As Home Genetic Tests Flood Market, Are Physicians Being Left Behind?

Kathy D. Miller, MD; Susan Domchek, MD


January 22, 2020

This transcript has been edited for clarity.

Kathy D. Miller, MD: Hi. I'm Kathy Miller, professor and associate director of clinical research at the Indiana University Simon Cancer Center. I'm joined by Dr Susan Domchek, executive director of the Basser Center for BRCA at the University of Pennsylvania in Philadelphia.

Susan, let's talk about where we are in genetic testing, specifically for cancer research. I know much of your work focuses on breast cancer, but we keep finding more and more genes for more and more types of cancer. Are we getting the tests to the patients who need them?

Susan Domchek, MD: That's a great question. I think it's always good to take a historical approach to this problem and to remind ourselves that clinically available genetic testing only came around in the late 1990s. In the early days, we had a setup that was consistent with how we approached genetic testing for Huntington disease, which was pretest counseling, genetic testing, and posttest counseling. We found that this model, although very effective, creates barriers. There were both perceived and actual barriers to patients in getting genetic testing, including meeting with the genetic counselor and cost, as well as general awareness.

As we learn more and more, we're expanding the population of people whom we think should get genetic testing. Currently, the guidelines say that we should test every patient with ovarian cancer, every patient with pancreatic cancer, every patient with metastatic prostate cancer, and then not quite all (though we could debate that) patients with breast cancer. That is just starting with some of our patients with cancer; we haven't even talked about those with other cancers, such as colon cancer.

We know that there's a large group of patients who should get genetic testing and have a high risk for a specific gene mutation. And yet, we don't actually test most of them. A really nice article from Allison Kurian and colleagues showed, just even a few years ago, that although we should test every patient with ovarian cancer, we're only testing about 30% of them.We have a great deal of work to do.

Miller: That's particularly disappointing, because the recommendations could not be more straightforward: If it's ovarian cancer, test. You don't even have to take a family history, look at a disease phenotype, or consider age—this cancer should be tested. With only one third being tested, that's a huge gap of patients who could benefit.

Domchek: Absolutely. We hope that now that there are very specific therapies in the first-line setting, which is new—for patients with BRCA1 and BRCA2 germline mutations, we absolutely should be giving a PARP inhibitor—that number is getting bigger all the time.

We know that we're particularly bad at genetic testing in underrepresented minorities. There are disparities in terms of socioeconomic class as well. We have to really ask ourselves, why is it hard for patients to get testing? Why is it hard for providers, and how do we make that easier? There's a great deal of research in this area, using everything from telegenetics to digital health platforms to videos in the point-of-care setting, that hopefully will make this easier. At the same time, it's really important to remember that genetic testing information, the reason we do it, is to better inform patient care. Genetic counselors are obviously a tremendous resource in terms of helping to interpret complex results or identify genes that may not be the standard genes that we might send, because there's a whole host of genetic syndromes out there otherwise. We need to be creative about the models that we use to get people genetic testing so that we expand the number of tests that we send.

'If We're Too Slow, the Market Will Fill the Void'

Miller: It is now possible for patients to just skip medicine entirely, send off a sample, and get the test results sent directly to them. What are your thoughts on the direct-to-consumer testing that is available? How does it compare with the medical tests that we would do?

Domchek: At least in the United States, we know that if we're too slow, the market will fill the void. Things have massively changed since 2013. Our technology shifted to next-generation or parallel sequencing, where we can sequence many genes at once—not only BRCA1 or BRCA2 or the Lynch syndrome genes, but as many genes as you want for about the same price. There were issues with patents, where suddenly, many companies could send BRCA1 and BRCA2 testing, and that just opened the floodgates. The United States doesn't really regulate this, so there are these direct-to-consumer opportunities. People can go around us to get this access. That can be considered a leveler, or it could result in harm. Honestly, right now, there could be a little bit of both.

There are different types of direct-to-consumer tests, one of which is 23andMe, which has specific authorization from the US Food and Administration (FDA) to test for the three common Ashkenazi Jewish founder mutations in BRCA1 and BRCA2. To be clear, they only test for three specific mutations of the thousands known about BRCA1 and BRCA2, and that information is given back to the patient. However, there is a statement made on the report that this should not be used for any kind of medical decision-making, which begs the question, what does that mean?

Miller: It also begs the question, why do it? If they have authorization from the FDA to test for those three mutations, if I was a clinician and somebody brought me that report and had a mutation of one of those three, why would I not feel good acting on it?

Domchek: It's a sticky wicket. That statement is in the language; it's not clear why, but it is. When people bring us those reports, we do get repeat testing.

As I mentioned, 23andMe has authorization to do the three common Ashkenazi Jewish founder mutations as well as a specific gene called MUTYH, which is an autosomal recessive for a polyposis syndrome, but it does not test for Lynch syndrome. So, the first point is that the 23andMe website does clearly state that they're not testing full sequencing for BRCA1 and BRCA2, but it is not at all clear that patients or providers understand that—at least that's been our anecdotal experience. If you are not of Ashkenazi Jewish descent, this testing is irrelevant. There is potential for false reassurance by not understanding what tests you actually had performed on yourself.

