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

Thank you, everybody. How do I follow that? I guess I have to energize you and get you all awake now. So my name is Daniel Harvey. I'm with the Angelman Syndrome Foundation. I've been on the board for a long time and I have been involved in the science of AS for decades. It's an honor for me to introduce the first speaker next session, Emil Kakkis from Ultragenyx. Emil is not just the President and CEO of Ultragenyx. He's the founder of the organization. They founded in 2010 with a focus on the development of treatments for rare and ultra-rare genetic disorders, so an area that we are very excited about here. They've had quite a bit of success. They have 4 approved therapies, 4, 5 different rare diseases, and they have multiple late-stage programs, including an ASO-based treatment for AS that is in the clinic that we have heard about some this morning, and we're going to hear about more this afternoon. Emil has a lengthy list of awards. And if I read them all to you, he wouldn't have any time to speak. So I won't mention those, but I did notice, as I was reviewing his file that he was a graduate of Long Beach High School a few years ago, about the same time I was in high school back then. So I'll turn it over to Emil for an industry introduction for this session. Emil?

Yes. Wilson, Long Beach ruins, yes. So I'm happy to be here to give you an introduction. I have only a short time, so we'll fly like a drone quickly through the whole span of things and hopefully give you a feel for a little bit about what we've learned about drug development and rare diseases. And if there's a word to talk about drug development diseases, it's resilience. Resilience is the word, as a parent waking up each day and dealing with what happens in the day, what happens when you deal with the diagnosis itself. It's what we deal was in drug development and rare because there's so much unknown, so much figure out in the fly. And so we'll tell the story a little about who we are at Ultragenyx, but a little about my own career that leads us to are now in trying to develop drugs for many rare diseases. This is just my financial disclosure. I will touch on GTX-102. It is an investigational drug, and it's not approved by any regulatory authority. So Ultragenyx, you heard a little bit, just a moment ago, 2010, me and my secretary, we're going to start a company. It always seems totally improbable, I did have a couple of million dollars. My wife let me take that to do this, which was very generous of her. That's a big chunk of what we had and the worry she had was like does all of it is going to go in the end. But you start with a vision to develop rare and ultrarare disease treatments and to be very focused on products, getting the patients, doing it quickly. And in that time, we've -- now in 14 years. Were 1,300 employees globally. We've gotten 4 approved products. We're working on 20 different diseases. Angelman is one of them. We work on 20 different ones. Some are like Angelman like CDKL5 deficiency, but there's a lot of others, all kinds of different treatments. And nothing more exciting, I can tell you as a scientist to be able to bring new treatments to patients, and we'll tell you a little bit about how that happens today. This is our mission, rare disease patients and not only just getting drugs approved, but really it's about leading the future of rare disease management, which we've changed the methods, improve the process, the regulation, the design of things, make it more efficient, make it more possible to treat ultrarare. And some of those issues will come up as I go through the talk. So just for anchoring, this is where I started Harbor-UCLA in Torrance, that was where my lab was in that building. It was very dusty. It was old. But the truth was you could do good research there, and there's a lot of good research that went there. But there wasn't a lot of money and when we're working on genetic diseases and rare diseases, there wasn't really kind of investment that goes on for a lot of other types of things. But the science was amazing, and we did good work there. That was that very beginning, but as they started working on a disease called MPS I, a lysosomal storage disease, a terrible disease, it's my first connection to patients. And this is Mark Dant with his son Ryan. Ryan has MPS I. He is 5 years old. He got diagnosed at 3, and his father is expecting him to pass away by the time he is a teenager and he was coming to me and say, "Can you save Ryan?" And he's handing me a check for $40,000, and that's not enough to do it. But there's a bond there, scientists we had a treatment idea that actually was working in dogs. And he says, "I'm going to devote everything we have to this." And that was my first connection to the personal deep connection to the struggle, right, and the battle. And he started raising more money, and we set off on the vision we had no idea if we're going to get there, but I knew I couldn't not try, and when we tried. Fortunately, a company brought in some money and we're able to treat 10 patients. These are 10 kids -- 9 kids with MPS I in this picture, and it was working. The treatment worked. What an amazing thing. But then the FDA said, well, that open-label trial is not enough. You need a randomized trial now. It became a subject of a 60 minutes episode and we got -- and then we got enough money now all of a sudden. We'd already spent like $30 million. I'm thinking that's