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

Okay. So now we have another patient and that patient has 2 videos. The first is a pre and post video of this patient eating again. And then the second one is of the patient walking. So let's go ahead and start the first video.

Okay. This is him eating before he was treated. And this is not selected video. This is what he looks like after every meal. He basically smears foot around, throws food, gets a little bit of in his mouth. If they want him to actually eat anything, they have to feed him. But this is the mess that it looks like. This family describes giving the child a bath after every meal. And this is real because when we have these patients in clinic, there's cheerios and fruit snacks and everything all over the place. So that's what eating looks like. It's disorganized and impulsive. And now here he is after treatment, and he's focused. He's thinking about making a motor plan. He's going to reach out, get the fork, put it in his mouth, eat the food off the fork and then give it back. And so that's a much more organized way of eating than what he was doing before. He wasn't able to grab the fork because his hands were too ataxic before he went on the drug. And then here he is, he's got food on his plate, but it's not getting thrown all over the place, and he's going to focus and get one chip out of the bag and eat it and then he'll go back in the bag. And he's actually able to just pick his food up off the plate now and eat it rather than making a giant mess and the family no longer has to give him a bath after every meal. So this is a substantial improvement in his eating, and we hope to see this continue to improve to the point where he's able to use the utensils independently.

Yes. Thank you for that. I saw improvement in both limb and truncal ataxia in this patient, ability to focus on motor planning, as you mentioned, and then the fine finger movement. So again, multiple domains really being impacted in one video. So I think the last video we have is of the same child walking pre and post drugs. Let's go ahead and play that one.

Okay. So here, he's going to be walking before treatment. You can see the broad base again. He needs to do that to maintain support because he's ataxic and the arms in the air kind of flailing around, but trying to maintain support. And then here he is after treatment where he's actually walking on a moving walkway. And you can see how much more narrow his gait base is. He doesn't need to keep his feet apart to maintain support and he can stop. Stopping and turning our times when Angelman kids fall, he stops, turns and walks backward on a moving walkway, which is very impressive for Angelman syndrome, and then he's going to turn around again and go the other direction. So this is really a big improvement in his stability and his balance and leads to decreased falls and increased ability for him to ambulate without the family having to take a stroller around. And you can see the change in his base after treatment. Again, something that you -- there's no placebo effect here. These kids don't even know to have a placebo effect, and this is a real motor finding that we're having.

So very similar to the first patient where you saw the wide base gait, ataxic gait, become narrower and smoother patient, this has implications in terms of the family needing to use either a walker or stroller, like you mentioned, implications on travel, right, for the families and then the safety aspect in terms of reduced falls.

Yes. Yes, we have families who don't even have to take the stroller on vacation anymore, which is very exciting for them.

Yes. That's very significant. Okay. So any last words in terms of the totality of the new skills that we've seen right now?

Yes. I think what I've seen has been amazing. I mean frankly, when we first dosed these kids in 2020, it was the middle of the pandemic. And these families were telling me like this drug is working, and it was only like a couple of weeks into the dose or 2. And I would be like, you guys are having a giant placebo effect, send me a video of that because I don't believe it. And then I would get these videos, and I was like, wow, because having worked in developmental disabilities forever, we expect changes in development to take a very long time. And this change in these Angelman kids, although they're not close to normal at this point, but it's very dramatic. And I can only hope that these changes build over years. I mean we have to remember, we're putting an ASO and correcting things genetically in a brain that can't learn as fast as a baby anymore. So the learning may progress over a very long time, and we may have to wait 10 years to really know how much these kids can be different from what's expected. But what we're seeing so far is honestly, more dramatic than anything I've seen in a developmental disorder with any treatment before.

Yes. Thank you for that summary, Liz. And I think the videos are a nice complement to the objective data that Emil presented to show you the totality of what we're seeing with GTX-102 in this trial. So again, I just want to thank you for everything you've done in the trial, your work in the trial and especially for treating patients and being sort of a guiding light for them. And of course, thank the patient families who provided these confidential videos and all the patients taking part in our trial and those taking part in other trials because we know how challenging that is, and we are all in for Angelman. So this concludes our section. Thank you.

Okay. Very good. Thank you. Dr. Olugemo and Dr. Berry-Kravis. So you've now seen the Angelman data, quantitative data and natural history comparison EG, clinical meaningfulness and also videos. Hopefully that gives you a sense of where we are and why we're confident about the fact we have an excellent drug in the GTX-102 and we're excited about the prospects going forward. So we've given you now talks on the OI data, Cetuximab, showing, I think, transformative effects on bone mineral density and fracture rate and some of the clinical insights and then we've given you the Angelman story as well. So we'll open the floor now to Q&A. Tazeen, I saw your hand first, and you get the first question.

Thanks, Emil. Tazeen Ahmad, Bank of America. So can we back up to the first presentation maybe for OI. I wanted to just explore the comments that were made by the physicians about parents being at some point, scared of putting their kids on the bisphosphonate. And do you have a sense of what percent of patients actually don't get bisphosphonate at all? And would they potentially be the ones that could be the first on borders in terms of treatment possibility in a real-world setting? And then the data that you've shown today was impressive and clearly shows that the earlier you start treatment, the better. How are you thinking about treatment for children as a result versus adults and what would be the benefit that adults in particular could see in addition to bone mineral density?

Yes. So on the question of bisphosphonate, we've been talking to a lot of people about it. It's rather interesting to see there are certain doctors have certain take in view. So some are treating everyone. Some are treating very few. It's actually highly variable between different doctors. So it's a little hard to say there's a decided number. A lot of the patients are in the bisphosphonates, but I can't tell you a precise number today. And honestly, clinic by clinic, we see a lot of variations. What I would say to you is that most of the time, what's happening, if the patients are having a lot of fractures, the bisphosphonate is the one thing they can offer them. But the fact that bisphosphonates, at least in the randomized studies has been where they've succeeded 2 out of 5 studies have been successful. There's about a 20% decrease in fracture. So it has some effect, it has some effect in some patients. But it's the only thing they have to offer. So I would say to you, it's -- it's not the majority, but it's a good fraction that are on bisphosphonates. And it's highly variable depending on the doctor in their view of whether it's helping them or whether it also can potentially bring up risks, as Dr. Gottesman was talking about, which is chronic use of bisphosphonate can actually create more problems in the bone in the long run. I don't think it's going to determine who wants to be on the drug. I think bisphosphonates will become relevant to it. I think the people are going to look for ways to make bone and whether you're on it or not. By the way, all the patients essentially in the trial have been on bisphosphonates in the past or a very large fraction. So we know that it works on top of whatever you've been getting in the past, although they're not on bisphosphonates in the trial. With regard to adolescent children, there is no question that children have a better effect, but I think it also have an important effect. And certainly, the adults in the ASTEROID study when we first started working with Mario, were contacting us wanted to get back on drug. I mean there was interesting impact on it. I just think that the time frame for improvement in bone mineral density and fracture Brazilians is -- takes longer. But I would point out to you that in the Orbit study, we are including patients 18 to 25, they are adults who are post pubertal, growth plates fused in the who have high fracture rates. So we will have data in the trial itself for adults, which will help support adult usage. I'm confident adults will need to get treated. I think from a standpoint of the speed of their effect, the pedes will be easier. And the trial, an orbit trial is going to be driven more off the pedes result, but the adults will be stratified and included in there and we'll have -- be able to speak to adult fracture reduction as well. Where do I go? Maybe Gena, I thought...

Well, there is one microphone in the back.

Oh, there is one back there. Okay. You are getting it. All right. I will let you pick.

Joel Beatty from Baird. What are the implications of today's data for the powering of the Phase III study for setrusumab? And also, are you targeting a certain amount of patients in the different age groups?

Yes. So do you want to talk about powering?

Yes. No, exactly. I think that's the important part with having access to the pediatric data coming out of Phase II. So again, the data we had available to power the Phase III part of that study was based on adults. So now with pediatric data understanding the higher fracture rate there, we really do think we're probably overpowered. We added 2 interim analyses, so we will have the opportunity to evaluate for that. So it's an event-driven study. So we're really looking at the separation of those 2 groups, those patients randomized to setrusumab versus on placebo with fracture rate. So we really do see a path for potential for an early win here.

Okay. Gena Wang from Barclays. So maybe Eric, follow up to his questions. So for these 2 interim analyses you're adding, you said that it's event drives. Can you share what are the event numbers that for the 2 interim analyses? And also regarding the age, do you have a set number of a patient that's 5 to 12 and 18 to 26 -- sorry, 12 to 18 and 18 to 26. So that's the first set of questions. Also, I have a second set of question, very quick on the GTX-102. So first, I wanted to say like this is, indeed, the videos are very impressive. I do have one question for Dr. Berry-Kravis. So from those videos, it's very powerful. And how -- from the clinical study perspective, how can we contextualize those change into quantifiable scores? And we did see many different scores, right, the Baileys. And today, we saw the first time AVA. And then what about the CGI? And if you wanted to, how much differences that could deem to be clinically meaningful?

