NEW YORK (GenomeWeb) – The first tranche of peer-reviewed papers related to the Pan-Cancer Atlas were published online today, with more expected to follow in five journals published by Cell Press.
Researchers associated with the Cancer Genome Atlas (TCGA) Research Network will publish more than 25 papers this week to discuss findings from the initiative. A multiyear collaboration between the National Cancer Institute and the National Human Genome Research Institute, TCGA has generated maps of genomic changes across 33 cancers.
The first crop of papers, all of which appear in the journal Cancer Cell, cover such diverse topics as gastrointestinal cancer, cancer aneuploidy, and oncogenomic long-noncoding RNAs.
In the first paper, "Comparative molecular analysis of gastrointestinal adenocarcinomas," the authors detail the analysis of 971 adenocarcinomas of the esophagus, stomach, colon, and rectum to identify their molecular characteristics.
Corresponding authors on the study include Vésteinn Thorsson, a senior research scientist at the Institute for Systems Biology, Peter Laird, a professor of epigenetics at the Van Andel Research Institute, and Adam Bass, an associate professor of medicine at Harvard Medical School.
As described in the paper, they characterized fresh, frozen tissue samples using a variety of approaches, including SNP array profiling for somatic copy-number alterations, whole-exome sequencing, array-based DNA methylation profiling, mRNA sequencing, microRNA sequencing, and, for a smaller number of samples, reverse-phase protein array profiling.
Their goal was to evaluate the molecular characteristics that distinguish gastrointestinal tract adenocarcinomas (GIACs) from other cancers, as well as to "investigate the molecular features of GIACs across anatomic boundaries to provide insight into the pathogenesis of these deadly malignancies." They were able to divide the GIACs into five molecular subtypes: those with a high Epstein-Barr virus burden (EBV), microsatellite instability (MSI), chromosomal instability (CIN), genome stable (GS), and those with hypermutated single nucleotide variants (HM-SNV).
According to the paper, the HM-SNV tumors had diverse immune features that varied by tissue and subtype, while CIN tumors had more fragmented copy number changes in the upper gastrointestinal tract. The genome-stable colorectal cancer subtype meantime was shown to be enriched for mutations in the genes SOX9 and PCBP1.
The authors argued that the results highlight how processes such as DNA hypermethylation and chromosomal instability could "manifest themselves in different ways across related tissues," and also underscore how the "consideration of molecular subtypes as well as organ of origin will be essential in the study and treatment of cancer."
Another study published today discusses the link between aneuploidy and a variety of cancer types. Aneuploidy, whole-chromosome or chromosome arm imbalance, is considered to be a pervasive feature in all cancers. In the new paper, entitled, "Genomic and functional approaches to understanding cancer aneuploidy," the authors detail the correlation of aneuploidy with TP53 mutation, somatic mutation rate, and expression of proliferation genes.
Matthew Meyerson, a professor of pathology at HMS, as well as principal investigator on the TCGA project, is corresponding author on the study.
As part of the effort, they first generated an aneuploidy score reflecting the total number of chromosome arms with arm-level, copy-number alterations in a sample. Then they clustered somatic copy number alterations (SCNAs) on each arm according to location and length. Clusters where the mean length of the alteration was greater than 80 percent were deemed positive for an alteration while those that made up less than 20 percent were considered negative.
They subsequently applied the approach to 10,522 cancer genomes spanning 33 cancer types from the TCGA pan-cancer data set. DNA copy number was obtained from Affymetrix SNP 6.0 arrays. Ultimately, they were able to determine the SCNA status of more than 400,000 chromosome arms and more than 175,000 whole chromosomes for the 10,522 cancer genomes.
Finally, they calculated an aneuploidy score that reflected the total burden, or number of events in each sample. According to the authors, the aneuploidy scores correlated highly with the fraction of genomes altered by aneuploidy. They also reported that aneuploidy was anti-correlated with the expression of immune-signaling genes. Chromosome arm-level alterations also showed cancer-specific patterns. "This study defines genomic and phenotypic correlates of cancer aneuploidy and provides an experimental approach to study chromosome arm aneuploidy," the authors wrote.
In yet another TCGA-related paper out today, the authors reported an overview of the epigenetic landscape of long, non-coding RNA genes in 20 cancer types. The paper is entitled, "lncRNA epigenetic landscape analysis identifies EPIC1 as an oncogenic lncRNA that interacts with MYC and promotes cell-cycle progression in cancer." Da Yang, an assistant professor of pharmaceutical sciences at the University of Pittsburgh, is the corresponding author on the paper.
As related in the paper, the authors looked at genes encoding lncRNAs across 6,475 tumors and 455 cancer cell lines. They observed a recurrent hypomethylation of 1,006 lncRNA genes in cancer, including the gene EPIC1, the overexpression of which is associated with a poor prognosis in luminal B breast cancer patients and enhances tumor growth in vitro and in vivo.
They also showed how EPIC1 functions like an oncogenic lncRNA by interacting with the MYC protein and promoting cell-cycle progression. "These discoveries expand upon the known mechanisms of MYC activation in cancer and pave the way to develop therapies that target MYC through its interaction with EPIC1," the authors wrote.
In summary, the authors maintained that the establishment of a detailed knowledge base of DNA-methylation-altered lncRNAs in cancer will "facilitate the identification of cancer-driving lncRNAs."