already an incredible amount of money. I mean there's only 200 patients maybe we're going to treat in the U.S. And so we managed to figure out how to get the Phase III done. And we got there but what I am learning is that this process here, you can see the first few years on the top, getting there with Mark and then treating. And then that picture there for the company called BioMarin, those are all the people involved in the filing for approval, 150 people, almost as many people working on the drug as they are going to get treated. And then I learned how many technical skill sets and things you needed is just -- and we spent $130 million, $150 million to get a treatment that's going to treat 200 people. That's kind of insane, right? But the truth is what are you going to do? That's what it takes. And I've learned very early on that telling parents, being altruist and saying, "I really love to help your kid, but I don't have money to get the treatment developed." There's no benefit to that discussion, right? It's only failure. So the truth is you got to get what you have to get to get it done, and it's hard to change the whole system, but that's what it is. So I wrote a book on this struggle. If you want to read about it, you can read about this book, Saving Ryan. But the most important part of the struggle is the fact that patients have a real big impact on developing drugs, just like your organization, ASF, will have such an impact. The good part you see at the end of the book is that that's Ryan. He is now 35 and he got married and I went to his wedding. He can get there. That's what we hope for everyone that has a rare disease, to get there. So this led me from academia to industry because I learned something. It doesn't matter if I care about rare disease patients. If I don't have the money, I can't do anything. In order to get the money, you have to go to a for-profit company where then can raise the money. We have to deal with the money part, but if we get enough money, then we can treat more rare diseases and figure out how to make this work, and that's what I've been doing. I spent 11 years at BioMarin and now another 14 at Ultragenyx. But in my time, I've been involved in 11 drug approvals of all kinds of different diseases. And it can get done. And these -- once approved, then kids with this disease, when they get diagnosed, again, they're not told, no, there's no treatment. I said, "No, there is a specific treatment. These are all specific treatments for those diseases." So what have we learned? Rare disease drug one is hard, as you probably know, small patient population, although Angelman has more than some of the others, I'll just be clear. One of the disease we have has only 20 patients on the treatment in the U.S. But it worked and we got it approved, but these are complex and variable. And after history has been collected and many participated, which is great. There's more natural history in Angelman than I've had in any program I've ever worked on, right? So that's good. You've done some good work laying the foundation. And the trials are extensive, and I'll talk a little bit about trial design and development. The problem is it's been getting harder not easier, that is as there have been some rare disease successes and the regulators and reimbursements and everyone says, oh, good, it's doable. Now, let's ask for this, ask for that and add more and more things and then ultimately, it gets more and more expensive. And I'm trying very hard to try to keep and manage that, the time, how long it takes and how long it costs. With brain disease and slower progressing as it can be more difficult because of this. But despite that, here's what it actually takes to file. Now, I put this, and you might wonder why am I showing you these numbers. I just want you to get a feel for what happens in the industry because I didn't know it until I got there. So if we want to file a drug for approval to the FDA, what does that take? Well, it takes something like 100,000 to 300,000 pages of documents, all right? So it's on the order of, let's say, 1,000 individual documents, 2,000, 200 to 300-page full, right, some of the documents have hyperlinks between different pieces. It takes program software to connect all the documents. I just want you to understand the scale of this, right? It is mind-blowing. And in there, all these little bits information the FDA somewhat wants, and that's what it takes. It takes then a lot of people, and that's what's going on. And I can't change the system exactly, but we got to do it. This is what it is. It's not going to stop us, though. It just means it takes a bit of work. So the exciting thing is that technologies are coming. They are actually things and there's multiple technology being applied to Angelman. That is exciting. What we've found is it's one, and we're very excited about what we're doing in that, but there's so much more going on. And fortunately, getting Angelman Syndrome is never what you hoped for, but it does have some good features for being treatable. And I will say, too, there's a lot of disease where the possibility treatment is maybe none because of what it happens with the kid that may be irreversible but it's not a neurogenetic disease. The biology is possible, you can make neurons talk to each other where that's possible. That was figured out and there's actually a fair number of Angelman patients. That means it's really more easy just not -- when we did a 12-patient trial for that one