Genetics & Cancer Research with Dr. Kevin Hughes

We inherit lots of things through our families. Sometimes they’re visible, like the color of our eyes or hair, and other things might not be so apparent, like how you react to caffeine or how athletic you are. These traits come from genes passed to us from our parents – we inherit two copies of each gene, one from each parent, and they act as the blueprints for our bodies.

The genes we inherit can also raise our risk for some kinds of cancer. All cancers are caused by mutations in our cells, and most of the time those mutations happen simply because of aging, or because of lifestyle choices like smoking. But sometimes - in about 5%-10% of cancers - mutations are handed down through families. If you have one of these inherited mutations, you can minimize your cancer risk by following established health care guidelines and taking a proactive approach to your health.

In this episode of Science Never Sleeps, we’re joined by Dr. Kevin Hughes, the Director of Cancer Genetics at the Hollings Cancer Center and the McKoy Rose, Jr., M.D. Endowed Chair in Surgical Oncology in the College of Medicine at MUSC. Dr Hughes is recognized nationally and internationally for his expertise in breast cancer, breast disease management, genetic testing and the identification and management of patients with hereditary breast cancer risk. His research focuses on developing tools that make cancer genetic testing simple, safe, and efficient.

Read The Transcript

Correction at the 11:40 minute mark: the statistic should be 1-2 people out of every 100 will have be identified with a gene variant at the population level.

Clarification At the 12:08 and 26:00 minute marks: Once a participant is positively identified, they are offered a free genetic counselling appointment and from there are able to follow up with the Hereditary Cancer Clinic if they desire or may seek follow up care elsewhere.

[00:00:07] Gwen Bouchie: From the Medical University of South Carolina, this is Science Never Sleeps, a show that explores the science, people, and stories behind the scenes of biomedical research happening at MUSC. I'm your host, Gwen Bouchie. We inherit lots of things through our families. Sometimes they're visible, like the color of our eyes or our hair, and other things might not be so easy to see, like how you react to caffeine or how athletic you are. These traits come from genes passed to us from our parents. We inherit two copies of each gene, one from each parent, and they act as the blueprints for our bodies.

The genes we inherit can also raise our risk for some kinds of cancer. All cancers are caused by mutations in our cells, and most of the time those mutations happen simply because of aging or because of lifestyle choices, like smoking. But sometimes, in about 5 to 10 percent of cancers, mutations are handed down through families. If you have one of these inherited mutations, you can minimize your cancer risk by following established health care guidelines and taking a proactive approach to your health.

In this episode of Science Never Sleeps, we're joined by Dr. Kevin Hughes, the Director of Cancer Genetics at the Hollings Cancer Center, and the McCoy Rose Jr. MD Endowed Chair in Surgical Oncology in the College of Medicine at MUSC. Dr. Hughes is recognized nationally and internationally for his expertise in genetic testing and the identification and management of patients with hereditary cancer risk. His research focuses on developing tools that make cancer genetic testing simple, safe, and efficient. Stay with us.

[00:01:51] Bouchie: Dr. Hughes, welcome to Science Never Sleeps.

[00:01:53] Kevin Hughes, M.D.: Thank you, nice to be here.

[00:01:54] Bouchie: You have enjoyed an incredible career prior to joining MUSC. You spent 20 years at Mass General. You're a professor emeritus at Harvard Medical School. Tell us a little bit about how you got started in research and how you ended up here at MUSC.

[00:02:12] Hughes: Oh, well, I trained in surgical oncology and worked in the liver surgery and breast surgery and other surgeries of cancer origin. During the time of my work, it was always interesting to me to know more about what was going on. There's a lot to learn in cancer. There still is a lot to learn in cancer, but when I started out, there was really a lot to learn. So doing research was just a way to learn more about these areas, to try to advance the field, to try to make treatment easier for patients and more effective. And you can't do that without research. So, I got involved in it in that way. And I was at Mass General for 20 years, as you had said, and the opportunity arose at MUSC, and this is a great institution, and I was happy to come down here.

