Welcome back, everyone.
Welcome.
Today we have the pleasure of speaking with Emil Kakkis, President and Chief Executive Officer of Ultragenyx. Emil, to start, just to level set, you are sitting in front of key phase III Angelman data in the second half of the year on the potential path to profitability in 2027. In that context, how is Rare positioned from here in the second half of the year and then over the next few years, and what are the company's key priorities?
Certainly. We certainly are happy to be here and be talking with you today. We started the year off with a difficult problem of missing an OI at phase III, and that was tough. We have a lot more than OI in the program, in the company. Angelman is the big driver for what people are most interested in. It's the biggest upside product. We've done a lot of work in preparing for that phase III. We put out some data at the last quarterly call, which talked about the long-term phase II data. What that phase II data shows is that patients continue to gain ground over three to four-year periods. They have durable effects, and that the safety is excellent, and there's no cumulative safety concerns or issues arising.
It also tells you something even more important, which affects us commercially, in that patients and parents are interested in keeping getting intrathecal infusions every three months for years for their kid to see the benefit they're getting. There's very few parents who keep going for intrathecal injections for a drug for no reason. It tells you something that people are staying on the treatment. It tells you this is a viable treatment that's really helping their kids and is worth it. That's what we got out of that. We did a lot of work on our phase III and operationally, which was bringing forth sites that we had already tested in that phase II study, so we knew they could do the test, we knew what the patients were like, and we had ironed out any particular executional issues.
We brought in a third-party tester for the Bayley, which is one of the primary endpoints, to make sure that was done to perfection. We've supported the sites as needed now to be able to handle the workload. It's gone well, and we're really excited about being able to enroll in six, seven months and to now be looking at patients finishing the study. We have already had patients finished in crossover, and we continue doing that process. Last patient in was last July. Four to eight weeks takes to sometime this July and June-July. Then there'll be some weeks to months of cleaning, locking, preparing, analyzing data before we release. We haven't put out a specific timeline.
I think we're in very good position with the phase II data showing us long-term durable benefit and long-term gain, and a phase III study that's been executed well. Finally, we've added the Aurora study, which will look at the older and the younger patients and the other genotypes to fill out the story together, I think putting us in good position. It puts us about a year ahead of everyone else, and I think we're pretty encouraged. That positions us well for big upside. If you talk about revenue and cash, the two gene therapies probably have a more immediate impact. We've been investing in the launch of both programs, expensing product that we've created and put on the shelf, and some of our burn in the last couple of years has been to make product sitting on the shelf.
It's not burn that's gone. It's sitting there as product. Our PDUFA dates are in August and September. So far, reviews are going fine. If we get both approvals, we'll have two PRVs, we'd expect, which have been selling for very significant sums of money, which will add to our cash balance sheet and put us in good position. In addition to that, we'll be launching both of those programs. Sanfilippo is an urgent disease. Of course, there are many patients interested in getting treated, and so we had a lot of inquiries both in the U.S. plus also ex-U.S. It's a lot more like ZOLGENSMA for SMA in terms of the urgency. We think that will drive uptake fairly quickly for Sanfilippo syndrome. For GSD1a, it is an urgent disease, maybe not as urgent as Sanfilippo.
They're not losing their brain, but the idea of taking starch every few hours forever is tough for patients. They want to get off that treadmill and put away some of the fear they have that every day they could die if they miss a dose of their starch. Both those, I think, are programs that will do well, and I think our gene therapy franchise is not valued at all, I think currently. If you think about that much cash brought in plus the revenue, the swing between this year and next year will be very substantial and put us in position to launch Angelman well. Finish 2027 with potentially profitable year with three new products launched, then other products in late-stage development, which I think put us in a really good position for the company going forward.
I look at this year as we get to third, fourth quarter as being kind of transformative for the future of the company. I think on top of the base business, which is earning in the mid $700,000, $700 million range, has been growing by 20% a year last few years. I think we're real well set up as a global company to take advantage of those three products launching globally, and which we own the full rights to. That puts us in position to really transition to profitability, but soar past it. I feel good about where that is. It would have been nicer to start with a positive OI study this year, but we take the challenges that are put before us, it is biotech and it is hard.
