Good afternoon. Thanks so much for joining our 45th Annual Healthcare Conference. I'm Stacy Ku, part of the biotech team, with my colleague Vish Shah, and we're very happy to be hosting Saurabh Saha and Mario Accardi from Centessa Pharmaceuticals. Thanks so much for joining. Also, thanks to Kristen Sheppard, who's in the crowd. So for some background, we initiated coverage of Centessa recently in January. We're very optimistic for the orexin receptor 2 agonist class opportunity. I think it's going to be a revolutionary class of drug and should transform the sleep space. We've been following the story for some time now, so happy to welcome you both back for our second fireside chat. Obviously, a very well-deserved transformation of the company following proof-of-concept results, and looking forward to digging in. So just first, let's take a step back and walk through the different considerations of ORX750.
Why is it so important to have selectivity to receptor 2 and increased potency?
Yeah, well, thanks, Stacy, for the invitation, and knowing that you have a PhD in orexin biology, I probably have not been as intimidated before. We should be asking you the questions, but look, it's a tough biotech climate out there, I'll say, but we are fortunate at Centessa to be in a very unique position. We have now the focus of the company as almost a pure-play orexin company that's focused on a target that's probably one of the most well-validated in neuroscience in decades. On top of that, we've just come off last year, where we've had announcements on our data from our phase 1 study showcasing that we have what looks to be a potential best-in-class molecule with regard to efficacy and a tolerable safety profile, and on top of that, we're in a market that I think is largely underappreciated, just the rare hypersomnia market.
I hope we have a chance to discuss that. We're at a substantial market where there are very few players in the space right now, and we feel we have not only a best-in-class molecule, but a first-in-class opportunity in idiopathic hypersomnia and NT2, with nearly $500 million of cash and a cash runway into mid-2027, as well as a number of catalysts in 2025 that we have planned from ORX750 and our follow-up molecule, ORX142. We really are in a fortunate position to be very unique in biotech today, and so with regard to selectivity and potency, I'm going to turn it over to Mario, who is the gentleman responsible for leading the team to generate such great chemical matter.
Thanks, Saurabh. Hi, Stacy. In terms of how we've designed 750 and all our follow-up molecules, as you know, we've anchored these and managed to design them using very sophisticated structural biology and structure-based drug design capabilities that have allowed us to really understand the orthosteric binding pocket of the receptor and our medicinal chemistry team to really have a very clear roadmap of how to activate the orexinergic neurotransmitter system whilst maintaining very, very high selectivity of this GPCR. This has translated into a molecule, our first candidate, obviously, 750, entering phase 2, NT1, NT2, IH, that is almost equivalent to the native peptide orexin A and is around 10,000 times more selective at orexin 2 compared to orexin 1. That selectivity at orexin 2 versus 1 is important, as we've seen some preclinical studies to suggest that orexin 1 could potentially be implicated in some reward-seeking behavior.
By targeting orexin 2 only, what we've seen is the ability to completely restore excessive daytime sleepiness and completely suppress cataplexy with other orexin agonists. We feel very, very confident about having a molecule that is very selective at orexin 2.
As we then kind of obviously move on to what we've seen in your phase 1 proof-of-concept data, do you want to talk about what you think confirms best-in-profile competitive positioning? And were you surprised by how well your molecular design translated to the results? Of course, important is efficacy, but just as important is the side effect profile. What do you think is attributing to that selectivity and potency as you think about side effects?
As Mario mentioned, we wanted to generate a small molecule that nearly phenocopied the function of the endogenous orexin A peptide. The goal was to have a potency very similar to the orexin A peptide, as Mario mentioned, but also when administered and taken orally, this drug can be given once daily. It had a rapid onset of action, but yet have the effect sustained over the course of a day. A patient can stay awake for however long they wanted during the course of the day. The selection criteria was even far more nuanced than that. We wanted to make sure that we had a PK profile that essentially mimicked the orexin A peptide itself.
