Taysha Gene Therapies, Inc. (TSHA) Earnings Call Transcript & Summary

Good morning, and welcome to Taysha Gene Therapies Angelman Syndrome Investor Day. [Operator Instructions] Today, we will provide an overview of Angelman syndrome, hear from a patient advocate and discuss the TSHA-106 program and clinical development strategy. Joining the call today is R. Session, II, Taysha's President, Founder and CEO; Dr. Ben Philpot, our key opinion leader, guest speaker from the University of North Carolina at Chapel Hill; Dr. Kimberly Goodspeed, Assistant Professor in the Department of Pediatrics and Neurology at UT Southwestern; Dr. Allyson Berent, Chief Scientific Officer of the Foundation for Angelman Therapeutics or FAST; and Dr. Ryan Butler, Assistant Professor in the Department of Psychiatry and Pediatrics at UT Southwestern; and Dr. Suyash Prasad, Taysha's Chief Medical Officer and Head of R&D. Next slide, please. Before we begin, please note that this presentation will include forward-looking statements within the meaning of the safe harbor provisions of the Private Securities Litigation Reform Act of 1995. These statements may include the expected timing and results of clinical trials for our drug candidates and the regulatory status and market opportunity for those programs as well as Taysha's manufacturing plants. Please see the Slide 2 of the accompanying presentation and Taysha's SEC filings for important risk factors that could cause the company's actual performance and results to differ materially from those expressed or implied in these forward-looking statements. Taysha undertakes no obligations to revise or update any forward-looking statements to reflect events or circumstances after the date of this conference call, except as may be required by applicable securities law. Next slide, please. I'd now like to turn the call over to our President, Founder and CEO, RA. Session, II. RA?

Thank you, Kim. Good morning, and welcome, everyone. Next slide. Today, we are excited to do a deep dive into Angelman syndrome, a severe neurodevelopmental disorder, for which there are no approved treatments. We will discuss in detail our 2 strategies for treating this disorder, the UBE3A gene replacement approach to mimic maternal UBE3A allele expression and the vectorized RNA-mediated knockdown approach to unsilence the paternal copy of the UBE3A gene by targeting the antisense transcript, responsible for silencing the gene. We are very excited to share positive clinical data for the gene replacement approach, which was published last Friday in the Journal of Clinical Investigation Insight. We are honored to begin our event with a presentation from Dr. Kimberly Goodspeed, Assistant Professor in the Department of Pediatrics and Neurology at UT Southwest. As a child neurologist and neurodevelopmental specialist, her research focuses on natural history and biomarker development in rare genetic variance of neurological disorders and we are pleased to have her provide an overview of Angelman Syndrome, which will provide a deep understanding of the disorder and help set the stage for our treatment approaches. Following Doctor Goodspeed, we are very excited to hear from Dr. Allyson Berent, Chief Scientific Officer of the Foundation of Angelman Therapeutics, Director of the Angelman Syndrome biomarker and outcome measure consortium, and Co-Director of the International Angelman Syndrome Research Council. As a mother of a child with Angelman syndrome, she will provide a patient and caregiver perspective on the burden of disease, which will give real-world context to some of the therapeutic endpoints and current management approaches. She will also discuss key natural history data. We will then welcome key opinion leader, Dr. Ben Philpot, Associate Director of the UNC Neuroscience Center at the University of North Carolina at Chapel Hill. He is a keen and distinguished professor in the Neuroscience Center and Department of Cell Biology and physiology at UNC and conducts research on early-stage development treatments for Angelman syndrome and other neurodevelopmental disorders. Dr. Philpot will delve into UBE3A gene replacement approach to treat Angelman and we'll discuss the encouraging data he published on prior. We will then transition into a discussion of our vectorized short hairpin RNA knockdown approach to unsilence the paternal UBE3A allele with Dr. Ryan Butler, Assistant Professor in the Department of Psychiatry and Pediatrics at UT Southwest. His research focuses on gene therapies for genetic disorders and the evolution of molecular components of the nervous system. Dr. Suyash Prasad, our Chief Medical Officer and Head of R&D, will follow with a discussion on our clinical development strategy for our investigational programs. Next slide, please. Please feel free to read about our speakers at your leisure. We are very excited about our Angelman programs, where proof-of-concept data support further study of these therapeutic alternatives as a potentially safe and efficacious treatment for Angelman syndrome. With 2 treatment strategies for correcting the core deficit in the disorder and unsilencing the paternal allele, we are well positioned to eradicate this monogenic disorder and to be a world leader in the discovery of treatments for Angelman syndrome. We expect to initiate IND/CTA-enabling studies in early 2022. Next slide, please. We have a full agenda this morning. So I will now turn the call over to Dr. Goodspeed to provide an overview of the genetics and clinical features of Angelman syndrome. Kim?

Thank you, RA, and thank you for having me. I'm very excited to be here today and be part of this really innovative conference and I'm excited that there's some treatment opportunities on the horizon potentially. So next slide, please. Here are my relevant disclosures for today. Next slide. So my job today is just to introduce the disease. So Angelman syndrome is a genomic imprinting disorder. And as RA already mentioned, it's a pretty severe neurodevelopmental disorder. It's caused by deletion or loss of function of the maternally inherited copy or allele of the UBE3A gene. This maternal specific inheritance pattern is due to genomic imprinting of the UBE3A in neurons specifically. And this means that the maternal allele is expressed and the paternal allele is silenced. Absence of the UBE3A gene or protein product leads to abnormal neuronal connectivity in the developing brain. So we can think of this as a synaptic plasticity or connectivity disorder where neurons aren't speaking to each other properly. UBE3A is a ubiquitin ligase protein and its job is to tag proteins for degradation within the cell. In the absence of UBE3A, the control of this cell signaling responsible for synaptic formation and plasticity is dysregulated. So if I can turn your attention to the diagram on the right. In column A, the column on the left, there is the normal state where UBE3A tags certain proteins, in this case Ephexin5 for degradation. That leads to letting inactive [ ROA ] remain inactive and allows for synaptic formation. In the absence of UBE3A we don't have this tagging of the Ephexin5. So it goes on to activate this [ ROA ] compound and ends up suppressing synaptic formation. Next slide. And so as mentioned here, just demonstrating in a slightly different way, E3 ligase, or UBE3A, plays a really important role in the proteasomal activity and regulation of active and inactive proteins. And these are important or brain development and normal function and normal connectivity within the brain. Mentioned in a different way, UBE3A conjugates the polyubiquitin genes to specific lysine residues in its substrates, and this regulates expression and function of those proteins. Deletion and loss of function mutations in the maternally inherited allele will result in Angelman syndrome. However, maternal duplication or even triplication of the chromosome 15q11 to 13 region is also associated with an autism phenotype, and variable spectrum of neurodevelopmental disability. This is also considered dup15q syndrome. Impairments in ubiquitin-mediated protein degradation can lead to deficits in the development and the maintenance of synaptic connections which are super important for the developing brain to function properly and progress through typical developmental milestones. Next slide. So what do patients look like? So I want to give a brief clinical overview and let Dr. Berent go through quite a bit more of the details, especially living with the patient and caring for a patient and being integrally involved in many of the natural history studies. But as a brief overview, this is a relatively common among the rare disorders. So it has an estimated incidence rate of approximately 1 in 15,000. These are based on relatively small studies, and larger studies are definitely needed to get a more precise number. But if you extrapolate this out, that estimate equates to approximately 500,000 individuals diagnosed worldwide, though actually getting that genetic confirmation and the actual patients were identified may not quite reach that number. Patients invariably present with developmental delay within the first year of life, so they're missing their milestones to walk on time or talk on time. They typically have diagnoses such as global developmental delay or intellectual disability. Some patients may be misdiagnosed as having cerebral palsy or static encephalopathy. Their language skills tend to be a lot more impaired than their motor skills, though they do have a significant motor phenotype and motor disability. Most of these patients are never able to speak verbal words, but they may be able to use signs or other augmentive-assistive communication devices such as a tablet or a picture exchange system. And families will consistently tell us that their receptive language skills are stronger than their expressive skills. So they can understand spoken language. They can follow directions, though it is probably still limited in comparison to their age matched peers. And as I mentioned, they still do have a significant motor phenotype. Approximately 10% of these children will never gain the ability to walk independently and many may require a wheelchair or other assistive ambulation device in the community or for long distances. Behaviorally, these children are often described as having a happy demeanor, an old terminology you may come across was Happy Puppet syndrome, but they can also have maladaptive behaviors and behavioral challenges, including irritability, aggression, pinching, hitting, writing that make it difficult to function in a classroom or to participate in therapies or follow directions of families. Many children will have stereotyped movements, so maybe hand flapping, clapping, rocking repetitively or hyperkinetic movements or ataxia. They often are described to have an unsteady gait, wide base, sometimes hands up held high for balance. Sleep disorders are extremely common in this condition as they are in many neurodevelopmental conditions and can range from difficulty falling asleep to difficulty maintaining sleep. The vast majority of these children will have seizures, though the severity and the semiologies can vary quite a bit. And they typically have onset of their first seizure in the toddler years, though that's not 100% true. As these children age, they develop even more of a motor phenotype and many will develop an intention tremor or even a resting tremor by adolescence or adulthood, which can also be quite debilitating and impair their ability to self-feed or use their hands functionally. Next slide. And as I mentioned, epilepsy is extremely common. 80% to 90% of these individuals will develop epilepsy by the age of 3. And that's defined as one or more provoked seizures with an abnormal EEG. Even if children do not have clinical seizures, most of them will still have an abnormal EEG, and this serves as a potential biomarker that could be used in future studies. As I mentioned, the variability in the semiology and the severity of the epilepsy in terms of how well it responds to medications can be quite variable among different patients. There is somewhat of a genotype phenotype correlation here. However, with the deletion subtype tends to have a more severe phenotype in most aspects of the disease inclusive of epilepsy. They tend to have more severe seizures or more difficult to manage epilepsy. The seizure types can vary quite a bit, ranging anywhere from atypical absence seizures, which are prolonged staring spells that have a characteristic EEG pattern associated with them. They can have myoclonic seizures, which are brief muscle jerks. And if they happen in the legs can sometimes cause falls. The classic tonic-clonic, which is the generalized tonic-clonic seizure type, which most of us think about with full body stiffening and rhythmic jerking. They can have infantile spasms, which is a unique seizure type of infancy where children tend to have a flexion of the torso forward and arms may fly up and they tend to cluster but they're very brief. Kids can also have an atonic spell where they have sudden loss of tone and can fall. These can also lead to severe injuries of the head or face. And then these children can also have nonconvulsive status epilepticus. So that means that they clinically don't have any motor symptoms, may or may not be responsive at the time and have this prolonged abnormal EEG activity, electrical activity that is consistent with EEG definitions of seizure. And going back to the EEG, many groups have demonstrated this over time. And it's been a pretty consistent finding among most patients with Angelman syndrome, so much so that a well-trained epileptologist can open an EEG and often say, "Oh, this looks like a patient with Angelman syndrome." And this is typically high voltage rhythmic slowing in various regions of the brain and could potentially serve as a target engagement or biomarker that could be followed for a clinical response to an intervention. Next slide. Now how do children get diagnosed? There's a number of different ways. Many children probably these days are getting picked up by chromosomal micro array or a genetic epilepsy panel. I'm sure many on the call are familiar with the efforts that Invitae has had, and in fact, different companies are sponsoring these efforts to improve access to genetic testing in our country. And the Invitae behind the seizure panel is probably one of the instrumental places where children can be identified as having the UBE3A sequence variant. However, if they have a deletion or have normal methylation, additional studies may be required, including the chromosomal micro array to look for deletions and duplications or DNA methylation test to check for those genetic imprinting. You could also do a FISH, or fluorescence in situ hybridization, to go for a targeted mutation or particular allele. Next slide. And thinking back to the genotype-phenotype correlation. So this is adapted from a really nice review that came out this year, where they looked at the different segregated patients into either having a maternal deletion or a nondeletion genotype. And across the board, you'll see that the patients with a deletion tend to have a more severe phenotype. That's more severe developmental delay, requiring higher support needs to maintain as close to independence as possible, requiring quite a bit of support in the classroom and at home. Very limited language abilities and more severe seizures and also a lower response to behavioral therapies or behavioral strategies and interventions. Interestingly, the sleep phenotype, which tends to affect the entire family is relatively similar among all patients with Angelman syndrome. Next slide. And how do we take care of these patients? So as RA mentioned, there's no treatments that are specifically available for patients with Angelman syndrome as of yet. And we primarily focus our treatments on palliative measures and trying to alleviate as much suffering as possible and make children as independent as possible. These children tend to have a normal life span but they are very rarely, if ever, able to live independently. Therefore, they need constant supervision and supports throughout the entirety of their lives. So the economic burden has not been studied outright. I think we can assume and make some predictions that the overall cost of caring for an individual with Angelman syndrome from birth to death is probably quite high. And again, we're only striking at the different symptoms and doing as much support of care as possible. This often means that families are quite burdened with a very busy schedule with physical therapy, occupational therapy and speech therapy, possibly adding an applied behavioral analysis therapy, or AVA therapy, which tends to target autistic traits and autism symptoms. They're also often on a hefty regimen of polypharmacy, requiring medications to treat seizures, possibly behavioral problems, sleep issues and maybe even medications for their movement disorder or their tremor as that gets -- as they get older and that becomes more debilitating. Consequently, many of our medications that treat behavioral seizures can also exacerbate tremor. So we're constantly in this balance of trying to reduce the side effects of our polypharmacy while trying to make life as easy as possible for families and patients. Even with all of this, there's still significant unmet need, which I can't do anything about as a physician these days. So their motor impairment, their speech and communication impairment, I have methods to augment and support their natural ability as much as possible, but I have nothing to do to change the course of that disease or really the severity of that impairment. Their maladaptive behaviors can somewhat be managed with behavioral therapy and medications, but often not to an adequate degree and not without the added cost of side effects. Their sleep problems can often be very difficult to manage, and our medications can cause significant sedation that can also impair their ability to function during the day. And there's absolutely nothing for their cognitive impairment apart from just a good supportive environment through their schools and their therapies and their families. Next slide. And so with that next slide, we're going to turn things back over to Kimberly Lee. And I'll be happy to take any questions, but know that following me is Dr. Allyson Berent who will be diving even deeper into the phenotype and the family...

Thank you, Dr. Goodspeed. We'll now open the lines for Q&A here.

Your first question comes from Gil Blum of Needham & Company. Are all the pathologic accumulating proteins in Angelman known? Could you hypothetically degrade those proteins directly?

