Great. I want to welcome everyone to the Piper Sandler Healthcare Conference. My name is Biren Amin. I'd like to welcome our next company. We have Centessa Pharmaceuticals and their President of the Orexin program, Mario Alberto Accardi. Welcome, Mario. Great having you. I know we've talked a lot over the last several months. Clearly, a lot of exciting developments and data sets coming out of the company. Maybe you know, before we kinda dig into the program, can you just talk a little bit about the Orexin Agonist platform, the potential differentiating attributes, and then we can kinda go into each of the programs?
Absolutely. Hi, Baron. Hi, Ron. Great to be here. Thank you for inviting us. Taking a step back, Centessa Pharmaceuticals is a clinical-stage biotech company predominantly focused on what are the most exciting targets we've seen in neurosciences, which is orexin biology, orexin agonism. It's an extremely exciting time for us at Centessa. We recently provided an update to our phase II study, in the first few cohorts, across more or less 50, 55 patients across narcolepsy type 1, type 2, and idiopathic hypersomnia. And it really, we're the first company to have shown a robust data set across all three indications with clinically meaningful, obviously statistical significance, efficacy and endpoints across all three indications.
And this is really a pivotal milestone for us because it really helps cement the position, as a company, as really a pure play, orexin company and leveraging this data set for the next stages of the clinical development. And I think you, in the introduction, you rightly said, we have created a platform here because we're not stopping at our lead asset, ORX750. We have additional molecules in the pipeline. We have ORX142 currently in phase I and entering patient studies in Q1 of 2026. We have ORX489 commencing phase I studies in, again in Q1 of 2026. And these are some of the most potent and selective orexin 2 agonists in clinical development today. And this franchise really enables us to move even beyond the rare hypersomnias.
This is why the narcolepsy type 2 and the IH components are a phase II update. It's so critical for us because we think it really helps de-risk a lot of the other non-orexin loss indications across neurodegenerative and psychiatric conditions. For us, it's really, I think, the beginning of a new chapter here as we build on the program and the assets.
So clearly, a lot of broad potential across neuropsychiatry outside of narcolepsy, NT1, NT2, and IH. But just, I guess, to focus on that opportunity as it relates to ORX750 on NT1, NT2, IH. What are your thoughts in terms of just, you know, number of patients that are available for each of those buckets? And then clearly, you know, I think everyone views NT1 as de-risked, you know, given the data sets that's been around for the class and with other orexin agonists. But talk to us a little bit about your data sets in NT2 and IH and how that's potentially de-risked into those cohorts.
Yeah, absolutely. I think there were a lot of question marks as to will an orexin agonist ultimately work outside of narcolepsy type 1. Remember, in NT1, an orexin agonist is essentially providing a functional cure for the disease. You're mimicking this missing peptide called orexin A. And, we're very excited that already on our second dose of 1.5 mg , we demonstrated a potentially best-in-class profile with greater than 20-minute change from baseline, half of the patients above 30 in absolute terms in the MWT. ESS down from 19 to five, right, was a huge, huge delta. Really speaks about the fact that these patients are doing very, very well, probably also in the later parts of the end of the afternoon. Cataplexy suppression down to 0.13x incidence rate ratio. So, really very, very good efficacy profile and very well tolerated, right?
No visual disturbances in the 1.5 mg cohort. Bearing in mind, this is just the beginning for us because we're dose-escalating higher. There's still 50 patients as we speak right now that are currently being dosed at higher indication across all three indications. The NT2 and the IH cohort were, I think, de-risking in so many ways because it's, I think, the first robust data set that showed that you can actually achieve meaningful efficacy also in indications where there isn't a loss in orexin tone, which, as you can imagine, is just de-risked so many other indications. And in NT2, at our second dose of 4 mg, we achieved a greater than 10-minute change from baseline. And today, there are no drugs approved, right? Standard of care does not have shown a greater than 10-minute change from baseline. So it just bodes very, very well.
ESS in NT2 at the 4 mg dose down from fifteen, sixteen all the way down to eight. So again, restoring that normative component, that normal indicator according to Epworth Sleepiness Scale. And then in IH, again, at the first dose level, 2 mg showed clinically significant and statistically significant clinically meaningful and statistically significant in the very first dose on the MWT and other efficacy measures. So really demonstrated best-in-class pro-potential across all three rare hypersomnias. We have one drug to go after the three indications, which is important, right? There's so much overlap sometimes between these indications. So it's very important to have a single asset for all three, and ultimately, obviously, very, very well tolerated.