That implied having a PK with a narrow Cmax to Ctrough ratio, minimizing the area under the curve that's lost above the concentration that you need to have of the drug to stay awake. What we've learned from our phase one studies is, yes, we've achieved nearly maxing out the MWT in healthy volunteers that are acutely sleep deprived, giving the drug at 11:00 P.M., and then in the early hours of the morning, seeing these subjects awake and awake and alert is almost science fiction in the most difficult setting, because these subjects have extremely high sleep pressure. They also have their circadian rhythm kicking in at that point as well, where it's probably a perfect storm in terms of trying to keep someone awake. With ORX750, we were able to do that.
That, coupled with the fact that I think it frankly surprised us that given the data that we've seen from others, that we didn't have the same kind of AE profile that I think people were expecting from an orexin agonist. They probably were expecting that as you increase dose, you would see a cluster of those on-target AEs emerging over the various doses, from insomnia to pollakiuria to hypersalivation to blood pressure increases. You see those all kind of clustered together as you increase in dose. In our case, when we went from 1 mg to 2 mg to 2.5 mg to 3.5 mg to 5 mg, and then doing the sleep studies at 2 mg and 3.5 mg and 5 mg , what was apparent is that we see beautiful linear PK from a small molecule that frankly, in my 20 years, I haven't seen a molecule with that beautiful PK.
Linear PK is super important. And we've also seen that we have linearity in terms of the effect size, in terms of the responses we're seeing with MWT in a placebo-adjusted manner. As we increased dose, we saw increases in being able to keep subjects awake. However, what we didn't see is any appreciable differences in the on-target AE profile compared to placebo.
And of course, our KOLs tell us that insomnia and pollakiuria is likely on target for receptor 2. And at the higher doses, we did see some instances just lower in comparison to maybe some of the other programs, which kind of gives us a little confirmation that your drug is doing what it's supposed to be doing. Talk about maybe very briefly before we move on to the phase two, what AEs were you absolutely looking to avoid and have not seen so far in your program?
Yeah, so we wanted to avoid cardiotox, hepatotox, any kind of visual disturbances, hallucinations. I think some of these that you've seen with other orexin agonists, again, our goal was to design a molecule that could have a narrow Cmax to Ctrough. And if some of those truly are on target, then having a lower Cmax always helps. But we honestly don't know what they are, whether they're on target or off target, because we haven't seen them.
So maybe talk about the doses that you've selected for your phase 2A basket trial. Maybe talk about the number of patients. Obviously, there's different numbers of patients selected for different indications, but I do think it's a nuance that's important to kind of discuss.
Yeah, what's interesting about our phase two is that it was well-informed from our phase one, and I think this is often maybe missed, in that we generated essentially what is the primary endpoint data that someone would have for registration studies, including one of our competitors. I mean, that is their single primary endpoint, which is the MWT. We generated that data in the most difficult population to treat, which is healthy volunteers, in our phase one setting, so when we went into phase two, we did not go in blind or cold into that design of that study or the number or the dose selection in that study, so we were informed basically by the PK, and I've mentioned this before.
Some of you have heard me say that we are able to predict almost to a nanogram per mL in terms of the concentration we need to stay above to keep people awake in the MWTs, and if you drop below that by one or two units, you are effectively probably not going to be able to stay awake, so it is pretty incredible at how precise you can be in terms of predicting when someone's going to be able to stay awake, at what point in the day, based on the concentration of your drug, so based on all that data that we generated in the phase one setting, we were able to then extrapolate into phase two and decide, OK, 1 mg for NT1, which is equivalent to maybe 3 mg in a healthy volunteer POC setting.
You get some accumulation of drug, which you see with most small molecule drugs. It's probably around 3.5 mg . At 3.5 mg in our healthy volunteer POC setting, we saw an MWT. I think it was 34 or so. Then similarly, extrapolating our NT2 and IH doses at 2 milligrams, that's equivalent to somewhere between 2.5 and 3 milligrams, which is where we saw a 32-minute MWT. If you look at the data we generated in the phase one setting, it very much informed the dose selection in the phase two setting. In the phase two setting, we didn't predefine doses. We can select, after unblinding each dose cohort for each of the indications, they're all running separately at their own pace.
