uniQure N.V. (QURE) Earnings Call Transcript & Summary

Good morning, everyone, and welcome to uniQure's Virtual Research and Development Day. I'm Maria Cantor, Chief Communications Officer at uniQure. We appreciate you taking time to be with us today. Before we get started, please note that we'll be making a number of forward-looking statements this morning. We suggest that you take a moment to review this slide which contains our forward-looking statements. These statements involve risks and uncertainties, many of which are beyond uniQure's control, and actual results could differ materially from those that are in these statements. For a detailed description of these risks and uncertainties, we encourage you to review the company's most recent Form 10-Q filed with the Securities and Exchange Commission as well as the company's other SEC filings. Regarding our agenda this morning, we will begin by reflecting on uniQure's long history of leadership in gene therapy and presenting our vision and strategy for the company over the next several years. We'll provide a program update on AMT-130 in Huntington's disease and then have a 15-minute Q&A session. We'll follow this with a short 10-minute break and return for presentation on the expansion of our research pipeline. We'll have another 15-minute Q&A session and transition to an update on our hemophilia B program. After this presentation, we'll have a Q&A session before concluding remarks. For our research analysts, if you would like to submit a question, please use the Ask a Question box. We will answer as many questions as time allows. We also note that biographies on each of our presenters are available for your convenience. Now it's my pleasure to introduce Matt Kapusta, our CEO.

Thank you, Maria. And on behalf of the uniQure team, I'd like to thank you all for joining us here at uniQure's 2021 Virtual R&D Day. To start our presentation, I'd like to go back to the beginning, uniQure's mission and purpose. 23 years ago in 1998, uniQure was founded with a singular mission of delivering curative, onetime-administered genomic medicines with the potential to transform patients' lives. We're not seeking to modify symptoms and we're not looking for marginal improvements. Our quest is to reimagine health care and to truly transform lives by harnessing the power of genetics. While a lot has changed during the 2 decades since the founding of uniQure, our mission has remained constant. And in the nearly 7 years I've been at uniQure, I've truly never been more excited than I am now about our future and our potential to deliver on our mission. During the next 3 hours, we'll be covering a lot of ground with a focus on defining uniQure's strategy for transforming patients' lives as well as for building value for our shareholders. This strategy includes not only the reimagination of our pipeline after our recently closed transaction with CSL Behring, but also the reimagine how and where we deliver genomic material, what genomic cargo can be delivered and how gene therapies can be manufactured in a reliable and cost-effective manner. One of the blessings at uniQure is that we have developed over many years a highly valuable and unique platform that serves as a powerful engine for developing transformative medicines. This engine has very broad applicability, and as such, we have defined core areas on which we're focused, including CNS diseases, liver-directed disorders and cardiovascular, muscle diseases. In each of these areas, we have dedicated years of painstaking research to understand the clinical unmet needs of patients suffering from these disorders and how we can optimize our gene therapies to address their specific diseases. Core to our strategy is to continue to leverage this powerful AAV engine and the technology platforms that we've established to develop innovative genomic medicines within these key focus areas. With hundreds of gene therapy companies and more popping up seemingly every day, I'm often asked, what makes uniQure different? The answer is our people and our heritage of innovation and ingenuity. Innovation and ingenuity are not just words at uniQure, they are imprinted in the DNA of every single one of our employees. For more than 2 decades, uniQure has been demonstrating Ingenuity as a pioneer in the field of gene therapy, helping to usher in a renaissance in the field. We were the first company in the world to achieve regulatory approval of an AAV gene therapy. We were the first to have a commercially licensed AAV manufacturing facility. Our Huntington's program was the first onetime-administer gene therapy to enter into clinical testing. And we are the first company to clinically demonstrate, with our AAV5 gene therapy, the ability to achieve positive clinical outcomes in patients with preexisting neutralizing antibodies. uniQure's capabilities are also unparalleled within the field. For over 15 years, we've been developing and optimizing our ability to produce CGMP-grade gene therapies at commercial scale. This includes gene therapies that have been used across 8 different clinical studies in more than 100 patients. Many of our talented scientists and techs have been at uniQure for 10 to 15 years or more and have helped us establish strong IP and know-how, covering our manufacturing processes and methods as well as novel tools and enabling technologies. Nowhere was this ingenuity more evident than our recent successes within hemophilia B, a program that, as recently as 2017, was written off as dead. However, the uniQure team never gave up and truly believed in AAV5 gene therapy to be best in class. So in 2017, we endeavored to engineer a next-gen product combining a variant transgene with the virtues of AAV5, and we were able to convince regulators that it was highly comparable, allowing us to move directly into a pivotal study, something that had never been accomplished previously. This reengineering simply would not have been possible without our strength in product development and manufacturing. Within 8 months of unveiling the program in 2017, we initiated 2 clinical studies, including our HOPE-B pivotal trial. And today, we are announcing 1-year follow-up data on all 54 patients in the study. Perhaps most importantly, earlier this quarter, we announced the closing of a $2 billion license and commercialization agreement with CSL Behring, a global leader in hematology with commercial operations in more than 80 countries around the world. This transaction represents one of the largest single-product deals in the space to date with an upfront payment greater than uniQure's entire market cap back in the days of 2017. Our Huntington's program is also emblematic of uniQure's innovative approach. We spent more than 6 years designing and developing miQURE, our proprietary gene-silencing technology. And with the first AAV gene therapy to enter clinical testing for Huntington's, a disease which affects up to 100,000 people in the U.S. and EU alone. We also incorporated an innovative Phase I/II trial protocol, including a double-blinded sham control design, which was strongly endorsed by the FDA and has the potential to expedite our path to registration. We expect to have initial safety and biomarker data from this U.S. study later this year which, combined with the data from our EU Phase I/II study, is expected to generate multiple additional data readouts over the next 2 years. No doubt, while we're very proud of our accomplishments to date with hemophilia B in Huntington's disease, what I'm really most excited about is the reimagination of our future pipeline. With the CSL deal behind us, we now believe we have the financial resources to aggressively expand our pipeline. Specifically today, we have approximately $700 million of cash on hand that we expect will be sufficient to fund operations into the back half of 2024. We're also eligible to receive more than $300 million in near-term, higher-probability milestones related to regulatory submissions and first commercial sales, which would extend the cash runway through 2025. And finally, we're eligible to receive royalties up to the low 20% of net sales and another $1.3 billion in additional commercial milestones that have the potential to extend our cash runway beyond 5 years. Leveraging this strong financial position, we are highly focused on expanding our pipeline and in particular transforming our portfolio of clinical-stage programs. Today, we have 2 exciting programs in the clinic that we believe can continue to drive value for patients and shareholders. But by 2026, our goal is to have 8 to 12 clinical commercial programs, plus a roster of next-gen preclinical programs. We recognize this is really ambitious, but believe it's achievable given the strength of our platform, our people and financial resources. We also recognize every project may not succeed, but we are taking strides to improve our R&D efficiency and effectiveness by doing 2 things: One, working to reduce our time lines to the clinic by leveraging a more modular product development process and platform; and two, selecting future programs that have balanced risk/reward profiles. In terms of balancing risk and reward, there are several criteria that we use at uniQure. First and perhaps most importantly, we want each pipeline candidate to have a line of sight to being best and/or first in class. If we can't envision a path to best and/or first in class, we won't waste time and money. Or if the data generated doesn't support this thesis or if the competitive landscape changes, we will have the courage to cut bait or find a different approach. Our past decisions around hemophilia A and B are examples of this. Second, we want each of our target staff human validation. We're not in the target discovery business, and it's important that biology we target is well characterized. Third, where at all possible, we want to be able to leverage validated platforms to derisk our programs. And this is why AAV5 has been such a workhorse for uniQure. And lastly, we want to have greater emphasis on larger indications. Not only can we maximize our impact on the greatest number of patients, but we can also maximize the commercial opportunity of our successes. As I mentioned, enabling technology platforms are an essential part of our strategy. We think of these technologies in 2 broad categories, with the first being how and where we deliver DNA material. When I first joined uniQure, the focus of the gene therapy field was largely on first-generation AAVs with relatively crude administration techniques. Today at uniQure, we've established and continue to develop novel AAVs, potent specific promoters, improved formulations and optimize administration techniques in order to solve some very important questions, including: How can we increase expression to improve efficacy? How can we target delivery to specific cells to improve biodistribution and safety? How can we achieve broader transduction of an entire organ? And very importantly, how can we achieve redosing to enable dose titration and durable efficacy for pediatric populations? The second category of enabling technology addresses what DNA constructs we are delivering and for what purpose. Most gene therapy companies focus exclusively on replacing missing proteins or enzymes. But at uniQure, we're constantly thinking about how to broaden the applicability of gene therapy to more patients and more diseases. After years of research, we believe we've established the leading onetime gene-silencing technology that we call miQURE. And we're applying these learnings to establish a knockdown and replace platform that we call GoQURE, as well as an exciting vectorized antibody platform called AbQURE. Each of these platforms significantly broaden the applicability of gene therapy, and Ricardo will discuss later in the presentation how we are mapping them to our product development efforts. The final part of our strategy relates to the manufacturing of AAV gene therapies. Manufacturing has always been important within the pharma industry, but for gene therapy, it's existential. First, the manufacturing process and methods are inextricably intertwined with the product itself. Second, it's very expensive to manufacture gene therapies. And quite frankly, the industry today is not capable of addressing large indications in a cost-effective manner. And lastly, it's really, really hard to manufacture gene therapies as biological platforms are inherently variable, difficult to scale and challenging to characterize. At uniQure, we spent 15 years developing a robust, highly optimized manufacturing platform that can be modularly applied across programs and scaled efficiently. We use the same platform from the earliest stages of research all the way through commercialization to maximize consistency and comparability throughout the development process. And in order to support larger patient populations, we're now working on plans to scale our process even further into bioreactors that are capable of producing thousands of liters of drug substance in a more cost-effective manner. In order to maintain our leadership position, support CSL's commercialization in hemophilia B as well as expand our pipeline, we will continue to invest in our capabilities and infrastructure. And so today, we're announcing the construction of a second manufacturing site in Amsterdam that is already underway and will be capable of producing CGMP material in 500-liter bioreactors by 2022. We're also constructing a pilot plant in Lexington to complement our manufacturing capabilities and make tech transfers and scale-ups easier and more efficient. And finally, we're establishing new research capabilities in Lexington that will work seamlessly with our Amsterdam colleagues to support pipeline expansion. We'll be going through a lot of material in today's program, so I want to call out some new information that we'll be covering. First, we'll be presenting new 52-week data from our HOPE-B pivotal study on all 54 patients and providing a regulatory update. At a high level, the 52-week data demonstrates sustained increases in FIX activity across all patients in the study, continued reduction of bleeding and no correlation of clinical outcomes with neutralizing antibodies up to 700, which we believe will cover 93% of all patients. Regarding our Huntington's disease program, we announced last week the commencement of dosing of our second higher-dose cohort and now have 12 patients enrolled in the study. Also, we continue to expect to initiate dosing of our European Phase I/II open-label study in the second half of 2021. We've received CTA clearance in the United Kingdom and expect clearances in Germany and Poland in the third quarter. We will also be unveiling 4 new research programs today, including a very exciting acquisition that we'll discuss later in the program. Additionally, we'll be providing updates on our Fabry and SCA3 programs, including some new preclinical data. And finally, we'll be announcing several new enabling technology platforms and how we plan to leverage them to support our pipeline strategy. As I mentioned, our people at uniQure are our most valuable resource. And presenting today will be a group of talented and esteemed speakers from uniQure's R&D organization in Lexington and Amsterdam, including: Ricardo Dolmetsch, our President of R&D; David Cooper, Vice President of Clinical Development; Melvin Evers, Vice President of Research; Paula Miranda, Senior Scientist in our Liver Group; Ying-Poi Liu, Associate Director in our Adult Neurology Group; and Astrid Valles-Sanchez, who's an Associate Director in our Adult Neurology Group as well. So without further delay, I'd like to hand the floor to Ricardo who'll be providing more details on our R&D strategy.