I feel good about where we're going, and I feel like, I think it should be a good year for us, and I think at our current valuation, I think it should be a clear opportunity for people.
All right.
Apart from taking-
No, that was helpful
a very long period of time.
Thank you for laying all that out. I guess maybe just quickly on the path to profitability before we dig into Angelman and some of the other pipeline products. You have two potential PRVs that you mentioned. I guess, how do you plan to time the monetization of those to bridge that gap to profitability potentially in 2027?
Well, we certainly plan to sell both of them, the exact timing could depend a little bit on what makes sense. Certainly, we could sell one this year and one next year to manage the burn. I think Howard and I will have talked through it, but I don't think we want to overthink it. I think cash and the balance sheet are what we need to do, but you could look at how that affects profitability next year. I'd also point out if Angelman's positive, that's another potential PRV, which could be in 2027. That itself could put us where we want to be, and I would say to you, we could do it this year, next year.
If we are able to achieve $200 million for each of those, which is what people have been getting close to at this point, it's a substantial increase from where we were modeling in our own plans. With the reauthorization, another third one, could potentially bring us $500-$600 total to the balance sheet. It's a pretty big deal without any dilution. I think that puts us in a good place from where we started the year with the quarter, we had $534. If we add that on top of the gains from revenue, it puts us in very good position to maintain an adequate balance sheet and hit profitability in the timeframe we're looking at with additional cost controls we've put into place.
All right. Maybe just digging into Angelman, ahead of the phase III data in the second half as you laid out. What do you need to see from the primary endpoint of Bayley-4 cognition in order to see a successful study? What are the kind of key risks here?
Well, we have looked at the modeling of the power of that endpoint before. The data we just put out in phase II suggests that at 12 months, we could see somewhere close to 10 points of improvement. I think we modeled a smaller number than that. With the standard deviation we saw, we were seeing 80%-90% power or greater, depending on what you assume for the control group. If you assume what we see in natural history or in studies, then we have 90 %+ power. If you assume the background could be two to three times higher, a little bit less than that.
The history of control groups, whether it's the levodopa trial in Angelman or the natural history study, they really haven't gained more than a point on Bayley, we feel pretty comfortable that there isn't going to be a big placebo effect going on there. Therefore, I think we're well above 90% power to detect the magnitude of change we're seeing. I think we're in a good place. We've put out a little more data on this power before, we feel comfortable with the strength of the study. That said, it is neurology is always hard, there's always things that go wrong or things that go off, we have to be conscious of that. We put in the multi-domain responder index for two reasons.
One, it's a more powerful assessment of all the domains that are affected, I think a better alignment of what patients are thinking. They're not thinking about one endpoint. None of us think about any disease we have as one point. We all think of multiple problems, right? The mDRI is a better alignment of what patients think about what are the big five issues that affect their kid. Also importantly, it's a very powerful analysis tool and gives us more power. If there were any problem with Bayley, mDRI is more powerful and will capture more endpoints, which help balance out if there's any risk. We're allocating 20% power to the mDRI, 80% to the Bayley. 20% is enough for the mDRI to be successful. It doesn't need very much power. We expect it to be able to beat that level readily.
That's how we've designed it. We hope that it'll help put us in better position to win one way or the other, if one or the other is positive, we believe we'll still be able to file for approval with that. We expect both to be positive.
All right, I guess maybe just taking a step back, could you just talk to the rationale for choosing the primary endpoint of Bayley-4 cognition as opposed to receptive communication, which is what Ionis has selected for their phase III study?
Yeah. I think they picked expressive probably.
Yeah. Expressive communication.