Well, you start with the OI. I will deal with the GTX-102.

Yes. So for the Phase III, we're not requiring a set number of patients in each of those groups. We are focused on younger patients who are showing an emphasis here, but we're not having a set number of patients for each age group. But we are seeing a good number of young patients enrolling in this study, which is important. As Emil pointed out, we do want to enroll some adults. We do want a broad label here. We do think it's an important treatment for children and adults.

Interim timing.

Yes. The interim timing, so we're going to submit the new statistical analysis plan with the FDA, so we don't have full agreement there. We will be looking, like I said, those 2 interim analyses there, but I think we'll disclose more detail.

Yes. I think one of the things to note, the way the trial is designed is that the unblinding plan would be when there was 100% of the fractures required to inform the positive result based on the original power design. So it's a number of fractures the amount of time and it could anywhere be from 1 to 2 years of time exposure. What we'll do in the interim is have looks that are earlier than having 100%, right, less than 100% of the data so that will allow them to occur significantly earlier during the -- as the fractures are cumulating, but the precise date will not be known. It will be based on fractures. We won't discuss those interims if we're doing them either. They will be done. But the point would be if the power of the study is dramatically higher and we hit persuasive statistical significance early, we'd want to end early both for the ethical nature of treating kids with this disease as well as because we'd want to move it toward filing as promptly as possible. So that's on the interim timing. We would obviously want to stratify for age to make sure we're not having old in one another. So -- but let me talk about GTX-102. Now videos contextualizing -- videos are really supplementary information to look at the quantitative, but to create videos themselves are very difficult as end points, right? People have looked at trying to develop them, but it is a challenge. What we are looking at is a quantitative endpoint. We're using the videos to give you context to the quantitative. If we so -- show daily growth motor improving or the age and severity assessment of gross motor improving, we're saying here's the meaning of it in terms of what that means for videos. So I think it just adds color to it, but I wouldn't say to you that it's going to -- it gives us a sense that what we're seeing is clinically important, that is if we can acquire and demonstrate this level of change in a Phase III trial, that this is a treatment that parents are going to want for their kids, right, that this is a treatment that is important to them, that they can walk without falling and these are things they can accomplish. And if you look at the -- the reason we showed you that emerging developmental skills thing is we try to put it in the parents, that's the parent's perspective, what can my kid do that never did before, just to show you that a lot of the kids has 5 to 7 things even at this lower dose enrollment period of that trial. They had that many first-ever accomplishments. I think that's really the point of it. It's less about the clinical trial design that it is about deciding how important is this treatment. And that's why we brought it to you today more about the clinical means of it. So the Bailey is -- has the most historical data. It has -- it's quantitative. The Angelman severity assessments, I think, are much more angel-specific. And so for certain ones, they are better. We have other domains like the -- there's another apparent behavior scale, which we could potentially use. And we are looking at more end points than what I've shown you in the expansion. And our goal is to get a read from FDA what they think will be important and be able to go to them when we have actual expansion data and really nail down what that means. But the CGI is a possible way of going, but the CGI we've said many times is a relative scale. It's not an absolute scale. There is no clinical movements built into it. It is just a relative scale of response. And for that reason, it has, in my view, generally hadn't been accepted. They did accept it recently. It was accepted in the RET program, but not by itself. They had a behavior scale in there too. Remember, there was 2 things. So that means it wasn't standing alone. I would just say to you, I think CGI scores have a utility, but being able to move the Bailey to me is show real treatment effect because the Bailey is hard to move, right? Just go on, where is the next mic? Yigal has it. There's also one person up here.

Emil, Yigal Nochomovitz from Citi. I had 2 on OI and 2 on AS. On OI, could you give us a breakdown of the 4 patients that did have fractures. Can you say which cohorts they were in, in the age groups, 5 to 12, 12 to 18, 18 to 26. And then second, with regard to the Phase III end point, you showed essentially almost as good data with lumping in all the fractures, including the fingers, extremities, face and skull. So just -- can you just explain in a little more detail why it seems like the FDA is more insisting on just the fertility fractures and not including everything, but your data suggests it's as good. And then on AS, working the natural history, could you just comment on how the 64 patients were selected. Was that done with some sort of propensity match score or was it hand curated? And then finally, for Dr. Berry-Kravis, you mentioned that the kids don't even know about a placebo effect. So I'm just wondering to what extent they even understand that they're on a therapy for their disease. If I need to repeat those, I will, but those are for me.

I got them. So you wanted to break out the 4 patients. The truth is these are just 4 patients out of -- overall it's a small number. The truth is the 4 patients had fractures, half the fractures occurred before 3 months, only one fracture occurred after 7 months. So part of the thing you have to understand what we're seeing there is they're starved in the beginning of their treatment course, and I think that's really important. I don't know if you have the break out. I don't have the break out in my mind. Do you have the breakout.

No. I mean we're really looking at them. We're not seeing differences. I mean we know, for example, the young man [indiscernible] and he was in the 40-milligram per kilogram dose. The 14 year-old female and the other patients we looked in the wheelchair, they were in the 20-milligram per kilogram, but the exact breakdown for those 4 patients who did experience fractures, I don't have off the top of my head.

We did look at the 4 patients from a BMD standpoint, and they were near the mean or median of where everyone was. So their BMD response was actually very similar. I think a lot of it has to do with -- they were trauma or accident-related. And it was most of them were early in the core, half of them were early in the quarter. So I would say to you, it's surprising how few we had overall. That's the thing that's kind of surprising that the effect is that fast to be shut down. And I think that's because the new bone being made is made where the bone is weakest, and that's why the strengthening actually can be quite rapid because of the way bone is produced.

Yes. And I think fingers, toes, face and skull is an interesting conversation that really was at the request of the FDA. I think some people are trying to make the argument that those types of fractures may not be as clinically significant. I think to anyone who fractures a finger and face or a skull would argue otherwise. But I think with the results we're seeing, it's something we can align with the FDA on, and we'll certainly look at fractures from both perspectives.

We're going to do all fractures too. The truth is it didn't matter. They just feel like those fractures are more accident-related not due to weakness of bone kind of thing. But honestly, you could see it, it worked both ways. It didn't matter. So natural history, the 64 were just based on age and deletion type. That was a selection with -- the agent selection type. There was no other. And then the issue of do they know they're being treated. Well, they're Angelman kids, there are pretty developmental delays. I think to try to determine if that somehow would affect what they do, I'd say, is pointless. They can't know what that means to them, right? They're still learning their basics. Their Angelman kids are like 2-year olds, right, pretty much. So they're not going to understand it. I'm sure they're happier when they're calmer and they're having fun, but I don't think they have any idea what's happening. Let's go on to the next question. Who has the mic here?

Brendan Smith from TD Cowen. I wanted to ask a little bit about some of the potential seizure data that you have. I wanted to see, first, if you're actually having the patients and their families maintain a seizure diary on the Phase II. And if there's anything you can comment on there and potentially how you're thinking about that heading into the Phase II meeting with FDA? And then also on Angelman, kind of looking at the different domains in which you're seeing improvement. Obviously, you see good in cognition and communication and a little bit different in some of the other endpoints, fair enough that these are heterogeneous patients. But if that maybe suggest that the drug is getting a little bit more consistently into the cortex versus maybe some of the deeper brain structures we should kind of think about that?

Yes. So on seizures in Angelman, most of the kids are in good control, and we did not have an issue. Now there's one of the kids Patient 5 clearly had seizures a problem when he came off, but we haven't -- we haven't been charting a real effect on seizures that we've been able to see. Clearly, we will obviously be monitoring seizures, but I would say -- Camille, do you have something to offer.

So the trial is designed in a way that patients have to be stable on their antiepileptic drugs and have a stable seizure count. So we don't really expect to see any dramatic change, and we're not measuring seizures as an endpoint for efficacy for that reason based on the way the current trial is designed. Now we are -- we do have a seizure diary that collects seizures. We also collect information on sleep and the seizure counts that we're collecting are primarily for safety.

Thanks, Camille. Those are important. Another question was on the domains and Cortex. I don't think that -- we did -- the product has been valued at least in the nonhuman primate size brain showed pretty good penetration. Certainly deep nuclei maybe not as much, but I would say you'd be hard-pressed to try to correlate clinical endpoints with exactly. One of the things that is important to us is GTX-102 did get to all the brain regions. And we actually did not see a distinction, and I don't think the clinical data tell you a distinction. For example, the motor walking, it's very much deep brain controlled motor walking, right? That is -- right walking is part of -- is a deep brain coordinated function. So I actually think that -- I don't think you could draw a connection there. Who's got the mic?

Hello. This is Michael Riad on for Morgan Stanley. For GTX-102 in Angelman given the benefits on several domains, can the MDRI be applied to the national history data? And if so, how could that look across the age groups?

We certainly could do that. It'd be easy. If you looked at what we saw, none of them got close to the threshold for none of them. So they'll all be blank if we did it, and then we could show a beautiful statistical comparison. And I guess we could do that if we wanted to enhance the analysis. But because you could see the green line, all the graph, you could see one of the other natural space or crossing threshold of clinical leanness. So that would have been fine. It would have been one other way to do it, certainly. Someone here that's been -- had their hand up and then where is the other one? Okay. Who's got it now? Do you want to go?

