome region 1p36.22 were down-regulated in tumors from patients with an unfavorable outcome compared with a favorable. DNA methylation was shown not to be involved in the down-regulation of gene transcripts. In a more comprehensive analysis of 1p36, four genes, ERRFI1, PIK3CD, RBP7 and CASZ1, were up-regulated by epigenetic treatment. Bisulfite sequencing revealed that DNA methylation most likely was not involved, suggesting for the potential involvement of other epigenetic mechanisms such as histone deacetylation. Missense mutations were identified in PIK3CD and ERRFI1 and the down-regulated mRNA expression of PIK3CD and CASZ1 was detected in high-stage NB. CASZ1 plays a role in neural development and is therefore an interesting candidate for further study. In a genome-wide analysis of DNA methylation, a group of methylated genes for which we showed gene expression was affected by epigenetic treatment was selected for further analysis. A selected group, e.g. SCNN1A, PRKCDBP and KRT19 could be used to distinguish between patients with an unfavorable outcome from those with a favorable one. Whole-genome copy number analysis of NB tumors identified homozygous deletions in the CDKN2A and RBMS3 genes. Moreover, copy neutral loss of heterozygosity was rare, but could be detected in three chromosomal regions. Tumors with MYCN amplification and those with 11q deletion displayed very different genomic profiles. The 11q-deletion group had significantly more chromosomal breaks than the other group, indicative of an 11q localized chromosomal instability gene (CIN). This group also had a significantly higher age at diagnosis. The groups defined by 11q deletion, MYCN amplification and 17q gain were the only groups associated with poor patient outcome. Conclusions: Whole-genome profiles add valuable information about genomic aberrations, which are important prognostic factors in NB. Aberrant DNA methylation can be a very early event in tumor development as well as in tumor progression. It is therefore of great importance to learn more about both the genetic and epigenetic profiles of NB. This thesis has added to the current knowledge in these regards and has also identified important genetic aberrations, as well as aberrantly methylated genes. In the future, these aberrations could possibly be used in patient stratification, as biomarkers or as targets for therapy.