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

Okay. Welcome, everyone. It's my pleasure to host this fireside chat with Ultragenyx Pharmaceuticals and very pleased to have Emil Kakkis, CEO and founder with us today. Thanks for joining us.

Maybe we can start by just having you give us your latest vision for the company and point out the key initiatives you have undergoing and how they will be producing updates for all of us over the next -- over the balance of the year.

Thanks, Joseph. Thanks for having us here. So Ultragenyx now is 14 years from founding, me and the secretary and a couple of million dollars. And we're now a company this year, crossing $500 million revenue with 4 commercial products globally. We have 6 late-stage programs including 4 pivotal studies currently. And this year will be a big year for those studies. We have a program in Angelman, which will be reading out data here in April at the American Association for Neurology, which is coming up. We have data from a Phase III study in GSD1a -- we're also unwinding this year. We have data from a Wilson disease, Phase I/II part of the pivotal study, first part, the dosing part will be reading out as well. We had data earlier in the year from MPS IIIA gene therapy showing a nice effect on brain function in those kids with MPS IIIA. And we expect to have a further update on our Osteogenesis Imperfecta Phase II study longer term data in the second half of the year. Going towards the end of the year, we also potentially have an interim -- first interim look at our OI program, Phase III study, which are both programs for OI and we are fully enrolled. So we're a company both growing revenue and having a lot of late-stage programs and a really important inflection point of the company, heading towards profitability, we think, in around '26, with revenue growth from the existing products being the majority of that and the new products, putting us in a new accelerating revenue growth stage. So I think it's a very exciting point in time to be in the rare space and the high success factory and the ability to commercialize these programs has made us an incredible success for Ultragenyx.

Yes. You've had a great run so far, and we're looking forward to these updates. And 1 of them is very near term. We saw today a placeholder abstract, implemented for the GTX-102 data at AAN. So is there anything you can say about what we might see there as a bit of a curtain raiser without stealing your thunder.

Yes. So we've announced at our Analyst Day last year that we'd have 20-plus patients with the day 170 data in the Angelman program. This is an antisense oligonucleotide, which induces expression of the UB3A gene. And we've shown at Analyst Day that these Angelman patients with significant development delay we can see improvements in cognition, receptor communication, sleep behavior, and gross motor function. And what we'll be looking for from this data set is a further group of patients all treated with the same dose level, now more than 20 patients together data that will be at day 170 so a little bit earlier from what we showed at Analyst Day will allow us to look at those same 5 domains using the key endpoints we've shown before. And so a lot of what you saw at the Analyst Day will be replicated there. We'll also include some EEG data and safety data. But we're so far very encouraged and excited about the Angelman program what we're seeing so far and its meaningfulness. And what we showed last year at Analyst Day shows that this drug has a clinically meaningful and important improvement the patients want and show that continued treatment showed continued gain of function over time. So we're really encouraged by that. The Phase II data in April, we'll touch on where we are with regulatory which is where a lot of the interest is and our expectations ahead to end Phase II around midyear time frame. And our goal would be to get to a Phase III start later in the year.

Okay. Great. And if -- given the loading doses are now going to be higher than the prior data cuts, I'm wondering, do you expect more of an effect? And if the effect is the same because it's a shorter time frame. Is that still a win?

Well, it depends on the end point, but we'd expect the same or higher effect. The early part of the dosing before I remember was [indiscernible] and around the [indiscernible]. So their dosing is higher than that. So that should help improve the effect. But it may be more evidence some endpoints than others. Day 170 is relatively early. And we'll compare it to day 170 at the other doses, so you can look at it. But I think the other thing is it's a lot more patients right? There's a lot more and I think having a lot of patients consistently showing a pattern, I think, is an important part of what we help to show.

Yes. Okay. And then given prior studies in AS have suffered from some placebo effect, I think what we saw at the Analyst Day was pretty striking. But I'm just wondering if you're implementing any strategies in the clinic in order to help mitigate any potential placebo effect just because we hear that these patients are constantly gaining skills. And there's obviously some significant hope in the community. So I'm wondering if you can enroll for any of that.

