For those who don't know me, my name's Akash Tewari. I am a pharma and biotech analyst here at Jefferies, and I have the pleasure of hosting Richard Pops, CEO of Alkermes. As a bit of background, Alkermes is a topic for us. We have a lot of interest both in the base business, but also in their Orexin pipeline, in narcolepsy and other sleep disorders. Richard, maybe I'll hand it to you for some introductory remarks, and then we'll get started.
Good. Is this mic on? Oh, yeah, it is on.
Yeah.
So good to see you, Akash. Thank you. And I'd, Akash is being modest because I think he has probably been the most intellectual leader in the Orexin space for the last year or two. So I think, I'm looking forward to the conversation because I think you probably have more background than most on the relevance of the most recent data that we put forward. This is a really interesting moment for Alkermes because, as a matter of fact, this week will complete what has been about a year-long project of spinning out our oncology business in preparation for 2024, where we'll guide to Alkermes as a standalone, profitable, growing CNS enterprise with a really interesting pipeline asset in 2680. 2023 for us was a complicated year. Going into it, we knew we had some housekeeping to do.
We had an outstanding litigation with Johnson & Johnson, and which was worth circa $1 billion, which we resolved favorably in our favor in the middle of the year. We had an outstanding litigation on VIVITROL and a litigation with Teva, so people didn't really have a clear sense of how to model that product in our portfolio in 2023, 2024, 2025, 2026. We settled that favorably in September, so that's sorted. We hadn't even been in a human being this time last year, just beginning in human beings with our Orexin program, and we wanted to advance that as fast as we could in 2023. We wanted to complete the spin of this oncology business, which was a big program. So we also were launching a drug called LYBALVI.
We had just had our first year of sales of $97 million. We wanted to roughly double that this year. And we felt like if we could do all those things, and that was a full menu, as we moved into 2024, we would be able to guide to a standalone, clean CNS company with a launching pipeline, a launching product in LYBALVI and a very interesting and maturing pipeline. So that's exactly what I think we've been able to pull off, and we're looking forward to 2024.
Great! Well, thanks so much for those intro remarks. I really do appreciate it. So I wanted to start off with, you know, obviously, over the last few years, there was also the long-term, you know, midterm guidance that you gave out, you know, where you were initially, it was 25% margins by 2025. Now, that's been moved up, even earlier, given some of the, you know, positive surprises you've had on your base business. As you think about next year and the ability to, let's say, hit those 25% margins, ex, you know, the, the J&J settlement, it seems like you have some levers here, right? Whether it's DTC spend for LYBALVI, whether it's, you know, other, other areas where you might be able to dial up or dial down on R&D spend.
On a general basis, what is your confidence on the ability to hit 25% operating margins, ex the J&J settlement, next year for Alkermes as a standalone business?
First of all, let's talk about why you do it.
Yeah.
It's interesting as you get one of these companies to a point where you actually have recurring revenues of a substantial level, not just sporadic from licensing or one-time milestone payment, but we have an ongoing economic enterprise. It's funny, you spend a lot of years just trying to do science and make drugs, and pretty soon then you realize you have to run this business as an efficient operating business with as an economic enterprise. So establishing profitability targets, we saw as almost like corporate hygiene to say, "You know what? We love spending money on R&D, but you have to do it to a limit." So setting up these targets actually forces more efficient capital allocation within the business.
Even on the R&D side it forces you to prioritize R&D spend. So my confidence in hitting those profitability targets is extremely high because we will modulate the expenses to hit the profitability targets. Having said that, what are the priorities? We want to fully fund LYBALVI launch because LYBALVI deserves, you know, you want to bigger that launch. That's why DTC, you know, we won't dial it back in pursuit of hitting a margin target because that's a long-term driver for the company. And Orexin, we're gonna fully fund that program 'cause we've passed the stage gate now, and we'll talk about this at more length. For us internally, in terms of allocating capital to Orexin, we've crossed that stage gate that says we should floor it now.
Yeah.
We should go as fast as possible.
Understood. So fair to say, despite your commitment to spending on Orexin, despite your commitment to DTC for LYBALVI, you remain extremely confident on the 25%?
Correct. And to the point where we would say we have additional capacity in the P&L to expand R&D as necessary, as opportunities present themselves.
Understood. So can you talk to us about, you know, like, it sounds like maybe DTC spend for LYBALVI on an annualized basis would have been maybe in the $70 million range.
Annualized would be more like 100.
Hundred? Okay.
Yeah.