Thank you, Matt. So as Matt said, our company is focused on diseases of the central nervous system, the liver, heart and muscle. And underlying all of these is our AAV technology engine. Before I unveil our new pipeline, I want to take you through some of the new developments we've made as part of this technology engine. So the technologies that we're developing at uniQure are designed to address some of the major challenges in AAV gene therapy. So one of the main challenges has been how to deliver AAVs, how to deliver them to the right cells, how to get enough cells transduced, how to dose through preexisting neutralizing antibodies, how to redose if patients don't respond appropriately and how to dose in a way that is noninvasive. We're also -- so we've developed technologies to address these major issues, and I'll get to them in one moment. We are also focused on developing the next generation of cargoes. The cargoes, of course, are the guts of a gene therapy. They are necessary for effectively knockdown, the efficient delivery of genes for the replacement and editing of the genome as well as for the effective delivery of biologics. And we have some exciting development there -- developments there as well. And then finally, the third pillar of our technology strategy deals with the challenges of manufacturing. And as Matt said, gene therapy is in the early stages of developing its sort of manufacturing technologies, and we are the leaders in this. And our ambition is to develop new technologies that increases the robustness and reliability of manufacturing, decrease the cost of goods, so we can bring our gene therapies to a much larger population. So I'll focus today on our delivery and cargo technologies. And let me start by telling you a little bit about some of the delivery technologies I'll be talking about today. So I'll talk about 3 technologies. QUREDose, which is our set of technologies for dosing through neutralizing antibodies and redosing. And I'll tell you some proof of -- some exciting proof-of-concept experiments that are enabling us to try to redose in the clinic. I'll tell you about our new generation of capsids, which we call Smart AAV. These are antibody-enabled capsids that can go to specific tissues and specific cells and across the blood-brain barrier. And then I'll tell you a bit about QURE-HDL, which is our technology for improving the transduction of the liver by taking advantage of high-density lipoproteins. So let me start with QUREDose. So the problem of neutralizing antibodies has been an issue in gene therapy since the very beginning. Many patients, as you know, have neutralizing antibodies against many AAV serotypes. And 5 or 6 years ago, we discovered that AAV5 was privileged in this regard. And in our hemophilia program, we have been able to dose through preexisting neutralizing antibodies. We are now extending this to be able to redose. To do this, we have done experiments in which we have administered 1 dose of gene therapy in an AAV5 vector. And then we have reduced the neutralizing antibody using plasmapheresis before redosing with a second gene therapy. And this works. And this -- at least in a nonhuman primate. And so what you see in the graph is that we can actually get beautiful gene expression when we redose. And we think that this establishes proof of concept for being able to try this in humans, which we plan to do in 2022. We think that this enables a wide variety of new applications. It allows us to precision-dose transgenes and to retreat patients with partial responses as well as treating children. It should also improve the biodistribution of our gene therapies. And we believe it is truly revolutionary. The next technology I wanted to cover is our new generation of AAV capsids. These are -- we call these Smart AAVs. So these smart AAV capsids combine the advantages of AAV5 with antibody-directed delivery to get cargo across the blood-brain barrier and to improve the transduction of cells, like microglial cells that are hard to transduce. We are taking single-chain antibodies derived from camelids, like llamas, and we have introduced them into not all the CAP proteins, but a subset of the CAP proteins in AAV; and we're using this to direct the transduction -- or to direct the AAVs into specific cells and into specific compartments, which allow the viruses to cross the blood-brain barrier. The next technology I wanted to cover focuses on our liver platform. Many AAVs go to the liver, and you would think that this was a solved problem. But in fact, we know from our studies, as well as the studies of others, that AAVs normally transduce just the cells that are around the portal vein. It is very difficult to get the whole liver transduced. And so we have been working hard to improve the ability of AAV5 to transduce the liver, and we have developed a new capsid. This capsid contains a binding moiety that binds to high-density lipoproteins which are taken up by liver cells. And it is somewhere between 10 and 15x better than naked AAV5. And this should have 2 very tangible effects: One, it should reduce the amount of AAV that we have to give patients; and secondly, it allows us to transduce the whole liver, which opens up a whole set of diseases for us to work on. So those are our delivery technologies. Let me tell you a little bit about our cargo technologies. So again, you've heard about miQURE, and Matt talked about that, which is our ability to deliver microRNA safely and effectively. I'll tell you about 3 new technologies: LinQURE, AbQURE and GoQURE. So LinQURE is the name we've given a new technology which allows us to deliver multiple microRNAs in a single AAV. Again, we have been working for a long time to optimize the scaffold that allows us to deliver a single microRNA, but we have extended this so that we can now deliver multiple microRNAs. You would think that this would be straightforward, but in fact, there have been many things that we need to optimize. As many of you might know, delivering microRNAs can be toxic if you don't have the right kind of scaffold. And in addition to this, if you use the scaffold over and over, you will not be able to manufacture the gene therapy. So we have optimized this, and this offers a number of advantages. Having multiple microRNAs makes it much easier to knock down a gene. It also allows us to knock down multiple genes in a pathway, which of course, opens up a whole set of new indications. The next technology I wanted to cover is what we call GoQURE. And GoQURE gives us this ability to actually replace a gene. So we're combining our miQURE platform to knock down a gene with the ability to express a gene in the same AAV. Now again, the details really matter. In principle, this is a straightforward thing to do. In practice, it is really difficult to get an AAV that efficiently produces an mRNA leading to a protein and also efficiently produces a microRNA. But we have achieved this. We have found ways in which we can do this safely and effectively. And then finally, I want to tell you about AbQURE. So AbQURE is the technology that allows us to deliver therapeutic antibodies. So we've already shown with our hemophilia B program that we can deliver Factor IX, and we want to extend this from the liver. And so we want to extend this to delivering therapeutic antibodies. We think that this offers some major advantages in the sense that it kind of changes the biologics world so that we can do one-shot delivery of biologics for serious diseases. And so what you see in the graph is that we have able to generate AAVs that can deliver antibodies efficiently. These antibodies bind to their targets, they're efficiently secreted. And this is true for both liver cells and cells in the central nervous system. So our technology platform really enables our pipeline. And today, we are unveiling actually 4 programs, but I'm only going to tell you about 3 of them because the last 1 is a surprise. So you've heard, of course, about hemophilia and Fabry and Huntington's and spinocerebellar ataxia if you've been following uniQure. And today, we're going to tell you about Parkinson's disease, ALS and Alzheimer's. And each one of these takes advantage of our technology platforms to develop these transformative gene therapies. I also want to take a moment to tell you a little bit about how we select our indications because this is a key part of our DNA. So we are developing gene therapies that are best in class, that maximize our opportunities but also balance risk. And so we start with indications where there's a large unmet medical need. We work on targets that are validated in humans, either through genetics or through human clinical experience. We make sure that there's a large addressable population. We have a very safe delivery vehicle, and we complement that by having very safe targets. We make sure that we can address the target technically, both preclinically as well as clinically. Of course, we want to make sure that we can actually carry out a clinical trial. And finally, we make sure that we're well differentiated from the competition. And as you'll see, that is a characteristic of all of our pipeline. One of the things that we're doing, as Matt mentioned, is we're extending our pipeline to much broader indications. And of the 3 new indications you'll hear about today, Parkinson's, Alzheimer's and ALS, all of them affect tens of thousands of patients. And so I think this is really the next chapter in gene therapy as we move from ultra-orphan and orphan indications to much common diseases. So let me finish by showing you our new pipeline, which is here. And just pointing out the 3 programs that -- the 3 new programs, again, that we will be introducing, Parkinson's disease, ametropic lateral sclerosis and autosomal dominant Alzheimer's disease. So with that, let me give the floor to our Vice President of Clinical Development, David Cooper, who is going to give you an update on our Huntington's program.

Thank you, Ricardo. My pleasure to be talking this morning about the Huntington's disease program at uniQure. By way of background, Huntington's disease is an autosomal dominant inherited disorder, which means you have a 50% risk of getting HD if your parent has it. There are about 25,000 patients each in the U.S. and EU who have symptomatic HD. It's initially described based upon a characteristic chorea, a movement disorder; dystonia in coordination; ataxia. Later, rigidity and bradykinesia or slow movements contribute to functional impairment. But I think what we've learned a lot over the last decades of longitudinal natural history research is that cognitive and behavioral symptoms may occur early. They could be some of the first subtle signs even before someone has the movement disorder, and this can date back to people in their 20s. It's a progressive course from onset at about age 45 to death over 10 to 15 years, and there are no disease-modifying treatments at the moment. So how does the neurodegeneration occur? So it starts out with this abnormal CAG repeat in the Huntington or HTT DNA. This ends up causing a repeat that comes in exon 1 of the mRNA and gets expanded in the protein into this polyglutamine tract in the protein. This abnormal protein then aggregates and causes neuronal degeneration. And I think the important thing for understanding HD and our approach to HD is this does not occur uniformly in the brain. This starts in the striatum, which we're showing here on the right. In the darkest color orange, that's #1. It then spreads to the somatosensory and motor cortex, frontal lobe, parietal lobe, occipital lobe. So we have to think about a treatment in terms of also treating the areas that are the most affected. So how are we doing that? What is AMT-130? This is a replication-deficient AAV5 serotype. It codes for a microRNA that targets the Huntington mRNA at exon 1. And I'll explain in the next slide why this is important. Ultimately, it blocks the expression of the Huntington protein. We've been working on this for many years. We started out with cultured human neurons. We went to 5 rodent species, including HD rats, 4 types of HD mice; and then brought this into large animals to understand how to really administer this through MRI-guided stereotactic approaches, both in nonhuman primates and in a large animal model of HD, the transgenic minipig. We've had 56 of these stereotactic approaches that have been done in the large animals, the nonhuman primates and the minipigs to get us ready to go into the clinic last -- that we did last year. And again, this is a onetime injection of AMT-130 into the striatum focusing on the caudate nucleus and the putamen. So I mentioned exon 1. The HTT or Huntington DNA doesn't just create a full-length mRNA and full-length protein. It also has these fragments that get formed through aberrant splicing. One of the most important that's been identified is the exon 1 fragment, which has been associated as well in human autopsy studies of patients with HD as being very toxic. So because of where our microRNA is blocking the mRNA, we're able to block both the exon 1 HTT mRNA and the full-length HTT mRNA. So we continue to generate longitudinal long-term data now out to 36 months in the transgenic minipigs. This shows, going from left to right, that we're able to get stable expression of the microRNA that we're able to see in the CSF. In terms of the brain regions, shown in the middle, we're able to show that we have very consistent knockdown over time. This shows 6 and 12 months in the minipigs, showing that particularly the area we're targeting the most, the striatum, the caudate and the putamen is knocked down at that period. And then also on the right, neurofilament, which we know that neurofilament in any patient that has a brain surgical procedure will show a transient spike. But this gives us some idea that this returns to baseline within a period of months, and that we should be able to see that return to baseline as a marker for ultimately tracking their long-term baseline neurofilament levels against the natural history. So there's a lot of talk about total Huntington knockdown versus selective, a little selective knockdown. This is our approach really, we think, optimizes the benefit/risk trade-off. We're focused on neurosurgical delivery to the most relevant brain regions. We're not targeting 100% knockdown, or the entire brain. We're really targeting, as I said, the sources of where HD pathology is going on. In the striatum, we're targeting 50% to 75% knockdown, and that's where we're infusing the gene therapy. And in the cortex, where we'll have retrograde and angiograde transport, we're targeting 25% to 50% knockdown. So what evidence do we have that suggests that knocking down wild type and mutant is safe? Well, if we look at adult animal studies, as shown here, in the first row with the cartoon, you can see that wild-type inactivation in adulthood does not really result in any kind of phenotype in a mouse. If you have 1 wild-type gene and 1 mutant gene, you end up with a Huntington's mouse, which occurs in adulthood. If you have partial reduction in mutant Huntington or both wild-type mutant Huntington, what you show is that there's, in adulthood, that slow disease progression and a delayed onset. So in the animal studies, partial reduction of wild-type Huntington in adult rodents and NHPs was generally self -- safe and well tolerated. So what do we know about humans? Well, one way is to look at people who have mutations. So if you look at people who have mutations who don't have 100% Huntington protein expression, what does that tell us? So there's 1 woman who's been characterized with 1 normal gene and 1 disrupted gene. So presumably, no more than 50% normal level. She developed normally. She's 46 at the time of last publication. She has not had an abnormal phenotype. She has a child who has the exact same abnormality and also remains asymptomatic. So what do we know about whether this mutant Huntington expression is actually creating some functional Huntington activity? Children who have heterozygous variations and very significantly decreased Huntington function showed early neurodevelopmental disorders, such as Rett-like syndromes, but don't get a Huntington phenotype. However, when we look at adults who have homozygous CAG expansions, 2 abnormal genes, they also have normal development until onset of HD, and the age of onset and preonset symptoms were similar to those who have 1 wild-type gene and 1 variant allele. So they had normal development, and the disease progression is ultimately no different from heterozygous HD. So these findings indicate that variant HTT must have some normal function during development and that the polyglutamate expansion is a toxic gain of function variation. So that brings us to our Phase I/II study. So the schematic on the right shows the Phase I/II study. It's a double-blind, randomized imitation or sham-controlled surgery. We're testing 2 levels of AMT-130. Cohort 1, which we completed this year, has 50% striatal, 25% cortical knockdown. Cohort 2, 75% and 50% striatal and cortical knockdown. It's administered by a onetime bilateral stereotactic neurosurgical procedure. We use MRI-guided convection enhancement, which I'll describe in a few slides. The follow-up is 12 months blinded and then 5 years overall, and we're doing this in 12 centers in the U.S., including 3 sites performing surgery. This shows the key inclusion and exclusion criteria, and I want to highlight a few. This is an early stage HD. And typically, that's been a DCL=4 or motor manifest HD. But we've actually, with the FDA's advice, been able to expand this to DCL=3 or multidimensional. And I mentioned cognitive and behavioral symptoms are important as well as motor. So if patients in the study present at screening and have a combination of motor, cognitive and behavioral symptoms that suggest that they've had symptom onset, they're eligible to be enrolled. And this is the first trial to really go into that earlier population. Also important is this is the first trial that has MRI minimum volumes for the putamen and caudate that need to be met. And that's both for surgical safety, but it's also so that we know that there's enough substrate in those targets that we think that we can exert disease modification. A lot of our exclusion also is very focused around ensuring that the patient does not have any contraindications to having the surgery. So when we think about the proof-of-concept endpoints in this Phase I/II, there are 4 groups. The first is biomarkers. I mentioned neurofilament earlier as well as mutant Huntington levels in the CSF. We're also looking at other exploratory markers and have samples of CSF saved for other exploratory analyses later. In terms of imaging, we're definitely looking at MRI as well as MR spectroscopy. Striatal volume will be key looking for atrophy, but also whole brain ventricular volumes, et cetera. In terms of clinical parameters, we're looking at total motor score and total functional capacity as well as composite UHDRS scores and other measures. But we're also taking a quantitative look using Q-Motor as a quantitative assessment of motor function, which assesses finger, hand and foot tapping, grasping and lifting. So we've learned a lot from the surgical procedures that we've done over the last year. This is the first gene therapy infusion that's been done in HD patients. It's also the first gene therapy in the brain that's been done with 6 infusions, 3 sets of bilateral infusions. And it's the first convection-enhanced infusion into the caudate nucleus. So a lot of firsts here. We're very thankful to be working with a group of international experts. We have a neurosurgical committee, which actually represents all of our dosing sites in U.S. and Europe and actually goes on a combined call every time we do surgeries to do trajectory planning. And then we do a postop follow-up call to discuss lessons learned. And we very much appreciate their feedback and have adapted the procedure as we've gone to understand best practices. So these are some images showing what the actual procedure's like. We use this very complicated, computer-guided software that basically allows us to place these little frames on the head, align them to a very specific trajectory and put a very thin microcatheter into that very specific area of the brain and then use the MRI to guide where we're delivering the gene therapy, how fast we're delivering the gene therapy as well. So I'm going to show you in a video in a second, but really what's the key of this MRI-guided convection-enhanced delivery is we're controlling the depth of the catheter within the organ here shown for caudate and posterior putamen. We're taking MRI images as the infusion is going on. We can loop and make minutes into seconds and basically condense a set of 20 or 30 MRI images into a single looped video that gives you some idea of how we watch this infusion. Eli, if you can show the video, please. [Presentation]

