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

Good afternoon. Thank you, everyone, for joining us. Really pleased to have the Ultragenyx team here with us. We have Emil Kakkis, Founder, President and CEO; and Mardi Dier, CFO.

With that, Emil, data sets that are going to start to appear in the second half onwards, let's start with the Angelman program. As we look to that midyear efficacy and safety data update from the Phase I/II study, can you remind us of the metrics that will be considered, the number of patients we'll see data on? And what you would interpret as a positive signal for the program?

Sure. So we'll be releasing data on Cohort 4, which is less than 8 group; and Cohort 5, probably the majority of patients. We will have -- the hope is to get the majority of patient Day 128 data. And what we would be able to show you is where they've gone with their dose exposure and dose titration. As we said, several have already reached 5 doses, that is 4 doses plus a maintenance dose. And we'll provide data on the CGI score for Angelman syndrome, where are they at in that overall and for individual domains. And then we'll present data on the Bayley receptive/expressive communication, the Vineland and the ORCA, which is a communication score. And we might have some data on the EEG, but it depends. But the idea is to have enough data to say, is the drug safe? And what are we seeing activity? And the titration began, remember, at relatively lower doses, moving into the range where we said before between 5 and 10-milligram doses where we're seeing efficacy. And we're in the nonhuman primate, we see -- by the time you get to equivalent of 10 in nonhuman primate getting your maximum knockdown of the antisense message. So we feel we're moving right into that range. And we've noted already that we're seeing clinical activity. And the clinical activity that was consistent with what we saw before in prior 5. So that's the body of the clinical safety and efficacy data. We'll give a regulatory update and also a strategy update on where we're going forward from here on the development program. But we've been very encouraged with what we're seeing. And we think that this change in administration strategy has solved the problem that we were seeing.

So just a follow-up on that. So are you comfortable at this point on the safety profile that you truly have solved it? And I guess the second question is, what is clinically meaningful when we think about these measurements you're going to get?

We do think we have solved the problem. We've had no lower extremity weakness. We've not reported any AEs, SAEs. So we feel very comfortable that with that administration strategy, you're not going to see the, what I call, the local irritation effect. It's simply making sure you're mixing the drug and not letting it sit down at the bottom of the spine. So by the way, that is how other people are administering; all goes the same. So it's not such a surprise. In terms of clinical meaningfulness, CGI scores are relative change score, right? It's just, are you much better or very much better. The Bayley and other scores are a little more objective scores of a set of activities that you have to achieve. And so those will help gauge whether you're seeing clinically meaningful results, right? And those scores have normative scales that you can compare to and would tell you, are you outside what we considered as statistically significant change in those scores, right? So it would be those individual scores where you get a little more clarity on clinical meaningfulness. But I'd also point out that what we're seeing is an ongoing thing. When you decide how the Phase III is going to be, what we're really trying to determine, are the patients separating from what you would think as no change to determine that the drug is the right dose, the ultimate effect though would allow -- allowing that change to continue for a while. Even in the first 5, when we stopped the dosing, the patients were still gaining ground for several months after. They were still gaining function. So we'll have to design the Phase III so we load the patient appropriately and give them enough time for their brains to adapt and improve and their bodies to learn, right, more of the functions. But so far, we're encouraged and for me, first time in my career to see developmental function being gained by applying a treatment. It's something special. It's something I didn't think you could see, but it is happening, and we're very excited about it.

Can you talk about the functional benefits that you are seeing? And if there is a kind of a delayed response to that, what that -- as the body tries to acclimate. How long does that take? And I guess the third part here is, given you've brought the dose down, how does that impact the earlier functional benefit that you've seen?

