Ultragenyx Pharmaceutical Inc. (NASDAQ:RARE) Q4 2024 Earnings Call Transcript

Ultragenyx Pharmaceutical Inc. (NASDAQ:RARE) Q4 2024 Earnings Call Transcript February 13, 2025

Ultragenyx Pharmaceutical Inc. misses on earnings expectations. Reported EPS is $-1.39 EPS, expectations were $-1.32.

Operator: Good afternoon, and welcome to the Ultragenyx Fourth Quarter and Full Year 2024 Financial Results Conference Call. At this time, all participants are in a listen-only-mode. At the end of the prepared remarks, you will have an opportunity to ask questions during the Q&A portion of the call. It is now my pleasure to turn over the call to Joshua Higa, Vice President of Investor Relations. Please go ahead.

Joshua Higa: Thank you. We have issued a press release detailing our financial results, which you can find on our website at ultragenyx.com. Joining me on this call are Emil Kakkis, Chief Executive Officer and President; Erik Harris, Chief Commercial Officer; Howard Horn, Chief Financial Officer; and Eric Crombez, Chief Medical Officer. I’d like to remind everyone that during today’s call, we will be making forward-looking statements. These statements are subject to certain risks and uncertainties, and our actual results may differ materially. Please refer to the risk factors discussed in our latest SEC filings. I’ll now turn the call over to Emil.

Emil Kakkis: Thanks, Josh, and good afternoon, everyone. 2024 was a pivotal year for the company as we advanced 6 late-stage programs in serious genetic conditions, most without any approved therapies, while also expanding access and growing revenue from our four commercial products worldwide. In the middle of the year, we increased our guidance for total revenue and now confirm that we exceeded the upper end of that range for 2024. After filing our first BLA for Sanfilippo gene therapy earlier than planned, we now expect to have a second BLA for GSDIa gene therapy submitted to the FDA in mid-2025. And if both are approved, we would have 6 commercial products on the market. Combine that progress with the expected Phase 3 data on UX143 in osteogenesis imperfecta and expected full enrollment of our Phase 3 for GTX-102 for Angelman syndrome, we are set for a strong year of value creation greater than any year in our company’s history.

Our international growth in 2024 was particularly impressive. We successfully launched EVKEEZA in Europe, Canada and Japan, while broadening access to our other commercial therapies in Latin America, Canada and Turkey. We’ve also been working within these regions to establish clinical trial sites to prepare drug submissions with regulatory authorities and bring our therapies into more geographies through named patient programs. We’re looking forward to another year of strong global revenue growth in 2025, supported by multiple products in launch mode globally. This progress sets us firmly on a path toward full year GAAP profitability in 2027. In January, I discussed in depth our priorities for this coming year, so I will use our time today to focus on our UX111 program for Sanfilippo syndrome, which has the potential to be our next approved product and our first commercial gene therapy program.

Last week, we presented important new clinical data at the WORLDSymposium in San Diego that were also included in our BLA submission last December. We filed for accelerated approval based on the substantial and sustained decrease in levels of heparan sulfate and cerebral spinal fluid, or CSF HS, following treatment with UX111. CSF HS is what I call a disease-caused biomarker because it’s directly responsible for disease pathology and progression. It’s not just a random measure associated with disease, it is the disease. These new data show that the sustained reduction in CSF HS exposure is statistically associated with significant continued growth in the Bayley III Cognitive Raw Score for the subdomains of cognition, receptive communication and expressive communication when compared to natural history data, which show a decline during this age period.

Importantly, we also saw that older children with more advanced disease at the time of treatment were able to retain clinically meaningful functional abilities, including communication, ambulation and self-feeding following treatment with UX111 when the Bayley score is not an effective measure. As we stated in the release, we know from caregivers, clinicians and others that stabilizing disease so that a child can retain or even slow down the loss of key skills like walking independently, communicating and self-feeding has a profound impact on their quality of life. The early we treat, the better is the long-term outcome. And in the long run, newborn screening will identify patients at birth and potentially enable an optimal treatment outcome.

Whether treatment from birth or later in life, these gains or retention of function with UX111 treatment are remarkable compared with the progressive loss of function expected in these patients. These results give us – give me confidence that UX111 will be a successful product once approved with the potential to make a meaningful difference for patients with Sanfilippo syndrome Type A and their families. Our UX111 program also serves as a strong example of how we are leading and driving changes for the field. Our progress in Sanfilippo and the progress of other companies in the MPS field is made possible by FDA’s willingness to accept CSF HS as a primary biomarker endpoint. Last February, we joined patient advocates, regulators, academics and industry representatives in a workshop hosted by the Reagan-Udall Foundation to discuss qualifying biomarkers in support of rare disease regulatory pathways.

This was an opportunity to take decades of work by academic researchers and clinicians in neuronopathic MPS diseases and put together a body of information to provide support for leveraging accelerated approval to change the paradigm for drug development in these diseases. Given my long history working on treatments for MPS diseases, including 4 of the 5 currently approved enzyme therapies in the U.S. This was an incredible achievement and opens up the possibility of accelerated development for broader – for the broader rare disease community, especially those impacted by metabolic diseases of the brain. We’re pleased and thankful to see the FDA’s focus on the rare disease over the past year with the advancement of first-ever treatments, some of which were at risk of being shelved entirely.