[00:02:56] Bouchie: We are so glad to have you here, for sure. So tell us, you talked about how much we currently know and still need to learn about cancers. And one of the things that we really have been able to understand a little better in the last few decades is the role of genetics in cancer. Can you talk a little bit about what we do currently understand about the role of genetics and the role that they play in cancer and cancer diagnosis?

[00:03:23] Hughes: Sure, we've known for a long time, 100 plus years, that cancers can run in families. But it was never quite clear why it was running in families that way. As the discovery of the DNA structure in 1952 by Watson and Crick made it available that we now knew that there was a structure to DNA. And once you knew the structure, you could start to understand how it functioned and then we knew there were these things called genes, but we didn't really have a good concept of what they really were. But once you have the structure of DNA, once you understand what the code was of life essentially, people began to study the families that were very affected by cancers and then to identify specific genes in these cancers that caused this family accumulation.

And over the years we've now identified at least 80 plus genes that we know that if they're not functioning properly the patient's at higher risk of cancer. And these genes are passed down, there are mutations in the genes that go from generation to generation. Each time there's a 50-50 chance that the child would get one of these bad genes from one of the parents. And knowing that there are these genes, knowing that they run in families, studying these families with these mutations, it then became apparent that if you knew which mutation was in a family, you knew which mutation was affecting a given member of that family, you knew what cancers that gene could cause, you knew how often those cancers occurred, that opened up the possibility of prevention or finding the cancers earlier.

So, we've learned what these genes are, we've learned what their spectrum is, how many types of cancers they can cause, and each one causes a different set of cancers. We know how much they affect the family and then can be, we can know how aggressive to be. For one that causes a 90% risk of cancer, we're extremely aggressive. For one that causes a 20% risk of cancer, it's a bad risk, but we're not quite as aggressive. So, we've learned to modify how we manage patients, our strategy for those patients, based on which gene it is and how penetrant it is and what types of cancers it causes. Genetic testing used to be extremely expensive, now it's cheap. So, with it being so inexpensive, with people understanding what's going on, knowing what we can do for these patients, it's become almost a commodity at this point in time. Genetic testing has become easy to do, it's inexpensive to do, patients are much less afraid of it, doctors are much less afraid of it, thank God. So, it's just become something, it's a normal part of medical care at this point in time.

[00:05:52] Bouchie: Right, right. And I want to point out something that you said, which I think is important and we can continue to talk about, which is that to find out that you have one of these genes is only looking at risk. It doesn't mean anything necessarily. It's not a diagnosis of cancer.

[00:06:12] Hughes: Absolutely correct. So, if you have a mutation in these genes, your risk of cancer is higher. Your risk of specific cancers is higher based on the gene. How high your risk is is based partly on the gene as well. But it does not mean you will get cancer. We treat all patients with mutations as if they might get cancer, and therefore we make sure we screen them very carefully. In very high-risk situations, we sometimes remove the ovaries or remove the stomach or other things if we have to, but that's unusual. Usually, it's in more intensive screening for the cancers that we wouldn't be screening for otherwise or not screening as intensively. So, knowing this information is extremely valuable for the patient, it's lifesaving, and the more we do it, the better off we're gonna be.

[00:06:55] Bouchie: And that's really where the Hereditary Cancer Center comes in. So, you know, you talked about this being a familial, it's a family situation because we're inheriting these things from parents. So, the Hereditary Cancer Center looks at these genes that can be passed on by parents. So, can you talk a little bit about the role of the Hereditary Cancer Clinic at Hollings Cancer Center?

[00:07:20] Hughes: Sure, so genetic testing, even though it's been around commercially for 25 years, it's not been used very effectively or in large enough numbers. So, a large number of patients never found out they had a mutation. And we were very invested in figuring out how to test patients and how to set up systems to test patients, how to find patients with mutations. We didn't really give enough thought to what do we do once we find them. So, we're one of the first in the country, there are about a half a dozen others, where we see the patient with a mutation and then figure out what is the right strategy for that specific patient. And then institute that strategy and then follow them over time to make sure that the strategy is followed.

