Jun 19, 2018
Jane Ferguson: Hello, welcome to Getting Personal: Omics of the Heart. It is June 2018, and this is podcast episode 17. I'm Jane Ferguson, an assistant professor of medicine at Vanderbilt University Medical Center, and a proponent of precision medicine, genomics, and finding ways to prevent and treat heart disease.
Jane Ferguson: This podcast is brought to you by Circulation: Genomic and Precision Medicine, and the AHA Council on Genomic and Precision Medicine.
Jane Ferguson: For our interview this month, early career member, Jennie Lin talked to Beth McNally about science and careers in genomic medicine. We'll have more on that later but first I want to tell you about the cool papers we published in the journal this month.
Jane Ferguson: First up, Orlando Gutierrez, Marguerite Irvin, Jeffrey Kopp, Cheryl Winkler, and colleagues from the University of Alabama at Birmingham, and the NIH, published an article entitled APOL1 Nephropathy Risk Variants and Incident Cardiovascular Disease Events in Community-Dwelling Black Adults. This study was conducted in over 10 thousand participants of the Reasons for Geographic and Racial Differences in Stroke, or, REGARDS Study. They examined associations between APOL1 variants and incident coronary heart disease, ischemic stroke, or composite CVD outcome. Because there are coding variants in the APOL1 Gene that are only found in individuals of African ancestry, these are hypothesized to contribute to the disparities in cardiovascular and renal disease in African Americans.
Jane Ferguson: The authors found that carrying the risk variants was associated with increased risk of ischemic stroke, but only in individuals who did not have diabetes, or chronic kidney disease. They hypothesize that because diabetes and kidney disease already increase CVD risk, the variant does not have an additional effect on risk in individuals with existing comorbidities. But, it contributes to small vessel occlusion and stroke in individuals without diabetes.
Jane Ferguson: They also found that the magnitude and strength of the association became stronger in a model adjusted for African ancestry, suggesting an independent effect of the APOL1 risk variants.
Jane Ferguson: While future work is needed to study this more, this is an important step in understanding the complex relationship between APOL1 and disease.
Jane Ferguson: Next up, Daniela Zanetti, Erik Ingelsson, and colleagues from Stanford, published a paper on Birthweight, Type 2 Diabetes, and Cardiovascular Disease: Addressing the Barker Hypothesis with Mendelian Randomization. The Barker Hypothesis considers that low birthweight as a result of intrauterine growth restriction, causes a higher future risk of hypertension, type 2 diabetes, and cardiovascular disease. However, observational studies have been unable to establish causality or mechanisms.
Jane Ferguson: In this paper, the authors used Mendelian Randomization as a tool to address causality. They used data from the UK Biobank, and included over 237,000 participants who knew their weight at birth. They constructed genetic predictors of birthweight from published genome wide association studies, and then looked for genetic associations with multiple outcomes, including CAD, stroke, hypertension, obesity, dyslipidemia, dysregulated glucose and insulin metabolism, and diabetes.
Jane Ferguson: The Mendelian randomization analysis indicated that higher birthweight is protective against CAD type 2 diabetes, LDL cholesterol, and high 2 hour glucose from oral tolerance test. But, higher birthweight was associated with higher adult BMI. This suggests that the association between low birthweight and higher disease risk is independent of effects on BMI later in life. While the study was limited to a well nourished population of European ancestry, and would need to be confirmed in other samples, and through non-genetic studies, it suggests that improving prenatal nutrition may be protective against future cardiometabolic disease risk.
Jane Ferguson: Laura Muino-Mosquera, Julie De Backer, and co-authors from Ghent University Hospital, delved into the complexities of interpreting genetic variants, as published in their manuscript, Tailoring the ACMG and AMP Guidelines for the Interpretation of Sequenced Variants in the FBN1 Gene for Marfan Syndrome: Proposal for a Disease- and Gene-Specific Guideline.
Jane Ferguson: With a large number of variants being uncovered through widespread sequencing efforts, a crucial challenge arises in their interpretation. The American College of Medical Genetics and Genomics, and the Association for Molecular Pathology put forward variant interpretation guidelines in 2015, but these were not tailored to individual genes. Because some genes have unique characteristics, the guidelines may not always allow for uniform interpretation.
