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

Excellent. I'm Dan Ahlquist, I'm the healthcare specialist at BofA based out of New York. I thank you for joining us at the first day of the London Healthcare Conference. Very pleased today to have from Ultragenyx, Eric Crombez, Chief Medical Officer; and Joshua Higa, IR. Maybe to start, we'll kick it off with a general question about overview of the commercial clinical programs as the company focuses on the rest of the year and early 2025.

Yes. So on Ultragenyx, we're about 14 years old now. Really founded by our current Chief Executive Officer, Emil Kakkis, to really focus on rare disease and really those rare diseases, really high unmet medical need. So really at a great stage for the organization. We do have some mature commercialized products out there that have been successful and continue to grow. But then certainly a really busy pipeline of drugs that have been coming through. I've personally been with Ultragenyx for the past 7 years. And in that tenure, we have not had anything fall out of the pipeline, which means there's been a lot of maturity with a big stack up in our pivotal trials with really looking at 3 filings in the next 12 to 18 months.

Great. So let's dig into some of the programs. So first and foremost, let's dig into Angelman. Can you provide us an overview of the GTX-102 program, discussing the complexity of the neurodevelopment disorder, current unmet need for safe and efficacious treatment, the mechanism of action, and then we'll dig into a little bit around the end points because I know there was a lot of debate around that before you have the data earlier this year.

Yes. So Angelman, again, really exciting program that we're really looking forward to getting into Phase III by the end of this year. So GTX-102 is an ASO and antisense oligonucleotide. And what that drug does is it knocks down what's called imprinting. So with Angelman syndrome, where you're really looking at a unique situation where you inherent 2 CACT to the gene, one from each parent. And normally, in all of us without Angelman syndrome, your paternal copy of the gene you inherited from your father's imprinting. It's methylated in lockdown, not producing protein, which means all of the UBE3A, which is the protein we're talking about here is expressed from the mother's allele. If you have any type of mutation where you're not producing UBE3A, it results in Angelman syndrome. This is a protein that's involved with synaptic communication but also somewhat unique to a lot of neurologic disorders, you're neurons to remain intact. They're not dying off. They're not accumulating damage. They're there and ready. And if you can provide a functioning copy of the UBE3A, the synopsis can start communicating with each other, allowing these children to develop. So that's exactly what we're trying to do here with this ASO, we're trying to knock down the increasing from the gene copy from the father to allow for fully functioning normal UBE3A protein to be produced, travel to the neurons, allow these synopsis to start communicating with each other, allowing these children to grow and develop on their own curve. That's what we've been looking at in the results. We have been recently publishing. We had big presentation end of last year with a data update earlier this year, really looking at all of those typical developmental domains that you would look at in all children, whether it's cognition speech, behavior, gross motor, fine motor and really looking for these children to start gaining and sustaining the ability to attain new skills, they're developing for the first time.

And what would you say in having conversations with KOLs are the biggest priorities for the patient's family in terms of what they want to get from therapy? Is it sleep, communication, cognitive behavior, gross motor skills, et cetera?

Yes. I think it's probably a little bit of a mixed bag. We do talk to the regional organization. We talk to the families. And I think to some degree, people have their own choices. I would say of those domains, probably where you see a meaningful amount of variability is in sleep. You have a meaningful group of patients who do not follow asleep, they do not stay asleep. And if you think about what a newborn does with the family with those type of things, imagine that lifelong over a year. So that's not only debilitating for the patient, but the entire family. So for those families, sleep can be very, very, very important. But then certainly, when you're seeing all of these patients, gain in cognition, gain fine motor, it's not a 2-hour feeding process with a 0.5 hour cleanup. They can start using utensils. We've showed a video of a patient's cooking with their mother. Gross motor, they're now able to emulate better. They can walk on even surfaces, they can walk on sand. So for me then, all of these patients making progress of all the -- across all these domains is important to that.

And maybe you had referenced some data present end of last year, recently had some data at AAN. Maybe just discuss the kind of clinical function data that you saw, how it compared to the expectations. Just give us a little sense of that presentation.