Well, I think one of the things to keep in mind is that all the Angelman syndrome patients aren't really the same. In the Orbit study and some of the other programs, they are treating the severe deletion patients, the missense, UPDs, all the same. But the missense patients tend to be mild or can have more progression on their own. And so the [indiscernible] to those types is less severe. We're focused only on severe deletion type. Those types do not gain ground in natural history, there's no placebo effect. They don't really -- can't really gain ground. And I think -- so I think that's 1 thing. The other thing is very important in placebo effect issues is the time of the study. Obviously, it was 12 weeks. It's like near peak when it's placebo. We're showing you data over a longer period of time. It's very hard to maintain a placebo effect over many months or our expectation in Phase III will be more like 8 months to 1 year. The other thing, in addition to that is how the assessments are being done, certain assessments like the Bayley are done by a third-party psychologists that has to visually measure things. It is actually very hard to move the needle on that test. It's a lot easier to get parents to report that things are better. And there's also more prone to bias. But the Bayley scores, which we've anchored on our pathologist administered testing, it's very hard to move and when they move and they move as much as we've moved. This is something that's always just in and of itself more fundamentally real. So we feel good about what we're measuring, what we're seeing, the magnitude of effect, the sustained improvement over time that gives us confidence that we're not seeing a placebo effect that is certainly much different from what you would see. And I would say when you look at what developmental milestones kids is that there's no placebo that will cause the kid to swim across the pool and not drown, just not a placebo effect type of thing, right, that you can see. So I would say we have enough confidence in what we're looking at is something that's fundamentally different than what you normally see. But we're in neurology, neurology is prone to issues. So we're all cognizant of that. But I think the magic that we're seeing, I think it's so large. I just think there's no way this is placebo and it's certainly greater than natural history. What we hope to show you with the new data is more than 20 patients for the data will allow you to see a bigger chunk and see the consistency and the variation of that will help, I think, give us confidence about what we got going forward.

Yes, that makes sense. And you also showed some nice objective data in terms of EEG, suppression of delta wave activity, which is suggesting that there's something right, a real benefit there. How much are these patients approaching what's neurotypical? Can you put that EEG change into proper perspective for us?

Well, the delta power certainly has declined, but it's certainly not neurotypical yet at that point. There's still more movement to go but it's a significant movement of our baseline.

And then how much is age a factor in terms of who you'd like to enroll in the Phase III in order to give yourself the best likelihood of seeing a benefit? Have you noticed any patterns when looking at the younger patients who have been treated with GTX-102 so far versus the older ones?

There were -- when we started a lot of opinions that older kids wouldn't respond at all and only young kids respond, and we were trying to press to 1- or 2-year olds. Here's the thing that's very unusual that actually patients of all age seem to be responding. They may not all respond exactly the same, but all of them are responding. And I can't tell you there's a really precise ways that the 48-year-old are better than the 9- to 17-year-olds, even though people thought they would. But the truth is people over-concluded from mouse models of what would happen to the brain that after a certain age the mouse model do not respond as well. But mice and humans brain-wise are not the same. Human brains are much more plastic, much more dynamic, right? And I think what we're seeing is that even kids that are teenagers can improve in a lot of ways, receptive, cognitive. And there's more variation between individuals and there is among the age at least at this point. Because we look hard at that, okay. And I would say we're going to continue the Phase III the same 4- to 17-year old. I don't want to go less than 4 because we're doing lumber punctures, placebo. I should think that -- I always imagine the parent, I am going to take someone's kid 2-year-old and put him in a placebo-controlled trial. I think this is not right or okay, we're doing it. But -- so I think 4 to 17 is a good range. And our expectation would be to do a study in young kids, it's not placebo-controlled and the basket study that we look at some other types of genotypes. If we do one high-quality placebo-controlled study and show the efficacy is real, then we can just validate that, that's happening in other groups without having to run placebo-controlled trials over and over again for something like this. And that's our strategy moving forward. So far, I think FDA has been in agreement with that. We'll test that out further.

