Aaron talks to Dr. Tatiana Foroud about her career and her current work in genetics and genomics. Her work spans from the early days of searching for DNA markers for rare disorders to today's search for genetic causes and potential treatments around Alzheimer's disease. Dr. Foroud's career relates a story about how the technology and techniques have developed over decades, and how these breakthroughs could lead to new treatments. We'll also get into the explosion in home genetic testing, and how this trend has changed public perception of genetic testing and treatments.
The Healthcare Triage podcast is sponsored by Indiana University School of Medicine whose mission is to advance health in the state of Indiana and beyond by promoting innovation and excellence in education, research and patient care.
IU School of Medicine is leading Indiana University's first grand challenge, the Precision Health Initiative, with bold goals to cure multiple myeloma, triple negative breast cancer and childhood sarcoma and prevent type 2 diabetes and Alzheimer's disease.
Dr. Aaron Carroll: Welcome back to the Healthcare Triage Podcast. This Healthcare Triage Podcast is sponsored by Indiana University School of Medicine, whose mission is to advance health in the state of Indiana and beyond by promoting innovation and excellence in education, research, and patient care. IU School of Medicine is leading Indiana University's first grand challenge, the Precision Health Initiative, with bold goals to cure multiple myeloma, triple negative breast cancer, and childhood sarcoma, and prevent type two diabetes and Alzheimer's disease. Today we're going to talk about genetics and genomics and lots of other "etics" and "omics" like proteomics, and to clear that all up we're going to be talking to one of the world's experts in this area, Tatiana Foroud. She's the Joe C. Christian Chair of Genetics and the Chair of the Department of Medical and Molecular Genetics at Indiana University School of Medicine. Tatiana, welcome.
Dr. Tatiana Foroud: Thanks for having me.
Dr. Aaron Carroll: We like to start off by talking to our guests about how they got to this position, because you're clearly chair of this department but there has to be a road between what did you want to do someday and then what do you do now. And then let's talk about what you do.
Dr. Tatiana Foroud: Glad to tell you. I grew up in Connecticut and went to a small college there, and I went in as a biology major. I wanted to be a doctor. Not a doctor like I am, but an MD doctor, and midway through realized I didn't want to do that and also picked up a math degree. So I got a degree in biology and math. Had no idea what to do with that but knew I wanted to go to graduate school and found a program in bio math at UCLA. And it was far away from where I'd grown up, so I went there for three years. Got a master's degree there and learned that I loved genetics. And so I came to Indiana University because there was a really well known individual who did something called statistical genetics, which let me put together love of genetics, biology, and math. And so I came here and got my PhD, and then I never left because I loved it here.
Dr. Aaron Carroll: It feels like at some point the genetics were right on the horizon and the next thing you know we were waist deep in it. So was it at the beginning of that swing or was it already moving into that area?
Dr. Tatiana Foroud: It's a great question. When I was in graduate school, we knew about a lot of syndromes and people were really just starting to figure out how to map genes that cause disease. So I worked in a really rare disorder, and the way we used to study it is we had worked with many families. It's what's called an autosomal recessive disease, so your parents don't have it but the child has it. And so we would scour the world looking for families that had this really rare disorder, and then we would go to the laboratory, we would isolate DNA, or genetic material, and then about every couple of weeks we would be able to study a specific spot on the DNA. It was called an RFLP.
So we would march through the genome, which we now call the genome, one marker at a time, getting one marker every couple of weeks. So if you were lucky, you figured out what chromosome your gene of interest was on maybe in the first couple hundred. And sometimes it took people much longer to even figure out where they needed to be looking.
Dr. Aaron Carroll: Did they just start at one end and go to the other?
Dr. Tatiana Foroud: It was a pretty non-systematic, and so sometimes you'd be lucky and be able to test two markers at the same time, but people, their friend would have some markers you could use and they'd say, "Oh, why don't you use those in your lab?" It was incredibly a cottage industry at that time. And so when I came to Indiana University they just discovered this thing called di, tri, and tetranucleotide repeats, this great new discovery that lets you have markers that were much more what we call informative, so when you studied one you had a chance of learning something from it. Before, you weren't even sure you could learn something when you tested it. And that was the beginning. But the idea of sequencing the genome, we didn't even... They were just starting the genome project. It was something that seemed like science fiction.
Dr. Aaron Carroll: If you don't mind my asking, when was this? What year?
