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

All right. Well, good morning, everybody, and thank you for joining us at the 44th Annual TD Cowen Healthcare Conference here in Boston. I'm Yaron Werber, on the biotech team, and it's a great pleasure to moderate this session with Ultragenyx, along with my colleague, Brendan Smith. With us today, we have Eric Crombez, who is the Chief Medical Officer and EVP at Ultragenyx. Eric, good to see you.

Good to see you.

Thanks for joining us.

So lots going on from Ultragenyx this year. This is really going to be a year of pipeline, both Phase III pipeline and Phase sort of II pipeline advancing forward, and it's going to carry us into the end of the year into the OI data and then into next year into the OTC data. So one of the programs that has gone a lot of attention is Angelman syndrome, a program that's close to my heart. It's an important indication. I guess the first question, as you think about the data that we're going to see in Q2, I am thinking it's Q2, can you give us a sense how many patients worth of data are we going to see? I don't know how much you can say about what dose levels are in maintenance? And how many of them are from the U.S. study that's been expanded versus European study?

It's one large study that has been expanded several times. So just as a refresher, we started with our dose escalation cohorts, and that was really to find the dose and establish initial safety for Phase III. But after working through those dose escalation cohorts, we decided that it was a good idea to expand into what we're referring to as the dose expansion cohorts, and those are the letter cohorts. So overall, we've enrolled 74 patients in the study in its entirety -- dose escalation cohorts, along with expansion cohorts. And 53 of that 74 are dose expansion cohorts. Originally, in the dose escalation cohorts, we started with loading doses between 3.3 and 5 and then went up from there with the expansion cohorts because again, we are looking to confirm the dose we want to take in [ 5 ] into Phase III. We started a little bit higher than that. But in all dosing cohort, the maintenance dosing has been between 10 and 14.

And what determines whether it's 10 or 14? Is it just patient tolerability?

Yes. So I think we wanted to start a little bit slower, start to titrate up. I mean, I think when you think molecularly about this disease and what we're trying to do, you really do want to knock down the imprinting of the parental allele in its entirety. So I do think driving towards 14 is the best way to get as much knockdown as possible, and that really sets these children out on their developmental path.

And in the maintenance, it's 14 mg. Is it Q4 months at that point?

Every 3 months.

Every 3 months, and that's going to be consistent the way it's going to be?

Yes. So that's where we are, and that's what we're looking forward with the data readout. We had our last data update in October at Analyst Day, we like what we saw there. But we do also want to be open to quite a bit of data that's coming out of these expansion cohorts to decide on the final dose and dose regimen that we'll take forward into the Phase III.

And so in Q2, we'll see the 53 from the expansion? Or do we also see the 21 from the escalation?

Yes. So we will have long-term follow-up data, and that's important. But a really big focus on those dose expansion patients will have around 20 patients at day 170, around 10 who will reach day 254 and then shorter duration for the rest of them, but will be very informative both on the efficacy and safety side.

Okay. And one of the questions initially, and I think it was answered somewhat in the data since then, it was very provocative in the first 5 patients; is how fast would the onset of action be, right? Like clinically, we thought that it might take even 60 days, maybe the 90 days. And of course, we saw early on, it was pretty quick. When you're thinking about impact on different endpoints, which endpoints are kind of more sensitive early on to change versus ones that need lower therapy?

Yes. And I think, again, to me, going back to what's happening molecularly, helps inform that because we do knockdown in printing very quickly. So that paternal allele is expressing a protein, these synapses start communicating in a much better way. It's been interesting to see what's moving first. It seems like sleep is moving first. And sleep is very interesting for -- because for those patients where they have a lot of sleep difficulty, which is very difficult for them, but also very difficult for the family; they seem to see improvement very, very quickly. Some are claiming differences in sleep with the first dose. But then I think when you're looking at some of the other tests that are really looking at development and developmental milestones, it makes sense, it may take a little while to see a signal there.

