false
Catalog
Genetics and Arrhythmias: Beyond Mendel's Peas
When Genetic Burden Reaches Threshold (Presenter: ...
When Genetic Burden Reaches Threshold (Presenter: Connie R Bezzina, PhD)
Back to course
[Please upgrade your browser to play this video content]
Video Transcription
And we'll introduce the first speaker. Connie Bezina is going to be talking about when genetic burden reaches threshold. She's coming to us from the Netherlands. And we will leave and go sit down. Good morning, everybody. I think it's loading. So the title of my talk is When Genetic Burden Reaches Threshold, and it's all about genetic architecture and how our understanding or our vision of the genetic architecture of inherited cardiac disease has evolved in the last years. So we think that there is a spectrum of genetic architecture underlying different forms of genetic susceptibility to southern cardiac death. At one end of the spectrum, we have pure monogenic disorders, where the inheritance of a very large effect genetic variant by itself largely determines risk. I think there are very few of these disorders. So in pedigrees such as this, where you see a one-to-one relation with the inheritance of the genetic defect and having the symptoms associated with that mutation. And disorders that we can consider in this category would be disorders like Timothy syndrome, the Kalmodjian-Lenopathy's, Djervalensis syndrome, where you have genetic defects that because of they're so strong and so penetrant will also are evidenced from birth or early childhood. A great bulk of the inherited cardiac disorders or of the more common ones are probably can be classified as near monogenic. So these would be disorders where a large genetic defect is operative, but that genetic defect, its penetrance is modulated by the inheritance of additional genetic factors. So over here, you have this large effect variant here, which is shown in yellow, and the effect of that variant is modulated by the co-inheritance of a few other modulators. Disorders like such as this are probably, we believe, are disorders like the Lanketi syndrome and catecholamine-induced polymorphic VT, where you do see familial clustering, but within families, you see variability. So over here, we have mutation carriers who have the phenotype and mutation carriers such as this one over here, who would have the familial mutation but won't have the disease or would have it in a less severe form. On the other end of the spectrum would be the oligogenic or polygenic disorders, where in this case, you need to accumulate a large number of variants to reach the threshold of the disease. I have to highlight that I'm only talking about the genetic underpinnings of these disorders, but likely, at this end of the spectrum over here, the polygenic inheritance would also have a very large component of environmental effects. For instance, over here, I show Brugada syndrome and earlier polarization syndrome as examples of this end of the spectrum, but there would be, for instance, VF in the setting of heart failure or VF in the setting of ischemia, where you would have this kind of inheritance but also a very, very large contribution by environmental factors, and that makes those genetic factors so difficult to track down. So monogenic disorders are caused by very, very rare variants, highly penetrant variants, which are at this end of the spectrum. So what we see here is the effect size of genetic variants that are present in the germ population plotted against their frequency. So the genetic variants causing Mendelian disease are sitting on this end of the spectrum. We believe that genetic variants that are operative in the setting of oligogenic and polygenic inheritance are situated at this part of this distribution over here. So we would have the common variants that have a small effect size or common in germ population, so they cannot have a large effect size because of purifying selection, and also we would have the low-frequency variants of intermediate effect. Now both of these, we believe, are important for oligogenic and polygenic inheritance, but simply because of the fact that these are so more easy to genotype in large numbers because of issues of costs but also statistical power, these common variants are the ones that have been mostly studied in the setting of polygenic and oligogenic inheritance. So how do you do this? One design is a case-control design where you would genotype your cases and your controls for tens and thousands and millions of SNPs spread throughout the genome, and then basically at each point, at each single nucleotide polymorphism, you compare the frequency between the cases and the controls. This is how you visualize the data in the form of a Manhattan plot where each dot on this plot represents one single nucleotide polymorphism, and there is always this magical threshold over here at a p-value of five