Yes. So these are the patients for the Phase I/II part of the trial, we had our initial dosing cohorts, which we call our number of cohorts and then the dose expansion cohorts, our letter cohorts. And again, certainly, last year, it was great to see really meaningful improvements, but really important to see this year that patients are not plateauing. They are actually on their own developmental curve. They are continuing to gain new skills, and that's exactly what we found over time.

Got it. You mentioned that you're going to be starting the pivotal Phase III Aspire study by year-end. Can you show some of the feedback the FDA gave you regarding the proposed trial design? And what is still any factors to assume to that trial?

Yes. So we've been talking to the agencies, not just the FDA, but other key agencies as well throughout the Phase I/II period. So a lot of conversation on endpoint study duration patient population. So certainly, going into the end of Phase II, we had a very good idea of what we wanted this Phase III to look at. The only meaningful change we added coming out of the end of Phase II was as opposed to going from a full, full blind. We adopted a sham control. And the only key difference there is, instead of injecting artificial CSF, instead of drug to those patients, we will be doing a pinprick to duplicate what the injection site would be in actively controlled patients. So it's technically a sham-controlled study, but otherwise, still aligned randomized study. We're looking at 120 patients, cognition as a primary endpoint, roughly around 48 weeks in duration.

Okay. And you're running a separate trial as well, the Aurora trial to evaluate the drug in other Angelman genotypes as opposed to one larger study. Can you maybe talk us through the thought process there of that additional separate study?

Yes. And we -- as a principle, we don't like to, if at all, avoidable test new things in a Phase III trial. So the Phase I/II was based on patients between 4 and 17 years of age with full deletions. That is your most severely affected group of patients. You're going to see a consistent response, consistent phenotype. And that group of patients makes up between 70% to 80% of patients with Angelman syndrome. So we think that's important. But certainly, we have a drug like this with a disease like this that with absolutely no treatment. You don't want to leave patients behind. So what that second study does is brings patient population to 2 years of age, and then allows us to study patients up to 64 years of age. We're not going 65 and above just because there's really not patients there, and you need to do a lot of additional work if you're in the geriatric age range. In addition, then we bring along those other genotypes. So they are missense mutations, uniparental disomin printing defects, which again is your other 20% to 30% of the population. So between those 2 studies, ages 2 to 64 across all of the genetic changes affecting the center patients.

Got it. Okay. Maybe kind of shifting gears to the primary endpoint. You mentioned cognition. It's mainly for cognition. What would be a clinically meaningful full result for that endpoint?

Yes. So obviously, it's important. We -- yes, we get a lot of patient input on that. I guess, we talk to investigators, Yes, we talk to the agency there. And really, what we're looking for at that week 48 time point is really replicating what we saw in the Phase II there. So that type of change, really everyone without debate degree, that is a clinically meaningful change. And that's really how we powered the study, how we select our endpoint and study duration. If we were powering the studies solely on cognition, we wouldn't need anywhere near 120 patients, but we do want a very big global footprint here for these studies. We do want a lot of patients. We do want a robust total safety data because we've decided to include 120 patients on the study.

And maybe just thinking about how Bayley-4 can be dependent on patient engagement and motivation, what's your expectation for the placebo response. And the reason I asked the question is there a risk that it could have a higher response than the clinically meaningful flat-point cutoff since investigators, patients, families, et cetera, know what the endpoint is going to be.

Yes. I mean I think it's something we need to be aware of. Certainly, we've talked a lot about this. Certainly, we would need to do all of this testing when patients are fresh when they're first coming in the clinic. Certainly, we need all of these assessments done before you introduce sedation. As part of dosing, we are making a change of, I think, a great addition to the Phase III, where we're using an outside agency for the Bayley testing. So these are not just specialists who we do a lot of training within the hospital that people who really focus as their day-to-day job on the minister of the -- administration and scaling of the Bayley. So really driving for consistency and response there. And then for all disease, but for rare disease, particularly, we always talk about the benefit of natural history rarely do you actually have a robust natural data set. We have 2 different databases here that allow us to really understand what's going on in the natural history. And really somewhere between 2 and 3 years of age. These children really stop developing, they stop gaining new skills. It's a really flat line by natural history with the Bayley-4 cognition.