And if I remember correctly, as you think about the data, it almost seemed that behavior and cognition had a more outsized effect, let's say, than motor, and there wasn't as much of an effect on seizure. Can you maybe just talk about the data?

Yes. So we are still really looking at kind of those 5 buckets that we've been talking about. We are very interested in cognition language. And I think looking at language, both receptive and expressive, is important. I think we get very focused on expressive language when patients are saying words, children are starting to develop words. But receptive language is very important because for these families, that means that the child is going towards something or in danger, they can call out and give direction, and the child will respond to that. So receptive language is important. Then, I guess, we're looking at behavior, sleep and gross motor.

Okay. And as you think about different endpoints, gross motor was always one that's a little harder to show an effect, either as a walk or is a fine or gross motor. Seizures are important. There is a seizure diary. Can you talk about the data so far?

Yes. So seizures are important. They are an important part of this disease. Seizure as an endpoint is very, very challenging. It's always hard, no matter what the indication you're looking at. But we are looking at seizure. We're looking at seizures through diary and EEG. We will continue to do that in the Phase III trial. But looking a little bit beyond seizures, the EEG data is very interesting and very important. And it looks like it might be a very important biomarker, which will help guide dosing and really help on the efficacy side of things.

There's a delta wave of change?

Yes.

Because these patients are coming in stable, with stable background.

From a seizure perspective, yes.

So it's a little hard to show an effect when they're stable.

Seizures are always hard.

Yes. And as you think of motor between fine and gross motor and the Bayley's or a walk test, can you talk about what you are measuring?

Yes. No, it's important. And we are using the Bayley. And again, those are similar to what you do with -- as children go to the pediatrician, you're monitoring normal development. So that's where we're looking to see with these children, is they're really gaining very limited skills over the years they've been alive. And with the [ uninfringing ] of that paternal allele, we're looking at them to sort on their own developmental curve. And the Bayley is great. That's what we'll do. And that's why I think it was important to add on the videos at Analyst Day. We were only able to show those videos in the room, but we were able to really show before and after with how are these children feeding. They're just completely ataxic and disorganized with how they're trying to use utensils, feed themselves, food is everywhere and then seeing over time them then really being able to use utensils, feed themselves and take care of themselves; and then gross motor, like looking at these children walking on hard surfaces. There was one child who apparently was traveling a lot, there's lots of videos at the airport, but really going from a very wide-based ataxic gait to a very much more normal gait, being able to walk on sand. So being able to see that and what that really means not just from a Bayley score, but what it means in real life to these patients.

And when we see the data at that point, are you going to show the CGI as well, the CGI-I data?

Yes. So we're really looking -- so obviously, all these endpoints remain the same, similar to what we were doing at Analyst Day. I think it's fair to say, we're really looking at that data set we released at Analyst Day as a template, but we'll have the totality of the data coming through.

And as you're thinking about -- it sounds like you've already commenced a discussion with the authority about the Phase III trial design. What would be the primary? Is there a chance to do a Multi-domain Responder Index? Or is this going to -- it's not validated yet?

Yes. So I think that was why we wanted to start the conversation with the FDA and EMA early. We wanted to understand endpoint-by-endpoint and then collectively with something like the MDRI, what is the book of work they want to see to bring that forward as a primary or even a key secondary. So I think that was helpful. To me, the MDRI for this type of indication really could be an ideal endpoint because it is in our logic indication, there is variability. And when you're looking at an MDRI, which is just really laying out these important domains, defining upfront what is clinically meaningful improvement and then just doing a simple scoring, that, to me, works for that type of indication. You're right, there is not as much support as the FDA may like to see. But I think importantly, whether it's the primary or a key secondary, we will be bringing forward the MDRI.

Yes. And if it's a key secondary, the primary, I imagine, would be a CGI?

Yes. We're trying to be open. I mean, certainly, the data that came through last fall was informative, it was directional. We do want to be open to quite a large amount of data that we're working through at the moment in helping us that drive final selection for primary endpoint and then certainly engaging with the FDA at end of Phase II.