Right. Makes sense. Okay. Great. And then speaking of the FDA, have you discussed the end points yet with them? I think you had previously guided for a meeting in the first quarter. So I'm wondering if that's happened yet. And if you can give us your thinking on your latest thinking on endpoints that you're steering towards because Bayley has a lot of natural history data, but then you've been doing a lot of innovative work on developing other things like the MDRI...

So we did have our meeting in Q1. First of all, the FDA invited us to come even with incomplete Phase II data, which is unusual. Usually, they make you wait until they have all the Phase II data. So I consider this a step forward and they're looking at how to be collaborative, help us. We had a meeting, we discussed all the endpoints. I can say is they're comfortable with the 5 domains that we're focusing on. We will get the other 2 domains, which is fine motor and express communication as well. They're fine with the 5 we're focusing on. And I think we talked through the MDRI strategies and other endpoints. I can say that the end points we're measuring are endpoints they could agree with. And they did give us a flexibility to look at those endpoints and questions and be comfortable removing questions if we thought they weren't useful or effective. And it's very unusual to see that much flexibility on how to take the data we have. So the end points we're measuring are ones we can use, right? Now with regard to using MDRI, MDRI is a new method. It's not really an endpoint. It's like an [indiscernible] technique. So we're using the same endpoints you're just combining all 5 into 1 analysis. Whether we do it primary or secondary is not decided, but in either place, we will give us the benefit of showing the clinical benefit kind of assessment, right, across multiple domains in patients. So we'll continue to talk with them. It could be primary, it could be secondary. And any of the 5 endpoints that would be in MDRI could be a primary. And so we'll have that discussion when we have more of the data in front of us with them. But we had a good starting place. It was collaborative, I think flexible, and I think we're on the right track, and I feel comfortable we're going to find an appropriate endpoint for the program.

Yes. Okay. And you are very much trailblazing, but there's a couple or 1 closely related agent still in development alongside you from Biogen, Ionis, the Roche program was, I guess, discontinued or terminated, suspended. So -- and you all use different chemistry. And I'm wondering how you see the competitive landscape shaping up because I think I've heard each company say that theirs is the best and the most potent. So what is your view on that?

Well, I'd be different, we're the best, but the truth is that just talking about safety of chemistry is not really smart. You have to talk about potency, too. So the MOE chemistries that are being used have less toxicities than LNA, locked nucleic acid chemistries, but they also have less potency. And so the net of that has to determine what works. So the -- when we compare MOE to LNA in our own seen sequence algorithms, the MOEs are 1/10 to 1/100 as potent as LNA. Is that big a difference. So with that big a difference in the chemistry, safety starts to disappear as an issue because if I have to use 100x as much, then what's the benefit, are we going to gain new chemical toxicities from the pack. So if you had equal potency but better safety than I understand, but that's not what we're seeing, right? So there's a trade-off on potency. Right now, we found a spot where we are potent in the 5 to 14 range with an LNA and that we can do that with good safety profile we're comfortable with. So I think we're found in a very good place. And I think that's a very potent ASL like SPINRAZA is a very potent ASL, right, at the 12 mg dose, right? That range -- I think that's when ASL work when they're highly potent. A lot of the other programs, even if they're using MOE or -- and actually Roche is using LNA, they're dosing was much higher and I think they had more challenge of doing that. So I think the potency is going to be the driver of the choice here, and we have safety of this LNA is sufficient to be able to do the dosing that we need to do. So we have the right therapeutic window, which I think is what's important. But we always pay attention both to the chemistry, the backbone chemistry, but there's also sequence issue, safety issues. We have a safety question that the lower extremity weakness problem that we have seen. We seem to have managed it now. It looked like it was more like a local inflammation, irritation kind of thing. But with the Trendelenburg and Flush strategy for administration, we seem to have reduce that dramatically. And we feel pretty good about where we're at in terms of being able to administer our drug safely. I'm happy to see how the data would be great to compare drugs if people would present the actual data, but I think we haven't seen it yet, but we're putting our data out. So anyways, I'm waiting and seeing.

Yes. Great. Okay. Well, let's switch gears. We've only talked about one of your many programs. But setrusumab and OI brittle bone diseases making waves too. So have you been able to complete enrollment in Orbit yet, I think that was...

