Good afternoon, everyone. My name is Akash Tiwari. I'm a pharma and biotech analyst here at Jefferies. We have the pleasure of hosting the Alkermes Management Team. Richard and Blair have joined us. Richard, why don't I give it to you for some brief introductory remarks, and then we'll get started.
Great. Thank you. Good to see you, Akash. Good to see everybody. I will make forward-looking statements and say what Blair will, too. So I always ask you to take a look at our risk factors that we describe in our periodic SEC filings, because they try to enumerate the risks that we face in our business. So with that said, it's an interesting moment if you compare where we were last year at this moment at this conference to where we are today, where I think that the process of simplifying our business over the last couple of years has really come to fruition. It's a very, very simple story now about Alkermes, a pure-play, profitable neuroscience company with a strong commercial portfolio, and what has emerged over the last year into a really exciting pipeline asset in narcolepsy.
But I think also what's clear right now is that this orexin biology is going to lead to other therapeutic opportunities as well. So we think of the orexin opportunity for ALKS 2680 as the core of the bullseye, where we're replacing a deficient neurotransmitter effectively with a small molecule. But the pharmacology is much more interesting and perhaps promiscuous than that. So we're really happy and proud of the work we've done. Three or four years ago, we were a much more complicated business with vestige manufacturing and royalty income, with an oncology business, with a nascent portfolio of drugs from the orexin. And I contrast that to where we are today, where I feel like the business is really easy to understand and hopefully poised for its next major leg of growth. So I'll let you take it from there.
Thanks so much, Richard. And I'm going to start off obviously. We'll talk a lot about the orexins. But I do think that right now, I do get a lot of questions on the base business valuation. I thought it was interesting on the Q3 call. You kind of soft-guided what EBITDA for 2025 looked like. And I think it came out lower than maybe some investors had expected. I think the Street was certainly modeling, let's say, more like $350 million-$400 million. You were guiding a lower amount. But part of that was increased spend because you wanted to invest in the orexin and IH, but also other moving parts of the business. So let's put it this way. A common question your team gets all the time is, what is the NPV of the base business of Alkermes ex the orexin?
When someone goes to you and says, you know, Richard, how can this be more than $1 billion, $2 billion if the EBITDA for next year is $200 million-$250 million? Why should the NPV of that base business maybe be different than what the Street's expecting? Why should we not be just simply putting a multiple on 2025 EBITDA?
First of all, it's become abundantly clear to me that people are not buying Alkermes stock because of the NPV of the base business. I'd say that 80% of the conversation we have with investors these days is around orexin and the potential. Because, as one investor said, the error bars around our commercial business are fairly narrow. But the error bars around the potential value of the orexin franchise are massive. And that's typically what animates biotech investors. With that said, I think the principal valuation difference that you would see in our long-range plan versus perhaps the more conservative sell-side ones is probably around VIVITROL. And VIVITROL, because we have a single ANDA filer in Teva with an entry date we've granted them in 2027, a number of investors model that as an LOE.
So you'll see a square wave decline in the anticipated revenues from VIVITROL in 2027. And that's quite different than how we model it for a couple of reasons. Number one, with a single entrant in a complicated market with a high cost of goods, there's no economic reason that we can see for someone to come in and destroy the market. And VIVITROL was so unique. I think it was launched in 2006. And it's in its, what, 18th year of launch. It's still growing. And it's still a very idiosyncratic product that helps a lot of patients. So we've transitioned the business almost entirely now to alcohol. And we're just touching the beginnings of that market of medical intervention for the treatment of alcohol disorder.
I'm hopeful that if we get other entrants into the space, after 18 years, there's only one ANDA filer, because the technology is incredibly difficult to make a substitutable VIVITROL. If there's only a single one, I think we model it more like a biosimilar with a much more protracted shoulder on that curve into the '30s. That alone, on an NPV basis, on a cash-on-cash basis, can make a big difference in the valuation.
Understood. Makes sense. And then maybe just on LYBALVI, obviously, I covered Bristol, COBENFY. I think it's a pretty remarkable clinical profile. It's funny, a lot of what people say about LYBALVI is what people say on COBENFY, where I can get olanzapine-like efficacy, but maybe not have the same amount of weight gain. Do you expect to see impact on your business with LYBALVI, given the COBENFY launch? Or is the market just so big that really both products can really exist in parallel with each other?