So as you think about the levers you have with your P& L once the Mural, you know, is spun out, can you talk to us about maybe some, you know, what, what, where is the operating leverage when you look at the standalone Alkermes P& L, that maybe isn't getting appreciated right now a lot by investors?
Well, what's fascinating, if you look at street models for 2024 and beyond, many of them have not even incorporated the spin yet. They haven't, they haven't rationalized the expenses. So I think when we guide in February for the year, it'll be a really nice moment because it'll should consolidate people's perception from a financial perspective, what the business looks like, which I think is, is quite optimistic for us, given the, the way that LYBALVI is going with the resolution of VIVITROL now and with the consistency and growth in, in ARISTADA. So we have a, we have a really nice flywheel of. and the transition has been profound. Recall that if you look at our P&L 10 years ago, it was all royalty income and manufacturing income from partnered products.
So we've been able to pass that baton now to the proprietary products business going forward. So the way, the way I think about DTC is not particularly novel because there's so much experience in this category, i.e., the atypical antipsychotic category, and in bipolar and schizophrenia. DTC is a reliable way of driving revenue over time. It's essentially a migration from the SG&A line one year into the revenue line a year, a year later. This is why everybody does it. So the, the word that I hear from investors from time to time, which I think is the, which is wrong, is inflection. When's the inflection coming?
Yeah.
You don't see an inflection. Basically, it's the same revenue curves just slightly rotated up, and you capture all that area under the curve over the life of the product. And so you spend on DTC because it works, and the metrics that enable it to give you that confidence both are historical precedents and also leading indicators like website traffic that you can measure moving up dramatically in response to television campaigns, and then activation of the website, as well as aided and unaided awareness that you can measure in the market. So there's plenty of things that tell you that you're on the right track.
Understood. Now, so it's been funny with LYBALVI , you know, typically gross-to-net going up is a bad thing for a company. The fact that you didn't go up is apparently the problem. But it seems like to me, like, you know, you have a carrot-and-stick approach. You're not gonna make a deal where you give up price without the requisite kind of volume gains, right?
Right.
Like, it seems like.
That's right
That you thought you might have done it in the back half of the year. You didn't do it. You alluded to guide into next year, right? And I guess the question I would get a lot if I was a PM is like, look, are they going to guide as if the gross-to-net on LYBALVI increases in 2024, right? Like, it sounds like not necessarily in 2023. It doesn't sound like there's gonna be a big change in the commercial portfolio, but into next year, you've had a year of DTC spend, you've had a year of clinicians using the drug, patients asking for it, right? The whole idea is, like, create that kind of grassroots demand and then have it pay off with gross-to-net.
Do you anticipate next year, you to engage with payers, in order to get broader access, right?
Yeah. The answer is yes, but I think we model it now that there shouldn't be some step function where you say.
Yeah
One quarter, everything changed the way. So the dynamic is as Akash indicates, which is in a category, if you launch an orphan drug in a rare disease without any therapies, you're in control of the payers. The payers are gonna basically take what you give. When you launch an atypical antipsychotic into a market with 15 generics, the payers are in charge, and the way they start is they say: "We're not gonna f-- provide access to your drug, notwithstanding how good or bad the medicine is. We're just not gonna do it because we have 14 other drugs." And you say, "Well, how about we get some formulary status?" And they say, "Well, how about a 60% rebate?" And we say: How about no?
Then it becomes essentially a test of how good the drug is. What we say is: We're willing to tolerate that lack of restriction to build notoriety and visibility of the drug in the marketplace. And what happens is physicians start prescribing it, fighting their way through the access barriers, and you go from selling zero to selling $97 million in your first year. Now that it's starting to move the needle a little bit. Second year, you double that. You're hopefully around $200 million circa. Now there's $200 million of commerce going on without any rebates coming back into the payers. They're like, "Wait a minute, every non-rebate-eligible life that we reimburse, we're losing money." So now, instead of at 60, they come in at a different level, and you have negotiating leverage.
So we are literally enduring restricted access right now in pursuit of maintaining higher gross-to-net, knowing that it's a longer game. Ultimately, the gross-to-net, we will make accommodations with these.
Yeah
Over time, but we hope to do so where the volume is co-compensating for whatever discounts you're making, and it's a more gradual process.
Understood. And, you know, like if you look at proxies for DTC, it's like you start to spend, and then maybe two quarters later, you see like You do see kind of an inflection in some of the other prior examples. I'm not sure it applies here with what you're describing. So in terms of your ROI on DTC spend right now, like, can you give any metrics in terms of what it's looked like so far?
No, you.
And then what.
You actually don't, you can't calculate the.
Yeah.
Financial ROI until 1 year into it.
