NEW YORK (GenomeWeb) – Name: Elizabeth McNally
Title: Director of the Cardiovascular Genetics Clinic and
the Institute for Cardiovascular Research, University of Chicago;
A.J. Carlson Professor of Medicine and Human Genetics, University of Chicago
Experience: Clinical training in internal medicine and cardiovascular disease, Brigham and Women's Hospital;
Fellow in genetics, Children's Hospital in Boston and the Howard Hughes Medical Institute
Education: MD and PhD, Albert Einstein College of Medicine
The field of cardiology may be an optimal clinical setting to incorporate next-generation sequencing, according to Elizabeth McNally, director of the cardiovascular genetics clinic at the University of Chicago.
Discussing her thoughts in a recent Science Translational Medicine editorial, McNally said that certain facets of cardiology and cardiovascular disease enable NGS-based testing to be introduced in a more straightforward way than complex diseases like cancer, and could provide guidance on best practices for adopting the technology in medicine more broadly. McNally elaborated on those thoughts in the following conversation with Clinical Sequencing News, which has been edited for length and clarity.
Can you tell me some background about yourself and what prompted you to write the editorial in Science Translational Medicine?
I run a cardiovascular genetics clinic, which I've done for about 15 years. In the last 10 years or so, we've had a steady influx in the availability of genetic tests for a variety of inherited forms of cardiovascular disease. Some are disorders that affect the heart muscle like cardiomyopathies, some affect the vascular system like Marfan's syndrome or inherited aortic diseases, and there are some that cause irregular heart rhythms like long QT syndrome and other inherited arrhythmias. As the years have gone by we now test more and more genes in a panel. The most extreme [example] is for cardiomyopathy testing, where we're now ordering a test for about 80 genes.
When we do that, we get more and more answers for people and are able to use that information in their care, both from a prognosis standpoint to identify people who are at risk for developing it, and in some cases to take preventive steps. It's been very interesting to watch this evolution of clinical genetics that's been quietly going.
How has genetic testing in cardiology changed over the last 10 years, and how will it change going forward?
Ten years ago, we were doing a bunch of single genes, one at a time, which wasn't very efficient. Those used Sanger sequencing. The first [multi-gene] test was done in an array-based format looking at around five genes. It took a couple of years before next-generation sequencing took place in panels. And that's what happening now. You can see the natural evolution, because with the time and cost and effort that it takes to do 80 genes now, one can easily do an exome. I think that will be the direction that it goes — to do broad-based testing, but then to do analysis of the subset of genes that you're most interested in.
How are cardiologists responding to the availability of these multi-gene panels? Are they embracing the technology or has it been a challenge to get them to adopt the panels?
There are a number of barriers to using [next-gen sequencing] in clinical care. One is getting the cardiologist to order the tests. Most are not comfortable ordering them because a lot of the times the interpretation can be complex. It often requires working with the entire family to get at the bottom of what some of the mutations really mean. It requires people that are knowledgeable both in genetics as well as in cardiology.
One of the things we're interested in at the University of Chicago is helping people get more comfortable and learn how to do this. From an education standpoint, that's going to be something that we work into our training programs.
In your editorial, you mention that testing based on next-gen sequencing may be especially suited to the field of cardiology. Why?
I think the interpretation is a little easier in some of these cardiology disorders. Cancer is a complicated area to work in because you have a host genome — the person who has the cancer. And the cancer has its own genome. And then the cancer often has many genomes across the tumor that change in space and time.
We don't have to deal with quite that complexity in the cardiovascular world. We're dealing with relatively straightforward germline mutations. And [the genome] is pretty stable within the heart or the vasculature system so we don't have to struggle quite as much with interpretation.
You also mention that many of the genes that are involved in cardiovascular disorders are expressed only in the tissues involved in those disorders. Can you elaborate on the importance of that?
A number of genes that we work with in cardiology, [for example] the genes involved in hypertrophic cardiomyopathy, those genes encode the structural proteins of the heart. So they are very highly expressed in the heart. In addition, there's also wonderful structural data known about them and there's incredible conservation of these genes and proteins across species and evolution. So it makes it much easier to interpret single amino acid changes and predict whether they are likely to be pathogenic.
