Ep19 August 2018

Getting Personal: Omics of the Heart

Jane Ferguson:

Hello. Welcome to episode 19 of Getting Personal: Omics of the Heart, the issue from August 2018. I am Jane Ferguson, and this podcast is brought to you by the Circulation: Genomic and Precision Medicine Journal and the American Heart Association Council on Genomic and Precision Medicine. Before I dive into the papers from this month, a reminder that early bird registration for AHA Scientific Sessions runs until September 4th, so go register now if you haven't already to take advantage of reduced rates. The meeting will be held in Chicago from November 10th through 12th, and it's the first year of the new three-day meeting format. It's already promising to be a really great meeting, and I'm hoping to see a lot of you there.

The August issue has a number of really interesting papers. First up, Gardar Sveinbjornsson, Eva Olafsdottir, Kari Stefansson, and colleagues from deCODE genetics-Amgen report that variants in NKX2-5 and FLNC cause dilated cardiomyopathy and sudden cardiac death. This team leveraged available DNA samples from the Icelandic population to carry out a genome-wide association study in 424 cases of dilated cardiomyopathy and over 337,000 controls. They applied whole genome sequencing to all of these samples, allowing them to identify common and rare variants. In total, they tested over 32 million variants.

They found two variants that were significantly associated with DCM at genome-wide significance, a missense variant in NKX2-5 and a frameshift in FLNC, both associated with heart failure and sudden cardiac death. Further, the NKX2-5 variant was associated with atrioventricular block and atrial septal defect. Although these variants are rare and not documented in other populations, they are significant contributors to familial DCM in Iceland. Because of the unique population structure of Iceland and known genealogy, the researchers were able to trace the NKX2-5 variant back to a common ancestor born in 1865. They traced the FLNC variants to a common ancestor born in 1595.

While the specific variants identified in this study may not be present in other populations, they are located in genes with known relevance for cardiac function. NKX2-5 encodes a cardiac transcription factor, which is required for embryonic cardiac development, and other variants in this gene have been associated with cardiac dysfunction in other populations. FLNC encodes filamin-C, a muscle cross-linking protein. Variants in FLNC have previously been ascribed to associate with myofibrillar myopathy, muscular dystrophy, and cardiomyopathy. This study adds to our knowledge of the genetics of dilated cardiomyopathy and supports screening for NKX2-5 and FLNC variants, particularly in the Icelandic population, which would allow for early intervention and monitoring in carriers.

Staying with the topic of dilated cardiomyopathy, Inken Huttner, Louis Wang, Diane Fatkin, and colleagues from the Victor Chang Cardiac Research Institute in Australia report that an A-band titin truncation in zebrafish causes dilated cardiomyopathy and hemodynamic stress intolerance. We actually talked to Dr. Wang about this research last year when he was presenting this as a finalist for the FGTB Young Investigator Award. You can go back in the archives to episode 10 from November 2017 if you'd like to hear more.

Titin mutations are responsible for a large number of cases of dilated cardiomyopathy, but there are also individuals with titin mutations that remain asymptomatic. This group used zebrafish as a model of human titin mutations and generated fish with a truncating variant in the A-band of titin, as has been identified in families with DCM. They found that homozygous mutants had a severe cardiac phenotype with premature death, but that heterozygous carriers survived into adulthood and developed spontaneous DCM. Prior to onset of DCM, the heterozygous fish had reduced baseline ventricular systolic function and reduced contractile response to hemodynamic stress, as well as ventricular diastolic dysfunction.

Overall, the mutant fish displayed impaired ability to mount stress responses, which may have contributed to development of disease. Extrapolating this to humans, this could suggest that hemodynamic stress may be a factor that contributes to timing and severity of disease in individuals with titin variants. Hemodynamic stress can be exerted by exercise, pregnancy, and other diseases contributing to ventricular volume overload. Modifying these hemodynamic stressors in at-risk subjects could potentially help to modulate the severity of DCM phenotypes.

Moving on to the topic of coronary artery disease, Vinicius Tragante, Daiane Hemerich, Folkert Asselbergs, and colleagues from University Medical Center Utrecht in the Netherlands report on druggability of coronary artery disease risk loci. This group was interested in using results from genome-wide association studies for CAD to identify new targets that may be amenable for drug repurposing. They used results from published GWAS for CAD and created a pipeline to integrate these loci with data on drug-gene interactions, chemical interactions, and potential side effects. They also calculated a druggability score based on the gene products to prioritize targets that are accessible and localized to increase the chance of a drug being able to find the target without affecting core systemic processes or housekeeping genes.

Their pipelines allowed them to identify three possible drug-gene pairs, including pentolinium to target CHRNB4, adenosine triphosphate to target ACSS2, and riociguat to target GUCY1A3. They also identified three proteins to be prioritized for drug development, including leiomodin 1, huntingtin-interacting protein 1, and protein phosphatase 2, regulatory subunit b-double prime, alpha). While these predictions were all made in silico and need to be extensively tested in clinical trials, the pipeline did identify many current therapies for CAD and myocardial infarction, including statins, PCSK9 inhibitors, and angiotensin II receptor blockers. These positive controls support that this method can successfully discover effective CAD therapies.

