Solving diagnostic challenges in inborn errors of immunity

And welcome everyone. We are very excited to be having this discussion with you today. As you all know from Tina's introduction, both Andy and I are full time paid employee employees of Blueprint Genetics and Quest Diagnostics to quickly review the learning objectives we're going to cover today. We will first review the diagnostic utility of genetic testing in inborn areas of immunity and talk about the importance of this testing within the diagnosis, the diagnostic work up for these patients. We will then dive even deeper into a very important part of genetic testing, which is the variant curation and classification aspect. And we'll discuss the various lines of evidence that are used to accrue that data and then present it to the clinician. And then finally, we have a couple of case challenges that we'd like to review with you that really demonstrate how variant curation helps achieve A genetic diagnosis and the importance it can play in patient management. So diving right in, I want to 1st review inborn errors of immunity, which I know many of you are probably incredibly familiar with. But these are conditions in which a patient has genetic variants that result in abnormal function of the immune system. These variants can either be inherited from parents or they can occur new or de Novo in an individual. And while the individual syndromes can be relatively rare, collectively these syndromes affect at least one to 2% of the world's population, which to put in numbers, that's over 80 million people worldwide. So this really is a significant health concern and it's great to see that there is a lot of ongoing research within this area, both to identify these syndromes and help these patients throughout the world. There is the International Immunological Society that categorizes these syndromes and organizes them. I've included the table from their most recent publication in 2022 here just so that way you can see where and how all of these different syndromes are kind of organized by this committee and I won't go through all ten of them. But it is important to note that even though there are 10 distinct categories, there is a tremendous amount of overlap in the symptoms that patients can experience across these categories. The most common of which are susceptibility to infections, autoimmunity, auto inflammatory diseases, allergy, bone marrow failure and or malignancy. And if we dig now a little bit deeper into the genetic causes of inborn errors of immunity, we know these are very numerous. The IUIS has described over 485 distinct genetic causes to date. And I always love reviewing this figure that I've included on the slide, which is just a graph that shows the number of genetic causes that get added, which with each publication of the IUIS list going all the way back to its inception in 1980. And you can see that with each one, more and more genes get added, which I think really emphasizes the ongoing research within this aspect of medicine. And it also shows just how many new genes and how much there is still to learn because even between 2019 to 2022, there were 55 new genes added. And we're waiting anxiously to see the new list should be coming out soon. It's usually published every, you know, two to three years. And so we'll have to see how many new genes are added this time around. And as these genes continue to be described, we see that it's not just whole new syndromes that are being added to the list. In fact, many of the new genes that are added are additional genetic causes of syndromes that have already been described. And a great example of that is chronic granulomatous disease, which we know has six different genetic causes even within the same syndrome. And that really can pose a challenge when it comes to identifying the exact etiology for a patient, because even if you have a well defined syndrome, you still really may rely on genetic testing to actually identify the exact etiology. And so it's important just to note that and it really does speak to the fact that genetic testing has become a quintessential tool in the diagnostic work up for patients with inborn Arab of immunities. Unlike most testing that is done for these patients, which can be non specific or just require subsequent rounds to really zero in on the specific phenotype for a patient, genetic testing really can expedite that etiology and that kind of confirm that diagnosis for a patient. So hopefully then you can move from the diagnosis aspect of their journey to treatment more quickly. And I love this study by the Jeffrey Modal Foundation, which is a large organization in the United States that promotes research in inborn areas of immunity. They actually sent out a survey to a large group of clinicians who were doing genetic testing and using it as part of their work up for a patient. And what they found is that the clinicians reported that in over a third, close to 40% of the patients that had genetic testing, it actually altered the diagnosis. And so that could be either there was a very broad differential and the genetic testing allowed the clinicians to 0 in on the specific diagnosis for the patient, or it could be a case where there was a clinical suspicion of one type of syndrome. But then actually after the genetic testing, it completely changed and went towards a new diagnosis based on the genetic results. And so of course you can understand how powerful that could be in your own clinical practice as well. It makes sense that if the diagnosis was altered, then