Yes. This is Priyanka Grover on for Anupam Rama of JPMorgan. You had previously mentioned when answering the first question that there is a 20% of fracture reduction rate with bisphosphonates, though the literature is a little all over the place on this. How confident are you on this assumption as it relates to putting setrusumab fracture rate data into context?

Well, the thing with fracture rate reduction, you have to look at the powered study that actually did a randomized power design to look at treatment effect sites. Because if you look at a small study of 20 patients, and they do something to come up with numbers, then you're dealing with SKU data, right? So to really get a good treatment effect size, essay should use the 2 out of the 5 studies that were randomized, controlled with enough power to really do a proper ascertainment. Those studies give you about 20%. I think the truth is that all the patients in this trial, all those fracs you're talking about, those are all people on bisphosphonates. So whatever it is, they're all fracturing still. So I'm really not concerned about bisphosphonates as a control group. It is not that good at controlling fractures, it just not, isn't. And the truth is to strengthen their bones, you need bone anabolic effect. It's not a question of preventing absorption. It's a question of laying down bone where the bone is weak. And these kids just don't. So I'm actually -- really, we don't have much concern now. The trial itself is placebo-controlled, right? So control group could have been on bisphosphonates and would have bisphosphonates in their bones, but that could be true in both groups. But that effect would be waning as time goes on.

We do have that second Phase III study to support Orbit, and that is a head-to-head setrusumab to bisphosphonates. So we will be generating our own data there.

We have no fear about that one. The drug setrusumab's effects will be so much greater on BMD and done the right way, it will be, we think, much superior. I think who's got a mic?

Emil, Dae Gon from Stifel. A couple of questions. Starting with OI. If we look at the trial conduct itself, I think you mentioned -- or one of the docs mentioned that parents have gotten so used to fractures. How frequent are these patients being called in to measure their radiographic fractures just so you don't miss any of those in real world? In terms of AS, if we look at the patient's 14, 15 in the Cohort 4 through 7, the extension, I believe, I think they were the -- they had the least amount of improvements. And I think bone of the remarks was at age 4, your curve kind of flattens out. Can you comment on the 14, 15 patient age and what their current status is if we can kind of get some clarity on that, that would be great.

So the first question was on the -- yes, go ahead.

So important question because we do know and we've heard very consistently that parents are underreporting fractures in their children out in the real world. So we're really trying to tell -- ask them to err on the side of caution if you're having the type of trauma, that would have historically generated fracture come in for x-rays. So yes, we do have set intervals where they're coming in for evaluation in clinical exam, but more importantly, we're really communicating and less is where training comes into play here is that when you do have that type of trauma, you are coming in for x-ray evaluation and we're really looking not at just suspected fractures, but for our primary endpoint, radiographically confirmed fractures.

And they are coming to the clinic monthly. So they are being seen monthly in the way because of the treatment. So it does assure that we're capturing things. There will be a skeletal survey at 1 year, which allow us to look for subclinical fractures and look for healing fractures as well. So you asked about the 14, 15, age 14, 15. I don't think I want to get into the details. These patients is a little bit different. The older patients may respond differently than the younger patients, but the truth is there's a lot of variability in what these patients are, but I can't tell you there's anything you're going to be able to glean from understanding who they are. I would say we've seen older patients respond, younger patients respond. I think it's expected the younger patient will respond better in certain ways. But the older patients, some of them have been, for example, better in gross motor and walking improvement mainly because their walking is so much poorer at baseline that you can see the improvement. So I think it's -- I don't think I would gain much insight from that. I think Liisa, are you next?

We have one at the back. I think she is having the mic.

Salveen Richter, Goldman Sachs. With regard to the Angelman's program, you plan to discuss endpoints with the FDA early next year. What would you see as a positive outcome here? Or what are you expecting in line with the receptivity you've had to date on the multisymptom domain?

Yes. Salveen, I think they -- at our last meeting, they seem to open to a lot of ideas. And that's partly why they want us to go talk to the clinical outcome assessment group and to talk with them. I think they were open to understanding this is a complex disease and first time ever treatment. And our discussion -- initial discussion is to pitch them ideas of how to go about this. My personal bias is toward the multi-immune responder-type index, only because it allows us to capture benefit different patients who have different situations. You heard 30% of the patients have severe sleep problems, but those problems are profound for that family. I'd like to be able to capture that in the primary endpoint. But it may not be everyone, the walking problems, language are pretty common for everyone. So I think that discussion is hard to know. They haven't really guided us exactly on one thing or another. We certainly not told us CGI, for example. They did not say that at all, right? That's not something we've talked about. I think people understand the Bayley and the stem and the others, and they're interested in understanding the supporting data on the endpoint and its relevance to Angelman syndrome. So we're doing that work to try to show how those quantitative endpoints relate and how they are valid for the use in Angelman. But right now I don't think they've given us a tip the hand 1 way or the other. I think it's really up to us, by the way, to do the right science and to present our case. And in the past, we've been able to get the right things done with the agency, but we have to present our case well. Next question. Right here, Liisa.

Liisa Bayko from Evercore ISI. I was just -- my first question is on setrusumab. Curious about the 4 patients that had some fractures. Was there anything different about those patients that they have the same level of bone mineral density after treatment or anything that you could identify why they might have responded differently? And then also, could you talk about the data in the context of means, I know you used medians there, and I'm just curious the differences...

Means, medians -- we've mentioned the 4 patients I did just a while ago. So those -- the bone mineral density of those 4 were the mean or mean -- they're like all the rest. Most of those kids had traumatic event, something happened that correlated, and half the fractures were before they had been on drug very long. So if you cut away half the fractures with that and then you have some traumatic events, and I think they respond to the same. So there's nothing wrong with them. They just got unlucky. They were slipped on black eye or whatever happened. There's different things that they did that fell on their arm and all that. So I actually think there's nothing different amount, Liisa. Do you agree with that?

Yes. I think looking at bone mineral density, we're not seeing the difference. We're not seeing them as slower to respond or weaker responders. The difference is time on treatment and then degree, really the degree of trauma there.

Good. I think we'll go with 1 more. I think we're at time -- we've burned through our time. Thank you for all the great questions. Is there 1 more? Where is it?

Kristen Kluska, Cantor Fitzgerald. I wanted to ask about the trends in what you're seeing after 12 months, especially given that this is a limitation for romo. And then how are you thinking about pricing early on here with the potential for off-label usage?

Yes. So you actually saw the 1 kid that Eric presented that had been on 17 months. I think the data -- remember, we have no fractures after 7 months in anybody, right? So there is a potential that this could dramatically reduce fractures in these kids in the long run, which would be amazing. The truth is that this -- for safety reasons, romo is limited to 12 months in osteoporosis. And that is all they're doing. They have a hard stop at 12 months and they discontinue. We're not. We're actually planning and we have not had any pushback from FDA of going chronic. Now we think once the kids reach a certain bone mineral density and once they have stopped having fractures, there may be a point at which we move them to a maintenance dose where it might be every 2 or 3 months, let's say. Based on the ASTEROID study where they stopped them at 12 months and then looked at withdrawal, BMD followed them. They are okay till about 2, 3 months, and then they started losing ground. The bone started going back the other way. So it looked like if you stimulate them every 2 to 3 months, you might keep them from a steady state. So that's the way we're looking at it. I think the competitive issue with romo, and romo works. Romo is another anti-sclerostin clearly works. Amgen is not interested in it. They have given that the nonexclusive license of the epitope patent tells you they're not interested because Amgen would narrow give rights to a patent they were not interested in. And that's a royalty free. It's like a license fee type deal, no royalty involved. And they are doing what they're doing because it's precisely required in their pit to do this, but I don't think they're putting it on label. So the question is, could people still use it off-label? And I think that is a potential. But they won't have the services we have in support of patients, co-pay support and the rest of it that we do. They won't have home infusion other approaches. So there are things we have to think about. But we've said as a company that we will be conscious about pricing. And we've always been a company focused on responsible pricing just as we did for Crysvita. And we'd have to look at how -- what the off-label risks are. And I believe that a reasonable price differential will not cause a problem. And with the combined improvement in services, we think it won't be a resistance, particularly since our drug will have been proven safe and effective in the indication, and that it will have long-term treatment effect data. It's also a fully human antibody, not humanized, fully human for chronic treatment is always better, right? The risk of immunological, we have no antidrug antibody issues. So you look at a number of those features and the care will provide to our patients. I'm less concerned about it. But I think you also have to be conscious of an extreme price differential could create more of a problem and what will be a responsible price company. When you think about how many patients are an unmet need. And if the effect we're seeing so far translates in Phase III, I can see this being a first-line tariff for basically most oli patients. So we see this as a tremendous win, and we'll manage the competition when we get there. Good. Thank you. I'm sorry we sure there are more questions, but we've been our time, it's time for break. So a brief break, and then we'll be back. Thank you. 10-minute break. [Break]