We can then decide how much to dose increase or decrease based on what we're seeing at either the 1 mg or 2 mg level or higher.
I want to talk about kind of the potential to unblind per cohort per dose, but this is a dose-ranging study. Our assumption is that you're going to go after multiple doses for the phase 2A within each indication, so how many doses do you think is necessary? And maybe as we think about NT1, I think we understand well, but as we think about NT2 and IH, is this a starting dose? Would you then move on to another dose? How should we think about kind of your flexibility in your phase 2A?
Yeah, so this is a dose range finding study. So the objective here is, and the goal is to find an optimal dose for registration studies. Really, this is very much the goal of this study. And if we take a look at the study design and that potential for unblinding at the end of the cohort and really pegging together PK, safety, tolerability, and a very holistic view of efficacy, not just MWT, but also ESS, KSS, patient-reported outcomes. We're looking at cognition, sustained attention, so many different measures. We were doing PSG and looking at the serum orexin in NT1. We're going to put all of that together and essentially help us inform the dose of the next cohort, which could be a dose up for cohort 2 or a dose down. Now, clearly, we at least want two cohorts because we want the separation between doses.
But we have flexibility to move in any direction.
I think I'd just add, it's really important that when we go into phase three, we have a nice dose range selected, because at the end of the day, you want to be able to treat every patient who has narcolepsy or idiopathic hypersomnia effectively. You don't want to have an artificial dose cap where someone then a dose cap on the label, for example, where someone needs more drug, but they just can't get it because whatever, their own physiology or they want to stay awake longer at night.
It's important to have a nice wide dose range to give patients and physicians the flexibility to dose up or down, which I think is unique for ORX750 in that because we have this TI that we've established in our phase 1 setting and we have linear PK, that we have the ability to have this potentially wide dose range to offer patients and the most patients who could benefit from this drug.
Again, it sounds like you are able to unblind per dose per cohort. So if there's one, I know enrollment is maybe still early. We're going to ask about that. But would you be able to kind of move one aspect of the basket trial ahead into phase three if you find the right dose, it's time to move on, or maybe NT1 versus NT2? You have that type of flexibility?
We have enormous amounts of flexibility in this phase 2 design. It's one IND that covers NT1, NT2, and IH. Each one is essentially their own study, and each indication has the ability for patients to be randomized either to four weeks continuous treatment, serve as their own control with the placebo-controlled crossover design, but also a withdrawal design in the study as well, where patients will get drug for two weeks, they'll be on placebo for two weeks, and they get drug again for two weeks, which could help us determine a potential future design in a phase 3 setting. Each study, each indication, each dose cohort, they can run at their own speed. They're not interconnected or dependent on either. It's essentially a horse race.
It could be a photo finish at the end with all three reading out where we need them to read out, or one could run ahead of the other. But we retain that kind of ability to see the data after each dose cohort, as was mentioned, but then decide whether we're done and we've hit the benchmark that we've set.
And I know we asked this question this morning, so hopefully you all are now ready to answer it or not answer it. But our assumption as a phase 2A is you're going to get blinded safety data as it's coming in. Is that a fair assumption to make?
There's a select group of people that are seeing the data, as with most any study in any company, and so we mentioned this before, that in 2025, we'll have the data from NT1, NT2, and IH.
Wonderful. And how's enrollment going?
There is tremendous excitement in the community right now with orexin agonist, as you know, given some great work done by those who have gone ahead of us, and we're noticing that patients and physicians both are recognizing that these drugs could potentially be functional cures. That's the word that's been used in the space, and look, I mean, when you have standard of care today giving you an MWT in the mid-single digits or high-single digits to a drug that nearly maxes out the scale and does it in a way where patients aren't feeling overly hyperstimulated, but feeling alert and not sleepy, I think that is something that many patients will clamor to be on a study like that, so I think it serves all of us well who are developing drugs in the orexin agonist space.
