Afternoon, and thank you for joining Guggenheim's 2026 Emerging Outlook Biotech Summit. I'm Debjit, and I'm one of the therapeutic analysts here, and my privilege to welcome our next presenting company, Centessa Pharmaceuticals. And from Centessa, we have the CEO, Mario Alberto Accardi. Mario, you've been on the hot seat for a few months now. Could you lay out your vision for the company, and how do you see 2026 evolve?
Yeah, totally, Debjit. Firstly, thank you. Thank you for having us. It's shaping up to be a transformational year for the company and for Centessa. Back when I started the company, Debjit, which was back in 2019, my vision was always to bring forward these orexin agonists, this very potentially very powerful biology, across multiple indications. So different assets, different PK profiles, really anchoring on the rare hypersomnias, narcolepsy type 1, a potential functional cure there, but really branching out into multiple other indications. Because fundamentally, this mechanism could potentially allow significant improvement in a number of these symptoms beyond just wakefulness and cataplexy suppression in NT1. And to finally get the opportunity to realize that vision here at Centessa is incredibly exciting. We are starting the year from a position of strength.
It was an incredibly positive year for us last year. I feel that now we have really laid the foundations for the rare hypersomnias. We showed back in November the potential best-in-class profile for ORX750 across narcolepsy type 1, type 2 , idiopathic hypersomnia. Now looking to start registrational studies in at least one or maybe all three indications this quarter. And we're really morphing into essentially a pre-commercial stage company on the rare hypersomnias. On top of this, as you know, we have a pipeline, some of the most potent and selective orexin agonists, all enabled by the structural biology platform that I put together back when I started the company to really take us into this breadth of other indications across multiple other neuroscience indications, and really morphing into a fully fledged neurosciences company.
That is very much part of the vision that I originally had for the program since the very beginning.
Thank you for that. So last year, when you had data in November, there was still another 50+ patients in dose escalation. Now that you're thinking about both once-daily and split dosing, have you significantly expanded the scope of the program, or it's still about 50+ patients?
Well, there's obviously a number of different cohorts, both QD and split dose, across all three indications that are running. They're not running in parallel. They're all staggered because they're... obviously, there's different enrollment and different patient numbers per dose cohort. And as part of our Q1 disclosure, right, this is our next update, the aim is to provide Phase II update from at least one of these indications. It really depends from the number of cohorts that are gonna be completed by the time of the disclosure of the update, and together with that, the commencement of the registrational studies. And it'll be also, as part of that disclosure, an opportunity also to convey my vision, my strategy for the company as we progress also the pipeline across these other indications.
Back to your questions around the 50 patients, we, although the plan is to start registrational studies for at least one, potentially all three this quarter, we may also be in a position where we may decide to add, post the disclosure, maybe an extra cohort, for example, in one of the indications. That's definitely something that is on the table, and we'll need to make a data-driven decision. And that comes from a position of not because we need to do that. I think we already showed a potential best-in-class profile, definitely in NT1. We also showed great NT2 data with a greater than 10-minute change from baseline at essentially the second lowest dose at 4 mg.
Back in November, we had only showed the first dose at 2 milligrams, but we already communicated that we had statistically and clinically meaningful impact on multiple efficacy measures in IH. So it really will have to be a data-driven decision. My goal as a company, Debjit, is, is for us to be not only best in class versus the companies that are clinical, in clinical development today. I wanna be best in class versus companies that are yet to come, versus orexin agonists that are yet to enter the clinic 5, 10 years from now, right? And I want to make sure that we use that opportunity now. Now is the time. We're ahead of the competition in NT2. We're ahead of the competition in IH.
So, very keen in making that happen because 750 has been designed from the very beginning to have a wide therapeutic index, right? That is one of the premise of the chemistry and the small molecule, and we wanna make sure that we harness the full power of this biology as reasonably possible.
So when you decide on which subset of... whether it's NT1, NT2, or IH, you plan to disclose data on, what's that threshold? Do you wanna push patients to the point they are, you know, become insomniacs? So you've basically hit MTD and then disclose the data because that's your max, or, how should I think about it?
Yeah, it's a great question, and I think the objective here, as part of the data disclosure, is to really only release data that we think is the end all be all, right, for that indication. If we think there is more room to play with another indication, then we'll push that specific data update for the indication out to a little bit later. You raise a good point, right? Which is a point we've discussed at length internally, is if there isn't a specific AE that defines the therapeutic index of the drug, right? If we can keep harnessing the power of this biology and really pushing those efficacy measures higher and higher, why stop? Right.