We were very close. Our Orbit and Cosmic are both very close. They are enrolling well. We're in good shape. The field is getting excited. When we presented data last year on the program, there was a lot of people that are like shocked because there's a fundamental misconception about OI that the mutated collagen makes the bone fracture, but that's probably not the biggest cause of fractures. The biggest cause of fracture that the mutated collagen results in a change in bone regulation. The kids don't make enough bone. Bone mineral density study in the Phase II study, the average with minus 2Z score, which is 2 standard deviation below the mean like 5th percentile, that's the average. Some kids were 4 standard deviations below. So they're not making any bone. So no wonder the bone is fractured, they're not making any bone because the mutation is leading to an altered regulation of the bone. What we're really doing with setrusumab is we're basically counteracting that turning on bone production again. And these kids make a lot of bone when you turn it on, which tells you that their bones are sitting and are waiting to do what we need them to do. The 5- to 12-year-olds will show bone mineral density improvement of 19% in 6 months, 19% in 6 months in the little kids. That's just an incredible amount of bone production. What we showed also was a 67% reduction in fracture frequency in just 6 months, minimum 6 months of treatment. So that blew people away and people were quite surprised because the doctors are now realizing that what I learned and what they learned about why these kids have weak bones is not really quite right. And the beauty of that is that there already is a solution that helps strengthen their bones, it does it in the right mechanistic way. And what setrusumab does, it turns on osteoblasts to become osteocytes, and osteocytes to make bone. But where they to get turned on and where they land, depends on where the bone is weak. When the bone is wiggling and moving, the surface of the cells become attractive to bone cells to land. And so your body has this built-in micromechanism to attract the bones towards the land where the bone is weak. So when we start making new bone mineral density, it's landing where the bone is weakest, which will strengthen the bone very quickly. In the study, within a few months, the fractures went down, right? We had -- after 5 months, I think there's only 2 fractures that were after 5 months, right? All the other fractures were before that so as soon as a little bit of bone is made the bones are strengthening and the fracture rate went down, 20 out of 24 patients had no fractures actually. So we think the mechanism is really the right mechanism, the strength of these kids. So we're pretty excited. The study is enrolled -- with the power of the study and the size of the effect, we reduced the sizes of the Orbit study from 195 to 150, we didn't need to do that, which would end the enrollment quicker. And because of the power of the effect we also added a couple of interim assessments would allow us to end quicker if the study is off scale. And it's 0.01 that we'd end at the first interim. And it's possible to do that. It's a very stringent requirement. But in order to get comfort with the FDA accepting a shortened study, we'd have to be able to show them that this was kind of an offscale effect that we could continue for ethical reasons. So we're excited about why I think this is going to be a transformative treatment for that disease. Studies are nearly enrolled, and we'd expect to get some more Phase II data later in the year, longer-term Phase II data, interim late in the year and next year. So it's coming along. It will be really our next big program, we'll have maybe a couple of gene therapies we think we could file. But within a year or so, we're going to talk about 3 BLAs.

Right. Yes. It's all happening. So can you talk about how you designed the interim analysis? I know there's a really high threshold for hitting the first and the second, even although it's much right easier than the first. So how did you choose these thresholds relative to what you have seen and think is possible in the Orbit Phase III.

Well, the first endpoint is going to happen when we hit 60% of the fractures approximately, which is towards the end of this year is expectation. At that point in time, there would be a number of patients who had maybe 8 months of treatment, right? So relatively less than the year. So for that reason, in order to get FDA comfortable with less than a year of actual -- the placebo control we felt that we had to have a stringent requirement, which would then spend no alpha. So we had a 0.01p value requirement. It's the primary endpoint, the clinical fracture endpoint will be evaluated. The DMC will see it, we will not see it if it's positive, we'll get informed. It should be targeted around the end of the year. And if it's positive, then we are preparing in case it is positive to be able to pull together a filing. You want to end the study cleanly, though you don't want to stop [ precipitously ] you'd have to finish all the patients and finish and cleanly. But we're trying to do these interims because we don't want to keep kids on placebo longer than they have to and we need to -- and get into a filing would be highly valuable for us. We'll do a second interim and a final analysis. The second interim may have more alpha spend, little less stringent because at that point, we'd expect everyone to have at least a year of treatment. And then the end is at what would be approximate 1.5 years of treatment, something of that time frame when we hit to 100% fractures so that's a basic design.