First of all, to make a comparative claim about relative efficacy versus olanzapine requires a study versus olanzapine. Talk is cheap, but I think it's all becoming real now as the drug enters the market and clinicians and patients begin to experience the real-world efficacy. I think put a pin in there, and we'll come back in a couple of years and see what the real-world efficacy actually is. To your point, the dynamics of the market are such that it's a switch market. And there are something on the order of 60,000 switches that occur in the atypical antipsychotic market every month. They're indicated only for schizophrenia right now. So they'll pick up only the switches that happen in the schizophrenia side of it. And sadly for these patients, the average length of therapy is only six months.
So I often say what comes in the top of the funnel in terms of potential patients, almost all those patients come out the bottom of the funnel. It just depends on how many different drugs they end up cycling through. It's a statement about the poor quality of care that these patients get. And as an aside, it's an argument in favor of long-acting injectables, by the way. So our view is that another branded medication in the schizophrenia space, encouraging payers in particular, that it's worth focusing on better efficacy for your patients, and it's worth paying for branded medications, that's a very good thing. We, from a purely competitive point of view, will need to make sure that we maintain share of voice.
That's the only thing that we worry about, is if there's another 300, 400 people out talking about a drug, we have to make sure that our drug is in that dialogue as well. So you heard us say we're going to expand our commercial footprint a bit into 2025, just to maintain that share of voice. But I think that the idea of better efficacy for patients with schizophrenia is a super good thing.
Understood. Now, you've taken kind of a grassroots approach with your commercial strategy on LYBALVI. We're like, look, we will engage with payer agreements when it makes sense for us, when the juice is worth the squeeze. We get the volume uptake without giving the price. You've made some payer agreements in 2024. As we think about 2025 and beyond and market expansion and agreements with more payers, A, how do you think gross to net will evolve as we get into next year? And then number two, how do you think volume will inflect as you get increased access here? Should we expect net-net on revenues an inflection for LYBALVI in 2025?
So I'll let Blair speak so I don't hog the mic. But we're really proud of the access strategy that we've employed from the launch. And it really is bearing out in 2024 as we roll into 2025. Blair, go ahead.
Yeah. And we spent a lot of time on the payer strategy, making sure that we were able to get contracts that are beneficial to the product and that will get us better access and get patients better access to drug. So if you look at the contracts we've made this year, we were able to do so while keeping our gross-to-net relatively flat. And that's kind of in the 29% range. As we move into next year, we expect to be somewhere in the mid-30s%, just by typical expansion of your gross-to-net that you have with these payers and additional contracts, as you said. I think as you look at the market evolving over the next little while, though, we've really contracted with most of the main payers in the retail space, two of the top three.
So I think we're in really good shape moving forward. And we'll start to see that volume picking up. In fact, we've already seen in the channels an inflection within those channels. But on a branded basis, I wouldn't expect any. I'd expect a steady growth of the overall market.
Continued steady growth. And just lastly on LYBALVI, before we get into the orexins, on IRA and Medicare Part D reform, I think people don't appreciate there's also a small biotech exemption. I think this is actually something ALKS has talked about. And so you don't necessarily have to pay that much on catastrophic coverage as, let's say, a large-cap pharma company. But talk to me about the impact you're expecting with the Part D reform for LYBALVI next year. How much of your business is Medicare Part D and that nuance of a small biotech exemption and how that benefits you?
Yeah. I think the small company exemption, what it does is it really allows us to phase into the program. So for companies like us, we didn't really have exposure to the donut hole prior to this setup, whereas big pharma did. So big pharma was sort of trading discounts. It's new for us. And so next year is the first year as we roll into the phase-in. So we'll have a 1% exposure to our Medicare Part D. And we expect this brand is going to do roughly a third Medicare, a third Medicaid, a third commercial. So it gives you an idea of the impact. So that phase-in will move forward over the next year, 1% next year, 2% the year after. And it just goes from there. So it's not a huge exposure to the company for the time being.