The second issue is that consumers can download their raw data from 23andMe and upload them into another online forum, such as Prometheus, and have their data analyzed. There have been publications suggesting that when they find mutations and other things—because it's looking at all sorts of genes using a sort of chip-based technology—about 40% of the time, mutations are not confirmed by other types of testing. We don't understand the problem there, whether it's the analysis or something else. You really have to understand what the test is doing and be extraordinarily cautious, if you're downloading your data and uploading them to another site, about what that analysis is.

Miller: There is a big potential for false-positive results of suggesting someone has a mutation.

Domchek: Right. Now, that won't be on the 23andMe report; that is only if you take your data and upload them to another site, but those are problems.

So, one type of genetic testing is 23andMe, which is truly direct to consumer. There is no physician involved. There is also a hybrid approach. There are several companies where you can go on their website and order a test. Even if you don't have a doctor ordering the test for you, they essentially will provide a doctor who will order the test for you. This is not your physician or a physician that you've met, but it enables you to order a test. Those tests actually are full sequencing and do give full results. But the question there becomes, how well do people understand what they're signing up for?

As I started with, this information is useful only when it's used in the medical context. I always encourage people, if they're going to use one of these methods, to please bring the test results back to someone involved in their healthcare. Whether it's positive or negative, it needs to be interpreted in the context of your family history. If you have negative testing on one of these direct-to-consumer tests, that doesn't mean you don't need more colonoscopies or breast MRIs. A negative test does not mean that you're off the hook in the setting of a strong family history.

Miller: For the hybrid model, do they make sure a physician's name is on the test results that are sent back directly to the consumer?

Domchek: Yes. The physician who signed my test—because I did do one of these—was an emergency department physician in Pittsburgh. I did not have a discussion with that provider.

Miller: I'm interested in the liability question. If I am moonlighting as the physician who signs off on these test orders, and a result comes back that is misinterpreted and leads to decisions that are not appropriate, do I have any responsibility for that?

Domchek: I don't know the answer to that question. It is a point I have raised, and I know that some of my colleagues are digging into that a little bit more. These companies, I'm sure, have lawyers who have investigated those sorts of things. It is an interesting issue. There are other laws too, including the fact that we can't practice medicine across state lines. There's a whole host of regulations. But, this idea of the virtual provider, I think, is one that's coming up, not only in genetics but also in many different capacities. What is the definition of a provider? I don't know the answer to that question.

Barriers and Opportunities

Miller: I want to ask you about some of the barriers for testing in a medical setting, and particularly for some of the insurers who require genetic counseling by people with specific training before they will provide coverage for a test. I think genetic counseling is hugely important, but that's a huge barrier.

Domchek: Fortunately, there aren't many right now who have that provision, which is good, because I think we should be talking about pretest education and moving away from the phrase "pretest counseling."

It also very much depends on the situation.

If you're an individual with metastatic pancreatic cancer and you're trying to get this information quickly to make medical management changes, then you can have a pretty short list of bullet points. For instance, we use a 7-minute video; that might even be too long for some individuals. You need to know the basics—this might change your care, and it may have implications for your family.

If you're a 25-year-old woman whose mother has a BRCA1 mutation and you're at 50% risk, I think you need a full counseling session about how this may affect you and your decision-making, because the more information you have going into this, the better.

We need multiple models, because different models are going to be right for different people.

I think what's gotten particularly complex lately is the fact that we have so many genes and there are some genes that are messy, such as CDH1, which is associated with hereditary stomach cancer, in which case we advise people to have their stomachs removed. That is something that people probably should hear about before they get their testing, because we absolutely have had people with no family history of stomach cancer who have a mutation in that gene and have to make unanticipated and difficult decisions.

People should know what test is being sent. We can't go gene by gene, but we can cluster the genes just to give people a sense of what it's about. I think that we have a large amount of work to do over the next few years.

Miller: We will get you back sometime soon to talk about how that work is going and whether we're making progress at getting testing to more of our patients who need it.

Domchek: As there are more therapeutic implications, I do think that we will get this done. Physicians and oncologists have always been good in that regard. If we know that we need this information to change the therapy, we're generally pretty good about figuring out how to get it done.

Miller: Thank you for joining me today. It's been a fantastic discussion about such an important topic that is changing almost every day.

Kathy D. Miller, MD, is associate director of clinical research and codirector of the breast cancer program at the Melvin and Bren Simon Cancer Center at Indiana University. Her career has combined both laboratory and clinical research in breast cancer.

Susan Domchek, MD, is executive director of the Basser Center for BRCA at the University of Pennsylvania in Philadelphia. She is a nationally recognized expert in the translation of genetic testing for cancer susceptibility into clinical care of patients in the areas of prevention, early detection, and cancer treatment.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.