I think part of the answer is probably what may end up happening is that your maximum tolerated dose may indeed be in the form of insomnia, right? As you start extending the duration of action to 16, 18+ hours, what is gonna happen is obviously, then you will reach a state of where you have very high rate of insomnia. Now, that's a good thing, because if you can get to that, it means that you've covered the full dose range to help control diurnal symptoms for all three indications. And at that point, you provide that bookends of doses and patients one day, together with their PIs, they can then find the optimal dose, right? That optimal dose that works for that specific patient.
It's much better to be in a position where you can reduce the dose down, right? And find the right dose for the patient than be in the opposite position. Be in a position where you cannot dose higher, right? For whatever reason, sort of limited absorption or risk of idiosyncratic liver tox, whatever that may be, right? And so to be in that position is far from ideal, because it means that patients will be in a position where they may end up going back on polypharmacy. We fundamentally believe that this is a monotherapy approach, right? There is nothing to suggest that orexin agonists will need to be co-administered with other mechanisms.
So why not, to your point, get to that maximum dose and then provide the bookends commercially for PIs and patient to provide that maximum flexibility.
Okay, so where does split dosing fit in this paradigm, given that split dosing—your competitors are pursuing split dosing, you've explored split dosing. Do you—are you doing it to get there, or do—you're doing it because you can do it?
So the whole premise for the design of ORX750 was obviously a QD drug. And we've proven, if you look at our NT1 data, 1.5 milligram, we're getting greater than 20 minutes change from baseline on the MWT. We're getting ESS down from 18 to 5, right? Normal is 10. We're getting down to 5. There's questions on the ESS around how patients are doing in the later part of the day, in the afternoon. The fact that we're at 5 suggests there's, there's very good diurnal symptom control with a single dose. Our weekly cataplexy rate, very clearly best in class, 0.13 incidence rate ratio, 87% reduction in cataplexy, weekly cataplexy rate. That means that again, we're getting that extended control of cataplexy right into the later part of the day.
So to that point, the potential advantage of a split dose is that now we're putting the patient in control of. Let's say there's a 5-milligram dose, right? 5-milligram dose, we split that up into a 2- and a 3-milligram. That second dose, right, in the real world, you're giving an opportunity for the patient to further modulate their duration of action, right? Imagine the patient has a commitment later in the evening, right, and needs to drive back home. These drugs work very quickly on and off, right? So symptoms do come back once the exposure dips below this so-called wakefulness threshold. So by introducing a split dose, you're allowing the opportunity to better modulate, better control that duration of action, right? It's really about providing maximum dose flexibility to patients.
Now, we're in a position where we have a QD drug, and we can also offer split dose, which is very different from a drug that needs to be dosed BID. They cannot go from a BID to a QD, and that's, I think, a very important differentiation that we need to make.
You sort of talked about being in a position to start the registration study in one, if not all, three. How do you prioritize which one to go after first, and how can you be first in class in NT2 without having the data out? If you're gonna start a registration study, is it gonna be NT2 ? Because you already have somebody in NT one, why not go with NT2 first?
So the aim for us is definitely to be first in class in NT2 and idiopathic hypersomnia, right? That really has not changed. And if you look at our disclosure from November of last year, at the 4-milligram, it was a very clearly best-in-class profile. Also, if you compare it to standard of care, there is no drug approved that has shown greater than 10-minute change from baseline on the MWT, right? We're showing greater than 10. ESS dropped to 8, so we actually normalized already at 4 milligrams in NT2. Could we take that dose forward in registrational study? Absolutely. It would be a great drug for patients, given the unmet need. However, we've gone much higher in dose since then, since the release. Why?
Because, again, we want to push the dose higher and harness as much efficacy as possible with, with ORX750. So back to your question, it'll be a data-driven decision. If we feel we have all the doses ready by the end of the disclosure, then we're gonna start the registration study across all indications. If we feel that we-- there is still maybe one cohort we want to add, right? Because again, because we, we can, not because we need to, but if we feel that we're leaving still some efficacy on the table, given we're still so far ahead of the competition, why not, why not go ahead? Why, why not do another cohort? And you see, this is the strength of our clinical development strategy with a Phase II....
that I think it's an IR challenge, because obviously, we're unblinding after every dose cohort, right? So some people in the company get to see the data and then make decisions for the next cohort. But it puts us into a great position, right? A great position where we get a chance to find that perfect, that optimal dose for registration studies. Dose selection for these indications, right? It's absolutely critical, right? You can have the perfect drug, the perfect chemistry, but you can still select the wrong doses. So it's all about dose selection. And our clinical development strategy really gives us a competitive advantage because we're not gonna be in a position where we'll need to continue dose exploration in registrational studies. I think that's really the aim.
