All right. Thanks, everybody. Happy to be moderating this fireside chat with Blair Jackson, CEO of Alkermes. I'm sure folks know the story decently well, but maybe we can start by just having Blair recap some of the key updates last week from the quarter, and then also looking ahead, what are the next events we should be expecting from this orexin program you have that people seem to be interested in?
Yeah. All right. Well, thank you so much. Look, I think it has been a really interesting year for the company. We have had a lot of things happen over the course of the year. We acquired Avadel. We had some positive phase III data on LUMRYZ in IH. We have moved our orexin programs into phase III. And then, just a few weeks ago, we announced some really interesting data in ADHD with ALKS 7290. It has just been a whirlwind of a year, and we are expecting that moving forward.
As we move into the next part of the cycle, I think we have something like six programs that are either medium or late stage reporting out over the next 12 months. It is in narcolepsy, it is in ADHD, it is in fatigue. We are looking to file our LUMRYZ application for IH. So there is just so much going on at the company.
It is just a really exciting time overall. And it is supported by a commercial business that is going so very well. We have generated over $1.8 billion of revenue this year. We are highly profitable. So, I think for us, as you think of the next little while, it is about execution. It is about pulling all these things over the line, making sure we have high-quality data, high performance from our teams, and so we are able to execute on the story that we have put in front of us.
Great. The next big readout here is coming in idiopathic hypersomnia. In other conversations we have had, you have talked about how you feel like in NT2, the data you generated might be like sort of a reasonable benchmark for what to expect. Maybe just sort of expound upon your thinking there and your expectations for this mechanism.
Yeah. I think it's interesting. Obviously, there hasn't been a lot of studies with orexin agonists in the NT2 or the IH populations. We're sort of leading the way in that category. I think with the NT2 study, we saw a really strong response, both on ESS, on MWT, multiple doses, and we see the IH population being very, very similar.
One of the things that drives that is there really is a little bit of a gray line in diagnosis between NT2 and IH. Patients oftentimes will get an IH designation because of a sleep test. If they go test again later, they find they have an NT2 designation. I think there's a lot of overlap between these populations, with the exception of one group. One group that's a little different in IH is the long sleepers.
About 25% of the IH population are long sleepers. So they have a lot of naps, but they're not very refreshing, so they suffer from excessive daytime sleepiness. Our expectation, though, is that this mechanism of orexin agonism can still drive wakefulness in those patients, because we see that across all the populations we've studied in.
Normal healthy volunteers in the SAD/MAD study, whether it's NT2 population, people who have intact orexin, if you give them super therapeutic levels of orexin agonism, you're able to drive that wakefulness. So, expectation-wise, the way we designed the study is similar power-wise, et cetera, to what we saw with NT2.
Yeah. Okay. Makes sense. I guess in this population, ESS is the more important endpoint, right? What do we know about that scale and sort of what the benchmark is for something that is clinically significant there on the benefit side?
Yeah. It's a good question. None of these scales are perfect, and ESS is one of those subjective scales used by patients. It's a look back of a week, and you have a number of different questions that they look at, which is sort of questions like, "Do you feel like you're going to fall asleep while you're sitting at a traffic light?
Do you feel like you're going to fall asleep in a restful situation or when nothing's going on?" It's really measuring how tired they're feeling and how that's impacting their life. So, for this scale, I think typically a 4-point change is considered to be clinically meaningful. I think 6 points is considered moderate. What's important, though, is that this is the scale that physicians use commercially.
When you go and you call on a doctor or the patient sees the doctor, they're assessing their performance on that ESS scale. I think what we feel pretty good about is we've seen an ability to move that scale across a number of different patient types.
In our NT2 study, which as I said before, I think is a little like this IH group, we saw at our highest dose that we were able to normalize patients from moderate to severe to normal. About 70% of them normalized on our once-daily dose. It does show that this scale's movable, that the mechanism's working in this type of patient population. We're excited to see what we can do in the IH group.