All right. Great. Thank you, everyone. We're now going to shift over to looking at our gene therapy programs. I'll do a clinical overview and then I'll hand over to Dennis to do a walk-through with our manufacturing platform. So this is our pipeline within a pipeline. To my knowledge, this is the largest and most advanced gene therapy portfolio out there, particularly in rare disease. These are our 6 active programs. Four out of 6 of these programs are in pivotal trials. This includes DTX401 for the treatment of glycogen storage disorder, our lead liver-directed gene therapy program in a traditional Phase III. UX111 that we brought into the portfolio about 1.5 years ago. This is a peripherally administered gene therapy affecting a neurologic indication and MPS IIIA. DTX301 for the treatment of ornithine transcarbamylase, so urea cycle defect. And then importantly, our newest liver-directed gene therapy, UX701, for the potential treatment of Wilson Disease. We are including this and counting this in a pivotal trial because with the learnings we have gained with these earlier programs, we were able to bring forward and come to agreement with the FDA and other agencies on a seamless Phase I/II/III design. That's important because that allows us to move forward without stopping for end of Phase II discussions with the FDA or scientific advice in Europe, really does accelerate developmental time lines and allows us to get these important treatments to patients as quickly as possible. With those important learnings, we were also able to diversify outside of the liver and beyond peripherally administered gene therapy. We're bringing forward UX055, a gene therapy administered directly into the central nervous system. This is for CDD or CDKL5 deficiency. And then again, really importantly, now looking at large muscle diseases with our newest preclinical program for UX810 gene therapy for the potential treatment of Duchenne muscular dystrophy. Quickly looking at these disease individually. Again, gene therapy for GSD1a is our most advanced liver-directed gene therapy program here. This is a disorder of energy metabolism. These patients are not able to break down glycogen to produce glucose during times of fasting. So this means that these patients get into trouble very quickly. Their glucose levels can plummet. They can start to have seizures, and these really can become life-threatening events, particularly in the pediatric age range. There is really no treatments available for patients today. And what they have found useful is the use of uncooked cornstarch. Because cornstarch in its uncooked form is very slowly metabolized by the GI tract. So it is providing some source of glucose during times of fasting. These patients are dependent on cornstarch for life, meaning they have to take it every day, several times a day for life, but also dependent on life because a single missed dose, again, particularly in younger patients, particularly overnight while they're sleeping, a single misdose can result in a fatal event for these children. What you're seeing here is a picture of what one of our adult patients packed to go on a long weekend trip. He doesn't trust that he can get cornstarch when he gets to where he's going, so he brings it with him. After treatment with DTX401, he now just brings a single box kind of as a safety blanket here, and really shows the impact this drug have. They're able to really lead a normal active life, travel, do things they weren't able to do without having to worry about how are they going to maintain their glucose levels. Patient in the Phase I study continue to do very well. All patients are in their fifth year of treatment. All patients continue to respond. And this reduction in cornstarch, while maintaining glucose levels, will be our primary endpoint for our ongoing Phase III study, which fully enrolled at the beginning of this year and with a data readout first half of next year. Shifting now to our first neurologic indication. This is a peripherally administered gene therapy but affecting predominantly a CNS disease, Sanfilippo or MPS IIIA. As I mentioned, we brought this program into our portfolio about 1.5 years ago. The ongoing transfer study was originally designed to be a Phase I/II study. The data out of that study is so compelling and really drove the reason why we wanted to bring the study in-house, allowed us to have negotiations with the FDA to consider this the pivotal data set, and we will not need to do a new study for registration. Original agreement was to follow these patients to their fifth birthday really looking at a clinical endpoint. By patients' fifth year of age, buy natural history, they really all have deteriorated to a meaningful degree. We do think this is a very good opportunity to look at a biomarker approach here. Heparan sulfate is what accumulates in these lysosome. It is what drives pathology and clinical signs and symptoms. And we are having ongoing conversations with FDA about a biomarker approach here, and we'll continue to have these discussions as we approach the end of the year. Our next liver-directed gene therapy program for OTC or Ornithine transcarbamylase deficiency. This is a urea cycle defect. Urea is made to breakdown ammonia. Ammonia comes from the ingestion of protein and the breakdown from muscle. We do not tolerate ammonias above the upper limit of normal. It's a very potent neurotoxin. So with this enzymatic deficiency, you can very quickly get into trouble with a metabolic stress, taking in too much protein, having a viral infection. And after treatment with DTX301, we're seeing 7 out of 11 patients being responders, being either having the ability to come off their current standard of care, alternate pathway medications and protein-restricted diet fully or to at least a 50%. These patients also and at least their fifth year of follow-up. So like these -- this program is also in a traditional Phase III program enrolling very well and looking to fully enroll that patient in that clinical study in the first half of next year. A little more time here with Wilson. We're really excited about this new also liver-directed gene therapy program. We really weren't able to look at these types of diseases with this type of incidence and prevalence without the advancement of our manufacturing platform and really bringing forward the Pinnacle PCL platform, building our manufacturing facility outside of the Boston Cambridge area has really enabled us to look at these larger indications. Wilson is a disorder of copper trafficking. Copper, free copper is also very toxic to the body. This genetic defect leads to the inability to transport copper out of a liver cell, out of a hepatis site, either into bile for excretion out of the body or for loading of copper onto ceruloplasmin, which is how it needs to normally traffic to other cells and tissue where copper acts as an important co-factor for enzymes. So we're really looking at this disease as a copper toxicity in excess of free copper, but also a functional copper deficiency because you're not able to get copper where it needs to be. Again, leveraging all the experience we've gained over the years, looking at a seamless Phase I/II study, because we do need to establish initial safety and selected dose for generating our pivotal data set, we're looking at stage 1. We are looking at 3 cohorts starting with 5e12, moving to 1e13 and then 2e13, then moving to Stage 2, which will be generating the pivotal data set. We recently had a data readout to support moving to Cohort 2 dose 2 or 1e13. And what you see here is these 5 patients listed out -- we listed out, we do enroll them sequentially so you see the time and study here. And with patients 2 and 3, longer on study really promising results here. We are seeing the copper biomarkers move in the correct direction. We are seeing data from a functional assay, which actually measures the direct loading of copper onto ceruloplasmin with chelators, you're simply binding copper out of circulation with zinc, you're preventing absorption from the GI tract. So if we're able to measure and see an improvement in the loading of copper on the ceruloplasmin, that really is directly showing the effect of this transgene. So seeing these patients, patients 2 and 3, completely off of standard of care, seeing the biomarkers moving in the right direction and importantly, having them telling us how much better they feel. We see this as a very promising signal. Patient 4 and 5, a little bit early on in follow-up, they are actively titrating. Patient 4 is currently recovering from COVID for -- so a little bit of a pause there. But with 33% and 50% reduction at 20 and 16 weeks, respectively, again, really promising results for these first 5 patients. We have fully enrolled cohort 2 at 1e13. We have all patients identified and deemed eligible for Cohort 3. So looking to quickly complete Stage 1, which will allow us to seamlessly move to Stage 2 and generate that pivotal data set for registration. Now I'd like to hand it over to Dennis Huang, our Chief Technical Operations Officer.