And on the secondary endpoint, can you help us understand the different components of the MDRI? Secondary, how did you decide on what the minimally important difference would be for each [ supplement ]?

Yes. So MDRI multi-domain responder index. So a very actually straightforward tool which I found surprising early on the agencies had some resistance to it. I think we've made a lot of progress. We had a really great conversation with the FDA a Phase II where they are getting on board with this. I think it is fair to say they don't like things that are new or novel and untested, but I think we've made a lot of progress. I think those tools are really great and important when you do see variability in a disease and an indication. And I think with most neurologic indications, you're going to see variability there. So you have all of your same domains, cognition, behavior, speech, et cetera, and you come up with your clinically meaningful change. And yes, we started that work, and that is a very specific conversation with the agency. What is a meaningfully clinically important threshold for each. And then it's really simply you score your individual changes in a total rollout number in that way if some patient has more problems to see another dozen, but then you have that balance with cognition, communication mode or what have you. So you rolled up scores all being open for. So not fancy statistics, nothing you can really play with there. It's just a really straightforward total scoring method.

Great. And then maybe finally, can you show details on how the lower extremity weakness AE was managed in the critical time line for onset resolution. I felt like in advance of that of the data, there was a lot of focus on safety and emerging from the data, there was a lot of focus on the robustness of the efficacy. So maybe just on that AE and just the safety profile overall.

Yes. And I think that really speaks to the importance of dose finding and stepping through your different doses and not getting ahead of yourself. So I think with the doses that were used early on and same radiculopathy and lower extremity weakness in the first 5 patients, and not at that point, understanding what we are seeing it, it really caused quite a bit of delay and certainly a lot of conversation. So we're really at the point now where we understand ASOs, we understand the chemical nature of these products and the potential for local irritation at the injection site. So you can imagine when you're doing an intrathecal injection, there is a point in time when you have the majority of that drug really at a localized site. That can cause local irritation, and we're seeing the lower extremity weakness really related to the loading doses, which are once a month. So I think there's a potential for a little bit of irritation to accumulate over those 4 doses resulting in the lower extremity weakness. So we added Lash, which is just a really simple way to try to get drug from a concentrated injected site into the CSR. We are using Trendelenburg positioning after dosing, which is just putting that patient at a site angle letting gravity help with distribution of the drug. The patients are recovering from sedation during that time anyway, so that's fine and no real additional burden there. So really just trying to get a very equal distribution as quickly as possible. And since we've applied those measures, the incident has come way down. One of the 3 new events we had seen it was only picked up on the retro spec because we saw elevated protein and then had additional conversations. And then the other 2 were very mild in nature, quickly resolved and all of the patients are very -- families are very eager to get back on to them. So I think it's something at this point thing is well understood and to the extent possible and indicated.

And would you say there's any specific risk factors that you've identified that might put child at more risk?

I mean, I think why some children are allergic to everything and some nothing. So I think some people's immune responses react differently to all these things. But in general, it's -- we really think it's just that local irritation, not the ASO.

Okay. Any questions from anybody on Angelman?

Yes. Maybe just a question on that side effect and how it impacts with dose. And do you think you need any [indiscernible].

Yes. So I mean, we're certainly impress and excited about the results we're seeing across all domains. I think to that question, we think it is important to try to get to as high of doses as we have agreed to within banding with the FDA as quickly as possible. We had to start a little bit slower and work our way up. And that's why in the Phase I/II, we really wanted to get patients on maintenance doses closer to 14 than 10 for exactly that reason. As we move forward in the commercialized setting, and we can look at larger sets of patients, I think there is potential to do individualized dosing and potentially going up on that in a meaningful way. To your point, if the children are starting to develop, you don't want to leave efficacy on the table. So if we can optimize, we think that's important.