We will continue our advocacy and engagement efforts to advance rare disease regulatory policy. With that, I’ll turn the call over to our Chief Commercial Officer, Erik Harris, to provide a more detailed update on the progress across our commercial portfolio.

Erik Harris: Thank you, Emil, and good afternoon, everyone. 2024 was a strong year for the CRYSVITA franchise. Both the Kyowa Kirin team in the U.S., Canada and Europe and our team in Latin America and Turkey delivered outstanding results and led to the revenue beat, raise and beat that Emil mentioned earlier. I’ll start with CRYSVITA in Latin America, where we lead the commercial operations. In 2024, our team generated approximately 290 start forms that led to approximately 250 patients on reimbursed therapy. We now have approximately 750 patients on commercial product in the region as the team continues to exceed our expectations. Similar to what we saw in the United States, once doctors in Latin America see how well their patients are doing on therapy, they frequently write prescriptions for their other patients.

Growth in the region has accelerated following successfully negotiating reimbursement from the Brazilian and Mexican authorities, the two largest payers in the region and continued expansion in other South American countries. In the U.S., our partner, Kyowa Kirin, is leading commercialization and had a strong fourth quarter, driven by growing underlying demand. Over the course of the year, the number of start forms generated exceeded our expectations as did the number of patients on reimbursed therapy. We believe this confirms there are still a large number of patients who could benefit from this therapy and gives confidence there are still meaningful opportunities to grow this product. Moving on to DOJOLVI. Growth of new start forms in the fourth quarter continued to steadily increase.

In 2024, our team generated approximately 120 start forms that led to approximately 105 new patients on reimbursed therapy across the U.S. This brings the total since launch in 2020 to almost 575 patients on reimbursed therapy. The split between pediatric and adult patients continues to be approximately 65% PEs and 35% adults. The number of new prescribers continue to grow in the fourth quarter with approximately half writing more than 1 prescription. For DOJOLVI across the EMEA region, revenue is a result of named patient sales requests. There are over 250 patients treated under MPS across 14 countries in the region. The majority of demand is in France, but we are receiving an increasing number of requests from other countries in the EMEA region, including the Middle East.

The demand for this product is quite strong in this region, especially given we are not actively marketing the therapy and simply responding to named patient requests. I’ll close with a few comments on EVKEEZA, which we have been launching outside of the U.S. over the past year or so. In the EMEA region, we have patients from all of the major countries, including France, Italy, Germany, Austria and the Middle East. We now have treated approximately 190 patients across 14 countries. This is a result of our commercialization efforts or responding to named patient requests as we continue to successfully navigate the country-by-country pricing negotiations. In Japan, the team continues to build on the launch momentum following the pricing and reimbursement approval that we received last year.

We expect Japan to contribute more meaningfully to the total revenue as we launch EVKEEZA and our future programs in this country. As I have mentioned on previous earnings calls, we continue to expect quarter-to-quarter variability in revenue, primarily due to uneven ordering patterns for CRYSVITA and Latin America, but we remain confident in the growing underlying demand for all of our products around the world. Our teams were excited to deliver strong results in 2024, and we are looking forward to potentially commercializing 2 or 3 new products in the coming year or so. In the U.S., our Inborn Errors of Metabolism team that is currently commercializing DOJOLVI and MEPSEVII look forward to being able to promote UX111 and DTX401 as transformational gene therapies once approved.

For UX143, we will need to build out the U.S. team in preparation for a potential launch, and we will be able to use our deep institutional knowledge from our successful CRYSVITA launch, where there is 90% overlap in call points. With that, I’ll turn the call to Howard to share more details on our financial results and guidance.

Howard Horn: Thanks, Erik, and good afternoon, everyone. I’ll focus on full year 2024 corporate results and our 2025 guidance. Starting with total revenue. For 2024, we reported $560 million, representing 29% growth over 2023. CRYSVITA contributed $410 million, including $249 million from North America, $135 million from Latin America and Turkey and $26 million from Europe. In total, this represents 25% growth over 2023. If you focus on Latin America and Turkey, where we are responsible for generating sales, this represents 78% growth over 2023. Turning now to DOJOLVI. It contributed $88 million, which represents 25% growth over 2023. EVKEEZA contributed $32 million as demand continues to build following launches in our territories outside of the United States.

And MEPSEVII contributed $30 million as we continue to treat patients in this ultra-rare indication. Total operating expenses for the year were $1.1 billion, which included R&D expenses of $698 million, SG&A expenses of $322 million and cost of sales of $77 million. Operating expenses included noncash stock-based compensation of $158 million. For the year, net loss was $569 million or $6.29 per share. As of December 31, we had $745 million in cash, cash equivalents and marketable securities. In 2024, net cash used in operations was $414 million. Importantly, in 2025, we expect reduced net cash used in operations compared to 2024. As you remember from prior years, we typically use more operating cash in the first quarter than in the other three quarters because it includes items like the payment of annual bonuses.

