Thank you so much for joining us today to learn about hearing loss and our experience testing for this indication. My name is Kim Gall, and I have the pleasure of co-presenting today with Dr. Jennifer Schlecht. So before we dive in, I wanted to take a step back and define what it is that we're talking about. So hearing loss is defined as any loss of the ability to hear as determined by audiometry. Deaf with a lowercase d refers to the total or near total loss of hearing, also defined by audiometry. And Deaf with a capital D is a community with a distinct culture and language shaped by the experience of having hearing loss or being deaf. So in terms of the frequency of hearing loss and deafness, we know that up to three in every thousand babies born every year will be deaf or have some degree of hearing loss. And almost 15% of children between the ages of 6 to 19 also have some type of hearing loss. So it's important to realize that this issue is not just limited to older individuals. And making the diagnosis early on and understanding the underlying etiology is key for reasons that we'll discuss in this webinar or in this talk. So there are many different aspects, both audiometric and clinical, that we use to describe hearing loss. And I've listed them across the top of this slide. Type refers to whether the hearing loss is conductive or sensorineural or mixed. Onset refers to when the hearing loss begins in relation to the development of language, so prelingual or postlingual. Severity refers to whether it is mild, moderate, severe, or profound. The laterality, of course, describes whether or not it is unilateral or bilateral in nature. And stability can be progressive, non-progressive, or fluctuating. And lastly, the configuration is the appearance of the hearing loss by audiometry. And we'll talk a little bit about etiology more on the next slide. So we know that there can be both genetic and non-genetic causes or determinants of hearing loss. The non-genetic causes or determinants are often secondary to things like infection during pregnancy or in childhood, malnutrition, or limited access to healthcare. And this slide indicates that those acquired or environmental forms of hearing loss account for approximately 20% of prelingual hearing loss. That might be higher if the patient comes from an underdeveloped country. But because there can be both these acquired or environmental causes of hearing loss in addition to genetic forms of hearing loss, it's important to make sure that you're ruling out non-genetic causes like congenital CMV, for instance, when you are assessing for the cause of hearing loss in your patient. So once you've done that, you then need to consider the genetic causes, which as you see on the screen, can be further grouped into syndromic hearing loss or non-syndromic hearing loss. And of course, the syndromic hearing loss means that in addition to actually having an impairment with hearing, there can be additional medical problems in different organ systems. And to date, we know that there are over 400 different syndromes that have hearing loss described as one of the features of that syndrome. With respect to non-syndromic hearing loss, that means that the hearing loss is isolated or not associated with any other medical problems. And this type of hearing loss tends to be sensorineural in nature versus conductive, and also has a number of genes at this point greater than 100 known to be associated with non-sensorineural hearing loss. Gene discovery does continue though, so we expect these numbers to increase over time. So it can be difficult when you have a patient before you who you believe has genetic hearing loss to determine whether or not it is syndromic or non-syndromic, as some of the syndromic features do not develop until later on in life or can be subclinical in nature. So if we are not sure whether or not the hearing loss falls into the syndromic or non-syndromic category, the workup for these patients can include things like diagnostic imaging, so MRI or CT scan, renal ultrasounds. It can include ECGs to look at the heart rhythm. There could be thyroid studies, urinalysis, ophthalmological exams, and so some of these clearly can be invasive and or burdensome. But if we confirm a molecular diagnosis in a patient and we find out that the cause of the hearing loss is non-syndromic, then we can avoid the investigations that I just mentioned. Or if we determine that it is syndromic in nature, but we understand now the syndrome in the natural history, we can target the investigations that need to be done and the ongoing surveillance for that individual. So I think we've established the clinical utility of a genetic diagnosis for hearing loss for the purpose of surveillance of comorbidities. But in addition to that, there are other benefits that a genetic diagnosis can bring, and I've listed some of them here. So it can include planning for future medical and educational needs, facilitating estimations for recurrence risk, both for the parents of the individual, the individual themselves, their family members, allowing families to better plan to have resources in place for the birth of a child that will be deaf or have hearing loss. And if the cause of the hearing loss happens to involve mitochondrial DNA mutations associated with the susceptibility to aminoglycoside toxicity, it is possible for relatives to avoid those medications and hopefully avoid hearing loss associated with that. So at Blueprint Genetics, all of our panels and single genes are run on an analytically validated clinical grade exome sequencing assay. And the sensitivity to detect our