disease, I had to move people from around the world to the United States for a year to run the trial. That's how hard it was to find patients. In this case, that's not true. And of course, a highly motivated patient community. So that keeps it going and you guys have been amazing. So a lot of treatment strategies are working, including in some neuro diseases. But what I would say to you is the number of successes in neurology disease has been less. Neurology is hard. It's because your brain is complicated, understanding and measuring it is harder. And therefore, there have been some great successes, like a couple of you here, but there's also been a lot of failed studies. So neurology makes it hard because it's so hard to measure and evaluate things. The brains are so complex. One of the challenges in all of this is that there's of variability. I'm sure you guys see all your other kids, you kind of know they have a lot of common, a whole lot of -- were different things and how they are. That variability becomes a challenge of trying to show someone better or not and how do you measure it. And it's part of what we think a lot about several areas I'm going to talk about now about drug design, which are important and what we do differently. One is about dosing, and the other is about endpoints and then we'll talk about trial design. Dosing, like how much drug do you give someone, becomes a very highly individual thing because each kid's body is just different and the drug may act the same way in all the different ones, but it's not the same degree. And so one of the things we like to do is use individualized titration strategy to try to get optimal, at least to get closer to optimal. And so finally, we've dosed with dynamic in visual titration. That's what we did in our early GTX-102 study. We switched to that strategy, we start you low, we titrate you up and we look at how you're responding and try to make a sense, are we where we need to be or not. And from that, kind of learn what's the range of doses, what is the likely response. It helps us go a little faster because otherwise, if you've set on a set dose, you may guess wrong and spend 2 years operating at the wrong dose. So creating this dynamic allows you to change and optimize on an individual case, and we also allow regimen optimization, too. So this leaves us some flexibility more quickly get to an answer. This is what we did in the first dose titration trial that we did. I won't go through all the data, but we had 4 loading doses and then the maintenance doses. We started a group at low doses, titrated through to see where they would go. And then we started another cohort, but this time, we started higher because we started to say, "Hey, you don't have to start as low. We you can start higher, and we start higher and move through." And what this does over it turns on all these patients are getting to about 12 to 14 mg doses. That's where it's coming out. There are some kids that respond to less, not everyone is the same. This gives us an understanding of how much variation there is and gives us a handle on how to think about dosing in the future. Now the one other tricky area is endpoints, how do you measure if you're better? One of the weird things is the difference from you going to your doctor versus us doing an experiment. So you go into your doctor and your doctor is evaluating your child with Angelman or something, you're not -- and even for your own health, you don't go to the doctor and he doesn't measure one thing on you whether you're better or not, right? You don't have a single primary endpoint on your clinical care, right, right? Who would decide everything on one? You'd ask a whole series of things and you pull it together and you come up with a feeling, "How you're doing? How is your disease doing," right? But then we go into a trial, we say, "No, no, no, we're going to do one primary endpoint, and we're going to decide everything on one thing. We're going to ignore everything else," that doesn't make a lot of sense, especially when you have a disease like Angelman, which everyone is different and everyone has different things. I've been struggling with us from the beginning because the statisticians make us do one end point. The truth is all of us -- no, there's more than one thing that matters. So why are we talking about one? And why are we just saying that one is the most important then others? How do you rate them? Each person might rate different one, right? It's the #1. So it's one of the real problems and I've been struggling with this for a while. We've come up -- the guy working with me on the Aldurazyme program came up with what we call the Multi-Domain Responder Index. I've been pitching it because I want you to understand it. The idea of this is another way to look at multi-domain, many different clinical problems a patient has and look at them together and assess how you're doing. And I'm going to explain this to you because I think it's kind of important, and we are going to use it in our Angelman Phase III, so you can understand it. But as I show you the data, I think one thing I've heard from patients is when they look at the table, look at the data, they understand what's going on in the trial. That, and they're looking at a graph of the line on it. So I think that's true. So I first developed this with MPS I. MPS I is a terrible disease. It has all kinds of physical problems; lung, heart, joint, everything -- every part of your body has a problem. And