Ultragenyx Pharmaceutical Inc. (RARE) Earnings Call Transcript & Summary

Welcome, everyone, to the 44th Annual JPMorgan Healthcare Conference. My name is Anupam Rama. I'm one of the senior biotech analysts here at JPMorgan. I'm joined by my squad, Rati Pinge, Joyce Zhou and Priyanka Grover. Our next presenting company is Ultragenyx. And presenting on behalf of the company, we have CEO, Emil Kakkis. Emil?

Thank you, Anupam. I'm certainly happy to be here at the JPMorgan conference again to give you an update on Ultragenyx. This is our forward-looking statement. We're a company that's leading the future of rare disease medicine. That means the company works on first-ever treatments and diseases have never seen a treatment before, developing new ways of measuring, studying and developing drugs for these treatments. And we try to help not only ourselves, but other companies as well as patient foundations, and it seems like everyone is doing rare disease drug development these days. We want them all to succeed, and we continue to drive forward in a number of different programs. Our differentiated strategy is really looking at potent biology in a bad disease and really picking well for those situations and picking the best mode for each disease. And then we try to be very creative in how we do development, looking at adaptive trial designs, endpoint choices and so forth. And it's often necessary to invent things and adapt to new diseases that have never been studied before. And finally, we commercialize. We believe in a very lean commercial model focused on patient find and patient support. We also have a very, I think, powerful method in post-marketing which allows to reduce R&D spend and allows us to turn over our programs and our teams over to the next pipeline opportunities. Now since going public in 2014, we have 4 commercial products approved, commercializing in more than 30 countries around the world and generating revenue that continues to grow, and we'll talk about it a little more. But we've also been able to turn over our team to 5 Phase II/III programs at the same time with 2 approvals coming this year. And that process is a part of how we're optimizing the rare disease drug development model in terms of trial design as well as in how we manage the post-marketing commercial launch. Now we've had some data on setrusumab recently, which was very disappointing. We missed in the trials, and I'm going to show you a little more about that data to get to explain what happened and what we know about our setrusumab program. So we had 2 randomized studies, and I'll go through these studies in detail, neither hit their primary endpoint, but both studies demonstrated substantial improvements in bone mineral density as expected from the Phase II studies. Additionally, we found improvements in vertebral fractures and the consequential effect of that, which is improvement in patient-reported pain and functional outcomes, which is what we also saw in Phase II, kids getting up, running around and doing better. So we saw data here, which we'll show you now about functional outcomes that I think are a real important part of what we see. And we'll try to understand a little more about the fractures, what we see, how it's working. And I'll hopefully give you a little bit of better understanding what went on in these studies because they are -- were so important to us in our evaluation. So I'm going to go into a little bit more, let's say, trial detail than a normal presentation because this is the moment when we need to do it. So there were 2 randomized Phase III studies. The Orbit Phase III was a placebo-controlled trial in patients 5 to 25 years old. This is the trial on the left, a traditional randomized trial, placebo-controlled. On the right is Cosmic, which is a different trial in 2- to 7-year-olds that is active controlled. Cosmic was intended to look at younger patients with more severe disease. and they need to be controlled with active. They could not be put in placebo for that reason. So this is the 2 designs. If you look at the patients that we enrolled in the 2 designs, in Orbit, they were generally balanced, but there were more type 3s, which are severe in the setrusumab arm and more type 4s in the placebo arm. And if you look at Cosmic, they're relatively well balanced. So there were 64% of the more severe type versus 54. So a little bit enriched in more severe type within the active controlled study. If you look at the pediatric subset, there are 67 patients in Orbit that were pediatric and similarly to the 69 in the other studies. So there's a significant overlap in the what I would call the prepubertal population of pediatric patients, and peds and teens together about 85%. So the majority of the study is less than 18-year-old pediatric. And this will become important as we go through the data. So who did we enroll in the studies? So when you look at the fractures in Orbit, the mean was about 3.2 in both groups and placebo and setrusumab had about a median of 2. So we enrolled the population we expected with a minimum threshold of 1. In the Cosmic study, we expected more severe disease, and we saw patients with higher mean fractures. And setrusumab had a higher median of around 4 fractures. So a more severe population in Cosmic, which is what we expect. And if you look at the stratification with fracture number, it was clearly shifted more severe in the Cosmic study than in Orbit. And this