New Genetic Variants Identified in Ovarian Cancer

Paul Pharoah, MD, PhD, director of teaching for the Department of Public Health and Primary Care, University of Cambridge

Paul Pharoah, MD, PhD

According to results from a large genetic search through the DNA of almost 100,000 patients with ovarian cancer, 12 new genetic variants were found that increase the risk of developing the disease and confirmed the association of 18 previously discovered variants.

Before this study, researchers had identified 27 common variants across the genome associated with ovarian cancer risk; however, some of these were only associated with rare subtypes of ovarian cancer.

“Until now, almost all ovarian cancers were treated in the same way, it didn’t matter whether you had mucinous, serous, or endometrial—everybody got the same treatment, more or less, even based on age and fitness level. By finding those big differences in the genetic risk factors, I think that provides an incentive to try to find more specific treatments for more specific types and personalize medicine,” explained one of the study’s leading researchers, Paul Pharoah, MD, PhD.

OncLive: Can you start by giving an overview of the study?

What are the next steps concerning this research?

In an interview with OncLive, Pharoah, director of teaching for the Department of Public Health and Primary Care, University of Cambridge, discussed the significance of his research and the potential for translating genetic discoveries into meaningful clinical benefits.Pharoah: This is a study conducted by the Ovarian Cancer Associations Consortium. We've been working together for over 10 years with a primary aim of identifying genetic variants that, when inherited, are associated with an increased risk of ovarian cancer. This project was a culmination of multiple consortiums putting together hundreds of thousands of samples and cases of patients with different types of cancer. Before we started this study, we had identified 27 variants across the genome that are reasonably common and are associated with the risk of ovarian cancer. We had identified another 12, increasing the number of those variants by about 50%. There are several directions this research might take. Although, we've now identified over 30 variants associated with ovarian cancer risk, these only explain a very small fraction of the inherited component of these risks. There is a lot we don’t understand and the ongoing search is to find more of these variants in order to explain as much as possible.

Although these variants are interesting from a biological perspective, they currently have limited clinical relevance. However, I am interested in the variants and genes that can potentially have clinical relevance and are more likely to be clinically useful.

What has this study ultimately taught us about ovarian cancer?

The second thing is while we've found more than 30 variants associated with ovarian cancer, we don't have any real understanding of the underlying biology and exactly why they cause an increase in risk. If a patient carries a particular variant, their risk gain is slightly altered for ovarian cancer. There is a big research effort going around worldwide to try to work out these molecular mechanisms but it's very challenging research for all sorts of complicated reasons. However, if the research is successful, that might have important implications for either disease prevention or treatments. If we understand a mechanism, we might be able to intervene to affect that mechanism and reduce risk or improve treatment of the disease.We know that ovarian cancer is a heterogeneous disease and not all ovarian cancer is the same. What this study has shown us is that there are some big differences in the genetic risk factors for the different types of ovarian cancer. For example, we found variants that associated specifically with a high-grade serous type of ovarian cancer. This has shown us that those different types of ovarian cancer are perhaps more different than we previously thought.

The implications of that is that until now, almost all ovarian cancers were treated in the same way, it didn't matter whether you had mucinous, serous, or endometrial—everybody got the same treatment, more or less, even based on age and fitness level. By finding those big differences in the genetic risk factors, I think that provides an incentive to try and find more specific treatments for more specific types and personalize medicine.

Although we don’t really understand the biology of what causes these variants, we have some hints because some of these variants line your genes and they're genes with interesting functions. For example, one of the variants is associated with the subtype called low-grade serous ovarian cancer, which is known to affect telomere lengths. Telomere lengths is one of those things that's associated to a more general aging process and if it is, what does that mean for cancer?

What are the most significant hurdles to our understanding of the genomic background with ovarian cancer?

Some of the genes that are near these variants we think are involved in embryonic developments. That in itself is quite interesting—if the embryo and the fetus actually have a long-term impact on your future cancer risk, that is profound and important. In a sense, the major difficulty for us is that we now know that most of these genetic variants individually have small effects and we suspect there are thousands. We've found [a lot] of variants associated with risk, the reality is there are probably thousands across the genome but they all have very small effects and the real challenge is how to find things that have such small effects.

What other ovarian cancer research are you involved with?

One way to do that is to have even bigger sample sizes but that's difficult in ovarian cancer because it is not that common of a disease, making it hard to get very large sample sizes. The other way is through better understanding of the biology and as I said, we are only at the beginning of doing that.Most of my work has been on this genetic susceptibility and just running these projects is a big job. In addition, we've got projects looking at how your inherited genetic makeup also influences not just if or when you get ovarian cancer but how that ovarian cancer might respond to treatment and clinical outcomes. For example, we know that if you inherit mutations of the BRCA gene and you get ovarian cancer, that cancer responds to some of the chemotherapies differently than patients without a mutation. Again, if we were to understand why those differences happen then we might be able to target our treatments better and improve. I am interested in any genetic makeup influences and responses to treatment and how each reacts with different types of chemotherapy and whether that helps us personalize the way we give chemotherapy in the future.

Phelan CM, Kuchenbaecker KB, Tyrer JP, et al. Identification of 12 new susceptibility loci for different histotypes of epithelial ovarian cancer [published online March 27, 2017]. Nature Genetics. doi: 10.1038/ng.3826.