New Targets for Therapy in Gastric Cancers Identified Via Molecular Classification

Researchers from The Cancer Genome Atlas (TCGA) Network have proposed a molecular classification system that separates gastric cancers into four subtypes.

Adam Bass, MD

Researchers from The Cancer Genome Atlas (TCGA) Network have proposed a molecular classification system that separates gastric cancers into four subtypes.1These new classifications could provide a basis for patient stratification and treatment with targeted therapies.

Gastric cancer was the third leading cause of cancer mortality worldwide in 2012, and aside from anti-HER2 (ERBB2) agents, no targeted therapies are available for its treatment.

Adenocarcinomas comprise the vast majority of gastric cancers. These are currently divided into subtypes using histological criteria according to the Lauren classification or the alternate World Health Organization system. The current systems, however, do not provide information useful in determining specific treatments for particular patients.

In order to develop a classification system that could guide therapeutic decisions, the researchers evaluated gastric adenocarcinoma tumor samples from 295 patients who had not received previous chemotherapy or radiation. The DNA, RNA, and protein from these samples were characterized using six different platforms.

After conducting a computational analysis of these molecular data, the study authors have developed a system of classification of gastric adenocarcinoma with four discrete subtypes: Epstein-Barr virus positive tumors (EBV), microsatellite unstable tumors, genomically stable tumors (GS), and chromosomally unstable tumors (CIN). These categories accounted for 10%, 20%, 20%, and 50% of the tumors in the panel, respectively.

“We had known that there was heterogeneity to gastric cancer,” said corresponding author Adam Bass, MD, of the Dana-Farber Cancer Institute. “However, we had not had the understanding of how so many of the deeper molecular and genomic features of these different cancers are so distinct. Moreover, we did not recognize the extent to which key genomic alterations that mark potential therapeutic vulnerabilities differ so substantially between these subtypes.”

These potential therapeutic vulnerabilities include frequent mutations of PI3-kinases in EBV-positive tumors, alterations of the RhoA signaling pathway in GS tumors, and amplifications of receptor tyrosine kinases in CIN tumors. The authors suggest using these subtype-enriched mutations to guide future drug trials in distinct groups of patients.

Daniel Catenacci, MD, from the University of Chicago, who was not involved in this study, noted that despite the tendency for molecular abnormalities to occur more often in one subtype than the others, they are not mutually exclusive. Therefore the presence of specific mutations would still need to be assessed in each patient.

“Current molecular diagnostic tools screen in a high-throughput manner anyway, so all tumors should be subjected to this profiling to see what is going on in that particular tumor,” said Catenacci. “Whether or not one subtype that is HER2 amplified would derive more benefit than a different tumor subtype is currently unknown, and is a subject of future research.”


Although questions remain about the clinical utility of these new classifications, there is a need for new gastric cancer treatments, and this study suggests many new therapeutic hypotheses. “At this point, it is essential to take these ideas forward and to test specific inhibitors in genomically-defined model systems and in clinical trials,” noted Bass.

  1. The Cancer Genome Atlas Research Network. Comprehensive molecular characterization of gastric adenocarcinoma. [Published online ahead of print July 23, 2014]Nature. 2014. doi:10.1038/nature13480.