Told not to start for another 20 seconds, so I'm just sitting here waiting.
We start already.
We're doing our own one-on-one. OK, excellent. We can start, I'm told. OK, excellent. Well, thank you, guys, for joining. Pleasure to have Alkermes Management here. I think I've been covering this doc now for, gosh, I don't know, 10 years or approaching something like that. So I feel like there's a lot of history. I first picked it up back when it was all a depression story. In any event, let's focus on things that are more important. I'll let you kick it off, Richard, and I'll jump right into it.
It's been a fun 10 years, Umer. Another 10 to go.
A lot of big ride.
So much of the story of our company has been this evolution, the passing of the baton between two completely different business models. We were a drug delivery formulations company working with big pharma to make improved medicines. And that was quite successful. But that led to a certain amount of top-line revenue capacity to segue then into creating our own products. And today, it's cool because August of this year, we actually ended our relationship with J&J on the long-acting antipsychotic Invega Sustenna in the U.S., which was our biggest product for many years, without missing a beat, because we've passed the baton now to our proprietary products business. So that provides a pure-play neuroscience company with a billion-dollar top line. It's highly profitable this year. But what's captured your attention and everybody else's attention is what the pipeline is.
And now, with the orexin program maturing in the way that it's maturing, it's not notional anymore. You actually diligence it. You can look at data. You can make decisions and try to assess the risk associated with what may be a really big new class of medicines. And I'm sure you'll want to talk about that.
So maybe let's jump right in. I want to do base business, all those things, perhaps later when we get to it. But let's jump right in on Orexin. You have a phase 2 going on in narcolepsy type 1 and a phase 2 going on in narcolepsy type 2 as well. Maybe just remind us, for starters, just the primary endpoint timelines, how you're thinking about it?
Just to orient folks, narcolepsy is categorized in these two principal domains, NT1 and NT2. The difference between them clinically, the patients present very similarly. The difference in NT1 and NT2 is NT1 patients typically are characterized by cataplexy. So with emotional highs or lows, they can have this form of paralysis. And physiologically, they have a deficiency of orexin. If you do a lumbar puncture and look in their cerebrospinal fluid, you'll find no orexin. So I often say nature has given us the perfect knockout model to understand what the orexin circuitry does in the brain. If you don't have orexin neurons, and it's a small bundle of 60,000 neurons or so in the hypothalamus, you have this excessive daytime sleepiness. And so NT2 patients share a similar clinical presentation, i.e., excessive daytime sleepiness.
But if you assay them, there will be presence of orexin in the cerebrospinal fluid in their brain. But something about that signaling in the brain is aberrant. It was a fairly logical conclusion that one would draw a priori that if you could replace the deficient neurotransmitter NT1 with a small molecule agonist of it, you would have a positive effect on the disease. Less clear was what would we do in NT2. That's been answered now empirically. We know now that both NT1 and NT2 respond beautifully to exogenous administration of an oral orexin 2 receptor agonist. The furthest afield so far would be what's called idiopathic hypersomnia, which is not characterized as narcoleptic, but often clinically presents very similar to it. These patients have orexin in their brains.
And we've shown in our hands with an orexin agonist that we can drive significant wakefulness in IH patients as well. So NT1, NT2, IH, we've tested those in a really elegantly designed 1b program, which led to data now that we've kicked off the phase 2 study in NT1 and a second phase 2 study in NT2. Both of those are activating sites and enrolling now. And we expect completion in the second half of next year. And I think Umer will give more precision around those enrollment curves, probably when we guide in February, because it's a little too early on that ramp to project within certainly where they're going to end up. But we're on track for the first half.
I guess what I was getting at was one trial is six weeks. Endpoint one is an eight-week.
Oh, I'm sorry.
How did you guys think about that?
The field is new. So with FDA, I mean, FDA is interested in understanding not just the safety and tolerability, but also the durability of the effect. So we are interrogating six weeks in one study, eight weeks in another study. Others have done eight weeks. And now others are doing 12 weeks as well. So I think what the whole field is trying to figure out, what the shape of that curve is over that duration period. Specifically, NT1s, it is such a severe disease, we wanted to minimize the time that a placebo arm would be off drugs because they have to wash off of everything. And there is a one in four chance in our phase two study of being randomized to placebo. And that is a long period of time to have no medicines at all if you have NT1.