OK. Understood. Maybe getting on to the orexins, and certainly, for our investment thesis, that's by far the most important thing. Talk to me about some of the NT2 data that you showed. Because I think there is this perception, oh, orexins, they'll have urinary issues, they'll have blood pressure increases, but at the doses, there were a lot of doses that you showed where I think you had one case of insomnia. I mean, you had a fairly benign safety profile in NT2 patients. Richard, I think a lot of investors have struggled to translate the safety profile depending on what your underlying biology is and who your patient is. Sleep-deprived, healthy volunteer data compared to NT2 data, compared to NT1 data.
When you looked at your NT2 data, what gives you confidence that you're going to have a clinically acceptable profile with a strong efficacy signal in that population?
I would say that in my experience developing drugs for the last many years, I've never seen so much pseudoscience applied to limited data sets as in this orexin space. It's remarkable. So what gives you confidence? Actually, patient data gives you confidence. And what the others in the field are observing, namely, Takeda has done a great service by really being in the lead and exposing large numbers of patients over multi-week periods to orexin 2 receptor agonists. So categorically, we know on target, these drugs are for a new class of highly active, centrally active molecules. The overall safety profile is remarkably benign. So what do we know is on target? We know that insomnia is on target. And we know it diminishes over a multi-week period. We know that polyuria or urinary urgency is on target. And according to Takeda's data, diminishes over time.
Generally, everything we're talking about so far is mild or moderate. And so what gives you confidence is testing your drug in patients with the disease at the doses you expect to test. Now, that's why you do phase II, by the way. And the way we do phase II is by picking doses that we model, bracket the efficacious doses that we think will be useful in these patients, and running patients over a multi-week period in those lanes without titration, without adaptation, to see what the dose response profile looks from a safety and efficacy point in those lanes. And then in phase III, we'll have another 180 patients' worth of data or 240 patients' worth of data to model for our commercial and final doses.
But I'd say categorically, at this point, we're quite impressed with how our relative profile that we've seen in NT1, NT2, and in IH.
Understood. Now, when we think about your phase II trial design, it makes sense. Your point is like, look, I want to get fixed doses. I want to understand how much we can push efficacy. And I want to get a better sense of what our clinical profile is. I do think like MWT, it's Wall Street likes it because you can look at efficacy. But in the real-world setting, not everyone's going to walk around 40 minutes MWT all the time. I think you're going to find that kind of golden dose where you can have a good side effect profile and a good efficacy profile. In your phase II study, OK, you have fixed dose formulations. But I kind of asked your team, what about for phase III?
Do you think you can build in a titration protocol so that you can really get patients onto an optimal dose? And how would that design actually look like? Is that something your team's considering?
So it's actually better than that. Because what we're doing in phase II, Akash, is that at the end of the fixed-dose period of six weeks, baseline assessment time zero, six weeks later, MWT comparison, primary endpoint, run your stats on that. Add another seven weeks on top of that. All patients go to the middle dose. So in NT1, it'll be six. NT2, it'll be 14. They stay on that dose for two weeks. And then they get to choose their dose. So we'll actually get real-world data answering that very question. Is maxing MWT what all patients want to do or not? And we'll be along the way measuring patient-reported outcomes in terms of the quality of their experience. So we expect by the time we make the call for phase III and commercial doses, we'll have a lot of information.
I do not right now think that we need to introduce a titration protocol into the phase III. I think we're better off probably just knowing which doses are the modal doses for people.
OK. Understood. Now, I'm going to actually go to a tangent that a lot of people maybe don't talk about, ADHD. And it's funny. You had an R&D day. And everyone's like, they don't talk about expanding orexins. And then you did. And then now you have to spend on it. People are like, oh, well, now Alkermes is spending money. It's kind of a Catch-22. But it is really interesting. If you look at Takeda's data at sleep on cognition in narcoleptic patients, they had a very strong signal in cognitive scores. When you look at the drugs that work for wakefulness promotion in narcolepsy, they're all repurposed antidepressants, right? Reboxetine, these are all basically old antidepressants. And they will make you feel kind of like a zombie long term. It's just not a necessarily pleasant patient experience.
What did you think about the cognition benefit that Takeda showed at Sleep? And then number two, why is ADHD a potential indication for an orexin? Why make that announcement at your R&D day a few months ago?