Dr. Tatiana Foroud: I was in California from 1987 to 1990, and so I came to Indiana in 1990.
Dr. Aaron Carroll: So what computers were they running this on back then?
Dr. Tatiana Foroud: Oh my gosh. You did it actually on your desktop. We had all these desktop programs-
Dr. Aaron Carroll: This is just unfathomable, yeah.
Dr. Tatiana Foroud: It's completely unfathomable. And so just to give you a perspective, and I tell graduate students when you wrote a paper, and you would either fax your changes to the other person or maybe you'd send it in FedEx. But the idea of using email to send tracked edits was unheard of.
Dr. Aaron Carroll: Who wrote the programs that would do this, even?
Dr. Tatiana Foroud: So actually, the program that I used... It's actually a great story. The project I worked on in California with this disorder called ataxia telangiectasia, and it was the graduate student that I worked with who had written the program that we used. It was part of his thesis. And he went on vacation and I got a marker while he was on vacation. And I actually got the marker that actually mapped the disease chain. And he was so sad. He said it was the one vacation he'd taken in graduate school. We worked on this project together. But that was... It was really a mom and pop kind of thing that you did.
Dr. Aaron Carroll: So someone would be trying to do some sequencing and then you would just...
Dr. Tatiana Foroud: Oh. No, no. Not sequencing. Not even sequencing. What you did is you would have what we call an RFLP. That means you would have primers on two different ends of this spot that you wanted to test. And so what an RFLP just means is there's a variation at that particular spot of DNA that changes whether or not you can cut the DNA there. So that's all you're looking for. So you're either looking for a big fragment of DNA or maybe two fragments of equal size. And that's all you were looking for on a gel that you ran out. You would then have to enter all the data for all the individuals in your family into your computer. So that would be your first job, was to take somebody who had scored gels. They'd looked at the gel, figured out what the two alleles were that each individual had. They would write it on a piece of paper. They would hand you that piece of paper. You would now enter it in the computer and then you'd run the program.
Dr. Aaron Carroll: This is getting us into, let's say, the early '90s. So was the next leap forward that we started being able to sequence the genome?
Dr. Tatiana Foroud: We weren't even there. So then we had this discovery of single nucleotide polymorphisms, these SNPs. And SNPs were really great because there are many of them throughout the genome, so you went from testing a couple hundred markers and trying to figure out where a disease gene was and really having trouble. I always describe it to students, it's like you have the world and you can figure out what state you want to be in but how do you figure out your actual address? It was really hard work and it could take you years to get there. So once you got these things called SNPs... They're called SNPs, single nucleotide polymorphisms, then you suddenly had the ability to test thousands of them.
So now you can really start to get yourself not just to a state, you can get maybe to a city and if you're really lucky maybe you get yourself onto a street. So it really, really sped up the way we could do this, and we didn't have to do them one at a time. They then developed chips, and it would give you back all the data from all the SNPs you wanted to test in the genome at one time. So no more entering them one at a time. You now would get back a file and you could analyze data.
Dr. Aaron Carroll: So is it that we had a sense of what it should look like and each time you were testing someone's genes to see if it looked like what we thought it should look like, to look for one that didn't? Is that what...
Dr. Tatiana Foroud: What you're doing throughout this whole process is a lot of this is about being agnostic. So when we first started doing genetics, we thought we'd know maybe what chromosome we wanted to look at. This process is very agnostic. You say, "I have no idea where in the genome the relevant gene is that I'm looking for, so I'm just going to March my way through in a systematic way like you were doing a survey and look everywhere." So what moves from going these different technologies is just how much real estate you really have to probe in depth. So with SNPs, you now get a really good feel and now maybe you'll start to do more in-depth work. What you get from sequencing is a really comprehensive look. The two kinds of sequencing we now focus on are what we call whole exome sequencing and whole genome sequencing.
The difference is how much real estate you look at. With whole exome, you just look at a couple percent of the genome. But it's not just any part of the genome, it's the part of your genome that codes for genes. So if you're looking for something that causes disease, you're looking for things that have dramatic effects. Maybe it makes it so that you can't make the protein, or you make it in an abnormal way. If you look at the whole genome, you look at all 3 billion base pairs, it's comprehensive, you're looking at everything that could be affected, but you have a lot more sequencing to do and a lot more work to do and it's sometimes harder to interpret what you find, whereas whole exome typically is a lot easier to interpret. So we've marched our way through in the complexity of genetics. We're simple people. We work our way through. We started... We worked our way through these RFPs, these single nucleotide polymorphisms, whole exome.