How important is it that you see a p-value less than 0.05 in this readout?
Well, it's an interesting question. Obviously, we designed the study and expect to hit statistics. I think what people need to recognize, though, is this is a phase II study, and we are going to be doing some learning here. To the extent we see anything unexpected, for example, perhaps the long sleepers perform differently than we had anticipated, if we're able to still design and power around that for phase III, we'll take those learnings and we'll move forward with phase III.
I think you don't want to be a slave to the statistics, because it kind of could mislead you. This patient group is a group that really doesn't have a lot of drugs available to them, where there's a huge unmet need there. The only branded drug in the space is, right now, XYWAV. I think being an orexin agonist could add a lot to that treatment population.
Yep. Okay. You have two phase III reading out next year in NT1 and NT2. Maybe one, do you want to talk about just how those studies are enrolling? Then two, in terms of your internal expectations for what these data may deliver on, are the prior phase II studies the right way for us to think about a benchmark here?
Yeah. So we kicked off our phase III program in the middle of the summer last year, and our expectation is it is going to take about a year to enroll these studies. So, thinking about middle of summer next year or so being done with them. Enrollment is going really, really well. I mean, we are through the summer shutdowns in Europe, so the European sites are up and running and enrolling patients.
That is a very important group for us because, one, it provides a large number of patients, but also the quality of data you get from these European sites is actually really, really good. They tend to have a better handle on the patient types and some of the testing methodology.
For the phase III programs, when you look at the overall size of the program and you look at what we are doing from a site perspective, it is really not that much larger than what we did in the phase II. So, we do not expect to see a lot of erosion of signal or a lot of variability because of those things that you see in some of these broader psychiatric indications. So I think the phase I and the phase II levels that we saw, we would expect that it would be something similar to that.
Yep. Okay. Maybe taking a step back in this space more broadly, as you think about your lead molecule, you obviously have a backup and multiple backups. You have Takeda there, you have Lilly. How should we think about the way the orexin space evolves and where the most likely points of clinical differentiation are going to be? I guess what would you say is more important at this point to success in this category long term? Is it clinical differentiation or is it being first in given indications?
Yeah. It's a great question, and I think it differs by different indications. I think when you look at the ability to treat the patient populations in diseases of hypersomnolence, it really is all about efficacy. It's all about being able to provide a path to a normal life for these patients.
What comes with that, though, is an ability to meet the lifestyle needs, an ability to meet the dosing where they are, making sure you can dose high enough for those that need highest dosing, providing optionality around dosing, once daily split dosing. I think that's all going to be very, very important, and I think one of the areas that is indicative of that is if you look at ADHD treatment, for example, it's kind of very similar analogy-wise.
You have short-acting and long-acting versions of things to meet their lifestyle so that they can operate normally in their life. I think as you get into some of these bigger therapeutic categories, depression and things like that, first mover advantage is going to be very important, and I think being able to get there with a drug that meets the TPP you're looking for, meets those needs, is going to be very, very important as well.
I think first mover advantage in those spaces is important. I think over the next number of years, we're going to see this space change a lot because Lilly, we know, is putting a lot of money behind their Centessa acquisition. I think there's an R&D day coming later this year, and hopefully we'll learn a little more there.
But with us, Takeda, Lilly, and we know a number of others are looking at the space, I think you're going to see a lot of investment in clinical trials in areas of neuroscience, in sleep disorders, and in maybe other areas like metabolism and I&I. So I think it's going to be a really exciting few years.
Yep. Okay, great. What's your expectation for how this Takeda launch goes in NT1? Do you feel like there's a pent-up demand from patients there who are identified, discontinued oxybate, and are ready to go?
Yeah, I think we've heard clearly from physicians, as we go to the sleep conferences and talk to the KOLs and the treating physicians, that there are patients who are waiting for this new mechanism to come. We estimate there's something like 100,000 patients who've tried oxybates and it just didn't work for them, and they're sitting on other therapies that maybe aren't giving them the best efficacy.
