Genetic Testing and Epilepsy-Who, What, Where, and When of Genetic Testing

Video Transcription

I'm going to start with a quick show of hands. How many people in the audience have sent genetic testing before on patients?  And how many of you have heard a talk on genetic testing and epilepsy before? Great. So I feel like it's a common topic, and I hope to add a little bit to that today.  So here are my disclosures. Disclaimers are that I am not a geneticist, and the genetics knowledge obtained here should  not replace consultation with a geneticist or genetic counselor. I'm a pediatric epileptologist with an interest in genetics.  Learning objectives are, at the end of this session, I'd like you to be able to decide which individuals with epilepsy might benefit from genetic testing, to appraise different  genetic testing modalities for epilepsy, and to develop a clinical pathway for genetic testing for epilepsy. So why perform genetic testing?  The genetic ideologies are very common in epilepsies. So in decades past, the view was that a majority of epilepsies were idiopathic in nature.  And now what we know is that group of idiopathic epilepsies includes those with single gene  ideologies, epilepsies with complex inheritance, as well as epilepsies that have modifiers, polygenic risk, and susceptibility alleles.  And we know now that there are many genes associated with epilepsy. This is a slide from Ingo Helbig, who we'll be talking a little bit later.  And this is from his paper in 2016, and I know there are updated versions. But basically, in decades past, there were sequencing studies.  And in the channelopathy era, most of the genes were found in ion channels. But with next-gen sequencing, and with more discoveries in copy number variants, we have  found that there are micro-deletions and many more genes involved in epilepsy. Sorry, I'm trying to advance the slide, but it's not working. Oh, there we go.  Also to remember is the genetic architecture of epilepsy. So when we send testing, it mostly covers Mendelian diseases in copy number variants.  That means single gene causes of epilepsy. And those are diseases on the x-axis here is allele frequency, y-axis is effect size of a variant.  And Mendelian genes have a low allele frequency and a high effect size. We should also remember that there are rare variants and micro-deletion syndromes that  have a moderate allele frequency and a moderate effect size. And then there are common variants, those that have a high allele frequency, low effect size.  But again, the testing that we currently have access to, you know, in the absence of polygenic  risk scores and things like that, are mostly for Mendelian diseases in copy number variants. So why perform genetic testing?  This can provide invaluable information about the explanation of the cause of epilepsy. It can end the diagnostic odyssey, lead to counseling for family planning, connect families  with disease-specific organizations, lead to targeted treatment, and open opportunities for research. There's also a lot of evidence now that genetic diagnosis influences treatment.  This is a long list of genes for which there are specific recommendations for treatment for each of those.  And there's now data that, in practicality, in the clinic, and this is stuff you can send to your insurance companies, that the genetic diagnosis leads to gene-specific treatment  recommendations in 80% of those with epilepsy at onset before the age of three. And actually, in practice, genetic diagnosis results in the adjustment of treatment in  72% of young children in one study, in 60% of children aged 0 to 18 years old. And very impressively, in that case series, 93% of those had at least a 50% reduced seizure  burden when they had a treatment change, and 12.5% became seizure-free. In adults, the yield's a little bit lower, and 17% of adults had a change in treatment.  The most recent study, and the largest study published just recently, showed that 50% of children plus adults had an adjustment in treatment.  And that might include adding a new medication to one that they're already taking, starting  a medication, referral to a specialist, monitoring for extra neurological disease, stopping a medication, changing a diet, referral to a clinical trial, a dose of medication change,  and then recommendation of surgery. Of those individuals that had a change in treatment, a large portion had an improvement  of seizure frequency, as shown on the graph there, where if the patient previously had daily seizures, they typically, or they often had decreased seizures, and some on the light  blue had no seizures. If they had weekly seizures, similarly, they would have decreased to monthly, annual seizures, or no seizures.  And those that had annual seizures often became seizure-free after that change in treatment. So what is on the menu of testing, and what test should we pick?  First and foremost, you should still start with your clinical impression and differential diagnosis, meaning in that patient that's in front of you, in the age group that they're  in, figure out what testing might be most beneficial for that patient. But the menu can include chromosomal microwaves, karyotypes, fish studies, methylation studies,  single-gene sequencing or Sanger sequencing, targeted mutation analysis, next-generation  sequencing panels, which are probably one of the most common epilepsy gene panels, exon-level  deletion duplication analysis, which is often included on epigene panels, whole exome sequencing with and without mitochondrial sequencing, and whole genome sequencing.  And we'll get into a little bit later testing algorithms and what's recommended. But to go into each of these in a little bit more detail, chromosomal microwaves had been  often the first line of testing, and that's fallen out of favor, because these detect deletions, duplications down to about 50 kilobases, and they can be a little bit higher  yield in those with epilepsy with developmental delay, and the yield's about 9%. The great things are typically covered by insurance, but they have relatively low resolution.  Gene panels sequence exons as well as the intron-exon junction of epilepsy genes, and they can range from 20 genes to hundreds of genes.  It efficiently tests for multiple genes, and the yield is about 20% based on systematic reviews. The pros are that it's an efficient way to get an early molecular diagnosis, but the  cons include access to gene panels, which is improving, and interpretation of the variant of uncertain significance.  With whole exome sequencing, you're sequencing the exons and intron-exon boundaries or junctions across the entire genome.  You can also do, some companies offer an in silico gene panel, meaning they technologically do a whole exome sequencing, but only analyze a subset of the 20,000 genes in the genome,  and those that are relevant to epilepsy. The yield is about 25%, which is higher than a panel, and it's nice because you can also  do reanalysis, meaning if you don't get a diagnostic result at the start, a year or two down the line, you can reanalyze those data to see if there's new information that  would then lead to a diagnostic result. The cons include interpretation of variants of uncertain significance, access, higher  yield, and limited, or sorry, that's not a con, higher yield is a pro, but there's limited insurance coverage as well.  We rarely do karyotypes anymore, but I should point out that the pros of karyotype are that  it can detect balanced translocations, whole genome sequencing can as well, but ring chromosomes are the one thing that only karyotype can detect, and none of the other tests can.  Whole genome sequencing, which will be hopefully more and more common, sequences the entire  genome and includes the introns and intergenic regions, it can get structural variants, it can get some repeat disorders, but not all of them, and the yield is about 48%.  So it's higher yield than a panel, and higher than exome, and reanalysis is also possible. The con is that there's a lot of areas of the genome in the intronic and intergenic  regions that aren't well characterized, so you'll get a lot of variants of uncertain significance and insurance coverage is an obvious issue still. Sorry, I'll just go back.  