That's a really interesting question. I do not know the answer to whether or not they are all known. I can hypothesize that trying to target and reinvent the wheel, so to speak, on being a functioning UBE3A without actually it being the gene. So possibly targeting those different proteins or small molecules would get us into a place where we're not adequately managing the levels of different proteins at different times. So I do not have data to support this, but in general, the brain is kind of like goldilocks. I tell this to my families all the time. They like things -- it likes proteins and different substrates at just the right level, not too much and not too little, and it's very challenging for humans to reinvent that milu biochemically.

Right. Your next question comes from Sami Corwin of William Blair. Are any other organs affected? And what's the impact of life expectancy?

Sure. Really, the vital organs like the heart, the kidneys, the lungs, the liver do not seem to be impaired in Angelman syndrome. It really seems to be central nervous system specific, especially brain specific. Their lifespan is anticipated to be normal. However, as we know, if you work in this neurologic space, it is possible for our patients to have a somewhat shortened life expectancy due to sensitivities to illness or succumbing to a pulmonary infection later in life, but this is not a disorder that is lethal in childhood certainly. These children live to be adults.

Your next question comes from Laura Chico of Wedbush Securities. How many existing Angelman patients do you estimate are currently diagnosed? The vast majority or might there be pockets that are not identified?

I think it's extremely likely that there are pockets that are not identified yet. And even in the developed countries like the United States or Europe, I'm sure that we are missing patients that are out there and just have not had that genetic test to provide us that confirmation. Angelman syndrome is one that was given an eponym and then later a gene is associated to it to UBE3A. But this phenotype is relatively similar to other genetic neurodevelopmental disorders. And so it's also possible there are children that are clinically diagnosed Angelman and haven't had their confirmatory UBE3A identification for a molecular diagnosis. As far as the actual estimation of numbers, I believe Dr. Berent has even better data on that than I do. So I will defer that question to her section and she may address it next.

Great. Your next question comes from Gil Blum of Needham & Company. Any information on the ratio of patients with deletions to those without?

Actually, stay tuned that Dr. Berent has a really lovely slide that gives an overview. Most of the patients are deletions, but she has a very nice slide that gives a wonderful overview of how they're distributed in a large natural history study.

All right. Your next question comes from Silvan Tuerkcan of JMP Securities. What are some of the novel approaches towards future treatments for Angelman's that you are aware of?

So there have been quite a few tried, mostly small molecules. The GABA system is also important in this disorder. So there's been a number of small molecules targeting the GABA system. And then there's a couple of great gene therapy options that are being discussed today that I find exciting and intriguing. So otherwise, most of the studies have targeted just symptom management with existing treatments rather than disease-modifying treatments.

Okay. Your next question comes from Joon Lee of Truist. What sorts of Angelman syndrome patients will qualify for KD? I'm not sure what that is, KD.

Ketogenic diet, probably.

Okay -- yes, of UBE3A-AS versus transgene approaches.

Well, I guess I don't know if I'm interpreting that correctly.

Maybe we'll request that he...

More clarification.

Yes, qualify that question. Okay. We'll return back to that. Next question comes from Yun Zhong of BTIG. Is there a correlation between EEG changes and improvement in other symptoms?

So to my knowledge, that has not been demonstrated yet. However, there is some correlation to severity of the EEG changes, especially the delta power looking at quantitative EEG abnormalities and severity of symptoms. So on that front end of discriminating between levels of severity of symptoms with the EEG that has been done.

Okay. Great. Thanks for that. Back to Joon Lee of Truist. So the question is what sorts of Angelman syndrome patients will qualify for a knockdown of UBE3A versus transgene approaches?

That's a good question. I'm not entirely sure of the answer too. So rather than speculating, I may defer that either to Suyash or RA or to other investigators that...

Kim, I'm happy to answer. Joon, thanks for the question. So I think the target population for either approach would be the entire population to be quite honest. What we're doing is going after a gene replacement approach to kind of mimic the maternal allele from an expression perspective. But then we've seen in the ASO approach some really nice data come out on unsilencing UBE3A ATS. And so essentially, what we're doing is vectorizing that approach and essentially allowing a more stable level of expression because it's being delivered potentially a more stable level of expression because it's being delivered in a vector more potent because it's being delivered in a vector, but essentially just trying to mimic that approach to restore a normal expression. One of the nice things about the paternal allele approach is because you're just basically mimicking natural expression by just knocking down the ATS transcript. So again, we think the population for either approach would be the entire population and for us, we're going to move both programs forward. So as we go through the clinical development and the preclinical development plan, both of these programs are moving in parallel. And ultimately, what we hope is both of them are successful and we can allow patients to decide which one would be more appropriate for them. But both approaches are going to target the entire population.

Thanks, RA. Your next question comes from of BTIG. Do proteins accumulate fast without UBE3A before diagnosis? And how important is early diagnosis? And then he has a follow-up question after that.

Sure. And I think it's important to remember, these aren't necessarily like our lysosomal storage disorders, where we have proteins accumulating and causing toxicity to the cell. This is more of a dysregulation of homeostasis within the cell and cell signaling. So to answer exactly how much protein is accumulating, I don't have that answer, but it's conceptually different from those loss of function in 1900[ genetic ] disorders where you have a toxic buildup of substrate. And I'm sorry, what was the second part of that question?

How important is early diagnosis?

I think in any disorder, early diagnosis is always important. So the younger they are, the more they're going to benefit from therapies. And by therapies, I mean speech, occupational, physical therapy and getting them into supportive care as soon as possible, getting seizures under control as soon as possible. I think that -- if I'm speculating about novel interventions and potentially disease-modifying therapies in general, in child neurology, the earlier you treat the better as in almost anything in medicine. So like -- would leave it there.

Kim, may I add to kind of what you just piggyback on what you're saying? I think absolutely, you're correct. Early intervention is always going to be better. What I will say is just some of the recent developments in Angelman from our colleagues over at FAST and Gene Therapeutics and Ultragenyx did show that by even intervening with older children, you can still actually see a significant improvement. And to be quite honest, we just really encouraging data and hats off to both groups for generating that data to be quite honest. Because I think that really gives us all hope going after this disease that you don't necessarily have to go in as soon as possible that even a patient that has had some accumulated disease could benefit from a potential intervention. And to your point, this is very different from the neurodegenerative diseases that we're going after where you're having this constant loss of neurons. And once a neuron is gone, you're not getting it back. And so I think it was really encouraging to see some really nice clinical data that hopefully we'll be able to generate here as well in the near future.

Thank you, RA, that's such a good point because these are connectivity issues. So it's -- the synapses are not formed properly, but that doesn't mean they can't form properly in the right setting. So I think that's a really important point that I placed over.

You're on mute, Kim.

Thank you. And he has a follow-up question or a second part to the question, which is below what point or threshold can the therapy deliver the best outcome?

In terms of like -- I suppose that's how much do we need to get in?

Well, no, I think the question is really at what point would a patient be too far gone to actually deliver a benefit?

I don't think we know that. And I don't think there is an upper limit of how old is too old. It's again, conceptually different from these degenerative disorders. Neurons aren't dying, they just aren't talking to each other appropriately.

Yes. I would totally agree. Again, I think this is much different from the neurodegenerative diseases. Again, similarly to Red syndrome, we think patients could have a significant benefit. In some aspect of the disease further in life, to be quite honest, understanding that not only are there -- this is a seizure phenotype of developmental phenotype cognitive, but there's also a movement disorder associated with this. And so again, I think if you would ask patients and caregivers, and we're lucky to have Dr. Berent on the line to give her comments. I think patients with -- and caregivers would probably see a benefit just improving one aspect of the disease, if not multiple. So I think to your point, Kim, I totally agree. I think there's probably a benefit that patients could receive even later in life. And because these patients tend to live a normal life span, I think all patients would be appropriate to treat.

Your next question comes from Sarah Kropp of UBS. Does the chromosomal deletion responsible for Angelman syndrome affect any other genes that may be influencing pathology?

It probably is affecting other proteins in terms of -- other proteins that are being degraded. We just highlighted some examples. But it is not a transcription factor. So there's always more knowledge coming out about epigenetic factors and interplay between genes. But I don't think of it as being a gene regulatory element as opposed to more of a protein regulatory element.

Allyson, do you mind providing some thoughts on that question as well?

Sure. So for -- there's different genotypes of course, of Angelman with the most common being the deletion. And in the deletion genotype, there's certainly UBE3A that is missing or non-functional. And there are 10 other genes outside of UBE3A on the deletion genotype that are missing, but they're haploinsufficient. So the imprint is on the paternal which really only impacts UBE3A [ biologically ]. But the other genes that are missing outside of UBE3A are only missing on the maternal. So there is haploinsufficiency. There is some data to support some of those genes like some of the GABA genes may impact seizure phenotype more severely, which might be why the deletion genotype has a more severe seizure and cognitive phenotype. But there are also patients that are documented that have some deletions to those other genes or mutations to those other genes without UBE3A that are considered neurotypical or [indiscernible]. So it's really hard to know the answer to that question without replacing UBE3A and seeing what's left over. And with that said, we actually just funded a mouse model as well as IPSC cell line in an organoid model looking at on a landing pad where we can replace each of the other 10 genes and UBE3A or without UBE3A and ultimately see in the mouse model, what phenotype is left when you have UBE3A but the other 9 genes that are missing. And what does it look like electrophysiologically from the electro physical phenotype? What does it look like when you replace UBE3A and the other genes are missing? And then you can replace each of those genes in more of a landing pad approach so that you can really see what the benefit is when you replace them and what the target would be if one of those specific 9 genes is associated with a significant phenotype. But the big answer is we don't know. The impact they have is probably minimal compared to UBE3A alone.

Thank you. Thanks, Kim.

Your next question comes from Dina Ramadan of UBS. What is the average age that Angelman syndrome patients present to the clinic and are diagnosed? And at what stage of the disease or patient age, would you want to administer a potential disease-modifying treatment to prevent or slow disease progression?

So it's getting younger and younger in terms of age of diagnosis as more people and more community physicians are thinking about genetic testing as kids are presenting with their first delayed milestones, maybe not sitting on time, not walking on time, not developing their language skills. They're getting a CMA done, they're getting genetic testing with their first seizure within the toddler years. So that age has been tracking downwards as increased access to genetic testing is coming about. So the question on age and timing, I think RA and I went through that fairly detailed, but just a brief recap of that is that there probably is not an upper limit of age at which you would not want to treat. But there's reason to think that the younger you go, the better your outcome may be, but studies need to be done to really tease that out.

Yes. Thank you. Your next question comes from Mike Ulz of Morgan Stanley. Given the broad impact of the disease, what are the key areas of improvement you would like to see from a treatment, speech, behavior, et cetera?

So pretty consistently, and I think Allyson will go through this as well, the communication piece is really profound. If patients can't communicate with their families or their caregivers, especially those people outside their homes that don't know them as well, that's really limiting to their level of independence and their ability to adapt to different environments. So communication is always going to be a big piece of that. The seizures, of course, are scary and do cause significant burden, but I think the speech and cognitive impairments are what families often point to because that's a real unmet need here.

Great. And great -- your next question comes from Joon Lee of Truist. Do you have a particular scale you like to use to monitor Angelman syndrome patients?

That's an interesting question. In community practice, there are a lot of different ways that we can manage patients and mark their progression or attainment of new skills over time. Patients are periodically getting psychological cognitive behavioral assessments. And those can vary from place to place. One skill that I do think is really important clinically that is new and was developed for this condition is the ORCA. It's the observer-rated communication assessment. And it has a really nice ability to capture all levels of communication abilities in this patient population all the way from localizations and use of assistive devices up to spoken language. And Allyson will go through that a little bit more in her talk as well.

Great. Thank you so much. Let's see here. Your next question -- and unfortunately, we're a little out of time here. Your last question comes from Dina Ramadan of UBS. What clinical outcomes and measurements do you view as most important in clinical trials as a physician treating these patients?

So I think it's a combination of a biomarker, if available, and a functional outcome measure that links to some degree of symptom management. And that's kind of a blanket statement. In this particular case, I think a great biomarker that we have a lot of data on already is that EEG biomarker. Again, subsequent talks will go into even more detail on that. But then linking it with a nice functional outcome measure. The challenge here is do you pick a performance measure? Or do you pick a parent-based measure, parent report measure? And the answer is probably both, right? So we're probably going to obtain something that is kind of a global cognitive scale that captures motor abilities and language abilities that the patient can do and perform at the psychologist and pair that with parent-rated scales, such as the ORCA that I just mentioned, adaptive skills like the Vineland and quality of life skills or clinical global impression scales. Something like that, that gives a nice round overview of biological target engagement functional improvement in a patient and perceived improvement by parents and caregivers.

Thank you for that. And this concludes our section on the overview of the disease. Thank you, Dr. Goodspeed for all your insights, and thanks for addressing all these questions.

My pleasure. Thank you for having me.

Thanks, Kim.

Now I'd like to turn the call over now to Dr. Allyson Berent. She's going to be discussing the disease burden and providing a patient and family perspective as well as she will be discussing a natural history of the disease as well. Dr. Berent?

Great. Thank you so much. Thank you for giving me the opportunity to speak with you today and represent the patient and the caregiver perspective in living with Angelman syndrome. Next slide. So my name is Allyson Berent, and I am the Chief Science Officer for the Foundation for Angelman Syndrome Therapeutics. I also serve as the Chief Operating Officer of Genetics Biotherapeutics. I'm the Director of the Angelman Syndrome biomarker and Outcome Measure Consortium and the Co-Director of InSync International Angelman Syndrome Research Council. I am a veterinary internal medicine specialists and interventional radiologist. So I come to this as a clinician. I've been involved in medical device development and translational research for the last 15 years on being the PI in over 25 clinical trials and medical devices. But my most important job and my most important role here today and speaking with you is I'm the mother to this beautiful little girl on the right, here name is Quincy, she's 7 years old and she lives with Angelman syndrome. Today, I'm going to share with you a short video highlighting a day in the life of just one family, and that's my family. And I want to remind everyone that there are tens of thousands potentially of other families that live a similar life some more complicated and some a little less complicated. But needless to say, I want to really share with you what it's like living with this profoundly impactful condition called Angelman syndrome. And I was asked to share with you about my journey as a parent, as a patient advocate at FAST and our learnings through my role as the Director of the Angelman Syndrome Biomarker and Outcome Measure consortium, which is a pre-competitive consortium that includes over 24 pharmaceutical companies working hand in hand with the foundations to develop meaningful and impactful end points for those living with Angelman Syndrome. Next slide. We say things are either possible or impossible. And there's no in between. It's a model my husband and I always say about work. And so when Quincy came along, we realized that curing Angelman syndrome is definitely not impossible, which means it's possible. And if it's possible, we're not going to stop until it's done. My name is Allyson. I'm the mother of 3 beautiful little girls. Our middle daughter, Quincy, has Angelman syndrome. Me and my husband are both focused on dealing with treating diseases in animals that there are not good alternatives for treatment currently and finding solutions for disease when there's not really very good options. We have a very good understanding of clinical trials and clinical research and how you can translate something from the bench to the clinic and to make a great contribution to medicine, whether it's veterinary or human. For pharmacy. We can see up here this is all of Quincy's supplements and medications besides the ones that [indiscernible] generator. So we have lots of syringes as basically as a sensory of all things for Quincy to make sure that she has everything actually needs. So Quincy is incredibly similar and incredibly different to any child her age. And having 2 sisters that are typical, it is very easy to compare her, and it's very easy to see the similarities and the differences. And I would say that the joys of Quincy are grander than a typical child. The sadnesses of Quincy are grander than a typical child. Any other kid can laugh, but it's never a laugh like Quincy. The joy brings to your heart when you hear her laughing because this is a classic Angelman trait, but they are happy, happy kids. Having an orphan disease is a lot less difficult when you have a community. The way that you can fix Angelman is that, one, you can replace the gene. So with gene therapy, you take a virus, basically take all the bad parts out of that virus and you put in the gene she is missing. It can take that gene. It can deliver it where it needs to go. And then the gene is functional and can produce the protein that genes normally make, it can produce the one it's missing. Today was a really good example of a kid who just wants to try anything. And it doesn't matter what ability or disability she has [indiscernible] as a kid. And she just wants to have fun. And there's a whole lot in there, and we just need to help her get it out.