I think our expectation is that Takeda's launch is going to go quite well. I think that's great for the space. It'll show the promise of what orexin agonists can do, and then our hope is as we come to market, people will recognize that there's some additional benefits we can provide as well, and we'll grow the market from there.
Yep. Okay, great. You also have ADHD reading out next year. Again, I know we had dinner yesterday, so forgive me for the redundancy, but I think one of the conversations in the investor community is sort of, how do we interpret this phase I-B data?
You're seeing this really promising signal for the two active arms, but I think the lack of disclosure on placebo has led people to infer that there either was some sort of big placebo effect or something that confounded the interpretation. As you think about this dataset, I know the initial goal of this study was really safety, but where do you sort of- How would you characterize this as it relates to real proof of concept or not?
Yeah. It's a good question, because I think it's important to recognize how we staggered these studies. The first study we designed actually was the phase II dose-ranging study that we're running right now. That's a large, randomized controlled study, multiple doses, and that's what study is going to be able to allow you to understand the effect size.
How big are you relative to, say, the stimulants or other drugs that are in the class? What we did is we added this parallel I-B study for safety purposes and for, one, for both proof of mechanism and dosing purposes. What I mean by that is we wanted to demonstrate the therapeutic window that we're able to generate with these assets, and we wanted to understand whether or not we were in the right dose range.
This is the first time orexin agonism has been used in an ADHD population in humans, and we wanted to be able to use a study to make sure, do we need to add another cohort? Do you need to add another cohort on the bottom end or the upper end? What we found is we were in the right dose range. I think to your question about how to view these results is, I think you need to be cautious on them because it is an inpatient two-week study, and it was not placebo-controlled.
I think as you look at the program, our goal here was to try to understand what metrics an orexin agonist can measure as it relates to ADHD. Is it going to work in hyperactivity, in impulsivity, in inactivity? All these sort of things that you are looking to try to treat, cognition, working memory, et cetera. What was very telling for us is that we saw a consistent response across the different measures, moving in the same direction, having a fairly robust response.
We were seeing the patient-reported outcomes report the same way as the physician-reported outcomes. We saw a CGI-S change of 2 points in a two-week period, which is very profound. I think when you look at the EEGs, which is an objective measure, it was all consistent with activation of the circuitry. We feel from a proof of mechanism, it was a very good study.
I think for us as study designers in the phase II that we are running, it gave us confidence that we are in the right range with the ongoing study, and then we should all wait for the placebo control study. That will tell us the effect size, and that is what it is designed to do.
Yep. As you think about the target product profile here, does this need to be as effective as a stimulant for this to ultimately kind of attain a big commercial market?
It is a great question, because I think, what is interesting when we talk to a lot of people is everybody has family members or knows someone who has an ADHD diagnosis, and you see it everywhere. It could be your children, it could be someone else.
I think it is well-known that stimulants are very efficacious, but what they carry with them is a lot of side effect burden. You see with children, a lack of weight gain. They end up going on drugs so they can gain weight. You see irritability, you see substance use in adults. There is more and more adults being diagnosed with ADHD. You see cardiovascular liability as well.
I think what people have been trying to do over the last decade, the real holy grail has been, can I develop a drug that has stimulant-like efficacy but has a much better safety profile? What we've seen over and over and over again is that the drugs are falling way short. One is they're unable to hit the efficacy level, but they're also bringing with them also a side effect profile that's not as attractive as you would like.
To your question directly, I think there's a lot of ways to win here in that, you could be between stimulants and non-stimulants from an effect size, but bring with it a breadth of response. Maybe we add cognition into the mix because we're an orexin agonist. Maybe we have a safety profile that is better than the others in the class. Obviously, our TPP, or at least our preferred TPP, has stimulant-like efficacy with a clean safety profile and a broader response. We'll see where we land.