This year, the first quarter will also include $45 million in payments for two milestones that were achieved in the fourth quarter of 2024, one for the initiation of our Angelman Phase 3 study and one for a sales milestone for EVKEEZA. We will continue to prioritize expense management and focus our investments on the execution of our upcoming commercial launches and advancing our Phase 3 programs. Shifting to revenue guidance for 2025. Total revenue is expected to be between $640 million and $670 million, which represents 14% to 20% growth over 2024. Drivers include increasing demand for our products in Latin America, continued penetration of the pediatric and adult XLH markets in the U.S. and growth from EVKEEZA in Europe and Japan. CRYSVITA revenue is expected to be between 600 – or excuse me, $460 million and $480 million, which includes all regions and all forms of CRYSVITA revenue to Ultragenyx.

This range represents 12% to 17% growth over 2024. As a reminder, in the first quarter of the year, we restart at the lowest tier of our royalty structure in North America. DOJOLVI revenue is expected to be between $90 million and $100 million, which represents 2% to 14% growth over 2024. As in prior years, our DOJOLVI projections represent a blend of faster growth in countries where we commercialize and lower growth in countries where we respond to named patient requests. With that, I’ll turn the call to our CMO, Eric Crombez, who will provide an update on our key clinical data readouts expected this year.

Eric Crombez: Thank you, Howard, and good afternoon, everyone. I’ll provide some brief operational updates on our late-stage programs and review our upcoming clinical milestones. Starting with UX143 for the treatment of osteogenesis imperfecta. The Phase 3 Orbit study is progressing well, and the safety profile is similar to what was observed in Phase 2. Based on the Phase 2 data we previously shared, we are confident that this study will show a clinically and statistically significant reduction in annualized fracture rate at either the second interim analysis or the final analysis. The Orbit and Cosmic studies will both have an interim analysis midyear after all patients have been on therapy for at least 12 months, and the alpha spend for each study will be 0.01.

If IA2 for Orbit is not achieved, the study will proceed to full study analysis in the fourth quarter. Moving to GTX-102 for the treatment of Angelman syndrome. Enrollment in the Phase 3 Aspire study is going well, and we are on track to complete enrollment in the second half of this year. Based on the strong interest from physicians and patient families for this transformative therapy and with our team’s commitment, I have confidence that we will be able to enroll the planned 122 patients in less than 1 year. In parallel, we are also working to initiate our Phase 2/3 study, Aurora, that will study younger and older patients and those with other mutations. Once this protocol is through the regulatory process, we will share additional design details.

Shifting now to our gene therapy programs and starting with UX111 for the treatment of MPS IIIA. Emil already shared the latest data we presented last week, and we expect to receive confirmation of our BLA acceptance shortly. This will also confirm our PDUFA date, which we expect to be in the second half of the year. The biomarker data and the newly presented clinical improvements in multiple domains show the impact this therapy can have for families and patients who otherwise face a lethal disease. I look forward to completing this work and to the launch of our first gene therapy product. Next, DTX401 for the treatment of glycogen storage disease type 1a. Work on the BLA filing is going well, and we are on track for FDA submission midyear.

We have talked about the decision to move manufacturing for this program to our facility in Bedford, Massachusetts. Finally, I’ll touch on DTX301 for the treatment of ornithine transcarbamylase deficiency. As noted in our press release today, we have completed enrollment in the Phase 3 study with a total of 37 patients randomized 1:1 to placebo or DTX401. We did amend the Phase 3 protocol to allow a blinded comparison of the ammonia reduction primary endpoint through week 36 and will now evaluate the reduction and removal of standard of care after unblinding given the reluctance of patients and doctors to stop alternate pathway medications in a blinded study. This reluctance stresses the severity of this disease with the irreversible damage caused by high ammonia levels and the need for a onetime treatment option that establishes the normal ability of the urea cycle to metabolize ammonia.

The primary endpoints are unchanged and are, one, response as measured by change in 24-hour ammonia levels during the first 36-week blinded period compared to placebo patients. And two, percent of responders who were able to remove alternate pathway medications and protein-restricted diet. This responder endpoint will now be assessed across the whole group of patients, including placebo crossover patients for up to 64 weeks. The new design increases the power of the study by assessing ammonia early where reductions occur rapidly and then assessing the diet and drug removal by evaluating all patients compared to the baseline in the longer follow-up period. These changes allow us to reduce cost and to shorten the overall time line by ending enrollment at 37 patients and making the study more patient-centric.

The protocol amendment reflects the real-time feedback we heard from patients and physicians about the fear of removing life-saving treatments in a blinded placebo-controlled study. As a patient-engaged company, we need to hear their concerns during trial conduct and now have made a better trial design with their feedback. I’ll now turn the call back to Emil to provide some closing remarks.

Emil Kakkis: Thank you, Eric. In 2025, we’ll continue building on the outstanding clinical and commercial execution that put us in place to transform into a leading rare disease company with a commercial financial engine that supports its clinical pipeline. We have a number of near-term catalysts, which Eric has already covered and expect to have three potential approvals in the next year or so. As a company, Ultragenyx has arrived, we are creating the paradigm for the next generation of rare disease companies changing the future for rare disease patients. With that, let’s move on to your questions. Operator, please provide the Q&A instructions.

Q&A Session

Operator: Thank you. We will now be conducting a question and answer session. [Operator Instructions] Our first question comes from the line of Tazeen Ahmad with Bank of America. Please proceed.

Tazeen Ahmad: Hi, guys. Thanks for taking my question. So mine is going to be on OI and how we should be thinking – I think you’ve talked about this many times. But Emil, if you could give some color about your confidence in the second interim read being sufficient to stop the study versus needing to go to the third interim. I know you’ve expressed overall confidence in the trial design and the likelihood of success. But we’d like to hear your thoughts about how you’re thinking about the importance of the study being stopped at the second interim. Thanks.