Okay.
At least. But you can rely on some of the historical precedents that are, that are informative.
Are there any precedents that you think might be reasonable to look at when you're thinking about DTC?
Well, I think it would range from 1.X-2.X.
Okay.
Yeah.
Understood.
Yeah.
Now, you know, that midterm guidance, I mean, it's scary. We're almost going into 2024. That is a whole another discussion, but, as you think about Alkermes going to 25-30, right? Do you expect it to be a growth company when it comes to operating and margin, right? Like, it's... How do you view not just, you know, hitting 25% next year or, you know, whatever you have for 2025, but when you think about the cadence of EPS growth from 2025 to 2030, you know, at the board level with you, how much of a focus is there to actually continue to grow profitability over the next five years?
Well, so much of it depends on what your investment opportunities are in the pipeline.
Sure.
I think that at the board level, management level, we think that there's a real differentiating feature of being consistently profitable with expanding margins. That's attractive. That attracts investors other than biotech specialist investors.
Yep.
So we think there's a virtue to that, particularly if the maturation of 2680 continues apace, then you can also begin to see what the next leg of growth comes beyond that. The other thing is, you've heard us say, and again, don't overinterpret this, we have invested mightily in this commercial infrastructure over the last several years. And for a while, the criticism of Alkermes was that your SG&A is too high relative to your sales. And our argument was, actually, that might be true, but the solution is not to cut SG&A because the infrastructure you need to commercialize products at this price point, when the government is the principal payer in the U.S.
where the gross-to-net are very high, where there are huge restrictions to access, where you need policy people, and national accounts people, and medical affairs people, and all these. Well, the solution was to put more revenue through it, and that's what LYBALVI proved. Because you see the financial leverage with LYBALVI. We only added, I think, 50 salespeople and a small increment of SG&A to launch LYBALVI. If we can continue to do that, if we can leverage this commercial infrastructure, then all the profitability dynamics change as well. So long answer to a simple question, we believe in sustained profitability. We also believe the explosive valuation changes that happen in these companies is based on new products.
Yeah, understood. Maybe now, switching to the orexin asset. Obviously, early data in Type 1 narcoleptic patients. Before we get into the scientific question, I will say from a distance, this always strikes me highly competitive, multiple players on the market, large cap and small cap. With the IRA, you can't take just one molecule forward, right? I don't think you can go. It wouldn't make sense to just go take one compound into IH and then NT1 and NT2.
Precisely.
'cause then you're gonna get negotiated within nine years.
That's right.
This seems like a large cap pharma development program or a large cap biotech development program, right? So you know, I remember we spoke earlier this year, and you mentioned that World Sleep would be one of these inflection points in terms of your confidence on the strategic value of this type of asset. So I'll ask you kinda the same question I did a few you know, a few months ago, which is: Now that you have that data in-house and you know that you're probably gonna have to take a portfolio of molecules forward here, what do you think about doing this on a standalone basis for Alkermes versus looking at this more strategically?
I think we can do this on a standalone basis, which doesn't foreclose the idea of doing it strategically as well, because there's a whole world. There's a lot of territories, there's a lot of indications. So just to elaborate on what Akash is saying, the biology that we're seeking to exploit here relates to the sleep-wake cycle in the brain, the natural diurnal pattern of going to sleep and then activating and firing up to become alert during the day. A lot of pharmaceutical research has gone on on the sleep side. We have major sleep compounds. This is really the first time we're interrogating the wakefulness, what's called the orexin/hypocretin pathway.
Mm.
And that probably the most explicit embodiment of it is in Narcolepsy Type 1, which is actually a deficiency of that neuropeptide. So rather than superimposing some new pharmacology onto an intact brain, you're going to a disease where you're missing the natural neuropeptide, and we're effectively replacing with a small molecule drug. But to the extent that that biology is more promiscuous, the idea of activating that circuitry can relate to other diseases, post-IRA, you're gonna do that with other molecules. The chemistry space is actually quite limited because this is, without getting into too much chemistry, it's a complicated thing to do, to take an orally active compound, cross the blood-brain barrier, go to a target tissue in the brain, and then agonize, turn on a complex multidimensional receptor.
So it's not gonna be a place where you're gonna see all kinds of chemical diversity all interrogating this. Alkermes has this, this repertoire, and we're gonna deploy it. So what we're doing in the labs right now, preclinically, is investigating some of these adjacencies to narcolepsy. But that's why, to your point, the 2680 data were actually really important for that stage gate. Do we have therapeutic index? Do we have potency? Do we have selectivity to interrogate this target? And the answer seems to be yes. So we're gonna, we're gonna, we're gonna go for it now.