We also get some questions around the timing and length of the study. Our best understanding is four weeks of treatment with two weeks of placebo. As we think about Takeda's study, they started with eight weeks length of study. Alkermes is at six weeks, and you all are at four weeks. You all mentioned this morning that TAK-861 is four weeks for IH, which was confirmed by the company. So kind of curious, do you want to talk about the efficacy, the signal, the decision-making around kind of your timelines and clinical trial design?
Yeah, if you look at clinical precedent, and again, Takeda have been the front runners here, the TAK-861 data is very repeating. There has been sustained efficacy between the four weeks ESS and MWT all the way out to the six months, and then it was presented last September. So that is a very, very, very strong signal of sustained efficacy over time in NT1. There's also some clinical data showing sustained efficacy of orexin agonist in narcolepsy Type 2 over an eight-week treatment period. So I think in narcolepsy Type 1, at least, Stacy, what tends to happen is that the system normalizes after a few doses, starts to normalize. And from what we've seen is that you tend to reach a steady state in terms of effect size within four weeks of chronic dosing. That's why we're very much anchored on that four weeks.
In narcolepsy Type 2 and idiopathic hypersomnia, we believe the first dose, a single dose, is actually probably very predictive of long-term effect size and long-term efficacy outcomes. I think it's just our study design is an extremely efficient study design in terms of timing to really get us that answer, help us answer that key question, which is finding that optimal dose so then we can quickly move forward.
What's important for NT1, NT2, and IH is that NT1 patients generally cluster, if you were to do an MWT, probably in the low single digits, there's not too much variability there. For NT2 and IH, one has to be very careful on how you select the patients and which patients you select and go into the study. I say this because ultimately, you want an effect size that you can actually extrapolate and make sure your drug is working as it should. What I specifically mean there is if you start at a high baseline MWT in IH or NT2, that can make that very difficult to draw a conclusion. If you notice, our study design allows for, in the screening phase, an MWT and an ESS that's required as part of screening.
So we're essentially picking a very homogeneous population of patients with a predefined cutoff for our MWT. And that will hopefully allow us then to generate the data in the NT2 and IH setting where hopefully it'll be interpretable. And it will be a large enough effect size with a few number of patients to be able to determine where we should be going in terms of phase 3.
It sounds like you're really focused on making sure baseline characteristics are well understood before trying to, let's say, interrogate efficacy.
That's correct.
Understood. So we're going to get our phase 2A data. And just to double-check, after that, going straight into phase three, is that correct? Is there any other efficacy or safety data that needs to be shown?
The plan is to move towards registration studies after this phase 2A. That is really the goal of this study, Stacy. I don't think we've given comments on how that study, registration studies, are going to look like, but that's very much the goal.
You all are going to be collecting longer-term safety for both phase 2 and phase 3. That is kind of the stated goal for now.
Absolutely. A very robust clinical development plan, and the goal is exactly that.
Understood. So this year, we're going to get a lot of updates from both Takeda, Alkermes, and you all. So maybe briefly, how do you think about the impact of read-through? Is it all positive de-risking for the class? And the reason why we ask this, what's your level of confidence for NT2 and IH? And what are the remaining questions for these two other indications?
Yeah, I think we've answered the question around our confidence in NT2 and IH. I think the beautiful part about being in this space is, as I mentioned before, is that we and others have demonstrated that this class of drugs is bona fide to work and keep people awake and alert. I think the market opportunity here is completely, as I mentioned before, underappreciated. I think folks have anchored on a few billion dollars because I think that's what the existing sales are for the companies that are currently in the narcolepsy space. But if you look at the number of subjects that are patients that have narcolepsy or idiopathic hypersomnia and the number that are actually treated, it's about 50% are diagnosed, a quarter are treated, and the vast majority aren't happy with the standard of care today.
Any slight movement in the diagnosis or treatment rate takes that number to north of $6 or $7.5 billion. And that's a conservative estimate for what is a so-called rare hypersomnia market. And that's just the near-term opportunity that these orexin agonists have. So our hope is that everyone in the space, the very few limited number of people in a very large market, continue to have great data. I think the old adage holds true, as rising tides raises all boats. And I certainly hope we all do well in the space.