that would immediately change the disease management and treatment. Because I, as I'm sure all of you are aware, the treatments for different inborn areas of immunity really is becoming more and more personalized. You have treatments coming out that are specific not only to a given condition, but a specific genetic cause within a condition. And so again, this really just speaks to the power that this testing can have in your diagnostic toolkits. And so when we think about that tool and the different options that you have, there are various different genetic tests. And so I'm just going to quickly go over them here. So that way we're all on the same page. I've listed some of the more common genetic testing options, basically from most targeted to most general. And at the top, we have single gene tests. And so most people will elect this test when there's a very specific syndrome that is on their differential. And often there's only just a single gene that is the cause of that syndrome, at least as of what we know today. A common example that's given for this scenario is something like cystic fibrosis, where there's a very clear phenotype that patient could meet criteria for, and then there is a single gene that causes that. The next and probably one of the most common tests that are ordered for patients with suspected inborn errors of immunity is a panel of genes, and this could range widely in the size and the number of genes tested. It can be everything from a single syndrome that has just a couple of known genetic causes all the way up to multiple syndromes within a general category. So a category such as all of the the genes on the IUI as guidelines. The next even more general of a genetic test is whole exome sequencing. And so rather than a curated panel of specific genes for a given, you know, syndrome or clinical indication, whole exome sequencing is going to look at all of the protein coding regions within the human genome. And the, the really the advantage of this type of testing is that not only is it going to, you know, test for variance within the genes, it's also going to allow for gene discovery. And so this could be identifying a variant in a gene that has incredibly preliminary evidence and help really expand our knowledge of conditions that could cause inborn errors of immunity. The final Test that is the most general is the whole genome sequencing. And this is largely still reserved for research spaces, although it is becoming more common clinically. And I think in the future we are going to see increasing utilization of this test both commercially and in the clinic. Like the whole exome sequencing, it does include all protein coding regions in its targets. However, it also includes the non coding regions, so the regions in between the exons of a gene. And similarly to whole exome sequencing, this technology allows for gene discovery, allows for sequencing, sequencing of genes that may not be on a curated panel, but it really does depend on kind of the platform of the whole genome sequencing, what exactly is included in it, and the interpretation that you'll get. And so when we think about the two most common tests right now within the field that really is the panels and the whole exome sequencing and digging a little bit deeper into the diagnostic utility of these tests, we know that there's actually a lot of overlap in the studies that have been published. The diagnostic yield ranges anywhere from 15 to 79%. And this really varies depending on the including criteria for the study, how it was designed and what they considered a diagnosis. And so it kind of comes down to your individual patient and clinical scenario, whether you know you want to go with a panel or whole exome. It can really vary depending on patient. OK, now with that, I'm going to transition it over to my colleague, Anne, who is going to give us more information about an incredibly important aspect of the genetic testing process, and that is variant curation. All right. Thank you, Terry. So this first slide is to provide a general overview on the type of information we use and how we apply it. When we create variants from patients who are tested with immunology related panels, first we collect all available evidence for the variant. We utilize external and internal knowledge about a particular gene of interest. Creating immunology related variants requires deep understanding and experience and at Blueprint Genetics we have a specific geneticist team who are specialized in inborn errors of immunity. This team has expertise in immunology related genes. They understand the disease penetrance and onset as well as the clinical variability and this helps to make decisions about variants. We also have an in house designed interpretation tool that organises any previously created information for a variant, allowing us to expedite our review. Next we consult various external sources of edit event evidence including peer reviewed articles to see if the variant of interest has been reported in an affected individual. We also check disease specific and gene related databases, Clinton guidelines when available for certain genes and also reference population databases. Additionally, we evaluate the data by using in silica prediction tools or splicing predictions when needed as well as functional studies when those are available. We also consider the segregation evidence related to the variant as there may be family studies available for the variant or specific variant has been shown to occur as new event also known as de Novo