Jane Ferguson: In their manuscript, the authors focused on variants in fibrillin-1 that cause Marfan Syndrome, and reclassified 713 variants using the guidelines, comparing those classifications to previous in-depth methods which had indicated these variants' causal or uncertain significance. They find 86.4% agreement between the two methods.
Jane Ferguson: Applying the ACMG, AMP guidelines without considering additional evidence may thus miss causal mutations. And it suggests that adopting gene specific guidelines may be helpful to improve clinical decision making and accurate variant interpretation.
Jane Ferguson: Delving deeper into FBN1 and Marfan Syndrome, Norifumi Takeda, Ryo Inuzuka, Sonoko Maemura, Issei Komuro, and colleagues from the University of Tokyo examined the Impact of Pathogenic FBN1 Variant Types on the Progression of Aortic Disease in Patients With Marfan Syndrome. They evaluated 248 patients with pathogenic, or likely pathogenic, FBN1 variants, and examined the effect of variant subtype on severe aortopathy, including aortic root replacement, type A dissections, and related death. They found that the cumulative aortic event risk was higher in individuals with haploinsufficient type variants, compared with dominant negative variants.
Jane Ferguson: Within individuals with dominant negative variants, those that affected Cysteine residues, or caused in-frame deletions, were associated with higher risk compared with other dominant negative mutations, and were comparable to the risk of the haploinsufficient variants. These results highlight the heterogeneity and risk of the FBN1 variants, and suggest that individuals with haploinsufficient variants, and those carrying dominant negative variants affecting Cysteine residues or in-frame Deletions, may need more careful monitoring for development of aortic root aneurysms.
Jane Ferguson: Lydia Hellwig, William Klein, and colleagues from the NIH, investigated the Ability of Patients to Distinguish Among Cardiac Genomic Variant Subclassifications. In this study, they analyzed whether different subclassifications of variants of uncertain significance were associated with different degrees of concern amongst recipients of genetic test results. 289 subjects were recruited from the NIH ClinSeq Study, and were presented with three categories of variants, including variants of uncertain significance, possibly pathogenic, and likely pathogenic variants. Participants were better able to distinguish between the categories when presented with all three. Whereas, a result of possibly pathogenic given on its own, produced as much worry as a result of likely pathogenic. The authors conclude that multiple categories are helpful for subjects to distinguish pathogenicity subclassification, and that subjects receiving only a single uncertain result, may benefit from interventions to address their worry and to calibrate their risk perceptions.
Jane Ferguson: Erik Ingelsson and Mark McCarthy from Stanford, published a really nice review article entitled Human Genetics of Obesity and Type 2 Diabetes: Past, Present, and Future. Over the past decade, we've had a lot of excitement, optimism, and also disappointment in what genome-wide association studies can deliver. Doctors Ingelsson and McCarthy do a great job laying out the history and the successes in the field of genetic interrogation of obesity and diabetes, as well as acknowledging where reality may not live up to the hype, what challenges remain, and what the future may hold. They also have a figure that uses an analogy of a ski resort to emphasize the importance of taking a longitudinal perspective. And I would argue that any paper that manages to connect apres-ski with genomics is worth reading, for that alone.
Jane Ferguson: Robert Roberts wrote a perspective on the 1986 A.J. Buer program, pivotal to current management and research of heart disease. Highlighting how the decision by the AHA in 1986 to establish centers to train cardiologists and scientists in molecular biology, has led to huge advances in knowledge and treatment of heart disease.
Jane Ferguson: Finally, rounding out this issue, Kiran Musunuru and colleagues, representing the AHA Council on Genomic and Precision Medicine, Council on Cardiovascular Disease in the Young, Council on Cardiovascular and Stroke Nursing, Council on Cardiovascular Radiology and Intervention, Council on Peripheral Vascular Disease, Council on Quality of Care and Outcomes Research, and the Stroke Council, published a scientific statement on Interdisciplinary Models for Research and Clinical Endeavors in Genomic Medicine.
Jane Ferguson: This paper lays out the field of cardiovascular research in the post genomic era, highlights current practices in research and treatment, and outlines vision for interdisciplinary, translational research and clinical practice, that could improve how we understand disease, and how we use those understandings to help patients.