Yep. Okay. That data is second half next year?
That'd be second half next year, yeah.
Okay. The molecule you've advanced into ADHD specifically, is there anything about that molecule that makes it better suited for ADHD?
I mean, all the molecules that we're moving forward are very different. alixorexton is different from ALKS 7290, which is different than ALKS 4510. What we've tried to do is develop them as best we can for the populations we're going into. As you know, we're going into a very broad population with ALKS 7290, both in ADHD and depression. So we wanted to make sure we had the widest therapeutic window possible. Early data suggests that we're in that range.
Yep. Okay. Makes sense. The other sort of indication or family of indications you've talked a lot about is fatigue. Maybe, one, talk about what gets you excited there, but then, two, what still needs to be elucidated around the actual right endpoint or the sort of optimal regulatory path?
Yeah. Fatigue's just a fascinating area for us to research, and in many ways is the reason we're in the orexin space to begin with. We've heard very clearly through the years that MS patients, for example, are heavily burdened by severe fatigue, and it can be more than 50% of the population. You also see that with Parkinson's patients as well.
Even when we were making drugs for MS, and remember, we developed VUMERITY a number of years ago, and most of the physicians wanted to talk about, "Are you doing anything in this space to help our patients?" Early on, we wanted to do a lot of preclinical work, a lot of testing to see, could this mechanism work in a fatigue model? It makes intuitive sense that it should. It drives the wakefulness and attention circuitries of the brain.
The challenge, though, is people don't really know what drives fatigue. Some of the models associated preclinically are loosely translatable. We ended up putting the PROMIS Fatigue Scale into all of our narcolepsy programs and in our IH program. What we saw from the first study is we were able to take moderate-severe patients, normalize them on the PROMIS Fatigue Scale.
That's a profound outcome, and I think that's something that we found particularly enticing as we started to think about where could you take this molecule. I think the question we had is, what areas of fatigue could benefit from this? Similar to like we talked with IH, our belief here is that super therapeutic levels can deal with fatigue, but what we needed to do is understand fatigue a little better.
We included two populations, the MS population and the Parkinson's population, and we did that because they experience fatigue in slightly different ways. MS fatigue patients tend to feel central fatigue. They feel exhausted mentally. They feel just a different feeling than how Parkinson's patients tend to describe it. They tend to describe it as muscular fatigue and these bone-tired muscles. What we wanted to do is say, "Are these really the same disease? Are they tractable by an orexin agonist? What is the endpoint that we want to use to test this?
Is it PROMIS Fatigue or is it more of a disease-specific scale?" That's the purpose of this phase II study, because that will allow us to then meet with the agency, plan our path forward, and then we can start to talk about moving it into broader fatigue areas, chronic fatigue, cancer treatment fatigue, some of these other areas that are much bigger. What we first need to do is start to get the nomenclature and the tools in place for the regulatory path.
In these initial studies, are you just looking at a whole host of different endpoints?
We are actually looking at a number of different tools. For example, we have the PROMIS Fatigue Scale, we have an MS Fatigue Scale, we have a PD Fatigue Scale. What we want to see is, as a tool themselves, are they sensitive enough and are they discriminating enough to use as a primary endpoint? Do they need to be combined? Would a combined tool or a composite tool make some sense? These are all the discussions we will have with the agency after we have seen the data.
Great. Do you want to briefly talk about the thesis in MDD, and specifically, I think, there is a logical population, right, where there are these comorbid issues with sleep or daytime sleepiness or stuff like that. But how homogeneous of a population is that, really, and how do you actually define the right cutoffs for that population?
No, it is an interesting question. I think we knew early on that we wanted to look at depression as an area for orexin agonism. It makes a lot of sense. If you look at a lot of the characteristics of depression, there is a lack of engagement, there is excessive sleeping, but it is not necessarily the whole population. As you know from the MDD trials, it is very, very heterogeneous.