So there's a study that showed that whole exome and epilepsy panels are more cost efficient  than chromosomal microarrays, so those chromosomal microarrays typically shouldn't be the first  line of testing anymore, and it's really pushing towards epilepsy panels and whole exome and whole genome sequencing.  But remember that there are special cases that aren't covered by your genetic testing, and why phenotyping the patients and getting a great family history is still really important.  Myoclonus epilepsy with ragged red fibers and PCH19 have specific modes of inheritance that might point you to the diagnosis.  As I mentioned, ring chromosome 20, you would need a karyotype for that diagnosis. The ARX gene has a recurrent 24 base pair duplication that's not well covered by exome sequencing.  CSTB gene, or Unrich-Limburg syndrome, has a repeat expansion in the 5 prime UTR that isn't covered by a lot of gene panels.  And familial adult myoclonic epilepsies often have large intronic repeats that aren't covered in genetic testing.  So there is a very important paper published just at the end of October, and finally I feel like we have a evidence-based clinical practice guideline for genetic testing put  out by the National Society for Genetic Counselors and endorsed by the American Epilepsy Society. And this is the recommended genetic testing strategy.  So to start with, you ask, does the patient have an epilepsy phenotype associated with a single gene or gene class etiology? If yes, you go to targeted testing.  And with that, if you get a diagnosis, you stop. If you don't get a diagnosis, you go to exome sequencing or genome sequencing.  If that's not diagnostic, you go down and get a CNV analysis. If that's not diagnostic, you reanalyze exome genome sequencing.  If there isn't a specific phenotype pointing you to a specific targeted testing, then it's recommended to do as a first line exome or genome sequencing.  And if that's negative, then go to CNV analysis. That's negative, go to exome genome reanalysis.  It's also acceptable to do a large multi-gene panel with deletion and duplication analysis. And with that, if that's not diagnostic, then you go for exome and genome sequencing.  So who should receive a genetic evaluation? In this guideline, the underlying message is that any unexplained epilepsy at any age should receive a genetic evaluation.  To break that down a little bit further, there are special cases or particular cases where it can be more indicated, more helpful.  So to confirm an identified or suspected genetic syndrome, like tuberous sclerosis, Gervais  syndrome, or familial self-limited neonatal epilepsy, there are features that might suggest  a genetic syndrome, like dysmorphisms, congenital anomalies, brain malformations, early onset epileptic encephalopathies, and epileptic spasms have a particularly high yield for  genetic diagnoses. In the case of myoclonic epilepsies, when a genetic diagnosis may affect treatment or predict prognosis would be important.  Of note, it's relatively low yield with genetic generalized epilepsies, which is kind of funny  because we call them genetic generalized epilepsies, but for the most part, they have complex inheritance.  It's also relatively low yield with non-acquired focal epilepsies, except with those with a family history or defined genetic syndrome.  I often send genetic testing if the genetic generalized epilepsy seems to be refractory to treatment.  It's important with epilepsy surgery evaluation, which I think the course will get into later. Adults with intellectual disability, the gene panel yield is 23%.  So often I feel like the adult epilepsy clinics have less genetic testing being sent, but something that we should think about, and it's known that there's not as high of a yield  compared to sending testing in children, but there's definitely cases that genetic testing is very helpful in adults as well.  In children with onset less than three years, the gene panel can be 37% yield, in spasms about 30%, and in neonate with epileptic encephalopathy, the yield is up to 83%.  So where should patients receive a genetic evaluation? There's a variable degree of genetic counseling and epilepsy genetics expertise, and I think  that's just a reality that we are grappling with and have to deal with as a society of epilepsy care providers.  To get a show of hands from people, I saw that a lot of people have sent genetic counseling, or sorry, sent genetic testing.  Can you put your hands up again if you've sent genetic testing from your clinic? And can you keep your hand up if you've used a genetic counselor with that testing?  How many have used a medical geneticist with genetic counseling with that testing? Great. And how many do the genetic counseling themselves? Great. Excellent.  So there's quite a bit of variability in how this is done, but I think it's very important, of course, as we heard from Emily, that genetic counseling be done.  So as a part of this clinical practice guideline, they state, genetic tests should be selected, monitored, and interpreted by qualified providers with pre- and post-test counseling.  And that's important for many reasons. The genetic counseling can help the patients and families understand the selection of the  tests, the potential outcomes, whether that be a diagnostic test, a non-diagnostic test, what variants of uncertainties might come out.  And also with post-test counseling, variant interpretation and what to do with that information, connection with clinical trials, connection with family and patient groups.  So in maybe an ideal world, a clinical workflow for genetic diagnosis would include starting with a new patient here, and then with those data, the phenotype data, you'd have human  phenotype ontology and a clinical workup. And if genetic testing is indicated, you do pre-test counseling, obtain that sample,  and run that genetic testing along with the phenotype data to try to figure out whether you have a diagnosis, re-evaluate the patient, have genetic counseling, recurrence risk,  and prognosis. And if that case is unsolved, you might go to additional testing. If the case has diagnosis, then those data would be shared, and that's a way that we  can get precision medicine much more accelerated in our society. If there's a known functional effect, then there can be functional studies, basic science  studies to figure out whether that's a gain of function or loss of function, for example, with the ion channel mutations.  And with that, that can lead to precision therapy, and again, back to more data sharing. So I hope after this, we've learned a few things.  One is that you should still choose your genetic evaluation based on clinical phenotype, and  that whole exome and whole genome sequencing should typically be first-line genetic testing for epilepsy, depending on availability and local standards, but a multi-gene panel is  the next best option. Individuals with additional findings, like intellectual disability, autism, dysmorphic  features, and early onset of epilepsy have higher yields with genetic evaluation, and that genetic testing is best performed by clinicians that have familiarity of genetic  evaluation and in partnership with genetic counselors. I think one of the big challenges for us is to figure out how to best have genetic counseling  as a part of our clinical workflow. And I can say for myself, I don't feel like I've always done the best job with genetic  counseling because I'm not a genetic counselor, and I'm not a geneticist, yet we tend to send gene panels a lot, and especially if we're going to start to be sending exome and  genome studies, and if those get covered by insurance, and that that's the standard of care for first-line treatment, we really need much more detailed genetic counseling involved  in our clinical practice. This is my soapbox moment. So the tough part is how to pay for genetic counselors, and there's state-to-state variability  in terms of licensing for genetic counselors, and the ability for them to bill, and the ability for them to get insurance coverage of that time, that valuable time they're spending  with our patients. And that financial piece sometimes is shifted onto the operational budget of the clinic, sometimes shifted to philanthropy, sometimes shifted to research budgets.  But I think one thing that we should and need to advocate as a society, along with the genetic counseling folks, is to make sure that we can get better genetic counseling for our  patients across the country, and potentially across the world. So thank you very much.