event in an affected patient. So overall, we gather all relevant evidence for the variant of interest and critically evaluate the quality and reliability of the existing data. We compare all available evidence to our Blueprint Genetics classification SHEEM that is based on American American College of Medical Genetics and Genomics guidelines for the variance we report to patients. We present the most significant findings in the variant description in a clear and easily accessible format, which then enables healthcare providers to interpret and apply to information effectively in patient management. All variants reported to patients must be classified. Variants are categorized into five classes, but we only report those in the top three, including pathogenic, likely pathogenic, and variants of uncertain significance. If a variant is classified as benign or likely benign, it is not reported to the patient as it is not considered clinically relevant. We carry a significant responsibility in accurately classifying variants as considering a variant that's likely pathogenic or pathogenic constitutes a genetic diagnosis for the patient, and this can directly influence clinical management decisions. This is especially true for patients with an inborn error of immunity. A genetic diagnosis may lead to critical medical decisions such as bone marrow transplantation, family planning or enrolment in a cancer surveillance program. Now that we reviewed the creation and classification of variants, we will walk you through three real case challenges to illustrate the impact of variant creation in genetic testing results and it's implications for patients and their families. This first case challenge may present the scenario that is very familiar to you. The healthcare provider must determine the most appropriate treatment for a patient and is turning to genetic testing for guidance in such cases. Genetic testing results can play a vital role in informing difficult clinical decisions and offering clarity in complex situations where there may be several treatment options. This is a young child with erythroplastopenia, lymphopenia, alopecia, and immunodeficiency, and the provider is hoping to decide if bone marrow transplant is the most appropriate treatment. Based on these results, they had ordered the primary immunodeficiency panel for this child. We identified A homozygous ADA 2 MIS SENSE variant in this patient, which is associated with adenosine DMNS 2 deficiency. The next question is what is the classification of a variant we are reporting for this child? As mentioned, our variant classification scheme for nuclear variants follows American College of Medical Genetics and Genomics guidelines. And in that scheme we consider first of all the phenotype match our patients clinical symptoms match well with ADA 2 related disease. Therefore this variant would be reported if it's classified as a variant of uncertain significance, likely pathogenic or pathogenic. Then we take a look at in Silicotol rebel and the rebel score for this variant is high, which indicates that this variant may associate with disease. Next we check the frequency of the variant. This ADA 2 variant is relatively real in general population databases and also in our in house database. Through a literature review, we find that this variant has been reported in multiple individuals with ADA 2 deficiency. In fact, this variant is the most frequently reported diagnostic ADA 2 variant in the European population. The variant has also been shown to segregate with disease in families. We also consider the functional studies which have shown reduced enzymatic activity and protein secretion for this variant. Variants reported for the patients need to be true positive variants with very high certainty. Therefore, we always carefully assess the quality of each variant call in our NGS data. If the variant quality does not meet our high threshold criteria, we confirm the variant with another orthogonal and independent method before reporting. So bringing all that evidence together and using our classification Sheen for nuclear variants, this variant was classified as pathogenic and the patient got a diagnose. But what does this diagnosis mean for the patient? First of all, this confirms A molecular molecular diagnosis of adenosine DMNS to deficiency. But how does this guide the clinicians goal on deciding treatment for our patient? So not all diagnostic variants in ADA to lead to the same clinical manifestations. Our review on of the available literature indicates that this specific variant, when present in a homozygous state, is associated with a significant risk of neurological and hematological symptoms. According to the same publication, 3 patients with severe symptoms underwent hematopolytic stem cell transplantation, which restored ADA 2 activity and LED to resolution of their clinical manifestations. Because on our in depth variant creation, we investigated this important evidence and provided the clinician with specialised knowledge about this specific variant. This information hopefully helped guide the clinician in developing an appropriate treatment plan and also highlights the crucial role of genetic testing in patient care. Additionally, now that they know the variant that causes the disease in the patient, family members can pursue genetic testing for this specific variant. This could help determine the recurrence risk for ADA 2 deficiency in siblings, which could be as high as 25%, and help screen family members for bone marrow transplant donor eligibility. The second case challenge is identifying and explaining a patient with an atypical phenotype. This is an adult patient suspected with primary immunodeficiency. The patient was tested with the primary immunodeficiency and primary ciliary dyskinesia panel, a synonymous splice region variant in Illinois. 