Okay. Very helpful and I am a stranger to rare bone diseases with all the experience for Crysvita. How does...

[indiscernible] Crysvita...

True. MPS IV, Yes.

[indiscernible], achondroplasia.

Exactly. There's a good track record here. Well, I know Mariah, who we spoke with earlier today, likes to use XLH and Crysvita as an analog for Setrusumab in OI. How do you feel about that, that as a proxy for the need and the market opportunity.

I think it's a reasonable proxy. I think OI is more severe disease than XLH even. So I think the unmet need and demand will be higher. The population for OI is a little bit larger. Some clinic or center say it's much larger, like 50% larger, but our estimate is around 60,000 [indiscernible], where it's 50,000 for XLH. So there are some parallels that make them comparable. I would think OI could be a little bigger and probably go a little faster.

That's helpful. And then switching gears. At the Analyst Day, we got to see some early data from your gene therapy in Wilson disease. Can you remind us what we saw and what you're excited to look for next.

Yes. So we released just the first cohort data from the program, which there were 3 patients that had enough time and those 3 patients, 2 to 3 had gotten off their chelators and had lowered urine, copper and off their chelators, it looked like they had a good effect. And we're showing signs of increased copper distribution occurring. That means the copper must have been loaded on ceruloplasmin, that was encouraging to see the copper distribution going up. And so it was a good sign that the therapy could work. It is early, though. It's the first cohort. We've now treated the 2 more cohorts. Third cohort was completed in the first quarter. We'll now need to wait for those kids to get through their first 12 weeks, at which point we'll look at urinary copper. And then if they've lowered their urinary copper level, we'll start titrating down their chelators, but the goal would be look at can we restore copper detoxification through the bio rather than dumping in the urine with the chelators and can we restore copper distribution and do that safely with that data from the 3 cohorts, and we'll have an ability to make a decision on Phase III dose.

And how long do you think you need to treat patients in order to see efficacy in terms of the neurological symptoms or the liver issues?

Well, the neurologic symptoms are complicated because some of them may be reversible, some not. But we think that within certainly 6-month period, people were having them should be some effect, but I think you could look at transplant cases to go to get a feel because people have been transplanted on Wilson when they have neurologic problems that don't resolve. And I think within a matter of 6 months or so, I'm sure it'll be a year will be better than 6 months, but it's that kind of time frame. The thing is that I would say is that not everyone in the study is going to have neurological problems. It's going to be a small subset because most people have been on the multiple meds. So try to control their neurologic side effects. But we think it's something we would look for, but I think it's going to be more dependent on -- study where you have a lot more patients to be able to make a conclusion. I think right now, the focus on can we lower urinary copper? And can we eliminate the need for chelators? Finally, can we improve copper distribution? It's the copper distribution improvement that I hope would remove the what is, I think, copper deficiency neurologic issues. In the first 3 patients of that cohort, they felt good and the ones that got off drug felt very good. So is that feeling good, meaning something, I don't know. It's open label. I guess everyone feels good open label but they felt distinctly better. I'm not sure what's happening neurologically, but we'll -- there's some indication maybe there's some effect. But I do think one of the things with Wilson is that we all think it's too much copper or free copper but actually, there's a bigger change in the fact that they don't deliver copper to their body normally. But we've never been able to fix that part. So we've always discounted it. It's kind of what we do. We tend the diseases where we can't fix it rather than look at it completely. And I think when you get deeper into it, I think restoring copper distribution is going to end up becoming an important part of the total picture. And it's why transplant helps Wilson patients when chelators can't. That's because of the need for copper distribution.

Interesting. Well, I think we're out of time, but I certainly appreciate the update, and we look forward to seeing the next data releases as they emerge. Thank you so much, Emil.

Thank you.