for those parents the same, heterogeneity, all kinds of different things going on. And what -- the idea of the multi-domain is trying to say, "Look, let's figure out what are the 4 to 6 clinical problems are really important, the ones that really matter to patients, right, and parents. So let's just score each one. Do you get better by a lot or go worse by a lot? Or if you're kind of a little up and down a little down, not a big move, right? We want to know who gets better a lot. We want to figure out the big wins, right, count those. If you're getting worse, that's certainly not good. So you count the big wins or big losses, and then you look across all 5 of those endpoints, it's kind of like a scorecard, right? You have 5 of your endpoints, did you win? How many did you win on? How many you not win on. It's a pretty simple concept. So we did this for Aldurazyme. Now this is a case we look at lung capacity, walking, sleeping, shoulder, function, visual eyesight assessing throughout what's the big win, what does that look like? What does a really big improvement look like? And then we score it. Now this is data from a randomized trial. On the Aldurazyme treatment side are a list of patients 24 through 45, and those are the 5 domains. The red boxes indicate they got better by a lot. And you could see on that side that the red is much more than yellow and yellow is I got worse. So you can look and say, wow, the reds beat the yellows by a lot, right? On the left side was the kids who got placebo. So there, you can see that the yellows are more than the reds, right? So you immediately get a sense that, for example, in visual acuity, the placebo patients have no change, but 5 of the patients had a substantial improvement in their vision which was very important to them. It may have been a small fraction, but it was -- for those kids, it was really important, their visual acuity, which was poor, all 5 that had poor vision, those were those 5 got better. So you begin to learn more about what's going on in this approach. Now we think it's a great way for analyzing Angelman Syndrome in a more powerful method to capture a better feel for how kids are doing. So in Angelman, of course, in multiple domains, if you do and know all the domains and we've done analysis using this in Angelman syndrome as another way of looking at it. It doesn't mean the other ways are not useful. It's just another way of looking at the same data that helps you understand it. So when we did it with Angelman, this is very complex 2 slides. On the left side, Cohorts 4-7, those are the dose titration patients at day 338. So that's 1 year of treatment. As you can see there, there are 4 domains being shown and you can see the result in the green. The green means they're better. And so you can see on the far right, all the patients, but two, one had a decline in one domain, one had a zero, then everyone else had a -- usually 2 to 5 domains improvement, 2 to 4 domain improvement. On the right side is the most recent data Cohorts A and B, but that's at 6 months, Day 170. In there, you can see all but 4 patients had a at least 1 domain, usually 2 to 5 domain improvement, but you can see where they got better and it gives you kind of a feel for how they're doing. The power of these methods is very high. These value change with just 12 or 20 patients. P values are less than 0 and 1, right? They're very strong. This pattern of change, it's a very strong statement towards the efficacy, and we feel very strong about this. And the FDA has now accepted we can use this as a key secondary endpoint. which is important. It gives you another way to look at your data, and I think an important way of understanding what's happening on a patient-by-patient basis in terms of benefit. So that's one thing, and I want to move quickly along because I don't have a lot of time. So finally, proving a drug works, the last thing I want to talk about. The FDA loves randomized, double-blind, placebo-controlled trials, this is what they like because they feel this is the rigorous way to do it. And sometimes it's not possible. Now for history control trials, a lot of people want. I know in some -- in the Angelman asked us about doing it that way, where we just have some other patients who were studied and we just compare how these guys are doing. The truth is it's hard to do this in Angelman that way because of a few things I'll touch on. But I just want to touch on this two ideas of the trial. Now I've been involved in programs that did do a natural history of control, but this is a disease called Batten's disease. These kids declined fast. They -- between age 2, they get diagnosed and they go into a horrible situation within 2 years, but they're going to -- they're almost to death. So it's a very rapid thing. You can't do a randomized controlled trial. The control arm would be assigned, right, to decline in death. In this incredible situation, then natural history -- it had been done historically, and it was so profound the FDA allowed them to do the treated group and just compare. And you can see the treated group survived. The other patients are being lost. It's not hard to see the delta. It's big. So you can do this kind of strategy, but it has to be dramatic. It has to be like irrefutable score. This is -- these are very profound effects. When you talk about something like Angelman, it's going to be much harder to do that. Natural history is very flat. So you think that's pretty good, if it's flat, then