This disease and rare diseases need someone -- each of these need someone like Allyson who really have dedicated their lives and people just don't understand. Unless you see behind the scenes, you don't understand the amount of work that goes into this. It is really a selfish purpose to help our only child. But so many people are going to help now that Allyson is a part of this community. I know that.

I just know that the end is in sight and there will be a treatment for her. So I live knowing that I'm going to help her make a bigger difference in this world, not just for her but for other kids like her and not just for Angelman syndrome, but hopefully many other disorders like Angelman syndrome. And so I think that has made me a more valuable human being in this world, where I was just a veterinarian with a couple of cute kids. And that's -- I feel like all I would have been a good daughter, a good wife. And what legacy do you leave with that? And I feel like because of her, our impact on this world will be greater. Next slide. As you can see, that was just a video that was made just to show a day in our life and a day in the life of a family that is doing everything they can for their daughter who was diagnosed with the disorder that they had never heard of. And that brings us to D-day, to diagnose this day. And I pick up the phone from her neurologist when she was only 5.5 months old. And I knew that there was significant developmental delay. I was told over and over again from 3 weeks of age to 5 months of age that she was -- had significant reflux and she was just a little bit delayed, but it was no big deal and she would catch up. And I was being overanxious as a parent and comparing her to my neurotypical child who was older than her, and there was nothing wrong with her. And I actually had to demand to get genetic testing because I believed in my heart as did my husband that there was something significantly wrong. And we demand genetic testing and ultimately at 5.5 months of age, I got a phone call and I pick up the phone and it's her neurologist and he says, "Dr. Berent, I have catastrophic news for you." And that's the first word I heard. And he said, your daughter has Angelman syndrome. And I said, "I'm sorry, how do you to spell that?" And I did the worst thing any parent should ever do is I looked it up on the Internet, and I Googled Angelman Syndrome. And unfortunately, the first thing I saw was Wikipedia. And you should never open Wikipedia when you're a parent playing doctor Google. But I did, and I saw all of the things that this child would never do. She would never walk, she would never talk, she would never live an independent life. She would never self feed, she would never go to college. She would probably never go high school. She would likely never get married, never have a friend, never have a partner and never live independently. And this is a 5.5-month-old baby, and you hear this. And in the first 30 seconds of this diagnosis, I just doomed this child for every dream I had for her while I was carrying her for 9 months. And while I was waiting for answers the first 5 months of her life. And then I moved very quickly to what will she be able to do? What's the prognosis for survival? This is a nondegenerative disorder. And so I felt if this is nondegenerative, I have a lot of hope that there is some time that we can help her. And I really had to figure out how do we give her the best life and advocate for everything she's capable of? And how do we ensure she's accepted in this world of disability so that she can live the most independent and fulfilled life because that's just what every parent wants for their child. And so I realized the first moments of her diagnosis that we had to advocate frequency and we needed to learn and execute in any way that we could do to change the trajectory of the life that she was just granted on Wikipedia. Next slide. And so I learned everything I could. And what I learned was that UBE3A is caused by the disruption of the expression of the UBE3A gene. Angelman syndrome is caused by the disruption of this gene, whether it's through a deletion or mutation or some other gene types that we'll talk about. And ultimately, that results in a decrease in the UBE3A protein ligase. So I learned that this is a protein deficiency and I learned it's nondegenerative. And I read every paper I could on what this means for her and how do we address this both from an upstream approach to a downstream approach. So as I understood the genetics behind it, which can just give a really nice presentation on, I realized -- and I learned that this is an imprinting disorder. So there's obviously a maternal and paternal allele for every copy -- every chromosome that we have. And on the maternal gene in all cells of the body except for neurons, the maternal UBE3A gene on the 15th chromosome is red. But in neurons, the UBE3A gene is actually silenced by something called the UBE3A antisense transcript, which Ben will go through in a lot more detail after me. But ultimately, with the silencing of this long non-coding RNA over the UBE3A gene on the paternal copy which only occurs in neurons, the rest of the body has a functional paternal copy of UBE3A. But all of us have a silent copy, paternal copy in our neurons. But individuals suffering with Angelman syndrome are also missing the copy on their maternal allele. So they essentially have no functional copy in their neurons, but they all have a functional copy peripherally from their paternal allele. They just don't have it on their maternal allele. And so really, this is a neurologic disorder. This is a disorder of neurons due to the fact that we have a silent copy paternally. And so they are essentially null. Individuals of Angelman syndrome can present with 5 main genotypes. The first being and the most common being the deletion. The deletion occurs in about 70% or so of the population. And that involves the UBE3A gene as well as potentially 9 to 12 other genes surrounding the UBE3A gene, as you can see in that image in the center. There's also the UBE3A mutation genotype, which is essentially just a miscoding of the UBE3A gene. And therefore, the UBE3A protein is typically truncated so it's not made. Sometimes it is made, but it's not functional so that there is a protein that is around, but it is not working properly. And then there's also this double imprinting disorder or uniparental disomy where there's 2 paternal copies of the gene. So you have this antisense transcript of both copies. So you're functionally silent on both sides the maternal and the paternal, instead you have 2 silent paternals. And an imprinting center defect is very similar to UPD in that, you create this imprint on the maternal copy so it too is silenced very similar to what you see with UPD. And so we really understand that the majority of the patients do have this larger deletion genotype, which is associated with a more severe phenotype. And then the next most common genotype and some point mutations. Next slide. And so how does this manifest? And I don't need to go through this again because Kim you did a really great job going over the clinical manifestations of AS, but they are severe, they are lifelong and they're incredibly impactful. And they're not only impactful to the patients, which is obviously the most important thing, but they're tremendously impactful to the families. And so obviously, individuals with Angelman syndrome have a universal lack of speech. If individuals speak, which is essentially not present in the deletion genotype, but in some of the other genotypes, they may speak 1 to 5 words. Very, very limited speech if any at all, most of the time there is none. They have life-threatening and debilitating seizures at times with significant developmental delay. They can be severely incoordinated and have significant ataxia, which is associated with falls and unsafe navigation around objects, in which case in all aspects of life, they essentially need a one-to-one support in order to ensure even if they're mobile, that they're safe when they're mobile. They have significant apraxia and dyspraxia. So because of that, they're very difficult to test in a clinical setting or in a school setting when it comes to taking tests. So we have to think about that when we talk about outcome measures and how we best assess these patients in clinical trials. Some individuals can have aggressive or disruptive behaviors. It's not very common, and it's more common in certain genotypes than others, but it definitely does happen and it's something that could be tremendously impactful to families if it happens. Significant sleep disturbances and severe and are very, very common with Angelman syndrome. Almost all individuals with Angelman have some sleep disturbance, some so severe that individuals sleep 2 to 3 hours a night, their entire life from birth until death which if you can only imagine the impact that, that could have on a family as well as the learning of that individual. Individuals have feeding issues, usually very early in life and can continue on into adolescence and adulthood. And overall, they just have a complete inability to live an independent life no matter what genotype they have. And so what impact does that have on the family, as you might imagine? It's been documented that families living with Angelman syndrome have an inability to maintain employment at times, have significant depression and anxiety, high stress, a significant loss of sleep, can feel socially isolated, can have a significant impact on family, both nuclear and remote family relationships. And they have a very difficult time caring for their other children when they're so focused on their individual with a significant disability. And overall, there's just tremendous fatigue within a household, which can have very significant impact on that family. Next slide. And so what really coming through the diagnosis was one of the first papers that I came across as I did my research was this paper on somatic mosaicism in patients with Angelman syndrome. And this is incredibly impactful. And this really guided my desire to realize how much impact we can have on this population. And so what this paper showed was that there's a very small population, probably around 1% that are mosaic. And so those individuals have some normal cells that are expressing maternal UBE3A and some cells that are -- and a majority of cells that are not expressing UBE3A. And you can quantify that peripherally. And when those are quantified, what has been documented is the expression of about 1% to 40% of cells that are appropriately expressing maternal UBE3A. And then that translates to probably in the brain. That's what you're seeing as well in terms of the number of neurons that might be expressed in UBE3A. And so when you look at the clinical score of those patients, if you look at the individuals that are mosaic, that score between 1% and 5% of normal functional UBE3A, they have few to no seizures, they are fully ambulatory. They may have a little bit of ataxia, but they are walking very independently, and they have some speech. If you look at individuals that are around the range of 15% to 25%, they have no seizures, their ambulation is essentially normal, and they can speak in some full sentences. They can be in typical classrooms. They need some supports, but they ultimately are living much more of an independent type of life. And so what we think about when we meet these individuals, which I've been so blessed to have met many of them, what you can see is that the difference between my daughter who have no UBE3A and a child with 5% to 10% is an absolute different child. And because of that, I have so much hope that expressing only 1% to 5% of UBE3A in this population can make a profound difference in the phenotype and the quality of life of not only the patient but their families, their siblings and their peers. Next slide? And so why are we here? Next slide? We're here, and I'm here representing the Foundation for Angelman Syndrome Therapeutics as well as the Angelman Syndrome Biomarker and Consortium. And I think what I really want to share is the role and the mission of the community. We're here as parents and FAST is certainly here as parents and professionals that are really dedicated to bringing transformative treatments to all individuals in the world living with Angelman syndrome, regardless of their age or their genotype with our ultimate goal, of course, is to reach the "cure" for Angelman syndrome. And we do this through a very aggressive research agenda while we collaboratively and in parallel support collaboration, understanding, readying and expediting. And so that's really where we are, is that we want to accelerate proof of concept through the entire drug development life cycle so that we can accelerate drug development in the safest and most aggressive way so that we can bring meaningful treatments to individuals living with Angelman syndrome, regardless of their age and regardless of their genotype. Next slide? And so really briefly, what we've done, FAST was found in 2008, and this is well before I joined the foundation in 2015. And so this is on the backs of so many individuals, a very highly functional, very supportive working Board that are full volunteers as well as on the backs of incredible researchers that have really supported the mission of the community of bringing meaningful treatments to individuals with Angelman syndrome. And through that time, FAST has spent over $25 million in research to support, bringing treatments to individuals with Angelman syndrome. That's been through a variety of measures through $3 million supported to creating novel animal models like a large deletion rat model, a large deletion mouse model, both including and not including UBE3A as well as a deletion and mutation pig model as well as funding over 20 laboratories to really accelerate therapeutic platforms so that we can really be platform-agnostic but disease-specific as we accelerate drug development for Angelman syndrome. And through that, we've supported over 9 gene therapy or disease-modifying strategies in order to bring this forward for individuals, I think with Angelman syndrome. Next slide. And so the road map to success for our community is to really think about it in 3 pillars. Firstly, how do we investigate therapeutic strategies, which we're going to talk a lot about today. So we're going to think about things like gene or protein replacement therapy, replacing that missing maternal allele, that missing maternal copy of the gene, whether we replace that through protein or whether we replace that through a gene replacement therapy. We also want to think about how we can activate that silent paternal copy of the gene. We have this beautiful long non-coding RNA that's just sitting there that is stopping the read of the transcription elongation of the paternal copy of the UBE3A gene. So if we can inhibit that [indiscernible] transcript, we can actually activate the paternal copy of the gene and give that a single gene copy that these neurons are missing. We also want to think about symptomatic treatments, and there was a question asked earlier on what about downstream targets and how do we address that protein excess with that ubiquitination that's actually been lost with that accumulation of proteins at the synapse. As Kim explained very beautifully, this is a synaptopathy. There is a missing communication and the homeostasis between neurons talking together is very disturbed in Angelman syndrome. And if we can address that synaptopathy and we can allow the communication between those neurons to be better, even if that's with some downstream targeting of those -- of those downstream proteins that are not being ubiquitinated or need to be ubiquitinated, then ultimately, we may be able to benefit this disease and the symptoms of this disease. And while we're doing all of that and investing millions of dollars into all of these therapeutic platforms, we also need to, in parallel, be thinking about how we're going to prepare for clinical trials. We don't want to have a drug product without having the preparation to be able to launch into a clinical trial. It is our job as a foundation and as a community to be the low-hanging fruit for any pharmaceutical company that wants to work in Angelman syndrome. And so how do we do that is we make it easy for everyone to be able to leverage all of the work that we have done to be able to utilize that for the benefit of their clinical program so that we can get the most meaningful and safe therapeutics to our kids. And we can accomplish that through obviously the animal model work that we've done as well as others through developing and supporting the Angelman syndrome biomarker and Outcome Measure Consortium, which we are very robustly invested in. And through helping all the pharmaceutical companies in this space really work toward clinical trial design and how we can best measure these symptoms in individuals with Angelman syndrome. An additional -- and additionally, we want to ensure that we are creating the best models for all of the genotypes. So we never leave a single genotype behind and all therapeutics can benefit all genotypes, where we figure out which therapeutics would benefit, which phenotype -- which genotypes in the best possible way. Next slide. And so this is just the landscape slide to show you where there are players in the field, which I know a lot of people wonder it and really want to understand where players are and where they're sitting. And so we have a lot of players on the AAV gene replacement side, the secretary protein gene replacement side. We have a lentivirus hematopoietic stem cell gene therapy approach as well as an enzyme replacement therapy approach. And then there's paternal gene activation, so stopping the stop. So we have ASOs and SMEs, artificial transcription factors, CRISPR, and then we're going to talk today about shRNA. And there's also some work being done in miRNA. And then, of course, there's companies working in downstream targets as well, looking at small molecules and ligands in order to address that synaptopathy. Next slide. And so this is just an example to show you after a gene replacement therapy, and these videos, I want to acknowledge [indiscernible] and Joe Silverman for sharing these videos with me. But ultimately, what you can see on the top of the view is an Angelman syndrome mouse before gene replacement therapy. So this is a hematopoietic stem cell gene therapy. But what you can see here is this is a mouse that's on a [ digit gate ]. So on a treadmill, and he really is incoordinated, and he really can't keep up. He can't get his back feet coordinating with his front feet and so he keeps hitting the bumper in the back. And then after gene replacement, you can see the coordination is incredibly improved. These animals are able to run and keep up with their wild-type peers. And ultimately, there was no difference in this measure neuro behaviorally between an Angelman-treated animal and a non -- and a wild-type animal without Angelman syndrome. Next slide. And so that brings us to with all of these promising therapeutics on deck and all of the support that the entire community is bringing forward to ensure that all of these platforms get accelerated through the drug development life cycle. What's really important to do at the same time and in parallel with all of this work is to really understand what matters to patients and caregivers. And I shared a lot with you what matters to me and Quinsey and my husband and her sisters, but let me just share some data. Next slide. And so we really support huge efforts to clinical trial readiness. And so there is a disease burden and unmet clinical need publication that was published by Ann Wheeler and her team at RTI as well as Boston Children's Hospital, which is an excellent paper. And then more recently, in 2020, the Angelman Syndrome biomarker Outcome Measure Consortium as well as Roche partnered in publishing a disease content model. And so that reference is right there. In addition to all of that, we also have an amazing natural history study that was funded from the NIH from 2006 to 2014, under the support of the folks like Len Hanan and Lindberg at Boston Children's Hospital as well as Radieschildren's and 6 universities throughout the country, supporting 302 patients in the natural history study and then the FDA furthered that work from 2018 to 2022. And there have been another 150 patients enrolled since 2008. So we have over 400 patients that have been enrolled in the natural history study. And that is a continuum of endpoints that are being assessed in patients, which I'll go into in the next slide. Additional data and natural history work is also now being done. We have a natural history study out of FAST U.K., FAST Latin America, FAST Italy, FAST Spain and FAST France. And so you can actually go back one. And so we also have a global income and syndrome registry that can identify patients around the world, and this was supported out of FAST Australia in 2016. Over 2,000 patients were identified outside of the United States, and now this is being translated into multiple different languages, hoping to leverage this work to really identify the demographics of the individuals around the world living with Angelman syndrome. We'll talk more about the Biomarker and Outcome Measure Consortium. And then I already mentioned all of the model development that we've been doing in order to try to really help support and understand the different genotypes so that we can be clinical trial ready for everybody involved in Angelman syndrome. And finally, on this slide, I just want to mention that there's also an Angelman syndrome infrastructure for that's doing a ton of preclinical testing for many pharmaceutical companies in this space to be able to look at the same neurobehavioral assessments in that -- with the same team to really understand the phenotype of Angelman syndrome and really look to see if any of the drug candidates act these different pharmaceutical companies are testing is being -- is working really being evaluated by experts in neurobehavior for Angelman Syndrome because those neuro behaviors in the rodent models are very unique and really should be evaluated by experts in the field to understand how to assess those phenotypic test in animal models. Next slide. And so this is the natural history study since 2006, 302 patients, mean age of 5.5 years. And all of the neuropsychologic tests that were evaluated in these individuals, including EEG. So there's a lot of work that's been done and over 20 publications have come out of this work. And now there's an additional study from 2008 to present, like I said, that's exploring the most utilized endpoint supported by the ABOM. Next slide? And so this is the global Angelman syndrome registry just for a visual to show you all of the clinical parent input, clinical input and modules that are being evaluated, that are supported by pharma industry in order to put information and questions in there the parents can fill out so that different industry partners can get answers to questions that they specifically want answered for their own therapeutic platform. Next slide? And so this is a survey that FAST planned in 2018 prior to the publication of the disease concept model, where we serve 332 parents and caregivers about what key outcomes will be most impacted with the treatment for individuals with Angelman Syndrome? What was their dream outcome? And as you can see, overwhelmingly, speech and communication were number one. Second to that were seizures and mobility, and then cognition and daily life skills. And that was completely corroborated with the disease concept model that was published 2 years later. Communication impairment was the #1 desire of parents to see improvement in their individuals living with Angelman syndrome. Next slide. And so really preparing all the stakeholders for trials, we were lucky enough to be invited to speak with the FDA through a patient-focused listening session in 2018. And we met with about dozens of individuals at the agency, both as a foundation and that we're present there with both CDER and CBER, Orphan Products, Rare Disease Staff as well as the clinical outcome staff. And we really introduced 5 patients and caregivers to the FDA and and they were able to ask questions. And the questions they asked these patients and these parents were what would be the most impactful treatment for you and your child? What would you see change with a meaningful therapeutic for your child? And the answer across the board by every single parent and they were not prepped was that they would like to see an improvement in communication and expressive language in their child. It doesn't matter if that's with gesture, with sign or with an AHD device, they would like to see ideally speech. But most people would feel that wouldn't be attainable, that if they could have any better expressive language and be able to communicate their needs that would be the most important outcome for caregivers. And so the FDA listened very carefully as we presented all of the natural history data that was published and all of the endpoints that had been available. And what they did was they suggested that we consider developing a very simplistic communication assessment tool to capture very basic but very important changes in communication for a child living with AS and the clear message from these parents supported that. And so when we left that meeting, they had sent an e-mail introduction, introducing us to a team at Duke University. And Duke then we partnered with to develop the ORCA, which is the observer reported communication ability measure for Angelman syndrome. That has been developed. It's currently in the final review process of publication. And since that time, the FDA has actually funded Duke University an additional $2 million to develop the ORCA for 14 other neurodevelopmental disorders because they have this measure now in their hands at and they see it for Angelman syndrome the value that it has and how sensitive it is to this population, and there's clearly an unmet need for these endpoints for other neurodevelopmental disorders. So the FDA was an amazing partner in helping to develop this tool and have taken this tool very seriously as an endpoint for Angelman syndrome. Next slide. And so that brings us to the Angelman Syndrome Biomarker and Outcome Measure Consortium. Next side. And so what we really do here is we talk to all of the stakeholders, and that includes the other foundations like the Angelman Syndrome Foundation, FAST Australia, FAST France, FAST U.K. and FAST Latin America. And we work together as foundations to talk with all of our pharmaceutical partners and stakeholders, including academics and clinicians, to really define what functional domains or priorities for clinical trials. And so we have -- it's a pretty competitive space that is fully funded by FAST currently. And it is a ton of work that's being done together with the Angelman Syndrome Foundation. And we are working to really leverage what pharmaceutical companies want. And the #1 tool that they are most interested in is a communication tool. And next to that is fine and gross motor and then a global tool like CGI. And then we're working with all of the pharma partners as well and developing biomarkers that are most appropriate in order to be able to measure target engagement and change early in clinical trials so that we can then use the functional assessment to pair them with an active and functional biomarker. Next slide. And so overall, we've developed a dashboard for all of the different focus domains that was through the disease concept model as well as the parent survey. And then we looked at all of the measures evaluated in Angelman syndrome and prioritize them in a hierarchy of what endpoints we should utilize for clinical trial development and clinical trial design. And so we've been working hand-in-hand with all of our partners in the space to ensure that all of the patients are able to be measured in the endpoints that would be most appropriate for the mechanism of action of their drug. And all of these measures have been funded by the foundation, therefore, are available to all pharmaceutical partners to utilize the data and to be able to use the measures and the endpoints at their will. Next slide. And so overall, the mission of our foundation and certainly of our whole community is that we need to collaborate, we need to understand, we need to ready and expedite. And there are numerous potential therapeutic platforms investing in Angelman Syndrome. And AS is unique with very promising clinical results to date in which we can attack this from many different angles. So where our gaps are is that we need to ensure that we have a meaningful transformative treatment for every single individual living with AS regardless of their age and regardless of their genotype and ensuring that we meet the needs of all of these individuals in the best way of course, the safest way, but also the fastest way. Next slide. And so I want to thank you for all of your time and I want everyone to remember why we're here. We're here for these beautiful thesis, and we're here to remind everybody that what we were told was impossible is likely possible. So I thank you all for your time.