So this gives us the ability to learn more about how well our guidelines are working, to make sure patients do what they're supposed to do relative to the guidelines and then minimize the morbidity and mortality of cancer. That's the whole approach. There are a lot of individual clinics for individual genes. So there might be a neurofibromatosis clinic that would follow patients with that problem. There's what's called a Von-Hippel Lindau clinic that follows people for kidney cancer or brain tumors. There are adenomatous polyposis clinics. Those follow patients for colon cancer risk. But there aren't, now you say it, there's maybe half a dozen clinics that just take anyone with a mutation in any of 80 plus genes and then set up a strategy specific to that patient and that's something I think is gonna become very common, we're trying to set the pattern for how to do that.

[00:08:50] Bouchie: And this is really through a team of care providers, you know, you yourself are a surgeon who would look at that portion of it, but we have folks like genetic counselors who are playing into this, who are fully helping individuals and families understand their risk and what that means for the next steps that they'll take throughout their lives.

[00:09:11] Hughes: Absolutely, yeah, this is certainly a team sport. So, the way the clinic is set up that if a person has a mutation, they see the genetic counselor, they understand more about the mutation, we start to talk about getting their family members tested because it's very important to do that. And then once the patient somewhat understands what the mutation is and what it does, they then see either myself or Jen Diaz, who's a nurse practitioner, in the hereditary cancer clinic as an extension of it, where they then get, again, the strategy specific to them. And the... Every time they come into the clinic, we remind them to get their family members tested.

Then we have a network of physicians very interested in hereditary cancer in each of the specialties, neuro-oncology, in colorectal surgery, in dermatology, et cetera. We just have doctors in each area who are very good at this particular problem. And our job in the center is to make sure the patient gets to the right doctor, gets the right management and then we follow them over time to make sure they don't fall off the radar. Then we continue this for life because that's what we need to do with these patients.

[00:10:14] Bouchie: Right, right. In the hereditary cancer clinic environment, who should be screened? And what are the types of results that you might receive if you are screened within that kind of a clinic?

[00:10:27] Hughes: So genetic testing is a two-step process. So, the first step is a genetic testing piece. And anyone who has a strong family history of cancer, they should have genetic testing done. And today we test for 80 to 142 different genes because the testing is inexpensive. We just test everybody for everything, it's easier. But to have it covered by your insurance, it has to be, you have to meet the criteria. So, with a strong family history, multiple relatives affected, young age of diagnosis, people in your family with more than one cancer, like bilateral breast cancer, breast cancer on both sides, or breast and ovarian cancer. If that's in your family, then getting testing is important and talk to your doctor about how to get that done. Either our genetic counselors or nurse practitioners or myself, we can test you for that.

But we also now are testing at the population level. And the In Our DNA South Carolina research program is set up to test anyone over the age of 18 for risk of breast cancer and colorectal cancer. Now this is for a small number of genes, but this is now a way to look at the population level with or without a family history to just test patients to see if they have a mutation. This is gonna pick up a lot more patients. The next level of that, when you do this test, about one out of 20 will have a mutation. And it's that one out of 20 with the mutation that then goes on to the hereditary cancer clinic for very specific management for that mutation.

And again, it doesn't mean you're gonna lose an organ, it doesn't mean we're gonna tell you that you're gonna die of cancer. That's certainly not the point. The point is that we know your risk is higher, we know where that risk is higher, we know how to try to find that cancer early, and in some cases, we know how to prevent it, either by medication or by surgery. And picking the right strategy for that patient, that's where the Hereditary Cancer Clinic comes in. After the testing, once a patient is positive, and then getting them tested. And then we also, in that clinic, be sure to test as many family members as possible. Because every member of the family on that bloodline is at risk and if we can test them, find their mutation, they also could be saved from a dread disease.

[00:12:28] Bouchie: And when we talk about the population level of this type of work, when we look at genetic testing to look at what's happening with cancer at a large public health type scale, that's when we really start looking at the genetic databases and how they can help identify maybe trends or things that are happening in the population as far as cancer goes. And this is one of the places where research can really come into play. And the databases can also help to identify perhaps different genes that we maybe didn't even realize in the things that they do. So, can you talk a little bit about the key to what's next in cancer research and how our work with genetics is going to be part of that?