But wouldn't that logic apply to NT2 as well?
NT2, FDA would prefer to see eight weeks and would prefer to see 12 weeks over time. So it's just how you sequence it all, because there's other trial designs that can guarantee people get access to drug in later studies as well.
Got it. Now, NT1 trial, your doses are four, six, and eight milligrams.
I'm sorry. Just to make that, based on data from others and in our own hands, we don't expect to see any diminution in efficacy between six and eight weeks. And data like these will help prove that. So we'll have a six-week time point. We'll have an eight-week time point, albeit in different populations. But it'll help inform whether there's any shape to that efficacy curve. And we don't expect there to be.
Got it. It looks like, based on prior evidence, three mg dose was pretty active. So six milligram might possibly be approaching the upper end of the efficacy. Is that your expectation? I know you have four, six, and eight. And four might be very active as well. But do you think six might be that optimal sweet spot?
That's why you do the experiment. I mean, we were active at one, remember. So there's a couple of things that go on in this category. For example, in single exposures, or even if you switch narcolepsy medicines, there's often a period of insomnia as patients accommodate a new medicine. So in single exposures, you might see a high level of insomnia like we did at eight milligrams, which we took as an efficacy marker. But it may actually be something that's really well accommodated by patients over a multi-week period. It's one of the interesting features of our phase 2 design that I don't think anybody else has done so far, which is that patients are washed out of all their medicines, and then they're randomized to one of three doses or placebo for a six-week period.
The primary assessment is from the baseline to week six on the MWT. Then there's an open-label seven-week period where all patients will go to the middle dose. And they get to choose then which dose they go to, which will answer an unanswered question so far, which is we tend to think, oh, more MWT is better. Let's max it out. Well, maybe not. Maybe some patients want to be on a lower dose. Maybe some patients want to be on a higher dose. How does that correlate with their patient-reported outcomes? And so we'll actually learn in a naturalistic setting what the right set point is. Our hypothesis had always been that you'll need variability in dosing. You'll need a range of doses to accommodate differences in the disease and differences in patients' preferences as well. And so our phase 2 will help inform that.
Then we'll have another, between the two studies, 160 patients' worth of data of patients in fixed lanes. And then in a naturalistic setting, then we have all the information to model for phase 3 and commercial.
Got it. Richard, are you still expecting NT1 trial? I remember you said it could be similar timing. But my sense was NT1 trial comes out at least a couple of months before the NT2 trial.
Today, if I had to guess, I'd say NT2 will finish before NT1.
Is that right?
We're the only player in NT2. Enrollment, site activation, everything is going fast. We're competing.
Faster than NT1?
Faster than NT1. And that's simply a lot of it's just the logistics of getting sites. Some of those sites are doing other studies. And so we're not competing for patients like we are.
Got it.
Now, both of them are effectively as we modeled them, but the NT2 is definitely ramping up quickly.
Got it. So I guess my question would be, because NT2 trial has the higher doses, 10, 14, and 18, one of the observations that could play out is a couple of instances of visual disturbances, perhaps, on the 14 and 18 mg dose. But before we go down that direction, could you remind us, how is the screening for visual function at baseline different in phase 2 versus phase 1?
Phase 1, there was no. I mean, that was just all comers, no baseline ophthalmic. And we don't know whether people have cataracts or intraocular pressure change. Who knows? In phase 2, there's a baseline ophthalmic assessment. And if anybody does report a visual AE, there'll be a workup from a qualified ophthalmic workup on it. Now, just remember, because people tend to overblow this, we saw one visual disturbance, quote unquote, and one in the NT2s and one in the IHs. We saw none in the NT1s. Both were. One of them was a blurriness of vision. That was mild and transient. And the other one was light sensitivity. Notice that the lights were brighter, mild, and transient. So in a phase 1 study, you're interrogating these patients on an inpatient setting for all types of phenomena.
And that's why you do it, to see if you can pick up any signals of it. Remember, in the parlance, when we're grading AEs, adverse events, there's mild, moderate, and severe. Mild means it's noted by the patient, but it doesn't affect them. And so when you see AE tables of new drugs, you're always looking for what's mild, moderate, severe, what might be dose limiting, what might be leading to discontinuation. So far, the overall category in our hands and in others' hands is that these orexin 2 receptor agonists are generally very well tolerated.
