Ellen Goode and her colleagues at the Mayo Clinic found a surprise while studying the human genome: four chromosomal locations with mutations that could lead to an increased risk of ovarian cancer. But the biggest surprise was a number of SNPs on chromosome 8 that seem to act through a different mechanism than other polymorphisms to influence cancer risk. Genome Technology's Christie Rizk spoke with Goode about her study, which appeared in Nature Genetics in October.
Genome Technology: What can this research tell us about ovarian cancer risk?
Ellen Goode: Alone, this research won't predict who will get ovarian cancer, but in combination with additional risk SNPs, eventually we could more accurately test for risk, and help higher-risk women with early detection and prevention.
GT: You found several variations on four chromosomes that could lead to an increased risk for ovarian cancer. Was there any connection to risk for other kinds of cancer?
EG: One of the regions is chromosome 8q24, which includes the MYC proto-oncogene, and in that region we found a SNP that is associated with risk of ovarian cancer. But in fact, there are other different SNPs in that region associated with risk of other cancers including breast cancer, colon cancer, prostate cancer, and bladder cancer. What's interesting is that it's not the same SNPs that are associated with each of these cancer risks. For ovarian cancer, it's a SNP over 700 kb 3' of MYC, and this is quite far from the SNPs 5' of MYC associated with other cancers. What we think may be happening is that these SNPs are acting upon MYC, and [the] regulation of this gene might be leading to modification of cancer risk. We don't yet know why different SNPs relate to different cancers, only that there's definitely a cluster of cancer associations in 8q24, and we think it's related to MYC.
GT: What are your plans for further research in this area?
EG: In the regions we've identified, we're going to be doing fine-mapping — meaning we're going to look at additional, denser SNPs in these regions — because we don't expect that each SNP that we've looked at is the causal variant, but that there's an additional causal variant correlated with it. We're going to go in closer within each of these regions. We're also going to look at interactions to see if any of these associations is modified by an environmental factor, or if any of these genetic loci interact with other genetic loci, and we will of course be doing functional experiments. Finally, we're going to try to see if there are other hits out there that maybe we couldn't detect because of [the] initial sample size of this study, so we have increased the number of samples with genome-wide markers and have more replication [studies] underway. Of course, all of this is only possible due to our very collaborative interdisciplinary consortium.
GT: Will this discovery eventually change the way that ovarian cancer is detected?
EG: What this will do eventually is enable identification of women at highest risk of ovarian cancer. For example, a woman with a family history could go to the clinic, be screened for mutations in the genes BRCA1 and BRCA2, and if those are negative, be screened for a panel of these types of markers. That wouldn't then predict who has ovarian cancer, but it would predict who we should be looking at more closely. It's an earlier detection effort; it doesn't mean they have the disease. It means they might be at increased risk and we have to target them for prevention efforts and early detection efforts.