Yes. So the functional benefit we saw before happened within a couple of doses. And remember, at the prior study, people started -- we started at lower doses, at 3 and 10 in the first group. Second group started at 10. But third group of 1 patient started at 20. That patient, one dosed at 20, had a response within the month. So at a high enough dose, you will see that. But even at the 3 and 10 doses, they were showing the activity and -- but because the titration was so fast, a little hard to know how the dose would have related had we just simply stayed dosing at 10, let's say, and repeated it rather than kept jumping. So we're going to have to learn a little bit more about that, but I think the time frame for seeing a change of some kind was a month or a couple months, right? But that process will continue to gain ground over more of a 4-, 5-, 6-month process. So if you think of the first -- patient #2, where we have the most detailed time course, because that one started in the low dose and went up, you can see them jumping up, particularly, just at the beginning, but also when we got 10. But their language gains were progressing over time. And even when they stopped dosing, they had gotten to a few words, but they ended up still progressing to get to 12 words over the next 2, 3 months, right? So in a period of a few months, they continued to gain ground. So the good part about that is that that's not 1 year or 2 years, right? So when we started the program, to be honest, we were thinking of a situation where we try to induce the gene and hope that over a long-term something would show -- that something would show. I didn't expect it to be within a few weeks to months that you start seeing changes, right. It just tells us how little we know about the brain and maybe how fast and plastic the brain can be. So I think that's very encouraging. But the ultimate outcome for a patient may still depend on 1 year or 2 years of treatment, right? We don't know how far it would keep going, right? Will it keep going? But my expectation is that the brain is learning, adapting and retaining now, which it didn't retain before, that we'd expect kids when they're properly knocked down and opened up that their trajectory will continue, which is, I think, the most exciting thing to think about is what it would be like when we actually keep kids on treatment for a couple years, right, not just 6 months.

Just the follow-up period with the patients that -- or the patients that we're going to see midyear, and again just wanted to go back to if you lower the dose, do you think there's going to be an impact here on efficacy.

Well, we definitely think there's a dose-dependent efficacy, right, because we saw it before, but we only -- we never gave enough of any one dose repeatedly to know what the accumulation effect would be on efficacy. So it's why it's a little hard to interpret. But what you're going to see is the earliest enrolled patients will have gone through 1, maybe 2 maintenance doses, right, which is every 3-month dosing, and would have titrated through and then the patients that are earliest maybe more limited data on the latest one. So it's going to be a pretty big range.

Can you explain the rationale behind each of the ex U.S. protocol amendments and then just -- how you -- well, I guess you've touched on this on the tolerability as you increase the dose, but where do you stand there as -- in terms of dose titration upwards?

Well, ex U.S., we haven't gone through the details of titration. The protocol -- remember, we set a ladder of titration. The protocol allows us to keep titrating until we get to a certain level of change. So patients will continue titrating into the range. We believe based on nonhuman primate that getting 10 milligrams for 3x or 4x equivalent should give you near-maximum outcome. And so that is right where we're at. So we know we went much higher before, but we also didn't give itself a chance to see. But when we saw the movement happening before, we just went zooming past it. So does that answer your question?

Yes. That's helpful. And just going back to efficacy, you've guided that a 2-point improvement in at least 2 domains of CGI scale is an important goal here. Is that really the bar that you're looking for with this first data set?

Right. The 2 -- plus 2 score in 2 domains for the CGI Angelman syndrome was -- for each domain, that was really a titration criteria. Like for this person in front of me, should I keep going higher. So once they got to that level, we'd stop titrating, right? So it's not really the end game of what the efficacy is. It's more of a choice of whether I should keep elevating the dose or not, right? So it's a titration criteria. We'll look through all the data, but we obviously would want a drug that made people much improved, right? So that certainly is important. Each patient is a bit different. Even in the first 5, some had terrible sleep problems; it got better. Some didn't have sleep problems. And so we have to look across the spectrum of domains to see what's improving. And we'll see where -- what the doses are that people get to, to make that call about whether we're at the dose we want to be. I believe with the new administration strategy, we're going to get to the dose. We will be at the dose range we need. So I'm very comfortable now with the change we've made that we're going to get there. So hopefully, the data then will provide that support. And then we'll expect to, moving forward, synchronize the U.S. and Europe or ex U.S. protocols around one approach, go into expanding cohort at the dose level that we've achieved and gain more ground understanding regarding the level of efficacy if you start at that dose level and now repeat it 4 times.

And turning to the regulatory front. Can you just talk about your plans for where they stand for aligning the U.S. and ex U.S. protocols and whether this midyear data update could impact the FDA's thoughts on the forward?

