The Multifaceted Experience of Multigene Panel Testing for Breast Cancer

Mary Daly, MD, PhD

Today we are speaking with Mary Daly, MD, PhD, chair of the Department of Clinical Genetics at Fox Chase Cancer Center in Philadelphia, Pennsylvania and the founder and director of Fox Chase’s Risk Assessment Program, which helps clinicians speak to patients about risk perception and risk-reducing behaviors for women with a history of breast and/or ovarian cancer. Dr. Daly recently penned an editorial that discussed how best to use multigene panel testing, which provides women with information on any mutations within their germline that may increase their risk of developing breast cancer.

-Interviewed by Anna Azvolinsky, PhDOncoTherapy Network:First, what are the options for women who should be evaluated for genetic predisposition to breast cancer and how are these multigene panel tests generally utilized?

Dr. Daly: I thought I would start by giving a very short history of panel testing and how it evolved. It started many, many years ago when it was recognized that breast cancer tended to cluster in some families, leading to the hypothesis that there may be a genetic reason behind it. It wasn’t until 1994 and 1995 when the first two breast cancer genes, BRCA1 and BRCA2, were identified. Shortly after that, commercial testing became available for women to test whether they carried one of these genes. These were directed towards women with personal histories of breast cancer and strong family history. The company that started the testing actually obtained a patent for testing for these two genes and so for many years, it was the only lab that was able to do testing. About 4 years ago, there was a challenge to this and the U.S. Supreme Court ruled that no company could actually hold a patent for human DNA, which opened up the testing world to other competitive labs. In those intervening years, however, several other breast cancer genes were identified in addition to BRCA1 and BRCA2 that contributed to the risk of breast cancer. So, when new commercial testing became available after the Supreme Court ruling, it included not just BRCA1 and BRCA2 as it had in the past, but also these other genes that may contribute to breast cancer risk, which is where the name “multigene panel” comes from. Over the last 3 to 4 years, these panels have grown to include not just breast cancer genes but genes for colon cancer, gynecologic cancers, and others. There are a huge host of panels to choose from, depending on family history and what best fits any given individual. We’ve all seen families where there is a history of not only breast cancer but also colon cancer or another kind of cancer, so these multigene panels have become very popular as a means of looking at a broad array of genes all in one test. Currently, unless we know that there is a BRCA1 and BRCA2 mutation in a family and we only want to test a new individual for that particular gene, the genetic community has moved towards using these multigene tests routinely and that has led to a lot of questions. Some of these newer gene mutations we are not as familiar with, and we don't exactly know how common they are in the population. Another factor is that we are not quite sure of the level of risk, and we are not quite sure how aggressive to be in preventive options. And so studies like the one we are going to talk about are important for trying to answer and address some of these questions.

OncoTherapy Network: You wrote an editorial on the use of these multigene panel tests, which was in response to a study that was published in JAMA Oncology. Can you briefly tell us about what that study found?

Dr. Daly: Yes. The study was done in cooperation with a commercial lab that has joined the testing world, which is an advantage because they have done thousands of tests, whereas any one institution may only have tested several hundred people. Right away there is a strength to the study in large numbers. The study authors initially had 65,057 women for whom a multigene panel test had been performed in the lab and they then narrowed it down to only Caucasian women and other exclusion criteria that I will talk about shortly. But, overall, they tested these women and found that 10.2% had a mutation in one of what we call the “breast cancer genes”, but if you exclude BRCA1 and BRCA2, there were still 6.18% of women who had one of these other genes. So over half of the genes that were discovered in this test were not BRCA1 and BRCA2, which strengthens the argument that we shouldn’t just be testing for BRCA1 and BRCA2 but that there is value in looking at this broad panel of genes from which we will find additional breast cancer risk genes.

OncoTherapy Network:What do you see as the way forward for the best use of these genetic tests? And are there any trials that are ongoing that are addressing the best use of these testing tools?

Dr. Daly: I want to say a few words about this study and other studies like it because these results point out some of the limitations of this approach, which can then lead to our ideas for future studies. The way that this study was done is that they took this data sample of women from the commercial lab, excluded certain groups of women, other ethnicities except Caucasians, and they also excluded any women with another breast cancer prior to their [current] breast cancer. So these exclusion criteria were to make the sample as close as possible to a control group.

Now, in oncology, we are used to doing clinical trials where we randomize patients to one treatment or another or to a treatment and a control placebo group, and this allows us to have a tight control over the characteristics of the treated vs. nontreated or the treated with one drug vs. another. The kinds of studies that are based on populations and not clinical trials are a little harder to develop and design a good control group. So we want to know whether the numbers they found in the study are just random variations in the population or if they are truly breast cancer genes, so they look at another group of women that are available to the scientific community of a population-based cohort in whom they tested for the same genes to see what the population prevalence of the mutations in these genes were and then compare that to the prevalence of the genes in this population from the genetic commercial labels.

The study found that there was a much higher proportion of women with mutations in these genes compared with the control group for several of the genes. This introduces some limitations that somewhat lessens the strength of the study, although for a lot of these genetic studies it is really the only way to perform them right now. The authors tried to match the inclusion and exclusion criteria to the control cohort that was available for them and in some of the cases the controls were not well matched, specifically, on age and ethnicity, which could introduce a bias into the findings. Future population-based studies would help to overcome these biases. That is, rather than going into a commercial lab where people preferred for testing are already selected to be at high risk, researchers would look at a random sample in the population to determine the frequency of these mutations and how that tracks with breast cancer incidence.

We would also want to see if these mutations have specific characteristics, if the women more likely to be of a younger age, more likely to have bilateral breast cancer, or have other family members with breast cancer, etc. Obviously, those kinds of studies are very expensive to perform, but they would be ideal for answering some of these questions. I am not aware of a study like that is actually ongoing. We do have some good, large studies, however, that are trying to overcome some of these limitations. One of these is a cohort called the Breast Cancer Family Registry, which includes several thousand women with and without a history of breast cancer, some with a family history and some without, and many with a history of other cancers, either in the individual participants in the registry or in their family members. There are a lot of variables that we can tease and then look at various mutation rates, so the more studies we have, the more data we can accumulate and extrapolate from these studies and determine a consistent finding vs. a random finding.

One other useful thing that came from this particular study is that investigators were able to eliminate, at least in this analysis, a couple of genes on these multigene panels that have previously been associated with breast cancer, but in this study they were not. That is very helpful when counseling women who may have one of these mutations in a particular gene-this study eliminated certain genes as breast cancer genes, so the patients that have them may not actually be at any increased risk for breast cancer. The more of these kinds of studies that we do, we will be adding to our knowledge base for each of these various genes, and of course it will be repeated, if it hasn’t been already, for other kinds cancers such as colon cancer, prostate cancer, ovarian cancer, etc. So we are gradually accumulating a large amount of information so that we can become more and more specific about the real risk and how we should act on it.

Oncotherapy Network:Thank you so much for joining us today Dr. Daly!

Dr. Daly: Thank you.