Jane Ferguson: Our guest interviewer today is Dr. Jennie Lin, an Assistant Professor at Northwestern Universities Feinberg School of Medicine, and the incoming Vice Chair of the Early Career Committee of the AHA Council on Genomic and Precision Medicine. As an aside, Jennie is a great person to follow on Twitter for insights into genomics and kidney disease, and as a bonus, she also posts the occasional dog photo. So she's well worth following just for that. You can find her on Twitter @jenniejlin. As you'll hear, Jennie talked to Dr. Beth McNally about her view on genomic medicine, and Beth also shared some really great practical tips for early career investigators building their independent labs. So make sure you listen all the way to the end. Take it away Jennie.
Dr. Lin: Thank you for tuning in to this edition of Getting Personal: Omics of the Heart, a podcast by the Genomic and Precision Medicine Council of the American Heart Association, and by Circulation: Genomic and Precision Medicine. Today I am joined by Dr. Elizabeth McNally, the Elizabeth J Ward Professor of Genetic Medicine, and director of the Center for Genetic Medicine at Northwestern University. Beth, thank you for taking time to chat with all of us.
Dr. McNally: Happy to be here.
Dr. Lin: As a successful physician scientist, you have been interviewed in the past about your life, your scientific interests, and advice for budding investigators. I don't want to rehash everything you have already stated beautifully in an interview with Circ Res, for example. But instead wanted to focus more on your views of genetic medicine and genome science today.
Dr. Lin: So you mentioned in that prior Circ Res interview that you started your laboratory science training and career during college, when you participated in a project focused on genetic variation among children with muscular diseases. What have you found to be most interesting about the process of identifying functional genetic variants back then, and also that on-going work now.
Dr. McNally: Well, I think over the years I've been doing this is the tools have gotten so much better, to be able to actually define the variants much more comprehensibly than we ever could in the past. And then also to be able to study them, and very much to be able to study them in context. And so I look at the revolutions in science that will cause people to look back on this era as the era of genetics. It began obviously with PCR, we couldn't have gotten anywhere without that.
Dr. Lin: Right.
Dr. McNally: And then you leap forward to things like next generation sequencing, and IPS cells, and now CRISPR/Cas gene editing. And to realize that the last three happened within a decade of each other, is going to be so meaningful when you think about the next few decades, and what will happen. So being able to take an IPS cell and actually study a mutation or a variant in context of that patient, the rest of their genome, is really important to be able to do.
Dr. Lin: Okay, Great. And so, where do you envision ... with taking say for example, this next gen. technology, CRISPR/Cas9, studying variants in an IPS cell, for example. How do you envision this really revolutionizing the study of human genetics for patients? And how far do you think we've come in fulfilling that vision, and what do you think should be our focus going forward?
Dr. McNally: I think broadly thinking about human genetics we're really very much still at the beginning, which I know is hard to say and hard to hear. But, we've spent a lot of the last 15 years very focused on that fraction of the genome that has high frequency, or common variation, through a lot of the GWA studies. With those common variants, we had a lot of associations, but relatively small effects of a lot of those, causing a lot of people to focus on the missing heritability and where we might find that hiding. And of course, now that we have deep sequencing, and we have deep sequencing where we've really sampled so much more of the genome, and from so many more people, I think we're just at the beginning of really appreciating that rare variation. And beginning again to really appreciate that 80-85% of the variation that's in each of our genomes is really characterized as rare.
Dr. McNally: And so we really each are quite unique, and that when we understand a variant we do have to understand it in the context of all that other variation. So computationally that's very challenging to do. Obviously requires larger and larger data sets. But even in doing that, you are not going to find exact replicates of the combinations that you see in any one individual. While I know everybody would love that we're going to have the computational answer to all of this, it's still going to come down to a physician and a patient and making what you think is that best decision for the patient, based hopefully on some genetic data that helps inform those decisions.
Dr. Lin: Right, right. So it kind of gets into this whole concept of precision medicine, which has gained a lot of popularity and buzz in recent years, and Obama has really brought it to the forefront in the public arena. You mention rare variants in ... finding rare variants in each patient, for example. And moving a little bit away from some of the common variants that we find in GWAs. What does it mean for a patient to have a rare variant and come see you in your cardiomyopathy clinic, is it going to be precision for that patient, or suing rare variants among many different individual patients to try to find function for a gene?