Yeah. I think their argument was that expressive communication is what parents want, but I would argue, I don't think parents want their kids to start talking but not having any thinking. That may be what's happening already for us with teenagers. Sorry, bad joke. There's no one here to laugh anyways. No. You can't have anything without cognition. The idea that parents want it, well, I think that if you look at the cognitions at the core, we think it's the most fundamental function. FDA has expected it. I would say to you we're evaluating all the other endpoints as well. We can look at all of them. We'll look at expressive, we'll look at receptive. Expressive does take time to evolve, and it also can be dependent a little bit on how much training or education.
I think it would probably work better if you're doing more work on educating kids to help them develop their language skills. I think all of these things are important. We picked Bayley-4 as being a more fundamental function. There's also history. We're also being using it in the Sanfilippo program. It's something FDA Neurology is familiar with. Those are some factors. I would not look at it as defining what's most important in the disease. It's what we need for a regulatory process. What patients will see is that endpoint, secondary endpoints, and they'll see the whole thing. I think patients can understand endpoints, whether they're primary or secondary, honestly, does not really matter to patients. They want to know what's happening. They're not going to pay attention to the regulatory rubric of statistical plans.
You touched on this briefly already, but just given the one-point improvement seen in natural history historically, and then the 10-point improvement that you saw in the phase I/II, I guess what's your level of confidence that the sham control arm in the Aspire study will not exhibit a significant placebo effect that could kind of compress the treatment effect size?
The main reason is the type of endpoint it is and the type of patient we're dealing with. The Angelman patients will be gaining function, but they don't necessarily have the understanding that they're in a trial. They don't really know what's going on. The placebo effect depends on you knowing something and having some belief in something. That's not really going to happen. We exclude all caregiver input, where the caregiver could have that placebo effect. They know, but that's being excluded. We think that will help. The other thing is that this is assessed by a third-party psychologist, not the parent, not the doctor, third party. The value of that is they're more objective. They're not connected to the family. They don't know the family. They're just doing the testing. Does that make sense?
They're, I think, going to be a little bit more objective, and they're professional in science. Those are the things we're doing to help control for it. If you look at the randomized placebo-controlled trials that have been done with Angelman, which is the one levodopa trial, for example, that was placebo-controlled. There was no Bayley placebo effect observed. That's not a natural history study, a natural trial. It didn't occur in that study.
That's helpful. Maybe just quickly on safety. In the phase I/II study, there were some cases of lower extremity weakness. Do you believe that these have been mitigated via the use of the Trendelenburg flush in the phase III?
Yes, we believe it has been alleviated. I think the issue we were having, based on the MRIs and everything else, was pretty clear that there was local concentration of drug that was causing local chemical irritation. As long as patients are put in Trendelenburg and the drug driven cranially to allow for rapid mixing, we seem to have eliminated that. We feel pretty comfortable. We have patients now for five years on drug, up to five years, not having a problem. If it was really a chemical, it was a drug-related effect, it should have either accumulated or recurred, but I think it was very much an administration localized effect. I think we've managed it through these changes.
As you mentioned, you're also running the AURORA study in additional age groups and non-deletion genotypes. Can you talk about the regulatory filing strategy once you have this data in hand, and then also your level of confidence in that data set, given this is a more heterogeneous population?
The AURORA study will help broaden the label past deletion, as you mentioned. Includes missense, UPD, ICD-type patients. It also includes under age four and over age 18. The idea is here to cover the rest of them. We haven't put forth what our filing strategy was going to be. We'll have our phase III data, we'll have AURORA data, and we'll come up with our exact planning. The 4- 17 age group deletion, we think is the majority of the market, are in that range and so forth. There are going to be some adult with Angelman. We'll get to them. What we do for the filing right now is still left to be finally decided. We think it most important to get the major market, and the AURORA study will give us knowledge about the rest of it.
If that study is not stat sig but shows a trend, do you think that would be enough to get the broader label?
Well, I think that it depends a little bit on what we're seeing.