So I'll start and let Blair give you some color on it as well. You start with the biology. So if you look at this bundle of orexin producing neurons and where they project into the brain, you can draw some certain hypotheses with respect to what they might have in terms of clinical benefits. And this idea of vigilance is pretty self-evident. Norepinephrine levels in the prefrontal cortex and other parts of the brain. And you say, OK, we can test that preclinically using a variety of techniques that we describe in that R&D day, both behavioral as well as microdialysis and EEG. You can see what you're doing in the brain. This stands to reason. I was interested that Takeda at the sleep conference focused their cognition on vigilance as the scale they were using, which makes sense. It makes sense. So we had a hypothesis.
Then we tested it preclinically. Blair, you take us back.
Yeah. I think as you look at the fact that orexin works on the wakefulness circuits within the brain, I think we had a very clear hypothesis that there are a number of different diseases that have that as a disorder that could benefit from this, obviously ADHD being one of them. And as we were looking at assets, we really were trying to compare to some of the standard of care that was out there, how could this work? And we felt that there'd be a very similar effect both between attention as well as what you see on mood. By upregulating alertness, what you're doing is that patient, there's more connections. They're feeling more engaged with the outer world. That improves their mood. That improves their cognition. That improves a whole host of other things. And the question was, how does that quantify relative to existing therapies?
To Rich's point, our focus on using microdialysis, behavioral models allowed us to really benchmark. We were really surprised by the effect that we saw across a number of different disease states. What we showed at our investor day was not necessarily the indications we're going at directly, but different domains that we can interrogate. We may find indications where these things intersect, where these intersect with sleep, where these intersect with other things. Frankly, we were hoping to limit down a little bit some of the areas that we could take the drug to to start to prioritize. We've been finding just so many new areas of research with this new biology. There's really been no other tool to test this in the brain up till now. Stimulants just don't get you there. They have a very different effect in the brain.
It's going to be a really exciting area for us as we move forward.
I just want to underline one point, Akash. I know you know this. But certain investors may not. So there's a pretty simple analysis one could do to say, if I improve someone's wakefulness, so for example, you fly to England for an investor conference. You don't sleep very well. You're not thinking quite as fast. You're a little grumpier. Your resiliency is a little down. If you just have a better quality of wakefulness driven by that wakefulness circuit, you're probably going to affect positively certain scores that measure attention as well as wakefulness. But it doesn't stop there. The point about the circuitry in the brain is that orexin is actually the master regulator of direct effects on these other domains. So it's a way to win twice. So I think to Blair's point, the question is picking our spots.
We don't have to pick our spots right now. Because the necessary prerequisite is putting these additional orexin compounds into the clinic, running the SAD and MAD studies, characterizing the doses. Then we'll let you know where we're going to go specifically.
Understood. Now, maybe talk about the selectivity of 2680. I think there's been some debate. Could some of the urinary side effects be OX1-mediated side effects? When I look at your compound, you're about 5,000-fold selective. That's very similar to the TAK I.V. compound. There was no evidence, at least from what we saw from Takeda, that they had any on-target OX1-mediated side effects. But I'd love to get your take. How selective is 2680? Are you seeing any evidence of OX1 activation? And why do you feel like you have the right selectivity profile?
We're highly selective. We're somewhere between 10,000- 5,000 fold selectivity. I don't think we feel that it's OX1-mediated. I think if you look at the polyuria that you see in all of the trials, these are more reflective of the nervous system being upregulated. It's like when you wake up in the morning, your body's ramping up. You stand up, your blood pressure goes up to stand up. You also have this urge to urinate. That's the same system that you're regulating as you work with these molecules. And so I think that also is supported by some of the longer-term data that you saw with Takeda, the multi-dose data, where you see that start to attenuate over time as that starts to tolerate out. I don't think it's a direct kidney impact.
Understood. Now, in NT2, I think the one point that our team is trying to understand, if you look at the TAK-861 data, obviously it was under dose. They can't dose much higher because of concerns on liver toxicity. But I think it was the seven-mg dose. They did at an earlier time point kind of have a 9-minute placebo-adjusted delta on MWT. As you got to a longer time point, that benefit started to attenuate. And I think there is this question about maybe in NT1, the biological hypothesis totally makes sense. But Akash, maybe I'm worried in NT2 and in IH, where patients have more normalized levels of hypocretin at baseline, that an orexin wouldn't have a durable effect. So A, what have you heard from Takeda in terms of their backup drugs and whether they're pursuing those in larger indications?
And then number two, what gives you confidence that the signal here will be durable in these broader populations?