Staying on the topic of drugs, Kishan Parikh, Michael Bristow, and colleagues from Duke University report on dose response of beta-blockers in adrenergic receptor polymorphism genotypes. Two clinical trials have reported pharmacogenomic interactions between beta-blockers and beta-1 adrenergic receptor genotype in the setting of heart failure with reduced ejection fraction. In a retrospective analysis in almost 2,000 subjects from the BEST and HF-ACTION studies, the authors analyzed whether genotype at the Arg389Gly polymorphism in beta-1 adrenergic receptor, or an indel in the alpha-2C adrenergic receptor interacted with drug dose to affect mortality and hospitalization.

They found that ADRB1 genotype affected mortality in response to drug dose with less all-cause mortality in high versus no or low-dose beta-blockers in individuals homozygous for arginine at position 389, but not in individuals carrying a glycine at that position. In individuals on high-dose beta-blockers, genotype did not affect outcomes, but there was a significant difference by genotype in all-cause mortality in individuals on no or low-dose beta-blockers. These data support the guideline recommendations to use high-target doses of beta-blockers in HFrEF.

Switching gears towards precision medicine and genotype-guided approaches, Laney Jones, Michael Murray, and colleagues from Geisinger were interested in the patient's perspective. In their paper, Healthcare Utilization and Patients’ Perspectives After Receiving a Positive Genetic Test for Familial Hypercholesterolemia, they explored the impact of providing genotype test results for familial hypercholesterolemia to subjects participating in the MyCode Community Health Initiative. In MyCode, exome sequencing is conducted in participants, and results are returned for pathogenic and likely pathogenic variants in genes representing actionable conditions based on American College of Medical Genetics secondary findings and recommendations.

It is estimated that 3.5% of MyCode participants will be carriers of such variants, and this number may increase as more variants are discovered. In this pilot study, the authors screened for individuals with mutations in LDLR, APOB, or PCSK9, consistent with FH. They identified 28 individuals, of which 23 were eligible for inclusion in the study. Only five of the 23 subjects had previously been diagnosed with FH. Receipt of genetic test results led to change in medications in 39% of individuals. 96% of the subjects had previous LDL measurements, but only four subjects had ever met LDL goals. After genetic test results, three individuals met their LDL goals.

Seven individuals consented to participate in interviews about their experience. Almost all of these subjects already had a personal or family history of high cholesterol or heart disease, and all subjects felt that they were being adequately treated. Only three of the seven subjects mentioned using diet and exercise to control their high cholesterol, with most individuals being relatively unconcerned because they felt their medication was effective in controlling disease risk.

While the numbers studied here are too small for any statistical testing or inference, the paper describes the results from the interviews, including some excerpts from patients, which really highlight the complexities of returning results and of helping patients understand what their results mean. Given increasing genetic testing and returning of results, studies like this are really important to help us figure out the most effective ways to communicate results and support patients and their care providers.

Also from a patient-centric perspective, we have an article from Susan Christian, Joseph Atallah, and colleagues from the University of Alberta in Canada on when to offer predictive genetic testing to children at risk of an inherited arrhythmia or cardiomyopathy, the family perspective. This article considers the timing of cascade testing to predict inherited arrhythmias and cardiomyopathy in children of affected individuals. European and North American guidelines differ on when or if they recommend genetic testing in children.

In this study, surveys were circulated to foundations and patient groups to solicit familial perspectives on when genetic testing should be offered to children. In total, 213 individuals responded. In the case of long QT syndrome, 92% of respondents thought testing should be offered before the age of five, while 77% of respondents thought genetic testing should be offered before the age of 10 for hypertrophic cardiomyopathy and arrhythmogenic right ventricular cardiomyopathy.

Overall, the potential benefits of genetic testing, including guiding therapies, sport participation, and decreasing worry were ranked more highly than potential risks of discrimination or increasing worry that could occur from genetic testing. Overall, the responses indicated that families would welcome the option of genetic testing for at-risk children from a young age and support initiating early discussions with families to explore costs and benefits of early genetic testing.

Finally in this issue, we have a review from Paul Franks and Nicholas Timpson from Lund University and the University of Bristol entitled Genotype-Based Recall in Complex Cardiometabolic Traits. This review looks at the increasing practice of selecting samples or individuals from larger cohorts or biobanks based on their genotype to carry out additional studies. The article focuses on examples of such genotype-based recall studies in cardiometabolic disease, highlights approaches and new methods, and discusses the ways these types of studies can be used to extend and supplement randomized trials and large population-based studies.

As always, you can find all the articles, accompanying editorials, and video summaries online. Our website recently underwent some redesigns and has moved. You should be redirected if you have the older site bookmarked, but you can also find us directly at ahajournals.org/journal/circgen. Also, thanks to everyone who participated in the Twitter poll last month. You were pretty evenly split on what you want to hear in the podcast, but please continue to leave suggestions and feedback on what we're doing and where we can improve things. That's it for the August issue of Circulation: Genomic and Precision Medicine. Thanks for listening, and tune in next month for more.