different types of variants is publicly available on our website. I will point out that our assay does perform very well across variant types, but in particular, we've demonstrated that the assay can detect single exon deletions in addition to large micro deletions and micro duplications. The panels that we offer in our ear, nose, and throat space are divided as you see on the screen. So we offer a comprehensive hearing loss panel, in addition to a syndromic and a non-syndromic hearing loss panel. And then we also offer several condition specific panels. All of the genes in the syndromic and non-syndromic hearing loss panels are included in the comprehensive hearing loss and deafness panel. And the mitochondrial genome is also included in the comprehensive hearing loss and deafness panel, the syndromic hearing loss panel, and the non-syndromic hearing loss panel. Further, we have customized our assay to include up to 112 non-coding variants that are disease causing associated with hearing loss as cataloged in HGMD or ClinVar. So the number of these non-coding variants obviously varies with the panel size that you order. With the comprehensive hearing loss and deafness panel, all 112 would be included. And then if you order a smaller panel, it depends on the genes on that panel, how many of these non-coding variants are included. So I wanted to walk you through our most recent experience with a patient cohort with hearing loss. So this was a cohort of 898 patients consecutively referred and analyzed in retrospect. Demographics of the cohort were roughly equal in terms of sex, and almost 80% of the individuals who were tested were under the age of 18 years at the time they were tested. Infants and children, you can see here, were the most frequently represented. So in that patient cohort, we had 269 patients out of the 898 or 30% who received a molecular diagnosis. And the diagnosis in this particular cohort is defined as a primary finding that's consistent with the patient's phenotype. The variant is either pathogenic or likely pathogenic. And there's one variant present in genes responsible for autosomal dominant or X-linked conditions, or two variants in genes associated with autosomal recessive conditions. So in the cohort, what we noted was the diagnostic yield was significantly increased among those infant samples. So over two-thirds of our infants were received diagnoses compared to our older children or the adults in the cohort. And the majority of the individuals received a diagnosis after having a comprehensive hearing loss and deafness panel. Fifty-four genes were responsible for the diagnoses in the 269 patients. And those genes are listed in the table on the right-hand side of the screen. The DFNB1 locus made up almost a quarter of our diagnoses, while STRC was responsible for almost 10%. And the genes that I've highlighted in pink are there just because they're not always included in a standard assessment of hearing loss, but in our cohort accounted for about 15% of our diagnostic yield. So just an important consideration when selecting testing for your patient. In our group, we saw that CNVs were present in 16% of patients who received a diagnosis. So that means the CNV contributed to or was responsible for the diagnosis in 43 patients. And the CNVs ranged in size from very small, so 160 base pairs in the ADGRV1 gene to up to 200, sorry, 2.75 megabases. Of those 52 deletions that we observed, about 36% were intragenic, so on the smaller side. And perhaps even more interesting is that, again, almost a third of the deletions were big, so greater than 34 KB in size. And again, this is an important consideration when looking at the testing capabilities of the panel that you're going to select for your patients is the ability to detect different sizes of copy number variants. And then of the deletions that we detected, 65% were deletions involving the STRC gene. Not surprising as this is a very common mechanism of disease affecting this particular gene. So I'll just pause here to show you one of our interesting cases. This information that we received is very straightforward and typical of the type of case that would be referred to our laboratory. This is a two-year-old who has sensorineural hearing loss and otherwise reported to be well. So this patient had the comprehensive hearing loss and deafness panel, and our results demonstrated that the patient is heterozygous for a whole gene STRC deletion, which was confirmed by qPCR, and this deletion was classified as pathogenic. In addition, we identified that the patient was hemizygous for a single nucleotide variant in exon 21 of STRC. This variant was also confirmed with orthogonal testing and also classified as a pathogenic variant. So this STRC is responsible for the second most causes of non-syndromic hearing loss, and also is the most common cause of mild to moderate hearing loss. It's second to GGB2 and GGB6 in terms of causes. So, it's an interesting gene. It encodes the protein stereocillin, and its role is in the outer hair cells, in the inner ear, and it is spread over 29 exons, and the entire gene is actually tandemly duplicated in the genome, with greater than 99% homology across all of the exons. And to complicate matters even further, the first 15 exons, and those are the ones that I have indicated in black on the screen, actually share 100% homology with the pseudogene. So, how does this sequence homology impact genetic testing by short-read next-generation sequencing? Well, in general, our confidence in the read alignment decreases as sequence homology increases. So, if those sequence reads align equally well to