Emil Kakkis: Great. Well, we believe the second interim certainly has a much greater chance of hitting than the first one because it’s a 0.01 threshold and because the patients will have had at least 12 months of treatment to allow the groups to separate. So there’s certainly greater confidence in the second interim. But we feel confident one way or another the trial is going to end and we’ll be successful this year. Just – we’ve been watching the Phase 2 patients. They’re doing great, and we feel very good about how Phase 3 conduct is going. So we have more confidence in the second interim than we did in the first. But it’s either the second interim or in the study, but we have a product that’s going to work, and we’re excited about that product and bringing it to OI patients.

Joshua Higa: Thanks. Operator let’s move to the next question.

Operator: Thank you. Our next question comes from the line of Salveen Richter with Goldman Sachs. Please proceed.

Salveen Richter: Good afternoon. Thanks for taking my question. Just a follow-up to the first one here. Maybe help us understand if it doesn’t hit on the second interim, what would those reasons be? Or what are the risks to that? And then how long would we have to wait for the final analysis? Thank you.

Emil Kakkis: Well, we said the second will be midyear. For it not to hit at 12 months, it’s usually in rare disease, it have to do with the amount of variation in the number of fractures. If there’s a lot of variation, there’s a wide range of patients baseline fracture rates because we have some type 3, 4 and then type 1 patients. Large variation could create some challenge. But I think that so far, we feel like the trial is proceeding as expected. So if it doesn’t hit in the second interim, we’d expect to release data by the end of the year on the final assessment for the trial.

Operator: Thank you. Our next question comes from the line of Anupam Rama with JPMorgan. Please proceed.

Unidentified Analyst: Hi guys. This is Priyanka on for Anupam. Thanks for taking our question. Once Orbit hits, will you immediately prepare to file or wait for the Cosmic analysis to finish and add it to the application? Thanks.

Emil Kakkis: If Orbit hits, first of all, we will have the data clean and locked in and there won’t be a need to follow any for any longer. So it will be more quickly to reach final analysis and to file. And so compared to interim mouse 1 [ph] there’s probably just a one quarter delay in filing off the quarter two. So it should come more quickly after that result.

Joshua Higa: Thanks, operator. Let’s move to the next one.

Operator: All right. Our next question comes from the line of Gena Wang with Barclays. Please proceed.

Gena Wang: Thank you. Sorry. I will ask another OI question. Given the study complete enrollment on May 1, 2024, is it fair to assume June, July time frame, you should share the second interim update? And then how would you communicate with the investor in the case the study didn’t hit the stats? And in the case the study hits stats, would you need to take extra time to analyze the data and therefore, share the update a little bit later?

Emil Kakkis: Yeah. So we will be managing the second interim like the first. A DMC will meet and look at the data. And if they don’t inform us that it hit, then we know it didn’t hit. And we will expect to communicate to the Street where we’re at, at that point in time. You asked whether it’s fair to consider June, July. I guess that’s possible, but we haven’t set the time. We said midyear is in that range, but we’re not committing to an exact time. We’re trying to keep you guessing a little bit, Gina. So did I miss – did I get your question?

Gena Wang: Yes. So if you hit it, that means like you would share with us a little bit later since you need to take a few more weeks to analyze the data.

Emil Kakkis: Yes. But because the database is locked, we would – the DMC would see the data, we would find out it hit. We would then get to unblind and look at the data ourselves and relatively sooner after that we would expect to put out results. It won’t be like the first time where we had to go 2 months more, remember, we had to take 2 more months. So it should be relatively quicker to come out with the actual result – top line results anyways.

Gena Wang: Okay. Great. Thank you.

Operator: Thank you. Our next question comes from the line of Yaron Werber with TD Cowen. Please proceed.

Yaron Werber: Great. Thank you. And so I also shockingly have a 143 question. In the Orbit study, are you stratifying just remind us type 1, 3 and 4 between the 2 arms? And then secondly, when you look at the primary fracture rates, do you have a secondary looking fracture rates by type, underlying type? Thank you.

Emil Kakkis: Yeah. So in general, we do stratify, but it’s mainly for its overall fractures and its age. But I’ll let Erik talk about the way we’re approaching it.

Erik Harris: Yeah. So because the primary endpoint is annualized fracture rate, you want to stratify by fracture rate. So while that definitely will kind of encompass the different types there, the strict stratification is based on fracture rate coming into study.

Emil Kakkis: So type 3 and 4 may have a higher fracture, but we’re focusing on that. Doing it by the 3s and 4s and 1s, we didn’t – it didn’t look like that was going to be the right way to go as fracture rate was a better way. So we are also looking at ages, so there’s an age balance between the groups. And regarding the other endpoints, we are looking at total fractures, not just the fractures minus fingers, toes, skull. Those total fractures are an endpoint. And the subset between subtypes, I’m sure we’ll do analysis sensitivities on that in there, but it’s not a formal secondary endpoint.

Operator: Thank you. Our next question comes from Yigal Nochomovitz with Citi. Please proceed.