probably becomes important when we start looking at fracture reduction. Now in Orbit, because it was placebo controlled, we had to include a rescue arm because patients did not want to enroll in a trial that could go as long as 2 years, be off their bisphosphonate treatments. So if they hit a certain number of fractures, they could exit the study; 19%, 19.5% did exit the study. The placebo, though exited with type 3, type 4 patients, essentially, as you would expect, at nearly twice the rate of what the treated patients. So there's clearly a differential between which patients came out. Now their AFR would be calculated from their first 12 months in for those patients that exited. So you could still calculate AFR, though it might affect other endpoints that would happen at other points. Cosmic, there were no rescue criteria because both arms had treatment. So it's important staging of what we see. Now when you get into the trial itself, what we saw is substantial improvement in bone mineral density in the Orbit study, essentially 1 Z-score change, exactly what we saw in Phase II. So the drug is working, and you can see very tight error bars working as you would expect. On the Cosmic side, a substantially better improvement of bone mineral density over bisphosphonates, clearly working better. And we believe the type of bone mineral density you make with setrusumab will be a better quality of bone because you're making bone, not just preventing resorption. So the bone mineral density data replicated exactly what we expected. Now if we dive in the Orbit fracture story, this is where things started to change. If you look on the right, you can see the cumulative fracture distribution curves, and you can see the groups are not really separating whichever way you look at it without the primary endpoint, without vertebral fractures and fingers and skull and the others where you look at total fractures, very similar. And because the story comes really -- if you look at the placebo group, based on the negative binomial model, it was only a 0.55 fracture rate estimated. But the median, if you look above is 0, which means of 52 patients, 26 patients had no fractures in the study. So this gives you very little power now to tell the difference. The fracture frequency in the UX143 group was actually fine, 0.71. It was not very high, considering what the pretreatment fractures were at. So both study arms were lower. The placebo is slightly favored, but that's only with the primary endpoint, looking at all fractures are about the same. So the question is what's going on here in the study with fractures. This was the part that was a surprise. Now we know from the Phase II study patients felt better and got more active. So one of the question is did they get more active and cause more fractures, that we won't know for sure, but we did look at how they did, and Orbit is a placebo-controlled trial. And the POSNA-PODCI is a pediatric orthopedic assessment. If you look at the POSNA-PODCI graph, you can see the setrusumab-treated patients had a statistically significant improvement in pain comfort versus placebo. And in sports activity, also a strong trend for improvement. The other scale, the Patient Global Impression Scale also showed in the same peds plus teen subset that statistics have an improvement in bone OI pain and a statistically significant improvement in daily activities. So both in pain and functional activity in both -- 2 different measures in a placebo-controlled trial are showing the drug is working. So you might wonder, well, if the fractures aren't better, why are they doing better? I think it could be because they're doing better and they're running around. And we did see that in Phase II that some kids had fractures, but we still should have seen some differentiation, we didn't. But this is the Orbit story, we think this shows an activity, but we didn't show the fracture differential. So now let's talk about Cosmic. In Cosmic, the story is a little different. The fracture rate was higher. If you look at the mean or median, you could see there was 2.6 was the mean for the bisphosphonate group. So several fold more fractures going on in the study. If you look at the cumulative distribution graphs on the right, you can see that they're separating quickly, right, after initiation of trial. There's a little turn up in the graph that happens to be when we do a 12-month skeletal survey. That's why a bunch of fractures are added because we see the fractures on a survey that weren't seen before and they're added. But you can see the separation of the 2 lines gets even greater. So this is separating in the way we would have thought, right? This is what we were expecting. When you do the negative binomial model, it shows a 21% difference, but still not significant, but it's -- the pattern looks more like what we would have expected to see. Now when we look a little deeper at these fractures, what we saw is this. You can see on the left of the table here, it's just the total number of fractures seen. And you can see there's less fractures in the 143 group. But if you look at the vertebral factor, there's a very substantial difference, about a 59% reduction in vertebral fractures. And beyond vertebral, if you look at all of the trabecular bone including the scapula and also some of the fingers and feet that have more trabecular bone, all of those had the 59% reduction. So clearly, the bones that have created