Yes. So you're with CEDAR, right, because it's an ASO. How open are they for accelerated approval of endpoints?

So we're not going forward with accelerated approved range. We don't think it's the right place to do that. Obviously, that's in great contrast to Sanfilippo, where you have these type of metabolic pathways, where you have an enzymatic break, that is the accumulating substrate that is what's toxic. Really, that is the right place to be using a biomarker for accelerated approval. I think with these types of very complicated diseases and with this mechanism of action, I think we really want to bring forward a clinical endpoint in a truly blinded Phase III study to really ensure we understand both safety and efficacy here.

And one of the -- in this study and in all studies right now in Angelman for the benefit of the patients, there's no control, which is appropriate; I guess my question is, how do you ultimately power the Phase III?

Yes. So we will power the Phase III once we do agree with the FDA and EMA on that primary endpoint. We will power it based on the data from the Phase I/II study.

Based on what you think would be a desired outcome or based on what you think would be a delta versus placebo? Because it's sort of interlinked, and that's...

Versus placebo.

Yes. And in -- what's considered clinically meaningful for CGI? Is it at least 1 point? Or is it you really want to see 1.5 or and 2?

Yes. So it will be interesting to see if the FDA really pushes us there. They have not for other programs like GSD1a, like OI. So they haven't directed us in that direction about setting a clinically meaningful bar. But again, we do need to have that conversation on what we're going to bring forward for our proposal for the primary endpoint.

So any change versus placebo is the way it would be powered and [ hopefully ] clinically meaningful.

Yes, I think we've been saying, and I think it is fair to say, roughly between 100 and 120 patients is probably reasonable [ because ] it's efficacy, but also you want a total safety database. And both things will drive patient number.

Okay. We touched on -- we just started touching about GSD1a. What -- and we're expecting that Phase III from the glucogene study to read out, again, also first half. it sounds like it's going to be Q2. The primary percent change from baseline cornstarch intake, what's clinically meaningful?

Yes. So it is important that in discussion with the FDA, and we had a lot of conversation with the FDA on this, interestingly, not on what is really the primary endpoint because this is really -- it's a disorder of energy metabolism. These patients aren't able to maintain normal glucose levels. So I think it's very clear that what -- that's important for these patients. The conversation was really on how were we going to measure glucose through CGM, fingerstick, controlled fasting challenges. So again, the FDA did not directionally require us to set a firm threshold on what's clinically meaningful. But I think fair to say, and this goes back to your -- their powering question. If we see what we saw in the Phase I/II, this will be a very positive Phase III study.

And remind us of what we saw in the Phase I/II [indiscernible] baseline.

Currently, most recent data cut for those patients and those patients are -- some of those patients are entering their fifth year follow-up, 66% reduction in cornstarch from baseline.

Right. And you wouldn't expect placebo to have any meaningful change?

No. And that's it. I mean, these patients, you really spend time -- certainly, during the pediatric years, you do start to reach steady state going towards adulthood. But you find what works for them. You certainly follow them, and they may make small adjustments. But by and large, that becomes their fixed regimen.

From a commercial perspective, also to support pricing because cornstarch is obviously not expensive. It's obviously fairly [indiscernible]. How are you thinking about -- what are the key secondaries you need to hit as well?

Yes. So really, this really all does come back to reduction of cornstarch, but in the background of maintaining good glucose control. And that's why we think this layered approach with continuous glucose monitoring because that gives us around the clock, it's important, yes, to know what's going on, while these patients are exercising changes with feeding, but very importantly, what is happening overnight. So that gives us a tremendous amount of data that can be supported by fingerstick data, and then bringing these patients in occasionally for a controlled fasting challenge gives us really layered understanding of what's happening with the glucose control of these patients.

Okay. And from a manufacturing standpoint, once you have the data, I believe, there's going to be a little bit of a gap before you file for tech transfer reasons. Can you talk about that, what you're going to be doing?