Thanks, Eric. So we'll pivot here a little bit and move from the clinical view of things and begin talking a little bit about the technology because as you may know, gene therapy programs are inherently highly technical. And also being a more cutting-edge type technology have shown in various programs moving forward here that the technology or the CMC side of things can actually be an impediment to drug development. So as Eric mentioned earlier, many of our older programs use the HEK293 suspension process or system. And we currently run those for our GSD programs and our OTC programs indicated there DTX401 and 301. What else be spending most of my time talking to you about really is our Pinnacle platform. Pinnacle PCL platform, which stands for producer cell line. And that really is the way forward for us as a business, as a company and really enables a lot of different capabilities that we believe will really lift up the whole area of AAV gene therapy. With regard to Wilsons, this is the first program where we've really turned it on, if you will. And built and enabled a really tight integration. As Eric mentioned earlier, the Wilsons program is really the prototype with which we are going to be moving forward, both from a clinical design perspective, but then also from a technical base. So as an example, 1 of the things that we're really keen on with this Pinnacle platform is certainly reduced cost of manufacturing. That typically has been a very significant cost element within the development cycle, but also maintaining that all the way through not only the clinical development, but through the commercialization. And I'll talk a little bit about that and certainly much more detail. But the other aspects of it is for the CMC program to match that integrated approach that we're taking, we have not had to do any sort of scale changes as part of this Wilson's program. We started out at that 2K and so you see a picture of that over there at the 2,000-liter scale. That has enabled us to really minimize the supply chain challenges that you typically would see or doing scaling up and making CMC changes to the life of the program. So fundamentally, what we'll be sharing here is this process and this technology really enables us to expand our worldwide patient access ability to dose at much higher levels, it probably wouldn't be possible. And then really just much larger market opportunities. So we've not only invested in the technology, but we've also invested in a lot of internal capabilities. And so when we acquired Dimension Therapeutics in 2017, we already had a very established research and development area. And so we've pumped in a lot of effort and energy into those groups and have expanded their capabilities in the Boston area. Subsequently, we then added GMP manufacturing capabilities in QC Laboratories, which also tends to be a bit of a bottleneck for many of these gene therapy programs in 2019. And then, as Eric mentioned, you see the nice shiny picture of our new facility, our GMP manufacturing facility, which just came online earlier this year. And so this facility has not only drug substance, but drug product capabilities. It is designed to both produce our HEK293 platforms, but also just as importantly, our Pinnacle platforms. And again, this is based on the 2,000-liter scale. Though part of the reason why we selected that scale is really tied to the fact that for rare diseases, we probably don't need that pick. But from a technology standpoint, it's much -- it's capable of going much higher, and I'll show you that shortly. So comparing these 2 different processes, this is sort of one way of showing the differences here. As I mentioned before, the commercial scale is really critical because going back to the changes that you typically do during a clinical development, we're starting to add up to commercial scale. So we're minimizing the need for changes. So CMC type changes that can slow down programs very significantly during the course of that, and that's one way of us minimizing and making a very time-efficient and cost-efficient development life cycle. The second piece here really refers to how stable this cell line is. And I'll go through in a little more detail about why we -- or how this PCL has developed and what makes that important in terms of supporting our programs. Quality attributes are really key here as well. And just I'm sharing just one example here where in the Pinnacle PCL platform, because of the way it's engineered and sort of leveraging this AAV5 approach that we're talking about here, really creates much higher quality full capsids coming out of the bioreactor, okay? And so typically, in a -- maybe HEK293 process, we'd see something in the 20% to 25% coming straight out of the bio reactor for the Pinnacle PCL process, we start out at a much higher level. So you can imagine that starting at a higher level, cleaning it up and doing all the things you do just creates much better economics from the get-go and reduces fundamentally, and this shows in the last row here, the cost of manufacturing. This is a -- can be in a very expensive therapy to produce, to manufacture. And certainly, we expect to get much lower cost of manufacturing with our Pinnacle PCL platform, very comfortably getting below the 100k per patient dosing range. And in fact, for the Wilson's, we were able to achieve that even during the clinical phase as I mentioned. So let's delve a little bit. I've just got 1 slide that looks like a cartoon. But this particular slide just articulates a little bit the elements that come together in our Pinnacle producer cell line platform, that's really geared towards high titer and creating, therefore, commercial viability. All starts on the far left there with the producer cell line, and that's by definition, what this process is. This work was first started or first described in the mid-1990s. So this has been around for a while, actually by our gene therapy, Chief Scientific Officer, Reed Clark. And so he did that work initially as an academician in Ohio State, and we licensed it from there. And combined with the producer cell line with this AAV5 helper virus. And what that does for us is it creates a very simple formula, if you will. Now compare that or contrast that with the ATK-293 where you add DNA plasmids to it instead of the helper virus. We're -- and all those plasmids create a lot of additional cost because they are not particularly cheap. But then also from a thermodynamic standpoint, probably creating some challenges with regard to creating full capsids, which is what is represented here in the far green -- on the far right-hand side in the green is the product that we're after. So a big quality attribute that I mentioned before are the percent full. And so the bigger the number, the higher the number, the more is a better thing for patients and so forth. This slide here really indicates the improvements that we've made over the years since acquiring Dimension in 2017. One of the things I also want to mention on this slide is that we first introduced clinically this PCL, the early forms of it in 2015. So this is a clinically validated sort of methodology or approach. And obviously, you're hearing about the Wilson's program, which is also sort of illustrated here with the yellow 2019 technology. But the main point here with this slide is since the time that we've taken over this technology, we've really invested a lot and made some significant improvements in that time. One of the examples that I'll be sharing with you is related to the Duchenne program or the AAV10 program. As part of the AAV10 program, we really had a number of challenges because of both the prevalence population, higher dosing. And so this particular challenge was identified pretty early on. And so one of the things we appreciate about our PCL, our Pinnacle PCL platform is the fact that it abides by or really takes proven what I would say, biotechnology. So if I go back to my earlier career, in recombinant proteins and so forth, a lot of those technologies batch-fed fermenters and all those kinds of things. Many of those technologies or improvement strategies are sometimes called in the field, intensification approaches apply directly to what we're doing here with the Pinnacle PCL platform. And so we feel very confident that using those proven approaches that we'll be able to enhance very quickly, and we have shown very quickly improvements in yield and tighter and quality, all the things that you're looking for in a low-cost therapeutic. And so we've developed what we call our APEX process, which is an AAV Perfusion Enhanced eXpression system, and this has now been applied to our AAV10 process. And so a couple of slides or a couple of panels here on the left-hand side show you some of the changes that we've made and the improvements we made using this perfusion system. The benefits of this really are, as I said before, the engineering control of this, the scaling of this. And so I mentioned it's regularly easily achieved much higher in traditional biotech programs instead of 20,000, going up to 10,000, 12,000, 25,000. And so in principle, this -- because the supplies is very relevant to the -- our system here, could be scaled very easily for larger indications in therefore, economies of scale and again, reduce COGS. So while the Wilson's program is definitely the technology associated with was very good. We've subsequently added additional improvements to this, and that was the example that I was sharing with you around the Duchenne program and then potentially even larger indications or larger dose indications. So hopefully, been able to share a little bit of some of our excitement around how we believe this Pinnacle platform, Pinnacle PCL platform can really transform this area of AAV gene therapy to touch on not only liver disease, neurological and musculoskeletal. This is something that we've invested a fair bit and they're now just to the point where we are able to sort of visualize and see that path forward based off of our internal capabilities. And my final slide here really just talks about the fact that if you compare this to other technologies in the past, monoclonals, recombinant protein technologies in the '90s and 2000s, standardization and of those things, has really helped us move that technology forward, and more importantly, better helping and treating our patients more quickly. More recently, the mRNA with COVID vaccines is now, I would argue, has now become the standard for those types of treatments. And the PCL platform, the Pinnacle PCL platform that we've developed really has a lot of the right characteristics. It's clinically validated, we've demonstrated it can be cost effective, certainly efficient from a time -- development time line without making changes during that life cycle, high-quality and scalable product. So I'll -- those are conclusions of my remarks. I'll turn it over to Emil to talk a little bit more about an application of this Pinnacle platform.

