Cancer gene found hiding in plain sight

A mutation that may be driving as many as 20 percent of endometrial and colorectal cancers has come to light this week, thanks to a study by researchers from the Broad Institute of MIT and Harvard and Dana-Farber Cancer Institute. The researchers describe finding the connection between the gene...

RNF43
RNF43

A mutation that may be driving as many as 20 percent of endometrial and colorectal cancers has come to light this week, thanks to a study by researchers from the Broad Institute of MIT and Harvard and Dana-Farber Cancer Institute.

The researchers describe finding the connection between the gene RNF43 and these cancers earlier this week in Nature Genetics.

At first the team found it hard to believe that previous large-scale, comprehensive analyses could have missed such a significantly mutated gene, so they decided to reexamine past data. They found that RNF43 was in fact frequently mutated in previous studies, but had been unreported.

The authors of the Nature Genetics paper suspect that these mutations were left out because of where in the genome they occur – in microsatellites, which are spans of repetitive sequence that are often misread by genome sequencers. Algorithms that are used to analyze sequencing data account for these common errors by filtering them out.

“Because these microsatellites are a frequent source of sequencing errors, it makes it much harder to detect real events that happen within these parts of the genome,” said Eran Hodis, a co-first author of the study, “but once we spent a little more time looking at this, it was clear that these mutations were not at all an artifact of the sequencing.”

These mutations in RNF43 affect the Wnt signaling pathway, which has long been associated with cancer growth, through a mechanism that was only recently identified. Serendipitously, drugs targeting that mechanism are already in clinical trials The authors say the finding opens the door to testing these drugs in cancer patients who have mutations in RNF43.

“This study shows there’s a value to revisiting what we think we know,” said co-first author Marios Giannakis.

“At the Broad, we are always refining our algorithms, and we’re continuing to do sequencing. It was the sensitivity of these new algorithms, combined with the opportunity to examine the data with a closer eye, that led us to identifying this important gene,” Giannakis said.

Hodis is a graduate student at Harvard Medical School and MIT and works in the labs of Broad core member Aviv Regev and Broad senior associate member Levi Garraway. Giannakis is a postdoctoral fellow in the labs of Garraway and an instructor at the Gastrointestinal Cancer Center lead by Charles Fuchs of Dana-Farber. Garraway, who is also an associate professor of medicine at Harvard Medical School and Dana-Farber Cancer Institute, and Fuchs were senior authors of the paper.

Other Broad researchers who contributed to the work include Xinmeng Jasmine Mu, Joseph Rosenbluh, Kristian Cibulskis, Gordon Saksena, Michael S. Lawrence, Eli Van Allen, Bill Hahn, Stacey Gabriel, Eric Lander, and Gad Getz.

Paper cited:

Giannakis, M et al. RNF43 is frequently mutated in colorectal and endometrial cancers. Nature Genetics. Oct. 26, 2014. DOI:10.1038/ng.3127