the most bone mineral density are having the best effect. The long bone difference then there wasn't as much. The long bones may take longer, perhaps. But we see a very substantial effect. This is still a trend P-value 0.08 at this point. If you look at the fracture, they are non-morphometric in the spine, you could see a 94% reduction. These are true fractures of the vertebrae, are essentially getting eliminated. So we think those are quite important. But this is what we see, did see an activity of the drug that bone mineral density improvement is resulting in improvement in trabecular bone, particularly vertebral fractures. So when you look at safety in the program, the safety was good. There were only a couple of patients in each arm that came out from safety in Orbit. So relatively no problems in safety. Safety was fine for -- in both studies. So what do we see going on here. Overall, we think the data suggested the drug is active. It's clearly improving bone mineral density in a profound way, in a way we think is the best. How is that translating? Well, the bones that are trabecular are clearly having a reduction in fractures. The long bone is less so. And this is the pattern we're seeing. But those changes are resulting in kids that have less bone pain or more active. And we've clearly seen a number of cases of kids that don't walk well. They are walking much better and getting about. So we think the drug is clearly active in this disease. But the question is -- the real question is why the AFR missed, and that's the challenge we have to deal with. We continue to do more work looking at the data, but this gives you an idea of what we've seen. I think it replicates a lot of what we saw in Phase II, but again, the AFR did not hit what it need to hit, and we'll have to continue looking at all the data to understand it better, but hopefully that gives you a sense of where we're at with setrusumab at this point in time. So let's talk about what else is happening. It's a company in the pipeline. Sanfilippo syndrome, a very horrible disease, one of the worst progressive CNS debilitating disease. And we've been developing a gene therapy that was given to us by Abeona. And the data we've seen on it shows a biomarker restoration of lysomal function in the brain combined with a substantial and persuasive statistically significant improvement over natural history for these patients in multiple neurologic endpoints. The data were strong. We received the CRL last year for CMC issues, which were very resolvable. We are resolving them. We expect to file early this year, and we would expect to be able to get approval based on the data we have and what we've shown in this program. Now one of the questions in gene therapy is are these going to exceed because I think people think gene therapies have failed. Some have done well. The entire success of gene therapy is highly predicted by one key factor. What is the level of patient urgency. If you know patient urgency, is there another treatment? Is there no other treatment? Are these kids going to die or have risk of dying or have substantial impact? Sanfilippo, these kids are going to die and the parents know it. This is a situation of a high urgency. We believe this will be one where patients will want to get treated, and our goal will be to make it accessible to as many of these patients as possible. We'll work through that filing and look for an approval, hopefully, around the middle of the year. The review in a refile is about 6 months. Second program, we already have completed a BLA filing in December is for DTX401 for glycogen storage disease. This is also an urgent see. These kids are every few hours eating starch to keep their glucoses up from falling. If they miss starch at night, they can die. It is running on a treadmill, an emotional and stressful treadmill all their life. When we enrolled these studies, we had great demand to enroll in the studies. We have -- when we had to cut off the enrollment, we had people very upset. There's a lot of demand in this one as well, urgent. We've shown a randomized controlled data, which were excellent with reduction in cornstarch need and that continues to improve with time. But it's also associated with excellent impression scale scores in these patients and particularly 95% of the crossover patients showed that they had improved. We think this treatment will do well. We think the urgency is high. And it's really the first time they've had a treatment that truly corrects the underlying problem of their liver disease. That filing is in, and we'd expect assuming we move through validation, we'll get through to a PDUFA date would be about 8 months' time from filing. GTX-102 is the other major event of the year. Angelman syndrome, we have a very potent antisense oligonucleotide that turns on UBE3A expression. We've had data on over 60 patients in Phase II showing profound developmental improvements, improvements in cognition, receptive communication, sleep, motor, fine motor, and the Phase III trial enrolled very quickly and finished enrollment in July last year. It's a 48-week study. So we'd expect in the second half of this year to be the Phase III data come out. Now we're doing 2 studies. The first study on the left is a placebo -- I'm sorry, a sham controlled but blind -- double-blind study, 129 patients, ages 4 to 18, the same age range that we used in Phase II. So we feel we understand that population. The