Thank you, Dr. Berent for your insightful presentation. I would now like to open up the call for Q&A.

Your first question comes from June Lee of Truist Securities. How would you describe the risk-benefit tolerance of Angelman syndrome patients and families? And do you see ASO as possibly an augmentation strategy for gene therapies?

So I can -- I think RA was going to answer most of the questions here, and I can briefly answer that and then pass it off to you, RA. But I think that where the community is, is that we have really been told for so many years that all these animal models are being treated and symptoms are being rested in all these animal models. And humans are not mice and mice are not humans. And trying to understand what therapeutic is most meaningful and most impactful to patients is really going to be determined by the patient. And so ultimately, I think that the clinical trials are going to tell the story. And I think any therapeutic platform is going to have their story to tell, and we're so excited with all of the great news out there for the multiple clinical trials that are ongoing and the future ones that are starting. And so our community is just so excited to see the fact that everyone believes that any age is possible and any genotype as possible. And the data will speak for itself. And we're excited to get out of the decade of being a mouse with Angelman syndrome, and we're just so excited to finally be in the decade to be a human with Angelman syndrome and really see a transformative difference in these kids and adults, of course.

Thanks, Allyson. Yes, just to piggyback on what Allyson mentioned, typically, we don't ask our caregivers questions on our investor calls. But just to piggyback on what she mentioned. I think to take June second question around augmentation strategy. I think we've seen something similar play out in SMA, where we had 2 approaches. At first, there was an ASO approach and then a gene replacement approach came on behind that. And I think ultimately, what we're starting to see play out in clinical practice, particularly in SMA, is this co-administration both using both mechanisms, whether it's a augmentation strategy on the Delta 7 gene SMN2 or the gene replacement strategy. And so I think if we want to hypothesize, I think that you would have the ability to do that. I do think there needs to be more elucidation around overexpression of UBE3A because you want to make sure that you don't overexpress the gene because there could be some issues with that as well. So I think more research needs to be done, but certainly, it could be approach that we certainly see playing out in other areas where you use ASO and gene therapy together.

Great. Thank you, RA, and thank you, Dr. Berent. And as RA mentioned, we will not be addressing questions. Dr. Berent will not be addressing questions. So thank you, Dr. Berent, for all your insights, and it's very helpful, very moving, and thank you for participating. And with that, I'd now like to turn the call over to Dr. Ben Philpot who will be discussing a novelty replacement for the treatment of Angelman syndrome. Dr. Philpot?