Unidentified Analyst: Hi. This is Rena on for Yigal. Thanks for taking my question. I just wanted to ask one on OI. I was just wondering if at the point it hits, whether it be the second interim or the final, if we would learn maybe retrospectively, like any kind of retrospective analysis you’d share on how close the studies were to hitting at either the first interim or the second interim, respectively?

Emil Kakkis: Do you want us to tell you what didn’t – what happened at IA1 or IA2 later? You have like personal bets with other investors about that to try to settle. I don’t know if we would show with the other interim. We haven’t decided that. I think what matters if the study hits the study work and the drug works. And I’m right now not committing whether we’re going to show all the little bits and pieces of what happened. We’re pretty comfortable that – look, we – if you look back, we presented Phase 2 data, remember, at 6 months and showed you a 67% reduction with a p-value of 0.04. And then we did the same group of patients, but now at 14 months, right? And they had 67% reduction in fractures. But now they had a p-value of 0.014.

That is a reasonable model for what would be going on here that the groups are separating, you just need to go a little further to get the p-value where it needs to be and hit the threshold set. So I look at that Phase 2 data we’ve already shown you as kind of a model for what’s going on. But right now, we put out data – we’ll put out the data we have, and I don’t know if we’ll put out the rest to settle scores, but thank you.

Operator: Thank you. Our next question comes from the line of Maury Raycroft with Jefferies. Please proceed.

Unidentified Analyst: Hi. This is Farzin on for Maury. Just to clarify an earlier point, you mentioned that if the second interim hits, you do the final analysis quickly. So just wondering if you need to supplement the final data package with the 18 months data or not?

Emil Kakkis: No, if we hit at the second interim, then the package will include all those patients with at least 12 months of data. The range, though, would be like from 12 – there are some patients at that point would have 20 months, it’s 12 to 20 months of exposure time. So it’s a pretty long period of time, actually. It’s almost 2 years in all patients. So 12 – that will be, I think, plenty for that. I didn’t mention someone asked about Orbit or I mean, Cosmic earlier. Cosmic will be evaluated in parallel. And – but we won’t close out Cosmic if Orbit doesn’t hit. If Orbit hits, then we’ll look at Cosmic, but the data will be ready to be looked at. But we wouldn’t stop Cosmic if Orbit doesn’t hit because we need Orbit in order to file. But we wouldn’t hold out filing to get 18-month data. We’d file with the data we had. And we’ve had that discussion with the FDA.

Unidentified Analyst: Got it.

Operator: Thank you. Our next question comes from the line of Joon Lee with Truist Securities. Please proceed.

Joon Lee: Thanks for taking our questions. If the OI study goes to completion in 4Q, does that imply that the magnitude of effect – sorry, may not be as great as expected? And in that case, how competitive would setrusumab be compared to bisphosphonates?

Emil Kakkis: Actually, it would not mean it’s not as great. If you remember, earlier when we had 6-month data and 14-month data in Phase 2, they both had 67% reduction in fractures. What it has to do with is the 2 lines have to separate. So the biggest creation is there’s too much variation in those variations might cause a delay, but the actual rate separation could very well be 67%. It’s just you have a lot of patients that may have 10 fractures a year or 1 fracture year in the same study and some of the ones may not have fractured, for example, and then during for whatever reason. And so it’s really more about separating the 2 groups. But I don’t think it necessarily tells you what the percent reduction is. We think if you listen to some of the KOLs that 50% or greater reduction of fractures is considered really important.

And frankly, when we look at patients after a year, 15, 16 months of therapy, we’ve had some or longer, they are – many of them are not fracturing at all at some point. So we feel very comfortable that the long-term outcome here is greatly reduced fractures, whatever the number is. But I think the biggest issue is the variation in how much fractures are occurring in each group and that wide range that probably exists that will impact how the study reads out. We are using co-variables to manage that variation, but that would be the number one reason. So I don’t think you can conclude the drug is not working well if we go to the end. Remember, the original plan here was to do a 2-year study. And the only reason we felt we could go sooner is because the percent reduction was higher than we thought and that the speed of response was faster.

Those are the things that give us confidence that we can go earlier. But we’ve been moving this up from 18 months to 2-year study, right, down to what we’re talking about now, which is a 12 to 18-month time frame. So 18 months is still a win. And I feel confident whichever one happens that we have a drug that will be far better in bisphosphonates and certainly the best treatment for OI that’s available.

Operator: Thank you. Our next question comes from the line of Jeff Hung with Morgan Stanley. Please proceed.

Jeff Hung: Thanks for taking my question. I just wanted to clarify and make sure I understood correctly. You talked about how the Phase 2 data and the 0.014 – 0.14 [ph] But just for setrusumab, would the Orbit Phase 2 portion have hit with the second interim analysis criteria? And if not, how were the baseline fracture rates different from the Phase 2 portion? Thanks.

Emil Kakkis: Yes. So what I said was the Phase 2 data at 6 months last patient in, we had 0.04. And then with the 14-month data, we had 0.0014, right? So that was the difference. You’re asking how close does that reflect what’s going on. Well, the Phase 2 patients are fairly similar in terms of the entry criteria for fractures are the same. They’re made up of type 1, 3s and 4s. There’s – Phase 3 has somewhat more 3s and 4s, but not a dramatic difference. So it’s a very comparable population, age range, types included and baseline fracture requirements. So I think that those are reasonable ways to look at what Phase 3 should be happening. And so the only question has to do with how the variation in the population, how big is it and how much it moves in the time frame. But I think the data from Phase 2 are a reasonable model for what’s happening.