2 RG located in X chromosome was identified in this patient. IL 2 RG gene is associated with severe compound immunodeficiency. This variant is absent in Genomat which is a large red reference population database. It affects the last nucleotide of Exxon 7 within splice region and therefore effects splicing too. Splice loss was predicted by pangolin in silica splicing prediction tool for this variant. Literature review revealed that functional analysis of this variant demonstrated Exxon 7 skipping and a tenfold reduction in messenger RNA levels compared to controls. Based on the same literature, this variant has been identified previously as hemicycous in 2 male cousins affected with a typical X linked severe compound immunodeficiency and presenting with diffuse cutaneous warts and respiratory infections. Based on the available evidence, we classify this variant that's likely pathogenic. Coming back to our patient, while the clinical description we received was non specific, we were were very surprised to identify a diagnostic variant in IL 2 RG in an adult. As many of you are aware, X linked severe compound immunodeficiency is typically fatal in childhood without proper treatment. However, some hypomorphic variants in this gene have been shown to cause milder a typical form of the disease, including variable immunodeficiency and or autoimmunity. Interestingly, the variant identified in our patient has previously been reported in two related males with a typical disease presentation. This provides important prognostic information suggesting that the variant is associated with a milder form of the disease and may explain that primary immunodeficiency observed in our adult patient. When we took a closer look of the variant, this case proved to be even more intriguing than initially anticipated. This Illinois 2 RG gene is located in X chromosome and since this is a male patient, we would expect the variant present in all sequencing reads reflecting hemipsychosity since men only have One X chromosome. However, upon closer examination of the variant in our NGS data, we observed that it was present in only 73% of the sequencing reads from the patient's blood sample, indicating that 27% of the patient's blood cells do not carry this disease causing variant. This observation indicates that the variant is mosaic in the tested sample, which may help explain more about the patients are typical phenotype both due to the mosaicism itself and the variants known association with milder or atypical clinical presentations. Let's now pause for a moment to talk about mosaicism. Mosaicism means that an individual has two or even more genetically different sets of cells in the body. Different types of mosaicism can occur, but what we typically identify in our patient samples is post psychotic mosaicism. This means that a change in the individual's genome arises after the fertilization of the egg, meaning that it occurs in the developing embryo rather than being inherited directly from a parent. Here's a nice figure from a recent paper, and it illustrates what the different consequences might be in the developing individual of an acquired variation occurring at different times in embryonic development or in different cell types. Depending on when and where the variant occurred, you may get different levels of mosaic system between different tissues. If a variant on the X chromosome is mosaic in the male, it's out of frequency may be less than 100% in sequencing data depending on the tissue examined, reflecting that the variant is not present in all cells of that tissue. Post psychotic mosaicism may result in milder phenotype in patient as the pathogenic variant is present in only a subset of cells, potentially reducing the overall impact on organ systems. Since it's not present in all cells, post psychotic mosaicism can sometimes make diagnosis more challenging and may not be detectable in all tested tissues. For example, it might be present in skin but not detectable in a blood sample. However, upon reviewing the literature on both psychotic mosaic system in IL 2 RGT and we found that this patient could also be exhibiting A phenomenon known as somatic genetic rescue which results in different type of somatic mosaicism compared to both psychotic mosaicism caused by a de Novo event. Somatic genetic rescue refers to the spontaneous correction of a pathogenic variant in a subset of cells leading to a mixture of mutated and genetically rescued cells. This phenomenon leads to the presence of both phenotypically normal and abnormal cells as seen in both psychotic mosaicism, and may also lead to a minor phenotype in a patient than would typically be expected. However, compared to post psychotic mosaicism, in this type of mosaicism the disease causing variant may have been inherited from a parent. Therefore there may be a recurrence risk for the variant to occur also in siblings. A few cases of X linked severe compound immunodeficiency have been reported with these revertent mutations in the IL 2 RG gene revealing somatic mosaicism in these patients. And these patients had a significant, significant attenuation of the clinical phenotype. So what did we learn from this quite complex creation and what does this mean for the patient and