Okay, understood. And just maybe on another data set that came out of World Sleep that I think flew a little more under the radar was the TAK-861 healthy volunteer data. So I'm a little longer-winded question, but I wanna see if I'm crazy here, right? So we know the doses for TAK-861 and NT1, right? They're also 10 sub-10. We know that your doses-
You know them? You don't. We don't know them yet.
Well, at least on what they've disclosed on their European clinical trials page, at least.
Right, but we don't have the results of those data.
Right, we don't have the results. Correct. Assuming that they have good efficacy there, we know that ALKS 2680 is somewhere between 3-8 would be kind of your sweet spot for NT2 in terms of maxing out.
NT one.
Sorry, NT1, maxing out. So do you feel like the sleep-deprived healthy volunteer data that TAK-861 showed at World Sleep, at the 15 mg dose and the 40 mg dose, so doses that are much higher than what they had at least planned for NT1, there were no drug-related side effects at the 15 mg dose. They maxed out MWT. Is that a proxy for how perhaps your drug would behave as you go into an NT2 population, right? Like, does that logic make sense? Why or why not would the 861 healthy volunteer data at least be a read across to your compound?
I'd be careful of those multiple baton passes.
Mm.
I would say I would just wait for data. I think TAK-861's a very, very different drug than our drug. It's being dosed twice a day with a maximum of 10 milligrams because of a hepatic signal. It's demonstrating the pharmacology. It's demonstrating the powerful wakefulness aspects of it. But I think each of these different orexin agonists is going to be quite different in the clinic because we're asking a single molecule to do multiple things at once. Be orally bioavailable, to cross the blood-brain barrier, to not be a substrate for efflux pumps or any other thing. Potent, very potent, very selective, but also with a waveform that's consistent with the sleep-wake cycle.
Mm.
So it's PK, it's potency, it's oral bioavailability, it's blood-brain barrier permeability, all these things you've written about. It's. There's a lot of. So the probability of a bunch of different molecules all sharing the same features in the full elaboration in patient with the disease is very low. So I'm encouraged. One of the things about World Sleep that was encouraging is that it is literally world sleep. It happens every two years. All the experts are there. So we had a chance to sit down with all the experts. They're testing 861. They tested 994. They're testing 861. They're around 925. They all understand this biology and this pharmacology. So it gives you a real good sense of how your molecule stacks up vis-a-vis other ones in development.
I think, I think it is competitive because it's such a great opportunity, but I think there will be true differentiating features between the molecules.
Understood. Now, you know, it's interesting, there's this commentary from Takeda. I think even Jazz has alluded, maybe BID would actually be preferable 'cause they feel like you can optimize-
Or is it because their drug is BID?
Right. Well, so I think the reason they say that is, and this is, I think the deeper question is, okay, you had insomnia at the 8 mg dose, right? Seems like a non-target effect, seems like not a bad problem to have 'cause you wanna be efficacious first, right? But it's very interesting, you guys like kind of bookmarked the 3-8 mg dose. You kinda nailed it in terms of where you were maxing out efficacy pretty well in terms of your PK/PD modeling, right? Like, that is.
Yeah.
Kind of unusual.
Yeah.
So can you talk about, you know, EEG, your PK/PD? Like, what are you doing from a biomarker perspective in order to identify that kind of optimal therapeutic window? 'Cause the second question I often get from people is like: Look, Akash, we know that TAK-925, the 11 mg dose, when they went from acute to 7 days, you know, you saw a drop in efficacy. The 44 mg dose, you know, it, it was maintained, right? So, like, why is Alkermes so confident that they're gonna not see tachyphylaxis with the doses that they kinda tested? And to me, it seems like you guys are doing something with EEG or, or, or biomarkers that might be a little differentiated here.
So what gives you the confidence that the range that you kind of saw at World Sleep will actually be the efficacy that holds up with longer-term dosing?
So the foundational knowledge is what you said. In an earlier study of an intravenous compound, there was some evidence of some degradation in the maintenance of wakefulness test, not to zero, but to some decrement with extended dosing. That was seen at a lower dose, but not at a higher dose, suggesting that potency can, or dose can overcome that tachyphylaxis. So the answer to the question is, you learn it empirically.
Mm.
So what we love about having such a potent compound with such a wide therapeutic index, and remember, we did not hit a maximum tolerated dose in the SAD. So we feel like we have a long way to go. To the extent there is tachyphylaxis, that you can dose through it over time. So we have a very, very potent compound that's maxing out the MWT test at 8 milligrams, and the AE profile at 8 milligrams was very mild.