different genomic positions, then they're discarded. So, if you have a gene with 100% homology, like the first half of STRC, you're basically getting no unique reads mapped. So, you need a different strategy altogether to tackle exons with this level of homology in order to understand sequence variance in these regions. If we think about the last half of the gene, which has 98-99% homology, you can see that you do get some genes from the parent gene or the functional gene, sorry, some reads from the parent gene or functional gene, and also some reads from the pseudogene that are then aligned with the reference sequence. And so, in this case, it's really important to have orthogonal methods in place that can differentiate whether or not the variant came from the parent gene or from the pseudogene. So, I mentioned previously that in our cohort, STRC was responsible for almost 10% of molecular diagnoses. And of those diagnoses, 83% of the variants were deletions, but 17% were in fact missense variants. And those missense variants would have been variants that we were able to detect and confirm from the last half of the gene. Looking in our cohort, we also saw that almost 6% of patients had a very suspicious BUS, where segregation analysis may have been sufficient to reclassify the variant to likely pathogenic and therefore diagnostic result. Unfortunately, we don't have segregation data for all of the patients in this cohort. But if we did, and we were able to reclassify those variants to likely pathogenic, almost 36% of our cohort would have had a diagnosis or a likely diagnosis. To be transparent, we wanted to let you know about the limitations in the analysis of this cohort. As we are a commercial laboratory, we have updated our panels over time. So, there are different numbers of genes on the panel, depending on when that panel was performed. We've also changed our sequencing assays over time. So again, not all of the patients sequenced were sequenced using the same assay. They all did, however, receive sequencing and deletion duplication analysis as part of this testing. And lastly, I'll just mention that GJB6 deletion analysis, which I didn't talk about in detail, but is included at Blueprint Genetics, was only performed for those patients who were run on the CES assay, so 172 of the 898. And the mitochondrial genome analysis that Jennifer is going to talk about, again, was only included on the CES assay. So, with that, I am going to transfer control over to Dr. Schley. Thanks, Kim, for that great introduction. And yeah, we've started, Kim's talked about quite a few cases where we were thinking about nuclear encoded genes, but more recently, here at Blueprint Genetics, we've been thinking beyond the nuclear DNA. And particularly in 2020, we turned our focus to mitochondrial DNA and the mitochondrial genome, which encodes 37 genes within a circular genome that's approximately 16 KB, which unfortunately, because they're not located in the nucleus, mitochondrial variants cannot be detected with standard nuclear DNA sequencing. In addition, another important aspect to consider during analysis of mitochondrial DNA is a concept known as heteroplasmy. And that is the proportion of normal or variant mitochondrial DNA that is present in an individual. And this is actually a critical factor for determining the expression and severity of the disease that is caused by a variant. The goal representation of what mitochondrial, the mitochondrial genomes looks like with the 10 most recurrent point mutations that are commonly identified in the mitochondrial genome. I know that today our talk is mostly focusing on hearing loss, but it's important to just have a good reminder that mitochondrial disorders really impact many, many tissues. And that the phenotypes can range from fatigue and exercise intolerance to hearing loss, seizures, strokes, heart failure, diabetes. In many cases, these can be multisystemic, and yet in some cases, the phenotypes are limited to a single organ. Further complicating the diagnosis of mitochondrial disorders can be that these phenotypes can vary greatly even between family members, and can often be difficult to differentiate from other more common diseases. And there are a number of different mitochondrial diseases to consider, and we've just included a few that include, that are responsible due to both single nucleotide variants, but also to insertions and deletions and deletions within the mitochondrial DNA. And so, more recently, we've developed an approach to analyze the entire mitochondrial genome instead of just focusing on individual variants. And so, as part of this, the Blueprint Genetics team has developed a high quality mitochondrial DNA analysis pipeline using next generation sequencing that's combining deep uniform coverage across the entire mitochondrial genome, allowing for high resolution of detection for mitochondrial deletions and other mitochondrial variants at a varying range of heteroplasmy levels. Currently, there isn't really a gold standard for mitochondrial deletions. A gold standard is a reference assay, a sample set, or a validation study. And so, one of the goals of Blueprint Genetics has been to really set the standard for mitochondrial DNA testing. And so, we have carried out a large and comprehensive analytic and clinical validation study for mitochondrial variant detection. And so, let's just take a look at some of those results. Our NGS-based assay includes very high coverage with an average sequencing depth of over 18,000. This allows us to call heteroplasmic variants even at very low levels. However, throughout the entire mitochondrial