Jeff Hung: Okay. Thank you.

Emil Kakkis: Hopefully, that’s helpful.

Jeff Hung: Yeah.

Operator: Thank you. Our next question comes from the line of Joseph Schwartz with Leerink Partners. Please proceed.

Unidentified Analyst: Hey, all. This is Will on for Joe. Thanks for taking our question today and congrats on the progress this quarter. I have one on OI and then a follow-up on UX111. So just first for OI, wondering if you could provide any more color on the characteristics of the patients in the Phase 2 that had fractures while they were on treatment. Do we know the subtype of these patients? And how does that impact your thinking about the Phase 3, which has, just as you said, a few more severe patients? Thank you.

Emil Kakkis: Okay. So the Phase 2, the patients had fractures, surprisingly, they were not necessarily type 3s and 4s only. There were 17 type 1s, and they were some of the ones that had fractured were type 1. So it’s not like having type 3s and 4s made more fractures necessarily. The 3s and 4s tend to have higher fracture rates. But the ones that did have fractures were not necessarily like only type 3s and 4s, for example, it was – I think they’re mostly type 1 as I recall. So there’s no, I would say, a pattern there that would tell you something about how Phase 3. I think actually, there’s no pattern. What was probably more impressive is how much all 3 types respond very comparably in bone mineral density improvement. I think that was the most interesting thing that we saw. So you had a follow-up with 111?

Unidentified Analyst: Great. Yes, thank you. Just on 111, since it’s going to be the first gene therapy launch for you guys, how are you thinking about pulling learnings from other programs or experiences that had success or maybe not were that successful in their commercial rollout? And how does this kind of set you up for other gene therapy launches down the line, including 401? Thank you.

Emil Kakkis: We are a very dynamic commercial organization, and we’re used to launching new things in disease never been treated before. So what we’ve essentially done every single time. I’ll let Erik say a few words if you’d like. But the – we are going to be looking at what everyone is doing for sure. And our goal will be to get our work with payers upfront to get to understand the value proposition, the data we have in hand and to allow them and hopefully to create as rapid and prompt a pattern of review and approval because urgency and time really matter to these kids that a 6-month delay is not allowable. This is a very severe disease. So I would look at UX111 as being more like ZOLGENSMA and SMA in terms of life or death and no alternative treatment. I do think there’s an urgency. I think the payers will respect that. We just need to make them aware of it. And I don’t know, Erik, is there anything simple thing you can touch on how we’re going about the launch?

Erik Harris: Yeah. We’ve been in close contact with the treatment centers – potential treatment centers because remember, we’re in with those customers now with both DOJOLVI and MEPSEVII. And so we’re learning a lot about the coordination that will be required between the patients and the physicians for delivery of the gene therapy. So we’re ready. We’ll be ready to go.

Emil Kakkis: We’ll certainly want to learn for successful launches of ZOLGENSMA, and there’s a lot of people out there that have that knowledge. And I think there’s an advantage now coming out now with several products out there knowing what works, what hasn’t. And I think the truth about rare disease about gene therapy launches, it’s all about urgency. Disease doesn’t have a treatment. It’s lethal. Their adoption is going to be much faster, and we think UX111 fits that mold like ZOLGENSMA. So we’re looking forward to a good launch, and we’re excited about the opportunity of treating this disease. They’ve been ample [ph] patient waiting forever for something, and this is the first time it’s actually happening. And I know they’re grateful and excited about finally something being done for their kids.

Operator: Thank you. Our next question comes from the line of Kristen Kullka with Cantor Fitzgerald. Please proceed.

Rick Miller: Hey, everyone. This is Rick Miller on for Kristen. And thanks for taking our questions. We’re going to mix it up a little bit and ask about OTC deficiency. So just to better understand the amended protocol that you talked about today, when the patients in the Phase 3 are in the initial 36-week blinded period, are they still eligible for titrating down on ammonia scavengers and diet control? Or would that just be an option in the following unblinded period? And then based on the natural history for the disease, is there any good understanding that you have based on what a meaningful reduction in ammonia looks like and how this correlates with normalizing some of the clinical effects you see? Thanks.

Emil Kakkis: Sure. Well, we already have treated a lot of patients and know that none of them want to reduce. So we kind of know during the blinded period, none of them would reduce. So we’re – if we have the option or not, it doesn’t matter. The truth is it’s a little cleaner if no one would. But since they haven’t, it doesn’t matter. So we’re going to look at ammonia at baseline relative to later. What we know is we allowed up to 200 micromolar [ph] in the study. So there will be a number of patients that have significantly elevated ammonias, which on a chronic basis are not good. And these ammonias we’re talking about are 24-hour curves, right? So they’re mantra during the 24-hour cycle that go up and down. So we expect from our Phase 2 data that should drop within 6 weeks and very rapidly drop to normal range.