Well, the Canadian and U.K. authorities were very accepting of our strategy. In fact, they had no concerns about what we are approaching. And so that moved well. With the new data collected in hand and with additional data on immunological question the FDA was asking us, we will be able to approach the FDA now with to reopen the U.S. protocol on the same strategy being used ex U.S. And I think we have enough to get their agreement to that. We've dosed them at higher doses and with the administration strategy that was the right one. And we've been able to exclude immunological things as being a factor, which was one of their questions. So I think we've answered all their questions and the clinical data tell them that what we're doing is safe for patients.

And help us understand why the amended U.S. protocol so far started at such a low dose. Why they said no escalation and requested a comparator arm.

Well, in that discussion, and I was in that Type A discussion with the agency, they felt they wanted a 10x safety margin for the starting dose relative to the lowest dose at which the event was seen. That was 20. Therefore, we started at 2 in order to avoid another hang-up because we've had a series of filings with them. I opted, that was my decision, not to try to escalate the dose. They didn't tell us not to. I just said, just open the door, get this going. Because we're starting at 2, even if we were escalating the dose or starting from a lower place, it would take us time to catch up. So our view was let's get started, treat them with this dose. It will end up being like a low-dose cohort comparator, right, with the other patients, so we can compare what's going on at 2, what's going on at the higher doses. So it's like it's like a dose arm, right? The control arm, they just wanted us to prove that if you just look at patients and talk to them that you're not going to see changes like this. They, for some reason, hadn't really read the natural history information before, but there's a lot of natural history information, which was provided to them. But the truth is that Angelman patients with severe type don't really change over years; they don't change. So there's no way that just through excessive physical therapy or anything else that you get them to start saying words. Deletion-type patients essentially never gain language in their life, right? So if a deletion-type patient is saying words correctly, that is not anywhere in your natural history. So the 4 patients will give us like a control group. They didn't want to randomize it because they knew that it's -- they didn't want to have placebo [indiscernible]. So we just have been going along with that requirement. It's not really a control group, but it allows us to see whether we just reevaluate patients, what does the variation look like. In the end, the story is really going to get driven by the ex U.S. data, right? The ex U.S. data is where we're doing the optimal approach. We'll bring back the U.S. and get the U.S. arm straight. And I think -- in my discussion I think we've made progress with them. I think -- I feel like they're -- we are in a good position this time to move forward.

What does -- well, I guess, how are you thinking about the registrational path here?

Our approach in the registrational path is to trying to figure out if it's -- drug is working, then we need to design a path that gets us to filing, a submission as promptly as possible, right? Part of that is having long-term safety data. So our view is instead of just jumping to Phase III, it'd be better to expand the dosing to a larger cohort of Phase II patients to begin getting dosed immediately and collect long-term safety data. And you can do that very quickly by just expanding the current protocol, right? We'd also learn more about dosing, accumulation, safety, et cetera. So that group, if we expand, let's say, 40 patients on the top of the 10 or so, we'd have 50, maybe 60 patients that are on drug for a 2-plus year period, right, of time? During that expansion, we would then initiate the discussion with FDA, bring whatever interim data we have, talk through a design and develop that design. From that information, we drop in the dose when we're ready to make the call, the dosing algorithm and then provide a final protocol to them. Our hope would be to start Phase III next year. It's -- I think we can get there as long as we hit our marks and what we need to with regard to dosing. If we start Phase III, my hope would be to run a trial that's maybe 6 months in length, where we provide 4 loading doses and show the separate -- 4 loading doses and enough time afterwards to improve. It could be a 6-month trial. It could be 4 loading doses and 1 maintenance dose and then monitoring. And that is not a very long time for a treatment like this in terms of durability and efficacy. But if we had the 60 patients on Phase II, it would provide us the long-term data in a package. So you'd have a randomized controlled study, and you have this package of long-term exposure. The third thing we need to do is provide other supportive data for the missense mutations and uniparental disomy phenotypes, genotypes, and we believe we can do that in open-label format, that we do not need to run multiple placebo-controlled trials. We feel that would be not ethical. If you prove the work is very powerful and works in one, I think you could just verify that it works in the other genotypes and provide that supportive information. So that would be a package that would help us get the age range as well as the genotype variation with -- anchored by 1 high-quality randomized study.

