NEW YORK (GenomeWeb News) – In a paper appearing online today in PLoS Genetics, an international research team described their pathway-based gene expression approach for classifying gastric cancers.
Researchers from Singapore, the UK, Korea, and Australia used gene expression profiling and computational approaches to map signaling pathways in hundreds of stomach cancer tumors. Their study highlighted three main pathways that contribute to cancer function in the majority of stomach tumors tested. By looking at which of these pathways were activated in combination, the team was also able to classify the tumors and identify ties to patient outcomes.
"We identified three oncogenic pathways that were activated in over 70 percent of the gastric tumors we examined," lead author Chia Huey Ooi, a post-doctoral researcher at Duke-NUS Graduate Medical School in Singapore, said in a statement. "We also found that combinations of these pathways are significantly related to patient survival."
Stomach cancer is the one of the most common causes of cancer death around the world. The National Cancer Institute estimates that there will be more than 21,000 new stomach cancer cases and 10,600 deaths in the US this year. The disease is particularly common in Korea, China, Japan and other parts of Asia, where it usually goes undiagnosed until it has reached an advanced stage, the team noted.
Studies of gastric cancer so far suggest that it is very heterogeneous, varying widely in its cellular features and treatment response.
To better understand and classify gastric cancer, Ooi and colleagues assessed expression patterns in cancer samples and cell lines using Affymetrix Human Genome U133 Plus 2.0 and HG-U133A arrays. But rather than focusing on individual genes, the team used computational approaches to look for cancer-related signatures affecting entire pathways.
After validating this method in 51 breast cancer cell lines, the team turned their attention to gastric cancer, evaluating 11 potentially cancer-related pathways in 301 tumor samples from patients in Singapore, the UK, and Australia.
Overall, the researchers identified three pathways that are altered in some 70 percent of the gastric cancer samples they tested: a nuclear factor-kappa B pathway, a Wnt/beta-catenin pathway, and a proliferation/stem cell pathway.
The team then used their pathway prediction algorithm to screen a panel of 25 more stomach cancer cell lines. Indeed, they found, the cancer-related pathway clusters in the cell lines overlapped with those identified in the primary tumor samples.
"[T]he ability to perform 'high-throughput pathway profiling' opens up a number of interesting possibilities," senior author Patrick Tan, a researcher affiliated with Duke-NUS, the Genome Institute of Singapore, the National Cancer Centre, and the Cancer Science Institute of Singapore, said in a statement. "It suggests that pathway combinations, rather than single pathways alone, may play a more critical role in influencing tumor behavior."
Next, the team looked at how the pathways they identified related to gastric cancer proliferation and patient outcomes. While they did not see differences in outcome linked to individual pathways, the researchers found that they could glean patient survival information when they looked at the cancer-related pathways in combination.
For instance, they found that patients whose tumors had both elevated NF-kappa B and proliferation/stem cell pathway expression had significantly shorter survival times than those with lower activation in these two pathways — a prognostic indicator that was independent of tumor stage.
"[T]hese identified sub-groups are clinically relevant in predicting survival duration and may prove useful in guiding the choice of targeted therapies designed to interfere with these molecular pathways," the researchers wrote.
Along with the insights into classifying gastric cancer, researchers noted, a similar approach may also prove useful for profiling pathway activation profiles and their relationship to outcomes in other types of cancer.