Yes. So not too far out from the Boston Cambridge area in Bedford Hills, we built from the ground up our own gene therapy manufacturing facility, is completely dedicated to gene therapy. We can run both HEK and the Pinnacle PCL process out of there, and we are prepared to do that. We have a lot of capacity to expand. So we felt at this point, it was better to bring that manufacturing in-house, control cost of goods, reduce cost of goods and really control all of this manufacturing as we lead into approval on commercialization.

And are you using HEK now through your contract manufacturer, it's going to be the same process, just bringing that in sight? Or...

Same process. We've used HEK from the inning, we will bring that into our facility. We will launch with that. This program, along with OTC, was selected as our original inborn errors of metabolism gene therapy programs because it does have a relatively lower incidence and prevalence. We needed that Pinnacle PCL process to be able to move into things like Wilson.

Okay. Got it. And right. So Wilson's already internal?

Already internal and with Pinnacle PCL from the beginning.

Okay. And OTC and GSD are not with PCL [indiscernible], but you need to move to PCL, which allows you...

Yes, we don't need to move to PCL. So we will launch both products on HEK, and then we'll -- I guess it probably gets fair to say when is the right time to move to Pinnacle PCL, but we will start with the HEK process.

And so how long would it take to do the HEK transfer?

Not sure if I'm the right person to ask that question, to be honest with you. It probably depends on how we prioritize and push.

Okay. Understood. But it doesn't seem like you'll file this year because it will take a little time.

So we made the decision relatively recently to bring that in-house. So we are working through how quickly we can do that transfer. And I think with the data readout that will be coming, we'll be able to update on filing plans.

And PCL, just maybe describe that. And does that still use HEK?

No. So that's a HeLa process. So really, when we first started and I've been with our gene therapy programs for almost 10 years now, starting with Dimension. That was really the technology that was available to us, having really the [ farm ] manufacturing expertise as part of Dimension, bringing it into Ultragenyx, brought forward that Pinnacle PCL HeLa process. And again, that allowed us to evolve our portfolio and really look at diseases with much bigger incidence and prevalence.

Okay. Would you need to do a bioequivalence study?

So I think we always want to start with analytical comparability, I think, with gene therapy. And again, I started in the enzyme replacement therapy world, which is a different type of biologic and very used to analytical comparability. Those are much more challenging, I think, to show that type of comparability. So I think that's a very reasonable approach for a gene therapy tech transfer.

Okay. Great. Let me move into Brendan.

All right. So I do want to spend a little bit of time on OI, osteogenesis imperfecta. So maybe just quickly kind of remind us because you’re running 2 pivotal studies now, first, maybe why that makes sense? And if you're expecting either different response rates or kind of the quickness with which you could get to actual response in clinical studies?

Yes. So I think it's fair to say the core Phase III trial, the original plan was to use the data coming out of Orbit, our ongoing Phase II/III study. That was the data we presented at Analyst Day. And -- but while we were working through our partnership, bringing that study in-house, we were hearing a lot of conversation on the use of bisphosphonates. Do they work or don't they work? Where are they improved? Where are they not? So we thought it was a good idea to bring forward the second Phase III trial, COSMIC, so that's the head-to-head trial to bisphosphonate. So we really will be able to answer in a very definitive way what effect does bisphosphonate has? What is the difference between setrusumab and bisphosphonate? And we were originally hearing a lot of conversation, at least in the U.S., that for younger, more severe patients, they do want to start them on bisphosphonates early. So we lowered that age down to 2 years of age. So it does also broaden the population with that second study.

So what, generally speaking, is kind of the cutoff between when physicians feel less compelled to starting on bisphosphonate?

So it is – there is a tremendous amount of variability even within the U.S. and then you get outside of the U.S. And I think you put some people, you're going to hear very, very different but very clear opinions on whether you should or should not be doing that. So I think we are seeing a lot of regional differences, where some people are just – they don't want to use bisphosphonate, they don't think they work, versus some patients who were [ treaters ] who just think with these young children. But fair, that was the only thing available for a very long time, and I think people always want to do something versus nothing.

