Posted: August 14, 2017

View the TCGA study of pancreatic ductal adenocarcinoma here

TCGA study of pancreatic cancer reveals novel biological features

Pancreatic cancer is the 3rd leading cause of cancer death in the United States. It is aggressive, usually presents as advanced disease, and does not respond to most treatments. To better understand the genetic alterations that are responsible for this deadly disease, researchers with The Cancer Genome Atlas comprehensively studied 150 pancreatic tumors and published their results in Cancer Cell on August 14, 2017. Their analysis corroborated previously identified pancreatic cancer genes and subtypes, discovered novel characteristics, and highlighted genetic alterations that may help qualify patients for treatments based on precision medicine. 

Pancreatic cancer consists of very few tumor cells surrounded by thick supportive tissue. Tissue samples tend to contain a small proportion of cancer cells, making pancreatic cancer particularly difficult to study. To address this, the research team used special genomic techniques, including deep and targeted sequencing based on prior knowledge of pancreatic cancer, to successfully characterize tumor samples with a wide range of cancer cell content.

Using these rigorously generated data, the authors determined that 93% of the cancers studied had mutations in KRAS, a gene that encodes an important signaling protein involved in cell growth and cell death. They further identified that in some cases, two different mutations of the KRAS gene were present within a single tumor. Though the association of KRAS with pancreatic cancer was previously known, this is the first demonstration of simultaneous, distinct KRAS mutations in the same tumor in pancreatic cancer and across the whole TCGA dataset. Of the tumors under study that did not harbor a KRAS mutation, 60 percent had mutations in genes that encode other members of the same signaling pathway, highlighting the importance of the RAS-MAPK signaling pathway in pancreatic cancer. The researchers also identified a novel gene driver that was recurrently mutated in pancreatic cancer, RREB1, another member of the RAS-MAPK pathway.

The study further sought to examine potential therapeutic opportunities for pancreatic cancer patients. The authors used an algorithm called PHIAL to match genetic alterations present in the study’s samples with corresponding targeted therapies. They observed that 42 percent of patients had cancers with at least one genomic alteration that may qualify for current clinical trials. For example, 8 percent of the tumors harbored changes in DNA repair pathways, which have been shown to be sensitive to therapies such as platinum-based chemotherapy and poly-(ADP-ribose) polymerase (PARP) inhibition.

Finally, the authors utilized multiple types of genomic analysis, including analysis of regulatory RNAs called microRNAs (miRNAs) and long noncoding RNAs (lncRNAs), and chemical marks on DNA that regulate gene expression called DNA methylation, to identify and characterize prognostic subtypes of pancreatic cancer. They found that these attributes form two distinct clusters that correspond to the previously described basal-like and classical subtypes of pancreatic cancer. Together, these findings reveal important insights into the complexity of pancreatic cancer and highlight opportunities for precision medicine approaches. The Cancer Genome Atlas Research Network is led by the National Cancer Institute (NCI) and the National Human Genome Research Institute, both parts of the National Institutes of Health.