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

All right. We're going to go ahead and get started here. So next up, we have Chief Medical Officer from Ultragenyx, Eric Crombez. Eric, I'm just going to turn it over to you for a brief introduction of where the company stands today, and then we'll get into it.

Yes. So Ultragenyx founded very dedicated to developing and commercializing drugs for patients with rare disease that has been core to our mission from the beginning, and we are very much [indiscernible] to that core there. Certainly, a lot of success in the commercial setting led by Crysvita and very much supported by Mepsevii, Evkeeza and Dojolvi. But I think what a lot of people are looking forward to is really with the progression of our development pipeline and a lot of Phase III data readouts and filing in the near term. I think as far as value drivers in the near term led by osteogenesis perfecta.

Perfect. All right. So I think we have to start off on the setrusumab side. Maybe before we dive into that, what was the rationale to do the royalty financing at this point in time?

Yes. So I think really what drove us to, I guess, start thinking about doing something was with the complete response letter for our Sanfilippo program that pushed off the potential revenue from the sale of a PRV. So with pushing that off into the d, it made sense for us to think about topping off the balance sheet. We started a competitive process. It truly was competitive and comprehensive. Obviously, we know Omar, they know us. And I think we think, and I think I've heard consistently that it was a good deal. It's nondiluted, was not tied to the stock price on any given day. So really just the right deal at the right time and just a good opportunity to top off the balance sheet.

Makes sense. All right. So going into Orbit and Cosmic, are you still planning to show both of those readouts together at the same time?

Yes. And we've thought about doing them separately, staggering them by a little bit of time. But we think we've heard and very much understand that it makes sense to talk about both studies at the same time.

Yes. All right. So one piece that I've always found interesting is you've noted you think Orbit probably has a higher probability of success, I think, largely due to the sample size there. I think Emil thinks maybe Cosmic may have a higher POS just given the event rate. How do we square both of those?

Yes. So I mean, I think -- I guess what's underneath that message is between the 2 of us, we are very supportive and very confident in the success of both programs. And the truth is both studies were powered for success. And importantly, Cosmic is powered to show superiority. Just from a pure numerical stats-driven probability of success, it is very marginally higher if you look at just straightforward statistics. But if you believe with the younger, much younger patient population enrolling in Cosmic and the really great results we saw there, there is potential for a very power effect there. And I think that's what Emil is being anchored on.

Yes. That makes sense. I'm going to summarize a question I've got from investors in a lot of different ways, and we can ask different scenarios a lot, but I'll just phrase it one way, which is, let's say, Orbit is statistically significant. The effect size is on the lower side with the low to mid-30% and Cosmic misses on statistics, but numerically is still beating our bisphosphonates. Maybe it's in the 10% to 15% effect size ballpark. Is that a commercially viable profile?

So I think the easy top line answer is yes, that's commercially viable. I mean we've seen the use of bisphosphonates. And it's important to remember that when you're making this type of diagnosis with a disease that's severe being able to offer nothing is just a terrible position to be in. And I think that's why bisphosphonates has been used to the extent they have. That type of profile does show you are consistently beating bisphosphonates to be fair and honest, that is probably a floor of where my expectations are. I mean we powered Orbit to detect at least a 50% treatment effect with that 159 patients. And with the 67% response rate we saw consistently in Phase II, that gives you room for something unexpected to have happened there. We've heard treaters really consistently say a treatment effect size down to 40% is clinically meaningful to them. I know some will get down to 30%, but I think it's fair to say that's kind of a floor for any of our expectations.

Just from a statistical perspective, have you said what the minimum effect size is you think that would be detectable in the trial? Like where are you 50% powered roughly?

Yes. So with the 159 patients enrolled, that gives us greater than 80% power to detect that 50% treatment effect.

Okay. What's your current thinking on how long the bisphosphonate effects last?

Yes. So very confident that it is months plural. I think in my mind, and I think most people would agree that once you get to that 12-month mark, you're probably seeing meaningful washout of bisphosphonates. I think if you're thinking about a total complete washout of bisphosphonate, you're maybe talking about multiple years. But I think with all patients going to at least 18 months with final data readout, I think it is fair you are seeing real washout of bisphosphonates at that point.

Okay. I think in the Orbit study, I'm not sure if this is true for Cosmic also, but there is the ability to go on to receive rescue treatment. What are the criteria to do that? And how many people have gone on to do that?