Thank you. So that really gives you an idea as to how the surgeon is assessing the depth and the rate of infusion throughout this procedure. We've made a lot of progress in the last year. We dosed the first 2 patients just over 1 year ago. We were able to dose the next 2 patients, complying with the FDA's design for 3 months of data in October of 2020. We then completed enrollment of the first cohort in April of this year, and we are able to proceed to dosing the first 2 patients in this cohort in June. We've disclosed that we're also starting an EU study. This is an open-label study that we really are designing to augment our ability to detect a signal in Phase I/II and enable us to find a dose to go to Phase III. So we're including both a low-dose and high-dose arm in the European study that will augment our ability to compare the 2 doses as well as compared to sham. In total, we'll have 41 patients between the 2 studies. So what are the expectations? Well, it's always important to say there's a Phase I/II study. Demonstration of safety is key, demonstration and safety of administration and safety of both doses. The patients in the U.S. study will be unblinded by groups after they finish their 12-month blinded visits. We'll be following then the patients who have received AMT-130 for 5 years. Efficacy and biomarker data will be analyzed after each cohort and study is completed. I think what's very important, I talked about that this is a group of patients who are unique in that we have to have a minimum volume criteria. So a lot of the natural history that exists is for the larger group of patients who have Stage 1 or 2 disease. We're working with CHDI and IXICO and other partners to develop a very specific natural history data set to map our expectations for this group of patients who have a certain minimum volume with those clinical characteristics as well. And we believe overall that the Phase I/II studies will demonstrate disease modification to inform us on proceeding with the Phase III trial. So what are we going to know by the end of this year? Well, the first 4 patients will be unblinded by the end of this year. And remember, these are pairs of 2 in the initial part of the study. So that's 1 dose, 1 control. We're just unblinding the first pair that were treated a year ago, and then we'll do the same in the fall. So what will we know? Again, safety and tolerability is the key. What insights might we also get at that point? So chemistry biomarkers from CSF, neurofilament compared to baseline and control, mutant Huntington in CSF, markers of inflammation and immunogenicity, volumetric MRI compared to baseline control and natural history and probably some early ideas about functional MRS imaging, understanding how neuronal function compared to baseline and to control. And with that, we will then move into our first Q&A session of the day.

Thank you, David. And we now have several questions that came in for our first Q&A session. The first one is from Paul Matteis, our analyst with Stifel. And Paul asks, how should we interpret CSF mutant Huntington data for AMT-130? Should we be looking at the 30% reduction in minipigs as a benchmark for what we'd hope to see in humans?

David, do you want to answer that question?

Sure, I'm happy to answer that. I think the short answer is we would expect that, that's probably a best case for what we would see. The action of what we're trying to do is in the striatum, and the human brain is a lot bigger and the human CNS system a lot bigger than a minipig. So that we would anticipate that the signal will probably will be less than that in a human. So it's not one of the key things that we're looking for.

And a follow-up from Paul for MRI in the Huntington study. Are you largely focusing on deep brain structures like the striatum? What would you expect to see in the placebo arm regarding MRI imaging?

Hey, so you want to...

Okay. So we're applying a whole suite of imaging, and I think this is very important. We'll be looking at whole brain. We'll be looking at ventricular volume, white matter volume, gray matter volume, striatum. So we're looking very comprehensively at volume. We would expect, in the control patients or sham surgery patients, to see the same as the natural history. But again, this is why we're also looking at reanalyzing some of the original data from some of these longitudinal studies, to be able to understand what that natural history should be in volume loss in patients who start with a certain initial volume with Stage 1 disease. So it's important for us to establish what that baseline expectation should be for the control group.

Ricardo, I think this next question is good for you. It comes from Patrick Trucchio with H.C. Wainwright, who asks, can you discuss further what is meant by larger indications? Would this include larger orphan diseases? Or would these include prevalent diseases, such as type 2 diabetes?

Yes. We are -- I don't know if you can hear me here. But we are very focused on genetic diseases. And so we start with larger monogenic diseases, but we're moving towards polygenic diseases that have a strong genetic basis. So perhaps one day, we would go for type 2 diabetes. I think for now, the way in which we're doing things is we're starting with diseases like Huntington's that are clearly monogenic but are quite large indications, and we're moving to diseases like Parkinson's and Alzheimer's that have a strong genetic basis, but of course, have much larger populations.

Okay. And another question -- actually not another question. A first question from our analyst with Truist, Robyn Karnauskas came in. And it is for QUREDose. Is there an AAV viral load threshold where this doesn't work? Can other gene therapies copy this approach?

Yes. So we have developed a lot of IP around this. We think that there is something unique about AAV5, so it is something that is very difficult to do with other AAV serotypes. AAV5 is somewhat immunologically privileged. But in addition to this, we've had a number of insights that allow us to do this that are going to be quite difficult for other companies to copy.

And Ricardo, I think this is for you as well. Our analyst with Guggenheim, Debjit Chattopadhyay, asks a couple of questions. First, thoughts on the relative advantage of plasmapheresis over IgG-cleaving enzymes. For example, the simplicity and efficiency, if you can talk about that.

Yes, the IgG-cleaving enzymes are also very interesting, and we have also explored them and we've explored a few other ideas. In general, we are looking for ways in which we can reduce the neutralizing antibodies. The advantage of plasmapheresis is that it is widely practiced. It is in the clinic today. So it is something that we can do now. But as we move forward, we are exploring other ideas. And we have -- are establishing collaborations with companies that have some of these other technologies that are complementary to ours.

And Debjit also asked if there was a separate promoter associated with the microRNAs in LinQURE.

No. There is no -- it is a single promoter that drives both the microRNA and the gene.

Another question going back to Huntington's, I think as a follow-up. And this comes from Joe Thome, our analyst with Cowen. If you expect the Huntington knockdown to be less as you progress away from the injection site, what would have 50% striatal knockout equal in the CSF? David?

I think in terms of where we would think to see the CSF signal, it's probably more going to be in the high dose, where we have a 75% reduction. And again, thinking about the size of the brain in a human, and that's where we might be able to see a 25% or 30% reduction with a higher dose. I think with the lower dose, it's going to be challenging to see a reduction that will be over the baseline variation from the assessments.

Yun Zhong, our analyst with BTIG asks. Is it known at DCL=3, what the percentage of striatal neurons are still surviving?

I'm not aware of any data that specifically has looked at what exactly is the percent of neuron survival at that stage. I think people have definitely thought about the idea that we should be going earlier. And certainly, the community has been very focused on disease-modifying therapies being at a sufficient point in time that we can actually modify the disease. And I think that's a challenge across a lot of CNS, neurodegenerative disorders. But I don't think that we're really very clear as to where the -- what the exact striatal neuronal loss is at that stage. And there's a new classification system that probably will be proposed by the community in the coming years that will try and focus more on staging as a continuity that's more representative of that and less on the symptomatic occurrence.

Another question from Paul Matteis at Stifel. What exactly does disease, does -- yes, does disease modification look like in the Phase I/II trial? When you think you'll have enough data to really see disease modification regardless of biomarkers?

So I can take that. We think that disease modification really is stabilization of symptoms where they are at the moment at which the patient was dosed. So we don't believe that we will be able to bring back function because there is no bringing back dead neurons. But we do think that we can halt the progression of the disease, and we know that the patients progress significantly over the course of a year or 18 months. So again, what we expect is arresting the disease while the placebo patients continue to decline.

And a follow-up question from Robyn Karnauskas at Truist. Can you touch a little bit more on the functional MRI data? What can we really learn? And how much follow-up is needed to see an impact?

David, do you want to answer that?

Sure. So I think that's similar to what Ricardo just said, I think we're looking at for functional MRI that probably over a 12- to 18-month period, we would see some evidence of changes. I think the challenge in doing the first gene therapy infusion in HD is that we need to understand the time course of some of the postoperative changes that occur in the brain in those first months after surgery. And when those changes sort of reside as with the neurofilament level towards sort of the baseline of that patient again, and we're able to do very accurate imaging, what does the volumetrics look like at 12 months and 18 months? What does the functional imaging look like at 12 months and 18 months? And how does that differ from what we would expect from the natural history?

We have also shown preclinically that using MR spectroscopy, we can see changes in some of the secondary metabolites in the striatum. So that is also something that we will be looking at with MRS.

We have 2 questions related to neurofilament light, one from Gil Blum, an analyst with Needham; and another from Yun Zhong with BTIG. The first is, is 1 year from administration a sufficient amount of time for neurofilament levels to come down to normal ahead of the biomarker output? And related to that, do you have any idea how elevated the baseline NF-L level is? Is it elevated enough for a potential meaningful reduction? David?

So I think we need to differentiate what we see in minipigs, where the minipigs are dosed very early and develop a phenotype later in life. So their neurofilament is -- returns to a baseline that's relatively normal at the time of the surgical procedure within several months. In humans, what's been shown in the natural history data is that neurofilament is on a gradual increased trajectory in patients who have stage 1 disease. And this has been shown recently as well in the Roche observational study. So what we hope is, over a period of months after the surgery, that the neurofilament returns to that baseline trajectory. And the hope is, based on the natural history data, that sham patients and the dosed patients should start splitting apart at that 12-month point. We'll also be able, obviously, to follow the neurofilament labels -- levels in the patients who are continued to follow-up up to 5 years. So we can follow the natural history of those patients for well more than a year, against their expectations when they're dosed.

And then one other one related again from Debjit. Could you talk to the NF-L changes in the plasma versus the CSF, which might be a better marker?

So we're actively monitoring both. I think the -- we'll see how those correlate ultimately with all of the brain imaging and the clinical outcomes. And have the opportunity over looking at the data in the study to try and understand, for a therapy that's directed at the brain itself, which is a better marker in the long term for the patients that improved or have a stabilization compared to the sham arm.

Joe Schwartz, our analyst with Leerink, writes in that you've done some very good animal work for AMT-130 showing the robust knockdown of toxic microRNA caused by the DNA repeats, but I believe these were not HD models per se. So I'm wondering, how much is known about the damage that the mutant Huntington wreaks on the brain? And how much can be reversed in patients after it's been set in motion for many years? For instance, are you able to transduce all cell types which could be damaged in patients?

So I can take that. We have tested AMT-130 in a wide variety of disease models. Of course, all preclinical models have their limitations, especially when it comes to the, for example, differences in the size of the brain and the differences in lifespan. So Huntington's patients get their disease in their 30s, 40s and 50s, and mice get their disease much more quickly than that. But -- we have been able to reduce the expression of mutant Huntington. We have been able to reverse the phenotypes, even of quite severe models. And while the pigs haven't shown a phenotype yet, they will -- we are still looking at them. And I guess, we're cautiously optimistic that they haven't developed the phenotype. So it's not quite true that we haven't tested them in Huntington's models. I think it is very validated preclinically. Of course, we have to wait to see what we will see in patients.

Salveen Richter, our analyst with Goldman, writes in. While caveating that this will be informed by the natural history that we're collecting, what difference in the volumetric MRI and functional imaging do you expect would be clinically meaningful between the placebo -- between the -- not the placebo and the control, but AMT-130 and the control? Will the natural history data be released at the same time as the HD data?

David, do you want to answer that?

Sure. So I think it's hard to put a clinically meaningful tag on volumetric MRI data. I think it will be a good biomarker to let us understand specifically what happens in HD, particularly in the subset of patients who have bigger or smaller initial volumes within Stage 1 and how they progress differently, or whether they progress similarly. So I think that's one of the things that we're looking for in this pilot project. And we're excited to be collaborating with CHDI and IXICO on this. It's part of a larger project that CHDI is doing to reanalyze all of the TRACK-HD, PREDICT-HD, IMAGE-HD data sets to modern volumetric assessment technologies that we're using in current studies. So really updating the old images to the new analyses. We're committed to sharing that data as soon as we understand what that data looks like and have it analyzed. So probably by the end of this year and beginning of next year, we'll have some information as to what natural history does look like in subsets of patients with stage 1 disease.