[indiscernible].

As far as monitoring for the lower extremity weakness or for cognition or...

[indiscernible] I mean, is there any biomarker you can use to measure how much [indiscernible].

Yes. We think -- and certainly, it's been talked about quite a bit over the past couple of years, EEG can be very helpful. Certainly, we'll move a lot more quickly than the other endpoints, but certainly looking at cognition or motor is going to take some more time. So we do think EEG particularly adult power can be -- can potentially become an important biomarker there and help with dosing. To all of us, I think looking at Angelman have all been focused on EEG more recently. So I would say, yes, we are looking at that, but I think it's fair to say, I own this, and I think previously [indiscernible] was looking at that as well as is the field.

Excellent. Maybe turning the page to osteogenesis imperfecta. Can you provide an overview of the program in OI, how the unmet needing space rationale for cross an MOA and the most recent 14-month update?

Yes. So great. I mean OI, it's been exciting. It's been exciting to us for a long time. It's great to see the conversation really coming into balance between Angelman OI. Certainly, OI is first. It's coming first. We announced at the beginning of the year that those 2 Phase III studies were fully enrolled. So OI is a fully human antibody to sclerostin. Sclerostin drives your balance between bone building and bone breakdown for remodeling and those your osteoblast that build bone in your osteoclast that break down bone. And sclerostin keeps that balance. So children could grow normally. We as adults can constantly remodel our phone. What Cetuximab does by blocking sclerostin is really kind of quiet style in the osteo class. You're not breaking down, but more importantly, really drives what are called osteoprogenitor cells to become osteoblasts. So you have more cells, building bone, laying down new stronger, denser bone. And that's why it's important for us to be looking first at bone mineral density by Dexa. That's exactly we want to see. We want to see more mineralized denser bone. I think we want to see that leading to reduction in fractures. Osteogenesis imperfecto with at 1 time and still sometimes refer to as vertebral bone disease. I mean with type 1, these are patients who are fracturing with very minimal trauma, certainly something that would be a bump for us, or type 3s and 4s. Your more severe end of the spectrum. These are children who fracture during sleep, meaning just the force of twisting over as you're turning over in sleep is enough to fracture a bone. They can fracture from moving from the wheelchair to the car. So they are very, very fragile children. And with the results, we started talking about the end of last year, talked in more detail on a longer period of follow-up, bone mineral density continues to increase. We talk about scores with bone mineral density, and a really nice way to think about it is that is talking about standard deviations below normal and then your ability to move towards normal, and we're seeing patients at the most recent time mark, at least by mean and median a full standard deviation and improvement and that's with, on average, a 14-month follow-up period. And then importantly, individually fracture rates are coming down. They're going from fertility, nontraumatic fractures to dramatic fractures, disappearing completely in a good number of patients with an overall annualized fracture rate of 0 at this point.

And what percentage of these patients are on background by phosphates?

So not a lot of grades data on that. Bisphosphonates are not approved for use in the U.S. They are approved in some countries, I believe they are not approved in the U.K. They are approved for use in Japan. No really well-designed controlled Phase III study there. So we do understand from talking to people and patients that a meaningful number of patients are in bisphosphonates. Again, you never want to make a diagnosis as a physician or hear diagnosis as a parent and have absolutely nothing to offer. So looking at kind of what you can do from a meta-analysis perspective. I think it's clear that bisphosphonate help, they do help reduce the risk of fractures. But with the majority of patients coming into our Phase III and our Phase II on bisphosphonate, they were still as a group on average fracturing at least 1 time a year and some patients much more frequently.

Got it. Can you discuss any of the benefits you've seen on mineral density and the expedition for the 2 long-term bone health for these patients?