Thanks, Dennis. We thought it important to talk about the Pinnacle platform the gene therapy franchise. I think AAV has been in a challenging place, but I think there's ways to make it more accessible, more cost efficient. And I think Pinnacle is one of the ways to do that. The question though people think that $3 million price point, who are you going to treat and how does that work? And I think that's because it's early. But we think with the platform, we can actually do something much greater. In fact, we think it's also possible to reach larger market disorders through the manufacturing improvements of the Pinnacle PCL platform. What I want to talk to you about another indication, a very large indication, which we now have discovered some insights on that we could potentially use with AAV. And so I will talk to you now. The last topic will be a novel gene therapy therapeutic approach to targeting beta amyloid accumulation that builds off some of our historical information. And this is a project that's gone on a very low key at the company, very small spend, but has come up with something very exciting, and the Pinnacle platform enables us now to consider potentially even as large indication as Alzheimer's. So to give you a summary, you may know in the past, we worked on Galactosialidosis enzyme replacement therapy that we stopped. But while we are working on it and working with Dr. D'Azzo at St. Jude on the enzyme PPC, we discovered something about PPCA, is that actually cleans Aß42 very efficiently. In fact, can actually disaggregate Aß42 and we were very encouraged by this activity. But from my own experiences as a scientist and as a lysosomal biologist, I always had the feeling that Alzheimer's and all the beta-pleated sheet proteins are due to accumulation of data plate sheet that the lines are not adequately cleaving in. And so when we know Alzheimer's is complex. We know now that beta amyloid accumulation is clearly a problem. It's a resurgent and the monoclonal antibodies now are -- clearly have an effect, but the effect is still modest, decreasing decline but not yet solved. But [ Ultragenyx ] lysosomes is actually a more parallel than you realize, and I'll talk through those today and show you why actually providing supplements of PPCA can actually clear amyloid and as an alternative using monoclonal antibodies and we'll show you data in the 5xFAD mouse today. Now this is obviously something that Ultragenyx would do. We're a rare disease company. We're not going in Alzheimer's right now, but the idea would be we could put this into another entity and actually have that be pursued. And we think, frankly, with the success we've seen so far in it, it seems like an essential thing to do. This needs to be tested and move forward. So let me talk through the data and then our plans going forward. So the amyloid hypothesis has been on again, off again, on again. I think it's clear the amyloid matters, and you can affect and have an impact on Alzheimer's disease on cleaving or getting rid of Aß42 oligomer. What's not been so clear to people is where does it really go? What are the problems? And in fact, if you look at a lot of the lysosome, or they have many of the same features, and there's clearly some relationship between the 2. And I think from the work I've seen done, there's actually very precise pathological alignment between what you see in Alzheimer's pathology in the brain and what you see in lysosomal diseases. My own mentor, Rose Neufeld, studied MPS IIIB and showed it had a tauopathy. It was very similar to Alzheimer's, capping and all the other pathologic features, Alzheimer's. So Dr. D'Azzo Sialidosis and showed massive A-beta plaque accumulation with the deficiency of neuraminidase. And if you look at all the stories lysosomal disease, you see this pattern. But more than that, if you look at Alzheimer's studies and research is more and more data behind the idea of live dysfunction in Alzheimer's and that these features are not true to unrelated. These are actually related because I think there is a relationship between the biology that's going on between the two. And so direct evidence to lysosomal disease now -- of lysosomal involvement in Alzheimer's has been shown in a number of studies, but Dr. Ralph Nixon probably has the best data and he recently published some work showing activated lysosomal function in Alzheimer's brain. And clearly, there's a shift your thinking about the evolvement of lysosomal involved in amyloid production. And so that connection, we think, is actually more important than people realize. And based on the therapeutic data we're going to show you, there's a distinct connection what's going on in lysosomal what you're seeing is plaque outside the cells. So let me talk a little about the PPCA enzyme. I'm giving you a short [ clause on that ] chemistry very quickly to it. PPCA, protective protein/cathepsin A. Cathepsin is a protease, it's lysosomal protease. It actually forms a complex with neuraminidase in the lysosome, something [indiscernible] discovered. PPCA itself is a carboxy peptodase. It cleans the C-terminus of proteins. It's highly selective for hydrophobic immuno acids, which are very much part of Aß42 at the C-terminus. It also recruits -- PPCA will recruit neuraminidase to lysosome and neuraminidase regulates lysosomal exocytosis in biology. So when you give PPCA, you're actually giving PPCA and new 1 together in a sense, and that's changing the biology of lysosomal degradation. Now we did some early work with PPCA. And if you look at the sequence at cleaves, it aligns very well what you see the C-terminus Aß42 is doing. And the C-terminus Aß42 is actually very important because the C-terminus forms like a lipid zipper that assembles A beta. You need the C-terminus. If you get rid of the C-terminus, it will not form, it will actually disassemble. And what we've shown in vitro, if you take Aß42 in vitro that we will cleave and if you look at the bottom left there, you can see LC mass spec where we chew in half of the A beta from the C-terminus with the enzyme. And we can show that if you put Aß42 in what's called an aggregation assay, amyloid aggregation assay on the right conditions, then steady forming amyloid with small amounts of PPCA, you can prevent the amyloid aggregation from occurring. In fact, you're digesting. If you look at the western blot below, oligomerization are being adjusted. But if you look carefully above, you can also see the high molecular fibrils are actually disappearing, too. What you're actually is just assembling these oligomers and digesting them both. This is very interesting for us. This is in vitro. Question is whether this would happen in vivo and what -- how would it relate to what's going on in Alzheimer's model. So we did some experiments. I won't go through all we've done a number of them, but I'll show you 1 of them just to give you a sample. The 5xFAD mice is an Alzheimer's model has 5 different mutations, it's the most aggressive Alzheimer's mouse, has the most rapid amyloid accumulation. And if you look at the data that's been published, no one of the monoclonal analyst have published data enough this mice because it's so severe. But we've taken this model as the most severe model and have done some work. The experiment going to show you, the mice were started on treatment at 14 weeks. They're given a single injection of AAV that express it PPCA, and then taken down 12 weeks later. And we've done this experiment in various forms and repeated a number of times, but what we can show at baseline, there's already significant amyloid beginning. And in the control animal, you see a massive between 5% and 15% of the brain surface by quantitation is amyloid now in just 12 weeks, right? But if you treat with PPCA, you can see a substantial reduction or elimination of amyloid the 3 mices and not the most are actually the ones that had the lowest expression. So if you achieve a level of expression, you essentially reverse it, and we measure this in various parts of the brain in very different ways. I'm just showing you 1 example. If you look at the immunohistochemistry on the upper left is the baseline before treatment, showing already amyloid at the 14-week point, significant amyloid already across the brain. And below is the vehicle-treated animal, a 12-week showing massive amyloid accumulation, right? The 2 treated animals on the right show the bright green is where PPCA is being expressed. And you can see in the upper section, essentially no amyloid remaining. The PPCA is made in those cells at certain part of the brain, the enzyme that gets secreted in a pro form. The enzyme then get retaken up by other cells essentially arm them with the peptidase that can cleave amyloid and we see a significant reduction or in some cases, elimination of amyloid. If you look more closely at the images, and we're zooming in here. On the left is sort of what vehicle looks like, and you can see the frothy plaque, not very much of it in the lysosomes or in the perinuclear area, but most of it extracellular. When we do the treated analysts, you start seeing particles of amyloid actually in the lysosome. And if you look carefully, you can see -- some of them are yellow. The ones that are yellow are having PPCA and the amyloid pieces are in the same compartment, right? That's how they turn yellow. So you can see the high expressing green cells that are making enzyme cross-correcting cells around them who are now able to process the amyloid, the particular amyloid and plaque. This is essentially, you're seeing the plaques essentially being digested. So the question here, and this is a crazy part is the plaques are outside the cell. And the enzyme, we're putting lysosome enzyme in the lysosome. So how is it that placing a lysosomal enzyme in the lysosome only active lysosome is [ remitting ] pack outside the cell. Well, the truth is a black outside the cell is actually coming into the cell and coming back out again, that has a path through the lysosome and the plaque are connected to each other. We just don't see that because we don't see the pathologic process, the plaque is the car crash. The brakes are broken inside the cell, the car is crashed and you look at the crash and then you're trying to figure out what happened. The problem is that the brakes inside the lysosome for cleaning amyloid and getting rid are gone, and that's why you keep dumping and have seen car crashes outside the cell. So we can actually put a lysosomal enzyme in the lysosome and affect plaque outside the cell. This should be illuminating on what's really going on. And a fundamental shift to thinking, and it fits well with Dr. Nixon's work and Dr. D'Azzo work, the lysosome is really important. The second point I'd say is where are the toxic oligomers that everyone talks about, where are they actually? Are they floating around? Well, I think they're actually in the lysosome. And the reason they're toxic is because they're assembling in the acidic environment or assembling into fibrils. They're rupturing lysosome killing cells, and that's why they're toxic. And then once they kill the cells, the plaques dumped in about side, gets taken up again. And I've shown you in our pictures, pieces of flat plaque back inside cells again. What we're doing with PPCA is giving the cells the extra tool of protease to help digest the plaque when they have it, allow them to eliminate the plaque. And this enzyme is particularly good for Abeta42 because of its specificity for the C-terminus, which helps disassemble plaque. So we're very excited about this result. But obviously, it's something that's hard for us to do. But I want to say one more thing about enzyme versus antibody. I'm an enzyme guy, right, and we have an antibody product, setrusumab, but enzymes are more powerful. And this is just a point to think about here. When you give a monoclonal antibody, I'm always impressed with the fact that they actually have to give more grams an antibody, then you have amyloid in that they're binding. That you have a 180-kilodalton protein to buy 2, 4 kilodalton peptides. That's kind of a strange ratio, right? You actually have to give more enzymes. So the problem with monoclonal is that you have to give a lot of them, even if they buy oligomers, it's still more protein in the antibody. And so you're getting inflammation by this antibody because it turns on macrophages, the inflammation causes area and you have a limit to how much amyloid can you bind up, which is why you have to treat things early. But what if you had an enzyme? Well, the enzyme treat cleaves 35 Abeta a second. One enzyme of 65-kilodalton will cleave thousands of peptides. The stoichiometry is just dramatically better, dramatically better for what would be needed to achieve treatment and more advanced amount of amyloid. How does that compare? Well, if you compare what we're doing with PPCA, and the monoclonal antibodies that have been approved, you can see that we are working in the 5xFAD mouse, the worst most rapidly progressive amyloid producing mouse, and we achieved greater than 60% reduction in amyloid in that most difficult model. If you look at the others, they're only working in the 2x model, which is much slower. It takes months to get accumulation and their best is 30% to 50%, that's the difference between an enzyme and an antibody in terms of potency. It opens the door then to perhaps an enzyme delivered with a gene therapy could be a way to arm the cells with enough enzyme to actually give them the ability to process their own amyloid, which they're now unable to do. So our Pinnacle PCL platform puts us in a position to do this. We actually can get the costs down to maybe $10,000 to $15,000 a patient for a single-dose therapy. And imagine getting 1 infusion instead of over a 5-year period, having something like 100, 200 infusions, right, for monoclonal antibody, a onetime treatment. If we can get the cost in that range and optimize in us deliver to get the activity required, it does open the door to a large market CNS indications with this platform. So summary, as I said, Ultragenyx is not going to go into the Alzheimer's field, don't worry. But it's too important to finding not to do -- not to happy done. So the right way to do it is to put it into its own entity. We're calling Amlogenyx. We'll have a majority interest in it. We're seeking outside funding currently and are entertaining, talking to a number of investors. Our hope is to finish a Series A by the end of the year, it will have our IP in it, our platform, and our ability to help manufacture it. The program will be cash-neutral genetics. So I won't alter our commitment to our key value drivers and our burn, goal targets, et cetera. But I couldn't sit by and watch something this exciting and not actually put it into play and then make it potentially available. And I can see this just being a leap ahead in how you might treat Alzheimer's far better than diffusing monoclonal antibodies. So that's that story. That was surprising, right? A little excitement, a little buzz. Should be. So we'll go right into closing remarks. It's been a long day, I realize, but it's got to be an exciting day. We've talked about 3 major programs, and I just threw another huge program at the end. But transform treatment for UX, UX-143 to choose Osteogenesis Imperfect. The doctors are drought drop, the amount of improvement is just staggering because they didn't think that making more bone with this patient would do that, but it is doing that because the bone is not bad bone. The bone they make is still good bone. They're just not making enough. It may not be perfect bone, but it's good bone and the patients are strengthening, potentially transformative, if we are able to achieve eliminating fractures as a common occurrence that is transformative, especially if you imagine doing it from a 1- or 2-year old. The Angelman story has had a lot of ups and downs, but it's been all up for us. It's just down for other people. So I don't know, we hopefully try to give you a little more of what we've seen and feel, right? So when I say there's time and dose-dependent improvements that you say, "Oh, it was true, there is time- and dose-dependent improvement." And wow, this is powerful because we're changing the quantitatives, we're exceeding the natural history. We show you quantitative EEG improvements, and we've shown how that's clinically meaning clinical meaningful threshold. And we've shown how it's meaningful for patients in terms of doing first ever developmental accomplishments. It's a dramatic result. And honestly, I would have never thought anyone could get treated, but now I feel it's going to be treated, and we're excited about to be part of that. The 701 Wilson Program Pinnacle platform in the clinic. We had great data in the animals, but you never really know. And our first cohort data was so exciting and particularly for Eric, and also Sam Wadsworth, who was the originator of that program. We just had to lay it out for you. We weren't planning to put out cohort by cohort, but we decided it was too exciting not to say something about it because we've been bursting with the excitement as they've come off their chelators, and it looks like they're distributing copper again. So the potential now to actually treat Wilson the right way, restoring copper distribution as well as generic toxic copper is exciting, particularly in context of Alexion moving -- Alexion-AZ moving away from their key layer, which we felt was never going to be better than OP layer is just another one. I think it opens the door to us to be the leaders in the Wilson disease treatment area. I think the data we have is putting us on good track. With the low dose working, we have great safety, that's why we're in a very good place to having a great treatment for Wilson. So we look forward to both updates on OI Phase II next year. Wilson to the rest of it. So we're on a pathway to profitability, too. So let me -- I think there's one more, which is our catalyst. So catalysts, we've shown today the facts data. We'll have some more Phase II data coming, which will be year-long data essentially. And we'll let you know where the Phase III is and what's happening with Phase III. We've given you the extension data today on GTX-102. We promised you expansion data for at least 20 patients through at least 6 months in the first half of next year. It should be a nice substantial piece of data. It would be more reflective of what the dose that we'll be using in Phase III and we expect that to be a very successful update. It's going well so far. UX-701, you heard Cohort 1 data. We'll have Cohort 1 through 3 in the dose decision in the first half of '24. DTX-401 unblinding. And if that's not enough, we also have MPS IIIA discussion with the FDA. I've been active in the policy front. You'll see me writing editorials, you've seen me advocating at FDA. We're making progress. There's a lot of move in Sanfilippo space and other MPS CNS space. Breaking that door down, opens up gene therapy and other treatments for the brain and a lot of inborn eras it's why it's so important. Peter Marks is aligned with us. We're going to make that happen. We're going to push them on this and get our program through, but hopefully, do more than just for us, but for the whole field of treating these horrible, rare diseases. So that's our update for today. We can go into Q&A. Thank you for coming.