And now, when we can, we do whole genome sequencing. And all of this is driven by cost. So as the cost of these technologies go down, our ability to be really thorough and investigate everything is now increasingly possible.
Dr. Aaron Carroll: But I'm still always interested. We get the genome. Here's the list of all the base pairs. But then do you compare it to an expectation? And what is that expectation? Who has the genome we say, "That's normal and that's what we need to compare-"
Dr. Tatiana Foroud: That's actually a great question. We used to call it reference genomes, and one of the challenges of that is exactly what you just said: how do you pick the right reference? Because all of us carry some, we'll call them mutations, some variants that can lead to disease. And so in that case, that variant isn't a very good reference. There's also a lot of variation due to race and ethnicity, and so many of our original reference genomes were all from individuals of European ancestry. So they weren't particularly informative when you looked at individuals who are more diverse.
So what we do now is we talk a lot more when we look at our sequence, not so much how does it compare to a reference? But can we predict what different variants, whether or not they might have an effect? So we move from saying, "This is good and this is not good," to predicting. What variants, as you go up and down the genome, we typically look for things that are rare because we expect things to be deleterious, to not be as common, because typically they would not be as likely to lead to your survival.
So we typically focus on rarer variation, especially when we're doing exome. And we ask, could this cause any kind of deleterious effect? And we'll ask about all rare variation that we see. So we do consider reference. When you think about reference, it's not a reference individual but we'll talk about reference populations. So we know the frequency of just about every base pair in reference groups of genomes, so people of European ancestry, people of African ancestry. One of the challenges is if you start working in populations that are rare or are ethnic or regional isolates, we don't know as much about what's common or rare in those populations.
Dr. Aaron Carroll: Right. It seems like not only would the cost of the doing the sequencing be prohibited, but it just seems like it would be an enormous amount of computing power.
Dr. Tatiana Foroud: That's where the dramatic advances in terms of computing power that have occurred over the last decade have really made this all possible. So in order to be able to analyze the sequence of one individual, that's one capability. To analyze the sequence of thousands of individuals, that's why we have cloud computing in many cases or, like at any Indiana University, we have Big Red and we've had various technologies where they've really tried to upgrade our ability to have high throughput computing. It's one thing to store it and it's one thing to analyze it. So increasingly those are two different kinds of computational resources that you need.
Dr. Aaron Carroll: How much space does it take to store a genome?
Dr. Tatiana Foroud: We've gotten clever in genetics. We can store it in different ways. You can literally just store the raw files that come off the sequencing or you can process them and turn them into a file that just says, at each base pair, "What do I have?" As you increasingly process that file, you often turn it into something smaller. So if you start thinking about things like precision health and how could you ever use this in healthcare, you can't be using these really raw files. That's going to be useless. What's every physician going to do, go back and try to run a computer program to figure out what things mean?
We have to increasingly come up with formats where we summarize the information from the genome that we think is important for healthcare. The challenge we have is that we're learning at such a fast pace that if we do that annotation now and we were to look back at this in five years, we'd say, "Oh my gosh, I missed all these really important things." So a lot of what we do right now is even analyze data that we generated a couple years ago and reinterpret it because we've learned so much in that intervening period.
Dr. Aaron Carroll: It seems like there's a real potential too, given that we probably have stored material from years ago and now we probably have future, forward-moving medical data at that point that by looking at stuff from the past and adding it to data that we already have in cohort studies, that you could start to make some movement as opposed to just looking forward.
Dr. Tatiana Foroud: You're right. There's two things to think about. One is, how did you get that material and did the individual consent, give approval for that to be used for that purpose? Some of the material that we have wasn't consented in that way, so that gives you sort of a problem of how you can use it and how it can be linked to future medical records or other information. The other piece is how that material was collected. So when I do families, I'll often have people offer me things like the hairbrush of a deceased person and they'll say, "But on TV that hairbrush was enough to find the killer." And I'll say, "You know, depending on what you want to do with that DNA that you're going to get from that hairbrush, it may or may not be possible to do what we need to do."