And just the last couple of questions before we go into a break. This comes from Suji Jeong, our analyst with Jefferies. Will you be able to measure exon 1 fragment in the CSF? And again, a follow-up on -- do you think 1 year of follow-up will be enough to see changes in the putamen and caudate volume?

I can take the question on the fragment of Huntington. So we have developed an assay. The levels of the exon 1 are very low, so we are not completely sure that we will be able to measure it. But the assay is now at the point at which we're validating it so that it will be useful in humans. And then the next part of the question. Do you want to take that, David?

Sure. I think the challenge is really that we're talking about a very small number of patients, right? So being able to compare, even at the first cohort, 6 patients who were treated with AMT-130 versus 4 patients who weren't at 12 months may not be really a sufficient number to be able to see that difference. However, what we do want to do is be able to compare those 4 sham patients, and overall, the 10 sham patients against this new natural history data set that we're working on. And then for the patients who are followed for 5 years with the AMT-130, we certainly have the opportunity well past 12 months at all of the subsequent appointments and visits, most focused in the first year and second year, to really understand how that's different than the natural history and how it's different than sham. And we'll be looking at comparing it in both ways to try and understand what the signal is, particularly given that, I mean, these are still very small Phase I/II programs. In combination with the U.S. and EU studies, I think we'll have much more ability once we have 12 months data across all of that data set. And then 18 and 24 months for the patients who were treated, we have a lot more ability to look at how it's different from the natural history.

And this last question really relates to that. This comes from Maneka, who is our analysts with Evercore ISI. Do you have a sense for the scope of data that will be needed for a registrational study for AMT-130, patient numbers, endpoints, et cetera, similar to the antisense Phase III study? Or could there be a more rapid path?

I can answer that. So of course, it depends a lot on the magnitude of the effect that we see in this Phase I/II study. So in principle, if the effect is very dramatic, as it sometimes is with gene therapies, then we might consider a study that might -- that is just a little bit larger than the study, or perhaps even some conditional registration with this data, given that there is no treatment for Huntington's disease at the moment. But it's also possible that the effect won't be as dramatic. In which case, we will then have to power it based on the signal that we see. We don't expect it to be quite as large as the studies that people have conducted with the ASOs, only because we have -- we are recruiting a much smaller set of patients, and we are taking, whereas they recruited patients that were a much broader population of patients, and we have many more measurements. So yes, so I don't have a specific answer to you. I can't say it's going to be 50 patients or 100 or 300. But what I can tell you is that it's quite likely to be -- it's likely to be much smaller than what people have tried to do before because we expect a much larger effect.

Great. So on that note, thank you very much, David, Ricardo and Matt. We're at our point now where we want to take a quick break. We'd like to return to our program within 10 minutes. So by 9:45 a.m. Eastern Time, we will reconvene. And at that point, we'll be going into details on expanding our research pipeline. So we'll be on a break, and we'll be back to you soon. Thanks. [Break]

Welcome back to uniQure's Virtual Research and Development Day, everyone. I'm Maria Cantor, Chief Communications Officer. And at this point, we would like to switch over and focus now on the topic of expanding our research pipeline as we move forward. I'd like to introduce Dr. Melvin Evers, who is our Vice President of Research, who is going to kick off this section. Melvin?

Thank you very much, Maria. Welcome back, everyone. I'm excited to give you an update today about our Spinocerebellar Ataxia Type 3 program. So first, a short introduction about the disease, Spinocerebellar Ataxia Type 3 or also known as a Machado-Joseph disease is the most commonly autosomal-dominant inherited ataxia. It has a prevalence of 1 to 2 per 100,000 in the total population, which means roughly 7,000 patients in the U.S. and Europe. The disease manifests itself around mid-life and get worse over time and while severely impacting the quality of life. The symptom that manifests itself first is ataxia also described as drunken sailor's gait, dystonia and muscular atrophy. And when the disease progresses, the patient lose the ability to communicate, they get wheelchair-bound and become paralyzed. After this path downhill, the patient will eventually die because of this disease. So the cause of Spinocerebellar Ataxia Type 3 or SCA3 is a CAG trinucleotide repeat expansion in the ataxin-3 messenger RNA. This CAG repeat expansion results in a protein. That translated protein, it also has an expanded polyglutamine repeat expansion, similar to Huntington's disease. This expanded ataxin-3 protein acquires toxic functions, gets aggregated and results in neurodegeneration. This neurodegeneration is most prominent starting in the brainstem, upper spinal cord and cerebellum, as you can see here in the MRI picture. As the mechanism of disease is similar to Huntington's disease, also for Spinocerebellar Ataxia Type 3, we could leverage on the proprietary miQURE technology, which has proven to be safe and efficacious. And in this case, we use the miQURE that is engineered to bind to the ataxin-3 messenger RNA, and by binding resulting in a degradation of the ataxin-3 messenger RNA and thereby reducing the toxic ataxin-3 protein. To mention is that we do not distinguish between the mutant and new alpha allele as knockout of ataxin-3 has been shown to be well tolerated and by targeting both alleles, all patients have the target secretes and are thus eligible, in essence. So after selecting a lead candidate, that was also shown to be efficacious in patient-derived neurons, we performed a dose escalation study in a lentiviral model of SCA3. This lentiviral model has the protein aggregates as a hallmark, seen on the left-top panel with these black dots. After a onetime treatment with AMT-150, we saw a clear reduction up to a prevention of these aggregates, as seen in the Ferris panels as well as in the graph. Subsequently, we also looked at the neuropathology seen here with this big lesion. So in untreated lentiviral rat models are big white lesions. If we dose with AMT-150, we see a dose-dependent decrease up to alleviation of this neuropathology in this particular lentiviral SCA3 mouse model. Next, we assess the AMT-150 administration by directly administering it in the cerebrospinal fluid. In this case, we use the cisterna magna approach. After assessing effective genome copies, microRNA expression and ataxin-3 lowering in the brain areas most relevant to the disease being this brainstem and the cerebellum, we moved to a transgenic SCA3 mouse model. This transgenic SCA3 mouse model that we use is an homozygous acute model and already construction has a dramatic phenotype as can be seen here on the right panel looking at the total motor activity comparing the rat, the diseased animal and the black, the wild-type animal. That's very clear phenotype at birth. The AMT-150 was injected 1 time in the cisterna magna at 8 weeks of age. And as you can appreciate that although this is a dramatic phenotype, we could still see a partial improvement of the phenotype, as was measured again by the total locomotor activity. Well, to bridge from a small rodents to humans, it's usually important for the proper translation to the clinic that we also looked in the larger animals. And that, in this case, we make the use of nonhuman primates that were administered one time with AMT-150 directly in the cerebrospinal fluid. We made use of 2 doses, a low and a high dose and 3 animals per group and we assessed from different brain structures the genome copies as this can be seen on the Y-axis. First of all, we saw a clear dose response. So the higher the genome copies that were injected, the higher genome copy we found back in the various brain structures. And interestingly, the highest reduction of AMT-150 was found in the areas being most affected by the disease being the brainstem, cerebellum and spinal cord. Small. As a take-home message from the update on the AMT-150 on the Spinocerebellar Ataxia Type 3 program is that AMT-150 is shown to lower ataxin-3 messenger on the protein in patient-derived neurons. That is ataxin-3 lowering and reduced the neuropathology in a lentiviral SCA3 mouse model that discusses functional improvement in mice after cisterna magna administration. Also in a larger animal, we have shown that cerebrospinal fluid delivery of AMT-150 results in transduction of the brainstem, spinal cord and cerebellum, the areas most affected by Spinocerebellar Ataxia Type 3. At the moment, the IND-enabling studies are ongoing, which are the last piece of the puzzle to finalize the writing of the IND and to start the clinical development. With that summary, I would like to give the word to Dr. Paula Miranda, who will give a -- provide an update about the Fabry program.

Thank you, Melvin, and good morning and good afternoon, everyone. I would like indeed to introduce you to Fabry disease and our program with 191. So Fabry disease is a lysosomal storage disease. It's an x-linked genetic disorder affecting more severe male patients. It's deficiency of alpha-galactosidase A, standing for GLA, with the prevalence estimated of 1,000 to 3,700 to 80,000 live births and a Fabry population of 15,000 patients, both in U.S. and in Europe. These patients suffer from a severe variety of symptoms that goes from fatigue to hearing loss, neuropathic pain and cardiac disease, renal failure and are at high risk of stroke. In fact, this is the highest unmet medical need of these patients. Just because of this disease, the efficiency of this enzyme, alpha-galactosidase A, which is responsible for the degradation of globotriaosylceramide, Gb3 and lyso-Gb3. This -- the absence of this enzyme or the deficiency of this enzyme, there is no degradation of the substrates. Then you have a systemic accumulation of the substrates within the cells throughout various organs in our body. As for example, in the lysosomes of endothelial cells or in the kidney and heart. As you can see in the bottom right corner of the slide, you have histological staining of the Fabry mouse kidney. And in the Fabry mouse model, you have an accumulation of Gb3 that is depicted here in a brown-like thing and inclusion bodies. And this also happens in the patients and eventually leads to kidney failure. There is a standard care, but unfortunately, it's not efficient. It's a suboptimal therapy. It's a biweekly enzyme replacement therapy with a recombinant GLA. However, this recombinant GLA has a limited tissue penetration and biodistribution. There's a poor cross-correction which hampers the proper substrate clearance. So disease progresses despite current treatment, which is one of the reasons why we are interested, of course, of finding an alternative for these patients with such a high unmet medical need. So we developed AMT-191, which is an AAV5 GLA, with AAV5 encoding for alpha-galactosidase A in its transgene and driven by a liver-specific promoter, which is proprietary to uniQure. We -- for the development of this therapy, we did several assessments in different preclinical studies. And one of the preclinical studies, you can already see the data on the right side of the slide, in which we have a panel of histological stainings of the kidney once more. The top panel is actually well type, so representing a healthy situation, then a Fabry-treated and -- sorry, a Fabry disease model and then Fabry-treated in the bottom part of the panel. Gb3, which is a substrate that eventually you want to get rid of is indicated by a red staining. And hopefully you can see it, that its deposits in the Fabry mouse model, both in the cortical and the medullary region. And if you look into the medullary region upon treatment, this staining is cleared, meaning that you do not have Gb3 presence or accumulation when we treated with an AAV5 GLA. Going to guide you through other Fabry studies that we also have worked on. And in this study, this is actually upon a single administration, with systemic administration, of dose rating finding study. And in the top-left corner of the slide, you can see that we could deliver with 3 different doses of vector, we could efficiently deliver that to the liver, vector DNA copies, especially at mid and in a high dose, which you can see always throughout the presentation as the color for orange bars. Then just right below, you have histological staining with GLA protein in the liver, which is highly expressed than is with a brown staining throughout the liver. And we -- and then on the right side of the plot, you can see that in the top part, you have GLA activity in the liver tissue that we could assess and, of course, relate to the protein expression as well. And most important of all is that this GLA is not only present in the liver, but is also expressed and is secreted into the mouse plasma. And this is the slide that you can see or the graph that you can see in the bottom part of the slide. With -- we achieved ASO GLA activity with 3,000-fold increase compared with baseline plasma levels. We did not only assess the activity, but we also assessed the amount of protein that is present in the plasma and which correlates very well with our activity. This is very important because, of course, this will be the GLA that will be reaching the different organs with the main afflictions. So next, of course, we wanted to look into if we can reduce the substrates, the ones that are causing the disease. So you will see here the different graphs in which representing the levels of Gb3 and lyso-Gb3 substrates both in the kidney, that's in the top-left corner of the slide, and then just below, you have the levels in the heart. Throughout, you see that the knockout levels in the disease animal are really high, and they can be significantly reduced already even at the low dose when treated with our AAV5 GLA or AMT-191. We went further and we also looked into potential phenotypical correction, which it means that we would be looking into the pain perception. It's known that these mice, they have less sensitivity to pain perception. So we performed an experiment that's a non-perception which is in a hot plate -- And you lay the animals into this hot plate with increasing temperature. So they will take some time to react. But of course, the healthy animal will react faster. And this is what you can see in the darker gray bar. Then in the light-gray bar, you can see that they take longer to react. And of course, we removed the animals. But when we treated it with AAV5 GLA already at the mid dose -- but also at a high dose -- but already at the mid dose, we can achieve the same type of reaction as you would have in a healthy control. So proving that we also have phenotypical correction improvement in pain perception in these mice. Once again, of course, we wanted to go further. So we also looked into large animal models using our AMT-191. And in the top left corner, you can see that we can deliver high copies, again, with the systemic administration. We can achieve high copies in the liver and in sustained amount, so for 14 days and 56 days. Just below, you have another plot that we also assessed the GLA activity and this is 4x higher than in the treated animals compared with control naive animals. Then on the next part of the slide, you'll see that we also assessed, of course, the GLA activity in the plasma, which is again, high and sustained levels. We also looked into if they would develop anti-GLA antibodies, and we did not find any anti-GLA antibodies being raised upon treatment. And then the most important part of this slide, so the most important data is actually from the bottom plot on the middle, which is our cross-correction to the heart. So meaning that you can see here is the GLA activity. And to the AAV5 GLA or the AMT-191 treatment, you have doubling of GLA activity in these animals in the heart. So this, again, once more, this activity in GLA comes from the liver and can then be cross-correct into the heart, which is, again, one of the main unmet medical need in these patients, and which every Fabry treatment wants to achieve. On the next slide, I'm giving you -- we're giving you a therapeutic overview or landscape of Fabry disease. So I'm showing you a table here in which it has in the first column, the enzymatic replacement therapy approach, which is the current therapy. Then next to it, you have AAV5 GLA so it's AMT-191. And following on next columns are different gene therapy approaches with different serotypes of AAV and different viral vector. This all actually have -- we all are exploring alternative therapies with different serotypes and different promoters, some specific to the liver; others, overall. And some therapies actually also did need to have an immune suppression as a cotherapy which is, of course, something that you want to avoid. And in our case, we do not need it. We can do a single treatment, but in case, of course, if needed, we also can leverage our platform and do a re-dosing. Besides and most importantly, we also know that our AAV5 has low prevalence of neutralizing antibodies throughout the human population, around 92% of the human population, then ensuring that we will have a larger patient population that we can treat with our AAV5. In summary, AMT-191 is shown to be high and sustained GLA protein and activity levels in the plasma in both small and large animal models. Our AMT-191 also results in a phenotypic correction in the Fabry mice with a pain perception. And also that the expression from the liver results in GLA activity both in kidney and heart in mice and nonhuman primates. Once again, one of the main goals of any therapy for Fabry. As AMT-191 key program milestones, we expect to have GLP by the beginning of 2022, a GLP tox study in nonhuman primates and an IND filing in 2023. So now I would like to hand over to Astrid Valles-Sanchez, and she will be introducing you to Parkinson's disease program. Thank you.