Mm.
So the hypothesis is that as you move into NT2 patients, the whole dose response curve shifts. You'll need a bit higher dose, 2-3x, to drive wakefulness, but also your side effect profile will shift alongside that. But even if it weren't to some extent, we have such good tolerability at 8 already, we feel like we have room to explore. But the answer will be in the data, and so I wouldn't believe anybody who tells you they know this a priori based on EEG or animal-
Yeah. Understood. Now, in terms of, you know, we see with a lot of these G protein-coupled receptor targets, frankly, that there is even- you see, you might see tachyphylaxis on efficacy, but you also see blood pressure, polyuria. You see a waning of that-
Initially.
in the TAK-994 paper. So if you were to kind of hypothesize how that evolves as you go from acute dosing, where patients have never had orexin, to your point, like they're completely deficient of that neuropeptide, to something where, you know, NT1 patients have been taking it for, let's say, 7 days or, you know, a month or longer, why... You know, should we be so quick to write off the 8 mg dose as, you know, somehow not having a tolerable safety profile? How do you expect that insomnia profile to change with longer term dosing in one patient?
It's funny, at World Sleep, Akash, I don't know if we talked to you about this, but sitting down with all the experts, the resounding statement was, "Do 8 milligrams as your test in phase II," for that very reason.
Yeah.
You'll probably see an accommodation on the side effect profile. To the extent you saw tachyphylaxis, you'd be able to, you'd wanna explore that, particularly in phase II , 'cause in phase II, it's not a phase III. You're looking to establish those parameters of the dosing. So I think we're quite confident now that we'll be testing 8 milligrams in the phase II study.
Understood. Now, in terms of the data that you're gonna have in Q1 2024, level set that for us, right? Like, the goal in NT1 was max out MWT, show a dose response, identify a therapeutic window. Is the goal in IH and NT2 also the same thing, max out MWT, see, see a dose response? Is that a fair expectation in those populations?
I think it's different in the NT2s because it's a different phenotype.
Right.
The pretest hypothesis was that in NT1s, there would probably be a fairly tight distribution around the dose because people are deficient in orexin. You're giving them orexin for the first time to a naive system. They should have a fairly consistent response. That's exactly what we saw. NT2 is a much more kind of Gaussian distribution of orexin tone.
Mm.
So if we test four or eight patients, we may see much more variability in the response. We don't know. We're blinded on that yet. We're enrolling those patients now. But that's where therapeutic index becomes so important, because in the real world, I believe, in NT2 and IH, people will titrate to efficacy. And if you have tolerability across a wide spectrum, you have the ability to do so. But if you're limited to, you know, X dose per day or some other dosing limitation, you won't have the flexibility to do that.
Understood. So last question. You mentioned you're blinded on the data, and yet you were able to kinda make that comment, "We think our NT2 dose is gonna be 2-3x.
That's based on modeling.
In our NT1." Okay, so that's based on your modeling?
Yeah.
And any color on the range of doses that you're exploring in NT2? Is it gonna be similar in terms of this, you know, four doses,
Yeah, it's roughly. I actually don't know the... I can't recall the, but it's effectively that. Take the NT1 doses, multiply them by 2-4x, I think is where we probably are, and that's what we'll be testing. And we'll see. Like you said, we were really lucky or smart on the NT1s. We picked that 1, 3, and 8.
Yeah.
Hoping to get a minimally effective dose and a maximally effective dose. That's exactly what we got, so we happened to just center that correctly. Whether we're so fortunate in the NT2s is to be determined, but we can always adjust the dose with more clinical experience.
Maybe lastly, in terms of moving into phase II and beyond, right? Can you walk through, like, could we see a, a trial initiated for you guys in the first half of next year?
Sure.
Okay.
Yeah, we're flooring it right now. So coming out of World Sleep, we effectively had enough information to sit with the KOLs and design the phase II. Those protocols are being written right now. We're activating CRO. We'll be going to FDA, we'll be going to ethics, IRB submissions. You know, the whole machinery of the multinational, multicenter phase II is now being ramped up.
For NT1, NT2, or just NT1?
Yes. So we'll see.
Yeah.
If it's faster to segregate the two so that we can run independent protocols, we'll do that. Depends on how, probably how fast we get clarity on that NT2 dose.
Understood.
In this first group of patients. So that's, that's still to be determined.
Understood. I know we're out of time. I will get you guys out of here. Thank you very much, Richard. This is awesome.
Thanks for showing up today, everyone.
Thanks again for everyone for joining us.