So we’d expect that the patients that have high ammonia are clearly abnormal range that clearly impair cognitive function, right, above 80 or so, should normalize. And that’s why we think we’ll have a lot of power. So I think that the degree of ammonia change in those that are elevated will be clearly from abnormal cognitive impairing type range and into a normal range where you expect greater clarity of thought and function. We’ll be monitoring those things as well. The problem, of course, with removing their drugs is they’re so afraid of crashing and because it can’t happen suddenly that we’re just reserving that after they get unblinded. And so we’ll actually have the same endpoints and really the same trial. It’s just that we’ll unblind people after 36 weeks and then they can start titrating and – but we’ll get blinded data, controlled blinded data for that period, which we think will be clinically meaningful ammonia reductions.

I don’t know if there’s anything else I missed, Erik, that you’d add.

Erik Harris: Well, I would just want to draw the parallels to GSDIa. I mean we had a lot of reluctance there in the blinded period to be aggressive with the reduction of cornstarch. But once all patients rolled over to long-term follow-up in an unblinded setting, we saw further reduction in all patients and very closely mimicking what we saw in the Phase 2. So I think we’re confident we’ll see something similar here in this Phase 3 and get much closer to a duplication of what we saw in the Phase 2, which was done in an unblinded manner.

Emil Kakkis: And the percent of responders is complete responders, so they have to get off their drug and diet control to be considered a complete responder, which is what the FDA preferred. So I think having all patients be assessed that way, I think, is a – will help the power of the study because it’s actually more patients and doing it within patient comparison is a lot more powerful than doing a 2-group comparison. So thanks for the question on OTC and breaking up the OI domination. Next.

Operator: Thank you. Our next question comes from the line of Luca Issi with RBC Capital. Please proceed.

Luca Issi: Great. Thanks so much for taking the question. I guess we’ll revert back to the mean here and maybe ask a question on OI. What’s the latest thinking on duration of therapy? Assuming this trial is successful, do you think that the FDA will cap the duration of therapy to 12 months as they have done for romosozumab in osteoporosis? Or do you think this will be a chronic dosing with no cap? So that’s one. And then second, very quickly, can you just remind us why for Angelman running a sham control trial is the better way to go versus a placebo-controlled trial? Thanks so much.

Emil Kakkis: Sure. So on the duration is one of the first things we did in taking over the program is not make the assumption that 12 months makes any sense for these patients. And what we are learning already is that the assumption that 12 months that the treatment stops working essentially, you don’t need to go longer and that you then lock it in with bisphosphonate. That model is true for maybe 80-year-old women with osteoporosis. But when you’re talking about these kids, their bones want to resorb the kids that stopped – the adults have stopped treatment before that were OI patients in the ASTRID study started losing ground. And it’s pretty clear to us that OI is just different. And when we look at patients long-term treated, they continue to gain ground in BMD for quite a bit of time.

And our view is that we should be treating until optimal clinical effect and our expectation is that will become a chronic treatment. And it’s possible down the road that they get – could go to maintenance therapy where they get less frequent dosing. But it’s pretty clear that 1 year is – you’re not done after 1 year and that most of these kids probably need at least 2 years of continuous therapy, if not, and chronic therapy is our expectation. So we’ll – the FDA is aware of that, that concept, that change, and we’ll have to help them understand why it’s the right way. I think the data we have is already telling us that that’s right, that the patients are continuing to benefit. And when you have a bone mineral density, for example, it’s minus 2 to minus 4 Z-score, while they gained a lot of ground in 1 year, they’re continuing to gain ground in the second year, clearly linearly.

So we’re very comfortable that we’re making the right call here. And it’s one thing that is important as a company is that we don’t just follow what everyone says. We’re looking at what’s really going on and look at the science and the science says these kids need to get treated longer, and we’re going to continue to do that. And we’ll work with the agency in the long term. Would there be a maintenance later? Perhaps. But I think at this point, we’re talking about chronic therapy for this disease, and that’s the best way to probably maintain the right balance of bone production versus bone resorption, which they need that constant stimulation to make that set point right, not like an osteoporosis patient, which they are artificially trying to up their bone production, but they don’t have an intrinsic problem like these kids do, right?

So when you have an underlying genetic cause, I think it’s just a different biologic situation, and we think chronic treatment is the right answer.

Luca Issi: So second question on Angelman.

Emil Kakkis: The Angelman sham versus placebo, we actually proposed placebo originally, the FDA at the time said there had considered it was potentially unethical to have to do sham treatment or placebo treatment. We opted not – and we had heard something similar from EMA. So we opted to go with the sham just because we felt there was an ethical or regulatory burden there that we have to go through, not them, just them but IRBs. But I think you can do the study either way. I don’t think it’s going to matter. Sham has more work, right?

Erik Harris: Yes. I mean the real difference between sham and placebo is we – in the placebo-controlled patients, we will not be injecting artificial CSF. So that really is the practical difference…

Emil Kakkis: In the sham design.

Erik Harris: In the sham design. So it’s really not a big significant difference between the 2 designs, but you don’t need to administer artificial CSF.

Luca Issi: Great.

Operator: Thank you.

Emil Kakkis: So we’ve managed all the prospects of blinding and execution for that. It’s a little bit more work to do sham, but it is we felt I didn’t want to battle a pill with IRBs or EMA or other regulatory authorities regarding the issue of injecting placebo and going – having every injection via an injection in these kids. So – but I don’t think it’s going to matter and won’t make a difference in the meaningful – meaning of the treatment. And we feel comfortable that the design is going to get us where we need to go.

Luca Issi: Super helpful. Thanks so much.

Operator: Thank you. Our next question comes from the line of Liisa Bayko with Evercore ISI. Please proceed.