We are looking at MDRI in that study, by the way, in those endpoints, and it's pretty powerful. When we had even 12 or 13 patients before and after comparison MDRI, we saw strong statistical significance. It's pretty powerful with that tool, not just the one point. We are looking at Bayley in the really young ones, of course, but I think the methodology will make it a little better for us. MDRI, you might say, Well, it's not blinded. How does that hurt you? The value of MDRI in an unblinded situation or open label is that you don't count anything unless it's a really big change, right? If it's a little change, like a little biased, you're not sure, that won't count. You have to have a big change. The MDRI has a natural filter for small changes.
The only thing that counts are big changes. If they get better, they're getting better in a big way. I think you can be more confident that that is something real, right, as opposed to something that's just biased on how they're using the tool. Those are some features. I think it will be enough. I think in the past, it has been enough. In Crysvita, for example, when we originally filed that program, we had our trials were in the 5- 12, and there was this older patient trial. We didn't have anyone under age five. We submitted four patients with data under age five, four total. We had 13 who had been treated, but four with six months of data. We were able to extend the label with the exon 5-12 data and with supportive data just for patients for six months.
We were extending the label down to age one in that filing. It was open label supportive, extending the data to an age group. We've recently done that very thing and was successful, and it was really only four patients, but the four looked very good. They fit, the safety looked fine. The 13 got doses, drug worked the same. I do believe that it has to look good in those patients. If it's not distinctive, clear that it's working, then maybe it won't work. It will work if it works the same it has been working and does well on the under-fives, I think we'll be pretty confident that we can put that forth and allow the adaptation.
We specifically decided not to put patients through another randomized control trial, because to put all these subtypes, all these smaller subtypes into randomized trials for each type would have created a lot of cost and put a lot of burden. We put the burden on a group of kids from 4- 17 to prove cause and effect of this drug. For those kids, we just show the effect again. We don't have to prove the relationship because we've already proven that this is a cause and effect relationship. Does that make sense? Now we just need to prove that they have the effect and they have safety.
Got it. Maybe we'll pivot to the upcoming gene therapy launches, the first of which is in the next couple of months, potentially. Could you just talk to the early launch dynamics for both of those, both GSD1a and then Sanfilippo syndrome type A, including maybe pricing, and then also which patients you think will be early adopters here?
Yes. First of all, of course, assume that we get approved. I appreciate your optimism to talk about launch and launch dynamics. We always have to give the FDA its due that they have something to say first before we launch. We feel like review is going fine, we'll see. We'll get done. The launch dynamic for the two are probably similar. I think for Sanfilippo, it's probably the easiest one to talk about. Sanfilippo syndrome is a horrible neurodegenerative disorder, kids are losing brain cells every week that they wait. They have no other treatment, between the age of two and six, they lose a lot of their developmental function. By 10 years old, they're bedridden, they may stay in a bedridden state with G-tubes in and out of the hospital, non-responsive for five or 10 years.
It's absolute horrible situation. There is no parent in that situation that would not want to get treated as soon as possible, even if they're five or six, because we've had patients in that age range who may have lost speech already but are walking, feeding themselves, interacting with the family, who have stabilized. We've shown five patients that were older that stabilized their function, they continue to participate with their family and continue to self-feed, walk, and so forth. We think the therapy can work in that group, the younger the patient, the better the outcome, we think there'll be great urgency not to wait. The longer you wait, the less function you may have. We think that one dynamic will be urgent, once approved, there'll be people clamoring to get treated promptly. We're hearing about patients now.
They're very active social media. We're also getting inquiries from Europe about the inpatient treatment, and the Middle East as well. We think there's a lot of interest in the first-ever treatment for a neurodegenerative disease. I think it's a pretty easy call. The treatment is a relatively safe 3E13 infusion dose. It's not a very high dose. We don't have the kind of complex safety issues that we've seen with some of the others. The downside effects, I think, are modest compared to the potential benefit. That's the dynamic for Sanfilippo. For GSDIa, it's an urgent disease. It's not like your brain is going, you can survive on cornstarch treatment, keep yourself. It is not a great survival. It is a difficult life because you're taking starch every few hours, all day, all night, about a pound of starch a day.