Whitney Ijem from Canaccord Genuity. On the Pinnacle platform, for the producer cell line, have you said or can you talk about what type of cells is that just an adapted 293 liners? Or is it a different cell line that's there? And then on the Wilson's program, exciting update. Any thoughts on the first patient or any kind of characterization you can give on that patient or the treatment in the context of the results presented today?

Sure. I'll touch on the Pinnacle PCL cell type thing. And Eric will give a comment on -- a brief comment on the first patient. So the Pinnacle PCL cell line is a HeLa cell drive, that's been known if we put that forward. It's not necessarily an original line. This line has been through years of evolution and selection and so forth, but it is derived from that line. The 293 lines are actually very difficult to turn into producer cell lines. And there are certain cells able to do and others not. To carry the AAV genes in your nucleus puts a burden on a cell to do it right. That's what the tricky part of developing. They're certain there can be other lines other than HeLa, but that's the one we're currently using. I don't think there's anything else you have for that. Okay. Eric, briefly what's happened in the first patient?

Yes. So dose 1 cohort 1, so we certainly don't need to see a home run with all 5 patients. But the truth is, we did see biomarkers start moving in the right direction. She was able to start tapering, but we are still using a prophylactic course of steroids to control the increases in LFTs. She wasn't able to tolerate the side effects she was having with those steroids. So we found out afterwards that she really wasn't taking her steroid. So I think we may have lost some transgene there. We'll continue to follow her over the long term. But I think it does still stress that we do see these elevations in LFTs and controlling that response is important.

Yes, I think the response to the level is not really that high, but it was some. But so far, actually, the safety has been excellent. So I think that's not 1 patient. There's always 1, but we're excited seeing 4 out of 5 at a low dose because our target dose was really the 1 or 2e13. So we're excited about that. Next question?

This is Michael Riad for Jeff Hung in Morgan Stanley. A quick question. For Pinnacle PLC, to what extent can the platform accommodate large-sized gene deliveries? I think you had mentioned UX-801 for Duchenne. Is it able to take like the full length dystrophin?

It's AAV. So AAV has to limit the 5-kilo base kind of limit. So you can't do full dystrophin. And so there's a limit. Our A-10 program is a Microdystrophin. It's actually 1 we licensed in from another company, but it's doing micro. The solution for Duchenne. I think Micros clearly have a role in helping to in Sarepta AdComm product approved, but looking bigger is going to require more innovative approach, more innovative.

Yes, just a few questions on the Wilson's program. With those 4 patients that did have an initial clinical response, was there any correlation between the reduction in the standard of care, medication and the reductions in the copper biomarkers? And then on the manufacturing, can you comment on what percent of the Wilson's market you could supply with the PCL process?

Sure. I'll let Eric, do you want to do the Wilson thing and maybe...

So I think it's fair to say, yes, there's a correlation, but there's almost a little bit of a forest correlation because we need to see those copper biomarkers moving in the correct direction to start titrating. So yes, we do. But it is important to be seeing these biomarkers moving prior to starting tapering, and that does drive tapering as well.

And regarding the Pinnacle PCL platform in terms of ability to treat the market. If you recall, the Wilson's is not the APEX process. And so we do have another arm of sort of lever to pull if we needed to grow the sort of efficiency of our manufacturing process. So at this point, using sort of the rough numbers we have now, we should be able to cover most nearly all of it.

Even with the current process already commercial grade, COGS are excellent, we can better than other programs. In fact, the Wilson program, COGS are about 80% of the other program COGS. In fact, the whole Wilson program costs less than the other programs because the manufacturing is so much cheaper. So even at the current process, the COGS are excellent to go commercial with we could dial it up 1 more step, if we needed to, to get another few fold more productivity. Huidong, yes?

So maybe follow that question, you just come at 80% of the cost. I think the HEK-293, if we're using, say, Roctavian dose 5e13, estimate cost is like $200,000, $300,000, but that was the numbers out there. I'm pretty sure every company has a very specific numbers there. So is that the baseline you were talking about when you're seeing 80% of the cost?

Yes. Roctavian remember is the [indiscernible] virus.

Yes. Sorry, sorry. Maybe I should use DMD as a better example. That's HEK293 adherence cells and a 1e14.

And you're saying they're spending $200,000?

Yes, $200 million to $300 million. That was the numbers out there.

Yes, I'd be surprised if that's true. Well, I just thought that there was like a $1 billion contract manufacturing deal that was done to support that product. So I'm not sure how that lines up. But anyways, she's asking 80% off or what. Well, she is looking for hard numbers.

Well, I think I did give a hard number, roughly hard number, at least less for the Duchenne program, I do believe we should comfortably underneath $100,000 per patient with similar dosing as I think it's 2e14, if I'm remembering right, for.

Okay. Like in terms of manufacturing cost, there are numbers?

Correct.

You definitely know how much GVC you're talking about, right? The indication is important.

Yes, it's tied to the dosing and all that stuff. .

Okay. Second question very quickly on the Abeta42, that was very impressive data, selective degrader although I think PPCA does have other activities as well, right, the M&As and also the Easter is. So have you done the whole, say, protein profile to analyzing what other proteins did or did not get affected?