So that's the other piece, is what do we have left? And so one of the things that people have really done a lot of work on is, for example, supposing someone had surgery and there were what we call blocks or tissue that remain. For many years, it was very difficult for us to get DNA from that kind of material because it's what we call fixed material. Increasingly, people realize how important it is to be able to get DNA from material like that, so there's been a lot of work, methods to get good quality DNA from tissue like that. Because exactly what you're describing, if you could study that and do all this new genetic studies that we have, you could learn a lot. So we've got a little bit of work to do in terms of being able to use everything that we have from the past. We have to ask, was it consented? How did we get it? Did the person know? And was it collected in a way that allows us to do all these new technologies?
Dr. Aaron Carroll: I was even thinking you could probably draw some material from me today if I happen to be in an electronic medical record for 30 years.
Dr. Tatiana Foroud: We have studies where we do exactly that, where we ask... We have a study called the Indiana Byway, and that is exactly. We ask the individual to consent to this research study. The research study collects up usually a blood sample from the individual and asks to link it to their electronic medical record. And we say that allows us to link to past information, the history of information in the past on the electronic medical record, what we know now, and what might be learned in the future. And the advantage of that is with one sample you can link to all of that information. So if you were to do sequencing, for example, on that sample, you can link it to all that information. You can look at outcomes, you can look at risk factors. You can ask, "Could we have predicted that this individual would develop that disease? And knowing that, could we use that to potentially intervene in the future?" So research studies let us ask those questions and design those studies so that we can then learn from those and hopefully implement them in clinical care.
Dr. Aaron Carroll: So what's the rate limiting step stopping us from doing all of that right now? It seems like we're there. We can sequence the genome, we can crunch the numbers, we have billions of people.
Dr. Tatiana Foroud: One of the things you have to ask is it's a financial consideration when you think about it for a healthcare system, that cost benefit ratio. Can I learn enough from studying that genome to be able to justify the cost of getting that blood sample, getting the DNA, performing that sequencing, and then interpreting it and storing it? And increasingly the answer to that is yes. There's a lot of things that have to be built in order to make that part of a healthcare system. How will you explain to someone, whether it's a physician or a patient, what the results of that sequencing means? How will we act on that? When should we act on that? And so there's a whole piece there, and increasingly that's exactly what Precision Health Initiative is trying to do, is figure out, how do we build that infrastructure? How do we use it? When do we use it?
And so we've done that in part with targeted disorders. But the Indiana Biobank thinks beyond that and says, "Let's get ready to do this on a broad scale." And part of the reason I talk about that is that's a study that I lead along with Dr. Anantha Shekhar, who is the principal investigator of the-
Dr. Aaron Carroll: And so is that more broadly like, "Let's gather up as much DNA as we can and crunch the numbers?"
Dr. Tatiana Foroud: We are. So we have samples right now from about 40,000 individuals. And what we're trying to do is optimize, how could we discuss this with patients being seen at IU Health? How could we do this in an efficient way so that we could scale that up and do, we'd like to think, 100,000 to 300,000 individuals?
Dr. Aaron Carroll: Is there a number in your head where you're like, "This is how many we're going to need in order to have enough data to actually-"
Dr. Tatiana Foroud: Yeah. I think our target right now, 100,000 to 300,000. And again, that's not to say those are the only people we want to be able to do, this is to say, "How do we figure out the feasibility of this, figure out if we should do this broadly, should we do this in targeted individuals?" And then make a case to the healthcare system and say, "Is this something we should be doing routinely?"
Dr. Aaron Carroll: This is clearly all the positives. What do people worry about? What prevents them from just lining up and getting involved?
Dr. Tatiana Foroud: I think for research studies, in some sense they are low risk because the way we currently, at this very moment, have the Indiana Biobank, an individual does not learn any results of the studies that are performed. There's a movement now, because I think a lot of people do want to learn from this information, to be able to return information back. And then the question is, what information do we return back? The American Society of Human Genetics, American College of Medical Genetics, has developed actionable genes. There's a series of genes where, if you found important mutations or changes in the DNA in those genes, you'd say, "We could do something about that." Many of them are tied to cancers. Many of them are tied to heart conditions where, if you could intervene and identify a person, you could start to do interventions. You could do assessments, you would do regular evaluations. You could identify when you needed to intervene.