Right. Okay. So you have some potentially readout -- some potential readout this year. But really based on these interim analyses, so maybe give us a sense of what can you tell us about the interim analyses, and what the threshold is? And how you kind of dictate whether or not the arms are separating fast? What's the real kind of [indiscernible]?

Yes. And we are [ reducing ] power because, again, that was a Phase II/III study, so we really needed to use the data that was available to us at that time to power that Phase III part of the study and that was really data available in adults. So bringing forward that Phase II data in children with very formative had a chance to take a deeper dive into natural history and talk to experts. And it was really that much higher fracture rate coming into Orbit, the quicker response rate with the reduction in [ fracture ] rate that allowed us to take another look at how we were [ performing ] in that study. And that's when we made the decision to add those 2 interim analyses that are based both on time of follow-up and on efficacy and really, really focused on that 67% reduction in annual fracture rate after at least 6 months of follow-up that we saw on the Phase II data.

Okay. So based on kind of where you are in enrollment across the two -- well, I guess, the [indiscernible] study, I mean, what's kind of giving you confidence that you would potentially hit that [ interim ] this year? Is it based out in the Phase II data?

Yes. So all of that was built on that data coming out of the Phase II because that was 24 patients' worth of data that with that type of degree of reduction, bringing that annualized fracture rate essentially down to zero as a whole. And then looking at the type of fractures that we were seeing in those few patients that did have fractures predominantly being traumatic fractures, gives us a lot of confidence in how we set up those interim analyses.

I mean do you have a sense -- I mean, this is a very stringent first interim, right? You have a second interim. Do you have any kind of sense, timing-wise, when after the first one, you would potentially be conducting second term?

We do, but we have not yet guided on that, but it is fair to say they're all -- it's based on both time of follow-up and then efficacy target.

Yes. Okay. I guess in terms of timing, let's say, the data whenever reads out this year or next year, is the plan to kind of combine both studies? Do you expect them to read out around the same time? What's kind of the general thinking in terms of when you could file? Do you need both together to do that?

So I think it's fair to say we don't need to. I think you can say that the Orbit Phase III study would stand on its own. I did think it was very important to bring those studies in together. So we really kind of doubled down our efforts on both of those studies in the second half of last year to get them enrolled on similar timelines to be able to bring them forward as a package to support registration.

Okay. I mean I guess, have you consulted with FDA how long after a potential top line readout you would need to follow the patients? Is there a certain amount of time you need to build a safety database here before you could file?

Yes. So I think it's fair to say we had original buy-in on that Phase III design with these interim analyses. We obviously needed to submit these updated statistical analysis plan. So we're working through that, but -- very good conversation so far. I think, to me, it's nice. I trained as a biochemical geneticist, and I've been in this area, I have never seen this type of clarity and signals. I mean, you just can't. With these types of reduction in fracture rates, with this type of disease supported by BMD, supported by biomarkers, it's just -- to me, it could not be more clearer.

I guess maybe one last one on OI. When you kind of look at the commercial opportunity between adults and peds, I mean how big do you see the market? And I guess, where would you expect this drug to fall, respectively in both?

Yes. I mean it's always tough to [ pose ] that question to me because as a physician, as a treater, if these patients are fracturing, they can benefit from this drug. And I think people will make the argument that some adults are having a much lower fracture rate, but they're still fracturing, but they've also become very protective and very sedentary. They don't do anything. And that's not going out and playing sports or doing any activity. They become very sedentary, very protective any type of fracture. So that really becomes a quality-of-life issue. So to me, yes, these children who are fracturing annually multiple times a year, yes. But even these adults, that's what drives everything on their mind day to day, to me, they can benefit.

All right. I think we're going touch a little on OTC.

Yes. Maybe before we go to OTC, just on [Technical Difficulty] from Abeona that you brought in with the Phase I/II data. It had good activity. There was some data in specific subsets. So -- and I think you have been talking to FDA about both manufacturing plan and accelerated approval plan. Can you give us a sense of where you are?