primary endpoint is the Bayley score, which is a validated neurologic score. But we're also supporting that with applying about 20% of the power toward a Multi-domain Responder Index, which is what we believe should be the future of neurologic disease evaluation. Neuro is often harder because it's complex, it's variable. The multi-domain allows us to capture 5 domains of effect and look for major changes in 5 domains and accumulate that efficacy. And therefore, heterogeneity is managed, and you're able to find the improvements where they exist in patients. And we've shown in the Phase II study data that the MDRI is quite powerful and important. So those are the major parts of that development program. We've added now another study called Aurora, and this is to expand exposure to other genotypes and ages, a number of different ones I won't go through today, but the idea is to help expand who can be treated safely and what does their efficacy look like. The randomized trial is obviously the important driver for a pivotal file for a filing. So the second half of this year is what we'd expect. Finally, I'll touch on Wilson disease, the gene therapy. We've shown some very interesting early results. We wanted to increase the dose to get to a dose level where we could essentially eliminate the need for other treatments in Wilson disease with this gene therapy. And while there's always a question about how many patients would need a gene therapy versus their chelators. We're now looking at some of the data from the biopsies. What I've been shocked to see is how many of the well-controlled Wilson patients have tremendous liver injury going on, which make me believe that maybe the chelators are not as good as people think. That maybe a larger fraction of patients should be on it. And we presented at a liver disease meeting last year -- late last year that some of the patients with liver inflammation did have a decline in liver inflammation with the proper canalicular expression of the transporter that we are delivering by the gene therapy. That was encouraging, where Cohort 4, which is at a higher dose with improved immune modulation as will be coming out later this year. So Wilson, I think, is a great prospect for changing the future of a disease with a truly specific treatment. Now with all the commercial -- with all the clinical activities and catalysts, we've also been building a global commercial business. We just announced our -- preannounced our numbers, which are we beat our consensus with expected $62 million (sic) [ $ 672 million ] to $674 million, which is a 20% growth rate. And you can see we've been steadily growing our commercial business in these 4 products globally, adding countries and growing each country. I think this is a strong base business for the company and a base value point. But we'll also be able to leverage our commercial global experience as these new products come forward and be able to launch through this existing commercial franchise, which is part of our original vision, that in rare, to really truly succeed, you need to be able to capture revenue across the world, not just in one territory or another. You need to capture rare across all territories, if possible. We're going well, and I feel good about the growth of this base business going forward. Now with the setrusumab challenge, we are planning expense of -- significant expense and headcount reductions. With that in the Sanfilippo CRL that we received, both of them provide a challenge to our financials, but with the strong revenue, double-digit growth we're having, combined with some expense and head count reduction we'll have to make, we will invest, of course, in the launches that we're planning. But with these expense reductions, will put us in line to continue our goal, which is heading to our profitability in 2027. And we expect -- we have $735 million in cash, plus we expect 2 PRVs will occur before the sunset clause in September, which allow us to be able to bolster our cash position. Now the PRV should get reauthorized, but if it's not, we still have a chance to get these 2 PRVs in time. Together with our expense reduction, we put us on path to get to profitability. That strong base business then is giving us the engine to get where we need to go, 2 gene therapy potential launches, the Angelman program also coming up, Wilson and whatever we end up doing with setrusumab going forward. So we're excited about what we've been able to do as a company. We'll get through the certain current challenges, but these are the catalysts for the year. Two gene therapies, which would come on around the middle of the year, Wilson data later in the first half and Angelman data in the second half of this year. So it should be a rich year of both regulatory and clinical milestones. So we've had a history of consistent and strong revenue growth with our base business. We have a number of catalysts in our programs. And I think with the revenue growth, the new launches planned, we are heading to profitability in 2027, which is, of course, a rare place to get in any biotech company. So I'm proud of the work that the team has done to get us there and whatever challenges we have, we will overcome them and move forward. And I'm excited about the prospects of how much we have going on. I mean, 2 BLAs in 1 year, plus other programs plus growing commercial revenue. It's a strong story for Ultragenyx. Thank you. So now Anupam, questions.