There are a number of disorders, I study Alzheimer's disease, where knowing that an individuals in increased risk, there's actually nothing we do right now. And so one of the questions about performing these studies and motivation of individuals is often people will say, "I'm interested in knowing about things that I can do something about. I actually don't want to know something about these things that I can't do anything about it. I don't want that sword hanging over my head," many people will say, whereas other people will say, "I really want to know that because I might make life decisions based on that." So part of that is having a discussion with individuals so they can make that decision.
Dr. Aaron Carroll: I want to back up a second. What is the difference between genetics and genomics?
Dr. Tatiana Foroud: Great question. Depending on where you look, you might get a slightly different definition. If you look at some research, they'll say genetics is the study of heredity. It's a very limited, "I'm looking at one gene." And some people would say genomics is when I start thinking about the genome and I look at the big picture. Some people use them a little bit differently and will say, "Genomics is when I start to think about RNA and epigenetics." And I think about when many people use this word other-omics. Let's start with DNA and genetics. DNA, most people think about their double helix and think about just our sequence, our 3 billion base pairs. When you think about epigenetics, what you're talking about are changes that can occur on the DNA that will change how much of a certain product of a gene that you make or don't make.
The changes we're looking at are often called methylation, and they can be caused by the way the DNA gets conformed, making it easier or harder for it to be copied and transcribed, a variety of different mechanisms. If we talk about transcriptomics, we're asking about not how much DNA we make but how much expression we have of the genes that are coded by the DNA. So one of the things that a change in DNA sequence can do is it can change the protein you make. It could also affect how much of the protein you make, where you make it. So transcriptomics often asks, how much of that product of that gene do I make? Does it vary by different kinds of cells? Does it change depending on if I look in the brain or in the liver? Does it change in different cells in the brain? Does it change because something's occurred in the DNA that makes it more likely a person has disease from that?
Dr. Aaron Carroll: When we say other-omics, though, because I hear proteomics or... What are all of those?
Dr. Tatiana Foroud: What people are really trying to tap into are all the added knowledge that we have. We've spent all this time learning about sequencing the genome. Proteomics is asking about, what proteins are we making and in what abundance, what frequency, and does it vary? Does the amount vary? Does what gets made vary? And we can ask that again in different kinds of samples. I personally study in a blood sample from plasma, but you can also do proteomics in cerebral spinal fluid. You can do it in a variety of different kinds of tissues. We often think about proteomics as what I call a biomarker. So suppose you get one of these individuals. From genetics you know that they're at increased risk for a disease, but what you really want to know is, are going to develop the disease?
And much of the work that we're trying to do is to identify what changes might occur before an individual ever has symptoms of the disease. And I study neurodegeneration, so we're really interested in that because what you'd like to do is recognize an individual that's already started on their path toward that disease so you could intervene before they ever have symptoms. Proteomics, transcriptomics are all possible ways that we can measure things in the blood or through another sample that we can ask at a regular frequency, just like you go for blood work, is anything changing that we need to be concerned about that suggests this individual is approaching disease onset?
Dr. Aaron Carroll: Do we have anything that we've gotten to the point where we're like, "This is something that should be rolled out and we should be doing in clinical care?"
Dr. Tatiana Foroud: So if you think about, at a very basic level, things like assessing prostate. Those kinds of assessments are really early forms of cancer biomarkers. What we don't do very well is do that, for example, in neurodegeneration. So I'm very focused on that because many people don't want to know about their risk for Alzheimer's and Parkinson's because they say, "I can't do anything about it." And so these are the beginning steps to say, "If we could identify biomarkers that you could measure in your 50s to say not only might you be at increased risk genetically but we also think you're at increased risk because we see things changing." And being able to test something in the blood is a lot easier than having to do brain scans and things like that. So what we're always looking for are things that are less invasive, more cost efficient.
Dr. Aaron Carroll: So what other areas are people working on? It sounds like some of this is screening. It sounds like most of what I think we've discussed is screening. Yeah.
Dr. Tatiana Foroud: What I do is screening but there's also just basic biology and understanding, what's happening in this disease? What's changing? What's causing those symptoms that we observe? And proteomics, transcriptomics can all be great ways to study that.
Dr. Aaron Carroll: This is interesting because most of the times when I hear precision medicine, at least when it's talked about publicly, most of the times it's "Oh, we want to run a test to tell which of these drugs will work best for you." Is most of the work in precision medicine on pharmacogenetics or is it all of this?