family? First and foremost, it confirms the molecular diagnosis of IL 2 RT related a typical X linked severe combined immunodeficiency. We also discovered that the observed mosaicism of the variant may influence the clinical phenotype of the patient. The observed mosaicism in the patient could be due to both psychotic mosaicism or somatic genetic rescue. However, based solely on this sample, we cannot determine the exact origin of the mosaicism. Testing additional tissues or a maternal sample may provide more insight into its origin. If this is somatic mosaicism caused by somatic genetic rescue, there could be a potential recurrence risk for siblings and their offspring. Depending on the nature and timing of the genetic rescue. Both types of mosaicism, post psychotic mosaicism and somatic mosaicism caused by somatic genetic rescue can contribute to a milder than expected phenotype in the affected individual. In conclusion, the identified variant and its mosaic status can significantly influence the predicted clinical phenotype, offering valuable insights to the clinician in understanding the patient's presentation. Now I will hand things back to Tori, and she will present to us the final case challenge. Thank you so much, Andy. OK, everyone. So our last and final case, this is a clinical challenge, particularly for the lab, is how do we handle a newly defined syndrome in the context of inborn errors of immunity. And so to go through this, we received an order for a child who was affected with general lymphadenopathy, splenomegaly, recurrent infections, periodic fever, recurrent inflammations, low T&B, cell levels, leukopenia, anemia, thrombocytopenia and phthalator thrive. So a pretty complex phenotype. This clinician had elected to order the our comprehensive immune and cytopenia panel. And this is actually our largest panel within our immunology offering. It includes over 640 genes and those genes are a combination of very, very well established genes within the immunology space and then also genes that are more rare and may have more preliminary evidence. They may have only been described in a few patients. And that's intentional because we really are trying to reach into those rare kind of syndromes and hopefully increase the diagnostic yield for patients that have a diverse phenotype like this patient. And so when our geneticists reviewed the data, they identified A homozygous variant in the Ras Group 1 gene. And when they first identified this variant, it was actually the first time that we had seen a variant in this gene. The variant alone was very molecularly suspicious given that it is a frameshift variant. And so that in prompted our geneticist to investigate this gene and this variant further. So they began to look for evidence of the rasp Group 1 immunodeficiency and this gene was actually first reported in 2016 in a patient. So this is relatively very recent. When you think about how rare these conditions are, only having a few years of studies really is just a small amount of data in the context of a genetic syndrome. And so the IUIS had only described this gene in 2018, and it was categorized as a disease of immune dysregulation. And like I said, at Blueprint Genetics, we hadn't actually previously reported any findings within this gene. So at the time of the variant discovery, we not only had to go through and review evidence for the variant to confirm, you know, its classification, but before we even did that, we actually had to take a step back and look at the gene itself and 1st establish that there is truly a gene disease association that would be clinically relevant for this patient. And so our process for curating a gene disease association is very similar to the variant curation process that Anne has so well described previously. We first look at the cases that have been reported so far. We look at their phenotype. We look at the types of variants that were identified in affected individuals. And for this particular gene, at the time we identified this variant, there were only ten reported cases. It was just a couple of families that had some siblings that had variance within this gene and a cohesive phenotype. So once we've established that, OK, there is a phenotype, it seems to be consistent across at least a few families, we then look at the gene and the protein functional studies to kind of logically understand how this gene could be implicated in an inborn area of immunity. And we found that this gene is very important for the Ras map K pathway, which is essential in T cell and B cell activation and proliferation. And so logically it made sense that a decrease in function in this protein might have an in a impact in the immune system and lead to an inborn Arab immunity. And so following that, we also just check what variants exist within this gene and how common are they within population databases. Just to know, you know also what is the disease mechanism, what types of variants contribute to disease in patients with variants in this gene. And so after we performed that curation process, we did conclude that there is a strong disease gene association between Ras GRP one and an inborn area of immunity. And so it would be relevant for us to report variance within this gene if there was good phenotype overlap. And so then now that we have the gene disease association established, now we turn our attention back to the variant and we perform the same steps that obviously Andy described. We look to the fact that this variant was not published or not found in population databases. We