Thank you very much, Paula. So I'm pleased to introduce to you our program targeting alpha-synuclein for the treatment of Parkinson's disease. Parkinson's disease is the second most common neurodegenerative disorder after Alzheimer's, affecting between 0.5% to 1% of the population aged between 65 and 69 years of age, and basing to 1% to 3% in those age 18 years or older. With an increased aging population, with the prevalence and incidence of the disease are expected to increase even further. Parkinson's disease is clinically diagnosed by the presence of bradykinesia accompanied with resting tremor or rigidity. Next to motor deficits, Parkinson's is characterized by non-motor symptoms which are very debilitating and includes autonomic dysfunction, sleep disturbances, pain, but also psychiatric and cognitive disturbances among many others. The progressive deterioration that the patients experience throughout the disease are very debilitating and have a severe impact on the quality of life and those of their caregivers. At present, there are no disease-modifying therapies available. Familial Parkinson's disease represents between 5% and 20% of the disease populations and many since have been associated with the disease. Most of them corresponding to the PARK family of genes. PARK1 corresponds to SNCA, which is the gene that encodes for alpha-synuclein. Mutations, single nucleotide polymorphisms, duplications or triplications of this gene have been shown to lead to accelerated pathology and early disease onset. Interestingly, alpha-synuclein pathology is not only linked to the genetic forms of Parkinson's but is applicable to the whole population. Alpha-synuclein is enriched in the neurons and localizes primarily in presynaptic terminals where it plays key physiological roles in external transport and neurotransmitters release. On the pathological conditions, alpha-synuclein is prone to aggregate, misfold and aggregate throughout intermediates like oligomers and fibrils. And these aggregates lead to toxicity in the cell. Aggregated alpha-synuclein is the main component of Lewy bodies, which are present in Parkinson's disease, but also in other alpha-synucleopathies such as multiple system atrophy and Lewy body dementia. This alpha-synuclein aggregates have the capacity to spread from cell to cell through prion-like mechanisms. And this is one of the mechanisms which I believe we linked to the increased extent of the neuropathology with disease progression. Neurodegeneration in Parkinson's disease affects neostriatal of amniotic secretes and other neurotransmitter systems in the brain. And actually goes very much in parallel with a Lewy body pathology which typically starts in the brainstem and extends to mid-brain and cortical lesions. At uniQure we propose complementary approaches to reduce alpha-synuclein toxicity in Parkinson's disease. First, we have designed miQURE and LinQURE approaches, the liver artificial microRNAs and reduced expression and production of alpha-synuclein messenger RNA in protein. Second, we are developing an AbQURE approach to deliver alpha-synuclein antibodies in order to block the cell-to-cell transmission of alpha-synuclein aggregates. The final goal is to combine the lead microRNA and antibody candidates in a single vector using GoQURE in order to maximize the disease-modifying potential of our proposed therapy. The design criteria of our miQURE candidates have been tailored to ensure safety and general applicability to the broad Parkinson's disease population. First, we have designed candidates, which started all 4 alternatively sliced alpha-synuclein transcripts present in the brain. We have also avoided targeting beta- and gamma-synucleins because of their redundant functions and lack of association with the disease. Finally, for general applicability, we have avoided directing the common mutations or SNPs, as indicated on the panel on the right. We also have very stringent criteria for in vitro lead candidate selection, which include: correct processing and lack of saturation of the endogenous RNAi machinery, support an endogenous alpha-synuclein messenger RNA and protein lowering and lack of off-target effects in relevant neuronal cell models. Here on the right, we show the typical results of our screening showing a dose-dependent lowering of alpha-synuclein for all of our candidates with respect to control-treated cells. We have performed initial studies transducing neurons in which we have shown an effective processing of microRNA and expression. This is shown on the left panel for one of our candidates, and in mRNA. In the inset, we have determined the measure of microRNA isoforms and also their expression in comparison to endogenous microRNA. MicroRNA is showing that their levels are very well within endogenous microRNA levels. And this is again key to show avoidance of saturation of the endogenous RNAi machinery. This correct processing and expression leads to efficient lowering of alpha-synuclein messenger RNA in the mid panel and protein in the right panel for, in this case, 3 of our test candidates with respect to control. We have also performed initial in vivo studies in several models which absolutely support the in vivo target engagement and phenotypic ask rescue of our approach. On the left panel, we show a first study in a Parkinson's disease rat model expressing the A53T mutation which we show effective lowering of alpha-synuclein messenger RNA lymphatic vibration using our miQURE candidates. Remarkably, combining 2 candidates at the same time, in this case, candidates A and B, show a synergistic effect, which actually supports the use of LinQURE for this program. We have also evaluated potential therapeutic rescue, in this case, in a C. elegans Parkinson's disease model. We evaluated the motor phenotype in this model after treatment with miQURE, showing correction of the phenotype for all the candidates tested and all the 3 time points test that are shown here on the right graph. For an efficient reduction of alpha-synuclein spreading in the brain, we propose an AbQURE approach to deliver alpha-synuclein-specific antibodies and block cell-to-cell transmission of the antibodies in the brain. Our data show high affinity of our lead candidate antibodies as shown in the mid panel with the results of alpha-synuclein binding in lyso. Additionally, our lead candidates show efficient secretion in vitro, spread here in the right panel, when comparing 2 -- choosing the peptides with antibody candidate #1. The high binding affinity and efficient secretion support the mechanism of action of this approach moving forward in this program. In sum, we propose 2 complementary approaches to reduce pathological alpha-synuclein expression and spread in Parkinson disease. First, we have shown that our miQURE and LinQURE link candidates are correctly processed and expressed in vitro and lead to a dose-dependent lowering of alpha-synuclein messenger RNA and processing. Next, we have tested a pure candidates in vitro showing high alpha-synuclein binding and efficient secretion. Finally, initial studies in in vivo Parkinson's disease models show target engagement and phenotypic corrections. The next key program milestones for our AMT-210 program for Parkinson's disease targeting alpha-synuclein includes the finalization of proof-of-concept studies in Parkinson's disease further models, the start of GLP tocology studies in nonhuman primates in order to complete the clinical package supporting an IND filing. And with this, I would like to hand it over to my colleague, Dr. Ying Poi Liu, who will be presenting our program for ALS. Thank you.

Thank you, Astrid, and good morning and good afternoon, everybody. I'm happy to present our program on amyotrophic lateral sclerosis, ALS, as well as AMT-161. So first, some words on the disease. ALS is a progressive and ultimately fatal autosomal dominant disorder with the age of onset between 40 and 60 years, with a median survival from diagnosis between 2 to 4 years. Yearly, there are 5,000 new ALS patients being diagnosed in U.S. and Europe with an incidence of 2 per 100,000 people. Approximately 10% of the ALS cases are familial with C9orf72 mutation being the most frequent and accounting for roughly 1/3 of all familial ALS cases. Typically, the patients suffer from symptoms including muscle weakness, atrophy and spasms, language dysfunction, swallowing problems, neuropathic pain. Eventually, they become paralyzed and often they die because of respiratory failure. So what is the pathology behind C9orf72-induced ALS? The pathology lies in 6 nucleotide repeat expansion containing 4 Gs and 2 Cs that resides in the C9orf72 gene. And this repeat is also present in healthy individuals, but up to 30x. However, patients can harbor as much as thousands of these repeats. And these repeats are causing a toxic gain of function via 2 mechanisms. One is RNA toxicity and the second is toxicity due to production of toxic dipeptides. And in both these cases, this will result in degeneration of upper and lower motor neurons in the spinal cord, brainstem and motor cortex of the ALS patients. On the right side, you can see a figure where the mutation reside into the C9orf72 gene and resulting in either the RNA toxicity or the dipeptide protein toxicity. In case of the RNA toxicity, the C9orf72 RNA with the repeat expansions are being expressed. And typically, they form atypical secondary structures, thereby forming RNA aggregates, so-called RNA foci, which are toxic to the cells. And in case of the dipeptide protein toxicity, the C9orf72 mRNA containing the repeats are being translated into these dipeptides in a non-ATG dependent translation mechanisms. So at uniQure, we are developing an AAV gene therapy targeting only the C9orf72 mRNA harboring these repeat expansions and leveraging on our miQURE technology. And for the delivery, we will make use of the AAV vector. So upon microRNA expression, we will only degrade the RNA from the C9orf mutant allele and thereby reduce the RNA foci and dipeptide proteins in the cell. So first of all, we addressed whether we could in fact deliver the AAV factors to the right areas that are affected by ALS. And therefore, we have injected to and on nonhuman primates into the cerebral spinal fluids and look to the vector DNA distribution in these animals, which you can see in the left graph. On the Y-axis, you see the vector DNA levels, whereas on the X-axis difference with brain and spinal cord regions. And we can see that in both animals, white spread spinal cord as well as cortex vector DNA can be found in these animals, which are the key areas where the ALS patients are being affected. And in the middle graph, we can see upon transduction, we clearly also see microRNA expression in these areas of the animals. And at last, the last graph, we can see there is a clear correlation between microRNA expression versus vector DNA copies that we detect. And the second question we addressed is how efficacious are our AAV5 miQURE vectors in a disease model. So we injected intracranially our AAV vectors into a transgenic mouse model that encodes the human C9orf72 gene with repeat expansions and first of all looked at micro expression, which you can see on the left graph. And as appreciated, the animals that received the AAV5 miQURE vectors, they show high miQURE expression as depicted in the orange and the blue graphs in both the cortex as well as in the striatum, whereas the animals that received the control AAV5 GFP did not show any miQURE expression. In the middle graph we can appreciate that upon miQURE expression, there is a significant lowering in C9orf72 mRNA compared to the control group. And most importantly, we also see a significant reduction in these treated animals that overexpressed the miQURE, a significant reduction in the RNA foci, which are the pathological markers in this model relative to the GFP-expressing animals. So to summarize my talk, AMT-161 targets mutant allele of C9orf72 and cerebral spinal fluid delivery of AMT-161 results in widespread cortical and spinal cord distribution in nonhuman primates, which are the key areas affected in ALS. And administration of AMT-161 results in lowering of C9orf72 mRNA, and most importantly, the RNA foci are also reduced in this transgenic model. So the next key milestones for the ALS program will be proof-of-concept studies in disease models, initiation of GLP toxicology studies in nonhuman primates and ultimately, we would like to have an IND filing. So with that, I give the words to our President, Dr. Ricardo Dolmetsch, to talk on -- about Alzheimer's disease.

Thank you, Ying Poi. So Alzheimer's disease, of course, is the pandemic before this pandemic. It's an existential threat to our health care system. At the moment, there are 5 million to 6 million patients with AD, and it costs us -- well, it will cost us more than $1 trillion to treat by 2050. Unfortunately, this has not gotten any better even with the recently approved treatment. There has been a revolution in Alzheimer's disease beyond the approval of the first therapeutic. And that is the fact that we have, for the first time, identified modifying mutations. So for a long time, we've known about mutations that cause the disease. So we know, for example, that the APOE gene and specifically the APOE4 gene predisposes people to get Alzheimer's disease. We have other mutations like presenilin mutations and mutations in the A beta gene that could cause -- that can lead to the disease. What is new is that we have found that there are alleles in specific genes and in particular, in the APOE gene that can actually protect you from the disease even in the presence of other mutations that give you the disease. These are modifying mutations. So the approach that we're taking for Alzheimer's disease is to target the main genetic cause of Alzheimer's disease or the most common genetic cause, which is APOE4. So APOE4 is the highest genetic risk for late-onset Alzheimer's, 40% of patients with AD have an E4 allele. The APOE4 allele gives you a 95% chance of getting Alzheimer's disease. And our approach is to knock down APOE4 and express a protective variant of APOE. And this is made possible by the discovery that not only is APOE2 protective, but they're -- even more interesting, there are mutations in another variant called APOE3 that together can protect even in the presence of mutations that lead to early-onset Alzheimer's disease. So we're very excited about this. Now one question, of course, is how does APOE actually work? And we don't know for sure. What we do know is that it's a key regulator of cholesterol transport and metabolism in the liver and in the brain. We know that it binds to a variety of cell surface receptors. We know that it modifies the inflammatory cascade in the brain, specifically microglial cells. And that by modifying this inflammatory cascade, it has the potential to affect the disease even relatively late. We think that it's likely to be more effective than, for example, removing A beta. So our approach is to take advantage of our GoQURE platform to express protective APOE variants and use the miQURE to knock down APOE4, which we know to be deleterious and then to test this in patients with severe early-onset forms of the disease. The next milestone, we're in the process of selecting a lead and conducting proof-of-concept studies, which we will be doing over the next year, and we will begin GLP toxicology studies in nonhuman primates before proceeding to an IND filing. And with that, I will pass the word on to our CEO, Matt, who is going to tell us about an exciting new development for our pipeline.