Yes. No, it's interesting. And I guess it also speaks to the potential success. So if we can normalize all of these patients bone mineral density and they're not fracturing without success, there is a disease out there that it's called osteosclerosis, where your bones are becoming too dense, and that can be a bad thing. So that was always a theoretical risk. And again, we just want to make sure patients aren't getting desner and denser and denser. Again, that means we have a successful product. So we're monitoring for that, especially our Phase II patients, they've been in for quite some time now. We're not seeing that. If we need to drop these patients from monthly dosing to every 2 months, in some cases, on an individual basis, we will do that, but we've not yet seen that.

Okay. So it could be a dosing alteration if you saw that. Can you walk us through the current Phase III trial design, time line for your planned interim looks to determine if the study can be stopped early?

Yes. So again, both of those studies and that's Orbit, which is our double-blind, placebo-controlled part of the Phase II/III Orbit study, again, fully enrolled at the beginning of this year. That second study Cosmic is our head-to-head study to bisphosphonate. So at the end of the day, if anyone wants to have the debate, which is better, what is the difference, we can have a data-driven conversation there. That second study also enrolled patients down to 2 years of age versus 5 years of age in Orbit. So it does give us a younger patient population there. Orbit was originally designed to be the study for approval. It stands alone as an approvable study. So Orbit leads. We did plan based on the success of the Phase II results, 2 interim analyses. The first one, and again, it depends on you're talking about beginning end and data readout will be around the end of this year. Very high bar. We didn't want to spend a lot of alpha there. So we're really looking at Alpha spend of 0.001. But we wanted to give ourselves the chance for an early win, if we see a really clear separation between the placebo controlled patients continuing to fracture and these actively treated patients not fracturing. And if that interim analysis for Orbit is successful, that triggers an interim analysis for Cosmic. I don't think that Cosmic would be a tough sell as a stand-alone study for approval. So it kind of follows orbit. It's important again to understand the difference between bisphosphonates, but that would follow a successful interim analysis 1 for Orbit. If interim analysis 1 does not hit for Orbit or in the case that it does hit in Cosmic, then cosmic would continue. We can file an orbit alone. If Orbit IA1 is not successful, we would go to interim analysis 2, which we've been saying is a few months after nterim analysis 1, and then kind of similar thing there. If it hits, it triggers interim analysis for Cosmic. If both hit, great. We think there's a high probability of success for 1 of the 2 internet analysis hitting. If in the case, they don't hit, again, that's fine. We'd always plan to follow these patients to the end of next year, and that would be a full study readout. So regardless, we're really looking at data readout and '25 and then triggering your filing time -- filing in a verbal time line from there.

Is there any scenario where you may hit on an interim but actually continue running the study to generate that larger safety data base, which will be very important in the real world. I know you mentioned that you intend to follow the patients out, but would you hit the interim and then just continue running without disclosing it?

it hits, no. Just because to me, I'm not going to spend alpha going to do it. And again, what's important for this is we've been doing this within the Phase II for quite some time now. There's no real meaningful safety signals that we're worried about. So very, very clean database. Again, we understand this drug. We understand the mechanisms of action. So yes, we're spending off we're going to use it.

All right. And maybe just given '25 is going to be a very important year for OI thinking ahead to how commercial market could look. How much do you think you could leverage your existing Crysvita sales force infrastructure being a successful commercial company in terms of initial launch expectations?

So we've had quite a bit of success launching in the bone space. Crysvita was the program that relaunched in the U.S. [indiscernible] really quite successful [indiscernible] there's a whole lot of that [ flavor ] that we could leverage. We would have to expand the field team a little bit back to where we were with Crysvita launch condition to commercialization response. So there's maybe some small incremental SG&A increases to support [indiscernible].

Any questions on OI? We'll move over to Wilson disease. Can you give us an update on your Wilson disease program? What do you think or what should we expect to see in terms of updates in the second half?