Dr. McNally: It's a great question. So I think the first way we approach it is, it depends who's asking the question. So if it's somebody who comes to me who has cardiomyopathy, or has a family history of cardiomyopathy and sudden death, that's a very different question to ask what's going on with their rare variants, for example in cardiomyopathy genes. Now if you translate that over to, I have a big population of people, I don't particularly know what their phenotype is, and I see rare variants for cardiomyopathy, those are two fundamentally different questions. So we very much know a lot about how to interpret rare variants for cardiomyopathy in the context of a patient or a family who has disease, and I do emphasize the latter part of that, the family, working with families and seeing how variants segregate within families. We interpret that very differently, and I think it's appropriate to interpret that very differently in that context. And that's completely different than again, going against what is the regular population, notice I'm not calling it normal population-
Dr. Lin: Right.
Dr. McNally: ... but the general population that's out there. The first step in doing that is the list of the ACMG, American College of Medical Genetics, actionable genes. So this is an interesting question in and of itself. It's 59 genes, of course that list is too small, and it should be bigger than that, and ultimately that will happen. But to take a population based approach to those actionable genes, and looking across the population, finding someone who's got variants in, lets say our favorite genes MYH7, MYBPC3. Knowing what that risk means on a population level is very different than knowing what that means in the context of a patient who comes to you, who has that variant, runs in their family, and has clear disease. Those again, really two different questions, and we have to come up with what's the best practices on that, how to answer either of those questions.
Dr. McNally: I think the first step working with patients and families who have known disease and have clear variants that segregate with disease, I think its very powerful. I think we've probably got close to a good decade of doing that already. It's incredibly useful for those patients and families. It helps us reduce their risk. It helps us treat them early, it helps us manage their arrhythmias. There's no question that that information is incredibly valuable, but we're still learning how to process that across the population, and how to answer that question for people who are coming who don't already have disease.
Dr. Lin: Right, right. That makes sense. And I guess that kind of plays into a follow up question about whether or not we need to test, or think about every variant of unknown significance in lab, and ... the-
Dr. McNally: It's a great ... You know again, you always have to very carefully consider the context in which the question's being asked. So again, if you're talking about a relatively normal population, well, walking, healthy person, and you're seeing variants of uncertain significance, that's a very different question than somebody who's coming in to you with cardiomyopathy and has a highly suspicious variant of uncertain significance that falls right within the head domain of MYH7. We know a lot about that, and we can do a lot of interpretation in that case.
Dr. McNally: However, I would say that to put too much value on what we do in the research lab ... Just putting a regulatory hat on for a second and thinking about it, there's nothing from a regulatory standpoint that really validates what we do in the research lab, to say that we can really fairly adjudicate a VUS or not. We can't do that, that's over-valuing what we do in research lab.
Dr. McNally: So I think, how do we consider variants among certain significance? I think it's really important to recognize that it's exactly that, it's a variant of uncertain significance. And so when you're a clinician taking that to a patient, you have to approach it from the standpoint of saying, this is a variant of uncertain significance. Which means we don't know whether it's pathogenic, but we also don't know that its benign. Because I think right now what we've seen, a lot of clinicians, and even researchers, fall into the path of this believing that variant of uncertain significance is the equivalent of benign. That's not true. It is simply ... That is a rare variant, and we don't know whether it's pathogenic or non-pathogenic. And hopefully overtime we will learn more to better assess that, and better provide the interpretation of what that means in the context of that patient.
Dr. McNally: It's a good conversation to have. It's important to recognize they're not necessarily pathogenic, but they're also not necessarily benign.
Dr. Lin: Mm-hmm (affirmative). So do you see a role, for example, when you see this variant of uncertain significance, is there a role to go back into lab, for example, and try to knock that mutation to IPSC's and test to see if its pathogenic? Or is that going a step too far?
Dr. McNally: In some cases, that is the right thing to do. Because genetics is so powerful, genetics doesn't only give you the association of a gene with an outcome, and GWAs was fabulous at doing that ... giving us a lot of variants, and often nearby genes, sometimes far away genes, but linking genes to phenotypes, and that's very powerful. But specific variants can actually tell you a lot about mechanism, about how a gene and protein actually function, and how it functions when it's broken. And so, particularly where you can gain a lot from the research front in understanding mechanism, then I think it's really powerful to take those things to the laboratory and to use that to learn about mechanism.