[00:13:03] Hughes: Wow, that's a lot of stuff. So basically, genetics is a basis of almost all cancer. So, all cancers are caused by genetic mutations. As you mentioned earlier, most of these mutations we pick up over the course of our lifetime. A small number of that, five to 10%, we're born with that mutation that then spawns other mutations to make a cancer develop. Understanding those genes better, understanding what they do better is kind of part of the next step. If you look at how we did this initially, you'd find a family where everybody had cancer, and you'd test that patient, and every family member, and they all had the gene, they all had cancer, and we thought the risk was 100% or some astronomical number. Well, when you test at the population level, you find families where there's no cancer in the family, and somebody has a mutation. And you start to realize that these mutations may not be as penetrant, may not cause as much cancer as you thought. And as we understand what level of risk there is, we can then ratchet down how intensively we screen the patient.

So, our next generation of research at that level is gonna be what is the real risk from this mutation and how well do our guidelines work? Because the guidelines are made by a group of professionals - physicians, nurse practitioners, genetic counselors, sitting around a table and saying, I think that's a good idea, let's try that.

That's not very scientific, but that's the best we have. But as you start putting it into practice, then you can identify is it right that a patient gets an MRI every year? Or do they need it every six months? Or do they need it every two years? Finding out that is gonna start to limit the amount of trauma we give to the patient to get the same outcome. So that's a big part of our research.

Other parts are learning new genes that we don't know about yet. Better understanding the variance in the gene. So just because you have a mutation in a specific gene, is every mutation the same? Does every one cause the same level of risk? That's a major research area at this point. And then we find mutations in genes that we don't understand. They're called variants of uncertain significance. And just as we said, they're uncertain. We don't know what to do with them. So, when we get that result, we tell the patient, you know, it's probably nothing. We're gonna keep an eye on you. We'll treat you based on your family history and consider this negative until we know better. But we're developing ways to try to sort out what those variants are without having to wait five years to gather another 100,000 patients.

So, there's a lot going on right now. And then the next step is gonna be really, who do you test? And what do you test them for? Right now, we test 142 genes. Well, there's 20,000 genes in the genome. And at some places they're testing all 20,000 genes. And then if you get to the 20,000 gene test, which is gonna become as cheap as 142 gene tests, so why not do them all, how do you manage all that information? All of us have 40, 50, 100 mutations. What do we do with all that information? We don't really know yet. And then it comes up, when do you test the patient? Well, we kind of say test at the age of 18, but there are some studies now testing it at the newborn level. So that's what we've got to figure out. How much do we test? Who do we test? When do we test? And how do we manage these positives? That's a lot to figure out yet.

[00:16:27] Bouchie: Does that also extend to how we're treating cancer as well? If we can better understand a patient's genetic makeup, does it make it easier, more efficient, better, safer to treat a patient for cancer?

[00:16:40] Hughes: Absolutely, that's a very large area of research and that's more on what's called the somatic mutation side. And that is in terms of what mutations does the cancer have that the patient wasn't born with. And so, we're looking for what's called targeted therapy. When a person has a cancer and they recur with that cancer and common medications that we normally use for cancer aren't working, we do testing of the tumor. We find mutations in the tumor that we didn't know about in that patient and identify that this patient can be treated with this certain drug and this patient can be treated with a different drug. And there's some overlap with the germline mutations that people with a BRCA1 or 2 mutation can have their breast cancer or ovarian cancer treated with what's called a PARP inhibitor, which doesn't really work outside of the mutation status of patients.

So, we're learning about mutations that the patient's born with. There's mutations in the gene, in the tumor itself that can target your therapy. And then there are gene mutations people are born with that change how medications affect them. So, a certain gene, not a mutation, but genes come in different varieties. Blue eyes and brown eyes are both normal. But some people have genes that metabolize drugs very quickly and other people have genes that don't metabolize the drug as quickly. And if you metabolize a drug quickly, a smaller dose may be more toxic. And so, you have to adjust your dose based on the type of genes you have.
We're just beginning to understand that. And we still haven't figured out how to put it in practice very well. But that's gonna be, I think, more at the computer level where you find out what mutations everybody has or what types of genes they have, what medications they're on, what disease they have, and then how do you adjust the medication and the dose based on their genetic makeup? That's extremely complicated, and it's gonna take computers to do that or artificial intelligence. We're not at the place yet where humans can do that in their head. And yet that's where we need to be.