saw that it was a frameship variance that results in a premature stop codon, which is very suspicious because loss of function was a described disease mechanism within this gene. And additionally, this particular variant was actually reported in one of the few families that had been described in this immunodeficiency. It was reported as homozygous in two siblings that had a phenotype that was very consistent with our patients. And so given all of that evidence, even though, you know, we were only talking about a handful of studies, because of the robust evidence, you know, process and the curation process, we were able to conclude that this is a relevant finding and, and a diagnostic finding for this patient. And when we think about that clinical synopsis that I was talking about and how this relates to our patient, we saw a lot of similarities between what had been published in the literature and what this patient was described as in the order for this test. And so we saw that there were recurrent infections, there were poor activation, proliferation and motility of T cells and B cells. These are in the reported patients, but that corresponds with our patients as well. There were instances of autoimmunity. There was also hepatomegaly and splenomegaly described as well as failure to thrive. The splenomegaly and the failure to thrive our patient also had. And then in addition, in some of the published literature, we learned that patients appear to have an increased susceptibility to the Epstein Barr virus infection and then subsequent Epstein Barr virus associated lymphoma. And finally, based on the studies that we've seen so far, this variance in this gene appear to be following a autosomal recessive pattern of inheritance. So this is really important because we do see good clinical overlap with our patient. And even though that this is a rare syndrome, like I said, there's still preliminary evidence, there is enough that we are able to give a diagnosis to this patient and kind of help them understand at least to this point, what the phenotype looks like. And when we think about the management implications of that, there was one study that included that hematopoietic stem cell transplant could be considered in these patients, particularly if they appear to be at risk for severe T cell deficiency. And additionally, I mentioned the increased susceptibility to lymphoma. And while there are not specific guidelines for this given immunodeficiency, there are general guidelines for lymphoma screening. And so the clinician may choose to apply them to their patients going forward. And finally, with the autosomal recessive inheritance, we know that the recurrence risk in full siblings is 25%. And so if they were, you know, if this family was to have more children, they can be aware of that. And then in addition, if they were to pursue a hematocritic stem cell transplant and there were any siblings who were to be considered donors, they could use genetic testing to determine their eligibility if they otherwise, you know, didn't have symptoms present yet for this condition. And so with this case, I really think that the big take home point is that even for conditions that are very rare and may have only been recently described, if you are working with a lab that has a very robust disease associate, gene disease association curation process and bearing curation process, there is still a wealth of information that can be communicated within a report. And you can still help patients achieve A diagnosis and figure out a treatment plan even when there isn't a ton of evidence. You just use everything that you have. And we're only continuing to push further into gene discovery in the immunology space with all the research that is going on. So I really hope this just highlights the power that a good curation process can have as a partner to you in the clinic. And with that, now I would just like to summarize some of the key points that we hope were conveyed throughout this presentation. As I know you all know, inborn errors of immunity are complex. They present with a wide range of overlapping phenotypes. There are numerous and an ever growing list of genetic causes that you have to be aware of. And it really emphasizes why genetic testing is the critical tool in the diagnostic odyssey of these patients and will also guide management. And what we hope to have emphasized is that the variant curation is really a quintessential aspect of that testing. It's not enough to just sequence patient's DNA and report the variants you see. It really is our duty as a lab to be curating these variants and to try and provide as much information as we can because that's what makes these. Result actionable. You know, the individualized curation can guide a treatment decision even by at the variant level. It's amazing that we're we're getting to the point where individual variants may have treatment decisions. It can help explain atypical phenotypes and help guide a patient through understanding their diagnosis in the context of different symptoms than what you may expect just based on the gene alone. And then of course, it can enable diagnosis in newly defined syndromes. And even when there isn't a ton of evidence, there is still enough that you can guide your patients and, and hopefully then just find additional evidence in the future. So we really want to thank all of you for joining us today. I hope you've enjoyed these cases. We always love looking through them. And so if you have any questions now, we would be happy to answer them.