What have been some of the biggest challenges in incorporating NGS-based testing in cardiology?
As we're doing bigger and bigger panels, we're often finding that there is more than one variant that is pathogenic. What we used to think of as single-gene disorders will, probably like everything else in the world, prove to be a little bit more complex than what we initially thought.
And that actually makes sense with what we see clinically because we may take care of a family whose members all carry the same mutation, but have a wide range in how they present. We think that some of that phenotypic variability is coming from second hits that are present in some people and not others and may explain why some people have more severe disease than others. As we do broader and broader testing, we may come to understand that better. If you're just looking at the test from the standpoint of an individual patient, the results of that genetic test can be complicated to interpret and that's why a lot of times we end up working with whole families so we can put things in context better.
What has been your experience with reimbursement for NGS-based tests?
It's actually getting better and better. It's not too bad and improving all the time. In cardiology, we're dealing with disorders that, in their worst case, are life-threatening. They can be associated with irregular heart rhythms or a ruptured aorta, and the nice thing in cardiology is there are things we do to prevent those things from happening. We monitor the disorders, offer surgery, put devices in, put people on medications. I think the insurance companies have really recognized that this genetic information is very valuable not only for the management of that individual patient … but also that genetic information is incredibly helpful for all of their family members. The first person in the family who gets the testing gets the broader-based testing, which can be more expensive because it's looking at a lot of genes. But then when we start working with the family members, we're usually doing more site-specific testing where we're just looking for the one or two mutations that have been found in the family. That's far less expensive and incredibly meaningful for managing relatives and identifying people early in the stage of disease and even making life-saving interventions. That's pretty hard to argue with from an insurance standpoint. And they've got that and see the value of this testing.
Earlier, you mentioned that the evolution of NGS would move into exome and whole-genome sequencing, but with analysis limited to a smaller set of genes. Why do you see that as the next progression and what are the advantages of that type of testing versus a gene panel?
Right now … a gene panel for 100 genes is reasonably expensive to set up and get it working well. You can do an exome now for far less money, and even a genome for less money in a research setting, than these panels.
If all testing relies on the same input test, whether it's a whole exome or a whole genome, from a reproducibility standpoint, that becomes easier to set up. You're doing the same test on the front end — the DNA sequencing. Then the analysis is just a question of making sure you're looking at the right genes and you can focus your analysis on the genes of interest. The cost of doing a gene panel can be anywhere from $6,000 to $10,000. That's the list price billed to an insurance company.
I think what we still have to prove with exomes and genomes is that we're covering the genes in question with good depth.
How are you using NGS-based cardiology tests at the cardiovascular genetics clinic at the University of Chicago?
We see patients and families who are referred to us from all over the place. We do the actual [NGS-based] testing through DNA testing laboratories; we don't actually do that here. We do have a DNA diagnostics lab at the University of Chicago and we'll do the testing wherever it makes the most sense and offers the genes we're interested in having tested, whether it's the DNA diagnostics lab at the University of Chicago, GeneDx, the Laboratory for Molecular Medicine, Ambry Genetics. There are a bunch of different companies that offer testing that are CLIA certified labs that can then bill insurance for the testing.
What do you think is needed for NGS-based to tests to be used more widely in cardiology?
Most cardiologists practicing these days have not really had much training in genetics. If the test comes back with a really obvious answer, then I think that's not too bad for a physician who really doesn't know much genetics, and the doctor can use it in the care of patients and hopefully then knows to screen family members.
Our big issue is that we have to get people trained in thinking about genetics. In medical schools and residencies, we have to start thinking about where that training comes in because it's going to be part of how all of them practice medicine. We do that here and hope to do more of it.
In addition, genetic counselors are incredibly important in this process and right now I think we probably don't have enough genetic counseling programs in the United States. I see a big increase in the need of well-trained genetic counselors. Cardiovascular genetic counseling is really almost a specialty now within genetic counseling. When genetic counselors get very familiar with the diseases in cardiology, that helps them do better interpretation and work with families better.