Dr. Tatiana Foroud: Pharmacogenetics and pharmacogenomics was the thing that was tackled first because it was easy to see how it could have significant clinical impact. Knowing that you prescribe a drug for an individual who will have deleterious metabolism of that drug and lead to adverse events, the healthcare system goes, "I want to know that because if that's life-threatening, that could have huge financial impact for us." So it's easy to motivate, even if something is a very rare DNA change, and this adverse event is very unusual, if it has really significant financial costs. Pretty easy to motivate people that that cost-benefit ratio is high. And we understood them because we knew what to look at. Remember we talked earlier about, how do you know what genes to look for? The thing about pharmacogenetics and pharmacogenomics is we know a lot about drug metabolism so we knew what genes to look at. So we made those discoveries really quite early.
Dr. Aaron Carroll: Right. This is just when the science communicator in me is the screaming in my head because I've written pieces when President Obama or others talked about, "Precision medicine going to change everything." And the example they always used was pharmacogenetics. And I was like, that works for like this much of the population. I'm holding my fingers very close together.
Dr. Tatiana Foroud: You are. Frighteningly close.
Dr. Aaron Carroll: A very tiny part of the population, and it's only for a very few specific diseases in a very few specific drugs, but what you're describing is a world of difference. And I'm like, "That's what they should be talking about."
Dr. Tatiana Foroud: I think the challenge of it is you can't intervene as dramatically yet. And so this is... I feel like even though we can sequence the genome, we're still in pretty early stages and so genetics only explains a piece of our variation right now. And so that's what everyone goes is, "If we study genetics, is that going to give us enough to really intervene on?" And I would argue, it's not just genetics, it's genetics coupled with other things.
Dr. Aaron Carroll: Yeah, but the change, I think, even in just a couple of decades, with what you're talking about, what you could do is so remarkable. It's made such leaps. Are there leaps like this you see coming down the pike or do you think we've gotten most of the way-
Dr. Tatiana Foroud: No, I think there's a lot of work. A lot of work is being done in an area that I don't do as much with, looking at machine learning, looking and really diving into the electronic medical records, whether it's with natural language processing. But really now going to another level of analysis. What we do now is we do genetics on these enormous scales. We don't talk about studying a couple thousand people. People look at studies that have half a million individuals. And so there you're really looking at patterns. Now, some people criticize those and say, "If you need half a million people to be able to find genetic effects, how important are they?" It's an early question to ask, and it's fair.
In order to get what we call robust and reproducible results, to have rigor, we need to have large samples. But then we can take what we learned from that and start to apply that and ask about those interactions. So once I identify people that have these genetic profiles that suggest they're at increased risk, now let me pair it with supposing we look at environmental exposures, we look at diet, we look at co-morbidities. We can start to tease that apart and find profiles of individuals that are increased risk for disease. And I think you're not going to use genetics in isolation, and I think that was the mistake-
Dr. Aaron Carroll: No, and I think you're right. Why do you need 500,000 people for all those reasons? Because it's going to be more than just the DNA, you're going to have to account for all of those other variables together. I get some of the concern of people who are like, "There's stuff I don't want to know that I can't do anything about." But the idea that we could somehow just pool data on just a lot of people, even anonymously, to try to get at some of these questions just seems, why wouldn't you do it? Of course I guess the anonymity is the problem.
Dr. Tatiana Foroud: Well, and I would say your answer to, "I don't want to know," will change if there's something that can be done.
Dr. Aaron Carroll: Oh, that's 100% of it. I totally agree. Yeah.
Dr. Tatiana Foroud: And so we have a number of studies where we encourage people. If that's something that you would like to know, now leverage that and participate in research studies. And the other thing to think about is potentially genetics in terms of what drugs will work. So if you're thinking about interventions, whether it's drug related or it could be behavioral, being able to use, whether it's genetics and other factors, to identify which individuals are going to be most appropriate. That's the other way we think about precision medicine, is, would you tailor what you would do in terms of an intervention based on some of that information?
Dr. Aaron Carroll: Right. How close do you think we are to be being actually able to use this in clinical care? Is this still like we're still trying to... We're getting the tools together and we're trying to find some good use cases, or we're going to be close to rolling this out for some things in the future?
Dr. Tatiana Foroud: I think you're not going to roll it out for everything, I think it's going to be staged. I think you're going to have certain conditions in which you're going to do this earlier than others, and it's really going to depend, again, on what you have in terms of intervention. So where are we in terms of therapies? We're not in the same place for a number of diseases. So I don't think it's going to roll out universally across everything. I think it's going to be very similar. Many people think it's going to impact cancer as it is right now first. You already have like precision tumor boards. You already have growth in that area. A lot of interest in terms of cardiovascular. But there are a number of areas that are just farther behind.