[00:18:38] Bouchie: Right, but that's the next frontiers, really, of where we're headed with all of this. Yeah, that's really great. So, your passion is to do a better job linking all of this together. And one of the big things, and you just sort of alluded to this idea that the human brain is quite limited in its capacity. And so, to take all of the data that's required to look at all of the facets of what it takes in order to really bring optimum health to a patient, we need to be doing that better. We need to have better systems for how we're looking at all of the data, because we're producing tons and tons of data at this point. So how do we bring it all together? So that's one of the things that you're really working on. What does that look like? Tell us a little bit about that first, just generally overall, and then maybe what that looks like in real world practice.

[00:19:31] Hughes: So medicine, like every other industry, has multiple databases that cover multiple things and often don't talk to each other. So, we were hoping that the electronic health record would bring it all together, but that hasn't happened. We still have very specialized systems for tracking data in pathology, other specialized systems for tracking data in the genes, other data for tracking people who have radiation therapy. And each of those areas, and that's just the beginning, need specialized software that helps them do their job better. It's like giving a carpenter a Swiss Army knife and telling them to make a cabinet. You can't do it that way. You have to give them the right saw, the right drill, the right pieces. So, each of these groups need software specific to them. But each of that piece of software was designed independently. So, then they don't talk to each other. So, the same information is in each of these places, but in different formats, in different codes, in different fields. So, we're working on trying to find ways to bring that data together by getting the databases all into one place. We call it a data lake. And then once it's in the data lake, making sure it's updated regularly, now all these are still independent, they're just in the same place now.

But then we start to map them to a single structure. And when they're in a single structure, now you have the data to start playing with to start understanding this better. And you start out by, as a human, trying to figure out what's the relationship of this problem to this cancer to this gene, et cetera. And so, it's more human driven and rules driven. But the next step is machine learning and artificial intelligence, where the machine starts to look at patterns that we're just not able to see. So, we can look at five or 10 characteristics of a cancer and then we get confused. A machine can look at a thousand characteristics and it doesn't care, it's just numbers to them. So, the future is really gonna be having the data, that's the first thing, having the data in one place in one format, and then having the machines able to start working with that data to find out what is the relationships that we're missing. And it's not gonna be machines or humans, it's gonna be machines and humans. Because they're gonna bring to the table a vast amount of ability in calculation and algorithm, but we bring to the table experience and empathy and all the human side of things. And together, augmented intelligence is more important than artificial intelligence. And that's where we're heading towards, is large data sets, very intensive analysis, but then looking at what does that really mean in the real world and that's going to be the future. That's the next decade or two.

[00:22:04] Bouchie: I love that augmented intelligence because it really is about giving the specialist, like yourself, a better tool in order to look at something and then make decisions about it. It's not about the data or the tool telling you necessarily what to do, but it's giving you a little bit more information to work with. It's giving you synthesis of 1,000 permutations where you just couldn't do that on your own.

[00:22:31] Hughes: Exactly right. And that's where we need to be. It's no different than you're driving a car. That doesn't mean that the car replaces you. It means that the car helps you get somewhere faster or more efficiently. So, it's using machines, whether it's a computer or a car or anything else to make you more efficient and effective as a human being. That's augmenting the intelligence.

[00:22:52] Bouchie: And what does it mean for cancer research specifically if we can pull these various data places together in the data lake as you mentioned. What does that mean for cancer research and where we're headed there?

[00:23:05] Hughes: So, I'll give you an example of Amazon, what Amazon does. So, Amazon takes all this information about what you buy and then looks at what the weather is and where you live. What's the environment where you live? What's the weather where you live and what's the weather on the day you bought certain things? So, you look at data as saying you got a whole bunch of data and the first question you ask is what happened? Okay, well people bought boots. Okay, then the second question is why did it happen? Well, it was raining or it's snowing. And then the next question is well, what's gonna happen? Well, when it snows, people are gonna buy boots. And the final question is how do you make it happen? Well, Amazon puts up a boot commercial in your area when it's snowing outside. So that's use of data that we are nowhere near in medicine.