I'll start out with the first couple of questions. But if there are questions in the audience, feel free to raise your hand and ask, happy to call on you. Emil, I was wondering if you could put the data that you just presented on setrusumab into context of what's kind of most important to patients, right? You hit stat sig on pain in both studies. Some of our diligence suggests that pain is one of the most important thing to these patients. Like how do we put all the new data into context?

Well, the pain data come from the measures from the same Orbit placebo-controlled study because the other study was open label, so you can't really do pain in an open label, but in a placebo-controlled trial with 2 different pain and with both hit. And I would say having studied it here and there, it's very hard to hit pain. So that's impressive to us. But when you ask patients, they have recently created a list, the OIF -- OI Foundation. Top on the list is bone pain. The fourth in the list is fractures. So that -- their view is that the pain, the functional activities and other things are actually more important than fractures. But the truth is, from a regulatory standpoint, the fractures are what was stated. So we have to now adjudicate between those 2 realities. We have an effect on bone pain and functional, which is important to patients. We had an agreement or understanding with AFR. So now we have to navigate in that world.

How do you think about next regulatory steps now that you have the totality of this data? Like is there a path forward, you believe, even if it's in a subset of patients such as pediatrics?

I think right now, our most important thing is to really understand more about the data and why it is the way it is. Long bone did not appear better. Is that because kids were more active and they fractured? Or is there something else going on there? So we need to understand all of that to understand it better. And then we need to do that before we end up going. If we go to the agency, we'll want to do that with a firm understanding of what we have in front of us. And I think there's a little more work to do before we make that move.

Questions from the audience? We've got an e-mail portal question here. It's more broad in nature, but would you consider divesting your viral vector manufacturing to improve your profitability?

Well, the truth is if you're going to be in the gene therapy area, it's better to retain that from a cost structure standpoint. And we're one of the usual companies because we don't have just one product. We actually have 2 products that were approved. We have others in development. So I actually think we're someone that can leverage a plant better than most. And I think at this point, it doesn't make sense. The cost markup and the reliability of contract manufacturing or gene therapy has been, I would say, problematic. And I think right now, we're better off not doing that. We've made the investment. It's behind us. We're now ready to turn the plant forward. I think it's much smarter to keep to the plan and manage our other expenses rather than the plant.

Maybe just going back to setrusumab now. You've presented us with these additional assessment since the top line. What are the additional assessments that you're hoping to conduct here in the next few months that could give you a better understanding of what happened, but also what is the potential of this product?

Well, we need to understand the various components of the fractures, why they happen, are they associated with trauma or not with trauma in the different subsets and different ages. So we kind of need to really understand that relationship well. The vertebral fractures improvement we're seeing is fundamental, and it's something that certainly was important to me, and I had talked about before, I think, Anupam, is that the vertebral fractures in type 3s and 4s are what caused them to become disabled and wheelchair-bound. And if you could prevent that from happening, it's an important thing, right? There's no question that's important. And I've talked about that specifically before, it's why we ran Cosmic. We ran Cosmic because I wanted to treat those little kids with type 3 and 4 to see if we could change that future. So there is something there to deal with, but we just have to figure out what does that mean from a regulatory standpoint. But we want to go in with the full knowledge of what we have right now, and that's what we're going to work on, understanding it fully, to appreciate why the vertebral fractures and trabecular are better. It makes sense. It's probably because bone mineral density improvement is faster in the trabecular bone. But what happens in the long bones, what happens in the cortical bones, is it -- will it take more time? Is there some way to look through the data and understand it? I think that will be important in trying to put forth a proposal once we understand all of that. So that's what we have to figure out. I think the symptom stuff is clearly showing that patients are feeling better on treatment. So I think that's a strong positive and important to patients. Certainly, the patients want to stay treated that we know.

Questions from the audience? Yes. I can repeat the questions.

Just a question on bone and cartilage genetic diseases in general as a therapeutic area? Are you still excited about that?

I think there are some many genetic bone and cartilage disease that don't have treatment. So we think it's a very strong area for development. And while most people shied away from bone and cartilage because getting drugs into the bone and cartilage is hard. What we're seeing is a lot of ways for us to manage bone and cartilage without necessarily have to get into the bone exactly like with XLH in controlling phosphate or in this case, controlling bone regulation with setrusumab rather than fixing the defect in collagen, we're actually just fixing the dysregulation that occurs that leads to not enough bone being made. So I think these are good areas for further work, and we are committed to working in the bone space for that reason.

I think we have a question over here.

Will we expect more data from setrusumab, like a long-term data -- long-term follow-up data?

Well, I would expect -- we have not a plan yet, but I would expect as we learn more about the data, just as we presented today, we'll provide an update on that data both longer term but also further deeper understanding of what we've done so far. So we're not done yet. We're going to be working that. But despite doing that, we are going to manage expenses and manage the burn, too. So we're not going to wait for that. We're going to be doing that because we need to. But we'll continue to work through and figure out what is the path for setrusumab, and we'll put out more data when we have more information on it.

There is a question over here.

Just curious, did you report the BMD gains for more cortical rich sites like femoral, neck or total hip?