Well, you don't really have to. I'll tell you why because it's enzyme-active in the lysosome. It doesn't matter what else it might cleave because it's in the lysosome and it works at acidic PH. So wherever else it is, it's not going to do anything. So only if the protein is in the lysosome, can it cleave. When it gets secreted, as I was saying, it's secreted and retaking up, it's in a pro enzyme form, it's inactive. It's secrete. When it gets secreted from an overexpressing cell it secreted with a pro piece on there that gets taken up by the [indiscernible] receptor taken in the lysosome, they cut off the pro form and activate it. Because of the lysosome activity, it's less of a concern that you're going to digest everything. It's similar, for example, look at BRINEURA for BioMarin's TPP1, right? It's an enzyme. It's in a pro enzyme form and injected in the brain. It was kids, they put 100 milligrams in there, huge amount of enzyme. It's a pro enzyme form, it doesn't digest anything until it gets into lysosome and activate. So only proteins that would be accessible to it are cleavable. Does that make sense to you? So you wouldn't worry about sort of emulsifying people's brain or something for that reason. But it's a question, I mean, it doesn't harm me to have too much? I think we're talking about several fold more, maybe it's 8 to 10 fold more. But the amount of the enzyme that's in your body normally is not very much to begin with. So 10 fold more seems like a lot, but it's actually on absolute terms, not a lot of enzyme. And most of the time, for example, on those histochemistry things, natural recurring loss of PPC, you couldn't see on immunohistofluore, it's not enough enzyme to see the signal. You see it only when it's over-expressed. So we don't think it's a problem. I think if you look at the mice at the end that are highly expressing, they're fine doing it, right? So I don't think it will be a problem, but it's an interesting question.

I just have 2 questions, maybe 1 on DMD since we've been tangentially talking about it. Emil, I just wanted to get a sense of where you guys are in development. I know it's still pre-clin but how does the current landscape potentially determine when you would enter the clinic, depending on if Sarepta does or does not get approved? Obviously, there could be a big advantage simply on way of manufacturing. But theoretically, what could be some additional advantages, not just to Sarepta, but to anybody else that's in the clinic? And then maybe to go back -- I'm sorry, to Angelman, I forgot to ask you earlier about what your thoughts might be for the Ionis program and how yours might be differentiated from theirs. And if you have a sense on when their data might be coming.

Well, we're still encouraged with [ ATM ] program. We have produced, the producer saw it so we can -- we have the ability to scale and the APEX thing that Dennis talked about is being applied to that program. So we can get the productivity way up and get the cost. So that part of the story is completed. We've been working on some other parts. We have some very interesting targeting work we're doing in collaboration, which I think will be the next step, which we won't talk about yet, but we're excited about. Our whole strategy in Duchenne was never to be first. We knew someone they were going to be first. The point would be to come in and actually leap to the next level of its efficacy. Because I think the efficacy, well, I think it's acceptable. I think it should get approved because some kids will never have another shot -- of that shot to say their lives because once they get past a certain age, their muscles are not going to be able to recover. We do think there's a lot of room for improvement and thinking less about the prevalent population. Think of it more as an incident population, new boys being born. The thing I also talked about is the $3 million price point or that range, how many countries are we really going to be able to do that, right? I think gene therapy should be for more than just U.S. patients with commercial insurance, right, that there's something wrong with that, right? And the Pinnacle platform puts us in a position of actually being able to do something large scale and global. So I think there's plenty of room, plenty of room for improvement. It's not our usual thing, by the way. We're not normally going -- we have to go after first ever. In this case, we thought we had a better idea. And so far, we're making great progress on the CMC side with the addition of the targeting piece. I think we'll be in a position to have a better product. We'll talk about it when we're ready. Oh, yes. The Ionis program. Well, the Ionis patented region, which was the original patented region, where Roche is also. That region is further 3 [ firm ] where we are. And our view of that region is it's just not as potent it's where we are. We've made ASOs from that region. They're not as potent. Ionis is using MOE version. When we've made MOE versions of ours, they're 10- to 100-fold less potent. It's just not as potent. So while the LNA types that we use have more risk of toxicity, the potency is so much greater that the therapeutic window difference matters, right? The truth is, if we're working at the 5- to 14-milligram range, the chemistry won't matter. If we had to go up to 100 or higher, then we'd have a problem. Right now, where we're working, we're a very good therapeutic range and it's so much more potent MOE that right now, I couldn't see any reason to go to that chemistry. So right now, we're not concerned on Ionis. I think we're in control of what will happen in Angelman if we get our job done. I think we'll be the first Angelman ASO. Next question?

My name is [ Asim ] with Joon Lee at Truist. On the Duchenne program, you mentioned that it will be a Microdystrophin. Could you elaborate on the nature of this molecule? For example, does it contain the nNOS domain? And then back on setrusumab, growing bones need both resorption and formation to accommodate growing organs at the skull for brain and rip cages for the heart and lungs. Do you think long-term use of setrusumab is safe and growing children? Or do you think there might be a time limit?

Okay. So for dystrophin, the version we are in does contain an opining domain. It is the same as the solid companies version that Microdystrophin and as partly, we said getting the space at that looked like the best Microdystrophin. And the nNOS clearly in MDX mouse provides much better long-term function because your muscle doesn't work well as it induces blood flow at the time of muscle activity, right, trustable fatigue. So we think the nNOS is definitely a piece that's important, ultimately, in the dystrophin story. On the question on setrusumab, your question was, if you're resorption important in growth, would there be a sake you should have in driving growth. What I'd say to you what's happening with setrusumab is, I would look at these kids as having a dial on anabolism versus reabsorption. Their dial is set, too much resorption not enough anabolism. We're basically dialing it up. It doesn't eliminate the ability to have resorption where it's needed, it does not. The anti-absorptive effect is modest. It's not like [indiscernible] in fact. [indiscernible] in fact probably are too strong in anti-absorption because that means there are times when you do need it. So if you have setrusumab on board alone, I think you'll dial up anabolism but you don't eliminate the ability to have important resorption. The question people have said, "Well, could you cause excess bone to be made?" Certainly, it's possible. And our view on the bone is we look at BMD over time, and we've talked about this with PIs. Once kids are achieving near normal bone marrow density and their fractures have stopped, we can back off on the treatment intensity to go every 2 or every 3 months and create a maintenance regimen. Our view right now is to work the mains regimen on an individual case basis to develop that regimen. We'll be looking at our Phase II patients and make that call and be the first patient, we probably switch on to a maintenance regimen to see how they work and monitor each kid and try to learn from an adaptive individual design rather than trying to do parallel analysis. With the Phase III patients as they come off will have a chance to look at maintenance dosing. But I think that would be the smart way is once they get to enough BMD to get the clinical effect we need then back them down to a level that would get what they need. But I'm actually not concerned about unsafe resorption. We're not driving it so hard that, that would cause that. But right now, we keep watching for it, of course, as always. Another question? Was there one?

Priyanka again for Anupam Rama at JPMorgan. I just had a scientific question regarding the PPCA beta amyloid interaction. Is there more color on the MOA in the sense that is it increasing phagocytosis within the cells to lead to the intake of the beta-amyloid? And is it almost, I would say, like is it almost dose-dependent where if you have more PPCA, does that lead to more increased beta-amyloid intake?

Yes. So what I think it's missing is that the beta amyloid is actually getting in the cells already, it's just coming in killing cells going out. It actually is coming into cells through autophagosome-type approach or just adhering an endosomal uptake. So I think you're seeing it being taken up. It's just not effectively [indiscernible] processed. So we're not optionalising increasing. What the monoclonals do is they do bind it and help it go in. But what the monoclonals have to do is induce macrophages to be activated to acidify their lysosomes to make their ability to digest the amyloid. So in order for the antibodies to work, you have to induce inflammation, which means you have to get close to the REA risk in order for them to work. In fact, if you don't have an acidic region, just getting it in the macrophage won't work. You have to turn on the macrophage, turn on the inflammation. So it's always a dance between too much inflammation and enough inflammation, right, to get to work. In our case, we're loading the cells to clear amyloid that is ending up inside. We didn't do anything to bring it in and the enzyme does not do anything regarding bringing amyloid in. It's coming in on its own. Except this time, it's staying particulate, not forming plague, not killing cells because we're starting to digest it when it gets in. So that's what's different. We have more -- there's more information we can look at on the sort of turnover of amyloid. But it's clear the plaque itself, people say is not a problem. The plaque is a source of amyloid that then contributes to uptake inside cells, right? So think of it as a source of a toxic fragments that are causing the problem. So that's the way we would look at it. Is that it? Very good. Well, I thank you all for coming. Hopefully, you've had a great day learning about the latest OI data, Angelman data, Wilson data, Pinnacle platform and maybe a little bit of Alzheimer's, for excitement, just in case you weren't expected, tell your friends. I feel -- look, Ultragenyx has an incredible inflection point as a company. I've shown you 6 late-stage programs, the most productive company in the business, and we look forward to your support going forward way to treat as many kids as we can. Thank you.