times 10 to the minus eight, which represent the Bonferroni-corrected p-value threshold. And anything above that line would represent a robust association, a genome-wide statistical significance. Now, the search for common genetic variants that modulate risk in the setting of polygenic inheritance has had most success in atrial fibrillation, and this has largely to do with the fact that it can be reliably scored, and it affects a very, very large portion of the general population, and people live with atrial fibrillation. So this is the large, latest mega-analysis of meta-analysis of GEO studies of case-control studies of atrial fibrillation. This represents more than half a million people with AF identifying almost 100 loci. We've been doing work on oligogenic inheritance in Brugada syndrome. We've published in 2013 these two chromosomal regions. A more recent effort extending the set to almost 3,000 probands has allowed us to identify nine additional loci, bringing the total of chromosomal regions that are now associated with Brugada syndrome in the setting of an oligogenic or maybe polygenic inheritance to a total of 11. We've been looking at the cumulative effects of these variants, so now instead of doing a case-control, typing one variant at a time, we calculate the burden of risk alleles within each individual, and then we relate that with case-control status. And what we see is that people who have very high burden of these Brugada syndrome-predisposing alleles have a remarkable increased risk as opposed to others, which are on the other end of the spectrum. So what I've been showing you, these two examples, are examples of case-control studies that have compared patients with arrhythmia to controls. What has also – so that would be making a link between the genetic variant directly and the ECG phenotype and or the arrhythmia. But another way to go would be to try to understand the genetic underpinnings of the intermediate phenotypes that underlie arrhythmia, such as electrical phenotypes, structural phenotypes, or perhaps even extracardiac phenotypes like hypertension. And once we have established those links, those SNP phenotype associations, then test those for association with arrhythmia. One area where a lot of work has been done is relating common genetic variants to electrophysiological traits as measured by the ECG. And this has been done for the QT interval, PR, QRS, heart rate, et cetera. And what I show you over here is the largest, latest GWAS for the QT interval in 75,000 people that identified 68 loci. So as I said before, common variants have a small effect size, so we need to start looking at them in a cumulative way and then try to start linking them to important phenotypes that could be of clinical relevance. So as I said, this link has been established for those SNPs underlying the QT. So what can we do with that information? And I think that one important study that starts opening the door for the clinical applicability of these polygenic risk score is a study that was published by Chris Newton Shea in Circulation in 2017. So what Chris did, he demonstrated that the genetic risk score based on the burden of QT-prolonging alleles could predict drug-induced QT interval prolongation. So what you see over here is an example of two individuals in this distribution. So we have this individual over here at the high polygenic risk score burden, who has a high, a larger degree of QT interval prolongation, being a high responder over here, as opposed to a person who has a low genetic burden of QT-prolonging alleles, who has a low response upon dofetilide in this case. So they showed this for multiple drugs. More recently, we've done some work on a similar concept, this time with cardiotoxicity for sodium channel blockade. We constructed polygenic risk scores based on SNPs that had been linked to QRS interval and PRS, PR interval in general population, and tried to see whether they could predict conduction slowing in the setting of sodium channel blockade. So this work was done by Rafik Tadros and my group in 1,400 patients that had been given azomalene because of suspected Brugada syndrome. And I show you here the data for the predictive ability of a polygenic risk score based on QRS SNPs. And we could demonstrate in a multivariate analysis that a polygenic risk score based on QRS-prolonging alleles could predict response to sodium channel blockade in this population. Of course, what we also did in this group, we also tested the predictive ability of a polygenic risk score based on SNPs predisposing to Brugada syndrome, and then refining it further based on other clinical parameters such as QRS duration at baseline or a family history of Brugada syndrome. And we showed that such a score could predict the development of a type 1 Brugada ECG in response to azomalene with a nice, perhaps clinically relevant predictive ability. So in