Scientists discover new DNA regions associated with three cancers

A massive gene-hunting effort involving hundreds of scientists has identified 74 newly discovered regions of DNA that are associated with breast, ovarian and prostate cancers — diseases that strike about half a million Americans every year.

The international project, known as the Collaborative Oncological Gene-environment Study, or COGS, nearly doubled the number of genetic markers known to be linked with the three cancers, scientists reported Wednesday. Their findings could lead to more effective ways to screen, study and treat these diseases.

“While these papers may not be what the average person is looking for — e.g., a cure for cancer — they have important near-term value” for medical researchers, said John Witte, an epidemiologist at UC San Francisco and co-author of a commentary on the work that appeared in Nature Genetics, one of five journals that published 13 studies detailing the COGS results. Witte was not a member of the collaboration.


For the last decade or so, cancer researchers have been working hard to ferret out the DNA changes associated with cancers and other common afflictions. They examine the DNA of large groups of people with and without certain diseases to see if any genetic variations can be linked to particular conditions.

One hope has been to pinpoint the people at highest risk for developing health problems. Another has been to identify the individual genes that cause illness in the first place so that scientists can better understand how those genes cause harm and use that information to develop better treatments.

But the work has been difficult because many of the variations linked to diseases are turning out to be more rare than scientists had originally thought, said Harvard Medical School genetics professor Raju Kucherlapati, who was not involved in the COGS group.

By examining hundreds of thousands of locations on the genome in 250,000 people with and without cancer, the COGS study was able to reveal more associations than smaller studies could. In addition to confirming many genetic markers that had already been found, scientists discovered 41 new regions linked with overall risk for breast cancer, three associated with overall risk for ovarian cancer and 23 tied to overall risk for prostate cancer.

Researchers also found additional variations linked to subtypes of breast and ovarian cancer, and noted that 16 of the new regions associated with prostate cancer were linked to aggressive forms of the disease.

Generally, the risks associated with the genetic changes were modest — about 10% to 20% more than the small risks faced in the general population.

To put that in perspective, Kucherlapati said, more than half of women who have the most dangerous mutations in the BRCA1 and BRCA2 genes will develop breast cancer during their lives.

But relatively small increases in risk can add up to become significant, said geneticist Rosalind Eeles of the Institute for Cancer Research in Sutton, England, co-author of a study in Nature Genetics that detailed the prostate cancer findings.

In Eeles’ paper, men who fell in the top 1% of risk were nearly five times more likely to develop prostate cancer than the rest of the men in the study. (According to the National Cancer Institute, 16% of American men will be diagnosed with prostate cancer in their lifetime.) Perhaps targeting such men for additional tests could improve screening, Eeles and her co-authors wrote.

Scientists said that looking for the small genetic changes — known as single nucleotide polymorphisms, or SNPs — in people who are already at high risk for disease could help physicians’ early detection efforts.

“If you have a family history, the SNPs can improve the risk estimate,” Eeles said, adding that much more research is needed before SNP screening would be ready for use.

In the meantime, both Witte and Kucherlapati said the COGS findings could help scientists study the genes that cause cancer, as well as other biological aspects of the disease.

The new data also underscore how much scientists still haven’t figured out about how the genes we inherit contribute to our cancer risk.

For instance, it is thought that there are more than 1,000 genetic regions associated with breast cancer. Yet only 76 common variants have been identified so far, University of Cambridge cancer researcher Douglas Easton and co-authors wrote in one of the studies published in Nature Genetics.

In all, the scientists said, about a third of the variants believed to be associated with the three cancers had been discovered.

What explains the remaining two-thirds is “obviously” the $64-million question, Easton said.