And you know, you mentioned the MWT improvements across NT2 and IH with NT2, you know, greater than 10 minutes. In IH, it was that significant, clinically meaningful. I don't think the company has disclosed what the exact delta was yet. When we talk to clinicians, you know, the question and even I think with investors, the question is, what's considered normalization on MWT? What is the company hoping to achieve with these profiles? Because, you know, you've got several companies in the space. They're all presenting data in different ways. And so I think everyone's trying to understand how to, you know, analyze the data that's, you know, that you can then potentially translate into the real world and, and what's important to clinicians as well as to patients.
Absolutely, so the approval endpoint is the change from baseline, right? So that's why it's very important to focus on that, especially in indications like NT2 where the baseline can be quite variable. It's not as homogeneous as a narcolepsy type 1 kind of profile on the MWT, and so very important to focus on the change from, from baseline. According to the American Academy of Sleep Medicine Guidelines, you need two minutes improvement on the MWT and two-minute reduction on the ESS in order to be deemed clinically meaningful, right, so again, that, that just speaks to what patients are wrestling with today in terms of standard of care, which by all means has a very, very important place, in, in the treatment landscape of these patients.
But I think it gives you an idea of just how well orexin agonists could do for these patients, right, by seeing these much larger change from baselines. Now, in terms of what is normal wakefulness, well, on the ESS, for example, you need less than 10 for patients to essentially be classified as normal, right? So we hit five in NT1. We hit eight in NT2, which is a great achievement in the first few doses of the phase II study. On the MWT, it's more difficult to know because there hasn't been really a lot of studies that have shown what is the MWT of the general healthy population. I was just in Italy visiting PIs last week. I would probably score, you know, an MWT of probably 15 minutes just now based on my jet lag.
So it really depends, Biren. I would say that there's literature that says it's about 20 minutes on the MWT. Other studies have shown 30. I think in reality, the higher the better, right? But that's why it's always important to really focus on the change from baseline.
Is there ever a concern of MWT being too high?
Could be. I think, remember, the MWT is capped at the 40 minutes. So the 40 minutes is a, an artificial cap, right? It doesn't mean that you need to score 40 to be deemed normal. I think you raise a really important point, which is the life of a narcolepsy patient does not end at 4:00 P.M., which is the last session of your maintenance of wakefulness test, right? There are four sessions, if you remember, over a period of eight hours. You start at 10:00 A.M. Last readout is at 4:00 P.M. It doesn't end there. That's why it's really important to look at other metrics: cataplexy rate, ESS, patient-reported outcomes.
The fact that you're scoring an ESS of five in NT1, that suggests that there's really good symptom control all the way up to the later parts of the afternoon because there's questions in the ESS around how patients are doing in the afternoon. And similarly for cataplexy, right? If you're a 0.13 incidence rate ratio, it just means that there's really good cataplexy control all the way out to the later parts of the day. So again, I think it's really important to look at different perspectives and around these different endpoints.
You mentioned the durability. And I think in the NT1 MWT, I guess, measured at two hours, four, six, and eight hours. And we kinda tend to look at, I guess, the average where companies have presented the average, not necessarily at six and eight hours. Can you talk a little bit about the durability of the 750 profile across those four time points in the clinical trial and the importance of that, and the fact that you are able to see activity later into the day?
So we haven't disclosed the individual time points. I don't think none of the competitors have, if I'm not mistaken, because it's obviously pretty sensitive data, as you can infer PKPD. But here's a hint. We have shown the individual time points in our phase I study in our sleep-deprived 5 mg dose at 11:00 P.M. showed very, very consistent sustained effect all the way out till 7:00 A.M. And if you look at the 7:00 A.M. delta, patients were on 10 minutes, which is a similar baseline to what you could expect, for example, in some of these NT2 patients and broader rare hypersomnias. And ORX750 had 35 minutes, almost 35 minutes.
So that speaks to the sustained efficacy over the diurnal part of the day because the drug has been designed, as a QD drug, first and foremost, and it has been designed to have a relatively flat Cmax to AUC ratio with an extended AUC in order to have that symptom control extend all the way up to the later parts of the day with a QD with a QD dose, right? We have full dose linearity across all doses tested, right? Also in our phase I study, we've gone higher than the 4 mg MAD, full dose linearity. And we have rapid onset of action and fully linear PK, which I think speaks to just the quality of the ADME PK properties of the drug.
Now, having said that, we do have also the option to use split dosing, right, in our phase II study, and we've used it also in our phase I study, not because we have to, Biren. It's because we want to introduce maybe additional flexibility, right, to your point around, you know, what is the good efficacy threshold? Well, if you put patients in control potentially of when to take a second dose, then that's just another potential competitive advantage we have in the toolkit because you can go from a QD drug to a split dose. You cannot go from a BID drug to a QD drug, if that makes sense.
Oh, that's interesting, and not every patient is the same, right?
Yeah, exactly.