I guess having the baseline ophthalmic exam, how could or couldn't it change the odds of, for example, these two cases? How do you guys think about that? They may have had, for example, light sensitivity at baseline. Would that be the idea?
Not necessarily. But remember, in that setting, in the 1B setting, they're getting the MWT on an inpatient basis every two hours. So the lights are going on and off. And so noting that the light looked brighter when MWT was over, that's an AE. That's a quarter as an AE. In the outpatient setting, over a six-week period, how much of that is it real? A. B. If it were mild and transient and not noted, it wouldn't even be reported. So you'll get a much more naturalistic setting, in a much more naturalistic setting, what the tolerability profile is.
I guess maybe to drill down that more. So you said they're getting it inpatient. It's every two hours. And light goes on and off. I totally understand that. Those were the two instances, one in NT2, one in IH.
Correct.
I do remember there were a few more cases in your healthy volunteers. Could you remind us?
In the MAD, in the multiple ascending dose study.
Were they not inpatient?
Yeah. That's inpatient too. And again, that was sporadic.
How many cases were there in the healthy volunteers?
I don't remember. It was a handful. But it was interesting because they were all graded mild. If I look at Sandy, they're all graded mild. And one patient might have had it one day and didn't have it another day. Right? So it wasn't consistent. There was no pattern that would have been recognized.
But was it also blurry vision and light sensitivity? Or was it something else as well?
I believe it was exactly the same, blurry vision or light sensitivity.
They were having the same light issue? There was no MWT measurements. Wait, I'm forgetting now. In the healthy volunteers, was there a light issue or no?
I don't remember, Sandy. I'm looking at you.
There was only one case of photosensitivity.
Oh, so I guess by that logic then, Rich, wouldn't that imply that so if there was no light on/off issue and there were some visual disturbances, wouldn't that mean that you could technically still have visual disturbances in the outpatient trial?
Absolutely. This could be an on-target phenomenon. Absolutely. But if it's mild and transient, so I often say, OK, let's make the argument that it's on target. What are the evidence that it could be on target? Number one, it occurred about an hour after dosing in both cases, and it was gone after about an hour, so it's sort of consistent with the PK Cmax profile. But with the Ns of 1, particularly in a 1B setting where you're interrogating for side effects, it's very difficult to draw conclusions.
Got it. OK, got it. The location of this ongoing phase 2, so you said it's outpatient. This is all U.S.-based trial?
No, U.S., Europe, Australia.
OK, got it. Excellent. And as it relates to drug administration timing and timing of MWT measurement, could you remind us what that is? Because I remember phase 1 was like 11:00 P.M. and then we went 1:00 A.M., 3:00 A.M., 5:00 A.M. So remind us on this one.
No, that's not us. That's sleep-deprived healthy that they're talking about, so we don't think that assay is actually particularly relevant to narcolepsy. We went right from the SAD/MAD right into patients, and so that's a really strong feature of our program, that the data we generate in patients with the disease and the sleep pressure that's generated in someone with NT1 is different than somebody who's sleep-deprived until 11:00 P.M. In our phase 2 study, patients will go home with their drug, either dose or placebo, take it every morning in the morning, and then midway, so they have a baseline MWT assessment. Midway through the six-week period, they're going to come into the sleep lab where we're going to look at their sleep architecture, and then we'll do an MWT at week six.
And by the way, they're taking the drug what time of the day?
In the morning.
In the morning, OK.
In the morning. So it's a true once-a-day dose. And the data so far have shown that we can maintain a really strong dose-dependent MWT over a multi-hour period. The notional half-life of the drug, if you were just to measure it, would be six-to-eight hours. But it's not particularly informative because the real question is, what's the threshold concentration? What's the target concentration at the neurons in the brain for the period of time? And we've seen Umer present on this that this dose proportionality allows you, as you increase dose, to effectively drive duration as well. And that's when the drug is well behaved in that way, you can use dose as a way of dialing in both duration and magnitude of effect.
Got it. OK. So maybe stepping beyond narcolepsy then for a second, I think one of the other aspects of the story, which looks extremely interesting, is some of the Takeda data shown in narcolepsy patients, but on executive function, sustained attention, et cetera, memory. Maybe, A, remind us of your strategy and your new programs that are entering the clinic on that. But also, is there evidence beyond this Takeda phase 2 update on those parameters as well? Or is that the core?