Thank you, Ricardo. So far during the presentation, we discussed 3 new programs that we are internally developing. When we announced our CSL transaction 1 year ago, we also stated we would be ramping up disciplined business development efforts to explore external innovation. In this regard, today we are extremely excited to announce that uniQure has entered into a purchase agreement to acquire Corlieve Therapeutics, including their lead gene therapy program to treat temporal lobe epilepsy. Temporal lobe epilepsy or TLE is a very, very large clinical unmet need. It's estimated that approximately 1.3 million people in the United States and Europe are affected by TLE, of which approximately 800,000 people do not respond adequately to anti-seizure medication. Together with renowned academic collaborators, including Professor Christophe Mulle at the University of Bordeaux and Professor Valerie Crepel at Inserm, Corlieve is developing a gene therapy targeting the kainate receptor, which has been demonstrated to play a key role in the pathology of TLE. In important preclinical studies, Corlieve has demonstrated strong proof of concept, including suppression of chronic spontaneous epileptic seizures. This highly compelling and strategic transaction expands uniQure's CNS pipeline, strengthens our leadership position within microRNA-based gene silencing therapies and provides us entree into a very large indication with a clear development path, established registrational end points and the potential for very rapid clinical proof of concept. The transaction also provides the platform to explore additional opportunities within the epilepsy space. With respect to the terms of the acquisition, uniQure will make an upfront cash payment of EUR 46.3 million at the closing of the transaction and quarterly shareholders will be eligible to receive an additional EUR 43.7 million of performance-based milestones through phase I/II development and EUR 160 million associated with Phase III development and regulatory approvals. UniQure has the sole option to pay up to 25% of future milestones in uniQure stock, and expect the transaction to close in the third quarter. Incorporating all uses of cash associated with the transaction, we expect to maintain cash runway into first half 2024, which represents a change for approximately 6 months compared to our previous guidance. And with this, I'll hand it back to Ricardo to go through some additional details with respect to Corlieve.

Thank you, Matt. So as Matt said, TLE is the most common type of focal epilepsy. It's associated with damage to the temporal lobe as well as hyperexcitability to the hippocampus. It's an acquired disease, often caused by brain injury or tumors or sometimes a prolonged febrile seizure, and it affects a large number of people, 1.3 million people have it. A very significant fraction of these folks are inadequately treated, about 400,000 of them with hippocampal sclerosis and another 400,000 that don't have any MRI abnormalities. Refractory TLE patients have a very poor quality of life and a significantly reduced life span. And the standard of care today is lobectomy or a laser tissue ablation. But only a small fraction of eligible patients actually choose to undergo surgery. Corlieve has developed an exciting program to target this. Targeting a glutamate receptor called the kainate receptor. So can kainate receptors are excitatory glutamate receptors that are epileptogenic and are believed to be aberrantly expressed in the hippocampus of refractory TLE patients. They drive the seizures through the recurrent excitation. So they form synapses onto themselves and thus drive the hyperexcitability of the hippocampus. And kainate receptor knockout mice are resistant to epilepsy in a preclinical model, a pilocarpine model of TLE. And that is -- you can see that in the graph on the lower right. So AMT-260 and AMT-261, which is a variant, are AAV gene therapies that delivered an engineered microRNA that targets kainate receptors. They dramatically reduce seizures in a pilocarpine seizure rodent model, which you can see in the graph on the upper right-hand panel. They also interestingly reduce seizures in human brain slices from patients with TLE, providing such human validation for the approach. So we're excited about this collaboration. We believe that this dramatically increases and enhances our pipeline. The next milestones are going to be a GLP tox study, a pre-IND meeting and an IND filing, which is scheduled for the coming years. So with that, let me bring you back to our new pipeline that now is expanded to include temporal lobe epilepsy, Parkinson's disease, amyotrophic lateral sclerosis and autosomal dominant Alzheimer's disease. We believe that this meaningfully changes the potential impact of uniQure, and we are excited about what this means for the future of the company. So with that, let me give the word to Maria as we have questions -- as we take questions from the audience.

Thank you, Ricardo. And now I'd also like to welcome Jon Garen, our Chief Business Officer, to join this Q&A session, as Jon was integral to the execution of the Corlieve transaction. We have several questions having to do with our research programs that we just spoke about. The first one comes from Difei Yang with Mizuho, and she asks regarding SCA3, if there are any signs of dose-dependent safety signals for SCA3 that we've seen.

So Melvin, do you want to take that?

Yes. Thanks for the question. And this is a very relevant question. And to split it into 2. At the moment, we are conducting a GLP toxicology study specifically designed to look at safety. Next to that, in the preclinical setting, we have also done a lot of work on the microRNA processing and the saturation of the endogenous machinery, also looking at safety from that aspect. There, we have not seen any issues with the microRNA. And again, for the safety aspect, currently, we have a GLP toxicology study up and going.

I should also add that we don't think that there will be target-related toxicity because a complete knockout of ataxin-3 is well tolerated in animals.

Next question comes from Yun Zhong with BTIG. Again regarding SCA3. Just a confirmation, is the intra-cisterna magna injection going to be the root of administration for our program? And what was the delivery method used in the nonhuman primate study?

Melvin, do you want to take that?

Well to start off with the latter part. In the nonhuman primates, we have tested both intrathecal as well as intra-cisterna magna administration. There we found that cisterna magna administration results in the most favorable transduction of the brainstem and the cerebellum being most effective. At the moment, we are in the process of the clinical development, and this is part of a discussion that we also need to have with the regulatory authorities on how to proceed in the clinic.

Turning now to Fabry disease. A question from Luca Issi, our analyst with RBC asks, another company was recently asked by the regulators to run a head-to-head trial versus Fabrazyme. What is our reaction to that news? And how should investors and analysts think about the implications for our program?

I can take that question. So we are -- we have always anticipated that we would have to run a clinical trial against the standard of care, which at the moment is Fabrazyme. So this does not, in fact, affect our clinical plans. I think for this other company, I think the surprise was simply that they had not run their trial against the standard of care, which would require them to redo their clinical development. We're at an earlier stage. So we don't think that, that will be an issue.

Okay. We have a question coming in now from Robyn Karnauskas at Truist. What is your take on the failure of some of the other alpha-synuclein approaches despite having positive preclinical data? How do you view the ASO approach for ALS versus gene therapy? And lastly, on ALS, given how much variability there is in disease progression, even with the familial form, how are you thinking about clinical development for ALS? So first with alpha-synuclein and then moving over to ALS.

So let's start with alpha-synuclein. So the therapies that have failed in alpha-synuclein have largely been biologics. So -- and that, to be honest, is not a big surprise. The concentration of alpha-synuclein in the periphery is extremely high. So if you deliver an antibody IV, the vast majority of it binds to the peripheral alpha-synuclein. Most of these companies proceeded into Phase II without any evidence of targeting engagement in the central neuro system. In fact, even preclinically, these anti-alpha-synuclein antibodies don't actually lower alpha-synuclein very effectively in the brain even of an animal. So it's really not a surprise that some of these have failed. However, the genetics is clear. Alpha-synuclein mutations cause Parkinson's disease, Lewy bodies contained alpha-synuclein. We firmly believe that by combining microRNA to reduce the expression of alpha-synuclein and an antibody to prevent the fusion, we are going to have a therapeutic effect. So of course, we don't know, but we are cautiously optimistic that this is a good approach, and it will be quite different from having to administer a large biologic peripherally and hoping that it gets into the central nervous system. So the next question has to do with ALS and what I -- what we think antisense oligonucleotides. Well, I think that the antisense oligonucleotides approach to C9orf is a reasonable approach. It's a disease of upper and lower motor neurons and ASOs administered intrathecally certainly do reach those targets. On the other hand, ASOs have to be injected once every 2 months, sometimes once a month intrathecally. If you have ever had an intrathecal injection, it's not the sort of thing you want to do again. So I think that a gene therapy that lowers C9orf is going to be very competitive relative to an antisense oligonucleotide that has to be administered much more frequently with the accompanying adverse events. Now the next question is, given the variability in the progression of ALS, how are we thinking about this clinically? Again, there is certainly variability in ALS progression. I think there have been a number of clinical study -- of clinical studies and clinical study designs that have actually shown that one can actually measure the progression actually quite effectively in a relatively small number of patients. There is some variability, but I think it should be possible to see especially a dramatic change in the progression of the disease. So -- but of course, the details of how we're going to do this, we'll wait until we have our conversations with the regulators.

A confirmation question coming in from Debjit. Will AMT-210 be administered directly into the putamen? Or will this be IV via AbQURE?

We still -- we're still -- we're in the process of evaluating those 2 options.

Turning back to Fabry. Difei at Mizuho asks, could you share your view on the AAV versus lentiviral approaches, pros and cons for delivery in treating Fabry disease? And how important is it to have expression in the CNS?

So Paula, do you want to take that question?

Yes, I can. So indeed, both AAV and lentiviral vectors are very different. As you know, one, AAV is not integrated; while lenti, it is. And in fact, they have an approach that is ex vivo, which is more cumbersome and also increases the chances of having more differentiated treatment between different patients. So they have to isolate hematopoietic stem cells, lentiviral transduction and then we transplant, which with AAV is only one single administration. So we see advantages on that. With the CNS, we are looking into, of course. It can be important. It's an important indeed assessment that we're also looking into it.

Yes. The one thing I would add to Paula's answer, is that the lentiviral approach, of course, requires a conditioning of patients. The conditioning regimen is not for the faint of heart. It has somewhere between 5% and 10% mortality. So while this is a really severe disease, I think there is always going to be an advantaged tool, an IV-administered onetime treatment relative to an ex vivo lentiviral integration, which carries with it, in addition to that, an oncology on sort of oncogenesis risk and a conditioning risk.

A question from Joe Schwartz with Leerink asks also about Fabry. Can you talk about the transgene that you envision delivering? Will it encode for a wild-type protein or fragment? And how will you be titrating the promoter and what cell types is it being expressed in? And lastly, does it need to get in, in order to clear toxic substrate?

Paula?

Yes. So thank you. We have a wild-type GLA. And with a proprietary liver-specific promoter that has shown in different studies are ready to be very strong promoter. We also have shown in our data, as you could see, that we can assure cross-correction with such high levels in the plasma of GLA. So I think -- I hope this covers your question. But in essence, we have a wild-type GLA, which is expressed by a very strong promoter. In terms of specific expression is in the hepatocytes. So it's restricted expression through the hepatocytes in the liver.

But I just want to make it clear. We expressed it in the hepatocytes, but we've seen very strong cross-correction in the kidney and in the heart and in other tissues. So obviously, the enzyme that is generated can be efficiently taken up by other cells, which is, by the way, not true of Fabrazyme.

A question came in regarding the Corlieve acquisition, and this is from Yun Zhong at BTIG. Does Corlieve have its own HEK293 cell manufacturing system? And with the acquisition, do you expect any future programs other than TLE to be based on the mammalian cell system?

Jon, do you want to answer that?

Certainly. The system that's currently in use by Corlieve is based on their partnership with REGENXBIO. So that's the system that would be continued forward with this program. And at the moment, we would contemplate what the platforms would be for future programs as well, whether it would be REGENX or others or our own.

Okay. Another question came in from Difei Yang at Mizuho. Most of the newly disclosed programs are in the CNS. Is it fair to assume AAV5 has advantages over other AAVs and CNS? And if so, in what ways?

Yes. AAV5, especially when administered directly seems to transduce neurons in ways that other AAVs don't do well. It has -- I think its main advantage actually is immunological, which is that patients don't have neutralizing antibodies or only a very small fraction of patients who are neutralizing antibodies and that the neutralizing antibodies that do exist are relatively low affinity so we can dose through them. It also has some significant safety advantages relative to AAV9 and AAVrh10. So AAV9 and AAVrh10 are commonly used in the central nervous system. They transduce neurons relatively well. But they have some toxicity issues at all doses. So we, therefore, think that there are some advantages to AAV5, particularly in the way in which we've modified it and the way in which we are administering it.

So regarding our new program in TLE, Joe Thome at Cowen asks, when do you expect that AMT-260 could enter the clinic for TLE? And do you expect the reductions in the seizure burden should be evident relatively shortly after administration?