We did a lot of work looking at neurotransmitter release and behavioral models in rodents, really to see, do you see an incremental benefit or even a synergistic benefit by using orexin agonism adjunctively with SSRIs and SNRIs? And we saw a very, very clear response in serotonin release in certain parts of the brain.
It gave us confidence to move forward, but one of the things that we are leery of always is the placebo effect in these studies running these large trials. What we are actually doing is we are looking at two populations in our phase II study. We are going to look at an MDD group, a normal group that you would normally see, but we are also going to look at a group who has excessive sleepiness.
Right now, our understanding is about 50% of the MDD population experience excessive sleepiness. That might be an interesting group for us to go after, and it may enrich our program and allow us for better statistics. It may be an opportunity with the agency and moving forward quicker. We are going to learn a lot from this study, and the hope is that it gives us a unique path to market, because look, orexin agonism is sort of untapped in a number of different areas, and MDD really lights up when we look at one of the core areas we want to try.
Great. Maybe a couple questions on the commercial business before we wrap up. As it relates to LUMRYZ, when you think about the outlook over the next kind of three to four years, I think there is really two sort of interesting questions on the risk side, and then there is some obviously opportunity to the upside. On the risk side, I mean, people ask you, I am sure, about generic oxybates and also orexins, right? To what degree do orexins eat into the oxybate-treated population? How did you get comfortable with those two issues when you purchased Avadel?
Well, two very different questions. I think when you look at the multi-source generics, I think one of the things to keep in mind is that they are generic to XYREM. They are not substitutable for LUMRYZ. Based on our experience working in a number of different areas, we didn't feel that there would be a class erosion as part of that.
Like a step edit or anything like that?
We didn't think that that's what was going to happen. In fact, that's what we've seen in practice so far. I think that's been really good. I think when you look at the acquisition of Avadel and LUMRYZ, we did a number of different things looking at, are you going to see erosion of the oxybates due to the orexins coming into the marketplace?
Our feeling was that while we would see some impact as orexins move into the market, especially on new patient starts, what we found from talking to physicians and doctors who use and are on oxybates, there is a level of benefit that they get from those that is quite distinct and unique, and a lot of them don't want to come off their oxybate.
What they would rather do is add an orexin to their regimen, treat with orexin in the morning and an oxybate in the evening. I think that's a really interesting paradigm, and that's something we'll be generating some data on as we move through our registration program with alixorexton, because I think if we can show benefits and what that brings, it'll provide some instructions to doctors and patients and payers to support that treatment methodology.
Maybe just more broadly on the base business, what do you think is maybe either most missing or said by the investor community or sort of like the most meaningful source of upside that you think gets overlooked?
I think there's a few areas in the base business where folks are, maybe they take a drastic point. I think in some cases, you'll see analysts and others kill the VIVITROL model right away. They assume it's a small molecule generic, and we have one ANDA that's in play. We've talked a lot publicly about how difficult it is to manufacture and work around VIVITROL. We know that Teva's run into some manufacturing challenges with that, so that's consistent with what we've said before.
We see a long tail with VIVITROL, and I don't think a lot of investors have recognized that. I think secondly, LYBALVI. I think I see a lot of models that people create where they have LYBALVI disappearing in a few years, which is just inconsistent with the Orange Book patents that we have related to LYBALVI, which go into the late 2030s. I think those two things kind of drive a lot of misunderstanding of the base business.
Yep. Okay. To wrap things up, Blair, anything else you want to hit on here?
No, I think we talked about the key things. I mean, for us, as I said at the outset, it is about execution. I mean, we have a lot of things that we have put in play over the last little while. We are going to learn a lot from the launch at Takeda in the next few weeks. I think between this, I think this whole space is going to evolve over the next few years, and I think you should watch it closely.
All right. Will do. Thanks very much.
Thank you.
Best of luck. Appreciate it.
Appreciate it.