Yes. So Wilson, it is one of our gene therapy programs, liver-directed gene therapy. So certainly pulling through the important learnings from OTC and GSD1a. Wilson is a disorder of copper trafficking. So without the functioning gene here, which is a transporter copper buildup in your hepatocyte until it starts to spill out of damaged hepatocytes, which is important because current standard of care is key leaders which can only pull a paper out in circulation after it's leaked out of the hepatocytes. We think gene therapy is an important step forward because normally what you want to do with copper in the hepatocytes, it's either pump it into the bio system for excretion out of the body or loading it on to a protein called to ceruloplasmin, which then allows copper to safely traffic in the body for the use of copper as an important pro packer for a lot of enzyme. And we think that it's the basis of some of the remaining neurologic signs and symptoms for patients who by measures are well managed aging. We think we need to get copper to the central nervous system for it to act as that cofactor. So we think an important difference there. We -- based on the learnings from our earlier programs, we're doing a seamless Phase I, II, III. And why that's important is because we're not going to use the time frame to Phase II in scientific advice, which honestly can be a minimum of a year in development time. We're really looking to take our learnings into what we're calling Stage 1, which is an equivalent of Phase I/II select our dose and then immediately seamlessly move into what we're calling stage 2, which would be equivalent to Phase III there. We have talked about an interim analysis data readout in the second half of the year, which is clearly where we're at, and that data set will allow us to make decisions for what will happen with Stage 2 of that program.

Can you maybe help us level set some expectations for the update, and we're going to be looking for specifically to got the dose selection decisions?

Yes. So I think very similar to the data readout we put out as part of Analyst Day last year. Certainly, the defense will be is now we've obviously now fully enrolled the 3 cohorts of 5 patients with 3 different doses. So it will be a fuller picture there. But what's nice about Wilson NAND because it is a disorder of copper. We can measure copper in a lot of different ways. You can measure it in extrusion in urine, where it spills out bound to ceruloplasmin, not bound to cereuloplasm in the plasma. We are doing liver biopsies, so we can stain for copper. So on the whole, we can really understand this as an important biomarker. And then as a clinical endpoint, and this is agreed to with the agencies, the reduction of chelators and think the reduction of standard care becomes your clinically meaningful endpoint.

Great. Any questions on Wilson? Awesome. Maybe we move to GSD1a. Can you discuss how you're thinking about the DTX401 program in GSD1a, why do you think efficacy can improve in the real-world setting?

Yes. So GSDIa, we kind of sample that we brought in for Evan. So the technical is our lead gene therapy program. As far as, I guess, our homegrown gene therapy programs, liver directed at GSDIa, and we had that Phase III data we had earlier this year, with really compelling clinically and statistically meaningful results there. So really a lot of confidence in that package. For GSDIa, we're looking at reduction of corn starch. And before the use of corn search, and this is a disorder of energy metabolism, these patients cannot break down glycogen to produce glucose during times of fastening. So during times of the fasting, their glucose levels dropped very quickly, just can lead to seizures and be life-threatening and sincerely life-threatening, particularly in pediatric f[indiscernible] reagents .Before these acorn structure was a universally fatal disease. These patients did not make it into adult or very far into adult. And cornstarch just in the GSDIa, it's solely broken down to glucose. So it provides that constant source of glucose, but still needing to be taken every 2 to 4 hours, and that includes overnight. So looking at the reduction of standard of care, in this case, corn starch as a clinically meaningful endpoint. But in a blinded fashion, when you do not know if a patient is on active drug on gene therapy or in placebo every one, and I think rightfully so it was very hesitant to reduce cornstarch. So we saw compared to Phase II, which is open label. You know what's happening here. All patients on gene therapy. We saw them come down very quickly on cornstarch in the Phase III because of the blind, it was gentle. And in the real world and also with unblinding of the Phase III as they've rolled over, we think -- we're probably going to get closer to what we saw in the Phase I/II everyone know what's going. There's no risk over reducing a placebo-controlled patient and really getting them into trial.

Great. Will you be presenting additional analysis from the Phase III?

So when we press released on the initial results, we said these are top line and we would be digging into a very, very robust data set there and that we would be talking about these results at some point this year.

And what's the current size of the tech transferring updates to your expectation of when you expect to file for regulatory approval?

And tech transfer as far as manufacturing?