Thanks, Kim. Next slide, please? So once again, I'm Ben Philpot. I'm Associate Director of the Neuroscience Center at the University of North Carolina. I'm also associated with the Caroline Institute for developmental disabilities and a professor in the Department of Cell Biology and physiology. Next slide?. Before I start, I'd like to list my disclosures, which is I've licensed the gene therapy to patient gene therapies, and I have a sponsored research agreement with them as well. Next slide. So I'm going to reiterate just a few things just because they're so incredibly important. So we know that Angelman syndrome is caused by mutations or deletions of the UBE3A gene. So it has a prevalence of around 1 in 15,000 individuals. And about 70% of these individuals have Angelman syndrome arrive because there's a deletion of the 15q11-q13 chromosomal region where UBE3A resides. But one of the clues that tells us that UBE3A is really the major player of the genes in this region is that about 10% of the cases of Angelman syndrome are caused just by mutations that are restricted to the UBE3A gene. So a mutation or deletion of just UBE3A gene is sufficient to cause the whole phenotypic spectrum of Angelman syndrome. Next slide. To remind you, there are unique genetic situations that UBE3A is in, on the top panel, you'll see a representation of the maternal, represented by Mat in red or paternal, represented by, Pat, in blue, chromosomal region of the stretch. And in neurons, UBE3A is expressed only of the material allele, whereas material allele, the UBE3A gene is epigeically silenced due to a transcriptional competition with a UBE3A antigen, which is just as long noncoding antisense that extends into where UBE3A is trying to transcribe. So on the bottom, you can see the situation in a neurotypical individual where UBE3A protein is produced only of the material inherited allele in neurons. And Angelman syndrome, this is mutated or believed there's no protein that's generated. Next slide, please. So today, we're going to be talking about 2 treatment strategies for Angelman syndrome, gene reactivation strategy and gene strategy. Next slide. I'm just going to give you a little bit of a precursor to what Ryan Butler has talked about. He's going to be talking about a gene reactivation strategy, in which uses an shRNA approach to knock down the UBE3A antisense. Next slide. And when you knock down the UBE3A antisense, what this permits is it permits UBE3A protein to be generated off that paternal inherit allele. Next slide. So in this way, UBE3A protein can be made of that dormant allele when it's reactivated with shRNA-mediated knockdown of UBE3A antisense. So this is a really exciting technology. But what I'm going to be talking about, next slide, is a gene addition strategy in which we're simply packaging UBE3A into an adeno-associated virus and replacing UBE3A expression through this virally delivered vector. Next slide. I wanted to mention that the proof-of-concept paper for this gene therapy was recently published in JCI Insight. So you can really look at that paper for a lot more details into the nitty pre. So I would encourage you to go look at that and look at thoroughly. Next slide, please. But there are a lot of things that we considered when trying to design our viral vector to treat Angelman syndrome. And one of them, of course, is just a target biology. And that's -- there's broad expression of UBE3A in the entire brain. So what you're seeing on the right are examples of UBE3A staining, which comes up as green in the mouse vein in sagittal section. And on the top, you'll see a neurotypical or wild-type mouse, where UBE3A is expressed on maternal allele and you can see it's expressed in neurons pretty much throughout the entire brain and in a relatively uniform distribution. But on the bottom, in our Angelman syndrome model mice, where we've deleted UBE3A just from the maternal allele, you can see that there is a near complete loss of UBE3A. And really what's happening is there's a relatively complete loss of UBE3A in neurons throughout the brain. And if you were to zoom in really closely, you could see that there's a little bit of expression left in glial cells is biolically expressed UBE3A. So this expression UBE3A is unique to neurons. Next slide, please? We think it's really important to compare the biodistribution of UBE3A in mouse and in nonhuman primate and even in human tissue. But what we're seeing is a pretty consistent story. So whether you're looking at a mouse or the nonhuman primate, which is the image on the right, I think -- believe this is a 3-month-old animal that we take this from, as you see broad and virtually ubiquitous expression of UBE3A and it's fairly uniform across different regions of the group. Next slide, please. And it's important to know what cell types UBE3A is expressed in. And what you're seeing here is just an idiocychemistry from a 2-week-old -- this is the cat, monkey. And the take-home message here is that UBE3A is expressed at very, very high levels in neurons and at low levels in cells. So we know that we really want to target to get reexpression UBE3A in neurons. Next slide, please. One of the interesting things about UBE3A is that subcellular localization changes across development. And we've seen this in mice, and we also wanted to confirm this in nonhuman primate model. So what you're seeing, if you look at kind of the middle columns where UBE3A is labeled, at either gestational day 100, gestational day 150, 2-week-old, 3-month-old is that the distribution subcellular UBE3A changes across this developmental time force. So for example, if you look at gestational day 100, you can -- you might be able to appreciate that there's a more cytoplasmic expression of UBE3A. But as you move towards a 3-month-old, UBE3A expression becomes much more nuclear. So that's why there are much more discrete but because it's UBE3A being labeled in the nuclei of neurons. Next slide, please. You can really appreciate actually this developmental shift in UBE3A subcellular localization. If you focus in on a gestational day 100 recess the CAC brain because there is a development of progression even in the layers of cortex. Because the cortex develops, so the deep layers grow first and then progressively add on layers and the more superficial layers at this time are more immature. So if you look at a more superficial layer like layer 3, so in that middle box or UBE3A's label, you'll see a more cytoplasmic distribution of UBE3A. And then even at the same age, if you look at layer 5, where the neurons are more mature because of this kind of inside out development of the cortex, you can see that there's a much more nuclear expression of UBE3A. So we really see that UBE3A comes increasingly nuclear expression as neurons mature. And this is a dynamic that is conserved from mouse to primates. Next slide, please. So a challenge in trying to think about a gene therapy for Angelman syndrome is that there are 3 UBE3A isoforms that exist. And unfortunately, they have a very complicated nomenclature because the nomenclature for the ice forms and mice doesn't line up with a nomenclature in humans. In mice, there are 2 isoforms that are expressed, in isoform 2, which is a long isoform and isoform 3, which is a short isoform. If you look at human, there are 3 isoforms that are expressed, 1 short form and 2 long isoforms. The short form is isoform 1 in humans and the long isoforms 2 and 3. As you can tell from this sequence map, there's a perfect conservation of the short UBE3A isoform during mammalian evolution between mouse and human. The long UBE3A isoform is also highly conserved. We think that these different internal extensions on the UBE3A isoforms tends to be really important, and they might mean distinct cellular functions such as a subcellular localization of UBE3A as well as access to unique binding partners that could change the functions of UBE3A and direct them to different [indiscernible] . Next slide. So one of the things that we thought was really important to do was to look at the isoform expression of UBE3A in both mouse and human across early development. So on the top, you can see the isoform fraction, and this actually point out [indiscernible] level, the short to long isoforms, the short isoform is in blue and the long isoform is in green. And you can see in the mouse, there's a fairly consistent roughly 3:1 ratio of the short expression of isoform being expressed to the long form being expressed. We see this across different brain regions, across forebrian, hindbrain and mid-brain and across early development, so from embryonic day 10.5 to post[indiscernible] , for example. In the humans there you see something very similar. So if you look at the isoform fraction of the short isoform 1 to long isoform 3, you can see that at the messenger RNA level, at least, it's roughly a 4:1 ratio. And you also can see that short isoform -- or sorry, long isoform 2 in humans is hardly expressed at all. But this 4:1 ratio at the mRNA level turns out around the 3:1 ratio at the protein level. But this 4:1 ratio at the messenger RNA level is really observed across development, so looking from the first second third from trimester embryonically all the way up to adult and across brain region, whether you're looking to hippocampus, prefrontal cortex or sensory cortex. So this favored expression of the short isoform to a long expression ratio, this conserve expression of the short to long isoforms and roughly 4:1 or 3:1 ratio we think is really important. The first of that is just that it's highly evolutionarily conserved irrespective of brain region or development stage. It's observed in mice, it's observed in nonhuman primates, it's observed [indiscernible] . And it's also important to note of the long human UBE3A isoforms, there is predominant expression of isoform 3 and isoform 2. Next slide, please. There's good reason to believe that expression of both the long and the short isoforms of UBE3A are important. And some of this evidence comes from the fact of a really elegant study from Wen-Hann Tan's Group in which they studied a family that had a start codon variance that got rid of the expression of the short isoform UBE3A and all that was left was especially the long isoform UBE3A. Now while these individuals presented with Angelman syndrome, the phenotypic severity was very much reduced. So this is one of the clues that while the short UBE3A isoform might be very important, and in its absence, you will get an Angelman syndrome phenotype. It's also important the long isoform of UBE3A also has something to add and contribute to a normal phenotype. Next slide, please. So in thinking about how we wanted to develop an AAV vector for on-target UBE3A immune Angelman syndrome, there were a number of things we wanted to consider. First of all, we wanted a broad distribution in neuron. So we really want to get UBE3A to be expressed in neurons, and it's not necessary that it will be expressed in other cell types. We also thought it was important to express both isoforms of UBE3A that are predominant. So as both the short and the long isoform of UBE3A. And not only did we want to express both isoforms, but we wanted to express them in their endogenous ratio of the short isoform to long isoform. We think that this could allow us to recapitulate the endogenous subcellular localization of UBE3A, which could be very important. So whether UBE3A is in the nucleus, 1 role there versus being out in the cytoplasma or even at synaptic sites. Another design feature is we thought it would be advantageous, particularly for, well, solely for preclinical studies to be able to trace virally-expressed UBE3A transient expression because this will help us identify the biodistribution of UBE3A getting through our transfering expression. Next slide. So this is a design that came up for our human UBE3A codon optimized vector that would allow us to express both isoforms of UBE3A. And basically, what we did is we leveraged different strong and weak sequences that would allow us to preferentially express the short isoform over the long isoform. So we use a human synapsin promoter because this is a target reexpressed on UBE3A neurons, and this would reduce peripheral toxicities that could arise from average delivery of the viral payload. This is a codon optimized UBE3A, which will facilitate tracing of the viral vector in recipient tissues. Originally, for the proof-of-concept studies, we packaged it in, which is an AAV9 variant, and this allows for a robust widespread delivery in the brain. We've delivered this viral vector ICD. So intracerebral metrically. This will allow us to get an early onset of expression. It will also be able to allow us to stay kind of the maximum opportunity for therapeutic benefit and also give us a good idea of kind of the worst-case scenario and be a really robust test of toxicity, which is something we wanted to assess early on in the state to make sure the viral vector that we're producing is not only efficacious but also that it has a safety profile. Next slide. So for our preclinical studies, we really have taken advantage of Angelman syndrome model mice because they have a lot of advantages for a preclinical platform to assess a viral vector design. As in humans, the Angelman syndrome model mice have a targeted disruption of maternal UBE3A. They also exhibit learning and memory impairments. They exhibit motor impairments. They have increased seizure susceptibility, enhanced EEG delta-wave activity and they exhibit microcephaly. So many of the same features that you observe in Angelman syndrome patient population you need to recapitulate many of these features in Angelman syndrome model mice. Next slide. So our current optimization really allowed us to have an efficient tracing of our UBE3A payload. And on the left-hand side, basically, what you're seeing is a sequence homology map between or codon-optimized UBE3A, which is labeled HGV3up on the top, our human UBE3A, monkey UBE3A, our mouse UBE3A on the bottom, and red indicates sequence divergence. And you can see that our codon-optimized UBE3A has a high degree sequence divergence from human UBE3A or mouse UBE3A. So this allows us to assess the viral payload that we've delivered. So if you look first at the images on the right, in the wild type mice, we're seeing NC2 hybridization for labeling for Rbfox3, which is new and it's a neuronal marker, or UBE3A. And in the wild-type mice, you don't see any signal for the Codon-optimized UBE3A. In the Angelman syndrome model mice where we've delivered our viral payload, you can see a very nice signal for our codon-optimized UBE3A. So it can really pick it up very nicely. The quantification of all this is shown in the mix where if you look at transcript levels of the codon-optimized UBE3A, you don't see any wild-type mouse, which is the first bar in gray. We've delivered AAD to wild-type to mice pick up the signal. An Angelman syndrome model mice, which is the next -- the third bar, you don't see any signal in the Angelman syndrome mice where we've delivered our AAV payload, once again, you'd see the signal. So we can really differentiate between UBE3A that we've delivered from our biovector from the UBE3A that's there. Next slide, please. One of the most important features of the viral vector is that we could recapitulate this endogenous short to long ratio, having roughly a 3:1 expression at the protein level. And to verify this, we took advantage of mice generated from Garzas lab, in which it knocked out the long isoform. So if you look where it as long only, these are mice that are lacking -- sorry, the lacking the short isoform, so they only have the long isoform expressed. And you can see the UBE3A band's picked up. It's only showing the long isoform. And on the Angelman syndrome mice that have been treated with AAV, which are flanking that one, you can see both the slightly lower molecular weight band and the high molecular band indicating the expression of both the long and the short isoform. When you quantify this, we are getting near to the 3:1 ratio of the short to long isoform that's expressed in the wild-type animals. So we're really excited by the fact that we've been able to recapitulate by leveraging these weak and strong, we've been able to recapitulate the endogenous ratio of short to long isoforms in UBE3A expression. And we think this is a very important design feature. Next slide. So to examine the biodistribution of UBE3A. In our initial studies, we delivered our viral vector into mice, just at the afterbirth, and we delivered it ICV. So intra-cerebellar ventricle. We then interrogate the expression of the viral vector in [ poster ] 10-, 15- or 25-day-old nights. Next slide. And what you can see here is that, we're already seeing expression of our viral vector and it's very strong already P10. And you can also see that expressions maintained at 5 and post-LA25. And you can see a difference a little -- if you've got a key night, you see a little bit of a difference in expression pattern. And you can actually tell that the UBE3A is becoming more and more nuclear in an expression profile as you move from to post-ALD25. And you can see this even better on the zoom in on the next slide, where at Post ALD 10, you can see that the virally-expressed UBE3A has a much more cytoplasmic expression. So of course, you see some of the nucleus but you see much more relatively out in the cytoplasm relative to adult animals, I should say. And post ALD 15 to 25, the virally delivered UBE3A is becoming increasingly more nuclear in its expression profile. So it doesn't mean that there's not any expression in the cytoplasm and in airside to 25, but it's just very, very reduced compared to this very intense localization of UBE3A in the nucleus. So our virally-delivered UBE3A is recapitulating precisely what we see during natural development. And that's that UBE3A goes through a developmental shift in its subset of localization such that in immature neurons, it's expressed in the nucleus as well as in the cytoplasm and it becomes increasingly shifted towards nuclear expression as you mature into adulthood. Next slide. So what I've shown you so far is that our viral vector has -- is producing the qualities that we think are desirable in a gene therapy. So we're getting expression in neurons. We're getting a broad distribution, and we are getting UBE3A to express both the short and the long isoform and in this roughly 3:1 ratio at the protein level. But what's the effect on the Angelman syndrome model mice? Next slide. So one of the things we've looked at is whether UBE3A can rest you a species appropriate behavior. And one of the things that mice are very good at is nest building. So normally, if you give a mouse a nest, give them nesting material, they breakup nest material and use it to nesting. So wild-type mice use most of the nesting material that's provided to, so roughly 60%. Angelman syndrome model mice use very, very little nest material. So that's closer to 10%. And when we gave the Angelman syndrome model mice, the mice that were treated with our gene therapy, you can see that we got a strong partial rest of this nest building phenotype. So we're able to exhibit rescue of a very species appropriate behaviour. Next slide, please. One of the challenges that individuals with Angelman syndrome are faced with is motor challenges. And you can study more motor performance in mice by simply putting them on a rotating rod that accelerates the longer the mice are on it. And you can see how long it takes for the mice to fall off. So their time on the rotating rod is an indication of how strong the motor performance is. So you can see in wild-type mice, they're observed -- they are very good at this rotating rod behavior, and they can stay on for many seconds. The Angelman model mice fall off much more readily. After the Angelman syndrome model mice were given the gene therapy, and I should also add after they trained on the rotarod behavior for someday, they were able to perform at wild-type levels. So this is really exciting for us, it showed that with the gene therapy, Angelman syndrome model mice can achieve a performance level that's on par with wild type mice. Next slide, please. Another phenotype that we observe in mice that's very devastating and common in Angelman syndrome is epilepsy. So individuals with Angelman syndrome are much more likely to have seizures. One of the features of seizures is that seizures beget seizures. So once you start having seizures, you become more likely to have seizures in the future. This is a phenomenon known as epileptogenesis. And it turns out that Angelman syndrome model mice have a very strong phenotype and epileptogenesis. And the way we study this is we take mice, and we put them in a chamber and then we give them an inhaled [indiscernible] . This is kind of indicated by this purple gas we have on the image on the left. We then getting this gas called flurothyl which is a GABA-A antagonist, and this will induce seizures in the mice. Now on the right, you see our experimental design where we give the mice flurothyl seizures for 8 days in a row, 1 a day for 8 days, then we wait 28 days. And at day 36, we rechallenged them. So normally, mice will become a little bit progressively more susceptible to seizures. Angelman syndrome model mice become particularly susceptible for seizures after this type of protocol. Next slide, please. And what we saw on the retest day, there is a huge difference in the time it took to have seizures between wild-type and Angelman syndrome model mice. So the Angelman syndrome model mice were much more susceptible to seizures. They had them much faster, much more readily. But Angelman syndrome model mice that were delivered to gene therapy, they did not exhibit this epileptogenic phenotype. So they're exhibiting seizures at the same level in the same way which wild-type were. So this is very exciting because it shows our UBE3A gene transfer rescues this very profound seizure phenotype in the Angelman model mice. And hopefully, we would see something very similar in the human population. Next slide, please. Now there are a lot of histological deficits that can be observed in the Angelman syndrome model mice and none is really more profound, at least none that we've come across yet, then the seizure -- the deficits we observed after inducing seizures in the Angelman syndrome model mice. So what you're seeing here, and I'll walk you through it, on the top, our [indiscernible] of the hippocampus, a wild-type mouse that's been given in vehicle. And NeuN on the left is testing for -- it's just a neuronal marker. Next is UBE3A levels in green, which UBE3A there because it's a wild-type mouse. And then next is WFA, which is a staining for something known as parieronal nets, which is just basically an extracellular matrix staying and then EMERGE all the way to the right. Obviously, in the wild-type mice treated with the gene therapy, we really don't see any differences. And I would say all these images are taken after the mice have gone through this epileptogenic kindling protocol. Next slide. However, if you look at Angelman syndrome model mice the are only treated with vehicle, you can, of course, see the absence of UV in green down at the bottom, but you also see this incredible phenotype of these huge accumulation of what's known as perineural net, it's -- extracellular matrices. And it's just night and day. So just with your eye, you can readily tell this gigantic genotype and the venders of the hippocampus is just right with the expression of perioral nets and its extracellular matrix complex. However, next slide. If you look at the Angelman syndrome model mice, with our gene therapy, not only can you appreciate the reinstatement of UBE3A expression, but you can also see that completely aggregates the net phenotype. So this is a dramatic recovery of a dramatic anatomical marker that occurs in epileptogenesis. Next slide. So you can just see the quantification of these studies here, where wild-type mice in gray that were treated with vehicle or wild-type mice that were treated with the gene therapy, shown in the black bars, show very little accumulation of the [indiscernible] nets. Angelman syndrome model mice that have just treated with vehicle, so a huge accumulation of payroll nets and in our mice were given the gene therapy, the Angelman syndrome mice that were gene therapy. We don't see this the accumulation of [indiscernible] . So not only are we recovering epileptogenic phenotype and preventing that epileptogenic gene therapy, but we're also preventing this anatomical appearance of net in our gene therapy treated model mice. Next slide. So just to conclude, what we show is that our human codon-optimized UBE3A really recapitulates isoform expression and this endogenous short-to-long ratio. We can recapitulate the subcellular localization of UBE3A so that early on in development, there's a relatively cytoplastic and increasingly more nuclear as we get later in development. We are able to recover really severe anatomical phenotypes such as that peroneal net phenotype as well as recover behavioral phenotypes, including ones that are very relevant to Angelman syndrome individuals that include seizures and motor performance. So this provides a very strong content that UBE3A gene addition therapy could provide a treatment Angelman syndrome. And I always like to provide note of caution that we still need to perform additional preclinical optimization, safety and efficacy studies what indications are that this is working really well and perhaps even better than our streams. Next slide. And I, of course, like to thank all the people that contributed to this work. I like to bold players that are either collaborators or lab members who did a particularly large amount of work for this project. Matt Jensen, who is joining us for the Q&A, is really the driver of this project. We also had, of course, very important collaborations with Steve Gray in his lab on viral vector design and Ype Elgersma and his lab for very important contributions for isoform-specific attributes of this UBE3A expression. I want to thank the Angelman Syndrome Clinic at UNC and Heather Hazlett, Mark Shen, I didn't talk about this work. We've been doing a lot of studies on biomarkers that can be used for eventual vector clinical trials. And also, of course, thanks funding source, really, the project was first funded and largely funded through funds from the Angelman Syndrome Foundation. I'd really like to thank them for their contributions to this -- into this project. So thank you for your time and attention. I will turn it over to Kim and ask Matt Judson to join us to help answer the questions.

