In His Own Words...
Thanks to the seminal work of BCRF grantee Mary-Claire King, PhD (University of Washington) and others, in the 1990s, we know that mutations in the BRCA1 gene -- breast cancer susceptibility gene 1 -- dramatically increase the risk of developing breast cancer. While our knowledge of the biology of this gene and its link to breast cancer has advanced greatly, researchers and clinicians continue to wrestle with the challenge differentiating those with the BRCA1 mutation who are at greatest risk of developing breast cancer versus those whose risk is lower. Approximately 35% of women with BRCA1 mutations will never develop breast cancer.
In a series of publications released in spring 2013, BCRF grantee Fergus J. Couch, PhD (Mayo Clinic) reported research findings that can quickly improve doctors' ability to determine accurately the likelihood of a woman with BRCA1 mutations developing breast cancer, through a simple test. Dr. Couch, whose work has been supported by BCRF since 2007, speaks about his findings and how they can impact future research and treatment.
Our study is trying to identify genetic modifiers of breast and ovarian cancer risk in BRCA1 mutation carriers. BRCA1 mutation carriers have on average a 65% lifetime risk of developing breast cancer, whereas the general population's risk is about 12%. As these numbers suggest, about 35% of BRCA1 mutation carriers never get cancer. We know that the age of onset ranges from very young -- from 20s and 30s -- to older, 60s and 70s. Our overarching question is whether there are other genetic modifiers that can control the age of onset or indicate whether individuals get cancer or not.
Our thought was to do a genome-wide association study (GWAS), which is an examination of many common genetic variants to see if any of them is associated with a trait. A breast cancer GWAS involves a very large number of breast cancer patients and their matched controls, meaning individuals who are generally from the same geographical area and perhaps the same age as the women with breast cancer. We look at the frequency of single nucleotide polymorphisms (SNPs), which are specific sites in the DNA that vary across the population (many millions have been identified). When a SNP is found to be much more common in the individuals with breast cancer cases than in the individuals with no evidence of cancer, then it generally means that the SNP is associated with breast cancer in some way and may even be considered a risk factor after further validation.
Risk Stratification for BRCA1 Mutation Carriers
We proposed to BCRF a number of years ago to do a GWAS on BRCA1 mutation carriers in collaboration with the Consortium of Investigators of Modifiers of BRCA1/2 (CIMBA), which is comprised of 60 groups around the world. Our plan was to look at the SNPs of BRCA1 mutation carriers who have developed breast cancer and those of BRCA1 carriers who have not, while also considering the age of the individual. We started with 2,500 BRCA1 mutation carriers and looked at about 500,000 SNPs from all over the genome. From that study, we came up with a single clear modifier of breast cancer risk in BRCA1 mutation carriers on chromosome 19p. This region has now also been associated with risk of ovarian cancer. From there, we expanded the study by combining forces with the Collaborative Oncological Gene-environment Study (COGS), a big international effort to study genetic risk factors for breast, prostate, and ovarian cancer that involved over 200 research groups. We ended up genotyping 16,000 BRCA1 mutation carriers with on iCOGS, a tool containing over 200,000 selected SNPs including about 32,000 from our original BRCA1 GWAS.
Analyzing our data and conducting additional studies, we came up with two new breast cancer regions of the genome, often called loci, that contain modifiers for breast cancer BRCA1 carriers. We also found two separate ones for ovarian cancer, which suggests that modifiers for breast and ovarian cancer in BRCA1 mutation carriers are very different from one another. One of the two ovarian modifiers turned out to be unique, in that it is not associated with sporadic ovarian cancer or BRCA2 ovarian cancer and has not been linked to any other form of cancer. In ongoing studies that involve combining results from the BRCA1 studies with results from triple negative breast cancer studies (see below) and estrogen receptor negative breast cancer GWAS, we have identified three more regions that influence breast cancer risk in these populations and do not appear to influence risk of other subtypes of sporadic breast cancer. In addition, we have checked whether other known genetic risk factors for sporadic breast and sporadic ovarian cancer influence risk of these cancers among BRCA1 mutation carriers and were able to confirm the involvement of a small number.
In total now, we have reported ten genetic risk factors for breast cancer and six genetic risk factors for ovarian cancer in BRCA1 mutation carriers. Using this information, we then moved to predicting the risk of cancer in individuals -- personalizing or individualizing risk estimates -- as opposed to just telling every BRCA1 mutation carrier that she has a 65% lifetime risk.
