Nearly 100 Genetic Variants Linked to Colorectal Cancer Risk

By Will Boggs MD

December 28, 2018

NEW YORK (Reuters Health) - Identification of 40 new genetic risk variants brings the total to nearly 100 variants that influence the risk of colorectal cancer, researchers report.

"We have now identified close to 100 genetic risk loci with about one-quarter harboring more than one independent risk variant," Dr. Ulrike Peters from Fred Hutchinson Cancer Research Center, in Seattle, told Reuters Health by email. "While each genetic risk variant has a relatively weak effect, if we combine these variants into a polygenic risk score, we can identify people at high and low risk. Those at high risk can be tailored for early screening or chemoprevention."

Genome-wide association studies have identified more than 50 loci related to sporadic colorectal cancer (CRC), but most genetic factors contributing to CRC risk remain undefined.

Dr. Peters and an international team of more than 200 researchers used results from some 125,000 individuals to expand the catalog of CRC risk loci and improve our understanding of rare variants, genes and pathways influencing sporadic CRC risk and risk prediction.

Overall, they identified 40 new signals, including the first rare variant signal protective for sporadic CRC near genes CHD1 and RGMB, and replicated 55 previously reported signals, they report in Nature Genetics, online December 3.

The researchers hypothesize that the rare allele confers a protective effect by lowering CHD1 expression, which is required for growth in cancer cells driven by the loss of the tumor suppressor gene PTEN, although they cannot rule out involvement of RGMB.

The new signals implicate lower-frequency variants, Krueppel-like factors, Hedgehog signaling, Hippo-YAP signaling, long noncoding RNAs and somatic drivers and support a role for immune function.

Heritability studies suggest that many rare and common variants have yet to be identified, as all the variants identified so far explain only about 20% of the variation in susceptibility to CRC.

"Another important implication of our findings lies in the potential for drug-target discovery," Dr. Peters said. "Each of the 100 risk loci points to specific genes, some of which can be targeted for new drug discovery. It has been shown that success rate for developing new drugs is substantially increased when based on human genetic findings like these."

"We have made substantial progress in understanding the genetic architecture of colorectal cancer, demonstrating that colorectal cancer is polygenic and that these findings can be used for personalized medicine," she said. "We still must do substantially larger experiments that also include whole genome sequencing to discover most genetic risk loci for colorectal cancer, which in return will provide a more complete picture of the underlying pathways and genes involved and will further improve our polygenic risk score."

"Our study was predominantly conducted in participants of European descent, which hinders the application of the polygenic risk score to non-European populations," Dr. Peters said. "Furthermore, adding more diversity to the study population will allow us to identify more genetic risk loci."

"We are currently working on a grant application to recruit diverse colorectal cancer cases and controls," she added. "This will allow us to not only improve discovery but also enable us to develop unbiased polygenic risk scores that can be applied across all major U.S. populations."

Dr. C. Richard Boland, Jr. of the University of California, San Diego, in La Jolla, who was not involved in the research, has urged caution in interpreting reports of "new" cancer genes and inherited CRC risk. He told Reuters Health by email, "This represents excellent basic research, but it will probably take a long time for any of this to reach an actionable state. If you have enough samples and analyze them enough, something will turn up as being statistically significant. So, it's not surprising that, using enough samples, they found some new risk loci."

"The authors point out that many of these risk loci are 'nearby' some cancer-relevant gene, but it would be interesting to know if you threw dice to choose genetic loci at random, how many of these would be 'nearby' some interesting gene," he said. "We need to do this work, but it is not useful clinically. Someday, maybe."


Nat Genet 2018.