We didn't report them out. We haven't studied them. In the original ASTEROID study, they did study all the bones. So from the ASTEROID study, and these are in adults, the lumbar spine measurements was about 10.9% in a year. But when you look at the cortical bone, it was about 2% to 3% range. Now the problem with cortical bone is that 2% to 3% means a lot because it's just a small tube of bone more will strengthen the bone. So you can't quite equate those numbers. With trabecular bone though, there is more surface area. So there's more places for bone cells to lay down more bone and maybe make more bone more rapidly, whereas in the tube of a cortical bone, you're dealing with one surface area of the tube and laying down bone. So it's possible it takes more time in the cartilage bone, but we don't have that true answer yet, but there is bone production in both types of bone, as you would expect. But trabecular bone is clearly more responsive and what we're seeing now is, let's say, a bigger effect in the trabecular bone.

A question in the audience.

Did you check the biomarker of formation and resorption, like P1NP and CTx on both trials?

We did look at the P1NP. I don't know about Cosmic, but in Orbit, we did P1NP, and of course, it increases as expected. You couldn't get the BMD improvement without P1NP being laying down. So it correlated the way you would think. So there is more P1NP and that makes more bone mineral density. But bone mineral density is the real end game. Think of it as the accumulation. P1NP is a transient measure of how much bone is being made today, but the BMD is accumulation of the net total bone. It's a better marker ultimately in showing what you're doing, but we did look at P1NP, and it gets -- it improves, increases.

Additional questions? I'm going to just switch gears now a little bit to GTX-102 in Angelman. Emil, you've had a wealth of data presented over the last several years, I think, maybe 4 or 5 updates, and it's kind of matured over time. If you had to highlight what are the couple of data points we should be thinking about in the totality of that data give you confidence going into Phase III with the Bayley-4 endpoint. What would they be and why?

Well, I think at ASF, we put forth the biggest summary of all the data on the Bayley there, and we showed a meaningful change in Bayley that you don't see in control patients, whether it's from the natural history or from the levodopa controlled study. There's less than 1 point change in the Bayley, and these kids really don't change in natural history. They're just flat. So to see 5 points improvement in the Bayley, which is a clinical meaningful threshold is, I think, something that's quite important and very difficult to do. And even less than 5 points is probably meaningful, but it's a continuous variable analysis for the Bayley. So we think that the control should be relatively flat. And it is though an international study. So we always have to consider when you run international studies, there's more complexity to that. So the Bayley, I think, is important. The cognition, by the way, is supported by the fact that patients have better attention span and better receptive communication, which makes it easier for them to deal with the Bayley cognitive testing. I'd still go back to the Multi-domain Responder though, because the data we've shown there has been p-value less than 0.001, even with just 13 patients earlier or later with the larger group of patients, it's very powerful. And with the heat map, you can kind of see how patients are getting better and majority of patients have 2 to 5 domains of improvement and in multiple endpoints. And I think it tells you the breadth of the treatment effect in a way that's more relevant to what patients want. The Bayley is a somewhat arcane tool that has regulatory standing, but truthfully, if I told you 5 points, most of you would know what 5 points mean, right? It's arcane. But if you tell someone that they are having clean improvement in their sleep or in their gross motor function, right? If we tell them -- those kind of things you see in the multi-domain, I think are a little more understandable. Now to support that further Anupam, we do have a development checklist that's closer to maybe what Taysha is doing with Rett, which is we're asking the question, have you hit these developmental milestones? Can you do these things? That development checklist we presented at one of the original Analyst Days where we showed data and you should go look at that because that tells you the meaning of things for patients. And look and see that the majority of patients who could not understand 1 or 2-step oral commands before are now able to understand the parent tells them a command and they can understand it orally. Normally, what parents do with Angelman syndrome is that they will demonstrate sitting to tell the kids to sit because it's like the kid doesn't speak your language, speaks no language. How do you tell someone what to do when you don't have language to share? So imagine now you can tell them sit down or stop, and they actually understand what that means. It's quite transformative for a family life to be able to do that and have the kid actually paying attention to you, first off, right? They actually are listening to you and responding to you. When you look at all the development checklist items that we put forth, you could see these are fundamental things that are quite important. So this is why our -- we have high hopes for this. Now we just have to navigate the regulatory waters of daily MDRI and the rest of it. But we're confident and we have a lot of patients continuing to get intrathecal injections every 3 months on a pump. You know that parents are not doing that for nothing, right? They're not going to keep doing that for nothing. So we have a lot of patients continue to get treated and doing well. So I think we're excited about the potential for Angelman syndrome treatment.

Any final questions from the audience? All right. Thank you so much, Emil.

Thank you all.