Video Summary

The presentation discusses the significance of genetic testing in epilepsy, emphasizing its evolving role from the view that most epilepsies were idiopathic to recognizing genetic etiologies. The speaker, a pediatric epileptologist, outlines the benefits of genetic testing, such as identifying causative genes, aiding in family counseling, and adjusting treatment plans. They provide an overview of testing methods like gene panels, whole exome sequencing, and whole genome sequencing, comparing their pros and cons. Despite genetic testing's critical role, challenges include interpreting variants of uncertain significance and limited insurance coverage for tests. The presentation outlines a new clinical practice guideline recommending testing strategies, suggesting genetic evaluation for unexplained epilepsy. The speaker stresses the importance of genetic counseling and the need for improved integration of genetic counselors in clinical workflows, also highlighting the financial and logistical challenges of providing adequate counseling services.

Asset Subtitle

Presenter:

Louis Dang, MD, PhD, is an Assistant Professor of Pediatrics and Neurology at the University of Michigan. As a pediatric neurologist/epileptologist, Dr. Dang has a particular interest in the care of children with severe, treatment-resistant epilepsies. Dr. Dang’s research focuses on understanding how genetic mutations result in epilepsy and neurodevelopmental abnormalities, in order to develop novel gene-directed therapies. Dr. Dang uses human stem cells to derive neural cultures that model fetal development of the human cortex. By studying human stem cells that contain mutations in epilepsy genes, he aims to determine the effect of those genetic mutations on neurodevelopmental processes and to uncover novel therapeutic targets.
◦LinkedIn: www.linkedin.com/in/louis-dang-neuro

Keywords

genetic testing

epilepsy

clinical guidelines

genetic counseling

testing methods