Yes. So I think, obviously, we had the Reagan-Udall Foundation meeting last week, and that was very, very positive, very supportive. I think everyone was very happy with the direction coming out of there. We've had a lot of conversations with the FDA. I think Peter Marks, particularly has been very vocal on that. So I think we are seeing a lot of support for this biomarker approach here. And then it feels like everything, we're bringing manufacturing along in the background. But our focus really has been able -- hasn't really been shifting from a clinical endpoint, which would potentially require following all patients to their fifth birthday. The data shows you need to treat early, that makes sense with this type of disease. So that becomes a very long follow-up. So being able to shift to this type of approach is -- it's important for rare disease, it’s important for accelerated approval. Most importantly, it's important for these patients.

And is it going to be a CSF biomarker?

Heparan sulfate, yes.

In the CSF?

Yes, yes.

So you need -- you'll still need to run a Phase III head-to-head against placebo or it's going to be an open label or to be aligned with the previous data?

Yes. So the FDA had agreed with Abeona that, that data set could be used as a pivotal data set. They had originally agreed on a clinical endpoint had -- they just had -- not had a lot of discussion on heparan sulfate as a biomarker. So we're shifting that conversation, but this is the registrational data package.

Okay. And would it support a broad label?

Yes. I think with [Technical Difficulty] you do see a lot of consistency. Unfortunately, these children untreated do not make it out of their [Technical Difficulty] in early 20s. So I think it's fair to say that this would be a very broad label that there's just nothing else you can do for these patients.

And then you're on a natural history in parallel? Or what's the confirmatory study?

Yes. So natural history data does exist. That's been an important data set [Technical Difficulty] comparative, and we've been continuing to work on that. I think it's fair to say, we may need to conduct some sort of confirmatory study, but in a post-marketing setting that's behind.

Okay. So tech transfer is again critical here. I assume you're bringing this in-house too?

So at this point, no, we are going to continue, and that is really how we're going to be looking at the gene therapy portfolio broadly as a combination of CMOs in-house, really prioritizing -- and then we do have a lot of capacity to expand, but really starting where we're at and then expanding from there.

Okay. And -- maybe just shifting to OTC. So that data is expected first half next year, the Phase III study, when the change in the primary endpoint is a plasma ammonia [ likely ], 24 hours at 64 weeks. And there is no bar for the primary, but what's clinically meaningful, in your view?

Yes. So with the work done with alternate pathway medication ammonia is considered a clinical endpoint. So that's a clinically approvable endpoint. And with ammonia, I mean, we really do not tolerate levels above the upper limit of normal at all. So you really need to get those ammonias within normal ranges. So that really requires you to take a noninferiority approach there. So clinically meaningful is keeping these ammonias within the normal range, but that's why we're also bringing forward this responder rate and really looking at how many patients can completely discontinue [indiscernible] pathway medication in those that can reduce standard of care by at least half.

By negative 50%, okay, and normalized. And you think about – from a peer perspective, a peer might say that some of these drugs are [indiscernible] expenses, some of them are going to go generic over time. How do you make an argument from a – to a payer to pay $1.5 million, $2 million?

Well, some patients in the U.S. are being charged over $1 million per year forever for alternate pathway medication. So I think, for gene therapy, if it's a onetime dose with these patients with -- within their fifth and moving into their sixth year follow-up with very good durability, I think that's probably an easier argument.

Okay. And again, there, too, you'll need to bring that in-house, right? So you'll be -- you'll have to do a tech transfer?

Yes. We don't have to. I think it's fair to say we likely will.

Yes. is this something you can start doing now, the tech transfer process, ahead of time?

I mean, I think you can. I think with gene therapy, again, you have so much stability there. It's not hard. So I think you just may make the decision and you do it. And I don't think you want to run 2 parallel processes for too long.

Right. Okay. Got it. Well, Eric, I think we're just about out of time. So thank you so much for joining. We really appreciate it. Good to see you.

Thanks.

Thank you.