They don't necessarily. It's not a one-size-fits-all dynamic, so it clearly offers optionality for patients that want that split dosing to be able to dose it because it QDs may not be, you know, may not be for everyone, I guess.
Exactly. And we think it's all about flexibility. So we could have already selected those levels of this phase II study of this update from a couple of weeks ago and be very comfortable of our potential best-in-class profile across a range of different indications. But we want to cater for 100% of the patient's experience. So what happens if there's patients that want to stay up until 11:00 P.M., until midnight, right? We want to be able to offer that with a drug with ORX750. That's why it's important to do that dose optimization in your phase II study so that by the time you start your registration program, which we're commencing in Q1 of next year of 2026, which could be in one indication, it could be in all three, we really need to wait to see how the data unfolds for the next 50 patients.
At that point, right, we will have covered the bookends in your execution as you said, to provide that maximum flexibility to patients because obviously there's different people have different commitments, right? Maybe patients need to drive back home. Maybe patients need to be up late at night one evening. And it's, it's important to provide that flexibility. It would be more difficult if you had a drug that simply could not be dosed higher for whatever reason, didn't have that therapeutic index, didn't have that horsepower to get you into an extended duration of action. At that point, you have a problem because then you need to introduce maybe multiple dosing regimens.
And so you've talked about, you know, you've dosed an additional 50 patients, you know, what's been disclosed.
Yeah.
At higher doses as well as at split dosing?
Absolutely. We've gone higher in QD, and we've also, looking at, split dosing.
When should we expect data from the [cohort]?
We, we will be providing an update in Q1 of 2026, right, along with probably some more information on registrational program design. But again, we still haven't given guidance beyond that. Like, we really need to see the data from the next 50 patients to then decide whether it'll be an update across all three indications, which cohorts, or, or, or maybe a single indication. So, we need to see how the data unfolds.
I guess, on the phase III registrational program that you're potentially embarking on in Q1 2026, will you initiate all three cohorts in parallel, or will they be kind of in a staggered fashion based on how the phase II finishes?
We have full flexibility. Exactly right. It depends from the next cohorts, right, that we'll be completing over the next weeks, months to see where we are in Q1 and see whether we're ready to start in parallel across all three or maybe stagger a few by a few weeks. We have full flexibility. Right now, we feel very, very confident around starting the registration program in Q1 and very confident around the profile of the drug. Right now, we really haven't seen, based on the disclosure, an AE that has defined the therapeutic index of the drug. So we want to continue, I think, really extracting as much efficacy as possible out of our ORX750 in order to really get to cater for 100% of patient needs.
On the split dosing, is there a time difference on from first dose to second dose? And what is that ideal time difference that the patient would have to take that second dose?
That's one of the things that we're looking at in our phase II study. So it's we're doing all of this work in order for us to be set up for success with the registration program because you really don't want to be playing around with different doses that you haven't explored in your registration program. We want to go in with a certainty that, you know, if we dose X, we're going to get Y, right? That's why we want to get out of the phase II study so we can go into the registration program with confidence and ultimately deliver a best-in-class asset.
What's the potential for AEs with the split dose regimen versus a QD regimen? Higher risk of insomnia, for example, given the second dose is being taken later in the day? Or do you feel like because it's split dosing, that you know that really shouldn't be a risk?
Yeah. I'm not too concerned about insomnia. And I think, Travers, you mentioned this before. You almost want to dose patients till they get insomnia because you can always scale back, right? You cannot do the opposite, right? You cannot dose higher if you have the drug that just doesn't have the right therapeutic index. On the split dosing, I wouldn't say that you're introducing additional risk of insomnia. You're just putting additional flexibility, right? In reality, then the time of the second dose can probably change, right, to better suit patient needs. As a byproduct of adding flexibility, obviously, because the split dose does show a reduced Cmax, right, obviously, compared to a QD dose, we will be looking at whether that changes anything from a tolerability perspective. Again, not because we have to, right?
The tolerability profile of ORX750 is extremely well tolerated. We only had one dropout, and that was due to polyuria, but the patient had a pre-existing condition associated with urinary incontinence. So we feel very, very good around the tolerability profile of the drug. But obviously, it's, as a byproduct, we'll also be looking at whether split dosing could introduce any other welcome features.
And what's the confidence level? I mean, phase II is a two-week duration study with treatment. When you move into phase three, it's likely going to be an eight-week endpoint. So the correlation between two-week on efficacy MWT to eight-week, what's your confidence level on that? And given you're going to have to have, I think, pretty good confidence to design the effect size for each of the phase IIIs?