Kim, before we get started, we're getting a lot of questions for this exciting work. And what we're going to have to do is just due to time limit the amount of questions. So we'll probably cap questions at the current questions that we have in hand. Ben, let us know how you and Matt are doing on time as we go through the questions. But quite exciting work. Obviously, people are quite interested in it. And so we'll do our best to get to everybody that currently has questions and while still remaining on time. So I just want to kind of caveat that for people. So for our analyst community and the investor community on the line, please, if you guys have any additional questions that we're not able to get to them, just send them directly to Kim and I, and we'll make sure that we're able to handle them with Ben and Matt.

Thanks, RA. and thanks, Dr. Philpot for the wonderful discussion. So let's start the Q&A session now. Your first question comes from Yun Yang of Jefferies. Given that there is no sign of Angelman syndrome at birth, how much efforts are underway for prenatal testing?

Yes. This is a great question. So since I've been in this field, which is almost 15 years now, the average age of diagnosis for Angelman syndrome has only been dropping, it's been dropping pretty precipitously. So it was around 2.5 years of age when I first started in this business, and of course, most of those individuals went undiagnosed and now it's around a year of age and dropping rapidly. There's a lot of effort in general to improve early diagnosis and a lot of individuals are working on very rapid tests that can be used for this. There are also groups that are working on noninvasive prenatal testing. In fact, there's some very exciting work from [indiscernible] Group, in which they found an noninvasive test to do prenatal testing. So I think every indications are that we're going to be looking early and earlier. And just to reiterate something that RA has mentioned, while we think -- feel generally thinks that the earlier we intervene the better, we also think that there's good reason to have lifelong benefit with intervention for Angelman syndrome. And a lot of that work comes from critical period studies done by [indiscernible] in which they reinstated genetically different time points during development in Angelman syndrome model mice. You get the most therapeutic benefit when you're reinstating EBT expression early in light. But even in adulthood, you can recover features such as long-term potentiation to a form of synaptic plasticity. And it's thought that LTP is an important, very modality and some serve learning and memory. So if you can recover long-term plasticity in an adult Angelman syndrome model mice, this bodes well, perhaps for cognitive recovery.

Great. And speaking of elasticity, your next question comes from Gil Blum of Needham & Company. Does plasticity change with age? And is there a thinking on the sort of support of therapy that would be required for reactivating or training the formation of new synapses?

Yes. I think that's a great question. So actually, the forms of synaptic plasticity are well known to change during age. So early on in development, the can use certain modalities of plasticity mechanisms of synaptic plasticity class didn't ship later in development. So this includes beautiful work for Rob. And even there's work that into essence, forms a plastic content yet again. I won't go to the specifics of what's observed those. But suffice it to say that we think that an important complement of UBE3A can be important for contributing to the different forms of plasticity and as they shift.

Great. Your next question comes from Yun Zhong of BTIG. Why is there no compensational expression of the paternal allele when the maternal allele is mutated?

I'm not sure I completely understand the question, but I'll try to answer what I think is being asked. So if there's a mutation of maternal UBE3A allele, the paternal allele remains epigenetically silenced in this long noncoding antisense. So there's not a way to have compensation in neurons of UBE3A expression. So the only way to be able to get expression on that paternal allele is to somehow either disrupt the antisense or to promote UBE3A promoter expression so they can override the antisense expression. So there's no compensatory mechanism to have paternal expression UBE3A compensate for loss of the material allele. So hopefully, that answers the question? If not, I'm happy to have it rephrased and I can answer it again.

Great. Thank you. And your next question comes from Michael of Morgan Stanley. What percentage of Angelman patients have mutations in their paternal gene?

Yes. It's roughly around 10% of individuals have mutation -- in the paternal gene?

Paternal, yes.

As far as I know, there's not any evidence. And Matt, correct me if wrong, but I'm not aware of any evidence for individuals having paternal deletions in addition to maternal deletions. So it's not that radiation of the maternal allele really is a promising approach, and that's something that we and others are pursuing in parallel to addition challenge.

Great. Thank you. Your next question comes from Joon Lee of Truist Securities. Given mosaicism in Angelman syndrome, could heterogeneous transgene delivery lead to some cells having too much UBE3A transgene and what would be the consequence of that?

Yes. So obviously, a very important consideration is that you don't want overexpression of UBE3A. And I think something that was really encouraging to us, at least when we had very successful delivery -- viral delivery of the vectors, intracerebral ventrically, the mean UBE3A expression in our virally delivered payload was very similar to the mean UBE3A expression that we observed in wild type or neurotypical mice. So we think we're really hitting the right levels. Now if you look carefully at our paper, you'll see that we did observe overexpression of UBE3A in the hippocampus, particularly in the dentate gyrus. This was largely due to missed injections -- at least partially due to missed injections, where we're going right into the parenchyma, which is something that we won't have to really deal with in the patient population because it's much easier to deliver the viral payload. And I don't know if Matt wanted to add anything to that as well.

No, I think that pretty much covers it in terms of delivery with respect to patients that would have mosaicism. Yes, they would be a little bit more susceptible to having cells that might overexpress the protein. And so overexpression is something we've been very sensitive to. But so far, our studies haven't shown any -- what modest overexpression we have seen. As Ben mentioned, the hippocampus has not proven to correlate with phenotypic recovery or lack thereof in so far as we've measured it. But it is something to be aware of.

Sorry, yes. Now I understand the question a little bit. They're asking if mosaic individuals who were treated, whether we'd have to worry about overexpressing those cells. And yes, it's something that we're very cognizant of, and we have a lot of studies to ongoing for us to understand the tolerance of UBE3A overexpression. But we have every reason to believe that we're in the right zone.

Great. And again, another question on expression level from Kevin DeGeeter at Oppenheimer. How would you address the balance of better biodistribution in the brain versus potentially overexpressing in certain neurons?

Yes. I mean I think, really, what we'd like to achieve is a very broad biodistribution of UBE3A. And I think with a single delivery of the payload that once again, we're -- our mean UBE3A expression levels are right where we want to have of in terms of the mean expression levels of what we're seeing neurotypically. So we think we're getting the right levels. And once again, I will let Matt elaborate.

Yes. And I think what Dr. Brent elaborated upon earlier during her excellent presentation is that we have some -- a lot of inspiration from the mosaic population that you don't need to get 100% reinstatement of UBE3A levels in 100% of neurons to have been some really meaningful therapeutic efficacy. So yes, if you're talking about the cost benefit of achieving perfect biodistribution versus overexpression, I think there's a very broad parameter space that we have to play with there.

Great. Thank you for that. And his follow-up question is for patients with UBE3A mutation that may remain partial protein function, would the dose level need to be adjusted to reflect the total UBE3A protein level after gene therapy?

Yes. That's a really good question. I think it's a good example of where individualized therapy might come into play because we have the ability to study individual mutations and understand whether they're ligase dead, whether they're just a change in the production of UBE3A protein, and then that might offer a way to tweak the gene therapy to best meet the needs of these individuals. But there are groups that include Jason Yi at Washington University St. Louis, Ype Elgersma, Mark Zylka, my colleague here, who are really studying what different mutations do to UBE3A. And these can be studied very well, both in vitro and in animal models. So it presents us with an opportunity for individualized therapy. So in individuals where there's absolutely no evidence of function of UBE3A or no UBE3A is produced, we could be more aggressive with the gene therapy. Whereas individuals where we see a partial expression of UBE3A function, that would lead to a dialing down. But of course, that's an eventual down-the-road goal. But once again, I think it's an opportunity for individualized therapy to intersect with a more generalized therapy.

Great. Your next question comes from Sami Corwin of William Blair. What would be the resulting phenotypes that either isoform was expressed at super physiological levels or if they were not expressed at the endogenous ratio?

Yes. I think the short answer to that is we don't know. The long answer to that is that -- once again, we're doing a number of studies to understand the tolerance of UBE3A overexpression. And all of our studies indicate that there's actually a pretty high tolerance of the system to UBE3A overexpression. But even with that, we're really not observing overexpression of UBE3A broadly as we expressed it. So the question about the different isoforms is incredibly important. And we have 2 clues as to their importance. One clue is this human clinical case study where these individuals only express the long isoform, but not the short. And they have Angelman syndrome, but it's reduced phenotype. So all that tells us is that the short is important, but it's also important to express the long. And then we have some really nice mouse studies with Ype Elgersma in which he's made mice that have deleted either the short or the long isoform. And what he's found is that expression of short isoform is particularly important for the phenotypes in agents with single model mice. But this is also tempered by the fact that 80% of the protein is made of the short isoform. So it's not clear whether it's just because he's expressing 80% of the protein when you have the short isoform and only 20% of the protein you only have a long. Or if it's because it's short is really much more important. So there are studies that are ongoing that are going to address this definitively. But we think -- I guess I would put it this way. We know that the long isoform plays a role from the human clinical studies. We also know that it's evolutionarily conserved for a long time, suggesting that there's an importance of both isoforms. And finally, I would just say, if you had a child with Angelman syndrome, would you want them to get a viral vector that only expressed one of the isoforms? Or would you want them to get both of the isoforms expressed at the right ratio? And I think the answer is that you want them to get both isoforms expressed at the right ratio.

Yes. Great. Thank you. Your next question comes from Yun Zhong of BTIG. Is the UBE3A protein from transgene also sufficiently distributed in the cerebellum, which should be important for motor function? The staining data show much stronger signals in the cerebral cortex.

Yes, I'll let Matt answer that. Go ahead, Matt.

Yes. So it's a good question. So there are 2 points I'd like to touch on with my answer. The first is the data actually regarding cerebellar involvement in movement disorders, at least as far as they've been studied in the Angelman syndrome mouse model, are actually pretty interesting. There's not a lot of involvement for Purkinje cells, this is relevant work from Ype Elgersma's lab using conditional genetic approaches. Whereas there seems to be, at least for the stimulator ocular motor reflexes and some kind of subtle movement control, a principal role for UBE3A in Golgi interneurons. Now it happens that with ICV delivery of our viral vector at this age, we are right now neglecting Purkinje cell neurons, but this could be addressed with intravenous approaches or maybe ICM approaches. But we are actually getting really good reinstatement in Golgi interneurons. So perhaps some of the really nice motor recovery we're seeing is due to reinstatement in that cerebellar subclass of interneurons. So yes, just to recap, I think we've got, again, a lot of parameters space to play with, with route of delivery and dosing, and some of the baseline motor recovery that maybe where there's room for improvement, I think we've got a lot of parameter space to address to achieve that improvement. But so far, things look optimistic.

Thank you. Your next question comes from of Jefferies. Between gene replacement versus paternal gene activation, do you have a view on which would potentially be a better approach for outcomes? And is there a potential for combining the 2 approaches?

Yes. I don't really view it as something that could be combined. I think they both have advantages and disadvantages. I think an optimal [ honesty ] approach of paternal UBE3A has a lot of advantages because you're even more safeguarded against overexpression. I think the path towards just gene replacement is simpler and more straightforward. So I think this is really why the reason that Taysha is pursuing both in parallel because ultimately, we don't know which will be the faster path and which will be better, but we have reason to believe that both are really, really good approaches. So I like the fact that Taysha has invested in both these technologies.

Yes. I would just piggyback on what Dr. Philpot mentioned. This is exactly why we're investing in both. We see an understanding of classic gene replacement as a kind of clear pathway. We've seen it work in multiple therapeutic areas. We've seen it work in multiple indications in the CNS. We've seen it work in our own hands. And so that's -- and obviously, this exciting work that was generated by Dr. Philpot's lab really kind of reinforces that. I also see the importance of the UBE3A-ATS targeting approach because we've seen the clinical data from an ASO. And we know just from an ASO perspective, we can improve on that by just having a simpler dosing regimen, just onetime dose, more stable expression and a little bit more potent delivery because that's just what you're going to get from an AAV. So for us, we're going to develop both, and we hope all of them work, to be quite honest. And then we have 2 therapeutic operates that we can let patients decide. Ultimately, that's the goal, is to allow patients to have maximum amount of choices and to give clinicians like Dr. Goodspeed, Dr. Philpot, the opportunity to choose what's best for that particular family.