Absolutely. Absolutely. It's a fascinating area of neuroscience, and I think in 10 years' time, we'll look back and say this was the beginning of a whole new area of neuropsychopharmacology. Because think about for how many years people were developing sleep medicines. Why do you develop a sleep medicine? Not to sleep better, so you're more awake during the day. This circuitry in the brain was only discovered like 20 years ago, so for the first time, we're using agents now to drive the circuitry in the brain, the natural circuitry that drives alertness, wakefulness, attention, so if you look at the way that these orexin neurons then project into other areas of the brain, it's very easy to develop hypotheses as to orexin 2 receptor agonists might affect attention and mood in particular, and we've shown some preclinical data on that as well.
So some of these features will derive simply from being more awake. But also, there's probably specific innervation in the brain that drives things like vigilance, attention, and things like that. So I think that the field is going to start with narcolepsy, which is the knockout model, replace the deficient neurotransmitter, recapitulate the natural state. The secret, of course, is doing it in a diurnal way. Because you can't just hit this receptor 24/7. You don't make a long-acting injectable. You need to follow the natural rhythm of the brain. And then I believe it's going to map onto certain psychiatric, neurological, and neurodegenerative conditions.
Maybe a little more specificity, what are those indications?
So we've shown data preclinically in pretty highly translatable models in ADHD and in mood. But as we map on that circuitry onto, there's three buckets that you would see us describe: psychiatry, where mood and attention are really central to it. Orphan diseases that are characterized by degenerative processes that lead to the phenotype of patients being excessively tired during the day. And then some of the chronic, larger indication neurodegenerative processes like MS fatigue or.
What about dementia and Alzheimer's? Could there be utility there?
That isn't the first place we would start. But I think that the most parsimonious explanations where we look at places where that circuitry is actually in the brain, this bundle of neurons is described the Orexin system as the master regulator of wakefulness and attention. So you wake, as you prepare to arise in the morning, what happens? Your blood pressure, your heart rate increases. You start preparing for the day. Your attention increases. Your wakefulness increases. And then that wanes towards the evening so you can go back to sleep. So think about all that circuitry activating, what that could map onto then in the cases of disease.
Got it. As you think about the new programs that are approaching the clinic now for these non-narcolepsy indications, do you have certain potency or PK parameters that you'd want to be optimizing for?
For sure. So the ticket to this party is very hard to get in terms of the chemistry. Because we're making orally active GPCR agonists. So these are G protein-coupled receptors in the brain. You typically need a significant amount of structure to agonize a GPCR. So it has to get out of your gut, into your bloodstream, into your brain. And then it has to mimic a PK profile that allows once a day or some favorable dosing regimen. It's not easy to do. And so absolutely, the sine qua non for all of these next candidates for us are the level of selectivity, potency, oral bioavailability. They can't be a pump substrate in the brain. They need to have right PK profile. And so you eliminate a lot of candidates before you ever get to the animal studies.
Got it. Maybe in the last minute or so, if I think about all the orals that have approached the clinic, we know there was Takeda's version 1, 994. We know there was a Jazz molecule and a Merck molecule. So at least three, of the six that I'm counting oral orexins that have approached, three have been terminated now formally. I guess any feedback on what happened on the Merck program? Because I would have thought they should have made progress. I know they had a bit of a PK issue. But I'm curious if there's any buzz around that program. Or was it safety related, I guess, is really where I'm going?
I didn't detect it. In contrast, they said it was terminated for business reasons.
Correct.
I think you and I had talked about when they put it in the clinic earlier this year, it was described as a twice-a-day looking at tolerability.
Correct. Three hours apart, right?
Yeah. So I think a number of pharmaceutical companies are going to be probing this pharmacology in their own hands to understand what it does. And then ultimately, there's going to be a handful that make it to the finish line. And I don't think all of them are going to make it to the finish line. Already, the data suggests that that's the case.
Did you guys have an internally generated version of the Merck molecule?
No.
Didn't have that?
No.
OK.
No.
Excellent. Anything we missed or we didn't touch upon just before we wrap up here from the audience? OK.