How would you define a win in NT2, given that the baselines are gonna be variable, right? It's not as tight as an NT1. You can have, although you have a cutoff of about 15 minutes in the Phase II study, you could have 5 minutes, 12 minutes. So do we look at a placebo-adjusted number, or trying to push patients to that normative wakefulness should be the criteria to define a win?
I would say in NT2, I mean, there's very few—if you look at standard of care, right? There, the effect size compared to, I think, what orexin agonist can do is still, it's probably very low, right? And that's why there is no drug approved today that has shown greater than 10-minute change from baseline on the MWT, unlike ORX750 on the Phase II study. And that was at second lowest dose of 4 milligram. So we think, and we hope there is plenty of room to do even better than that. But already that based on standard of care, already that effect size, already that response on the MWT and on the ESS is something that would be very, very meaningful, would be a very important place in the treatment landscape of patients.
But of course, we're going higher because in an ideal world, we can get to even better efficacy, right? Even greater absolute MWT numbers. Why not? That's really the purpose and the goal of the Phase II study.
How do you define normative wakefulness in NT2? Is it 20 minutes? Is it 40 minutes? What's that sweet spot that you would really like to land in?
Yeah, I think so I know there's some studies out there that have shown up to 20 minutes as the normative range. But in reality, to your point, because NT2 can be quite variable, and we actually have a very good inclusion criteria in our study. So, to avoid enrolling patients that are just not very sleepy, right? I think that that is something very important to mention. We're looking at sleepy patient populations, right, also in NT2. I think it's... The MWT is an important number, but it's also important to look at other endpoints, like the patient-reported outcomes, like the ESS. Again, dropping below 10 on the ESS, with what we've shown, down to 8, that suggests the patients are actually feeling normal. Now, can we get to lower than 8?
Right, that's part of the dose exploration that we've continued to do in the Phase II study.
In IH, what kind of magnitude of clinical benefit in ESS or IHSS would you like to get to?
So IH is where I believe there is probably the highest unmet clinical need, right? It's a very large population diagnosed and treated in the United States, about 120,000. But in reality, the numbers are a lot larger because the diagnosis rate is much lower compared to NT1 and NT2. And there's only one drug that is approved for IH, which is Xywav, so huge unmet clinical need. I think from an ESS perspective, I think the goal there is to, again, provide something that is obviously clinically meaningful. So you—the goal would, of course, to get as low as possible on the ESS again. I mean, as you know, the ESS in IH is actually the registration endpoint.
But what we're also capturing in our study is the Idiopathic Hypersomnia Severity Scale , 'cause we're very keen to understand how orexin agonists can impact specific symptoms of the disease, i.e., sleep inertia, for example, brain fog, right? These are all very specific symptoms to IH. So we're capturing all of that. Now, in the IHSS, although it's not a registration endpoint, it's an important endpoint. Ideally, dipping below 22 or having around a 10-minute change, delta on that would be, would be good. But we'll see. Again, the goal there is to continue dosing higher until we get a very meaningful clinical impact for these patients.
Are there any gating factors outside of being able to maximize the dose that prevents you from initiating the registration studies in terms of FDA interactions, et cetera?
Not at all, Debjit. We have everything we need to start registration studies. It's purely about selecting the most appropriate doses. As a matter of fact, we've had a number of interactions with the FDA, and what we can share is that we've had overall very positive feedback and overall alignment on the clinical development strategy. So we're, again, very excited to commencing studies this quarter. Let's see if it'll be in one or three indications. It'll be. It has to be a data-driven decision.
Given one of your competitors has sort of started a randomized withdrawal study for the European filing, are you planning to build that into your protocol right now for NT1?
Watch this space. Maybe it's something we already thought about. It's obviously a study design that EMA particularly likes. It's also a study design that's been utilized successfully in approvals for of drugs in idiopathic hypersomnia as well as narcolepsy... So again, I think this will be part of our Q1 disclosure. I'm very keen that now, given the maturity of the company, we're a company entering registrational studies. I think it's important to provide a little bit more color on the registrational study design and the strategy going forward, so something for this quarter.