We're still at the point of taking a small chunk of data and saying, well, this is what happened, and maybe predicting what might happen. We haven't got to the point of saying; how do we make things happen? How do we make the patient come in for her mammogram? How do we make the patient know what drug to take, or make sure she takes the right drug, or takes her drug regularly? Amazon's a master at that, and they use computers and artificial intelligence to do that, and we're nowhere near it. But that's the future, is if we can pull this data together, look at how it interrelates, and then how does that then drive getting patients the right treatment and then helping them understand why they need to do it and making them want to do it. That's the future.

[00:24:34] Bouchie: You mentioned In Our DNA SC, which is the community health research project that MUSC is running. Can you talk a little bit more about that for listeners who might want to get involved?

[00:24:46] Hughes: Sure. So, In Our DNA South Carolina or In Our DNA SC is a research protocol. Anyone over the age of 18 who's a MUSC patient can participate. They don't need any special counseling. They don't need anything beyond just signing an informed consent, understanding what the protocol is about. and then giving a saliva sample and then getting a result back. So, we're using the CDC, the Center for Disease Control Tier 1 genes. These are genes that everybody agrees should be tested in everybody. So, the genes are for breast cancer risk, BRCA1 and BRCA2, and then for colon cancer risk with what's called Lynch syndrome, and for hereditary hypercholesterolemia, which is an overabundance of cholesterol that increases the risk of heart disease.

So, testing for these genes, basically, the CDC has said everybody over the age of 18 should do this anyway. So, there are multiple programs going on around the country, and we have one of those programs going on. If you're over 18, if you're an MUSC patient, you go to MyChart and you can sign up for this study. If you're not an MUSC patient, become an MUSC patient to do the study. When you do this, you're going to get a result back that tells you do you have one of these genes. Now, if you have one of those mutations, you’ll see us in the hereditary cancer clinic, and we'll help to get your treatment organized and your management organized. If you don't have one of those genes, it does not mean you don't have a hereditary cancer. This is only the beginning. So, if you have a strong family history, you still wanna have regular genetic testing, not on a research protocol, where we test hundreds of genes, or actually about 140 genes, to look for genes that increase cancer risk.

But we want to at the population level; identify patients we may have missed otherwise. They don't have a strong family history. They can't get in to see a genetic counselor. They don't have time to come in. They don't understand they need to get tested. Their doctor doesn't tell them. That's why we're making this available to anybody. And again, we're not the only ones in the country. This is very normal. It's becoming a normal thing to do. So you can sign up, you get your test done. If you have a mutation, we'll help take care of you. If you don't have one, but you have a strong family history, then get commercial genetic testing, regular genetic testing so we don't miss anything.

[00:26:57] Bouchie: What do you mean when you say a strong family history? How would someone listening determine, oh, I have a strong family history?

[00:27:05] Hughes: Sure, so strong family history means that there are multiple relatives in your family affected with similar cancers. That's hard to know what's similar. So, a breast cancer and an ovarian cancer, they don't sound similar, but they run in the same type of gene. So if you have, make it simple, multiple cancers in the family. Young age of diagnosis, breast cancer in the 20s or 30s rather than in the 50s, 60s, 70s as we normally see. Breast cancer, or I'm sorry, cancer occurring in more than one organ. So, a woman who has breast cancer on both sides or a man who has a colon cancer and a prostate cancer. Multiple cancers in one individual, young age of diagnosis, multiple family members. That's what a strong family history means to us and most patients with that are eligible for commercial testing which is then covered by their insurance.

[00:27:54] Bouchie: We've been talking to Dr. Kevin Hughes with the Hollings Cancer Center about hereditary cancers and the importance of genetic testing. To learn more about the In Our DNA SC Community Health Research Project, visit www.inourdnasc.org or check our show notes.

Have an idea for a future episode of Science Never Sleeps? Click on the link in the show notes to share with us. Science Never Sleeps is produced by the Office of the Vice President for Research at the Medical University of South Carolina. Special thanks to the Office of Instructional Technology for production support on this episode.