How does natural history suggest as to the cognitive decline here in Angelman patients? And then just the last question on Angelman's. How do you believe your compound is differentiated versus other programs out there?

Well, with the severe deletion type, which is 70% of the patients, the natural history is pretty stable. They don't really decline, not distinctly. They generally stay pretty flat, not changing much. The missense patients do gain ground over time. They're not flat. They will gain function a little bit over time. They're more high functioning, but they're still limited. But I do think the ASO strategy would help in all types to make them improve. Now what's differentiating our ASO? Well, as we started, we normally, as a company, don't get into competitive situations. It's our strategy. I like to go where there's no one else with ultra-rare disease. In order to get the competitive strategy, we said we always had to have something that we feel would be superior, a distinct advantage. The GeneTx team who work with Scott Dindot had discovered aspects in molecular genetics of Angelman syndrome that was not publicly known, that was verified where the true 5-prime end of the message was and where the transcripts -- the various transcripts that existed, which are multiple, and that they're spliced. And the also really important discovery is that if you clip the transcript early in its transcription, you will cause the transcription to terminate. So an ASO design near the 5-prime end that is in a conserved sequence can actually destroy all the various forms of the message as well as clip transcription. And that combination is a more potent phenotype. The patented region they have is outside of the Ionis patented region because Ionis working with [indiscernible] patented the region, but the time they did it quickly, they did this huge sequence, but they just missed an end. And that end is the end that GeneTx acquired license to. So it's a unique sequence area that is exclusively for us. I know Roche are in the other area. And we believe you can make knockdown ASOs there but we believe the ones we make [ more here ] are much more potent, and that's our distinct advantage. And so we got into it because we felt that Dr. Dindot knows something that other people didn't know. And he wrote a tour-de-force molecular genetics paper, it's in the preprint server, that can tell you it's hard to get through because it's really detailed. But when you look through it in detail, he understands more about that locus than anyone, and that gives you the advantage of understanding how to tweak it to get this to do what it does. I think we have the evidence. Roche said they did 20,000 molecules, and I think they're great at this. And their best molecule worked at 24 milligrams. But with the GeneTx molecule at 1 milligram given 3x, we get the knockdown to occur, right? So 10 to 20-fold difference in potency. And I think that's the difference between where you are. And with the ASO world, as compounds ASOs are not drugs that have a lot of potential safety questions, right? However, if you can make it really potent, right, it could be like a scalpel and attack just the right thing. And I think for neurology, a really highly potent ASO that can touch just the right thing is the kind of control over the pharmacology of the brain that we need. And I think the GTX-102 is that type of molecule.

Great. Thank you. Mardi, where do you stand, I guess, from a cash runway standpoint? And how are you thinking about capital allocation just in these times?

Yes. Thanks, Salveen, good question. We get that a lot. We announced in the last quarter that we have $813 million in cash and cash equivalents in the bank. So we feel very good about our general cash position. But clearly, we're bringing a lot of programs to the development pipeline, offset, of course, by our continued growth of our commercial products with Crysvita and Dojolvi in particular. So we have sufficient cash on the balance sheet to get us through some key milestones in the next couple years, but we would like to bolster the balance sheet. We're always considering options of how we can do that. Fortunately for us, given the diversity and the number of levers that we have and what we've done in the past, we're looking at non-equity-dilutive solutions. And just to describe what we've done previously, we've done a tech transfer of our producer cell line platform technology to Daiichi Sankyo that brought in over a couple hundred million dollars. That was done in 2020. And prior to that, we sold the royalties to Crysvita in Europe for over $300 million. So there's lots of different ways that we can access capital outside of the equity markets. Given the general background right now, that's probably our preference.

One thing I would add here and that is sometimes the first quarter, you shouldn't straight-line the first quarter of cash because we have a plan -- it's just some onetime things that are going on, too.

Sure. Yes, of course, in 2022, in particular. We know people like to straight-line what's going on but in 2022, in particular, there are a number of onetime cash uses. In particular, we're finishing what we're continuing to build and finishing the build-out of our gene therapy facility in Bedford, Mass. So that is a pretty significant cash use. And then, of course, we had on our BD side, there's some onetime cash payments that get recorded in 2022 as well. So yes, good not to straight-line the net cash use going forward.