Yes. So I mean, I think there definitely is -- originally, there is the ethical argument. There's also a very practical argument that if you have patients who either you know are assuming were randomized to placebo and having a lot of fractures. We're talking patients with 10, 15 fractures in the first year of enrollment in the study. Is it fair to continue following them in a way and letting them continue to fracture at that rate? So yes, there's an ethical debate there, concern there. There's also the practical debate that you're going to start to see patients drop out if they're having that type of fracture rate. So the protocol has guidelines for when a PI can approach the medical monitor for a discussion on rescue. The medical monitor needs to approve every single rescue. So there was nothing automatic. There was nothing that just triggered them being transferred to this. So we did keep tight control of it. We haven't said exactly how many patients have been rescued, but I can say when designing the study and when powering the study, we account for dropouts, we account for patients being rescued in what we have seen is absolutely within our expectations for our original planning.

Okay. That makes sense. Makes sense. And that's true both for discontinuation, dropouts and the rescues, both of those pieces are within expectations?

Yes. And I think fair to say dropouts has been a nonissue for this trial.

Yes. All right. That's helpful. Just a few on the commercial side then. What's the latest thoughts on pricing?

So pricing for us, and I think the usual answer, it's too early, wait and see what will happen here. But I think we're in a very ideal situation where we have the experience from Crysvita. We have that play book. It was a very successful launch. So when we think about pricing for osteogenesis imperfecta, I think it is very fair to say that Crysvita is a very good model and where we're currently thinking about pricing.

What about duration of treatment, too? Like I mean, there is a paradigm that exists today where bisphosphonates are not treat forever type of paradigm. So how do you think that will evolve?

Yes. So I mean, to me, there is 0 debate here in my mind. I mean, understanding the biology, understanding how setrusumab works, this is a lifelong treatment. This antibody takes out sclerostin and that allows the osteoblast to continue to generate, make more osteoblasts, lay down more collagen, lay down stronger bone and stopping fractures. If you take that away, sclerostin will shift that balance back to it was prior to treatment and patients will lose ground. We saw that in ASTEROID, the original adult study done, and that study was originally designed for 12 months only. Patients did improve and then after stopping treatment, did lose ground again.

Yes. Okay. What about the type 1 versus type 3, 4 patients? Have you said the mix in the trial again?

Yes. So for Orbit Phase III, closer to 50% type 3s and 4 combined versus type 1. That's a little bit in contrast to the Phase II part of Orbit, which was 1/3 of patients, type 3s and 4s.

And what's the split of patients commercially? Like not just the prevalent population, but the actual patients that are diagnosed in the system managed today?

Types 1 versus [indiscernible]?

Yes.

I'm a phone a friend.

[Technical Difficulty].

That makes sense. Perfect. All right. I guess last question just on setrusumab, big picture since we're kind of right before the readout. I don't know if you're at any other conferences this week, but this might be the last one. Like any kind of closing thoughts on level of confidence, et cetera, you want to make as we're heading into that data?

Yes. No, and I think it is important, having not achieved success with IA2 that our confidence remains unchanged and higher. Probably success for full data readout remains the same, and this is how the study was designed and powered and originally designed to work there. I think, again, I fall back on the biology, the mechanism of action for this drug and the really consistent results we saw in Phase I/II across all patient types. We saw the increase in biomarkers supporting the increase in bone mineral density, supporting the reduction in fracture rate. And all patients are responding. They're all doing well and supported by a very, very clean safety profile.

Awesome. All right. Look forward to seeing that data soon. Let's switch gears. Let's go over to Angelman side. Are you going to share anything over the next, I guess, 9 months before we get data? Like can you make any directional comments on like blinded data variability baselines, anything like that?

Yes. We've started those conversations. I think the truth is for us internally, we understand how this drug works based on the 74 patients that were enrolled in the Phase I/II. For rare disease for us, that is a big Phase II study. So we feel like we do not have unanswered questions there. The truth is with Aspire fully enrolled earlier this year, now initiating enrollment in Aurora, the second Phase III study, it's 2 big global Phase III study that still does have relatively complicated dosing. We want to make sure the patients are hitting their visits. We're doing cleaning, everything is in the database and everything is as high quality as possible. So we certainly can. We've had that discussion. The question is, would that be beneficial to release that type of data? Or is it better to keep the team really focused on executing? Because once we get to midyear in the second half of the year with data release with positive results, we are then head down heading towards filing and that is a priority for us.

Yes, that makes sense. Just on the safety side, on the lower extremity weakness point. I mean, I'm sure this is something that you guys in the DSMB are monitoring very closely. Like what's the level of -- or I guess, the threshold for disclosure for anything that would be ongoing there?

Yes. I think generally, the disclosure, I think if we're just really talking safety for disclosure would be anything that's new or change, a meaningful change to what you've seen before. And I think with the time we took to enroll the additional 53 patients in what we call the dose expansion, and that included the trend Trendelenburg that included the flush with the understanding that ASOs from a chemical perspective can be irritating at the site of injection. So that's just flushing the drug away from site of injection. Trendelenburg is using gravity to help you do that. We saw in those 53 patients only a couple of episodes of new lower extremity weakness. They were both very mild. One was only picked up in retrospect, both resolved very quickly without sequelae. So we felt that we understand the safety event. We understand and we're able to mitigate it to a great degree. So I think it's something that is part of the benefit risk profile. We're looking at that in totality, but very much something understood and with our end of Phase II meeting, not even a topic of conversation with the FDA.

