High-coverage whole-genome analysis of 1220 cancers reveals hundreds of genes deregulated by rearrangement-mediated cis-regulatory alterations

Abstract

The impact of somatic structural variants (SVs) on gene expression in cancer is largely unknown. Here, as part of the ICGC/TCGA Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium, which aggregated whole-genome sequencing data and RNA sequencing from a common set of 1220 cancer cases, we report hundreds of genes for which the presence within 100 kb of an SV breakpoint associates with altered expression. For the majority of these genes, expression increases rather than decreases with corresponding breakpoint events. Up-regulated cancer-associated genes impacted by this phenomenon include TERT, MDM2, CDK4, ERBB2, CD274, PDCD1LG2, and IGF2. TERT-associated breakpoints involve~3% of cases, most frequently in liver biliary, melanoma, sarcoma, stomach, and kidney cancers. SVs associated with up-regulation of PD1 and PDL1 genes involve~1% of non-amplified cases. For many genes, SVs are significantly associated with increased numbers or greater proximity of enhancer regulatory elements near the gene. DNA methylation near the promoter is often increased with nearby SV breakpoint, which may involve inactivation of repressor elements.

Whole-genome sequencing of acral melanoma reveals genomic complexity and diversity

“…The exceptions were NRAS and CDK6 (both with only a trend for higher expression), TYRP1, and TERT. Upstream breakpoints have been reported to increase expression of TERT due to mechanisms such as enhancer hijacking in other cancer types 36,37 and are also proposed to occur in AM 7 . Breakpoints within the region 20 kb upstream of TERT were identified in 25 tumors, and 14/25 had no other TERT aberration (Supplementary Fig.…”

Chromosome structural variation in tumorigenesis: mechanisms of formation and carcinogenesis

“…Altered genes caused by structural variation may involve driver or passenger mutations through multiple mechanisms in tumorigenesis. Structural variations have functional consequences in tumorigenesis or clonal evolution through multiple mechanisms, such as CNAs, fusion gene rearrangements, and gene expression patterns (epigenetic alterations or inappropriate communication between genes and distal regulatory elements) [ 3 , 5 , 98 ]. Interestingly, the majority of genes with altered expression due to corresponding breakpoint events show upregulated gene expression [ 3 ].…”

“…Structural variations have functional consequences in tumorigenesis or clonal evolution through multiple mechanisms, such as CNAs, fusion gene rearrangements, and gene expression patterns (epigenetic alterations or inappropriate communication between genes and distal regulatory elements) [ 3 , 5 , 98 ]. Interestingly, the majority of genes with altered expression due to corresponding breakpoint events show upregulated gene expression [ 3 ]. Overall, although the vast majority of driver mutations occur in a protein-coding content, which makes up only 1% of the human genome, non-coding structural variations may be underappreciated mutational drivers in cancer genomes [ 99 ].…”

“…Conjoint analysis of structural variation and mRNA expression levels has been used to predict gene truncation affected by structural variation. In some tumor suppressor genes, including, but not limited to, PTEN, TP53 , RB1, NOTCH1, and NF1 , sequences or promoters can be frequently interrupted by translocation and inversion, leading to inactivation and the subsequent onset and progression of cancer [ 3 , 100 ]. In addition, gene truncation can also result from templated insertions or local n -jumps, which cause duplications of internal exons of this gene or insertions of exons from other genes, leading to a nonfunctional transcript [ 1 ].…”

“…The Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium of the International Cancer Genome Consortium (ICGC) and The Cancer Genome Atlas (TCGA) collaborated to analyze structural variants, genomic breakpoint cluster regions, and gene expression based on whole-genome sequencing data from 2658 cancers patients across 38 tumor types [ 3 , 4 ]. However, our understanding of the underlying molecular mechanisms of structural variation remains incomplete.…”