Thank you for that. Your next question comes from Salveen Richter of Goldman Sachs. What do you foresee as the greatest risk in translating from preclinical into human?

Yes. I think every gene therapy has risk. And so they're -- it's kind of well known what the main -- what main risks are outside of the viral vector payload. And those risks are largely innate and acquired immunogenicity and other things like that. So I don't think I'm going to address those. Outside of that, the risk is what we've talked about is making sure that there's not an overexpression of UBE3A. So to the extent that, that would be detrimental. But we've done a lot of work to try to ensure that we're not getting that and to also understand the safety aspects of having overexpression, kind of how tolerant the system is to having an overexpression. So all I can say is we have every reason to believe that we're right in the right area of expression levels. But that's something that's always going to be on our mind. And always thinking about the safety issues are very, very important.

And Ben -- I'll just add to what Ben mentioned. For me and for the team at Taysha, I think it was really important and what got us excited about this work, to be quite honest, was really the elegant payload design. And this is something that I think, again, just speaks to just the work that's gone into designing this vector from Dr. Philpot and Dr. Gray, was really trying to recapitulate the ratio both from a protein perspective but also an mRNA perspective of this endogenous ratio of short to long isoforms. And this is, to my understanding, the first vector and the only vector that has been able to successfully recapitulate that just from a payload design perspective. And so now I think to what Matt was talking about is really our -- so now we know we have the right payload design. So now it just really boils down to delivery, to be quite honest, and making sure that we're getting broad distribution. And we already know the distribution pattern, the biodistribution patterns of AAV9. But in some cases, you may want to do dual route of administration. Maybe it's intrathecal delivery with an ICM or an ICV delivery or maybe it's just classic intrathecal delivery, you're able to get enough expression. So as we go throughout the pharmacology work, we're going to be looking at that. And as we do our toxicology work leading to the IND, we'll be looking at that. But certainly, what got us excited to be quite honest, is this novel payload design and the fact of the matter that we're expressing both the short and long isoforms in the endogenous ratio, that's really groundbreaking.

Thank you, RA. Unfortunately, we're out of time now for our Q&A session. But thank you, Dr. Philpot and Dr. Judson for your incredible insights and for your time here with us today. And with that, I'd now like to introduce to you Dr. Ryan Butler. He's going to be talking about the RNAi gene therapy for Angelman syndrome. Dr. Butler?

Thank you very much for having me this morning. Next slide. Before I begin, just a few disclosures. I have a personal financial interest in Taysha through licensing and revenue sharing. UT Southwestern maintains a financial interest in Taysha due to Taysha's exclusive licensure of technology developed at UT Southwestern. Taysha has a sponsored research agreement for this research, and I am not being paid for this presentation. Next slide. So the previous talks, brilliant talks have gone into the details of the genetics of Angelman syndrome. I will be talking a bit about that as it relates to my program for the RNAi treatment. Go ahead. Next slide. So as was mentioned earlier, there is currently no cure for Angelman syndrome. Current treatment focuses on managing the medical and developmental issues. This includes antiseizure medication as well as therapies -- physical therapies and communication behavior therapies. Ultragenyx is currently undergoing Phase I/II clinical trials, which are showing promising effects. I want to point out that the approach that I'm taking here is scientifically very similar to the ASO approach with the added benefit that in theory, it should be a onetime injection, which could last for the life of a patient. Next slide. So this has gone over before. In neurons, the maternally inherited UBE3A allele is the only active allele. Paternal UBE3A is silenced through cell type specific imprinting. This relates to what I'm about to describe is that the -- in theory, the RNAi approach, the shRNA approach, could treat all of these different subtypes of Angelman syndrome so long as the paternal allele has a functional copy that's just silenced. Next slide. So to get a little bit more of the nuts and bolts of how this imprinting works. So there's an antisense DNA strand that is transcribed. It's a large transcript called a large non-coding antisense transcript. Most of it is for small nuclear RNAs. I won't go into that too much, but I will touch on it a little bit later. But at the end is the UBE3A-ATS or antisense UBE3A, and this is on the paternal allele. Now what's very attractive to UBE3A-ATS from a therapeutic perspective is there's really no functional role for it other than to inhibit paternal expression of UBE3A. And so you inhibit that, you should induce paternal expression of UBE3A. Next slide. So our strategy here is to reinstate paternal expression of UBE3A, and that we are going to be using a short hairpin RNA approach to target in the -- target a region between UBE3A-ATS as well as incorporating some of the SNORD regions. And this is based on prior studies, which had in a mouse model by the Bode Lab as well as Mark Zylka's lab and others as well, which have shown that this region can induce paternal expression of UBE3A. Go ahead, next slide. And so like I mentioned before, we have our potential target region, which incorporates the SNORD115 region through the region between UBE3A-ATS and that's based on prior studies. We also are cognizant of the other areas of the long noncoding RNA transcript, which is our critical regions for Prader-Willi syndrome. And so we want to effectively target the potential target region without compromising the regions which are critical for Prader-Willi syndrome. Go ahead. Next slide. So as we are -- there are really -- as we're trying to develop our most efficacious treatment using the short hairpin RNA approach, we are very aware that there are side effects, which are specific to RNA interference, one of which is off-target effects, which I will talk about a little bit later. Another one is shRNA-induced cardiotoxicity. And this is induced through competition with endogenous microRNAs. And this was done -- this was performed by work by Beverly Davidson's group. And basically, what they showed is that it's tolerated -- the short hairpin RNAs, effectively, short hairpin RNAs are tolerated in the liver. However, there are toxic effects in the heart. Go ahead. Go ahead, next slide. However, there have been studies -- follow-up studies to that, which have shown ways to mitigate this effect without compromising efficacy. And this is the approach we're taking to the construct design. And so if you encapsulate the short hairpin RNA in the construct with a microRNA scaffold, what you get -- specifically in this one, miR-33 scaffold, you can reduce the cardiotoxic effects without compromising efficacy. And so this is the approach we are taking to the design of our construct. Next slide. And so now going back to the other RNA interfering specific toxic effects that could occur, and that is the nonspecific binding. And so our shRNA is roughly 21 nucleotides or exactly 21 nucleotides. However, as few as 6 nucleotides can induce some level of off-target binding. And so to reduce this, we initially do a broad bioinformatic screen, and this actually produced roughly -- produced thousands of potential candidates, which could bind to the therapeutic target region, as I will call it. And so from these thousands of candidates, we did a second bioinformatic screen, which is to determine the probability of off-target binding. And so this is essentially a bioinformatics program, which will calculate other transcripts, which have similar homology or similar complementary targets to this shRNA. And so we have a threshold. And if it goes -- if it's beyond that threshold, it is no longer a candidate. And then step 3 here. We also are testing this in the monkey background. So our study here will require safety in toxicology in rhesus macaque or in a monkey and NHP model, if you will. And so we have to make sure that it actually is an appropriate sequence for binding in the monkey background as well as the human background. Go ahead, next slide. And so our strategy here, we have -- first, we start off with our in vitro study. So from these thousands of candidates which we initially had, our secondary bioinformatic screen looking for off-target effects was reduced down to about 40 candidates. And this is specifically for human-specific siRNAs. And so we will be testing these human candidates in induced pluripotent stem cells. And this is actually something from the Foundation Prangemen Syndrome Therapeutics, which Allyson has led. We received these induced pluripotent stem cells from Yale University, and we will be differentiating them into iPSC-derived neurons in testing our -- with our siRNA candidates as well as our plasmic construct on the Angelman syndrome-derived neurons. Very similarly, in a mouse model, there is a reporter line, which has a yellow fluorescent protein tagged to the paternal UBE3A. And so in this instance, if we are able to induce paternal expression, we would actually see glowing yellow neurons in culture. And so this compares with the Angelman syndrome iPSC-derived neurons. If we are able to induce paternal expression then in a neuron, which would normally have no UBE3A, we would actually see UBE3A. And so there's face validity in the in vitro model. Go ahead, next slide. And then moving on to in vivo. So like I mentioned before, we will have our human-specific -- monkey human-specific candidates within the miR-33 scaffold. We will take these and produce our AAV9 and we will do a safety test in nonhuman primates. For the mouse model, we -- as mentioned previously, there is a mouse model for Angelman syndrome. And we will be delivering this -- we have been delivering our AAV shRNA with an intrathecal injection to our Angelman syndrome mouse model. And among other things, after we do our behavior tests, we will also be checking for expression of paternal UBE3A and other nontarget -- off-target effects, which we want to make sure are not being affected. Go ahead, next slide. And so our behavior battery after -- sorry, intrathecal injection at day 7 to 10 after birth. That's really the earliest week in -- our lab can comfortably give an intrathecal injection to a mouse. And then 8 months and 12 months -- sorry, 8 weeks, 12 weeks after the intrathecal injection, we subject the mice to behavior battery, very similar to what Allyson and Ben have described previously. And there is some -- there is significant face validity between the mouse model of behaviors and the human condition. We also have the option of later on measuring EEGs in the rodent as well. Go ahead. Next slide. And so this is really the big results in the slide that I have here. And so this was our -- we had our 40 candidate -- roughly 40 siRNA candidates, human-specific siRNA candidates. And we did an initial in vitro screen in neuroblastoma cell line. And what we're looking for here ideally would be a decrease in the target, which is UBE3A-ATS, and a concomitant increase in UBE3A. We're more focused at this point on the reduction in UBE3A-ATS simply because the -- when we move on to the iPSC-derived neurons, we'll look a little bit further into the kinetics. Basically, how long it takes for UBE3A to be expressed following inhibition of UBE3A-ATS. So what we're saying here is that we have 7 top candidates, which we will be moving on for an in vitro screen in the iPSC-derived neurons. And from there, we will do all sorts of batteries of tests on those iPSC-derived neurons. Go ahead, next slide. And so this is just pilot preliminary data from our mouse model. And so like I mentioned before, this was an intrathecal injection of the AAV9 shRNA at days -- between day 7 and 10 of the age of the mouse. And so the top left slide is just a GFP tag. And so that what we're showing here is that the virus did reach the cerebellum. And so in the bottom left is the stain for UBE3A. And what we're showing is that there is some co-localization between the GFP-labeled neurons in the cerebellum, specifically the Purkinje cells of cerebellum and UBE3A. So we're seeing -- beginning to see some preliminary evidence of a reinstatement of the paternal allele in the mouse model. And with that, that's really the end of my talk. And so just to summarize, we do seek to utilize the AAV9 technology to stop the stop, similar to the ASO approach being performed by Ultragenyx and other labs as well with the added benefit that the construct will be constitutively active within the cells until we would have a permanent expression of this shRNA. And at the same time, we are trying to maximize efficacy of this shRNA. We are also making all efforts to reduce known side effects of RNA interference, including cardiotoxicity and off-target binding. So go ahead, next slide. And with that, I would like to give a very warm thanks to everybody in my lab. That's the left -- top left panel there, everybody involved with this project. Also collaborators, [ Berg ], Nemera as well as Kimberly who gave her talk earlier. And also Dr. Yong-Hui Jiang, who provided the iPSCs from Yale University. And with that, I will hand it over to Kimberly for any questions you may have.

Great. Thank you, Dr. Butler for your insightful presentation.

We'll start the Q&A session. And unfortunately, we will have time really to address only 1 question. So apologies to all who are asking questions. Dr. Butler, the question comes from Gil Blum of Needham & Company. Will there be safety data in normal mice to see potential overexpression phenotype?

We will be observing them for behavior deficits as well as survival. The big thing is the -- one of the interesting conundrums with this target region is there's not a lot of homology between the mouse and the human. And so the -- we have mouse-specific constructs, and we have human -- monkey human-specific constructs. And so for safety, we will be -- the big safety test we will be doing for -- as it's related to bringing this to the clinic, it will be in nonhuman primates. And that -- yes, we will be doing pilot safety studies in mice, but the big thing will be in the nonhuman primates with the human-specific constructs.

Kim, I'm going to just piggyback off of what Ryan has said. I think what's nice about this particular approach is that we have a pathway to the clinic because there's an ASO that's actually gone ahead of us using the identical approach. And so what we've shown is that if we're able to recapitulate the data in iPSC cells, human-derived -- patient-derived iPSC cells, couple that with safety data in nonhuman primates as well as well-controlled GLP rat tox study, we should be in a good position to meet all of the FDA's current demands of what it would take to take a program like this forward. There's also programs currently in the clinic that are using kind of a vectorized either shRNA, mRNA approach in other CNS diseases. So again, the pathway is relatively clear here. We're quite excited about this because, again, we've already seen some really nice efficacy data from our colleagues using an ASO approach. And so I think really the hardest work that Dr. Butler as well as our team over at UT Southwestern really needed to do was kind of to take this universe of 1,000 potential shRNA candidates and whittle it down to the number that we've gotten here. And so I think right now, we feel in pretty good position to now select our shRNA candidate and then move into those definitive NHP studies in order to enable a clinical trial start here relatively soon. So we feel pretty good about where we are from this program because of the work Dr. Butler has done with kind of taking that universe of thousands of candidates and getting to the number that we've got.

Great. Thank you. Let's take a couple more here. Dina Ramadane of UBS asked why is SNORD115 region the correct target? Are these SNORD regions conserved across preclinical models, mice and NHPs and in humans?

And Kim, just before Ryan answers, there are certain things that we're not going to be able to discuss because of IP issues, and we want to make sure that those things remain trade secret. But with that caveat, Ryan, please feel free.

So do you mind repeating the question? I heard the part about is a conserve between mouse and human and that is an easy answer, and it's very poor homology between mouse and monkey human. There is homology between monkey and human, thankfully. But not between mouse and monkey. What was the other part of the question?

Yes. And is the SNORD115 region the correct target?

So the SNORD115 target is -- as it was shown by Mark Zylka's group, that's actually their target using a CRISPR approach for gene editing. And they did a very elegant trap design and showed that they were able to induce paternal expression of UBE3A. And so that we incorporated the SNORD115 region into our larger therapeutic target region because of that.

Great. Thank you. And Dina has a follow-up question as well. Did you see any off-target effects in preclinical studies?

We are too premature right now to say that. We are still in the early stages of analyzing the behavior data. As of right now, we don't see any unexpected effects. We know what an Angelman syndrome mouse looks like and acts like. We also know what the wild-type mice looks like. And we don't -- haven't seen anything yet, but it's a little bit too early to definitively say that.

Just to add to what Dr. Butler said. Again, I think this is the beauty of being able to go after a vectorized siRNA approach is because you have multiple constructs to choose from, which is nice here. So although we haven't seen anything preclinically, if we did see something, you would just move to the backup construct. So not only are we moving 1 construct forward, we have a number of backup constructs to go with. And again, what we've been able to see from a clinical perspective is that there is a region to target and you could target that successfully without inducing any type of off-target effects. So for us, we feel pretty good about this. Again, it gives us a lot of comfort on why we feel so good about this approach.