you can be pretty confident they're getting better. It's real. The problem is the endpoints. We asked cognition, in Bayley, these are -- where we're asking questions, interpreting can be biased. I believe the data we have are very solid and very real, but people ask the question. They still ask a question, "Is placebo or not? Is it a real effect?" The type of endpoints we have to do in the developmental disorder make it very hard because you can't be absolutely sure that the patient in your study is like the natural history patient. Can we verify that? So it's very hard to get regulatory authorities to agree. Not only that, when we finally get approved now, it's not just the FDA, it's like all the reimbursers are saying, "Hey, I don't like the quality of your data," and they're looking for any excuse to say, "Hey, we don't think we want to pay for you treatment." I'll say to you, "No matter what, at Ultragenyx, if we get approved in the U.S., each and everyone are going to get treated and needs to get treated. No one insurance company can stop that, but I'm just saying to you, the better your data, the less trouble you have down the road, not just in the U.S. but everywhere." So we did meet the FDA and we've agreed on a randomized control design. At this point, they had suggested to us that we should do sham, not placebo. We're still doing the lumbar punctures. Sham means they don't actually give a saline injection to the controlled patient. They just -- they do the tap and get safety information for us, but they don't actually inject. I don't think it's truly different than safety. I think placebo would be better. However, the regulators feel that there is a question of ethical -- well, there's ethical and there was a sense that patients prefer a sham. So we're doing the Phase III sham. We don't want to delay and struggle through this discussion about how you do a control. It will be a randomized control. That means a control group will have procedures and tests that won't get treated. But when the 48 weeks is up, they will get treatment. So anyone in the trial will get access to treatment years before anyone else. So it's part of the contribution we need everyone to make in order to get this drug -- but we told the agency, and I'm telling you now, we're doing one controlled trial. And then the other types, we'll do something else. One controlled trial should be enough to prove that what you're seeing in these Angelman kids is real, all right? We prove it for you, but we don't have to keep doing that over and over again for multiple people. The endpoints with Bayley 4 cognitive test, it's the best test. It basically, and I think -- actually, I don't want to steal Kim's -- Kim is going to go more in detail this. So I'll leave that to him to talk about, but these are the basic end points. She's going to go through the Phase III trial. So right now, we're finalizing the protocol and planning the little pits, their addition. We're continuing to treat those Phase II patients. We won't take people off of the treatment. The key thing for some of you will be missense and other no-ndeletion type Angelman. We are going to do a study. We're planning the design right now. Our plan right now is that study is an open-label study, not a randomized trial for those patients. So you've waited but at least you won't be in the randomized trial, so there's a plus and upside to the wait. That trial will go on. The Phase III we have to get started quickly that, that trial will start soon after the Phase III in 2025, and that will help get patients missense, UPD and some of the other types, plus some age groups like what about adults, we'll include adults. We're going to manage really young patients as well. We'll collect data and the hope -- not proving it, but the hope would be to use that data to help support that when the drug file goes in that, that will have shown safety and efficacy in those studies to support the Phase III data in the main population. That's our strategy. And we've done that strategy in multiple of the drugs before. So what do the Angelman community need for the next couple of years? Participate in trials. I feel like -- it feels like we have a lot of people who want to be in the trial. So I don't know there's a problem, but is definitely a burden. We know that and we're going to try to manage the testing burden as best we can. We are tuned to it. We're paying attention to it. I think it's also going to be important for you to provide the caregiver perspective on what's important for Angelman families because when we show results, someone's going to say, "I don't care about that." But you might care a lot. They're not -- they're going to hear what that is, right? They need to hear that my kid is sleeping through the night, why is that a big deal versus not sleeping through the night.The people tell me, "Why does sleep matter?" I say, "Oh, it matters when you're not getting it." So I don't know why I have to explain that to people but apparently, they have a problem with this. So you'll need to be active, and there may be questionnaires and things come out. I'm just saying until they come, please fill them and do them because they need to hear your voice on why stuff matters and how much matters. And always keep us informed about your concerns. Don't hesitate, send me an e-mail or Kim or someone. Happy to hear everything, all right? So thank you for having you here. My time is probably a little bit long. Kim speaks fast, but thank you.