Yes. We think that our TLE program should be in the clinic over the next 2 years or so. We haven't guided any specific dates yet. This is something that obviously has just happened. And so we have to make sure that we fully -- we have conversations with the regulators to understand exactly what is required. One of the attractive parts of this indication is that we think that we can get a clinical readout relatively quickly, and it should be relatively unequivocal. Seizures, of course, are something that we can easily detect. I should say, we're also excited about this because preclinical models of seizures are actually very predictive of success in the clinic. So the fact that this works in this model of TLE is also exciting.

Going over to C9orf72, a question from Yanan Zhu at Wells Fargo. For this miQURE approach with C9ORF72, how did you achieve the selectivity for the mutant C9orf72 allele?

Yes, we haven't disclosed that yet. But we have achieved selectivity.

Okay. And a question from Suji Jeong at Jefferies. For Parkinson's disease, is developing GoQURE approach, the ultimate goal of the program? Or do you plan to develop lead candidates from miQURE, LinQURE and GoQURE and then select the final development candidate?

Well, we -- at the moment, we're excited about the combination of multiple mechanisms. But of course, in the process of development and when we settle on the final lead candidate, we will select the one that is most effective in the broader set of preclinical models.

Debjit at Guggenheim has a follow-up. At what dose would DRG toxicity be a concern with ICM route of administration?

We haven't seen DRG toxicity with AAV5. That is one of the advantages of AAV5 relative to AAV9 and AAVrh10.

Okay. Another question from Suji Jeong at Jefferies. For the Alzheimer's disease program, given APOE4 is a risk factor and not everyone with APOE4 has the disease, do you think replacing the APOE4 with the protective allele will have therapeutic benefits in patients?

Well, if you have 2 copies of APOE4, you have a 95% chance of developing Alzheimer's. So I think that if you're a heterozygous, you have somewhat reduced chance but still a significantly elevated risk. So we will initially develop a therapy for people who are homozygous. There are still tens of thousands, hundreds of thousands of those patients. And then we will test it in a broader population. We do think that this is likely to be efficacious in all patients ultimately based on the genetics.

Patrick Trucchio at H.C. Wainwright asks, regarding the Fabry program, what, if any, immune system regimen would you anticipate in humans? What learnings, if any, have emerged from other AAV programs in Fabry? And what gives uniQure confidence AMT-190 could have best-in-class characteristics?

So Paula, do you want to take that?

I can take that. So one of the best, well, things that we can say about AMT-191 is using AAV5, which has a high or low, in fact, prevalence of neutralizing antibodies in the general population compared with the other stereotypes. Besides, we also have a very strong liver-specific promoter. And we have shown in our data that we can cross-correct afflicted organs such as the kidney and the heart with our treatments. What else can I say? In terms of immune suppression, we show that we do not have neutralizing -- so antibody development against anti-GLA, which is indeed the case for other programs with AAV, already nonhuman primates. So we do not see this with our product. So I think these are the main advantages and what can turn our product into a best in class.

Yes, there is something sort of subtle here. Of course, patients that we don't have GLA, so if we introduce GLA, it principally could develop antibodies. We think that there is a reduced risk of that because we express something in the liver that can actually trigger tolerance. And so we're, again, cautiously optimistic that, that will be the case because that seems to be the case in preclinical species.

Okay. We have a few more questions. A question came in asking for the status of the research collaboration with Bristol-Myers Squibb.

Yes. So I'll take that one. Yes. So the research collaboration with Bristol-Myers Squibb continues to move forward, and it continues to move forward nicely. We've designated 4 programs. So we have 4 designated collaboration target programs with a lead program that we expect to move into IND-enabling studies. We announced, I guess, earlier this year, the receipt of a $5 million milestone payment regarding BMS' designation of a clinical candidate. So we're very excited about that. And they continue to be very enthusiastic about gene therapy and our collaboration of working together and are very excited to move these targets that are focused on cardiovascular muscle disorders forward into the clinic.

Okay. And a couple more questions before we move on with our program. From Judah Frommer at Credit Suisse. Do you have a sense for the magnitude of the C9orf72 lowering that is believed to convey clinical benefit in ALS? And again, any read-throughs this from other programs?

Sure. So Paula, you want to take this?

I thought maybe you want to take it.

Yes. So how much loss of -- how strong regulation do we need for clinical efficacy? So for C9orf72, we think that we have to reduce it by around 75%. This is based on the relative changes associated with alleles that are expressed that have a lower expression level. Yes, I don't know that there is any read-through at the moment from any of the other programs. So of course, we will continue looking at that. I would say one thing that we are excited about is the fact that we can knock down the mutant allele. So in principle, we can aim for complete knockdown. That should still be safe and well tolerated. Heterozygous mutation is completely well tolerated in humans and in animals.

Also on ALS, 2 additional questions, one from Paul Matteis from at Stifel, another from Luca Issi at RBC. How confident are we that C9orf72 is not a loss of function pathology, in which case, a knockdown approach may not work? And then why going after C9 ALS and not SOD1 ALS? Just wondering as one of our competitors has Phase III data for SOD1 reading out in the second half of this year.

Yes. So let me start with the second one. So SOD1 is also interesting, but it's a much, much, much smaller population. So C9orf is the most common mutation associated with familial ALS. But of course, we will also consider SOD1. So that's the first one. And the next question is, how -- why do we think that C9orf is a gain of function versus a loss of function? Well, full knockdown C9orf does cause a phenotype, but it doesn't actually cause an ALS phenotype. It causes a different phenotype. And heterozygous knockdown is well tolerated. So we don't think that it's actually a loss of function. And the fact that we can knock down just the disease allele, I think, is also going to be helpful in that respect.

And the last question from Robyn Karnauskas at Truist. Big picture, there's certainly a lot going on. Help folks understand how we plan to balance focus on later-stage programs with these various earlier-stage programs as we go forward.

Yes. I mean, I think the reality is that they do use different calories in the organizational apparatus, right? So yes, there's some crossover with respect to CMC resources, but a number of the programs that we discussed are in the research phase. And we think we have the bandwidth and capacity to drive these programs forward and to drive them forward quickly. Really the resources that are required for Huntington's and hemophilia B, those are more on the clinical operational side and -- or tapping into the commercial-scale CGMP manufacturing capabilities. So we do believe, over the course of the next year or 2 that we'll be able to drive all these various efforts forward and do them expeditiously.

Okay. So we're going to move on now to the next part of our program. We want to thank everyone in our research group for their participation and their presentations. At this time now, I'd like to have Dr. David Cooper rejoin us as he provides an analysis of the HOPE-B 52-week data. David?

Thank you, Maria. So I'm happy to be able to discuss now the HOPE-B 52-week analysis as well as the feedback we've received from regulatory authorities. So first, just to remind you about the HOPE-B study and its design. This is an open-label study of male adults, greater than or equal to 18 years of age with severe or moderately severe hemophilia B. So that's Factor IX of these of less than or equal to 2% who have been on continuous prophylaxis for than 2 months. The exclusion is other factors that might affect the efficacy and safety, including a history of Factor IX inhibitors, active hepatitis B and C, uncontrolled HIV infection. We measured anti-AAV5 neutralizing antibodies at screening, but we did not use this as an inclusion criteria. We were not using prophylactic immunosuppression. So this study had a screening visit. It had a lead-in phase of at least 6 months where the patients were seeing every 2 months, had phone calls on alternating months. And it's important again that this was at least 6 months period. The dosing visit occurred and patients got a single dose of 2x10^13 gcs per kilo. They were followed up weekly for the first 12 weeks, monthly for the next 12 months and then every 6 months. And of note, there's the first visit in that longer-term follow-up is at month 18, and that will be relevant when I discuss the FDA feedback. So we've been communicating in terms of the study having co-primary endpoints. We provided based on the prior agency feedback proposal for statistical analysis planned in the fall that included these co-primary endpoints. The FDA has recently provided feedback on the statistical analysis plan, again, with no review of data at this point in time. What the FDA has commented now is that they think there should be a single primary efficacy endpoint instead of co-primary endpoints that the annualized bleeding rate, or ABR, should be the primary efficacy measure for gene therapies. ABR should be assessed after all subjects have a stable Factor IX expression, and the analysis should account all bleeding episodes reported by the patient, not considering any investigator-adjudicated new or true bleeding events. So with this, that would change the primary endpoint to be 52-week ABR after stable Factor IX expression has been achieved compared to ABR in the lead-in period. So this means we will measure this now from week 26 or month 6 through month 18. And again, the importance of the last slide I showed were the first visit in that 6-month follow-up is at month 18. The secondary endpoints would then be Factor IX activity at 6, 12 and 18 months after dosing. The rates of total, spontaneous, traumatic and Factor IX treated and untreated bleeding episodes. Factor IX consumption compared with the lead-in period and the correlation of Factor IX activity levels with preexisting anti-AAV5 neutralizing antibody titers. In addition, we'll, of course, have safety endpoints. So to remind you, there were 75 subjects that were screened, 67 that entered the lead-in phase and 54 subjects that were dosed. We have now, since hearing from the FDA, closed the 52-week data set and locked it with preliminary analysis. So for the 54 patients, to remind you, their mean age is 41.5 with a range of 19 to 75 years of age. The majority of them, 82% or 44, has severe Factor IX deficiency, less than 1% levels. And importantly, the prescreening treatment was almost 60% of them or 31 out of the 54 were on extended half-life prophylaxis. Preexisting anti-AAV5 neutralizing antibodies, we're seeing in 23 of the patients or approximately 43%. In terms of what we provided previously at 26 weeks, their factor activity has increased to meaning 39% at month 6, representing a change from baseline of 37.8%. We see at 52 weeks now factor activity levels are maintained at 41.5% at month 12, a change of 40.3% from baseline. You can see in the figure on the right, where the mild hemophilia range is from 5% to 40%, where we were just marginally at that and now the mean is pushed into the non-hemophilia factor range. So we're very happy it's showing excellent stability of our data in 52 weeks. In terms of the neutralizing antibody effect, we haven't seen a clinically significant impact of preexisting neutralizing antibodies on the mean activity. We're now graphing that based on the mean activity of each patient in Factor IX levels between months 6 and 12 as opposed to just at a 6-month data point, which you can see on the left of this figure is all the patients who don't have the neutralizing antibodies at baseline. And again, there's a variability in the response that those patients have. And on the right, as you move across, our patients with increasing titers of antibodies, again, you can see that there's not really a very consistent response and certainly not a clinically meaningful response. The patient with the second highest antibody titer of 678 has a level that's just around 40%. So certainly, the titers themselves are not predicting what someone would happen in an activity level. In terms of bleeding, sustained Factor IX levels that we've seen were associated with significantly reduced spontaneous bleeding during their first year of follow-up. And I think it's important to remember here that the FDA new guidance is having to show all bleeding events even if they've been adjudicated by the investigator not to be a bleed or a continuing bleed, meaning, even if they did an ultrasound and found out that there was no blood in the joint, because the patient reported it, it has to believe we need to count it as a bleed. So the bleed numbers are different than they were in prior disclosures. In terms looking at that from a lead-in month 0 to 6 and month 7 to 12 basis, overall, there were 136 bleeds in the lead-in period versus 29 in the first 6 months and 26 in the next 6 months. It's important to recall, I said the lead-in was at least 6 months. The mean period of the lead-in is approximately -- is over 7 months on average. For month 0 to 6, we also excluded the first 3 weeks. So that's really only a 23-week period. So you really can't compare absolute numbers of bleeds. In terms of Factor IX treated bleeds, there were only 15 Factor IX treated bleeds in the first 6 months and 14 in the next 6 months compared to 118 in that lead-in period. Spontaneous bleeds with -- treated with Factor IX, again, was significantly dropped even in month 0 to 6 but, again, went from 6 to 2 in the next 6 months after that. So a significant reduction in the spontaneous treated bleeds as well as all bleeds compared to the pretreatment period. This shows the annualized bleeding rates, which were reduced on treatment compared with the lead-in period where they were on state-of-the-art prophylaxis with factor concentrates. During this post-treatment period, we have up to 2 years of follow-up in some of the subjects. 30% -- 30 of the subjects or 56% had no bleeds, and 39 subjects or 72.2% had no bleeds treated with Factor IX. So this shows the ABRs for lead-in, the ABRs for the first year of treatment, the percent reduction and the significance. So all bleeds were reduced by 67%. Factor IX-treated bleeds were reduced by 79.9%. Spontaneous bleeds treated with Factor IX reduced by almost 85%. Traumatic bleeds treated with Factor IX reduced by almost 83%. And joint bleeds treated with Factor IX reduced by 84%. And all those are significant to p less than 0.0001 with very low rates in year 1 for ABRs reflecting the few bleeds that occurred. We also, I said, looked at substantial reductions in Factor IX replacement as an endpoint compared with the lead-in period. At 52 weeks, 96% where 52 out of the 54 subjects discontinued routine prophylaxis and remain prophylaxis-free. In terms of Factor IX consumption, we looked at it 2 ways. One is the IUs per year or how many international units they use, which was reduced 96%, a reduction of some 250,000 IUs per year. We also saw a 96% reduction in the Factor IX rate of infusions per year with an adjusted mean that went from 73% -- 73 infusions per year to 3 infusions per year, and this includes prophylaxis to patients who remain on prophylaxis and again, significant to p less than 0.001 (sic) [ 0.0001 ]. In terms of post-treatment adverse events, 53 subjects had 408 adverse events, but 39 subjects only had 91 adverse events that were thought to be treatment related. The major treatment-related adverse events that we've been following are transaminitis. 9 subjects received corticosteroid treatment orally for transaminitis -- transaminase elevations. They were all discontinued on steroids prior to week 26. The Factor IX levels were preserved in the mild range, 8% to 39%. 7 subjects had infusion-related reactions at the time of dosing. One subject discontinued dosing midway through and received only about 10% dose and the infusions were completed successfully in the remaining 6. We haven't seen any inhibitors. We also haven't seen any statistical relationship between safety and preexisting neutralizing antibodies. The adverse events that are treatment-related in more than 5% of the population are listed here. They're very similar to what we saw in 26 weeks with minor changes in the order of frequency. So the conclusions from the very positive 52-week data that we've just locked and analyzed, mean Factor IX activity significantly increased to near normal levels of 26 weeks and was maintained at near normal levels at 52 weeks. The majority of the subjects in the study did not report bleeding after treatment. We saw significant reductions in bleeding and improvement in their ABR compared to their routine prophylaxis. 96% of the subjects discontinued prophylaxis, 96% reduction we saw both in Factor IX use and in the number of infusions. The most common safety findings were transaminase elevations requiring steroids and infusion reactions, supporting a very positive benefit/risk of treatment. Based on the FDA's feedback and on the statistical analysis plan, the final analysis will now be planned at 18 months with this 52-week period between months 6 and 18 to support the marketing authorization applications. With that, I will open up the next question-and-answer session.