Given that as of now, ORX750 looks like the only drug which can hit all three subsets, would you or have you thought of the BTD? Because some of your competitors already have it.
Absolutely. We're evaluating all potential regulatory measures to accelerate the development of the drug. Absolutely.
Assuming orexin agonists are commercially available, let's say in 2, 2.5 years from now, or 3 years from now, across all indications, where do you see or what do you think is the role of oxybates?
So sodium oxybates were first approved for cataplexy, right? They're very, very powerful cataplexy suppression mechanism and has an extremely important role for patients. Given the role, the mechanism of orexin agonist, I mean, it's not contrary to what people think, it's not a wake-promoting agent, it's not a stimulant. orexin agonists helps stabilize the boundary between sleep, wake, and REM. So you're essentially providing a functional cure for narcolepsy type 1 individuals. We fundamentally believe in the monotherapy approach, right? We believe all symptoms in narcolepsy type 1 should be able to be addressed with orexin agonists. The outstanding question is the disturbed nighttime sleep, and that is not a registration endpoint. I think it's something a few patients may complain about, but we're talking about very few patients.
If you already help with EDS and cataplexy, which are the main impairments of good quality of life, there's a lot of data out there to suggest, both preclinical and as well as clinical, to suggest that the lack of orexin is ultimately what's causing that sleep fragmentation at night, that wake intrusion, that disturbed nighttime sleep. So our thinking here is that if we can get some exposures of the drug that are sub wakefulness threshold throughout the night, right? To help consolidate the sleep architecture and sleep fragmentation, we fundamentally believe that that will essentially be the monotherapy approach. And that makes complete sense, right? Because if you look at the biology of orexin, it doesn't dip to zero at night, right? It dips about a third of the peak level.
This is something that, as you can appreciate, we're thinking about our own development strategy. We've got also different exotic formulations to try and achieve that if we think it's important. But in our Phase II study, apart from all these exploratory endpoints that we're looking at, in cognition, sustained attention, executive function, we're also looking at PSG. We're also assessing sleep quality.
All right, let's move to the next-gen asset, ORX142. We've got limited disclosures on the 89 healthy volunteers at the differentiated pharmacokinetics. Could you sort of elaborate on what did you or where are you going with that versus ORX750 or any of the competitor programs?
Yeah. So first of all, taking a step back, I fundamentally believe that this mechanism will help address many, many other symptoms beyond the symptoms that we're trying to resolve in the rare hypersomnias. Talking about fatigue, fatigue has a strong overlap with wakefulness, cognition, executive function. Some of the other companies out there working on orexin agonists have shown very strong data in improving sustained attention and cognition in narcolepsy type 1 patient population, which should translate very well for other indications. Orexinergic neurons branch out, have projections in different regions of the CNS, including the mood regions. So we think that the orexin agonist should also be able to treat symptoms in fatigue, executive function, mood, and other potential symptoms.
So I think here we're really talking about a potential GLP-1 type in neurosciences, where a lot of the validation comes from narcolepsy type 1, as it did for diabetes, for the GLP-1s. And then as we branch out and take a lot of the learnings that we're also amassing as part of our Phase II study, because we're looking at all these exploratory endpoints to help select indications for the pipeline. As part of this disclosure in Q1, we're also providing an update to the strategy and my vision there as we progress the portfolio. What I can share is that both ORX142 and ORX142, with ORX142 entering the clinic this quarter, very, very different PK to ORX750, completely different.
We have some, I think, very good ideas of how that will best fit for certain indications, but I think more about that in our Q1 disclosure.
So you think horses for courses strategy is more important than having one drug try and do everything?
First of all, there's benefits of having different PK profiles for different indications, right? That's something that is quite clear. The second part is obviously there's very different commercial dynamics, right? You're comparing, for example, the rare hypersomnia pricing and market compared to a very large market, for example, in fatigue and en masse or even in major depressive disorder . These are very, very different commercial dynamics, and it just makes a lot of sense to have distinct assets with also very different PK profiles to go after the specific indications. But this is definitely something that is part of our evolving strategy as a company, becoming really a neurosciences company. Starting registration studies, we're essentially pre-commercial. We're already thinking about commercial. We're already hiring in commercial this year.
And the idea here is really to help build this pipeline of multiple clinical studies that will help demonstrate that orexin agonists have the power to really help patients in many other symptoms beyond the rare hypersomnias.
Unfortunately, we have run the clock. I'm looking forward to the data update and getting the registration study started. Thank you so much, Mario.
Thank you, Debjit.
Really appreciate it.