It does seem though right now that there is a possibility of being opportunistic with partnerships. And you guys have always done a good job of looking for assets that are out there. Are you, I guess, doing that? Are you kind of looking around at what's out there right now with more of a willingness to partner potentially?

We haven't really needed to look around. Most people are coming to with their old bags and showing us what they've got. I mean there's a lot of people with good stuff out there that are in tough situations. And it's one thing for Ultragenyx. We're at a point financially where we've got greater strength. We did bring in the Sanfilippo program from Abeona, which was a deal which they were potentially going to terminate the program and we were able to pick that up with -- only with royalties and downstream. So that's 1 example of an asset that a lot of good money invested, we think it's a very -- we think the data are very strong on its effect on that disease, which is a lethal disease and a horrible disease. So that's one. We didn't expect to add it. We're going to be very selective and thoughtful because we don't want to dramatically increase our burn as well. But I do think it's a moment in time where a high-quality asset is something we should be looking at, at this moment and [ do it if ] we have to to put us in a position to gain value for shareholders.

Yes. And we continue to balance that. We're being very prudent in our spend going forward despite the pipeline. Sometimes it's a reprioritization or we really look how we can be efficient if we bring an asset in like Sanfilippo, which, of course, late-stage development. We have no upfront on the deal, which Emil said. So it was a good -- from a financial standpoint, it was a good transaction for us.

Where does cash take you until at this point on the run rate?

Yes, next couple years.

Wilson's disease, you have a Phase I/II/III that's ongoing. Could we get some interim data at any point?

Well, the Phase I/II part of the study will have 3 cohorts. And through agreement with the FDA, we have to stagger enroll each patient by several weeks, which drags out the time line for that part substantially. So we will expect, when we get through those cohorts, to put out some information about where we are on Wilson. But I don't think it's going to make it -- it might be next year. But I think it possibly could be after that, we'll have to know. I would say, in general, we're seeing that the world had long-COVID right now. And there is -- it's hard to get sites to go full bore because they have a lot of staff shortages and fatigue and other things. So we're going to work hard in keeping those programs running, and it is a bit of work because of the situation. But I think Wilson is a really important program for us. And I think the potential to really treat Wilson for the first time, not just sop up the extra copper but actually get to the problem, I think, gives me great hope that you could actually see an effective Wilson that is well beyond what current treatments can provide.

Commercial portfolio at this point. I guess, when you assess your trajectory that you've seen in these products, what are you thinking of these peak opportunities at this point?

So within our commercial portfolio, we'll continue to see growth of Crysvita, our lead asset, even with the transition of the marketing back to KKC in a year from optically from looking at the P&L, with the continued growth, the way the deal is structured, you should see continued growth moving forward in Crysvita. So that's been an incredible asset for us and will continue to do so. Dojolvi, which we launched during the pandemic, continues to see great uptake in terms of new patients, new patient clients, new physicians just prescribing Dojolvi in the U.S. So we're really excited about the trajectory of Dojolvi as well. I should say regionally, back on Crysvita, Latin America, we'll continue to commercialize ourselves, and we see a lot of momentum there. We're very close to getting full approval in Latin America. We're at the very final stages, and the number of patients that we continue to treat there continues to increase. We just crossed over 200 patients, which is fantastic. There's a great pipeline, mainly driven by Brazil, but the other countries are beginning to show some strength as well. So excited about the trajectory there. That will be a meaningful contribution to our top line moving forward. And then the pipeline, while we spoke a lot about Angelman but we also have our next bone disease in the pipeline with osteogenesis imperfecta. So a number of these studies will continue for the next few years, but mid- to mid-late decade, we should see an inflection point of revenue with these new products coming through, which will be an exciting time for Ultragenyx, right? So those 2, in particular, are quite large opportunities. Same time, the gene therapy programs will start coming through as well, which will be an excellent proof of concept with our 401 program and our 301, excellent proof of concept for the rest of the gene therapy pipeline behind it. So a lot of excitement coming out.

Great. Well, with that, thank you so much, Emil. Thank you, Mardi. Thanks, everyone. It's a wrap for our conference.