Dr. McNally: Sometimes you can do it to help adjudicate whether something's pathogenic or not, but again, I think we want to be cautious in doing that. Because what we do in the res ... I always like to say, "What we do in the research lab isn't exactly CLIA certified."
Dr. Lin: Right.
Dr. McNally: There isn't anything magical about what we do, but we definitely ... It is so powerful what's available out there in terms of the genetic variants, and teaching us about how genes and proteins interact. And so I think it is such a rich resource of information right now. The things I bring back to the laboratory, and get my students and trainees excited about working on, is usually where I think we can gain something new about mechanism.
Dr. Lin: Right, right, right. Since you are a role model physician scientist, and you think about questions in lab that will ultimately benefit your patients, and you are a genetic cardiologist. What are your thoughts on doing genome editing as a possible therapy for your patients? It's a little bit of a loaded question [crosstalk 00:21:51], it's a little bit controversial.
Dr. McNally: So I think, no doubt CRISPR/Cas9 gene editing is transforming what we do in the research setting. It's a fantastic tool. Is it a perfect tool? No. Anybody who has been using it a lot in the lab knows that it is much better than anything we've had before, but still quite limited in fidelity and efficiency. And so imagining that we are going to do that in patients is still pretty daunting to me. We do enough gene editing in cells to know that you have to select through an awful lot of cells before you get the one that has the exact variant you are trying to make. So that's not something we can tolerate in the human setting. But we're not there yet, we know that.
Dr. McNally: Many of the disorders I see clinically are things that are autosomal dominant due to very precise single base-pair changes. And so envisioning how we're going to correct only one copy of an allele and do in a very precise manner, we don't have those tools available yet. Now on the other hand, if you look at a disease like one of the diseases I spend a lot of time on, Duchenne muscular dystrophy, where the majority of mutations are deletions. It's X-linked, it's male, so there's only one copy of the gene, and we know a whole lot about the structure and function of the gene. We know that if we take out this other part we can skip around that mutation and make an internally truncated protein. That's actually a very good use of gene editing, because it only requires making deletions. They don't need to be very precise, and there's only one copy of it that you have to do the gene editing on.
Dr. McNally: So I see that being something in the near term that will happen, simply because the genetics positions it well to be something where that could be successful. The hard part is still how are we going to get the guides, and how are we going to get the Cas9 in safely into all the cells that need to be treated? And ultimately that lands us back at looking at what our delivery vehicles are. Which at this point in time is still viral delivery, and still has a lot of issues around can we make enough of it? Are people immune to it? So all those questions that come with viral delivery. So still lots of hurdles, but you can see some paths where it makes sense to go forward.
Dr. Lin: Very interesting. Okay great. Well thank you for providing your thought on human genetics and genome science. We're going to switch gears for the last portion of this podcast, and talk about your thoughts on career development issues for young investigators. At a recent AHA Scientific Sessions meeting, you participated on a panel that was assembled to provide advice to early career scientists. When you were starting out, what were some of the biggest challenges you faced when you were transitioning to independence and building your own lab, and what's your advice to those facing the same challenges today?
Dr. McNally: Well, even though I did it quite a few years ago, many of the things are still the same. Transitioning to independence, I think is easier if you pick up and move and start in a new place. I think it's much easier to establish your independence when you're not in the same place as your mentor. That said, we have many more people who now stay in the same place as where their mentors were and we have many more approaches towards doing that. So I think people are much more open to both possibilities as being ways of doing that. But at some level it still comes down to starting your own lab, and you hopefully have been given some start-up resources and you have to think about how to wisely spend them, and how to really get things going. I don't think this is changed either.