Yes.

So we really have the ability, and I think good fortunate, and I think a really smart decision to build from the ground up in a true greenfield space right outside of Boston, the manufacturing facility. So that really allows us to take control of our gene therapy manufacturing into our own hand. That's important from a quality perspective, also very, very important for our cost of goods. And the plan is over time to really shift everything into that facility. We made the decision to accelerate that move from CMOs, which was the original plan into our manufacturing facility. Obviously, you need to do the necessary regulatory steps and approvals to do that, but also gives us an opportunity to look at that cross-offer data from those Phase III of patients potentially increase the strength of that package, anything to filing. And we're saying we will be filing next year.

Yes. Yes, that transfer is going well.

Great. Maybe last couple on GSD1a. So how are you thinking about the pricing environment given the standard of care is dietary supplementation. And what's the value proposition of gene therapy in this disease?

Yes. So I mean, I think, yes, you can look at standard of care as cornstarch, but armor is not a therapeutic. It's food. And certainly, it's better than nothing, but it's not a therapeutic. So I think from pricing, we're really looking at risk to patients. These patients really feel under the gun, if you will. If they miss a dose of corn surge in this does happen in the pediatric age range really every year, where patients miss a dose of corners and can die overnight. So they are really tied to this cornstarch for live, meaning they have to take the duration of their life every 2 to 4 hours, but also it can be life-threatening if they miss a dose. So really looking at total burden of care for these patients impact to their like, but also really looking at benchmarking gene therapy is more only way.

And in terms of patient identification, how well identified are patients in the U.S. with this disease where patients treated and how large sales force would you be for launch?

Yes. So these patients, this is not one of those rare diseases where you talk about this diagnostic odyssey where patients are floating out there for years trying to get an accurate diagnosis because they are presenting in the pediatric age range that are presenting with really, really low glucose levels and with seizures and very, very sick. So you're going to need to work through that differential diagnosis. So I think by and large, they are diagnosed. They are known to centers of excellence because it takes a team right now, you need dietitians, nurse practitioners, nurses in addition to the physicians to make sure these patients are well controlled glucose monitoring 24 hours a day. So it's a big team, you need to keep these patients under control. So I think well known, well identified. And then for us, I think sales force wide, we can really depend on the team we built for FAOD, a lot of overlap, a lot of similarity there.

Okay. And maybe the last couple of minutes here, we can shift focus to Sanfilippo syndrome. As the pre-BLA meeting with the FDA being scheduled at this point, what will the key points of discussion be at the meeting? How soon after that meeting, will you look to update us?

Yes. So we've talked about filing by the end of the year, which means we certainly need to complete all of the necessary conversations ahead of that. Certainly, we will provide an update at the right time. We don't always guide exactly to every regulatory meeting. But certainly, when we hit key milestones, we'll make the updates.

And just to remind us the overall market opportunity for Sanfilippo?

Yes. So Sanfilippo, again, one of those diseases where these children at birth look normal, they appear normal. But actually very early on and with a few physicians probably around 6 years -- 6 months of age or younger, start noticing problems and they really start to stop developing. But in comparison in age, really drop off. They really start to lose skills and accumulate a lot of neurologic damage very quickly. So very high [indiscernible], nothing you can do for this to just a lysosomal search disorder. So with the gene therapy, really promising results that we saw early on, we have changed the conversation with the agencies from a clinical endpoint, which would have taken quite a bit of time to accelerated approval based on a biomarker heparan sulfate, which is great for this disease because that is directly what accumulates because of the enzymatic deficiency, and it is also what is the toxin causing the damage here. So if we can show clearance and reduction of heparan sulfate, that's a meaningful surrogate endpoint, and it does correlate well with -- again, with the Bayley looking at cognition here. So again, filing at the end of the year, we think there's probably between 3,000 and 5,000 patients in the territories, and we cover and we always think the U.S. is probably somewhere between 20% and 25% versus that number.

Excellent. With that, thank you, Eric. Thank you, Joshua. Appreciate it.