genome, all nucleotides in the mitochondrial DNA are covered at a minimum of 1000x coverage. And this uniform coverage is also beneficial not only for single nucleotide variants, but for calling of copy number variant performance. And so, what we can see along the bottom here is a graphical representation of coverage of the mitochondrial genome. The different colored sections represents the different mitochondrial encoded genes. And what we can see by the lines across the top is the average coverage for this particular sample with the dotted line across being the 1000x minimum coverage. Unfortunately, the legend on the side got, or the axis on the side got cut off. But what we can see is that throughout the entire mitochondrial genome, we can see uniform coverage that is above the 100, or the 1000x threshold. And what this means in terms of our capabilities of detecting variants in the mitochondrial DNA is it allows us to detect multiple types of variants, including single nucleotide variants and insertion deletion events. These are high sensitivity, even at lower heteroplasmy levels. And what's also really cool is the capability to detect quite large deletion events. And again, with fairly high sensitivity, even at low heteroplasmy levels, so approximately 99% of large deletion events were detected as low as 10% heteroplasmy level. Now, this is a hearing loss discussion. So, let's talk about what this means in terms of hearing loss diagnostics. So, we added mitochondrial DNA analysis to 30 different clinical panels that are available at Blueprint Genetics. We recently completed a retrospective review of over 6,600 different patient cases that, in which mitochondrial DNA analysis had been included, come check out my poster. I think it's 353 that is presenting this data. And when we looked specifically at the hearing loss panels, we saw that there was an additional 2% diagnostic yield when the mitochondrial genome analysis was included in those panels. And we identified diagnostic variants in a number of genes and saw both heteroplasmic and homoplasmic variants in the tested individuals. So, let's look at a case example. So, this is an, you know, a 36-year-old male who's reported to have bilateral moderate sensory neural hearing loss. He was diagnosed at a fairly young age, but there's an additional family history of lipodystrophy, hypomagnesia, hyperaldosteronism, and hypertension. However, previous genetic testing for lipodystrophy had been named. Had been negative in this individual. And so, they had requested our comprehensive hearing loss and deafness panel, which, as Kim mentioned, includes single analysis of single nucleotide variants, copy number variants, and now mitochondrial DNA analysis. And in this individual, we identified a homoplasmic variant in the MTPK gene, and this variant was classified as pathogenic. Variants in this gene are primarily associated with MRF. However, there's a significant variability, even within families, with symptoms ranging from severe Least Syndrome to isolated lipomas. And so, this is very likely the explanation for the patient's hearing loss and lipodystrophy. Whoop, forgot to advance. So, yeah, and MRF is myoclonus epilepsy with ragged red fibers. So, in this situation, we felt that this was a likely diagnosis for this individual. Now, as I mentioned, mitochondrial DNA analysis has been added at Blueprint Genetics, not just two panels associated with hearing loss, but approximately 30 other, or 29 other clinical panels. And so, in our retrospective review, we looked at a number of those other panels as well and saw that there were increases in diagnostic yield in several of those panels, ranging from as low as 1% up to just over 11%, in particular for our mitochondrial DNA depletion syndrome. Our optic atrophy panel had an increase in diagnostic yield by 9.3%, indicating that adding mitochondrial DNA can result in increased diagnoses for patients with suspected rare inherited disorders. But with that, I'm going to turn this back over to Kim to talk about an exciting new program that is happening at Blueprint Genetics. Thanks so much Jen. So I am excited to be able to take some time to introduce you to our collaboration with the KUOS to offer the Resonate program. This is a program that offers no-cost genetic testing for individuals diagnosed with auditory neuropathy and it uses our comprehensive hearing loss and deafness panel and includes post-test genetic counseling. At present it's available in the U.S. but we will be expanding it globally and currently it's a site-driven pilot program but we are expected to launch open access later this year. If you're interested or you have questions about this program we would ask that you please contact Blueprint Genetics. I've got the URL at the bottom of the screen here and we'd be happy to answer any questions or talk about what we might be able to do to set up your site. So in summary I hope we have demonstrated for you that the clinical utility of genetic testing for hearing loss is significant. Inclusion of genes like STRC, MITF, MP, ZP2 can increase the diagnostic yield by about 15% in our study. Inclusion of the mitochondrial genome also leads to important increases in the diagnostic yield and that high-resolution detection of both small and large CNVs is an important contributor to diagnosis in this patient population as well. So with that thank you so much for tuning in and please do be sure to stop by our virtual booth if you have any questions or would like to speak with any of us. Take care.

hearing loss

genetic testing

deafness

types of hearing loss

genetic causes

syndromic hearing loss