I don't know if you can imagine eating a pound of starch. It makes you sick thinking about it, right? They feel sick doing it, by the way. They do not feel good. They're all induced type 2 diabetics. Imagine you're making yourself sick. You have to keep doing it because you're trying to avoid running low. You don't know when you're going to run low. It's like the opposite of diabetes. You're trying to keep your sugar up. You have the worst problem that the starch doesn't last very long. You have to keep taking it. If you don't wake up at night, you could end up with a very low glucose comatose or not wake up or die. That is a scary thing to live with. There is a substantial urgency.
When we were rolling this trial, we had people want to get in. We even had a problem in one country that they didn't review it very quickly. We enrolled before this country got started. We canceled enrollment. Parents, patients were so upset. They called the regulators, who called us. Demanded we open the trial to let their kids in. We couldn't do it. We'd spent all the slots. It tells you something. I've never had regulators call us, demand that we open a trial in their country. I've never had that happen. That's how intense the feeling was. I know there's urgency.
If you deal with a problem every single day, every few hours, you think about that. You have to know that if I don't drink this starch, I have a gun to my head, and it might go off if I forget to do it. Just think about that stress of feeling that way. You want to get off that if you can. We think GSD1 will do well. There's more patients with GSD1 in the prevalent population than in Sanfilippo, though they have similar maybe incidence. The prevalence is maybe 1,500 or more in the U.S. for GSD1a and for Sanfilippo, there's maybe 300 or 400 patients around in the U.S. The urgency is very high. We think both programs will do well. We feel good about the potential for those programs to achieve meaningful revenue for us.
They may not be like other programs. If approved, we think they will be successful gene therapy products. I think be supportive of the whole field of AAV gene therapy.
I guess maybe just in Sanfilippo, given there is no standard of care currently, how do you think about patient identification and whether there might be a headwind there trying to find those patients?
Well, I think the challenge is finding them early enough to have the best benefit. Still, let me get one little water. The truth is you really need newborn screening to capture patients early because you really want one mother age two. The methodology for newborn screening is the same methodology they already have approved for MPS I and MPS II, which are being implemented in newborn screening. MPS IIIA is comparable, and there's a method. It's not hard to implement. It could be implemented. It should be. The newborn screening control system, the secretary's committee, is not operating. We have to work through Secretary Kennedy to help bring forth the need to start newborn screening.
In the meantime, what we're looking is to help support and make sure that patients are being included in panels for developmental delay or other early signs that you would express in Sanfilippo. Some patients are getting diagnosed earlier because of that any kind of developmental delay is being evaluated. We need to get newborn screening done, make sure panels are being used for those early developmental delay patients, and then use our network connection among the inborn error doctors to make sure that we're finding as many as we can. I agree with you, it's one of the pieces. Right now, I've said we've found about 300-400 patients with Sanfilippo already found in the U.S., and we know the age ranges, et cetera, of the patients out there. Will all patients get treated? Later-stage patients, maybe not.
If they're very advanced, not responsive, parent may decide that it's not right. We don't know what the label will do. We've treated people up to age eight or nine in the trial, so that range could be readily expected in the label. Whether it'll go beyond that, we couldn't be sure. I think patients are still ambulating and is probably a range that makes sense. We'll have to negotiate what the labeling shows, but among the patients we've found, I think that will keep us busy enough with supply for the time being.
I guess on that note, could you just speak to the capacity at your Bedford manufacturing plant and whether you have capacity to meet the expected demand for the first maybe 12 months of launch?
Well, unfortunately or fortunately, because of the delay, we've been accumulating more commercial product. While the launch last year would've been more supply-constrained, where our supply is continuing to increase and been running continuously, the drug substance actually made it at a contract manufacturer in Ohio, and we do the fill finish at Bedford for the Sanfilippo program. That program is continuing to run continuously and build a supply, and we think we should have plenty of supply to handle the demand based on what we think we need for the first 12 months. For GSD1a, we're doing both drug subs and drug product. Again, we should have enough supply to handle the needs. Of course, there are more prevalent patients of that disease, but we have enough for what we expect to be the population that will get treated in that timeframe.