Okay. Thank you, David. We do have a few questions that have come in. The first one from Paul Matteis at Stifel. If we could speak to the correlation or any correlation between Factor IX expression and steroids.

We don't necessarily see any correlation of Factor IX levels in steroids. We see a correlation of Factor IX levels with -- in the patients who had transaminitis. And when the transaminitis is addressed through the steroids, the factor levels stabilize. But we don't see specifically that the corticosteroids are impacting our assessment of Factor IX.

Kristen Kluska at Cantor Fitzgerald would like again for a reiteration on any correlation to factor activity and reported bleeds across those patients who experienced them. And also in light of the observations at month 6 and 12, what are our expectations for 18-month data?

So I'm happy to take that one also. I think our expectation is that spontaneous bleeds will again be reduced as we move further out into the follow-up period. I think probably the number of patients who are dosed with Factor IX for bleeding episodes will decline over time as patients and their doctors become more familiar with what their factor levels are and whether every minor trauma that you and I might experience and have some bruising whether all those still need treatment in a patient with hemophilia who now has an average level of 41% in the non-hemophilia range would need to be treated. In terms of further correlation with factor levels, I think we just would expect that the factor levels will continue stable over the next 6 months. And then I think, overall, we'll see either a stable ABR or declining ABR in the patient set that have sufficient levels.

Joe Schwartz from Leerink asks if there can be any commentary on what may have led to the FDA's change in preference for the final analysis to support registration of EtranaDez, if there's a threshold for ABR that they will consider necessary for approval.

So I can take that. So the FDA's change in the final analysis came before they saw any of our data. It seems to be -- well, it seems to be based on the idea that the best way of measuring the efficacy of gene therapy is to wait until the expression levels of the gene therapy have reached a stable level and that they don't want to include the time or soon after dosing while patients are reaching a steady state. I think that's what we can say. Is there a particular threshold? No, we don't know. But our assumption is that we've powered our study based on noninferiority to standard of care. These are patients that are under prophylactic -- that are getting prophylactic treatment, the state-of-the-art, basically the best treatment you can get today, and we're beating that.

Ellie Merle from UBS asks, what is the rationale for the distinction between the investigator-assessed versus the patient-reported bleeds? And how would we expect these to vary in the bleeds captured, if at all?

David, do you want to take that?

Sure. So the bleeds can be generally captured in 2 ways. The most -- the majority of bleeds are captured in patients with diaries, which is pretty standard in hemophilia trials. The investigators are asked to adjudicate those bleeds and report whether they think that the patient is actually reporting a bleed from their discussion with the patient or review with the materials and whether that bleed is indeed a new bleed or not the same bleed that's continuing over a period of days. So generally, the new and true bleeds, which we've reported before, as what we were considering our primary outcome, is a smaller subset that are basically bleeds that are within the patients' bleed lives that have then been discussed, assess sometimes clinically, sometimes by ultrasound, by the investigator and determined as to whether those actually represent events or not. So it's a smaller subset of the information that the patients are providing for the events.

Well, the total number of bleeds is around 15% to 17% higher than the new and true bleeds. And the FDA has asked us to capture all bleeds, not just new and true ones, I think because they want to take the perspective of the patient.

A couple of questions came in regarding BLA submission time lines. Luca Issi, how should we now think about filing time lines; and Yun Zhong, similarly, is it reasonable to expect that the BLA submission will now be in the first half of 2022?

Yes. The CSL is obviously going to be the party that's responsible for submitting the BLA. But after discussing this with CSL, we believe early 2022 is a reasonable date to expect submission.

Joe Thome from Cowen asked, do we expect that the EMA will have a similar requirement?

Very likely. The EMA and the FDA normally exchange information.

A question came in regarding, there seeming to be a weak correlation between factor activity and titer of pre-existing NAbs. If we were to draw a trend line, has the topic of preexisting neutralizing antibodies been included in the discussions with the FDA so far? That's from Yun Zhong with BTIG.

Yes, I can take that. So yes, we have been talking to the FDA for quite a number of years about the importance of neutralizing antibodies and what requirements it would have. There is a -- if you were to draw a trend line, there would be a very weak correlation, but the line would fit very poorly if you just look at it. So yes, there -- and there is likely an effect at the very high ranges, but that effect is not straightforward.

A question came in regarding whether we have initiated conversations with the EMA on the final analysis for registration.

We have had a -- yes, we've had a meeting with the EMA to discuss what would be required for the submission of EMA.

And going back to the data, is a refresher, a question from Gil Blum with Needham. Are there any specific details that we can provide context regarding the patients who not have the stronger responses to treatment?

David, do you want to take that?

I'm happy to take that. So as I said, there are 2 patients out of the 54 that remain on prophylaxis. One of those patients is the patient who only got a 10% dose because of the infusion reaction early on. So that patient only received a 10% dose and that patient has been on routine prophylaxis. The other outlier patient that remains on routine prophylaxis is the one with the neutralizing titer of 3,212, which is a pretty uncommon and a very high titer, which we would expect to pretty rarely see in the general population based on prevalent studies we've done on anti-AAV antibodies.

Yes. I mean this is an important point because the reality is that 52 out of 54 patients no longer require prophylaxis. And so really, the only 2 patients that we would argue have had a poor clinical response are the ones that David has mentioned. And so the -- with a mean in a non-hemophilia range with the highest patient really at 100% or slightly above 100% normal. Even the patients at the lower end of the range with the exception of those 2 are still driving a meaningful clinical benefit associated with the therapy.

I think to add to that also, Matt, I think it's important to realize that the patient with infusion reaction was very early in the study. And the fact that the other patients with infusion reactions were able to be able to successfully be managed and still be able to continue to get the full dose and benefit and be removed from prophylaxis suggests that, once approved, infusion reactions, we'll have some guidance around how to respond to an infusion reaction and likely that situation can be mitigated in the clinical setting.

We received a couple of questions from René Wouters at Kempen. The first is related to the patient who is still on prophylaxis. Do we expect that we would be able to redose that patient when that might become an option? And then the second question, if we can provide any additional color on how the observed frequency of bleeds compares to non-affected or non-hemophilia patients.

I can take the first question. So we are working very hard on establishing our redosing platform in the clinic. And patients who respond poorly would be good candidates for redosing. And I think that, that was going to have a meaningful impact on the commercial prospects for our candidates.

Happy to try and take the other part of that. I think the short answer is we really don't know. There haven't been a lot of studies that have looked at. If you think about what the common bleeding types reported in this treating group now are, it's mostly trauma. And that's, for example, a patient who dropped a log on his foot while out camping with the family. So I mean there are day-to-day traumas we all have. No one counts them. No one's really done an observational study of how much normally active adults have minor bumps and scrapes and joint injuries, et cetera. So it's kind of hard to know general population is. We know compared to the state-of-the-art prophylaxis that our lead-in period is very representative in largely EHL, extended half-life, Factor IX population. That ABR is very consistent with the rates in the clinical trials of extended half-life Factor IX products. So our lead-in group is very representative of the overall hemophilia-treated population on routine prophy, and we see a significant reduction from that in the bleeding rates in the patients treated with EtranaDez.

Another question. Do you think that the FDA guidance now will be applied uniformly across all hemophilia B gene therapy programs with a time to stable factor expression plus the 52 weeks?

Yes. We don't know what the FDA intends. I would say, yes, it seems as if that is the approach of the FDA, but of course, this is a question for them.

Yes. I mean the reality is there is nothing specific at all with our Factor IX kinetics. When you look across various therapies, there's nothing nuanced that was raised by the agency that would make us believe that we're the only company that they would be focused on this.

And a couple of questions, just clarifying questions again, Patrick Trucchio at H.C. Wainwright. Just confirming, does the change in the primary endpoint from co-primary endpoints to the single primary efficacy endpoint on bleeding impacts the powering of the trial at all?

No, it doesn't. As you saw, based on the fact that the significance was less than p 0.0001, the trial is very well powered on bleeds in addition to being, of course, extremely well powered on Factor IX levels. So the fact that we now have a single primary and the Factor IX is now a secondary as requested by the FDA really does not change the powering of the study.

And the original powering of the study was based on the ABR, noninferiority analysis of the ABR.

Another couple of quick questions as we begin to wrap up. Where do we believe the nearest competitor is with regard to completing their Phase III trials? Or how much of a lead does uniQure believe and CSL believe that we have with our program? And then related is, again, for clarifying, is the approvable endpoint at 12 months or 18 months?

The approvable endpoint is 52 weeks from 6 months. So it's at 18 months. How far ahead are we? It's difficult to know. We think that we're probably 6 months ahead. But it's very difficult to know how somebody else's program is going. And Pfizer has not been forthcoming with that information.

And the last question from Ellie Merle. Can we comment on the variability of Factor IX expression at 6 months versus at 12 months? With expression increasing overall, are you seeing a decrease in variability between patients at 12 months versus 6 months? Or is the range in factor expression level similar at both time points?

I'm happy to take that one. I think we're very early in this analysis, and certainly in the coming months or so, we'll have more data out. But I think the variability has been from the 6-month time period on all of the monthly data we've been looking at and now the every 6-month data, it has been very consistent over time.

Okay. And another quick question came in from Luca Issi at RBC. Just to clarify, can you talk about the mechanics of the statistical analysis at 18 months? Do you just need to show noninferiority versus the lead-in period?

That's correct. We need to show noninferiority in ABR versus the lead-in period.

And the very last question from Patrick Trucchio at H.C. Wainwright. Can we discuss the relative importance of the secondary endpoints from a regulatory perspective also, though, from the patient clinician and commercial payer perspective of the importance of those secondary endpoints?

David, do you want to take that?

Sure. We've certainly had many interactions with regulatory authorities over the past several years. We've also had preliminary interactions with other stakeholders mentioned. I think certainly, ABR is a primary concern. Factor IX levels are a secondary consideration. But I think all of the other things, including factory utilization being off of -- remaining off of continuous prophylaxis and how that ultimately impacts quality of life, will be things that various stakeholders and technology assessments will look at in addition to the primary endpoint of ABR. And I think for the patients, some of those are actually very much their primary concerns as to how their life will change with a potential new therapy.

Thank you, David. Thank you, Ricardo. And now at this time, I'd like to turn the program back over to Matt for our closing remarks.

Thanks, Maria, and a sincere thanks to all of you who stuck with us over the last 3 hours or so. In closing, I'd like to review some important key messages from today's presentations. First, our focus areas are clear, CNS, liver and heart and muscle, and they are supported by what we believe is the leading AAV engine in the field. Now with strong 52-week data to support our hemophilia B gene therapy, we strongly believe our product has been clinically demonstrated to be best and potentially first-in-class, and we remain confident that with CSL as a partner, we are poised to truly transform care for patients living with hemophilia B. Our program in Huntington's represents a very significant opportunity in a potentially best-in-class approach that will drive strong data flow from 2 Phase I/II studies in the U.S. and EU over the next 2 years, beginning with initial data on 4 patients at the end of this year. We have a line of sight to $1 billion of capital and cash runway that enable us to reimagine our pipeline and continue to invest in manufacturing capabilities in our technology platform. We've initiated plans to transform our clinical pipeline, including 4 new programs with the goal of having 1 to 2 new INDs per year and 8 to 12 clinical/commercial programs by 2026. These programs include larger indications that represent significant unmet needs and commercial opportunities such as Corlieve's lead program targeting temporal low epilepsy, which impacts hundreds of thousands of patients. We have established and are developing novel technology platforms that enable us to optimally address new indications with differentiated approaches, such as miQURE, GoQURE and AbQURE for vectorized antibodies, and we are significantly expanding our manufacturing capabilities to support commercialization within hemophilia B to expand our pipeline and to facilitate larger indications in a highly cost-effective manner, with the construction of a second manufacturing facility in Amsterdam underway and expected to come online in 2022. As I mentioned in my opening remarks, I've really never been more excited about the future of uniQure, our ability to drive value for our shareholders and the potential to transform the lives of many thousands of patients. And with that, I want to thank all of you again for your time and attention today. And we very much look forward to providing you further updates on progress in the new future. So long.