Dr. McNally: I usually tell people, don't just start in one area, if you can, start in two areas because things don't work, and sometimes things do work. In reality when you look at people who are successful, they're often working in more than one area. And so the sooner you start getting comfortable working in more than one area, that's a good thing. Now ideally, they should be areas that have some relationship to each other, and then feed each other in terms of information so that they grow off each other. But what does that practically mean? I always say, "Well if you can hire two people and start them on two different paths, that's a really good way to get going." And practical things like look at all different kinds of private foundations and things like that for getting some good pilot start up money to help develop new projects in the lab. And always be looking at how can these projects help me develop a bigger data package, that's going to put me in a good competitive position for example bigger grants and federal funding, and things along those lines.
Dr. McNally: Very much a stepwise process. People want to shoot for the moon and get the biggest things first, but sometimes just focusing on the smaller steps which are definitely achievable and building your path towards those bigger steps is the smarter way of doing it.
Dr. Lin: That's great advice. You also mentioned recently that young investigators should try to have as many mentors as possible. What advice would you have for, in particular, early career genomics investigators, for finding these mentors passed the postdoc phase? Some of us get introduced at the postdoc phase to maybe some other collaborating labs, but those are really collaborations of our mentors per se.
Dr. McNally: Well I think especially in the field of genetics and genomics, collaboration is key, and I will say one of the things that has changed over since I started doing this is there is a lot more understanding of the need to collaborate. Not so many years ago, it wasn't really an independent investigator went and started a lab, and it would be your trainees and the papers would have only those people on it.
Dr. McNally: I think these days, the best science is where you've tackled a problem from multiple different directions, one or two of those being genetics, genomics directions. And then sometimes there's other ways that you've approached that scientific problem. And by necessity, that usually involves collaborating with other people. And your role is sometimes to be the coordinator of all those collaborators, and that's where again you might be in a senior author position then doing that. But your role sometimes is to be the good collaborator. And so when I look at people being successful right now, seven, eight years in to running their own lab, I like to see that they've been the organizer of some of those, that they've collaborated with people who are even senior to them, and that they've established those good collaborations, but that they've taken the leading role in doing that. But also that they've had middle author contributions, that they've been a good collaborator as well.
Dr. McNally: And so, part of that is not being afraid to collaborate, and to recognize the value of doing that. And what's so great about doing that is you can collaborate with people at your same institution where you are, but you can also collaborate with people all over the world, and I think that's what we do. You go to where you need somebody who is using a technique or an approach that really helps answer the question you want to have answered. And so that's reaching out to people and really establishing again that network and good collaborators which you can do by a whole bunch of ways. You can do it by meeting somebody at a meeting, scientific meeting. You can do it through emails, phone calls, Skype, all sorts of different ways that you can reach out and collaborate with people.
Dr. McNally: It is easier than ever to share data and share ideas, but that negotiation of how to establish the terms of the collaboration and how to make it be successful is a critically important part of being a scientist. And what we now know when we look at the promotion process, is people who do that effectively, that's a really important mark of being a successful scientist, and marks them as somebody who should be promoted through the process. So great.
Dr. Lin: Yeah. No I agree. Certainly with the direction science is moving, it's definitely very difficult to work in a siloed manner.
Dr. McNally: Yeah. Well you won't get very far. You'll be able to have some really good first ideas, and show some proof of principle approaches. But to really, really address an important scientific problem, we know that you have to see those signals using multiple different methods. And once you have five different ways showing you that that's the right answer, then you're much More confident that you've gotten to the right answer.
Dr. Lin: Right. Alright, so I think we're going to wrap up. Do you have any other final thoughts for any other young investigators or genomics researchers listening to this podcast?
Dr. McNally: It's a great time to be doing genetics and genomics, and particularly human genetics, where we now finally have all this information on humans, and we'll have even more of it in the future. So I think humans are coming close to becoming a real experimental system.
Dr. Lin: Excellent. Alright well thank you so much for your time. It was a pleasure having you on this podcast.
Dr. McNally: Great. Thank you for doing this.
Jane Ferguson: As a reminder, all of our original research articles come with an accompanying editorial, and these are really nice to help give some more background and perspective to each paper. To read all of these papers, and the accompanying commentaries, log on to circgenetics.ahajournals.org. Or, you can access video summaries of all our original articles from the circgen website, or directly from our YouTube channel, Circulation Journal. And lastly, follow us on Twitter @circ_gen, or on Facebook, to get new content directly in your feed.
Jane Ferguson: Okay, that is it from us for June. Thank you for listening, and come back for more next month.