Maybe just on commercial efforts, I think you've mentioned before that you plan to leverage the existing Mepsevii and Dojolvi sales force. Given that these are gene therapy launches, could you just speak to the investments being made to manage the high-touch requirements of a gene therapy launch and the long-term patient monitoring that's required?
Well, there's several things we're doing. From the patient services side, we're creating a gene therapy guide. It's one of our patient services group that become point for helping coordinate all the different functions that are going on. We've created a separate group of the gene therapy treatment group, which is physician experts who are going to guide qualified treatment centers on both treatment and immune modulation or co-medication management so that they have an expert at hand at all times to help them. Our commercial team is also creating the qualified treatment centers who will be prospectively trained and ready to go at launch. Those things, there's a few extra people that are in the gene therapy treatment group. There's been some support for the MSLs required, and the patient services group will be set with some specialized people to help handle the unique needs.
There are a few places we need to bolster on top of the existing commercial group to manage specifically a gene therapy. The supply chain is also a bit different, it's following QA because we have flat pricing is expected. That means every group of weights is its own code and has to be packed. You actually have to create a supply chain that can handle creating a product for each patient's weight in real time.
Got it.
There's some people in the quality supply chain people that have to be able to do that. The idea is that we will have flat pricing for the gene therapy across the age groups, there is some logistics involved in doing that. Those are some places we have to do, but it's not a big lift. Not as big a lift as getting here.
Got it. Maybe just for the last few minutes, anything else you'd like to highlight from the pipeline? Also if you could just quickly touch on setrusumab, and whether you've completed those additional analyses of the phase III data and whether there's a path forward potentially in some subgroups or age ranges.
Well, we've done quite a bit of analysis. We talked about some of it at JP Morgan in the pediatric subgroup. We certainly see excellent improvement of bone density really in everyone. We saw improvements in pain, physical function, vertebral fractures in other areas. We've done a lot of analyses on that. I think we wanted to do it carefully and thoroughly, and there will be discussions with regulators around pathway. We'll probably put out information on what that pathway is, later in the year before we come out with Angelman data. Somewhere in that timeframe, we should have some understanding. We made a decision what we're doing. We'll know then. The goal is to try to find a way with existing patients and ongoing treatment to be able to come up with a path, to filing. We can always do another randomized trial.
We're trying to avoid that, which is a multi-year kind of a thing. We have a lot of data. We have two randomized trials and other patients ongoing treated. We hope we can work with regulators in coming up with an idea. We think the drug is working, but of course, we didn't hit the primary and secondary fracturing point, so that's a factor that will always be an issue for us to solve.
Anything else from the pipeline that you'd like to touch on?
Well, we've been incredibly productive. We have OTC phase III that was paused, that'll be ongoing. We have Wilson cohort forthcoming. Those are in play. We also have some things we've held back. If we hit our marks on our gene therapy approvals and launches, we have a program for creatine transporter deficiency we've had for a while. It's a pro drug of creatine, a really cool drug that we've made ourselves. It's actually a ring pro drug that has two clips, but we can deliver creatine to the brain very efficiently, and that's a very large market disease, maybe 30,000, maybe 50,000 patients. No treatment at all. Males and females both have affected symptoms that could come to the clinic. We're having some support from the patient group moving that forward. We also have a gene therapy for CDKL5 deficiency.
You might think another gene therapy, why? This is a neuro disease, intractable seizures. It used to be called atypical Rett syndrome. Like Rett, we expect it could have dramatic effects that are propelling the Rett programs forward. We think it's another good one. I just want to be clear, we are an engine of innovation. We have an incredible number of products we put, and those INDs have been sitting and waiting for the opportunity. We need to create the value, get the value, and drive forward and put some of these programs in play and show that we can be the most productive rare disease company in the business.
All right. With that, thank you so much for joining us. I guess we're just out of time here.