summary, many electrophysiological phenotypes have a complex genetic architecture. GWAS has started to uncover the cruel contributory genetic factors. Recent studies have so far focused on common genetic variation, but I think as the price of whole genome sequencing starts to decrease, we'll probably also hopefully start seeing studies that will also look at the low-frequency variants. And I think the challenge now on the next frontier is, how are we going to use this information to help us in recertification of patients? Thank you. So we have a couple minutes for questions, if anyone wants to step up to the mic. And our panelists. That was an excellent talk, and I think this is an important, the concept of GWAS, which we don't usually get to do in the pediatric population because we have such limited numbers. But my question is the applicability and your thoughts. I'm just going to give you an example, but the data from the GWAS from AFib studies, which is what we have. So if you think about that, if we've got, you know, multiple SNPs or alleles that we think are common and are going to additive, the child or the 30-year-old may present earlier, how does that affect our pediatric patient who presents at, let's say, 12 years of age with AFib? Is that someone who's got multiple, you know, SNPs? Or are we talking about a different modulation of a child with AFib? And how do you, you know, what do you think about that in terms of GWAS in our pediatric population? Yeah. So basically, maybe my diagram was to categorize. So I also think that within a particular disorder, you also have different genetic architectures. So that kid who presents with AF at 12 years of age will probably have a different genetic architecture. He will probably have a more penetrant, rarer genetic variant that would predispose to that. So, for instance, along the same lines, we've also been looking at the role of common genetic variation for susceptibility of Long QT syndrome. We've done a case control design, and then we compared the distribution of polygenic risk scores between mutation carriers and controls and Long QT syndrome patients, which are genotype negative. And what we observed also that those people, the genotype negative Long QT syndrome probands, have again an increased polygenic burden of these common variants. So I think even within the same disorder, we have different genetic architectures. So do you think this is the answer to the 20 or 25 percent of genetically elusive Long QT syndrome? Yeah. So also what we did, when you have genome-wide genetic data, you can also calculate the proportion of the variability, how much common variants explain variability. And what we found that in mutation positive individuals, common variants explain 10 percent of variability, whereas in mutation negative, they explain almost 20 percent of variability. So I think – and that's important for clinical management, because – Right. How would we treat them? Yeah. I mean, is this going to inform us on our therapies? It's important, because – for the relatives in particular, because they will be probably, as I understand it, at less risk, as opposed to – yeah. Should we be considering SNP arrays in our patients that are gene negative? Not yet. We don't know. We need – we know – yeah. We know only the 17 percent. We need to figure out the rest. I heard some moaning from the audience after that. You had a counselor's moan. Any questions from the audience? All right. Thank you very much for the excellent – Thank you.
Video Summary
The speaker discussed the genetic architecture of inherited cardiac diseases, specifically focusing on monogenic, near monogenic, and oligogenic/polygenic disorders. Monogenic disorders are caused by rare, highly penetrant variants, while near monogenic disorders involve a large genetic defect that is modulated by additional genetic factors. Oligogenic/polygenic disorders require the accumulation of multiple variants to reach the disease threshold. The speaker also mentioned the use of genome-wide association studies (GWAS) to identify common genetic variants associated with electrophysiological traits. These findings could potentially be used to create polygenic risk scores that predict drug-induced QT interval prolongation and other cardiac phenomena.
Meta Tag
Lecture ID
6683
Location
Room 203
Presenter
Connie R Bezzina, PhD
Role
Invited Speaker
Session Date and Time
May 09, 2019 10:30 AM - 12:00 PM
Session Number
S-013
Keywords
genetic architecture
inherited cardiac diseases
monogenic disorders
near monogenic disorders
oligogenic/polygenic disorders
Heart Rhythm Society
1325 G Street NW, Suite 500
Washington, DC 20005
P: 202-464-3400 F: 202-464-3401
E: questions@heartrhythm365.org
© Heart Rhythm Society
Privacy Policy
|
Cookie Declaration
|
Linking Policy
|
Patient Education Disclaimer
|
State Nonprofit Disclosures
|
FAQ
×
Please select your language
1
English