I wouldn't overinterpret the NT2 data for TAK-861. They're not at the dose. So whatever wobbles you see there, I don't think you can draw any inferences about what it means categorically for the class. Remember, in healthy volunteers at escalating doses, we saw quantitative changes in EEG bands associated with wakefulness. So we've always believed that nature has given us the knockout model in NT1. If you take away orexin from the system, what happens? You have narcolepsy. So replacing NT1 makes a lot of sense. NT2, you might think the bell curve of distribution of orexin tone is probably much flatter, much broader. So you'll have NT2 patients that look a lot clinically like an NT1 patient. And you have other ones that don't. So it stands to reason that doses may need to vary across that. IH, completely different phenomenon, right?
You might have a background of fairly normal orexin tone. So supraphysiologic levels of orexin seem to drive additional wakefulness in those patients as well. So in our view, it's always been an argument in favor of needing dosing flexibility. And so that's what we've seen so far. And phase II will bear that out.
OK. Understood. Now, I think there are obviously. It's a competitive space. There's other orexins in development, including from large-cap pharma. Can you talk about what you've seen develop in the space over the last three months or even in the last couple of days that makes you feel more confident that you have a pole position here?
Yeah. I think big pharma is a very interesting space. We know a lot of them are watching very carefully. And some of the notable ones that have jumped in, for example, Vertex has some patents in the space. We know Merck has been in the space for a number of months now or years now. I think we saw recently that Merck might be pausing their program or stopping their program, at least in the narcolepsy study. Still more to learn on that one. So really, the key players right now in the space are Takeda, ourselves. Centessa is developing their molecule. These are really the three marquee molecules to look at in the space. I think Jazz has a molecule they're developing but ran into some troubles. And it remains to be seen where they're going.
So when you look at this sort of three-horse race, I think Takeda, to Rich's point, has been fantastic about developing drugs and moving forward. They're dedicated to the space. Their lead asset is restricted to the NT1 population as a twice-a-day. We're the only other one that has clinical data in patients across all three patient types where we've seen positive results. So we think we differentiate quite clearly, at least on the data that we've seen to date. And then obviously, Centessa, we'll see where that comes in the coming months or year.
Understood. Richard, I'll ask you. I mean, this is getting to be a large-cap pharma type development program. You're going after multiple indications, multiple orphan diseases. When do you, as a management team and as a board, have a discussion about who's the right person to really develop this? Is it after phase II data when you've shown real proof of concept, durability of effect? Is it after phase III? I mean, do you think maybe putting this out to a strategic might actually be the best way to maximize value here?
I think you could proceed stepwise. I don't think you have to make the call right now. Because if the narcolepsy opportunity evolves the way that we hope it could, that is a gigantic commercial opportunity. We talk about an orphan indication. But there's 100,000 patients that are treated or diagnosed in something like 70,000 or 80,000 that are treated in the U.S. at a price point with the oxybates that is extremely high. So if you just model success in NT1 alone, that's a multi-billion-dollar opportunity superimposed on our valuation. That's a major valuation gapper for us. Layer on then NT2 and IH and see where you end up. If that all eventuates, we have the capacity to keep developing this molecule. So we're not waiting for anybody right now.
All the things that needed to be done to expand the opportunity for the orexin, we can do right now, which is, as I said, nominating additional compounds with slightly different pharmacology, putting them into the clinic, characterizing them, and then giving yourself a quiver of different arrows to use to go after different indications, and so we're not rate-limited by anybody right now. Now, if you end up with 10 indications in major depressive disorder and daytime sleepiness for jet lag or something like that, those are huge indications, primary care indications. We're a long way, way from that, and the analogy that I draw, and I don't mean to be hyperbolic, is to the GLP-1s. They start in diabetes, serious clinical indication. In so doing, you develop a baseline denominator of safety and efficacy in human beings over a few-year period.
Then you can begin to say, OK, the need to lose weight is more universal across other diseases. And you begin to expand the penumbra of people who get exposed to the medication. Wakefulness is a big deal in our society. So I think starting with narcolepsy and preparing for the future is the right way to do it.
Understood. So orexins for sell-side analysts, that's the trial I need you to run.
We'll use the $500,000 now.
Guys, thanks so much. I really do appreciate it. Thanks, everyone, for your time.
All right. Thanks, Akash.