Awesome. What about the dropout discontinuation rate for the study? I guess like when you were setting it up initially for powering, what did you expect?

Yes. I mean, I think as a general rule, we always -- typically, and this is probably true for most people, you're going to at least start without accounting for at least like a 5% dropout rate. With these types of studies, there are very, very low dropout rates. We saw it in the Phase II because this is a chance for these kids to develop, learn and gain new skills. The parents understand this Phase II data, and it really is the first chance they've had to see their children grow and develop and gain new skills. So it's essentially no dropout rate, and I expect it to stay very low over the course of these studies.

All right. Great. Are there any prespecified subgroup analyses in Aspire?

So prespecified, I mean, as far as the primary endpoint as part of top line, we're really looking at the patients in total and really just as a reminder because for Aspire, our main Phase III study, we're sticking with patients with full deletions as we evaluated in the Phase I/II. That is the great majority of patients, probably between 70% and 80% of patients with Angelman syndrome. More importantly, with full deletions, that means these patients are not making any UBE3A, which means they are severe, but also very homogeneous in their phenotype, making signal detection easier there. So if we were doing an all-in approach, all mutations into a single study, you would want to do the sub-analysis. We're taking care of those patient populations in Aurora. So from a primary analysis, it is that total patient population. Certainly, we'll look at pediatrics -- younger pediatrics from older, but it is really looking at them as a single group.

What were your expectations when you were setting up the control arm like powering versus the control arm?

Yes. So I mean, at its core, and I think natural history supports this, we all understand this treating community understands, particularly when you're talking about patients with full deletion, they have a very flat developmental curve. When we talk about neurotypical children, you're always looking for them to continue on their developmental curve, grow, develop, gain new skills. Their developmental curve is very, very flat. Certainly, we allow for some change there, some noise within that plus or minus as we're designing the study, but we expect that to remain fairly flat.

How often do you measure Bayley in the study?

So the Bayley is something you're putting them through. You always have to worry about learning skills and practice skills like this. So it's done with the 48-week primary efficacy analysis period. It's done 3 times.

Okay. That makes sense. I guess what's the evidence that you can kind of improve cognition communication, reverse the disease as opposed to prevent any type of progression?

Yes. No, and that's what's really unique and really such a unique opportunity here and why there's so much hope and potential with this drug because often with neurology, that's exactly what you're doing. You're hoping to stop deterioration or even just slow it down. And with Angelman syndrome, UBE3A affects the way synapses communicate with each other. So your neurons remain intact. You're not losing these neurons over time. So with the knocking down of imprinting from the paternal allele, you're expressing UBE3A. The synapses start communicating. These neurons are intact. So that gives you the potential to really have true development there.

Great. How many patients are coming on to the study with seizures? Like how -- what's the average frequency of seizures?

Yes. So I would say seizures is an important part of this disease. I can't tell you 100% of patients have seizures and what that frequency is. It hasn't been a big focus for us because we're not using a seizure approach. We're really looking to directly affect this disease and affect development there. So we will look at that. Most patients do have seizures. I think it could be a potential biomarker there to help with treatment decisions. But again, we're obviously very focused on the attainment of new developmental skills.

I guess, all the secondary endpoints also, how does the hierarchical testing work out there?

Yes. So hierarchical testing is just kind of a nice way to really try to control for false positives. So you're basically just ranking your endpoints in order. You're trying to do that kind of with a lens of importance, but with the development, you could argue that all of them are important there. So you're just putting them in a rank order, testing them one at a time. And as long as you're still hitting statistical significance as you move through your list, you keep moving through. As soon as you have an endpoint that does not achieve statistical significance, you stop testing further endpoints.

What's the order of the endpoints, do you know?

Yes. So for us, obviously, we're looking at cognition as a primary endpoint. That's important. And then we're looking then at communication, very focused on receptive communication, gross -- gross motor behavior and then sleep.

Okay. Really helpful. All right. Maybe in the last 30 seconds, we can just kind of wrap up and not really going to do a justice today, but I'm just going to flag 401, 111, 701.

Yes. So in addition to important programs, important value drivers for the organization with OI and quickly followed by Angelman, as a group, the gene therapy programs remain very important to us. And honestly, with the refiling, the response to the CRL with Sanfilippo and a 6-month review clock for that versus the standard review for GSDIa, which we said will finish submission this month, that puts them very close up to new PDUFA dates in the middle part of this year, and we're looking forward to and expect them to be very successful launches.

Awesome. We're out of time. Thanks so much for joining, Eric.

Thanks.