Dr. Aaron Carroll: So even when we talk about cancer, if we can delve into that specifically for a second. What can we do now that we couldn't do five or 10 years ago?
Dr. Tatiana Foroud: Here at Indiana University we have precision oncology. So what we can do for individuals who have failed initial treatment regimens, what we can do is ask, by sequencing the tumor, by sequencing the genome of the individual. And when we talk about the tumor we talk about both genetics, sequencing the DNA, and also sequencing the RNA, transcriptomics, what's being made. That is being used to ask, could we identify, potentially, what other therapy we should try next based on what we've seen in terms of what proteins are expressed and at what levels based on the variation we see in the DNA and the variation that is changed between the person's DNA in the rest of their body and the DNA of the tumor. So it's already being used. It's not a universal positive outcome.
Dr. Aaron Carroll: Okay, so in cancer. And what are other areas besides cancer?
Dr. Tatiana Foroud: I think there's a lot of interest in terms of cardiovascular. And one of the challenges of cardiovascular disease is there appear to be differences, at least in terms of genetics, among individuals of different ancestry. So we see different rates of disease but we also see that there may be different genetic profiles as well. So that gives you an added thing that you've got to work with, is you might be able to make some advances in some parts of the population but not in all simultaneously. And that's a challenge, too.
Dr. Aaron Carroll: How do you think that some of the public DNA testing sites, 23andMe, Ancestry.com, how have they changed the way the public looks at genetic testing or even thinks about or interacts with your work?
Dr. Tatiana Foroud: I think it's a great question. I think it has changed in a number of ways. First of all, I think it's created an interest in the population. I think people suddenly realize, "Wow, I see ads on this on TV. Is this something I want to do?" And so we do have individuals, for example, in our genetics clinics, and physicians all around will have people come in with their results. So I think it's created an awareness and an interest, which I think is great. I think sometimes there is a misunderstanding of the reports. That has really improved, I will say. In the early days I think there was more confusion around them. I think that's gotten better, but people sometimes just don't understand what that report means. And the fact that much of what you're learning is about risks. It's not about destiny, it's that your risk is increased or decreased, but not that you will or won't develop the disease.
So I think there's a great place in terms of a genetics, genetic counselors to try to help people interpret those reports. I think it's created also that interest in ancestry and maybe a greater understanding that the fact that ancestry can be seen and understood through DNA helps people understand that that can also interpret to medical conditions and your risk. So I think it's suddenly opened up our ability to say, "Depending on your ancestry, it may be that you may be at greater risk or less risk for certain diseases. The treatments you have may differ." And so it feels, I think, perhaps less about discrimination and maybe a greater understanding that this is about our ancestry and the genetics that's tied to it. So I think it opens up the ability to have that discussion as well.
Dr. Aaron Carroll: Okay. So you think more positive than negative?
Dr. Tatiana Foroud: I think so. There's always some. I think one of the things that will come up is this forensic DNA, and will this frighten people at all because they suddenly realize that there's some criminals that are being identified through these DNA databases? Will that frightened people from wanting to participate, whether in research studies or to learn this information? I hope not, but there are studies that suggest that pretty soon 90% of the European ancestry population in the US will be identifiable through information in existing databases.
Dr. Aaron Carroll: That's fascinating. This has been absolutely fascinating. And given how much has changed in just the last few years, I'm sure if we had your back in a couple years from now there'll be all new stuff to talk about, so I hope you'll come back.
Dr. Tatiana Foroud: Absolutely. It was a pleasure to be here.
Dr. Aaron Carroll: Thank you.
Dr. Tatiana Foroud: Thank you.
Dr. Aaron Carroll: This has been an episode of the Healthcare Triage Podcast, which is sponsored by Indiana University's School of Medicine, whose mission is to advance health in the state of Indiana and beyond by promoting innovation and excellence in education, research, and patient care. IU School of Medicine is also leading Indiana University's first grand challenge, the Precision Health Initiative, which we've talked about a lot today. With bold goals to cure multiple myeloma, triple negative breast cancer, and childhood sarcoma, and prevent type two diabetes and Alzheimer's disease.