Yes. So in our case, the primary analysis was done at two weeks, right, with a double-blind randomized crossover study. So every patient acted as their own control, which is why we had such high statistical significance in our study. We have now changed to a four-week parallel design, right? And we've slightly increased patient numbers in order to cater for the different design. And we did this change because initially, we really wanted to test things out quickly and efficiently. We needed less patients with the crossover study. As soon as we saw that enrollment was going so well, so much excitement to get on the drug, on orexin agonist, and then we moved to a four-week parallel. However, I think, Biren, the competitors have answered this question.
If you look at the two-week data set, right, across different endpoints in NT1 and NT2, once you hit two weeks, you have a sustained effect size all the way, way out through to week eight and beyond. So we feel very, very confident around using week two to select doses for longer-term, longer-term treatment.
You know, the company's talked about best-in-class profile. How does one define that? Is it based on MWT delta? Is it based on, you know, MWT, cataplexy, ESS, and/or tolerability, or it could be any of those combinations?
I think as a best-in-class profile, I think it's reasonable to look at the totality of evidence. You want to look at the different efficacy measures, MWT, ESS, weekly cataplexy rates. You obviously also want to look at tolerability, right? You put everything together, and right now, I think it's very clear where we lie in terms of potential best-in-class across NT1, NT2, and IH for that matter, right? We're the first company to really show a robust study in idiopathic hypersomnia. That was at the first dose level, so I set up 2 mg . So it's very much the totality of evidence. And then obviously, as part of our phase II study, we are also gathering a lot of, I think, really interesting data points. We're doing PSG.
We're looking at the PVT, the SSD, executive function, cognition, all of these other endpoints that obviously will then which I haven't seen yet, but will hopefully feed into also the dose selection for the registration program.
Any anecdotal evidence from the 750 program on cognition, concentration, focus, anything? You know, we talked about, you know, broader potential outside of narcolepsy into other neuropsych conditions. Anything that you can glean from, like, the secondary endpoints or tertiary endpoints that you could potentially apply to 142?
Spot on, Biren. And that's exactly why we're looking at all these other endpoints in our rare hypersomnia studies is to help de-risk the other indications. So by now, I think a lot of the other orexin agonists have shown, and I don't think it's a surprise that orexin agonists can show beneficial impact, beneficial effects across cognition, executive function, fatigue, potentially even mood. And it's not a surprise because the orexin system is implicated along all these pathways. And I think this is why we're so excited about our NT2 data set because it really does help de-risk a lot of the broader indications where there isn't necessarily a loss in orexin tone. So if you think of so many neurodegenerative and psychiatric conditions where there's mood, executive function, cognition, excessive daytime sleepiness, fatigue, right? EDS is very prevalent in a number of other indications.
We're really excited about progressing 142 and 489 across multiple of these indications. We really view the rare hypersomnia space as the tip of the iceberg in many ways. We think this class could really be the next GLP-1 of neurosciences.
For 142, you know, it's clearly gone through healthy volunteer studies. How does the PK profile compare when you look at the 142 data set to 750 phase I data?
I would love to tell you more because the PK profile of 142 is really exciting and I think it's a PK profile that we think will work very, very well in some of these indications where you want a long and prolonged exposure, right, maybe even stay way above this threshold to really help with some of these symptoms and endpoints that we just described so we haven't really given much more information than that. But all assets, I think, have very, very different PK profiles. They're all very potent, all very selective.
What we've assembled here, which is, again, thanks to a unique structural biology technology that we started back when I started the company that we licensed into the original company, really helped us to design these high-quality assets, which is really like finding a needle in a haystack from a chemistry perspective.
That's great. I guess, and then on that, it so it feels like from an IP standpoint, you've got strategic advantages. Is that fair to say given?
So all of the IP was homegrown from my medicinal chemistry team from the very beginning, and we managed to crack these high-resolution co-crystal Cryo-EM structures at the binding pocket. It's very tight, right? We know exactly what residues are needed to activate the orexin 2 receptor to drive orexinergic transmission. And we then built in all the right ADME PK properties on these really cleverly designed small molecules. And then we created essentially an IP moat around these small molecules in order to protect us. So we feel, again, very good about the IP. The IP runs all the way into past 2040, so and everything is homegrown.
Great. And then on 489, how would that be positioned relative to 142?
It's again directed to these neurodegenerative and psychiatric disorders where the PK profile may lend itself for specific indications versus 142, and at the same time, they're distinct assets, so that means very different commercial dynamics, and ultimately allows us to go after very different indications with different assets but using the same drug class.
Great. I think we're about out of time, but, you know, congrats on all the progress. I think, you know, it'll be exciting to see more data from 750 next year along with the phase III initiation. So a lot of hard work, and I'm sure a lot of work that goes into it. So congrats on all of that to you and your team.
Thank you, Biren. Thank you for having us.
Thanks.