Right. Thank you. And in the interest of time, we'll take 1 last question here from Joon Lee of Truist Securities. Is the region you're targeting with your short hairpin RNA in the similar vicinity as the region targeted by GTX-102 and Roche's ASO? And is there a lot of flexibility in UBE3A-ATS targeting? Or is it constrained by a need to suppress all potential isoforms?

I'm going to let [indiscernible] take that.

Yes, before we answer that, Joon, and I appreciate the question. Unfortunately, we're not going to be able to answer that first part. Kim, do you want to repeat the second part? I think we feel comfortable with Dr. Butler answering that.

Yes. Is there a lot of flexibility in UBE3A-ATS targeting? Or is it constrained by the need to suppress all potential isoforms?

What we're seeing with the different isoforms of UBE3A-ATS, at least with our candidates right now, is that it is inhibiting all of them. And so that's good. I'm not -- I know that there's differential expression of UBE3A-ATS during the development of the neuron. And so what we are hoping is that we will be able to target all of them, all of the different isoforms and get them as they arise.

What's interesting about this approach -- and just to kind of bring it home. What's interesting about this approach is -- all you're doing is unsilencing a natural copy of the allele. So not only are you going to be expressing all isoforms in the endogenous ratio because essentially all you're doing is stopping the stop, right? But you're also being able to control for overexpression because there's no ability to overexpress because all you're going to do is maximize the expression, the natural expression of the paternal allele, which would be the natural -- which would be the same as the natural expression of the maternal allele. So again, it's a really interesting approach from a safety and tolerability perspective because you -- essentially, you'd be unable to induce overexpression, at least through the mechanism of targeting the ATS region, because all you would be doing is maximizing the paternal allele. So again, I think this is why we like the approach. And one, we've seen clinical benefit of the approach with our colleagues and with the ASO. And so for us, it's about maximizing that clinical benefit, one-time dosing, more stable expression, more potent expression. And as Dr. Butler mentioned, lifelong expression.

Great. And we are out of time now. Thank you, Dr. Butler, for your insightful presentation and for really bringing home the vectorized knockdown approach.

You're welcome.

Thanks, Ryan.

And with that, I'd now like to turn the call over to Dr. Suyash Prasad, who will close out with the clinical development strategy.

Great. Well, thank you, Kim, and thank you to all our wonderful speakers. It's been a really stunning event. I've learned a lot just listening to the speakers, and I hope you're all as excited as I am about taking this wonderful science. And now what we have to do is take this wonderful science, move it through IND-enabling studies into clinics. And now the key thing that I'm going to spend a few moments to talk about is translating these wonderful scientific ideas and groups of principle into the clinical situation. Next slide, please. So as you've already heard, we -- you've heard a lot about the underlying biology of the disease, the genetics of the disease, the molecular pathology of the disease, the clinical signs and symptoms of the disease and in particular, a lot of the wonderful work that Allyson and the group at [ FAST ] have done in preparing the environment for clinical studies. So you see we have 2 approaches. We have the gene replacement strategy approach that Ben talked about, and we have the knockdown approach, i.e. unsilencing the silenced paternal allele to allow for normalized expression of the UBE3A protein across the range of isoforms as described by my colleague and friend, Ryan Butler. And as RA has already touched on, our intent is to move both programs forward in parallel as rapidly as possible through IND-enabling studies, through regulatory interaction and into clinical development with the intent -- unless there's a major show stopper with either, with the full intent of filing BLAs and commercializing both products so patients have a choice. And as we already talked about, it could well be that certain patients, depending on age and on genotype, may be more amenable to one approach than the other. That remains to be seen. But at the moment, we feel that the vast majority of Angelman patients, both approaches would be highly suitable. Next slide, please. So with regard to the first approach that was discussed, this is the gene replacement strategy approach that Ben described and there's a really nice paper that's published by Ben and Matt. I think it came out on Friday last week and we issued a press release about it on Monday. And I'd urge you all to go and read the paper. Some beautiful images in the paper and much of what was discussed in the paper, Ben talked us through, but it's a really nicely written paper. And this is a construct, as you can see. It's the full length UBE3A human DNA, with a neuron-specific promoter wrapped up in an AAV9 capsid. And one of the key aspects of this particular disease is the fact that this particular approach is that it recapitulates this 3:1 ratio of the isoforms, the 3:1 short to the long isoform, which is the -- which replicates what happens to the human situation. Most likely, it's the most optimal approach. You've also heard from Ryan about -- from Ben about how you actually get a very nice biodistribution. You saw a resolution of functional -- the functional epilepsy phenotype in the mouse model, and associated with that was just a pathological improvement which you saw with the beautiful images to stains that Ben shared earlier. So this approach, we anticipate will offer a rapid onset of UBE3A expression in the appropriate 3:1 isoform balance. This should happen in the vast majority of neurons. And we anticipate that this will be a long last effect and should hopefully lead to a resolution and amelioration of signs and symptoms of Angelman. Next slide, please. So that's the gene replacement therapy approach. So just to touch on once again Ryan's approach the knockdown of the UBE3A antisense transcript, i.e., this is a strip of RNA that silences the paternal gene. And so you're unsilencing the paternal UBE3A. As has already been alluded to, one of the beauties of this approach is that you unsilence the paternal gene, the risk of overexpression is minimal. Well, it's nonexistent in the vast, vast majority of Angelman patients. Because the maximum you can express is 100% levels of UBE3A. So the risk of overexpression, as I say, is inconsequential with this particular approach. Also, I think what's important, and which has already been touched on with some of the questions that we've received, and it's the fact that this essentially mimics the ASO approach. And Ultragenyx shared some really nice data on 5 patients that were dosed with the ASO approach. I know Allyson in the past had a lot of input in the initial work. And then Ultragenyx took over and ran with a clinical study. A very elegantly designed study, 5 patients really showed quite rapid improvement and significant improvements, clinically meaningful to the 5 patients who were dosed with that construct. The drug was given intrathecally and the drug was given on a monthly basis for a number of administrations. So really, really nice efficacy data. You'll all be aware, of course, there were some safety matters with this particular program in that several of the patients developed some significant lower limb weakness. And this was quite significant. It's the point why 3 of the children temporarily lost the ability to walk. This actually resolved very nicely over time. So I think in terms of balance of risk and benefits, that was a really nice -- it showed resolution of a wonderful improvement in clinical [indiscernible]. But there were some adverse events that the might need to be managed. The general thinking as that was related to the intrathecal administration we've done several times and some inflammatory effects in that particular area of the spinal cord. But really nice proof-of-concept data, I think, really lends credence to this short hairpin RNA approach that Ryan has been discussing. Once again, it will be onetime dosing. And we anticipate that there will be widespread transduction through the central nervous system, which is the target of interest in this particular case as you already heard earlier from Kim Goodspeed that that's really only the brain that was affected in that. And so that's because the -- outside the brain, the [indiscernible] doesn't actually work, it's just within the brain where UBE3A is the presented gene site. Next slide, please. So another important part to our approach to -- I think about clinical trials, as you're all aware, across our portfolio programs is the fact that we spend a lot of time talking to patients and families about what's meaningful to them clinically and functionally and from a disease burden perspective. And we do this really to help inform our regulatory interaction and to inform clinical development. Now we do have our workshops planned for the course of -- early in 2022. But actually a tremendous amount of work has been done by the group at FAST already. And so if we go to the next slide, please. And Allyson alluded to this in her presentation where they actually ran a very detailed qualitative study that listed the specific concepts of interest, concepts of note, concepts to cause issues with patients and families to really understand the humanistic burdens as much as possible. And this was mapped out in a really nice disease model. It was published in 2020. You can see the reference to the bottom right on the corner of this particular slide. And you can see that although typically, the disease is described as one where there's a lot of motor dysfunction and behavioral issues. But the thing that's most impactful to patients and families is the decreased speech. So the expressive, expressive language. Receptive language still seems pretty well understood by the children, but the inability to communicate their needs, their wants, their desires, that's particularly troubling to the parents. The seizures, of course, are also debilitating, but some of the behavioral issues are challenging, learning issues are challenging. And of course, there's a whole host of quality of life type matters such as gastric and spinal issues that can cause a whole host of trouble signs and symptoms, in addition to, of course, sleep, movement, et cetera. And it's therefore important for us as we design our clinical trial, as we talk to regulators about the design of the clinical trial, to weave those thoughts into our development plan. And in the next slide, you can see in a little more detail, actually some of these specific features. So I've talked about the expressive communication impairment. And it's why we're including a scale that, once again, was developed by Allyson and her team. Then there's the ORCA scale, which looks specifically at the communication issues in Angelman syndrome. We'll be looking at seizure frequency. We'll be looking at how frequently seizures are happening, how can we or can we not reduce the number of medications that patients are on? What triggers their seizures over time? What results -- what do the seizure pathology results into admission into hospital? And then related to that, of course, we've talked about EEGs and the importance of that neurophysiological assessment as a biomarker, in particular, looking at this concept of delta power and looking at that quantitatively as some of that may or may not improve with treatment. Next slide, please. So as I said, we routinely engaged with patients and caregivers when we try and build on some of the work has already been completed and try and do a bit more of that in the early part of 2022. Next slide, please. So in terms of clinical trials, we will likely be doing 2 clinical trials, 1 for each program. We're still in early stages. We started yet to get through the IND-enabling preclinical work despite having shown really nice proof-of-concept for both approaches. This is for the gene therapy -- gene replacement therapy approach. We'll be giving the drug intrathecally in our usual way. Of course, though, as the IND-enabling studies progress, if it seems more appropriate use in kind of a combined route or an alternative route, we will, of course, bear that in mind. But our expectation is an intrathecal dose of drug will deliver more than enough to the brain and the spinal cord. We'll likely get a starting dose and a higher dose in the dose expansion cohort. We will, of course, be covering with our usual regime of prednisone sirolimus or rapamycin. Next slide, please. And in terms of the studies about the gene -- the vectors RNA strategy, once again, a similar approach, whether the starting dose or a high dose. Because it's once again building with a [indiscernible] in a suppressive regime, and give the drug intrathecally. And we'll likely use similar endpoints across the range of -- across both studies. Next slide, please. As specifically talking about the endpoints, we will have a spread of endpoints in some multisystemic disease. You've already heard the key opinion leader perspective and what's important. You've already heard what the changes we've seen in the animal models. And you've already heard about the information we're getting the patients and families about what's important. But you can see here that we'll be looking at a number of disease-specific assessments in particular with CGI, the clinical load impression of improvement and severity as directed by Angelman syndrome-specific anchors, and this -- there's been some evaluation that's already been performed in this. We'll be looking at development and progression assessments, the babies, the [indiscernible], for example. We'll be looking at sleep. You've already heard that how sleep is very much -- is a real issue with these children, these patients and families. And we will be doing this in a very disciplined manner. Direct assessments, which seem to be a very, very helpful way of -- somewhat positive, but it's a very helpful way of looking at sleep. There are certain questionnaires with the global performance sleep [indiscernible] our baseline on the various samples there as well. There is some burden associated with that, but I think it's an important assessment. And of course, wearables are increasingly being used. I've touched on the seizure assessments already. We will be performing MRI on an ongoing basis from an imaging perspective, looking at the biomarkers. Specifically, [indiscernible] is a logical biomarkers. Now one of the questions I often get asked about this and other [indiscernible], what is the -- is there a serum biomarker? Is there a CSF biomarker? As yet, there is not. The likelihood is we're going to try and see if we can assess UBE3A in the CSF. Thus far, attempts a field have made to try and do that have not been successful. We're going to spend some time trying to do that. If we're able to, that will be great. If we're not going to do that, we will certainly reflect on CSF and perform a whole set of metabolomic and proteomic evaluation to look for explorating CSF-type biomarkers. Then the key biomarker of note at the moment is this neurophysiological EEG delta power one. And then in addition to that, we will do communication and of course, quality of life health utilization-type standards as well. Next slide, please. So in terms of next steps to the program. We will be performing mass pharmacology studies for both programs and expression distribution and safety data on confirmatory NHP studies. The preclinical progress of both will be a little bit different. Obviously, one is a gene replacement therapy, strategy approach. The other is the vector RNA approach. So what we'll look for from a toxicity perspective may be a little bit different. For example, in the case for the gene replacement therapy approach, we will be specifically looking for overexpression. One of the things with the vectorized RNA approach, as you heard from Ryan, is this risk of cardiac toxicity that is somewhat theoretical, but has been seen previously. So we'll be looking -- so the R&D enabled preclinical studies will have some differences between them. We're working on the international trial protocol. As you probably heard, we're already giving some thought endpoints in the study design. We're speaking with key opinion leads. We're running a couple of scientific advisory boards. One is at the end of this year, there'll be another one early next year. We have our patient focus groups in plan to gather some additional insight there during the early course of 2022. And of course, we'll be starting our regulatory interactions. I anticipate a nice cadence of regulatory interactions with the FDA and other non-U.S. agencies during the course of 2022. So on that note, I'd like to thank all my colleagues and speakers who've made this a very interesting, successful meeting. And I'm happy to take any questions as we close out this meeting. Kim?

Thank you, Suyash. Unfortunately, we are running behind and unfortunately, we can't -- we won't be taking any questions at this time. But thank you for your presentation, Suyash. I'd like to now turn the call back to RA Session for his closing remarks. RA?

Thanks, Kim. Next slide. We've really enjoyed sharing with you guys greater detail of our Angelman programs. Looking ahead, we will continue to focus on rapidly advancing our pipeline with many key milestones coming before the end of this year. Of note, we plan to release high dose data which is the 3.5x10 to the 14 total vg data from that cohort in our GAN study by the end of this year. We'll have preliminary safety and biomarker data from our Phase I/II GM2 gangliosidosis trial. We'll have that data before the end of this year. And preliminary CLN7 data, both safety and potentially some preliminary efficacy data from the first-generation construct by the end of this year. But we expect to have in total 5 open INDs or CTAs including for our Rett syndrome program by the end of this year, which we're quite excited about. Next slide. Again, we'd like to give a special thanks to all of our speakers today for their great discussions, research and, again, to all their respective labs as well. We appreciate everyone on the call taking the time out of their day to spend it with us to learn more about our approaches to Angelman syndrome. Again, we'd like to thank our Chief Scientific Advisor, Dr. Steven Gray, and his lab for the work that they've done, helping to enable both of these programs as well, to our partners at FAST and the Angelman Syndrome Foundation as well as to our partners at UNC, Dr. Philpot and his lab as well as Dr. Butler and his lab. Next slide. Thank you, everybody, for joining us, and we hope that you have a good rest of the week.