Liisa Bayko: Hi. Thanks for taking the question. A question on the variability in OI. In your work preparing for the trial and assumptions, can you talk about any sort of registry or database views you got on variability? Is there some way to describe that? And I guess, is there a trend like if a patient has a fracture, are they more or less prone to a fracture in a certain period of time? I just would love some color around that. Thank you so much.

Emil Kakkis: Well, we know from our Phase 2 baseline data that there was a pretty wide range. There were patients who had 0.7 or 1 fracture a year and some, I think they’re up as high as 6 or 7 fractures a year. Is that right, Erik?

Erik Harris: Yeah.

Emil Kakkis: So that’s a pretty wide range of baseline fractures. We know in the younger patients, they tend to have more fractures. So if you talk about the really young one, the younger you go, the more fractures they have. So those are the factors to age particularly. And with that wider range, then you have to imagine that you have to power the study to reduce fractures across a wider array of change. So we think that we’re well powered to do that in the design. And if you look at what happened in Phase 2 over after a period of time, really the majority of the patients have not had any more fractures, stopped having fractures completely. So – and certainly, as we put in our deck, there have been – one of the doctors said they’ve not seen a single fragility fracture since starting setrusumab.

So we’re comfortable that we’re going to be able to show the delta. I was just saying if you ask me why would interim 2 not work, well, it’s a question of variation is usually the thing in these studies. But we feel comfortable that with enough time, they will separate and you’ll get an important result for this disease.

Operator: Thank you. Our next question comes from the line of Ed Arce with H.C. Wainwright. Please proceed.

Unidentified Analyst: Hi. Good afternoon, everyone. This is Thomas asking a couple of questions for Ed. Thank you so much. So perhaps switching gears to 102, as you work to finalize the design for the Phase 2/3 Aurora study, can you discuss some factors between the 2 patient populations that would differentiate the study from Aspire? And then taking a bigger picture overall strategy with 102, do you also expect Aurora would have data to supplement BLA that we expect to be based on Aspire data? I expect that sometime in the second half of 2026.

Emil Kakkis: Okay. So on GTX-102, the main study, to remind everyone, the main study is Aspire is focused on deletion patients ages 4 to 17, which is the same population we study in Phase 2. So the Aurora study is simply to support labeling across different age groups and different genotypes. For under 4, we’ll have some patients treated open label. We’ll look at the same endpoints, but just show does under 4-year-old patients, do they respond comparable that we saw in our Phase 2 or in what’s in Phase 3. And our experience in rare disease world that you can expand the label for age simply by treating a small number of patients in an open-label format and collect data that’s comparable to your blinded study. For the – we will also do that for some older patients.

For the non-deletion types, we’ll look at patients in the main range and the point will be the same to look at the different genotypes, collect enough patients to show it’s safe to give it to them and that using comparable endpoints we can demonstrate a magnitude effect that is comparable. In our experience, that kind of data would allow us to support labeling. So our intent is to have cut of data from Aurora in time for BLA filing, but we will probably still continue to collect data in Aurora during the review process in case the agency wanted more. And the other advantage of Aurora is that we can also do it in other territories around the world, giving more doctors exposure to the treatment and to gain support for regulatory filings that might occur in other parts of the world.

So those are some of the factors of how Aurora will work.

Operator: Thank you. Our next question comes from the line of Yaron Werber with TD Cowen. Please proceed.

Yaron Werber: Yes. Thanks for taking my follow up. I just had a couple, Emil. The first one on 401. So given that you’re fully enrolled now and the primaries at 36 weeks and then you unblind, is there a chance that we’ll get the data, the first sort of the primaries this year? And then secondly, I think I misheard, but I just want to double check. It sounds like you’re saying in Cosmic, you’ll do an interim analysis now midyear as well. Is that indeed – is that going to be the first interim for that? And I think you mentioned the 12 months, but with a 0.01 alpha, but I thought that was referring to Orbit. I just want to double check the interim analysis on Cosmic.

Emil Kakkis: Sure. So I think you’re speaking of 301 for OTC Yes. So the 36-week data could provide data. We’re not saying when yet. It depends a little last patient in and having 36 weeks, et cetera. But it could be. We haven’t provided the exact timing of that data yet to you. For OI, when we run the interim on the Orbit study, we’ll run an interim on Cosmic at the same time. We’ll look at Orbit. If orbit is positive, then we’ll look at the Cosmic data. If Orbit is positive, then we’ll look at Cosmic and it also is being looked at a 0.01 p-value. If Orbit doesn’t stop, then there’s no reason to stop Cosmic at this point. So we’ll keep them going until we file. But we want to do them more in parallel. So the idea is to have both trials in the filing, which is the FDA’s preference and which would give us the full label.

We also think it gives us labeling against bisphosphonates, right, as well as versus placebo, right? So they both studies add value to the commercial launch of the program. So 0.01 for both studies.

Yaron Werber: Okay. Thanks.

Operator: Thank you. There are no further questions at this time. I’d like to pass the call back over to Joshua for any closing remarks.

Joshua Higa: Thank you. This concludes today’s call. If there are any additional questions, please contact us by phone or at ir@ultragenyx.com. Thank you for joining us.

Operator: This concludes today’s teleconference. You may disconnect your lines at this time. Thank you for your participation.