Looking at these data, we were able to identify groups with lifetime risks for breast cancer as high as 90% and as low as 28%, where it used to be 65% for everybody. For ovarian cancer, the lifetime risk ranges from 10% up to 90%. As results from our ongoing studies are incorporated into these analyses we expect to continue to improve our ability to individualize risk for BRCA1 mutation carriers. This should help these women make decisions about how they might manage their risk of cancer. They can decide whether to pursue prevention strategies -- be it watchful waiting, changing their diet and exercise, or undergoing prophylactic surgery -- and at what age. Every woman can change and can pick her own way of minimizing her cancer risk. Our work tries to help women make informed decisions by providing them with this information.
Much of this work has been done through CIMBA in parallel, with the BRCA2 GWAS work of Kenneth Offit, MD at Memorial Sloan-Kettering Cancer Center, who is also a BCRF grantee. Dr. Offit's work has resulted in the identification of several genetic risk factors for BRCA2 mutation carriers, many of which turn out to be different from the BRCA1 risk factors. Being able to compare notes and do these studies in parallel been informative for both of us.
Triple Negative Breast Cancer
In parallel, with the help of BCRF, we extended our studies into triple negative breast cancer, because about 75% of the tumors in BRCA1 carriers are of the triple negative subtype. This disease subtype accounts for about 15% of all breast cancers and is very aggressive. Because it lacks the three markers (estrogen, progesterone, and HER2) that current therapies target, triple negative disease cannot be treated with tamoxifen, aromatase inhibitors, or agents like trastuzumab. Triple negative patients are treated with chemotherapy, which can have significant side effects. It is important to figure out as much as we can about the genetic factors that influence development of triple negative disease so that we can have more focused prevention and/or therapy in the future.
We organized an international consortium of about 20 groups, called the Triple Negative Breast Cancer Consortium (TNBCC). We gathered blood samples from 5,000 triple negative patients and unaffected controls and did a triple negative GWAS. We have now identified three loci that only influence the risk of triple negative breast cancer and have no influence on other subtypes of sporadic breast cancer. We also looked at all of the known genetic risk factors from general breast cancer and were able to show that about 35 loci seem to have an influence on triple negative breast cancer. Through these studies we are hoping to explain how triple negative disease actually begins. Anything we can do to understand the risk of developing a triple negative, or even find genes that are involved in the development of triple negative disease, could benefit patients.
All of our studies are ongoing. In fact, there is a new initiative, in collaboration with the National Cancer Institute supported GAME-ON Consortium, to design the "OncoChip." This new genotyping array will contain about 600,000 SNPs that have been selected because of possible involvement in BRCA1 and BRCA2 breast and ovarian cancer, and sporadic forms of breast, ovarian, prostate, colon and lung cancer. The intent is to screen, working with Dr. Offit, as many as 20,000 BRCA1 mutation carriers and 10,000 BRCA2 mutation carriers using the new chip, with the goal of finding new risk factors. We think that by genotyping more samples and working with more SNPs, we can find many more of these risk modifiers. Some estimate that there are as many as 1,000 risk factors for sporadic breast cancer in the human genome. Going beyond that, we will be able to integrate the genetic risk modifiers with more classical risk models, such as breast density, exposure to estrogen, the number of children that people have, and diet and exercise. All of these various components can be combined with genetic discoveries to make an even more informed risk model to benefit patients.
In separate studies we are also involved in the hunt for new high and moderate risk breast cancer predisposition genes. In collaboration with Dr. Offit, and also with Katherine Nathanson, MD and Susan Domchek, MD at the University of Pennsylvania, who are also BCRF recipients, we are using exome and whole genome sequencing of high-risk breast cancer families to identify new breast cancer genes on identifying additional breast cancer genes and additional mutations that might be driving cancer cases in high-risk families.
Role of BCRF Funding
BCRF funding has played a very, very important role in these studies. My lab has always been focused on breast cancer genetics, following my involvement in the searches for the BRCA1 and the BRCA2 genes while I was training with Francis Collins and Barbara Weber. A number of years ago, when SNP association studies were just beginning, my lab wanted to expand into the GWAS field. We knew that we could do this BRCA1 study and had access to the samples needed to do this study, but we had a hard time getting the work funded. We approached BCRF and they had the foresight to support this really important study.
If we had not received the funding from BCRF, we and CIMBA would not have been able to do any of this work. Now, a few years later, we have made some tremendous discoveries that we think can really help patients make clinical decisions. BCRF is to be congratulated for that vision, in terms of seeing what was important and what would actually work out downstream. Continued support now will allow us to do this OncoChip study, where we think we can further answer the question of risk stratification, find more of these genetic risk modifiers, develop important risk models, and better understand the biology behind all these risk modifiers that are changing the risk of breast and ovarian cancer in BRCA1 and BRCA1 mutation carriers.