Wonderful. You talk about the opportunity for the orexin receptor 2 agonist class. I think we're close to $6 billion with $2.5 billion peak for you all. But obviously, let's put that in the context of maybe some of our KOL feedback, which has been, some say, very optimistic or maybe even overly optimistic, because they tell us 80%-90% of NT1 patients would try orexin receptor 2 agonists and 50%-60% of NT2 patients and IH patients would try kind of this class of drugs. So how do you kind of put your feedback from KOLs, your market research in the context of those comments?
Yeah, I'm glad you asked that question because it does deserve clarification. And I think people do need to go back and do some math and sharpen the numbers. We've seen the competitors in the space basically share similar data, I think, Stacy, to what you found. And we've spoken to others who have done the modeling. We've obviously done a ton of work here ourselves. And it does come out at a conservative number, about $6-$7 billion. I think there's much more opportunity there just given just the sheer total volume number of patients out there and those who will benefit from basically redefining standard of care in the space. So I think anchoring to where the market is today, a few billion between the incumbents in the space, is a starting point.
When you have a therapy that comes in that can completely upend where standard of care is today, I think you have to redefine your assumptions going in.
Do you think there's an emerging KOL appreciation for orexin receptor 2 agonists?
Absolutely. Absolutely. As I mentioned before, you see this generally as you're running studies in patients. You've heard it from others. Again, I like to say it, and it's been echoed, is these drugs are essentially like functional cures. They are a small molecule replacement essentially for a missing orexin A peptide. This is the type of excitement level that we're seeing in the physician community. Certainly, we're very attached to patient advocacy groups and patients in general who come and speak at our company. I think everyone is very hopeful that this is going to be the game changer in the sleep-wake space, specifically in the rare hypersomnias.
We would be absolutely remiss if we didn't discuss some of your pipeline agents, 142, 489. We do get questions on maybe the broader opportunity and potential next indications. But just maybe talk about these programs, differences in potency and selectivity, and what aspects of competitive companies, so like a Takeda or an Alkermes, what preclinical and clinical data have they shown so far that you think are most compelling?
Yeah, thank you for the question. We're extremely excited about what's coming next. This whole orexin agonist story started off as a functional cure, one of the best validated targets in neurosciences for narcolepsy type 1, then moved towards narcolepsy type 2, idiopathic hypersomnia, and the rare hypersomnias. But now finally, we're really starting to see a lot of momentum building up for what's next, which is essentially many, many neurological, psychiatric, neurodegenerative conditions where excessive daytime sleepiness, fatigue, even cognition and attention are some of the key, key symptoms where orexin agonists could have a transformational role. And we have a pipeline, as you say, of follow-up molecules. And I want to emphasize the word follow-up because there was nothing we wanted to improve with 750. There are no liabilities we really wanted to improve upon.
We really wanted to have distinct chemical entities and assets to be able to essentially further segment this very large market opportunity, which has very, very different commercial dynamics. Starting off with ORX142, our molecule for which we've given guidance, we're going to have sleep-deprived, healthy volunteer data already this year, and at some point, we'll probably be providing some more information about the indications for this molecule, and then we have ORX489 entering IND-enabling studies this year that, again, are very, very potent, our most potent orexin agonist, exactly equal potent to the native peptide. It is a remarkable medicinal chemistry feat, and of course, we have also more assets that we're working on underway, and what this does, Stacy, really helps us to segment the market opportunity and go after many indications such as Parkinson's disease, major depressive disorder, fatigue and MS, ADHD.
There are so many indications where orexin agonists could really have a dramatic role. It's not surprising that a lot of standard of care, like stimulants, for example, or modafinil, methylphenidate, are used in a lot of these indications, yet with limited efficacy. Watch this space. We've done our internal commercial assessment, and quite frankly, it's a number which is even quite difficult to quantify. We're looking at north of $10 billion alone.
Wonderful. So in the last few moments, we want to make sure we do pause for any potential questions. Well, with that, thank you very much for your time.
Thanks, Stacy.